::Abi\n\nSame as `RefFromWasmAbi::Abi`#### type Anchor = UriError\n\nSame as `RefFromWasmAbi::Anchor`#### unsafe fn long_ref_from_abi(js: Self::Abi) -> Self::Anchor\n\nSame as `RefFromWasmAbi::ref_from_abi`### impl OptionFromWasmAbi for UriError\n\n#### fn is_none(abi: &Self::Abi) -> bool\n\nTests whether the argument is a “none” instance. If so it will be deserialized as `None`, and otherwise it will be passed to\n`FromWasmAbi`.### impl<'a> OptionIntoWasmAbi for &'a UriError\n\n#### fn none() -> Self::Abi\n\nReturns an ABI instance indicating “none”, which JS will interpret as the `None` branch of this option. \n\n#### fn none() -> Self::Abi\n\nReturns an ABI instance indicating “none”, which JS will interpret as the `None` branch of this option. \n\n#### fn eq(&self, other: &UriError) -> bool\n\nThis method tests for `self` and `other` values to be equal, and is used by `==`.1.0.0 · source#### fn ne(&self, other: &Rhs) -> bool\n\nThis method tests for `!=`. The default implementation is almost always sufficient, and should not be overridden without very good reason.### impl RefFromWasmAbi for UriError\n\n#### type Abi = <JsValue as RefFromWasmAbi>::Abi\n\nThe wasm ABI type references to `Self` are recovered from.#### type Anchor = ManuallyDrop<UriErrorThe type that holds the reference to `Self` for the duration of the invocation of the function that has an `&Self` parameter. This is required to ensure that the lifetimes don’t persist beyond one function call, and so that they remain anonymous.#### unsafe fn ref_from_abi(js: Self::Abi) -> Self::Anchor\n\nRecover a `Self::Anchor` from `Self::Abi`. \n\n#### fn describe()\n\n### impl Eq for UriError\n\n### impl StructuralEq for UriError\n\n### impl StructuralPartialEq for UriError\n\nAuto Trait Implementations\n---\n\n### impl RefUnwindSafe for UriError\n\n### impl !Send for UriError\n\n### impl !Sync for UriError\n\n### impl Unpin for UriError\n\n### impl UnwindSafe for UriError\n\nBlanket Implementations\n---\n\n### impl<T> Any for Twhere\n T: 'static + ?Sized,\n\n#### fn type_id(&self) -> TypeId\n\nGets the `TypeId` of `self`. \n T: ?Sized,\n\n#### fn borrow(&self) -> &T\n\nImmutably borrows from an owned value. \n T: ?Sized,\n\n#### fn borrow_mut(&mut self) -> &mut T\n\nMutably borrows from an owned value. \n\n#### fn from(t: T) -> T\n\nReturns the argument unchanged.\n\n### impl<T, U> Into<U> for Twhere\n U: From<T>,\n\n#### fn into(self) -> U\n\nCalls `U::from(self)`.\n\nThat is, this conversion is whatever the implementation of\n`From<T> for U` chooses to do.\n\n### impl<T> ReturnWasmAbi for Twhere\n T: IntoWasmAbi,\n\n#### type Abi = <T as IntoWasmAbi>::Abi\n\nSame as `IntoWasmAbi::Abi`#### fn return_abi(self) -> <T as ReturnWasmAbi>::Abi\n\nSame as `IntoWasmAbi::into_abi`, except that it may throw and never return in the case of `Err`.### impl<T> ToOwned for Twhere\n T: Clone,\n\n#### type Owned = T\n\nThe resulting type after obtaining ownership.#### fn to_owned(&self) -> T\n\nCreates owned data from borrowed data, usually by cloning. \n\nUses borrowed data to replace owned data, usually by cloning. \n U: Into<T>,\n\n#### type Error = Infallible\n\nThe type returned in the event of a conversion error.#### fn try_from(value: U) -> Result<T, <T as TryFrom<U>>::ErrorPerforms the conversion.### impl<T, U> TryInto<U> for Twhere\n U: TryFrom<T>,\n\n#### type Error = <U as TryFrom<T>>::Error\n\nThe type returned in the event of a conversion error.#### fn try_into(self) -> Result<U, <U as TryFrom<T>>::ErrorPerforms the conversion.\n\nStruct js_sys::WeakMap\n===\n\n\n```\n#[repr(transparent)]pub struct WeakMap { /* private fields */ }\n```\nImplementations\n---\n\n### impl WeakMap\n\n#### pub fn new() -> WeakMap\n\nThe `WeakMap` object is a collection of key/value pairs in which the keys are weakly referenced. The keys must be objects and the values can be arbitrary values.\n\nMDN documentation\n\n### impl WeakMap\n\n#### pub fn set(&self, key: &Object, value: &JsValue) -> WeakMap\n\nThe `set()` method sets the value for the key in the `WeakMap` object.\nReturns the `WeakMap` object.\n\nMDN documentation\n\n### impl WeakMap\n\n#### pub fn get(&self, key: &Object) -> JsValue\n\nThe `get()` method returns a specified by key element from a `WeakMap` object.\n\nMDN documentation\n\n### impl WeakMap\n\n#### pub fn has(&self, key: &Object) -> bool\n\nThe `has()` method returns a boolean indicating whether an element with the specified key exists in the `WeakMap` object or not.\n\nMDN documentation\n\n### impl WeakMap\n\n#### pub fn delete(&self, key: &Object) -> bool\n\nThe `delete()` method removes the specified element from a `WeakMap`\nobject.\n\nMDN documentation\n\nMethods from Deref<Target = Object>\n---\n\n#### pub fn constructor(&self) -> Function\n\nThe constructor property returns a reference to the `Object` constructor function that created the instance object.\n\nMDN documentation\n\n#### pub fn has_own_property(&self, property: &JsValue) -> bool\n\nThe `hasOwnProperty()` method returns a boolean indicating whether the object has the specified property as its own property (as opposed to inheriting it).\n\nMDN documentation\n\n#### pub fn is_prototype_of(&self, value: &JsValue) -> bool\n\nThe `isPrototypeOf()` method checks if an object exists in another object’s prototype chain.\n\nMDN documentation\n\n#### pub fn property_is_enumerable(&self, property: &JsValue) -> bool\n\nThe `propertyIsEnumerable()` method returns a Boolean indicating whether the specified property is enumerable.\n\nMDN documentation\n\n#### pub fn to_locale_string(&self) -> JsString\n\nThe `toLocaleString()` method returns a string representing the object.\nThis method is meant to be overridden by derived objects for locale-specific purposes.\n\nMDN documentation\n\n#### pub fn to_string(&self) -> JsString\n\nThe `toString()` method returns a string representing the object.\n\nMDN documentation\n\n#### pub fn value_of(&self) -> Object\n\nThe `valueOf()` method returns the primitive value of the specified object.\n\nMDN documentation\n\nMethods from Deref<Target = JsValue>\n---\n\n#### pub const NULL: JsValue = JsValue::_new(JSIDX_NULL)\n\n#### pub const UNDEFINED: JsValue = JsValue::_new(JSIDX_UNDEFINED)\n\n#### pub const TRUE: JsValue = JsValue::_new(JSIDX_TRUE)\n\n#### pub const FALSE: JsValue = JsValue::_new(JSIDX_FALSE)\n\n#### pub fn as_f64(&self) -> Option<f64Returns the `f64` value of this JS value if it’s an instance of a number.\n\nIf this JS value is not an instance of a number then this returns\n`None`.\n\n#### pub fn is_string(&self) -> bool\n\nTests whether this JS value is a JS string.\n\n#### pub fn as_string(&self) -> Option<StringIf this JS value is a string value, this function copies the JS string value into wasm linear memory, encoded as UTF-8, and returns it as a Rust `String`.\n\nTo avoid the copying and re-encoding, consider the\n`JsString::try_from()` function from js-sys instead.\n\nIf this JS value is not an instance of a string or if it’s not valid utf-8 then this returns `None`.\n\n##### UTF-16 vs UTF-8\n\nJavaScript strings in general are encoded as UTF-16, but Rust strings are encoded as UTF-8. This can cause the Rust string to look a bit different than the JS string sometimes. For more details see the documentation about the `str` type which contains a few caveats about the encodings.\n\n#### pub fn as_bool(&self) -> Option<boolReturns the `bool` value of this JS value if it’s an instance of a boolean.\n\nIf this JS value is not an instance of a boolean then this returns\n`None`.\n\n#### pub fn is_null(&self) -> bool\n\nTests whether this JS value is `null`\n\n#### pub fn is_undefined(&self) -> bool\n\nTests whether this JS value is `undefined`\n\n#### pub fn is_symbol(&self) -> bool\n\nTests whether the type of this JS value is `symbol`\n\n#### pub fn is_object(&self) -> bool\n\nTests whether `typeof self == \"object\" && self !== null`.\n\n#### pub fn is_array(&self) -> bool\n\nTests whether this JS value is an instance of Array.\n\n#### pub fn is_function(&self) -> bool\n\nTests whether the type of this JS value is `function`.\n\n#### pub fn is_bigint(&self) -> bool\n\nTests whether the type of this JS value is `bigint`.\n\n#### pub fn js_typeof(&self) -> JsValue\n\nApplies the unary `typeof` JS operator on a `JsValue`.\n\nMDN documentation\n\n#### pub fn js_in(&self, obj: &JsValue) -> bool\n\nApplies the binary `in` JS operator on the two `JsValue`s.\n\nMDN documentation\n\n#### pub fn is_truthy(&self) -> bool\n\nTests whether the value is “truthy”.\n\n#### pub fn is_falsy(&self) -> bool\n\nTests whether the value is “falsy”.\n\n#### pub fn loose_eq(&self, other: &JsValue) -> bool\n\nCompare two `JsValue`s for equality, using the `==` operator in JS.\n\nMDN documentation\n\n#### pub fn bit_not(&self) -> JsValue\n\nApplies the unary `~` JS operator on a `JsValue`.\n\nMDN documentation\n\n#### pub fn unsigned_shr(&self, rhs: &JsValue) -> u32\n\nApplies the binary `>>>` JS operator on the two `JsValue`s.\n\nMDN documentation\n\n#### pub fn checked_div(&self, rhs: &JsValue) -> JsValue\n\nApplies the binary `/` JS operator on two `JsValue`s, catching and returning any `RangeError` thrown.\n\nMDN documentation\n\n#### pub fn pow(&self, rhs: &JsValue) -> JsValue\n\nApplies the binary `**` JS operator on the two `JsValue`s.\n\nMDN documentation\n\n#### pub fn lt(&self, other: &JsValue) -> bool\n\nApplies the binary `<` JS operator on the two `JsValue`s.\n\nMDN documentation\n\n#### pub fn le(&self, other: &JsValue) -> bool\n\nApplies the binary `<=` JS operator on the two `JsValue`s.\n\nMDN documentation\n\n#### pub fn ge(&self, other: &JsValue) -> bool\n\nApplies the binary `>=` JS operator on the two `JsValue`s.\n\nMDN documentation\n\n#### pub fn gt(&self, other: &JsValue) -> bool\n\nApplies the binary `>` JS operator on the two `JsValue`s.\n\nMDN documentation\n\n#### pub fn unchecked_into_f64(&self) -> f64\n\nApplies the unary `+` JS operator on a `JsValue`. Can throw.\n\nMDN documentation\n\nTrait Implementations\n---\n\n### impl AsRef<JsValue> for WeakMap\n\n#### fn as_ref(&self) -> &JsValue\n\nConverts this type into a shared reference of the (usually inferred) input type.### impl AsRef<Object> for WeakMap\n\n#### fn as_ref(&self) -> &Object\n\nConverts this type into a shared reference of the (usually inferred) input type.### impl AsRef<WeakMap> for WeakMap\n\n#### fn as_ref(&self) -> &WeakMap\n\nConverts this type into a shared reference of the (usually inferred) input type.### impl Clone for WeakMap\n\n#### fn clone(&self) -> WeakMap\n\nReturns a copy of the value. Read more1.0.0 · source#### fn clone_from(&mut self, source: &Self)\n\nPerforms copy-assignment from `source`. \n\n#### fn fmt(&self, f: &mut Formatter<'_>) -> Result\n\nFormats the value using the given formatter. \n\n#### fn default() -> Self\n\nReturns the “default value” for a type. \n\n#### type Target = Object\n\nThe resulting type after dereferencing.#### fn deref(&self) -> &Object\n\nDereferences the value.### impl From<JsValue> for WeakMap\n\n#### fn from(obj: JsValue) -> WeakMap\n\nConverts to this type from the input type.### impl From<WeakMap> for JsValue\n\n#### fn from(obj: WeakMap) -> JsValue\n\nConverts to this type from the input type.### impl From<WeakMap> for Object\n\n#### fn from(obj: WeakMap) -> Object\n\nConverts to this type from the input type.### impl FromWasmAbi for WeakMap\n\n#### type Abi = <JsValue as FromWasmAbi>::Abi\n\nThe wasm ABI type that this converts from when coming back out from the ABI boundary.#### unsafe fn from_abi(js: Self::Abi) -> Self\n\nRecover a `Self` from `Self::Abi`. \n\n#### type Abi = <&'a JsValue as IntoWasmAbi>::Abi\n\nThe wasm ABI type that this converts into when crossing the ABI boundary.#### fn into_abi(self) -> Self::Abi\n\nConvert `self` into `Self::Abi` so that it can be sent across the wasm ABI boundary.### impl IntoWasmAbi for WeakMap\n\n#### type Abi = <JsValue as IntoWasmAbi>::Abi\n\nThe wasm ABI type that this converts into when crossing the ABI boundary.#### fn into_abi(self) -> Self::Abi\n\nConvert `self` into `Self::Abi` so that it can be sent across the wasm ABI boundary.### impl JsCast for WeakMap\n\n#### fn instanceof(val: &JsValue) -> bool\n\nPerforms a dynamic `instanceof` check to see whether the `JsValue`\nprovided is an instance of this type. \n\nPerforms a zero-cost unchecked conversion from a `JsValue` into an instance of `Self` \n\nPerforms a zero-cost unchecked conversion from a `&JsValue` into an instance of `&Self`. \n T: JsCast,\n\nTest whether this JS value has a type `T`. \n T: JsCast,\n\nPerforms a dynamic cast (checked at runtime) of this value into the target type `T`. \n T: JsCast,\n\nPerforms a dynamic cast (checked at runtime) of this value into the target type `T`. \n T: JsCast,\n\nPerforms a zero-cost unchecked cast into the specified type. \n T: JsCast,\n\nPerforms a zero-cost unchecked cast into a reference to the specified type. \n T: JsCast,\n\nTest whether this JS value is an instance of the type `T`. \n\nPerforms a dynamic check to see whether the `JsValue` provided is a value of this type. \n\n#### type Abi = <JsValue as LongRefFromWasmAbi>::Abi\n\nSame as `RefFromWasmAbi::Abi`#### type Anchor = WeakMap\n\nSame as `RefFromWasmAbi::Anchor`#### unsafe fn long_ref_from_abi(js: Self::Abi) -> Self::Anchor\n\nSame as `RefFromWasmAbi::ref_from_abi`### impl OptionFromWasmAbi for WeakMap\n\n#### fn is_none(abi: &Self::Abi) -> bool\n\nTests whether the argument is a “none” instance. If so it will be deserialized as `None`, and otherwise it will be passed to\n`FromWasmAbi`.### impl<'a> OptionIntoWasmAbi for &'a WeakMap\n\n#### fn none() -> Self::Abi\n\nReturns an ABI instance indicating “none”, which JS will interpret as the `None` branch of this option. \n\n#### fn none() -> Self::Abi\n\nReturns an ABI instance indicating “none”, which JS will interpret as the `None` branch of this option. \n\n#### fn eq(&self, other: &WeakMap) -> bool\n\nThis method tests for `self` and `other` values to be equal, and is used by `==`.1.0.0 · source#### fn ne(&self, other: &Rhs) -> bool\n\nThis method tests for `!=`. The default implementation is almost always sufficient, and should not be overridden without very good reason.### impl RefFromWasmAbi for WeakMap\n\n#### type Abi = <JsValue as RefFromWasmAbi>::Abi\n\nThe wasm ABI type references to `Self` are recovered from.#### type Anchor = ManuallyDrop<WeakMapThe type that holds the reference to `Self` for the duration of the invocation of the function that has an `&Self` parameter. This is required to ensure that the lifetimes don’t persist beyond one function call, and so that they remain anonymous.#### unsafe fn ref_from_abi(js: Self::Abi) -> Self::Anchor\n\nRecover a `Self::Anchor` from `Self::Abi`. \n\n#### fn describe()\n\n### impl Eq for WeakMap\n\n### impl StructuralEq for WeakMap\n\n### impl StructuralPartialEq for WeakMap\n\nAuto Trait Implementations\n---\n\n### impl RefUnwindSafe for WeakMap\n\n### impl !Send for WeakMap\n\n### impl !Sync for WeakMap\n\n### impl Unpin for WeakMap\n\n### impl UnwindSafe for WeakMap\n\nBlanket Implementations\n---\n\n### impl<T> Any for Twhere\n T: 'static + ?Sized,\n\n#### fn type_id(&self) -> TypeId\n\nGets the `TypeId` of `self`. \n T: ?Sized,\n\n#### fn borrow(&self) -> &T\n\nImmutably borrows from an owned value. \n T: ?Sized,\n\n#### fn borrow_mut(&mut self) -> &mut T\n\nMutably borrows from an owned value. \n\n#### fn from(t: T) -> T\n\nReturns the argument unchanged.\n\n### impl<T, U> Into<U> for Twhere\n U: From<T>,\n\n#### fn into(self) -> U\n\nCalls `U::from(self)`.\n\nThat is, this conversion is whatever the implementation of\n`From<T> for U` chooses to do.\n\n### impl<T> ReturnWasmAbi for Twhere\n T: IntoWasmAbi,\n\n#### type Abi = <T as IntoWasmAbi>::Abi\n\nSame as `IntoWasmAbi::Abi`#### fn return_abi(self) -> <T as ReturnWasmAbi>::Abi\n\nSame as `IntoWasmAbi::into_abi`, except that it may throw and never return in the case of `Err`.### impl<T> ToOwned for Twhere\n T: Clone,\n\n#### type Owned = T\n\nThe resulting type after obtaining ownership.#### fn to_owned(&self) -> T\n\nCreates owned data from borrowed data, usually by cloning. \n\nUses borrowed data to replace owned data, usually by cloning. \n U: Into<T>,\n\n#### type Error = Infallible\n\nThe type returned in the event of a conversion error.#### fn try_from(value: U) -> Result<T, <T as TryFrom<U>>::ErrorPerforms the conversion.### impl<T, U> TryInto<U> for Twhere\n U: TryFrom<T>,\n\n#### type Error = <U as TryFrom<T>>::Error\n\nThe type returned in the event of a conversion error.#### fn try_into(self) -> Result<U, <U as TryFrom<T>>::ErrorPerforms the conversion.\n\nStruct js_sys::WeakSet\n===\n\n\n```\n#[repr(transparent)]pub struct WeakSet { /* private fields */ }\n```\nImplementations\n---\n\n### impl WeakSet\n\n#### pub fn new() -> WeakSet\n\nThe `WeakSet` object lets you store weakly held objects in a collection.\n\nMDN documentation\n\n### impl WeakSet\n\n#### pub fn has(&self, value: &Object) -> bool\n\nThe `has()` method returns a boolean indicating whether an object exists in a WeakSet or not.\n\nMDN documentation\n\n### impl WeakSet\n\n#### pub fn add(&self, value: &Object) -> WeakSet\n\nThe `add()` method appends a new object to the end of a WeakSet object.\n\nMDN documentation\n\n### impl WeakSet\n\n#### pub fn delete(&self, value: &Object) -> bool\n\nThe `delete()` method removes the specified element from a WeakSet object.\n\nMDN documentation\n\nMethods from Deref<Target = Object>\n---\n\n#### pub fn constructor(&self) -> Function\n\nThe constructor property returns a reference to the `Object` constructor function that created the instance object.\n\nMDN documentation\n\n#### pub fn has_own_property(&self, property: &JsValue) -> bool\n\nThe `hasOwnProperty()` method returns a boolean indicating whether the object has the specified property as its own property (as opposed to inheriting it).\n\nMDN documentation\n\n#### pub fn is_prototype_of(&self, value: &JsValue) -> bool\n\nThe `isPrototypeOf()` method checks if an object exists in another object’s prototype chain.\n\nMDN documentation\n\n#### pub fn property_is_enumerable(&self, property: &JsValue) -> bool\n\nThe `propertyIsEnumerable()` method returns a Boolean indicating whether the specified property is enumerable.\n\nMDN documentation\n\n#### pub fn to_locale_string(&self) -> JsString\n\nThe `toLocaleString()` method returns a string representing the object.\nThis method is meant to be overridden by derived objects for locale-specific purposes.\n\nMDN documentation\n\n#### pub fn to_string(&self) -> JsString\n\nThe `toString()` method returns a string representing the object.\n\nMDN documentation\n\n#### pub fn value_of(&self) -> Object\n\nThe `valueOf()` method returns the primitive value of the specified object.\n\nMDN documentation\n\nMethods from Deref<Target = JsValue>\n---\n\n#### pub const NULL: JsValue = JsValue::_new(JSIDX_NULL)\n\n#### pub const UNDEFINED: JsValue = JsValue::_new(JSIDX_UNDEFINED)\n\n#### pub const TRUE: JsValue = JsValue::_new(JSIDX_TRUE)\n\n#### pub const FALSE: JsValue = JsValue::_new(JSIDX_FALSE)\n\n#### pub fn as_f64(&self) -> Option<f64Returns the `f64` value of this JS value if it’s an instance of a number.\n\nIf this JS value is not an instance of a number then this returns\n`None`.\n\n#### pub fn is_string(&self) -> bool\n\nTests whether this JS value is a JS string.\n\n#### pub fn as_string(&self) -> Option<StringIf this JS value is a string value, this function copies the JS string value into wasm linear memory, encoded as UTF-8, and returns it as a Rust `String`.\n\nTo avoid the copying and re-encoding, consider the\n`JsString::try_from()` function from js-sys instead.\n\nIf this JS value is not an instance of a string or if it’s not valid utf-8 then this returns `None`.\n\n##### UTF-16 vs UTF-8\n\nJavaScript strings in general are encoded as UTF-16, but Rust strings are encoded as UTF-8. This can cause the Rust string to look a bit different than the JS string sometimes. For more details see the documentation about the `str` type which contains a few caveats about the encodings.\n\n#### pub fn as_bool(&self) -> Option<boolReturns the `bool` value of this JS value if it’s an instance of a boolean.\n\nIf this JS value is not an instance of a boolean then this returns\n`None`.\n\n#### pub fn is_null(&self) -> bool\n\nTests whether this JS value is `null`\n\n#### pub fn is_undefined(&self) -> bool\n\nTests whether this JS value is `undefined`\n\n#### pub fn is_symbol(&self) -> bool\n\nTests whether the type of this JS value is `symbol`\n\n#### pub fn is_object(&self) -> bool\n\nTests whether `typeof self == \"object\" && self !== null`.\n\n#### pub fn is_array(&self) -> bool\n\nTests whether this JS value is an instance of Array.\n\n#### pub fn is_function(&self) -> bool\n\nTests whether the type of this JS value is `function`.\n\n#### pub fn is_bigint(&self) -> bool\n\nTests whether the type of this JS value is `bigint`.\n\n#### pub fn js_typeof(&self) -> JsValue\n\nApplies the unary `typeof` JS operator on a `JsValue`.\n\nMDN documentation\n\n#### pub fn js_in(&self, obj: &JsValue) -> bool\n\nApplies the binary `in` JS operator on the two `JsValue`s.\n\nMDN documentation\n\n#### pub fn is_truthy(&self) -> bool\n\nTests whether the value is “truthy”.\n\n#### pub fn is_falsy(&self) -> bool\n\nTests whether the value is “falsy”.\n\n#### pub fn loose_eq(&self, other: &JsValue) -> bool\n\nCompare two `JsValue`s for equality, using the `==` operator in JS.\n\nMDN documentation\n\n#### pub fn bit_not(&self) -> JsValue\n\nApplies the unary `~` JS operator on a `JsValue`.\n\nMDN documentation\n\n#### pub fn unsigned_shr(&self, rhs: &JsValue) -> u32\n\nApplies the binary `>>>` JS operator on the two `JsValue`s.\n\nMDN documentation\n\n#### pub fn checked_div(&self, rhs: &JsValue) -> JsValue\n\nApplies the binary `/` JS operator on two `JsValue`s, catching and returning any `RangeError` thrown.\n\nMDN documentation\n\n#### pub fn pow(&self, rhs: &JsValue) -> JsValue\n\nApplies the binary `**` JS operator on the two `JsValue`s.\n\nMDN documentation\n\n#### pub fn lt(&self, other: &JsValue) -> bool\n\nApplies the binary `<` JS operator on the two `JsValue`s.\n\nMDN documentation\n\n#### pub fn le(&self, other: &JsValue) -> bool\n\nApplies the binary `<=` JS operator on the two `JsValue`s.\n\nMDN documentation\n\n#### pub fn ge(&self, other: &JsValue) -> bool\n\nApplies the binary `>=` JS operator on the two `JsValue`s.\n\nMDN documentation\n\n#### pub fn gt(&self, other: &JsValue) -> bool\n\nApplies the binary `>` JS operator on the two `JsValue`s.\n\nMDN documentation\n\n#### pub fn unchecked_into_f64(&self) -> f64\n\nApplies the unary `+` JS operator on a `JsValue`. Can throw.\n\nMDN documentation\n\nTrait Implementations\n---\n\n### impl AsRef<JsValue> for WeakSet\n\n#### fn as_ref(&self) -> &JsValue\n\nConverts this type into a shared reference of the (usually inferred) input type.### impl AsRef<Object> for WeakSet\n\n#### fn as_ref(&self) -> &Object\n\nConverts this type into a shared reference of the (usually inferred) input type.### impl AsRef<WeakSet> for WeakSet\n\n#### fn as_ref(&self) -> &WeakSet\n\nConverts this type into a shared reference of the (usually inferred) input type.### impl Clone for WeakSet\n\n#### fn clone(&self) -> WeakSet\n\nReturns a copy of the value. Read more1.0.0 · source#### fn clone_from(&mut self, source: &Self)\n\nPerforms copy-assignment from `source`. \n\n#### fn fmt(&self, f: &mut Formatter<'_>) -> Result\n\nFormats the value using the given formatter. \n\n#### fn default() -> Self\n\nReturns the “default value” for a type. \n\n#### type Target = Object\n\nThe resulting type after dereferencing.#### fn deref(&self) -> &Object\n\nDereferences the value.### impl From<JsValue> for WeakSet\n\n#### fn from(obj: JsValue) -> WeakSet\n\nConverts to this type from the input type.### impl From<WeakSet> for JsValue\n\n#### fn from(obj: WeakSet) -> JsValue\n\nConverts to this type from the input type.### impl From<WeakSet> for Object\n\n#### fn from(obj: WeakSet) -> Object\n\nConverts to this type from the input type.### impl FromWasmAbi for WeakSet\n\n#### type Abi = <JsValue as FromWasmAbi>::Abi\n\nThe wasm ABI type that this converts from when coming back out from the ABI boundary.#### unsafe fn from_abi(js: Self::Abi) -> Self\n\nRecover a `Self` from `Self::Abi`. \n\n#### type Abi = <&'a JsValue as IntoWasmAbi>::Abi\n\nThe wasm ABI type that this converts into when crossing the ABI boundary.#### fn into_abi(self) -> Self::Abi\n\nConvert `self` into `Self::Abi` so that it can be sent across the wasm ABI boundary.### impl IntoWasmAbi for WeakSet\n\n#### type Abi = <JsValue as IntoWasmAbi>::Abi\n\nThe wasm ABI type that this converts into when crossing the ABI boundary.#### fn into_abi(self) -> Self::Abi\n\nConvert `self` into `Self::Abi` so that it can be sent across the wasm ABI boundary.### impl JsCast for WeakSet\n\n#### fn instanceof(val: &JsValue) -> bool\n\nPerforms a dynamic `instanceof` check to see whether the `JsValue`\nprovided is an instance of this type. \n\nPerforms a zero-cost unchecked conversion from a `JsValue` into an instance of `Self` \n\nPerforms a zero-cost unchecked conversion from a `&JsValue` into an instance of `&Self`. \n T: JsCast,\n\nTest whether this JS value has a type `T`. \n T: JsCast,\n\nPerforms a dynamic cast (checked at runtime) of this value into the target type `T`. \n T: JsCast,\n\nPerforms a dynamic cast (checked at runtime) of this value into the target type `T`. \n T: JsCast,\n\nPerforms a zero-cost unchecked cast into the specified type. \n T: JsCast,\n\nPerforms a zero-cost unchecked cast into a reference to the specified type. \n T: JsCast,\n\nTest whether this JS value is an instance of the type `T`. \n\nPerforms a dynamic check to see whether the `JsValue` provided is a value of this type. \n\n#### type Abi = <JsValue as LongRefFromWasmAbi>::Abi\n\nSame as `RefFromWasmAbi::Abi`#### type Anchor = WeakSet\n\nSame as `RefFromWasmAbi::Anchor`#### unsafe fn long_ref_from_abi(js: Self::Abi) -> Self::Anchor\n\nSame as `RefFromWasmAbi::ref_from_abi`### impl OptionFromWasmAbi for WeakSet\n\n#### fn is_none(abi: &Self::Abi) -> bool\n\nTests whether the argument is a “none” instance. If so it will be deserialized as `None`, and otherwise it will be passed to\n`FromWasmAbi`.### impl<'a> OptionIntoWasmAbi for &'a WeakSet\n\n#### fn none() -> Self::Abi\n\nReturns an ABI instance indicating “none”, which JS will interpret as the `None` branch of this option. \n\n#### fn none() -> Self::Abi\n\nReturns an ABI instance indicating “none”, which JS will interpret as the `None` branch of this option. \n\n#### fn eq(&self, other: &WeakSet) -> bool\n\nThis method tests for `self` and `other` values to be equal, and is used by `==`.1.0.0 · source#### fn ne(&self, other: &Rhs) -> bool\n\nThis method tests for `!=`. The default implementation is almost always sufficient, and should not be overridden without very good reason.### impl RefFromWasmAbi for WeakSet\n\n#### type Abi = <JsValue as RefFromWasmAbi>::Abi\n\nThe wasm ABI type references to `Self` are recovered from.#### type Anchor = ManuallyDrop<WeakSetThe type that holds the reference to `Self` for the duration of the invocation of the function that has an `&Self` parameter. This is required to ensure that the lifetimes don’t persist beyond one function call, and so that they remain anonymous.#### unsafe fn ref_from_abi(js: Self::Abi) -> Self::Anchor\n\nRecover a `Self::Anchor` from `Self::Abi`. \n\n#### fn describe()\n\n### impl Eq for WeakSet\n\n### impl StructuralEq for WeakSet\n\n### impl StructuralPartialEq for WeakSet\n\nAuto Trait Implementations\n---\n\n### impl RefUnwindSafe for WeakSet\n\n### impl !Send for WeakSet\n\n### impl !Sync for WeakSet\n\n### impl Unpin for WeakSet\n\n### impl UnwindSafe for WeakSet\n\nBlanket Implementations\n---\n\n### impl<T> Any for Twhere\n T: 'static + ?Sized,\n\n#### fn type_id(&self) -> TypeId\n\nGets the `TypeId` of `self`. \n T: ?Sized,\n\n#### fn borrow(&self) -> &T\n\nImmutably borrows from an owned value. \n T: ?Sized,\n\n#### fn borrow_mut(&mut self) -> &mut T\n\nMutably borrows from an owned value. \n\n#### fn from(t: T) -> T\n\nReturns the argument unchanged.\n\n### impl<T, U> Into<U> for Twhere\n U: From<T>,\n\n#### fn into(self) -> U\n\nCalls `U::from(self)`.\n\nThat is, this conversion is whatever the implementation of\n`From<T> for U` chooses to do.\n\n### impl<T> ReturnWasmAbi for Twhere\n T: IntoWasmAbi,\n\n#### type Abi = <T as IntoWasmAbi>::Abi\n\nSame as `IntoWasmAbi::Abi`#### fn return_abi(self) -> <T as ReturnWasmAbi>::Abi\n\nSame as `IntoWasmAbi::into_abi`, except that it may throw and never return in the case of `Err`.### impl<T> ToOwned for Twhere\n T: Clone,\n\n#### type Owned = T\n\nThe resulting type after obtaining ownership.#### fn to_owned(&self) -> T\n\nCreates owned data from borrowed data, usually by cloning. \n\nUses borrowed data to replace owned data, usually by cloning. \n U: Into<T>,\n\n#### type Error = Infallible\n\nThe type returned in the event of a conversion error.#### fn try_from(value: U) -> Result<T, <T as TryFrom<U>>::ErrorPerforms the conversion.### impl<T, U> TryInto<U> for Twhere\n U: TryFrom<T>,\n\n#### type Error = <U as TryFrom<T>>::Error\n\nThe type returned in the event of a conversion error.#### fn try_into(self) -> Result<U, <U as TryFrom<T>>::ErrorPerforms the conversion.\n\nFunction js_sys::decode_uri\n===\n\n\n```\npub fn decode_uri(encoded: &str) -> Result<JsString, JsValue>\n```\nThe `decodeURI()` function decodes a Uniform Resource Identifier (URI)\npreviously created by `encodeURI` or by a similar routine.\n\nMDN documentation\n\nFunction js_sys::decode_uri_component\n===\n\n\n```\npub fn decode_uri_component(encoded: &str) -> Result<JsString, JsValue>\n```\nThe `decodeURIComponent()` function decodes a Uniform Resource Identifier (URI) component previously created by `encodeURIComponent` or by a similar routine.\n\nMDN documentation\n\nFunction js_sys::encode_uri\n===\n\n\n```\npub fn encode_uri(decoded: &str) -> JsString\n```\nThe `encodeURI()` function encodes a Uniform Resource Identifier (URI)\nby replacing each instance of certain characters by one, two, three, or four escape sequences representing the UTF-8 encoding of the character\n(will only be four escape sequences for characters composed of two\n“surrogate” characters).\n\nMDN documentation\n\nFunction js_sys::encode_uri_component\n===\n\n\n```\npub fn encode_uri_component(decoded: &str) -> JsString\n```\nThe `encodeURIComponent()` function encodes a Uniform Resource Identifier (URI) component by replacing each instance of certain characters by one, two, three, or four escape sequences representing the UTF-8 encoding of the character\n(will only be four escape sequences for characters composed of two “surrogate” characters).\n\nMDN documentation\n\nFunction js_sys::escape\n===\n\n\n```\npub fn escape(string: &str) -> JsString\n```\nThe `escape()` function computes a new string in which certain characters have been replaced by a hexadecimal escape sequence.\n\nMDN documentation\n\nFunction js_sys::eval\n===\n\n\n```\npub fn eval(js_source_text: &str) -> Result<JsValue, JsValue>\n```\nThe `eval()` function evaluates JavaScript code represented as a string.\n\nMDN documentation\n\nFunction js_sys::global\n===\n\n\n```\npub fn global() -> Object\n```\nReturns a handle to the global scope object.\n\nThis allows access to the global properties and global names by accessing the `Object` returned.\n\nFunction js_sys::is_finite\n===\n\n\n```\npub fn is_finite(value: &JsValue) -> bool\n```\nThe global `isFinite()` function determines whether the passed value is a finite number.\nIf needed, the parameter is first converted to a number.\n\nMDN documentation\n\nFunction js_sys::parse_float\n===\n\n\n```\npub fn parse_float(text: &str) -> f64\n```\nThe `parseFloat()` function parses an argument and returns a floating point number,\nor NaN on error.\n\nMDN documentation\n\nFunction js_sys::parse_int\n===\n\n\n```\npub fn parse_int(text: &str, radix: u8) -> f64\n```\nThe `parseInt()` function parses a string argument and returns an integer of the specified radix (the base in mathematical numeral systems), or NaN on error.\n\nMDN documentation\n\nFunction js_sys::try_iter\n===\n\n\n```\npub fn try_iter(val: &JsValue) -> Result<Option<IntoIter>, JsValue>\n```\nCreate an iterator over `val` using the JS iteration protocol and\n`Symbol.iterator`.\n\nFunction js_sys::unescape\n===\n\n\n```\npub fn unescape(string: &str) -> JsString\n```\nThe `unescape()` function computes a new string in which hexadecimal escape sequences are replaced with the character that it represents. The escape sequences might be introduced by a function like `escape`. Usually, `decodeURI` or `decodeURIComponent`\nare preferred over `unescape`.\n\nMDN documentation"}}},{"rowIdx":59167,"cells":{"project":{"kind":"string","value":"topepo_github_io_caret_index_html"},"source":{"kind":"string","value":"free_programming_book"},"language":{"kind":"string","value":"Unknown"},"content":{"kind":"string","value":"# The caret Package\n\nDate: 2019-03-27\nCategories: \nTags: \n\n# The\n\n `caret` Package \n2019-03-27\n\n# 1 Introduction\n\n The `caret` package (short for Classification And REgression Training) is a set of functions that attempt to streamline the process for creating predictive models. The package contains tools for: \n\n* data splitting\n * pre-processing\n * feature selection\n * model tuning using resampling\n * variable importance estimation\n\nas well as other functionality.\n\nThere are many different modeling functions in R. Some have different syntax for model training and/or prediction. The package started off as a way to provide a uniform interface the functions themselves, as well as a way to standardize common tasks (such parameter tuning and variable importance).\n\nThe current release version can be found on CRAN and the project is hosted on github.\n\nSome resources:\n\n* The book Applied Predictive Modeling features caret and over 40 other R packages. It is on sale at Amazon or the the publisher’s website. There is a companion website too.\n * There is also a paper on caret in the Journal of Statistical Software. The example data can be obtained here(the predictors) and here (the outcomes).\n * There is a webinar for the package on Youtube that was organized and recorded by <NAME> Jr for the Orange County R User Group.\n * At useR! 2014, I was interviewed and discussed the package and the book.\n * DataCamp has a beginner’s tutorial on machine learning in R using\n `caret` . \nYou can always email me with questions,comments or suggestions.\n\nThese HTML pages were created using bookdown.\n"}}},{"rowIdx":59168,"cells":{"project":{"kind":"string","value":"cPCG"},"source":{"kind":"string","value":"cran"},"language":{"kind":"string","value":"R"},"content":{"kind":"string","value":"Package ‘cPCG’\n October 12, 2022\nType Package\nTitle Efficient and Customized Preconditioned Conjugate Gradient\n Method for Solving System of Linear Equations\nVersion 1.0\nDate 2018-12-30\nAuthor <NAME>\nMaintainer <NAME> <<EMAIL>>\nDescription Solves system of linear equations using (preconditioned) conjugate gradient algo-\n rithm, with improved efficiency using Armadillo templated 'C++' linear algebra library, and flex-\n ibility for user-specified precondition-\n ing method. Please check <https://github.com/styvon/cPCG> for latest updates.\nDepends R (>= 3.0.0)\nLicense GPL (>= 2)\nImports Rcpp (>= 0.12.19)\nLinkingTo Rcpp, RcppArmadillo\nRoxygenNote 6.1.1\nEncoding UTF-8\nSuggests knitr, rmarkdown\nVignetteBuilder knitr\nNeedsCompilation yes\nRepository CRAN\nDate/Publication 2019-01-11 17:00:10 UTC\nR topics documented:\ncPCG-packag... 2\ncgsolv... 3\nic... 4\npcgsolv... 5\n cPCG-package Efficient and Customized Preconditioned Conjugate Gradient Method\n for Solving System of Linear Equations\nDescription\n Solves system of linear equations using (preconditioned) conjugate gradient algorithm, with im-\n proved efficiency using Armadillo templated ’C++’ linear algebra library, and flexibility for user-\n specified preconditioning method. Please check <https://github.com/styvon/cPCG> for latest up-\n dates.\nDetails\n Functions in this package serve the purpose of solving for x in Ax = b, where A is a symmetric and\n positive definite matrix, b is a column vector.\n To improve scalability of conjugate gradient methods for larger matrices, the Armadillo templated\n C++ linear algebra library is used for the implementation. The package also provides flexibility to\n have user-specified preconditioner options to cater for different optimization needs.\n The DESCRIPTION file:\n Package: cPCG\n Type: Package\n Title: Efficient and Customized Preconditioned Conjugate Gradient Method for Solving System of Linear Equati\n Version: 1.0\n Date: 2018-12-30\n Author: <NAME>\n Maintainer: <NAME> <<EMAIL>>\n Description: Solves system of linear equations using (preconditioned) conjugate gradient algorithm, with improved effic\n Depends: R (>= 3.0.0)\n License: GPL (>= 2)\n Imports: Rcpp (>= 0.12.19)\n LinkingTo: Rcpp, RcppArmadillo\n RoxygenNote: 6.1.1\n Encoding: UTF-8\n Suggests: knitr, rmarkdown\n VignetteBuilder: knitr\n Index of help topics:\n cPCG-package Efficient and Customized Preconditioned\n Conjugate Gradient Method for Solving System of\n Linear Equations\n cgsolve Conjugate gradient method\n icc Incomplete Cholesky Factorization\n pcgsolve Preconditioned conjugate gradient method\nAuthor(s)\n <NAME>\nReferences\n [1] <NAME> and <NAME>. “Function minimization by conjugate gradients”. In: The\n computer journal 7.2 (1964), pp. 149–154.\n [2] <NAME>. “The incomplete Cholesky—conjugate gradient method for the iter- ative\n solution of systems of linear equations”. In: Journal of computational physics 26.1 (1978), pp.\n 43–65.\n [3] <NAME>. Iterative methods for sparse linear systems. Vol. 82. siam, 2003.\n [4] <NAME>. “Iterative methods for solving partial difference equations of elliptic type”. In:\n Transactions of the American Mathematical Society 76.1 (1954), pp. 92–111.\nExamples\n # generate test data\n test_A <- matrix(c(4,1,1,3), ncol = 2)\n test_b <- matrix(1:2, ncol = 1)\n # conjugate gradient method solver\n cgsolve(test_A, test_b, 1e-6, 1000)\n # preconditioned conjugate gradient method solver,\n # with incomplete Cholesky factorization as preconditioner\n pcgsolve(test_A, test_b, \"ICC\")\n cgsolve Conjugate gradient method\nDescription\n Conjugate gradient method for solving system of linear equations Ax = b, where A is symmetric\n and positive definite, b is a column vector.\nUsage\n cgsolve(A, b, tol = 1e-6, maxIter = 1000)\nArguments\n A matrix, symmetric and positive definite.\n b vector, with same dimension as number of rows of A.\n tol numeric, threshold for convergence, default is 1e-6.\n maxIter numeric, maximum iteration, default is 1000.\nDetails\n The idea of conjugate gradient method is to find a set of mutually conjugate directions for the\n unconstrained problem\n argminx f (x)\n where f (x) = 0.5bT Ab − bx + z and z is a constant. The problem is equivalent to solving Ax = b.\n This function implements an iterative procedure to reduce the number of matrix-vector multiplica-\n tions [1]. The conjugate gradient method improves memory efficiency and computational complex-\n ity, especially when A is relatively sparse.\nValue\n Returns a vector representing solution x.\nWarning\n Users need to check that input matrix A is symmetric and positive definite before applying the\n function.\nReferences\n [1] <NAME>. Iterative methods for sparse linear systems. Vol. 82. siam, 2003.\nSee Also\n pcgsolve\nExamples\n ## Not run:\n test_A <- matrix(c(4,1,1,3), ncol = 2)\n test_b <- matrix(1:2, ncol = 1)\n cgsolve(test_A, test_b, 1e-6, 1000)\n ## End(Not run)\n icc Incomplete Cholesky Factorization\nDescription\n Incomplete Cholesky factorization method to generate preconditioning matrix for conjugate gradi-\n ent method.\nUsage\n icc(A)\nArguments\n A matrix, symmetric and positive definite.\nDetails\n Performs incomplete Cholesky factorization on the input matrix A, the output matrix is used for\n preconditioning in pcgsolve() if \"ICC\" is specified as the preconditioner.\nValue\n Returns a matrix after incomplete Cholesky factorization.\nWarning\n Users need to check that input matrix A is symmetric and positive definite before applying the\n function.\nSee Also\n pcgsolve\nExamples\n ## Not run:\n test_A <- matrix(c(4,1,1,3), ncol = 2)\n out <- icc(test_A)\n ## End(Not run)\n pcgsolve Preconditioned conjugate gradient method\nDescription\n Preconditioned conjugate gradient method for solving system of linear equations Ax = b, where A\n is symmetric and positive definite, b is a column vector.\nUsage\n pcgsolve(A, b, preconditioner = \"Jacobi\", tol = 1e-6, maxIter = 1000)\nArguments\n A matrix, symmetric and positive definite.\n b vector, with same dimension as number of rows of A.\n preconditioner string, method for preconditioning: \"Jacobi\" (default), \"SSOR\", or \"ICC\".\n tol numeric, threshold for convergence, default is 1e-6.\n maxIter numeric, maximum iteration, default is 1000.\nDetails\n When the condition number for A is large, the conjugate gradient (CG) method may fail to converge\n in a reasonable number of iterations. The Preconditioned Conjugate Gradient (PCG) Method applies\n a precondition matrix C and approaches the problem by solving:\n C −1 Ax = C −1 b\n where the symmetric and positive-definite matrix C approximates A and C −1 A improves the con-\n dition number of A.\n Common choices for the preconditioner include: Jacobi preconditioning, symmetric successive\n over-relaxation (SSOR), and incomplete Cholesky factorization [2].\nValue\n Returns a vector representing solution x.\nPreconditioners\n Jacobi: The Jacobi preconditioner is the diagonal of the matrix A, with an assumption that all\n diagonal elements are non-zero.\n SSOR: The symmetric successive over-relaxation preconditioner, implemented as M = (D+L)D−1 (D+\n L)T . [1]\n ICC: The incomplete Cholesky factorization preconditioner. [2]\nWarning\n Users need to check that input matrix A is symmetric and positive definite before applying the\n function.\nReferences\n [1] <NAME>. “Iterative methods for solving partial difference equations of elliptic type”. In:\n Transactions of the American Mathematical Society 76.1 (1954), pp. 92–111.\n [2] <NAME>. “The incomplete Cholesky—conjugate gradient method for the iter- ative\n solution of systems of linear equations”. In: Journal of computational physics 26.1 (1978), pp.\n 43–65.\nSee Also\n cgsolve\nExamples\n ## Not run:\n test_A <- matrix(c(4,1,1,3), ncol = 2)\n test_b <- matrix(1:2, ncol = 1)\n pcgsolve(test_A, test_b, \"ICC\")\n ## End(Not run)"}}},{"rowIdx":59169,"cells":{"project":{"kind":"string","value":"passerine"},"source":{"kind":"string","value":"readthedoc"},"language":{"kind":"string","value":"Python"},"content":{"kind":"string","value":"Passerine ORM 1.4.0 documentation\n\n[Passerine ORM](index.html#document-index)\n\n---\n\nPasserine[¶](#passerine)\n===\n\n| Author: | <NAME> <[<EMAIL>](mailto:j<EMAIL>)> |\n\nA generic object relational mapper (ORM) and data abstraction layer (DAL) primarily designed for NoSQL databases.\n\nYou can **install Passerine** by running `pip install passerine`.\n\nArchitecture[¶](#architecture)\n---\n\nPasserine is primarily designed for non-relational databases. Currently, the only driver shipped with the library is MongoDB. The next one will be\n**Riak 2.0** (a distributed database) and **MySQL** (a relational databases).\n\nThere are a few points to highlight.\n\n* The lazy-loading strategy and proxy objects are used to load data wherever applicable.\n* The ORM uses **the Unit Of Work pattern** as used by:\n\t+ [Hibernate](http://www.hibernate.org/) (Java)\n\t+ [Doctrine](http://www.doctrine-project.org/) (PHP)\n\t+ [SQLAlchemy](http://www.sqlalchemy.org/) (Python)\n* By containing a similar logic to determine whether a given entity is new or old, the following condition are used:\n\t+ If a given entity is identified with an **object ID**, the given entity will\n\tbe considered as an existing entity.\n\t+ Otherwise, it will be a new entity.\n* The object ID cannot be changed via the ORM interfaces.\n* The ORM supports cascading operations on deleting, persisting, and refreshing.\n* To ensure the performance, heavily rely on **public properties**, which does not have leading underscores (`_`) to map between class properties and document keys, except the property **id** will be converted to the key **_id**.\n* The class design is heavily influenced by dependency injection.\n\n### Limitations[¶](#limitations)\n\n* **Cascading operations on persisting** force the ORM to load the data of all proxy objects but commiting changes will still be made only if there are changes.\n* **Cascading operations on refreshing** force the ORM to reset the data and status of all entities, including proxy objects. However, the status of any entities marked for deletion will not be reset.\n\n#### Common for Non-relational or Distributed Databases[¶](#common-for-non-relational-or-distributed-databases)\n\n* Sessions cannot merge together.\n* **Cascading operations on deleting** forces the ORM to load the whole data graph which degrades the performance.\n\n#### MongoDB Only[¶](#mongodb-only)\n\n* For the MongoDB driver, as MongoDB does not has transaction support like MySQL,\nthe ORM has sessions to manage the object graph within the same memory space.\n* As **sessions** are not supported by MongoDB, the ORM cannot roll back in case that an exception are raisen or a writing operation is interrupted.\n\nAPI Reference[¶](#api-reference)\n---\n\n### passerine.common[¶](#passerine-common)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/common.py`.\n\n| Author: | <NAME> |\n\nThis package contains classes and functions for common use.\n\n*class* `passerine.common.``Enigma`[¶](#passerine.common.Enigma)\nHashlib wrapper\n\n`hash`(**data_list*)[¶](#passerine.common.Enigma.hash)\nMake a hash out of the given `value`.\n\n| Parameters: | **data_list** (*list of string*) – the list of the data being hashed. |\n| Returns: | the hashed data string |\n\n*static* `instance`()[¶](#passerine.common.Enigma.instance)\nGet a singleton instance.\n\nNote\n\nThis class is capable to act as a singleton class by invoking this method.\n\n*class* `passerine.common.``Finder`[¶](#passerine.common.Finder)\nFile System API Wrapper\n\n`read`(*file_path*, *is_binary=False*)[¶](#passerine.common.Finder.read)\nRead a file from *file_path*.\n\nBy default, read a file normally. If *is_binary* is `True`, the method will read in binary mode.\n\n### passerine.data.base[¶](#passerine-data-base)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/data/base.py`.\n\n*class* `passerine.data.base.``ResourceEntity`(*path*, *cacheable=False*)[¶](#passerine.data.base.ResourceEntity)\nStatic resource entity representing the real static resource which is already loaded to the memory.\n\n| Parameters: | **path** – the path to the static resource. |\n\nNote\n\nThis is for internal use only.\n\n`content`[¶](#passerine.data.base.ResourceEntity.content)\nGet the content of the entity.\n\n### passerine.data.compressor[¶](#passerine-data-compressor)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/data/compressor.py`.\n\n### passerine.data.exception[¶](#passerine-data-exception)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/data/exception.py`.\n\n### passerine.data.serializer[¶](#passerine-data-serializer)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/data/serializer.py`.\n\n### passerine.db.common[¶](#passerine-db-common)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/common.py`.\n\n| Author: | <NAME> <[<EMAIL>](mailto:jnopporn%40shiroyuki.com)> |\n| Stability: | Stable |\n\n*class* `passerine.db.common.``ProxyCollection`(*session*, *origin*, *guide*)[¶](#passerine.db.common.ProxyCollection)\nProxy Collection\n\nThis collection is extended from the built-in class `list`,\ndesigned to only load the associated data whenever is required.\n\n| Parameters: | * **session** ([*passerine.db.session.Session*](index.html#passerine.db.session.Session)) – the managed session\n* **origin** (*object*) – the origin of the association\n* **guide** ([*passerine.db.mapper.RelatingGuide*](index.html#passerine.db.mapper.RelatingGuide)) – the relational guide\n |\n\nNote\n\nTo replace with criteria and driver\n\n`reload`()[¶](#passerine.db.common.ProxyCollection.reload)\nReload the data list\n\nWarning\n\nThis method is **not recommended** to be called directly. Use\n[`passerine.db.session.Session.refresh()`](index.html#passerine.db.session.Session.refresh) on the owned object instead.\n\n*class* `passerine.db.common.``ProxyFactory`[¶](#passerine.db.common.ProxyFactory)\nProxy Factory\n\nThis factory is to create a proxy object.\n\n| Parameters: | * **session** ([*passerine.db.session.Session*](index.html#passerine.db.session.Session)) – the managed session\n* **id** – the object ID\n* **mapping_guide** ([*passerine.db.mapper.RelatingGuide*](index.html#passerine.db.mapper.RelatingGuide)) – the relational guide\n |\n\n*class* `passerine.db.common.``ProxyObject`(*session*, *cls*, *object_id*, *read_only*, *cascading_options*, *is_reverse_proxy*)[¶](#passerine.db.common.ProxyObject)\nProxy Collection\n\nThis class is designed to only load the entity whenever the data access is required.\n\n| Parameters: | * **session** ([*passerine.db.session.Session*](index.html#passerine.db.session.Session)) – the managed session\n* **cls** (*type*) – the class to map the data\n* **object_id** – the object ID\n* **read_only** (*bool*) – the read-only flag\n* **cascading_options** (*list or tuple*) – the cascading options\n* **is_reverse_proxy** (*bool*) – the reverse proxy flag\n |\n\n*class* `passerine.db.common.``PseudoObjectId`(*oid=None*)[¶](#passerine.db.common.PseudoObjectId)\nPseudo Object ID\n\nThis class extends from `bson.objectid.ObjectId`.\n\nThis is used to differentiate stored entities and new entities.\n\n*class* `passerine.db.common.``Serializer`(*max_depth=2*)[¶](#passerine.db.common.Serializer)\nObject Serializer for Entity\n\n`encode`(*data*, *stack_depth=0*, *convert_object_id_to_str=False*)[¶](#passerine.db.common.Serializer.encode)\nEncode data into dictionary and list.\n\n| Parameters: | * **data** – the data to encode\n* **stack_depth** – traversal depth limit\n* **convert_object_id_to_str** – flag to convert object ID into string\n |\n\n### passerine.db.driver.interface[¶](#passerine-db-driver-interface)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/driver/interface.py`.\n\n*class* `passerine.db.driver.interface.``DialectInterface`[¶](#passerine.db.driver.interface.DialectInterface)\nDialect interface\n\nIt is used to translate a generic query into a native query.\n\n`get_alias_to_native_query_map`(*query*)[¶](#passerine.db.driver.interface.DialectInterface.get_alias_to_native_query_map)\nRetrieve a map from alias to native query.\n\n| Parameters: | **passerine.db.query.Query** – the query object |\n| Return type: | dict |\n\n`get_iterating_constrains`(*query*)[¶](#passerine.db.driver.interface.DialectInterface.get_iterating_constrains)\nRetrieve the query constrains.\n\n| Raises: | **NotImplemented** – only if the interface is not overridden. |\n\n`get_native_operand`(*generic_operand*)[¶](#passerine.db.driver.interface.DialectInterface.get_native_operand)\nTranslate a generic operand into a corresponding native operand.\n\n| Parameters: | **generic_operand** – a generic operand |\n| Returns: | a native operand |\n| Return type: | str |\n\n`process_join_conditions`(*alias_to_conditions_map*, *alias*, *join_config*, *parent_alias*)[¶](#passerine.db.driver.interface.DialectInterface.process_join_conditions)\nProcess the join conditions.\n\n| Parameters: | * **alias_to_conditions_map** (*dict*) – a alias-to-conditions map\n* **join_config** (*dict*) – a join config map\n* **alias** (*str*) – an alias of the given join map\n* **parent_alias** (*str*) – the parent alias of the given join map\n |\n| Raises: | **NotImplemented** – only if the interface is not overridden. |\n\n`process_non_join_conditions`(*alias_to_conditions_map*, *definition_map*, *left*, *right*, *operand*)[¶](#passerine.db.driver.interface.DialectInterface.process_non_join_conditions)\nProcess the non-join conditions.\n\n| Parameters: | * **alias_to_conditions_map** (*dict*) – a alias-to-conditions map\n* **definition_map** (*dict*) – a parameter-to-value map\n* **left** ([*passerine.db.expression.ExpressionPart*](index.html#passerine.db.expression.ExpressionPart)) – the left expression\n* **right** ([*passerine.db.expression.ExpressionPart*](index.html#passerine.db.expression.ExpressionPart)) – the right expression\n* **operand** – the native operand\n |\n| Raises: | **NotImplemented** – only if the interface is not overridden. |\n\n*class* `passerine.db.driver.interface.``DriverInterface`(*config*, *dialect*)[¶](#passerine.db.driver.interface.DriverInterface)\nThe abstract driver interface\n\n| Parameters: | * **config** (*dict*) – the configuration used to initialize the database connection / client\n* **dialect** ([*passerine.db.driver.interface.DialectInterface*](index.html#passerine.db.driver.interface.DialectInterface)) – the corresponding dialect\n |\n\n`client`[¶](#passerine.db.driver.interface.DriverInterface.client)\nDriver Connection / Client\n\n`collection`(*name*)[¶](#passerine.db.driver.interface.DriverInterface.collection)\nLow-level Collection-class API\n\n| Returns: | the low-level collection-class API |\n| Raises: | **NotImplemented** – only if the interface is not overridden. |\n\n`config`[¶](#passerine.db.driver.interface.DriverInterface.config)\nDriver configuration\n\n`connect`(*config*)[¶](#passerine.db.driver.interface.DriverInterface.connect)\nConnect the client to the server.\n\n| Raises: | **NotImplemented** – only if the interface is not overridden. |\n\n`database_name`[¶](#passerine.db.driver.interface.DriverInterface.database_name)\nThe name of provisioned database\n\n`db`(*name*)[¶](#passerine.db.driver.interface.DriverInterface.db)\nLow-level Database-class API\n\n| Returns: | the low-level database-class API |\n| Raises: | **NotImplemented** – only if the interface is not overridden. |\n\n`dialect`[¶](#passerine.db.driver.interface.DriverInterface.dialect)\nDriver dialect\n\n`disconnect`()[¶](#passerine.db.driver.interface.DriverInterface.disconnect)\nDisconnect the client.\n\n| Raises: | **NotImplemented** – only if the interface is not overridden. |\n\n`index_count`()[¶](#passerine.db.driver.interface.DriverInterface.index_count)\nRetrieve the number of indexes.\n\n| Raises: | **NotImplemented** – only if the interface is not overridden. |\n\n`indice`()[¶](#passerine.db.driver.interface.DriverInterface.indice)\nRetrieve the indice.\n\n| Raises: | **NotImplemented** – only if the interface is not overridden. |\n\n`insert`(*collection_name*, *data*)[¶](#passerine.db.driver.interface.DriverInterface.insert)\nLow-level insert function\n\n| Raises: | **NotImplemented** – only if the interface is not overridden. |\n\n*class* `passerine.db.driver.interface.``QueryIteration`(*alias*, *native_query*)[¶](#passerine.db.driver.interface.QueryIteration)\nDriver Query Iteration\n\nThis is a metadata class representing an iteration in complex queries.\n\n| Parameters: | * **alias** (*str*) – the alias of the rewritten target\n* **native_query** (*dict*) – the native query for a specific engine\n |\n\nNote\n\nInternal use only\n\n*class* `passerine.db.driver.interface.``QuerySequence`[¶](#passerine.db.driver.interface.QuerySequence)\nDriver Query Sequence\n\nThe collection represents the sequence of sub queries.\n\n`add`(*iteration*)[¶](#passerine.db.driver.interface.QuerySequence.add)\nAppend the the iteration\n\n| Parameters: | **iteration** ([*passerine.db.driver.interface.QueryIteration*](index.html#passerine.db.driver.interface.QueryIteration)) – the query iteration |\n\n`each`()[¶](#passerine.db.driver.interface.QuerySequence.each)\nGet the sequence iterator.\n\n### passerine.db.driver.mongodriver[¶](#passerine-db-driver-mongodriver)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/driver/mongodriver.py`.\n\n*exception* `passerine.db.driver.mongodriver.``UnsupportedExpressionError`[¶](#passerine.db.driver.mongodriver.UnsupportedExpressionError)\nMongoDB-specific Unsupported Expression Error\n\nThis is due to that the expression may be unsafe (e.g., 1 = 2) or result in unnecessary complex computation (e.g., e.mobile_phone = e.home_phone).\n\n### passerine.db.driver.registrar[¶](#passerine-db-driver-registrar)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/driver/registrar.py`.\n\n### passerine.db.driver.riakdriver[¶](#passerine-db-driver-riakdriver)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/driver/riakdriver.py`.\n\n### passerine.db.entity[¶](#passerine-db-entity)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/entity.py`.\n\n| Author: | <NAME> <[<EMAIL>](mailto:<EMAIL>)> |\n\n*class* `passerine.db.entity.``BasicAssociation`(*origin*, *destination*)[¶](#passerine.db.entity.BasicAssociation)\nBasic Association\n\n| Parameters: | * **origin** (*object*) – The origin of the association\n* **destination** (*object*) – The destination (endpoint) of the association\n |\n\nNote\n\nThis class is used automatically by the association mapper.\n\n*class* `passerine.db.entity.``Entity`(***attributes*)[¶](#passerine.db.entity.Entity)\nDynamic-attribute Basic Entity\n\n| Parameters: | **attributes** (*dict*) – key-value dictionary |\n\nHere is an example on how to use this class.\n```\n@entity class Note(Entity): pass\n```\n*class* `passerine.db.entity.``Index`(*field_map*, *unique=False*)[¶](#passerine.db.entity.Index)\n\n| Parameters: | * **field_map** (*dict*) – the map of field to index type\n* **unique** (*bool*) – the unique flag\n |\n\nUnless a field is not in the map of fixed orders, the index will instruct the repository to ensure all combinations of indexes are defined whenever is necessary.\n\n`passerine.db.entity.``entity`(**args*, ***kwargs*)[¶](#passerine.db.entity.entity)\nEntity decorator\n\n| Parameters: | **collection_name** (*str*) – the name of the collection |\n| Returns: | the decorated object |\n| Return type: | object |\n\n`passerine.db.entity.``prepare_entity_class`(*cls*, *collection_name=None*, *indexes=[]*)[¶](#passerine.db.entity.prepare_entity_class)\nCreate a entity class\n\n| Parameters: | * **cls** (*object*) – the document class\n* **collection_name** (*str*) – the name of the corresponding collection where the default is the lowercase version of the name of the given class (cls)\n |\n\nThe object decorated with this decorator will be automatically provided with a few additional attributes.\n\n| Attribute | Access | Description | Read | Write |\n| --- | --- | --- | --- | --- |\n| id | Instance | Document Identifier | Yes | Yes, ONLY `id` is undefined. |\n| __t3_orm_meta__ | Static | Tori 3’s Metadata | Yes | ONLY the property of the metadata |\n| __session__ | Instance | DB Session | Yes | Yes, but NOT recommended. |\n\nThe following attributes might stay around but are deprecated as soon as the stable Tori 3.0 is released.\n\n| Attribute | Access | Description | Read | Write |\n| --- | --- | --- | --- | --- |\n| __collection_name__ | Static | Collection Name | Yes | Yes, but NOT recommended. |\n| __relational_map__ | Static | Relational Map | Yes | Yes, but NOT recommended. |\n| __indexes__ | Static | Indexing List | Yes | Yes, but NOT recommended. |\n\n`__session__` is used to resolve the managing rights in case of using multiple sessions simutaneously.\n\nFor example,\n```\n@entity class Note(object):\n def __init__(self, content, title=''):\n self.content = content\n self.title = title\n```\nwhere the collection name is automatically defined as “note”.\n\nChanged in version 3.0: The way Tori stores metadata objects in `__collection_name__`,\n`__relational_map__` and `__indexes__` are now ignored by the ORM in favour of `__t3_orm_meta__` which is an entity metadata object.\n\nThis change is made to allow easier future development.\n\nTip\n\nYou can define it as “notes” by replacing `@entity` with `@entity('notes')`.\n\n### passerine.db.exception[¶](#passerine-db-exception)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/exception.py`.\n\n*exception* `passerine.db.exception.``DuplicatedRelationalMapping`[¶](#passerine.db.exception.DuplicatedRelationalMapping)\nException thrown when the property is already mapped.\n\n*exception* `passerine.db.exception.``EntityAlreadyRecognized`[¶](#passerine.db.exception.EntityAlreadyRecognized)\nWarning raised when the entity with either a designated ID or a designated session is provided to Repository.post\n\n*exception* `passerine.db.exception.``EntityNotRecognized`[¶](#passerine.db.exception.EntityNotRecognized)\nWarning raised when the entity without either a designated ID or a designated session is provided to Repository.put or Repository.delete\n\n*exception* `passerine.db.exception.``IntegrityConstraintError`[¶](#passerine.db.exception.IntegrityConstraintError)\nRuntime Error raised when the given value violates a integrity constraint.\n\n*exception* `passerine.db.exception.``InvalidUrlError`[¶](#passerine.db.exception.InvalidUrlError)\nInvalid DB URL Error\n\n*exception* `passerine.db.exception.``LockedIdException`[¶](#passerine.db.exception.LockedIdException)\nException thrown when the ID is tempted to change.\n\n*exception* `passerine.db.exception.``MissingObjectIdException`[¶](#passerine.db.exception.MissingObjectIdException)\nException raised when the object Id is not specified during data retrieval.\n\n*exception* `passerine.db.exception.``NonRefreshableEntity`[¶](#passerine.db.exception.NonRefreshableEntity)\nException thrown when the UOW attempts to refresh a non-refreshable entity\n\n*exception* `passerine.db.exception.``ReadOnlyProxyException`[¶](#passerine.db.exception.ReadOnlyProxyException)\nException raised when the proxy is for read only.\n\n*exception* `passerine.db.exception.``UOWRepeatedRegistrationError`[¶](#passerine.db.exception.UOWRepeatedRegistrationError)\nError thrown when the given reference is already registered as a new reference or already existed.\n\n*exception* `passerine.db.exception.``UOWUnknownRecordError`[¶](#passerine.db.exception.UOWUnknownRecordError)\nError thrown when the given reference is already registered as a new reference or already existed.\n\n*exception* `passerine.db.exception.``UOWUpdateError`[¶](#passerine.db.exception.UOWUpdateError)\nError thrown when the given reference is already registered as a new reference or already existed.\n\n*exception* `passerine.db.exception.``UnavailableCollectionException`[¶](#passerine.db.exception.UnavailableCollectionException)\nException thrown when the collection is not available.\n\n*exception* `passerine.db.exception.``UnknownDriverError`[¶](#passerine.db.exception.UnknownDriverError)\nUnknown Driver Error\n\n*exception* `passerine.db.exception.``UnsupportedRepositoryReferenceError`[¶](#passerine.db.exception.UnsupportedRepositoryReferenceError)\nUnsupported Repository Reference Error\n\n### passerine.db.expression[¶](#passerine-db-expression)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/expression.py`.\n\n*class* `passerine.db.expression.``Criteria`[¶](#passerine.db.expression.Criteria)\nExpression Criteria\n\nSupport operands: =, <=, <, >, >=, in, like (SQL-like string pattern), rlike (Regular-expression pattern), indexed with (only for Riak)\n\n*class* `passerine.db.expression.``Expression`(*left*, *operand*, *right*)[¶](#passerine.db.expression.Expression)\nQuery Expression\n\n| Parameters: | * **left** ([*passerine.db.expression.ExpressionPart*](index.html#passerine.db.expression.ExpressionPart)) – the left part\n* **right** ([*passerine.db.expression.ExpressionPart*](index.html#passerine.db.expression.ExpressionPart)) – the right part\n* **operand** (*str*) – the generic operand\n |\n\n*class* `passerine.db.expression.``ExpressionPart`(*original*, *kind*, *value*, *alias*)[¶](#passerine.db.expression.ExpressionPart)\nQuery Expression\n\n| Parameters: | * **original** (*str*) – the original query\n* **kind** (*str*) – the type of the part\n* **value** – the parameter value only for a data part\n* **alias** (*str*) – the entity alias for a property part or the name of the parameter of a parameter part\n |\n\n*class* `passerine.db.expression.``ExpressionSet`(*expressions*)[¶](#passerine.db.expression.ExpressionSet)\nRepresentation of Analyzed Expression\n\n*exception* `passerine.db.expression.``InvalidExpressionError`[¶](#passerine.db.expression.InvalidExpressionError)\nGeneric Invalid Expression Error\n\n### passerine.db.fixture[¶](#passerine-db-fixture)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/fixture.py`.\n\n| Author: | <NAME> |\n\nWarning\n\nNot in used.\n\n*class* `passerine.db.fixture.``Fixture`(*repository*)[¶](#passerine.db.fixture.Fixture)\nFoundation of the council\n\nNote\n\nthis must be used at most once.\n\nWarning\n\nthis class is not tested.\n\n`set`(*kind*, *fixtures*)[¶](#passerine.db.fixture.Fixture.set)\nDefine the fixtures.\n\n| Parameters: | * **kind** (*unicode|str*) – a string represent the kind\n* **fixtures** (*dict*) – the data dictionary keyed by the alias\n |\n```\nfixture = Fixture()\n\nfixture.set(\n 'council.security.model.Provider',\n {\n 'ldap': { 'name': 'ldap' }\n }\n)\nfixture.set(\n 'council.user.model.User', {\n 'admin': { 'name': '<NAME>' }\n }\n)\nfixture.set(\n 'council.security.model.Credential',\n {\n 'shiroyuki': {\n 'login': 'admin',\n 'user': 'proxy/council.user.model.User/admin',\n 'provider': 'proxy/council.security.model.Provider/ldap'\n }\n }\n)\n```\n### passerine.db.manager[¶](#passerine-db-manager)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/manager.py`.\n\n*class* `passerine.db.manager.``Manager`(*driver*)[¶](#passerine.db.manager.Manager)\nEntity Manager\n\nThis is to manage the unit-of-work session.\n\n| Parameters: | **driver** ([*passerine.db.driver.interface.DriverInterface*](index.html#passerine.db.driver.interface.DriverInterface)) – the driver interface |\n\n`close_session`(*id*)[¶](#passerine.db.manager.Manager.close_session)\nClose the managed session\n\nWarning\n\nThis method is designed to bypass errors when the given ID is unavailable or already closed.\n\n`driver`[¶](#passerine.db.manager.Manager.driver)\nDriver API\n\n| Return type: | [passerine.db.driver.interface.DriverInterface](index.html#passerine.db.driver.interface.DriverInterface) |\n\n`get_repository`(*ref_class*, *session_id=None*)[¶](#passerine.db.manager.Manager.get_repository)\nRetrieve the repository.\n\n| Parameters: | * **ref_class** (*type*) – The reference class\n* **session_id** – The supervised session ID\n |\n| Return type: | [passerine.db.repository.Repository](index.html#passerine.db.repository.Repository) |\n\nNote\n\nThis is designed for Imagination’s factorization.\n\nNote\n\nWith an unsupervised session, the memory usage may be higher than usual as the memory reference may not be freed as long as the reference to the returned repository continues to exist in active threads.\n\n`open_session`(*id=None*, *supervised=False*)[¶](#passerine.db.manager.Manager.open_session)\nOpen a session\n\n| Parameters: | * **id** – the session ID\n* **supervised** (*bool*) – the flag to indicate that the opening session will be observed and supervised by the manager.\nThis allows the session to be reused by multiple components. However, it is not **thread-safe**.\nIt is disabled by default.\n |\n\n`session`(**args*, ***kwds*)[¶](#passerine.db.manager.Manager.session)\nOpen a session in the context manager.\n\n| Parameters: | * **id** – the session ID\n* **supervised** (*bool*) – the flag to indicate that the opening session will be observed and supervised by the manager.\nThis allows the session to be reused by multiple components. However, it is not **thread-safe**.\nIt is disabled by default.\n |\n\nNote\n\nThe end of the context will close the open session.\n\n*class* `passerine.db.manager.``ManagerFactory`(*urls=None*, *protocols=None*)[¶](#passerine.db.manager.ManagerFactory)\nManager Factory\n\n| Parameters: | * **urls** (*dict*) – the alias-to-endpoint-URL map\n* **protocols** (*dict*) – the protocol-to-fully-qualified-module-path map\n |\n\n`get`(*alias*)[¶](#passerine.db.manager.ManagerFactory.get)\nRetrieve the database by the manager alias.\n\n| Parameters: | **alias** (*str*) – the alias of the manager. |\n| Return type: | [passerine.db.manager.Manager](index.html#passerine.db.manager.Manager) |\n\n`register`(*protocol*, *driver_class*)[¶](#passerine.db.manager.ManagerFactory.register)\nRegister the protocol to the driver class.\n\n| Parameters: | * **protocol** (*str*) – the protocol string (e.g., mongodb, riak)\n* **driver_class** ([*passerine.db.driver.interface.DriverInterface*](index.html#passerine.db.driver.interface.DriverInterface)) – a DriverInterface-based class (type)\n |\n\n`set`(*alias*, *url*)[¶](#passerine.db.manager.ManagerFactory.set)\nDefine the database endpoint URL to a manager (identified by the given alias).\n\n| Parameters: | * **alias** (*str*) – the alias of the manager.\n* **url** (*str*) – the URL to the endpoint\n |\n\n### passerine.db.mapper[¶](#passerine-db-mapper)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/mapper.py`.\n\nNote\n\nThe current implementation doesn’t support merging or detaching a document simultaneously observed by at least two entity manager.\n\n*class* `passerine.db.mapper.``AssociationFactory`(*origin*, *guide*, *cascading_options*, *is_reverse_mapping*)[¶](#passerine.db.mapper.AssociationFactory)\nAssociation Factory\n\n`class_name`[¶](#passerine.db.mapper.AssociationFactory.class_name)\nAuto-generated Association Class Name\n\n| Return type: | str |\n\nNote\n\nThis is a read-only property.\n\n`cls`[¶](#passerine.db.mapper.AssociationFactory.cls)\nAuto-generated Association Class\n\n| Return type: | type |\n\nNote\n\nThis is a read-only property.\n\n`collection_name`[¶](#passerine.db.mapper.AssociationFactory.collection_name)\nAuto-generated Collection Name\n\n| Return type: | str |\n\nNote\n\nThis is a read-only property.\n\n`destination`[¶](#passerine.db.mapper.AssociationFactory.destination)\nDestination\n\n| Return type: | type |\n\n`origin`[¶](#passerine.db.mapper.AssociationFactory.origin)\nOrigin\n\n| Return type: | type |\n\n*class* `passerine.db.mapper.``AssociationType`[¶](#passerine.db.mapper.AssociationType)\nAssociation Type\n\n`AUTO_DETECT` *= 1*[¶](#passerine.db.mapper.AssociationType.AUTO_DETECT)\nAuto detection (default, disabled and raising exception)\n\n`MANY_TO_MANY` *= 5*[¶](#passerine.db.mapper.AssociationType.MANY_TO_MANY)\nMany-to-many association mode\n\n`MANY_TO_ONE` *= 4*[¶](#passerine.db.mapper.AssociationType.MANY_TO_ONE)\nMany-to-one association mode\n\n`ONE_TO_MANY` *= 3*[¶](#passerine.db.mapper.AssociationType.ONE_TO_MANY)\nOne-to-many association mode\n\n`ONE_TO_ONE` *= 2*[¶](#passerine.db.mapper.AssociationType.ONE_TO_ONE)\nOne-to-one association mode\n\n*static* `known_type`(*t*)[¶](#passerine.db.mapper.AssociationType.known_type)\nCheck if it is a known type\n\n| Parameters: | **t** (*int*) – type |\n| Returns: | `True` if it is a known type. |\n| Return type: | bool |\n\n*class* `passerine.db.mapper.``BasicGuide`(*target_class*, *association*)[¶](#passerine.db.mapper.BasicGuide)\nBasic Relation Guide\n\nThis class is abstract and used with the relational map of the given entity class.\n\n| Parameters: | * **target_class** (*object*) – the target class or class name (e.g., acme.entity.User)\n* **association** (*int*) – the type of association\n |\n\n`target_class`[¶](#passerine.db.mapper.BasicGuide.target_class)\nThe target class\n\n| Return type: | type |\n\n*class* `passerine.db.mapper.``CascadingType`[¶](#passerine.db.mapper.CascadingType)\nCascading Type\n\n`DELETE` *= 2*[¶](#passerine.db.mapper.CascadingType.DELETE)\nCascade on delete operation\n\n`DETACH` *= 4*[¶](#passerine.db.mapper.CascadingType.DETACH)\nCascade on detach operation\n\nNote\n\nSupported in Tori 2.2\n\n`MERGE` *= 3*[¶](#passerine.db.mapper.CascadingType.MERGE)\nCascade on merge operation\n\nNote\n\nSupported in Tori 2.2\n\n`PERSIST` *= 1*[¶](#passerine.db.mapper.CascadingType.PERSIST)\nCascade on persist operation\n\n`REFRESH` *= 5*[¶](#passerine.db.mapper.CascadingType.REFRESH)\nCascade on refresh operation\n\n*class* `passerine.db.mapper.``RelatingGuide`(*entity_class*, *target_class*, *inverted_by*, *association*, *read_only*, *cascading_options*)[¶](#passerine.db.mapper.RelatingGuide)\nRelation Guide\n\nThis class is used with the relational map of the given entity class.\n\n| Parameters: | * **entity_class** (*type*) – the reference of the current class\n* **mapped_by** (*str*) – the name of property of the current class\n* **target_class** (*type*) – the target class or class name (e.g., acme.entity.User)\n* **inverted_by** (*str*) – the name of property of the target class\n* **association** (*int*) – the type of association\n* **read_only** (*bool*) – the flag to indicate whether this is for read only.\n* **cascading_options** (*list or tuple*) – the list of actions on cascading\n |\n\n`passerine.db.mapper.``link`(*mapped_by=None*, *target=None*, *inverted_by=None*, *association=1*, *read_only=False*, *cascading=[]*)[¶](#passerine.db.mapper.link)\nAssociation decorator\n\nNew in version 2.1.\n\nThis is to map a property of the current class to the target class.\n\n| Parameters: | * **mapped_by** (*str*) – the name of property of the current class\n* **target** (*type*) – the target class or class name (e.g., acme.entity.User)\n* **inverted_by** (*str*) – the name of property of the target class\n* **association** (*int*) – the type of association\n* **read_only** (*bool*) – the flag to indicate whether this is for read only.\n* **cascading** (*list or tuple*) – the list of actions on cascading\n |\n| Returns: | the decorated class |\n| Return type: | type |\n\nTip\n\nIf `target` is not defined, the default target will be the reference class.\n\n`passerine.db.mapper.``map`(*cls*, *mapped_by=None*, *target=None*, *inverted_by=None*, *association=1*, *read_only=False*, *cascading=[]*)[¶](#passerine.db.mapper.map)\nMap the given class property to the target class.\n\nNew in version 2.1.\n\n| Parameters: | * **cls** (*type*) – the reference of the current class\n* **mapped_by** (*str*) – the name of property of the current class\n* **target** (*type*) – the target class or class name (e.g., acme.entity.User)\n* **inverted_by** (*str*) – the name of property of the target class\n* **association** (*int*) – the type of association\n* **read_only** (*bool*) – the flag to indicate whether this is for read only.\n* **cascading** (*list or tuple*) – the list of actions on cascading\n |\n\n### passerine.db.metadata.entity[¶](#passerine-db-metadata-entity)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/metadata/entity.py`.\n\n*class* `passerine.db.metadata.entity.``EntityMetadata`[¶](#passerine.db.metadata.entity.EntityMetadata)\nEntity Metadata\n\n`cls`[¶](#passerine.db.metadata.entity.EntityMetadata.cls)\nEntity Class\n\n`collection_name`[¶](#passerine.db.metadata.entity.EntityMetadata.collection_name)\nCollection / Bucket / Table Name\n\n`index_list`[¶](#passerine.db.metadata.entity.EntityMetadata.index_list)\nIndex List\n\n`relational_map`[¶](#passerine.db.metadata.entity.EntityMetadata.relational_map)\nRelational Map\n\n### passerine.db.metadata.helper[¶](#passerine-db-metadata-helper)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/metadata/helper.py`.\n\n*class* `passerine.db.metadata.helper.``EntityMetadataHelper`[¶](#passerine.db.metadata.helper.EntityMetadataHelper)\nEntity Metadata Helper\n\n*static* `extract`()[¶](#passerine.db.metadata.helper.EntityMetadataHelper.extract)\nExtract the metadata of the given class\n\n| Parameters: | **cls** (*type*) – the entity class |\n| Return type: | [passerine.db.metadata.entity.EntityMetadata](index.html#passerine.db.metadata.entity.EntityMetadata) |\n\n*static* `hasMetadata`()[¶](#passerine.db.metadata.helper.EntityMetadataHelper.hasMetadata)\nCheck if the given class *cls* has a metadata\n\n| Parameters: | **cls** (*type*) – the entity class |\n| Return type: | bool |\n\n*static* `imprint`(*collection_name*, *indexes*)[¶](#passerine.db.metadata.helper.EntityMetadataHelper.imprint)\nImprint the entity metadata to the class (type)\n\n| Parameters: | * **cls** (*type*) – the entity class\n* **collection_name** (*str*) – the name of the collection (known as table, bucket etc.)\n* **indexes** (*list*) – the list of indexes\n |\n\n### passerine.db.mochi[¶](#passerine-db-mochi)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/mochi.py`.\n\n#### Mochi Non-relational Database Library[¶](#mochi-non-relational-database-library)\n\n| Author: | <NAME> <[<EMAIL>](mailto:jnopporn%40shiroyuki.com)> |\n\nThis library is designed to work like SQLite but be compatible with MongoDB instructions (and PyMongo interfaces).\n\n### passerine.db.query[¶](#passerine-db-query)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/query.py`.\n\n#### [`passerine.db.criteria`](#module-passerine.db.criteria) – Query Criteria[¶](#module-passerine.db.criteria)\n\n*class* `passerine.db.query.``Order`[¶](#passerine.db.query.Order)\nSorting Order Definition\n\n`ASC`[¶](#passerine.db.query.Order.ASC)\nAscending Order\n\nalias of `ASCENDING`\n\n`DESC`[¶](#passerine.db.query.Order.DESC)\nDescending Order\n\nalias of `DESCENDING`\n\n*class* `passerine.db.query.``Query`(*alias*)[¶](#passerine.db.query.Query)\nCriteria\n\nNote\n\nThe current implementation does not support filtering on associated entities.\n\n`criteria`[¶](#passerine.db.query.Query.criteria)\nExpression Criteria\n\n`define`(*variable_name=None*, *value=None*, ***definition_map*)[¶](#passerine.db.query.Query.define)\nDefine the value of one or more variables (known as parameters).\n\n| Parameters: | * **variable_name** (*str*) – the name of the variable (for single assignment)\n* **value** – the value of the variable (for single assignment)\n* **definition_map** – the variable-to-value dictionary\n |\n\nThis method is usually recommended be used to define multiple variables like the following example.\n```\ncriteria.define(foo = 'foo', bar = 2)\n```\nHowever, it is designed to support the assign of a single user. For instance,\n```\ncriteria.define('foo', 'foo').define('bar', 2)\n```\n`expect`(*statement*)[¶](#passerine.db.query.Query.expect)\nDefine the condition / expectation of the main expression.\n\n| Parameters: | **statement** (*str*) – the conditional statement |\n\nThis is a shortcut expression to define expectation of the main expression. The main expression will be defined automatically if it is undefined. For example,\n```\nc = Query()\nc.expect('foo = 123')\n```\nis the same thing as\n```\nc = Query()\nc.criteria = c.new_criteria()\nc.criteria.expect('foo = 123')\n```\n`join`(*property_path*, *alias*)[¶](#passerine.db.query.Query.join)\nDefine a join path\n\n`join_map`[¶](#passerine.db.query.Query.join_map)\nA join map\n\n`limit`(*limit*)[¶](#passerine.db.query.Query.limit)\nDefine the filter limit\n\n| Parameters: | **limit** (*int*) – the filter limit |\n\n`new_criteria`()[¶](#passerine.db.query.Query.new_criteria)\nGet a new expression for this criteria\n\n| Return type: | [passerine.db.expression.Criteria](index.html#passerine.db.expression.Criteria) |\n\n`order`(*field*, *direction=<class 'ASCENDING'>*)[¶](#passerine.db.query.Query.order)\nDefine the returning order\n\n| Parameters: | * **field** (*str*) – the sorting field\n* **direction** – the sorting direction\n |\n\n`start`(*offset*)[¶](#passerine.db.query.Query.start)\nDefine the filter offset\n\n| Parameters: | **offset** (*int*) – the filter offset |\n\n### passerine.db.repository[¶](#passerine-db-repository)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/repository.py`.\n\n| Author: | <NAME> <[<EMAIL>](mailto:<EMAIL>)> |\n| Status: | Stable |\n\n*class* `passerine.db.repository.``Repository`(*session*, *representing_class*)[¶](#passerine.db.repository.Repository)\nRepository (Entity AbstractRepository) for Mongo DB\n\n| Parameters: | * **session** ([*passerine.db.session.Session*](index.html#passerine.db.session.Session)) – the entity manager\n* **representing_class** (*type*) – the representing class\n |\n\nA repository may automatically attempt to create an index if [`auto_index()`](#passerine.db.repository.Repository.auto_index)\ndefine the auto-index flag. Please note that the auto-index feature is only invoked when it tries to use a criteria with sorting or filtering with a certain type of conditions.\n\n`auto_index`(*auto_index*)[¶](#passerine.db.repository.Repository.auto_index)\nEnable the auto-index feature\n\n| Parameters: | **auto_index** (*bool*) – the index flag |\n\n`count`(*criteria*)[¶](#passerine.db.repository.Repository.count)\nCount the number of entities satisfied the given criteria\n\n| Parameters: | **criteria** (*passerine.db.criteria.Query*) – the search criteria |\n| Return type: | int |\n\n`filter`(*condition={}*, *force_loading=False*)[¶](#passerine.db.repository.Repository.filter)\nShortcut method for [:method:`find`](#id1).\n\n`filter_one`(*condition={}*, *force_loading=False*)[¶](#passerine.db.repository.Repository.filter_one)\nShortcut method for [:method:`find`](#id3).\n\n`find`(*criteria=None*, *force_loading=False*)[¶](#passerine.db.repository.Repository.find)\nFind entity with criteria\n\n| Parameters: | * **criteria** (*passerine.db.criteria.Query*) – the search criteria\n* **force_loading** (*bool*) – the flag to force loading all references behind the proxy\n |\n| Returns: | the result based on the given criteria |\n| Return type: | object or list of objects |\n\n`index`(*index*, *force_index=False*)[¶](#passerine.db.repository.Repository.index)\nIndex data\n\n| Parameters: | * **index** (*list, passerine.db.entity.Index or str*) – the index\n* **force_index** (*bool*) – force indexing if necessary\n |\n\n`name`[¶](#passerine.db.repository.Repository.name)\nCollection name\n\n| Return type: | str |\n\n`new`(***attributes*)[¶](#passerine.db.repository.Repository.new)\nCreate a new document/entity\n\n| Parameters: | **attributes** – attribute map |\n| Returns: | object |\n\nNote\n\nThis method deal with data mapping.\n\n`new_criteria`(*alias='e'*)[¶](#passerine.db.repository.Repository.new_criteria)\nCreate a criteria\n\n| Return type: | `passerine.db.criteria.Query` |\n\n`session`[¶](#passerine.db.repository.Repository.session)\nSession\n\n| Return type: | [passerine.db.session.Session](index.html#passerine.db.session.Session) |\n\n`setup_index`()[¶](#passerine.db.repository.Repository.setup_index)\nSet up index for the entity based on the `entity` and `link` decorators\n\n### passerine.db.session[¶](#passerine-db-session)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/session.py`.\n\n*class* `passerine.db.session.``Session`(*driver*)[¶](#passerine.db.session.Session)\nDatabase Session\n\n| Parameters: | * **database_name** – the database name\n* **driver** – the driver API\n |\n\n`apply_relational_map`(*entity*)[¶](#passerine.db.session.Session.apply_relational_map)\nWire connections according to the relational map\n\n`collection`(*entity_class*)[¶](#passerine.db.session.Session.collection)\nAlias to `repository()`\n\nDeprecated since version 2.2.\n\n`delete`(**entities*)[¶](#passerine.db.session.Session.delete)\nDelete entities\n\n| Parameters: | **entities** (*type of list of type*) – one or more entities |\n\n`flush`()[¶](#passerine.db.session.Session.flush)\nFlush all changes of the session.\n\n`persist`(**entities*)[¶](#passerine.db.session.Session.persist)\nPersist entities\n\n| Parameters: | **entities** (*type of list of type*) – one or more entities |\n\n`query`(*query*)[¶](#passerine.db.session.Session.query)\nQuery the data\n\n| Parameters: | **query** ([*passerine.db.query.Query*](index.html#passerine.db.query.Query)) – the query object |\n| Returns: | the list of matched entities |\n| Return type: | list |\n\n`refresh`(**entities*)[¶](#passerine.db.session.Session.refresh)\nRefresh entities\n\n| Parameters: | **entities** (*type of list of type*) – one or more entities |\n\n`register_class`(*entity_class*)[¶](#passerine.db.session.Session.register_class)\nRegister the entity class\n\n| Parameters: | **entity_class** (*type*) – the class of document/entity |\n| Return type: | [passerine.db.repository.Repository](index.html#passerine.db.repository.Repository) |\n\nNote\n\nThis is for internal operation only. As it seems to be just a residual from the prototype stage, the follow-up investigation in order to remove the method will be for Tori 3.1.\n\n`repositories`()[¶](#passerine.db.session.Session.repositories)\nRetrieve the list of collections\n\n| Return type: | list |\n\n`repository`(*reference*)[¶](#passerine.db.session.Session.repository)\nRetrieve the collection\n\n| Parameters: | **reference** – the entity class or entity metadata of the target repository / collection |\n| Return type: | [passerine.db.repository.Repository](index.html#passerine.db.repository.Repository) |\n\n### passerine.db.uow[¶](#passerine-db-uow)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/uow.py`.\n\n*class* `passerine.db.uow.``DependencyNode`(*record*)[¶](#passerine.db.uow.DependencyNode)\nDependency Node\n\nThis is designed to be bi-directional to maximize flexibility on traversing the graph.\n\n*class* `passerine.db.uow.``UnitOfWork`(*entity_manager*)[¶](#passerine.db.uow.UnitOfWork)\nUnit of Work\n\nThis Unit of Work (UOW) is designed specifically for non-relational databases.\n\nNote\n\nIt is the design decision to make sub-commit methods available so that when it is used with Imagination Framework, the other Imagination entity may intercept before or after actually committing data. In the other word, Imagination Framework acts as an event controller for any actions (public methods) of this class.\n\n`refresh`(*entity*)[¶](#passerine.db.uow.UnitOfWork.refresh)\nRefresh the entity\n\nNote\n\nThis method\n\n| Parameters: | **entity** (*object*) – the target entity |\n\n`register_clean`(*entity*)[¶](#passerine.db.uow.UnitOfWork.register_clean)\nRegister the entity with the clean bit\n\n| Parameters: | **entity** (*object*) – the entity to register |\n\n`register_deleted`(*entity*)[¶](#passerine.db.uow.UnitOfWork.register_deleted)\nRegister the entity with the removal bit\n\n| Parameters: | **entity** (*object*) – the entity to register |\n\n`register_dirty`(*entity*)[¶](#passerine.db.uow.UnitOfWork.register_dirty)\nRegister the entity with the dirty bit\n\n| Parameters: | **entity** (*object*) – the entity to register |\n\n`register_new`(*entity*)[¶](#passerine.db.uow.UnitOfWork.register_new)\nRegister a new entity\n\n| Parameters: | **entity** (*object*) – the entity to register |\n\n### passerine.db.wrapper[¶](#passerine-db-wrapper)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/db/wrapper.py`.\n\n### passerine.decorator.common[¶](#passerine-decorator-common)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/decorator/common.py`.\n\n| Author: | <NAME> |\n\nThis package contains decorators for common use.\n\n*class* `passerine.decorator.common.``BaseDecoratorForCallableObject`(*reference*)[¶](#passerine.decorator.common.BaseDecoratorForCallableObject)\nBase decorator based from an example at <http://www.artima.com/weblogs/viewpost.jsp?thread=240808>.\n\n`passerine.decorator.common.``make_singleton_class`(*class_reference*, **args*, ***kwargs*)[¶](#passerine.decorator.common.make_singleton_class)\nMake the given class a singleton class.\n\n*class_reference* is a reference to a class type, not an instance of a class.\n\n*args* and *kwargs* are parameters used to instantiate a singleton instance.\n\nTo use this, suppose we have a class called `DummyClass` and later instantiate a variable `dummy_instnace` as an instance of class `DummyClass`. `class_reference`\nwill be `DummyClass`, not `dummy_instance`.\n\nNote that this method is not for direct use. Always use `@singleton` or `@singleton_with`.\n\n`passerine.decorator.common.``singleton`(**args*, ***kwargs*)[¶](#passerine.decorator.common.singleton)\nDecorator to make a class to be a singleton class. This decorator is designed to be able to take parameters for the construction of the singleton instance.\n\nPlease note that this decorator doesn’t support the first parameter as a class reference. If you are using that way, please try to use\n`@singleton_with` instead.\n\nExample:\n```\n# Declaration\n@singleton class MyFirstClass(ParentClass):\n def __init__(self):\n self.number = 0\n def call(self):\n self.number += 1\n echo self.number\n# Or\n@singleton(20)\nclass MySecondClass(ParentClass):\n def __init__(self, init_number):\n self.number = init_number\n def call(self):\n self.number += 1\n echo self.number\n\n# Executing for i in range(10):\n MyFirstClass.instance().call()\n# Expecting 1-10 to be printed on the console.\nfor i in range(10):\n MySecondClass.instance().call()\n# Expecting 11-20 to be printed on the console.\n```\nThe end result is that the console will show the number from 1 to 10.\n\n`passerine.decorator.common.``singleton_with`(**args*, ***kwargs*)[¶](#passerine.decorator.common.singleton_with)\nDecorator to make a class to be a singleton class with given parameters for the constructor.\n\nPlease note that this decorator always requires parameters. Not giving one may result errors. Additionally, it is designed to solve the problem where the first parameter is a class reference. For normal usage, please use\n@singleton instead.\n\nExample:\n```\n# Declaration class MyAdapter(AdapterClass):\n def broadcast(self):\n print \"Hello, world.\"\n\n@singleton_with(MyAdapter)\nclass MyClass(ParentClass):\n def __init__(self, adapter):\n self.adapter = adapter()\n def take_action(self):\n self.adapter.broadcast()\n\n# Executing MyClass.instance().take_action() # expecting the message on the console.\n```\nThe end result is that the console will show the number from 1 to 10.\n\n### passerine.exception[¶](#passerine-exception)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/exception.py`.\n\n*exception* `passerine.exception.``DuplicatedPortError`[¶](#passerine.exception.DuplicatedPortError)\nException thrown only when the port config is duplicated within the same configuration file.\n\n*exception* `passerine.exception.``DuplicatedRouteError`[¶](#passerine.exception.DuplicatedRouteError)\nException used when the routing pattern is already registered.\n\n*exception* `passerine.exception.``FutureFeatureException`[¶](#passerine.exception.FutureFeatureException)\nException used when the future feature is used where it is not properly implemented.\n\n*exception* `passerine.exception.``InvalidConfigurationError`[¶](#passerine.exception.InvalidConfigurationError)\nException thrown only when the configuration is invalid.\n\n*exception* `passerine.exception.``InvalidControllerDirectiveError`[¶](#passerine.exception.InvalidControllerDirectiveError)\nException used when the controller directive is incomplete due to missing parameter\n\n*exception* `passerine.exception.``InvalidInput`[¶](#passerine.exception.InvalidInput)\nException used when the given input is invalid or incompatible to the requirement.\n\n*exception* `passerine.exception.``InvalidRedirectionDirectiveError`[¶](#passerine.exception.InvalidRedirectionDirectiveError)\nException used when the redirection directive is incomplete because some parameters aren’t provided or incompatible.\n\n*exception* `passerine.exception.``LoadedFixtureException`[¶](#passerine.exception.LoadedFixtureException)\nException raised when the fixture is loaded.\n\n*exception* `passerine.exception.``RendererNotFoundError`[¶](#passerine.exception.RendererNotFoundError)\nException thrown when the unknown template repository is used.\n\n*exception* `passerine.exception.``RendererSetupError`[¶](#passerine.exception.RendererSetupError)\nException thrown when there exists errors during setting up the template.\n\n*exception* `passerine.exception.``RenderingSourceMissingError`[¶](#passerine.exception.RenderingSourceMissingError)\nException used when the rendering source is not set.\n\n*exception* `passerine.exception.``RoutingPatternNotFoundError`[¶](#passerine.exception.RoutingPatternNotFoundError)\nException used when the routing pattern is not specified in the configuration file.\n\n*exception* `passerine.exception.``RoutingTypeNotFoundError`[¶](#passerine.exception.RoutingTypeNotFoundError)\nException used when the routing type is not specified in the configuration file.\n\n*exception* `passerine.exception.``SessionError`[¶](#passerine.exception.SessionError)\nException thrown when there is an error with session component.\n\n*exception* `passerine.exception.``SingletonInitializationException`[¶](#passerine.exception.SingletonInitializationException)\nThis exception is used when the target class contain a special attribute _singleton_instance not a reference to its own class.\n\n*exception* `passerine.exception.``UnexpectedComputationError`[¶](#passerine.exception.UnexpectedComputationError)\nException used when the code runs mistakenly unexpectedly.\n\n*exception* `passerine.exception.``UnknownRoutingTypeError`[¶](#passerine.exception.UnknownRoutingTypeError)\nException used when the routing type is not unknown.\n\n*exception* `passerine.exception.``UnknownServiceError`[¶](#passerine.exception.UnknownServiceError)\nException thrown when the requested service is unknown or not found.\n\n*exception* `passerine.exception.``UnsupportObjectTypeError`[¶](#passerine.exception.UnsupportObjectTypeError)\nException used when the unsupported object type is used in an inappropriate place.\n\nPlease note that this is a general exception.\n\n*exception* `passerine.exception.``UnsupportedRendererError`[¶](#passerine.exception.UnsupportedRendererError)\nException thrown when the unsupported renderer is being registered.\n\n### passerine.graph[¶](#passerine-graph)\n\nNote\n\nThis page is automatically generated. If you don’t see anything, this means this sub-module is not meant to be used. If you really want to know what it is, please check out the source code at `passerine/graph.py`.\n\n*class* `passerine.graph.``DependencyNode`[¶](#passerine.graph.DependencyNode)\nDependency Node\n\nThis is designed to be bi-directional to maximize flexibility on traversing the graph.\n\nGuide[¶](#guide)\n---\n\nThis section will guide you on how to use the library properly.\n\n### 1. Getting Started[¶](#getting-started)\n\nNow, you get through the boring stuff. Let’s write some code.\n\n#### Set Up the Connection and Entity Manager[¶](#set-up-the-connection-and-entity-manager)\n```\nfrom passerine.db.manager import ManagerFactory manager_factory = ManagerFactory()\n\n# Define the connection URL manager_factory.set('default', 'mongodb://localhost/sample_db')\n\n# Get the \"default\" entity manager entity_manager = manager_factory.get('default')\n```\nThe first three two is to set up the entity manager factory (`ManagerFactory`).\n\nThen, on the next line, we define the **default** connection to `mongodb://localhost/sample_database`.\n\nThen, use the entity manager factory to get an instance of the **default** entity manager.\n\n#### Define an Entity[¶](#define-an-entity)\n\nNote\n\nFrom this section on, **collection** (used by NoSQL DBs), **bucket** (used by distributed DBs) and **table**\n(used by relational DBs) are collectively the same thing for the library.\n\nNow, we have an entity manager to work with. Next, we need to define the data structure. For instance, we define two entity classes: **Character** and **Team**.\n```\nfrom passerine.db.entity import entity\n\n@entity class Player(object):\n def __init__(self, name, level, team=None):\n self.name = name\n self.team = team\n self.level = level\n\n def __repr__(self):\n attrs = {\n 'name': self.name,\n 'team': self.team,\n 'level': self.level\n }\n\n return '<{} {}>'.format(self.__class__.__name__, attrs)\n\n@entity('teams')\nclass Team(object):\n def __init__(self, name, location):\n self.name = name\n self.location = location\n\n def __repr__(self):\n attrs = {\n 'name': self.name,\n 'location': self.location\n }\n\n return '<{} {}>'.format(self.__class__.__name__, attrs)\n```\n1. `@entity` is to define the decorated class as an entity class. If all entities of this class will be saved in the collection, named **character**.\n2. `@entity('teams')` is similar to `@entity` except that the name of the destination collection is **teams**.\n3. The constructor must never take an identifier (e.g. `id` or `_id`) as a parameter.\n4. The name of the parameters of the constructor must be the same as the property.\n\nNote\n\nCurrently, data-object mapping is very straight forward. If the given data is\n```\n// mongodb: sample_db.teams team_raw_data = {\n '_id': 123,\n 'name': '<NAME>',\n 'location': 'Wakayama, Japan'\n}\n```\nthen, the data mapper will try to do somthing similar to the following code:\n```\nteam_entity = Team(\n name = team_raw_data['name'],\n location = team_raw_data['location'],\n)\nteam_entity.id = team_raw_data['_id']\n```\nThe proper data mapping mechanism will be introduced in later releases.\n\nYou may use a property getter or setter to restrict the access to the property if needed.\n\nNow, we have **Entity** classes and a working entity manager. What can we do next?\n\n### 2. Create a New Entity[¶](#create-a-new-entity)\n\nIn the previous step, we have an entity manager (`entity_manager`) and two entity classes (`Player` and `Team`).\nNow, we are going to create a new entity.\n\nLet’s create **one player**.\n```\nsid = Player('Sid', 5)\n\nsession = entity_manager.open_session()\n\nrepository = session.repository(Player)\nrepository.persist(sid)\nrepository.commit()\n```\nFrom the code, `sid = Player('Sid')` is to create an object.\n\nThe rest are pretty simple.\n\n1. We open a DB (pseudo) session with `open_session()`,\n2. find the corresponding repository, referred as `repository`,\n3. tell the repository to push the new object on commit with `persist()`,\n4. and finally commit the change to the database.\n\nNote\n\nAlternatively, you may forget `repository` by using `session` directly by replacing everything about `repository` with the following code.\n```\nsession.persist(sid)\nsession.flush()\n```\nwhere the session’s `persist` and `commit` are the same as the repository’s\n`persist` and `flush` respectively. **For the sake of tutorial, we will keep using the repository approach.**\n\nWarning\n\nThe database session is pseudo for unsupported databases.\n\nNote\n\nThe way we use `open_session()` here is to open an unsupervised session. The session connection cannot be closed by the supervising entity manager but the connection is still closed as soon as the process is terminated.\n\nAfter this process, you can verify with MongoDB CLI (`mongo sample_db`) by running `db.player.find()`. You should be able to see the result on the screen like the following (except `_id`).\n```\n{\n \"_id\" : ObjectId(\"abc\"),\n \"name\" : \"Sid\",\n \"level\": 5,\n \"team\" : null\n}\n```\n### 3. Find entities (Basic)[¶](#find-entities-basic)\n\nPreviously, we added an entity. Now, we just need to find it properly.\n\nHowever, just before we move onto the next step, let’s add more Player entities.\n```\n# Add more entities repository.persist(Player('Ramza', 9))\nrepository.persist(Player('Tsunemori', 6))\nrepository.commit()\n```\n#### Find many entities without constrains[¶](#find-many-entities-without-constrains)\n\nSuppose we want to retrieve everyone. There are two ways you can do it.\n\nFirst, just invoke the `find` method of `repository`.\n```\nresult = repository.find()\n```\nOr alternatively, you can create a new [`passerine.db.query.Query`](index.html#passerine.db.query.Query) via `repository`.\n```\nquery = repository.new_criteria('p')\n\nresult = repository.find(query)\n```\nYou should then see the same output (`result`) like the following:\n```\n[<Player {'level': 5, 'name': 'Sid', 'team': None}>, <Player {'level': 6, 'name': 'Tsunemori', 'team': None}>, <Player {'level': 9, 'name': 'Ramza', 'team': None}>]\n```\nWarning\n\n[`passerine.db.query.Query`](index.html#passerine.db.query.Query) must always be instantiated via [`passerine.db.repository.Repository`](index.html#passerine.db.repository.Repository).\nPlease do not try to instantiate the **Query** class directly.\n\n#### Find many entities with constrains[¶](#find-many-entities-with-constrains)\n\nSuppose we want to retrieve the information of the player named “Sid”. There are two ways you can do it.\n\n##### Hand-coded expectation[¶](#hand-coded-expectation)\n\nFirst, just hand-code the expected value.\n```\nquery = repository.new_criteria('p')\nquery.expect('p.name = \"Sid\"')\n\nresult = repository.find(query)\n```\n`query = repository.new_criteria('p')` is to instantiate a query object ([`passerine.db.query.Query`](index.html#passerine.db.query.Query)) where the alias of the expected entitiy is `p`. Then, `query.expect('p.name = \"Sid\"')` set the expectation that the attribute `name` of the expected entities must be equal to `Sid`. `result = repository.find(query)` will perform the query with the specification defined in `query`. If you execute the code, you should see the output (`result`) like this:\n```\n[<Player {'level': 5, 'name': 'Sid', 'team': None}>]\n```\nNote\n\nThe syntax for the expectation is ironically similar to SQL and DQL (Doctrine Query Language). It is by design to generalize the interface between different types of drivers.\n\n##### Parameterized expectation (exact match)[¶](#parameterized-expectation-exact-match)\n\nHowever, the previous `query` is not reusable with the dynamic data. We will create the Query object with\n**parameters** to allow use to reuse the same query object for different parameters.\n```\nquery = repository.new_criteria('p')\nquery.expect('p.name = :name')\nquery.define('name', 'Sid')\n\nresult = repository.find(query)\n```\nIn this snippet, you will see that in `query.expect('p.name = :name')`, `:name` replaces `\"Sid\"` and\n`query.define('name', 'Sid')` defines the value of the parameter `name`. This is equivalent to\n`query.expect('p.name = \"Sid\"')` as you will see the same output (`result`):\n```\n[<Player {'level': 5, 'name': 'Sid', 'team': None}>]\n```\nHowever, without recreating the Query object, if I re-define the parameter `name` with `Tsunemori` by adding:\n```\nquery.define('name', 'Tsunemori')\n```\nyou will now get the different result:\n```\n[<Player {'level': 6, 'name': 'Tsunemori', 'team': None}>]\n```\nNote\n\nThis is the recommended way to query.\n\n##### Range query[¶](#range-query)\n\nTo do range search, just like the previous examples, you can either hand-code the expectation or rely on the parameterization. The following example uses the latter:\n```\n# Range-query multiple query = repository.new_criteria('p')\nquery.expect('p.level >= :min')\nquery.expect('p.level <= :max')\nquery.define('min', 5)\nquery.define('max', 6)\n\nresult = repository.find(query)\n```\nAnd `result` becomes:\n```\n[\n <Player {'level': 5, 'name': 'Sid', 'team': None}>,\n <Player {'level': 6, 'name': 'Tsunemori', 'team': None}>\n]\n```\n##### In-set query (IN operator)[¶](#in-set-query-in-operator)\n\nTo do in-set search, unlike the previous examples, you can only rely on the parameterization. For example:\n```\nquery = repository.new_criteria('p')\nquery.expect('p.level IN :expected_level')\nquery.define('expected_level', (6, 9)) # or [6, 9]\n\nresult = repository.find(query)\n```\nAnd `result` becomes:\n```\n[\n <Player {'level': 9, 'name': 'Ramza', 'team': None}>,\n <Player {'level': 6, 'name': 'Tsunemori', 'team': None}>\n]\n```\n##### Regular-expression query (LIKE operator)[¶](#regular-expression-query-like-operator)\n\nTo do in-set search, unlike the previous examples, you can only rely on the parameterization. For example:\n```\nquery = repository.new_criteria('p')\nquery.expect('p.name LIKE :name')\nquery.define('name', '^Ra')\n\nresult = repository.find(query)\n```\nAnd `result` becomes:\n```\n[<Player {'level': 9, 'name': 'Ramza', 'team': None}>]\n```\nTo just retrieve everything, you can either do:\n```\nquery = repository.new_criteria()\nrepository.find(query)\n```\nor just go with:\n```\nrepository.find(query)\n```\n### 4. Associations and Relational Mappings[¶](#associations-and-relational-mappings)\n\nUsually data from different collections are related and may co-exist. As the result, even though some database softwares do not allow associations, Passerine allows the software-based relational mapping. This ORM supports all types of relational mapping either unidirectionally or bidirectionally.\n\nPasserine ORM utilizes two patterns to implement association.\n\n* Decorators (AKA annotations) to define associations.\n* Lazy loading to load data when it is requested.\n* Proxy object as direct result of lazy loading.\n\nIn general, the decorator [`passerine.db.mapper.link()`](index.html#passerine.db.mapper.link) is used to define association by mapping decorated fields to another classes by primary key (or object ID). The ID-to-object happens automatically during data mapping.\n\n#### Types of Associations[¶](#types-of-associations)\n\nFor the sake of the simplicity of this chapter, all examples are assumed to be in the module `sampleapp.model`, and begin with:\n```\nfrom passerine.db.entity import entity from passerine.db.mapper import link, AssociationType as t, CascadingType as c\n```\n##### One-to-one[¶](#one-to-one)\n\nSuppose there are two entities: `Owner` and `Restaurant`,\n**one-to-one associations** imply the relationship between two entities as described in the following UML:\n```\nOwner (1) --- (1) Restaurant\n```\n###### Unidirectional[¶](#unidirectional)\n\nUML:\n```\nOwner (1) <--x- (1) Restaurant\n```\nSuppose we have two classes: `Owner` and `Restaurant`, where `Restaurant`\nhas the one-to-one unidirectional relationship with `Owner`.\n```\n@entity class Owner(object):\n def __init__(self, name):\n self.name = name\n\n@link(\n target = 'sampleapp.model.Owner',\n mapped_by = 'owner',\n association = t.ONE_TO_ONE\n)\n@entity class Restaurant(object):\n def __init__(self, name, owner):\n self.name = name\n self.owner = owner\n```\nwhere the sample of the stored documents will be:\n```\n// collection: owner\n{'_id': 'o-1', 'name': 'siamese'}\n\n// collection: restaurant\n{'_id': 'rest-1', 'name': 'green curry', 'owner': 'o-1'}\n```\nTip\n\nTo avoid the issue with the order of declaration, the full namespace in string is recommended to define the target class. However, the type reference can also be. For example, `@link(target = Owner, ...)`.\n\n###### Bidirectional[¶](#bidirectional)\n\nUML:\n```\nOwner (1) <---> (1) Restaurant\n```\nNow, let’s allow `Owner` to have a reference back to `Restaurant` where the information about the reference is not kept with `Owner`. So, the\n```\n@link(\n target = 'sampleapp.model.Restaurant'\n inverted_by = 'owner',\n mapped_by = 'restaurant',\n association = t.ONE_TO_ONE\n)\n@entity class Owner(object):\n def __init__(self, name, restaurant):\n self.name = name\n self.restaurant = restaurant\n```\nwhere the the stored documents will be the same as the previous example.\n\n`inverted_by` means this class (`Owner`) maps `Restaurant` to the property\n*restaurant* where the value of the property *owner* of the corresponding entity of Restaurant must equal the *ID* of this class.\n\nNote\n\nThe option `inverted_by` only maps `Owner.restaurant` to `Restaurant`\nvirtually but the reference is stored in the **restaurant** collection.\n\n##### Many-to-one[¶](#many-to-one)\n\nSuppose a `Customer` can have many `Reward`‘s as illustrated:\n```\nCustomer (1) --- (0..n) Reward\n```\n###### Unidirectional[¶](#id1)\n\nUML:\n```\nCustomer (1) <--x- (0..n) Reward\n```\n```\n@entity class Customer(object):\n def __init__(self, name):\n self.name = name\n\n@link(\n target = 'sampleapp.model.Customer',\n mapped_by = 'customer',\n association = t.MANY_TO_ONE\n)\n@entity class Reward(object):\n def __init__(self, point, customer):\n self.point = point\n self.customer = customer\n```\nwhere the data stored in the database can be like this:\n```\n// collection: customer\n{'_id': 'c-1', 'name': 'panda'}\n\n// collection: reward\n{'_id': 'rew-1', 'point': 2, 'customer': 'c-1'}\n{'_id': 'rew-2', 'point': 13, 'customer': 'c-1'}\n```\n###### Bidirectional[¶](#manual-orm-associations-m-1-bidirectional)\n\nUML:\n```\nCustomer (1) <---> (0..n) Reward\n```\nJust change `Customer`.\n```\n@link(\n target = 'sampleapp.model.Reward',\n inverted_by = 'customer',\n mapped_by = 'rewards',\n association = t.ONE_TO_MANY\n)\n@entity class Customer(object):\n def __init__(self, name, rewards):\n self.name = name\n self.rewards = rewards\n```\nwhere the property *rewards* refers to a list of rewards but the stored data remains unchanged.\n\nNote\n\nThis mapping is equivalent to a **bidirectional one-to-many mapping**.\n\n##### One-to-many[¶](#one-to-many)\n\nLet’s restart the example from the many-to-one section.\n\n###### Unidirectional with Built-in List[¶](#unidirectional-with-built-in-list)\n\nThe one-to-many unidirectional mapping takes advantage of the built-in list.\n\nUML:\n```\nCustomer (1) -x--> (0..n) Reward\n```\n```\n@link(\n target = 'sampleapp.model.Reward',\n mapped_by = 'rewards',\n association = t.ONE_TO_MANY\n)\n@entity class Customer(object):\n def __init__(self, name, rewards):\n self.name = name\n self.rewards = rewards\n\n@entity class Reward(object):\n def __init__(self, point):\n self.point = point\n```\nwhere the property `rewards` is a unsorted iterable list of `Reward` objects and the data stored in the database can be like this:\n```\n// collection: customer\n{'_id': 'c-1', 'name': 'panda', 'reward': ['rew-1', 'rew-2']}\n\n// collection: reward\n{'_id': 'rew-1', 'point': 2}\n{'_id': 'rew-2', 'point': 13}\n```\nWarning\n\nAs there is no way to enforce relationships with built-in functionality of MongoDB and there will be constant checks for every write operation, it is not recommended to use unless it is for **reverse mapping** via the option\n`inverted_by` (see below for more information).\n\nWithout a proper checker, which is not provided for performance sake, this mapping can be used like the **many-to-many join-collection mapping**.\n\n###### Bidirectional[¶](#id3)\n\nSee [Many-to-one Bidirectional Association](#manual-orm-associations-m-1-bidirectional).\n\n##### Many-to-many[¶](#many-to-many)\n\nSuppose there are `Teacher` and `Student` where students can have many teachers and vise versa:\n```\nTeacher (*) --- (*) Student\n```\nSimilar other ORMs, the many-to-many mapping uses the corresponding join collection.\n\n###### Unidirectional with Join Collection[¶](#unidirectional-with-join-collection)\n\nUML:\n```\nTeacher (*) <--x- (*) Student\n```\n```\n@entity('teachers')\nclass Teacher(object):\n def __init__(self, name):\n self.name = name\n\n@link(\n mapped_by = 'teachers',\n target = Teacher,\n association = AssociationType.MANY_TO_MANY,\n cascading = [c.DELETE, c.PERSIST]\n)\n@entity('students')\nclass Student(object):\n def __init__(self, name, teachers=[]):\n self.name = name\n self.teachers = teachers\n```\nwhere the stored data can be like the following example:\n```\n// db.students.find()\n{'_id': 1, 'name': 'Shirou'}\n{'_id': 2, 'name': 'Shun'}\n{'_id': 3, 'name': 'Bob'}\n\n// db.teachers.find()\n{'_id': 1, 'name': '<NAME>'}\n{'_id': 2, 'name': 'Onizuka'}\n\n// db.students_teachers.find() // -> join collection\n{'_id': 1, 'origin': 1, 'destination': 1}\n{'_id': 2, 'origin': 1, 'destination': 2}\n{'_id': 3, 'origin': 2, 'destination': 2}\n{'_id': 4, 'origin': 3, 'destination': 1}\n```\n###### Bidirectional[¶](#id4)\n```\n@link(\n mapped_by = 'students',\n inverted_by = 'teachers',\n target = 'sampleapp.model',\n association = AssociationType.MANY_TO_MANY\n)\n@entity('teachers')\nclass Teacher(object):\n def __init__(self, name, students=[]):\n self.name = name\n self.students = students\n\n@link(\n mapped_by = 'teachers',\n target = Teacher,\n association = AssociationType.MANY_TO_MANY,\n cascading = [c.DELETE, c.PERSIST]\n)\n@entity('students')\nclass Student(object):\n def __init__(self, name, teachers=[]):\n self.name = name\n self.teachers = teachers\n```\n#### Options for Associations[¶](#options-for-associations)\n\nThe decorator [`passerine.db.mapper.link()`](index.html#passerine.db.mapper.link) has the following options:\n\n| Option | Description |\n| --- | --- |\n| association | the type of associations (See [`passerine.db.mapper.AssociationType`](index.html#passerine.db.mapper.AssociationType).) |\n| cascading | the list of allowed cascading operations (See [*6. Cascading*](index.html#document-guide/06-cascading) [`passerine.db.mapper.CascadingType`](index.html#passerine.db.mapper.CascadingType).) |\n| inverted_by | the name of property used where **enable the reverse mapping if defined** |\n| mapped_by | the name of property to be map |\n| read_only | the flag to disable property setters (only usable with [`passerine.db.common.ProxyObject`](index.html#passerine.db.common.ProxyObject).) |\n| target | the full name of class or the actual class |\n\n#### How to make a join query[¶](#how-to-make-a-join-query)\n\nFrom the customer-reward example, if we want to find all rewards of a particular user, the query will be:\n```\nquery = reward_repository.new_criteria('r')\nquery.join('r.customer', 'c')\nquery.expect('c.name = \"Bob\"')\n\nrewards = reward_repository.find(query)\n```\nAll features for querying is usable with joined entities.\n\n### 5. Updating, Deleting and Handling Transactions (Sessions)[¶](#updating-deleting-and-handling-transactions-sessions)\n\nSimilar to [Sessions in SQLAlchemy](http://docs.sqlalchemy.org/en/latest/orm/session.html).\n\nIn the most general sense, the session establishes all conversations with the database and represents a “holding zone” for all the objects which you’ve loaded or associated with it during its lifespan. It provides the entrypoint to acquire a [`passerine.db.repository.Repository`](index.html#passerine.db.repository.Repository) object, which sends queries to the database using the current database connection of the session ([`passerine.db.session.Session`](index.html#passerine.db.session.Session)),\npopulating result rows into objects that are then stored in the session, inside a structure called the identity map (internally being the combination of “the record map” and “the object ID map”) - a data structure that maintains unique copies of each object, where “unique” means “only one object with a particular primary key”.\n\nThe session begins in an essentially stateless form. Once queries are issued or other objects are persisted with it, it requests a connection resource from an manager that is associated with the session itself. This connection represents an ongoing transaction, which remains in effect until the session is instructed to commit.\n\nAll changes to objects maintained by a session are tracked - before the database is queried again or before the current transaction is committed, it flushes all pending changes to the database. This is known as **the Unit of Work pattern**.\n\nWhen using a session, it’s important to note that the objects which are associated with it are **proxy objects** ([`passerine.db.common.ProxyObject`](index.html#passerine.db.common.ProxyObject)) to the transaction being held by the session - there are a variety of events that will cause objects to re-access the database in order to keep synchronized. It is possible to “detach” objects from a session, and to continue using them, though this practice has its caveats. It’s intended that usually, you’d re-associate detached objects with another Session when you want to work with them again, so that they can resume their normal task of representing database state.\n\n#### Supported Operations[¶](#supported-operations)\n\n| Supported Operation | Supported Version |\n| --- | --- |\n| Persist | 2.1 |\n| Delete | 2.1 |\n| Refresh | 2.1 |\n| Merge | No plan at the moment |\n| Detach | No plan at the moment |\n\n#### Open a Session[¶](#open-a-session)\n\nAs you might guess from [*2. Create a New Entity*](index.html#document-guide/02-create-entity), Passerine always requires the code to open a session by:\n```\nsession = entity_manager.open_session()\n```\n#### Getting Started (again)[¶](#getting-started-again)\n\nThen, try to query a player called **“Sid”**, we created in [*2. Create a New Entity*](index.html#document-guide/02-create-entity)\nwith [`passerine.db.repository.Repository`](index.html#passerine.db.repository.Repository):\n```\nrepo = session.collection(Player)\n\nquery = repo.new_criteria('c')\nquery.expect('c.name = :name')\nquery.define('name', 'Sid')\n\nplayer = repo.find(query)\n\n# For demonstration only print('Sid/level: {}'.format(player.level))\n```\nThe output should show:\n```\nSid/level: 5\n```\n#### Update an Entity[¶](#update-an-entity)\n\nSuppose we want to update his level:\n```\nplayer.level = 99 repo.persist(player)\n```\n#### Delete an Entity[¶](#delete-an-entity)\n\nSuppose we want to delete the entity:\n```\nsession.delete(player)\n```\nNote\n\nUntil you commit the changes, other queries still can return entities you mark as deleted.\n\n#### Commit Changes[¶](#commit-changes)\n\nOnce I am satisfied with changes, commit the changes with:\n```\nrepo.commit()\n```\nor:\n```\nsession.flush()\n```\nNote\n\nWe will stick with `session.flush()` as `repo.commit()` is just an alias to `session.flush()`.\n\nNote\n\nTo discard all changes at any points, just close the session (mentioned later in this section).\n\n#### Refresh/revert Changes on One Entity[¶](#refresh-revert-changes-on-one-entity)\n\nOr, refresh `player`:\n```\nsession.refresh(player)\n```\nThen, if `player` is either **persisted** or **deleted**, to flush/commit the change, simply run:\n```\nsession.flush()\n```\n#### Close a Session[¶](#close-a-session)\n\nClosing a session is to end the session or discard all changes to the session.\nYou can simply close the session by:\n```\nentity_manager.close_session(session)\n```\n#### Handle a Session in the Context[¶](#handle-a-session-in-the-context)\n\nTo ensure all sessions are closed properly, you can open session as a context manager by doing this:\n```\nwith entity_manager.session() as session:\n ... # do whatever\n# session closed\n```\n#### Drawbacks Introduced by Either MongoDB or Passerine[¶](#drawbacks-introduced-by-either-mongodb-or-passerine)\n\n1. Even though MongoDB does not support transactions, like some relational database engines, such as, InnoDB, Passerine provides software-based transactions. However,\nas mentioned earlier, Passerine **does not provide roll-back operations**.\n2. **Merging** and **detaching** operations are currently not supported in 2013 unless someone provides the supporting code.\n3. Any querying operations cannot find any uncommitted changes.\n\n### 6. Cascading[¶](#cascading)\n\nThis is the one toughest section to write.\n\nMongoDB, as far as everyone knows, does not support cascading operations like the way MySQL and other vendors do with cascading deletion. Nevertheless, Tori supports cascading through the database abstraction layer (DBAL).\n\nWarning\n\nCascading persistence and removal via DBAL has high probability of degrading performance with large dataset as in order to calculate a dependency graph,\nall data must be loaded into the memory space of the computing process. This introduces a spike in memory and network usage.\n\nThis feature is introduced for convenience sake but should be used sparingly or accounted for potential performance degration.\n\nHere is a sample scenario.\n\nSuppose I have two types of objects: a sport team and a player. When a team is updated, removed or refreshed, the associated player should be treated the same way as the team. Here is a sample code.\n```\nfrom passerine.db.entity import entity from passerine.db.mapper import CascadingType as c\n\n@entity class Player(object):\n pass # omit the usual setup decribed in the basic usage.\n\n@link(\n target=Player,\n mapped_by='player',\n cascading=[c.PERSIST, c.DELETE, c.REFRESH]\n)\n@entity class Team(object):\n pass # omit the usual setup decribed in the basic usage.\n```\nNow, whatever operation is used on a Team entity, associated Player entites are subject to the same operation.\n\nChange Logs[¶](#change-logs)\n---\n\n### Version 1.4[¶](#version-1-4)\n\n#### Improvements[¶](#improvements)\n\n##### ManagerFactory[¶](#managerfactory)\n\n(Short for [`passerine.db.manager.ManagerFactory`](index.html#passerine.db.manager.ManagerFactory))\n\n* The alias-to-endpoint-URL map can be passed to the constructor to simplify the usage.\n* Shorten the name of each parameters of the constructor.\n```\nfrom passerine.db.manager import ManagerFactory\n\nmf = ManagerFactory(urls = {'default': 'mongodb://localhost/sample'})\n\nmf.get('default') # -> a Manager object\n```\n##### Others[¶](#others)\n\n* The default parameter of the constructor of an Entity class is now respected.\n\n#### House Cleaning[¶](#house-cleaning)\n\n* Made Python 3 (<3.4) the primary target.\n* Cleaned up the test folder.\n* Simplified the test running procedure.\n\n### Version 1.1[¶](#version-1-1)\n\n* (Support Riak 2.0)\n\n### Version 1.0[¶](#version-1-0)\n\n* A fork of Tori ORM (from Tori 3.0) with lots of little improvements.\n\nSpecial Thanks[¶](#special-thanks)\n---\n\nThis project is not possible without helps and guidance from <NAME> from [Doctrine Project](http://doctrine-project.org).\n\nIndices and Modules[¶](#indices-and-modules)\n---\n\n* [Index](genindex.html)\n* [Module Index](py-modindex.html)"}}},{"rowIdx":59170,"cells":{"project":{"kind":"string","value":"timber_ecto"},"source":{"kind":"string","value":"hex"},"language":{"kind":"string","value":"Erlang"},"content":{"kind":"string","value":"Timber.Ecto\n===\n\nTimber integration for Ecto.\n\nTimber can hook into Ecto's Telemetry events to gather information about queries including the text of the query and the time it took to execute. This information is then logged as a [`Timber.Events.SQLQueryEvent`](https://hexdocs.pm/timber/3.1.0/Timber.Events.SQLQueryEvent.html).\n\nTo install Timber's Ecto event collector, you will need to modify the application configuration on a per-repository basis. Each repository has to have the Timber handler attached. The best place to do it is in your application's `start` callback. It is important to note that the first argument must be unique, and the second is based on Ecto's default Telemetry event prefix. More information can be found in the docs for [`Ecto.Repo`](https://hexdocs.pm/ecto/3.0.6/Ecto.Repo.html).\n```\n# lib/my_app/application.ex def start(_type, _args) do\n # ...\n :ok = :telemetry.attach(\n \"timber-ecto-query-handler\",\n [:my_app, :repo, :query],\n &Timber.Ecto.handle_event/4,\n []\n )\n # ...\n Supervisor.start_link(children, opts)\nend\n```\nEach repository has a configuration key `:log` that controls whether Ecto logs query information. By default, it is enabled. In order to avoid duplicate logging, you will want to make sure it is set to false.\n```\nconfig :my_app, MyApp.Repo,\n log: false\n```\nBy default, queries are logged at the `:debug` level. If you want to use a different level, the `:log_level` option can be passed to the\n`Telemetry.attach` call:\n```\n:ok = :telemetry.attach(\n \"timber-ecto-query-handler\",\n [:my_app, :repo, :query],\n &Timber.Ecto.handle_event/4,\n [log_level: :info]\n )\n```\n### \n\n Timing\n\nThe time reported in the event is the amount of time the query took to execute on the database, as measured by Ecto. It does not include the time that the query spent in the pool's queue or the time spent decoding the response from the database.\n\n### \n\n Log only slow queries\n\nIf your queries are noisy you can specify a threshold that must be crossed in order for the query to be logged:\n```\n:ok = :telemetry.attach(\n \"timber-ecto-query-handler\",\n [:my_app, :repo, :query],\n &Timber.Ecto.handle_event/4,\n [query_time_ms_threshold: 2_000]\n )\n```\n```\nIn the above example, only queries that exceed 2 seconds in execution time will be logged.\n```\n[Link to this section](#summary)\n Summary\n===\n\n[Functions](#functions)\n---\n\n[handle_event(list, value, metadata, config)](#handle_event/4)\n\n[Link to this section](#functions)\nFunctions\n===\n\n"}}},{"rowIdx":59171,"cells":{"project":{"kind":"string","value":"php-cache"},"source":{"kind":"string","value":"readthedoc"},"language":{"kind":"string","value":"PHP"},"content":{"kind":"string","value":" This library provides a PSR-6 compliant bridge between Doctrine and a Cache Pool. The bridge implements the\n\ninterface. This is useful for projects that require an implementation of \n\n, but when you still want to use a PSR-6 implementation. \n\n## Installation¶\n\n```\ncomposer require cache/psr-6-doctrine-bridge\n```\n\n## Usage¶\n\n```\nuse DoctrineCacheBridge\\DoctrineCacheBridge;\n\n// Assuming $pool is an instance of \\Psr\\Cache\\CacheItemPoolInterface\n$cacheProvider = new DoctrineBridge($pool);\n\n$cacheProvider->contains($key);\n$cacheProvider->fetch($key);\n$cacheProvider->save($key, $value, $ttl);\n$cacheProvider->delete($key);\n\n// Also, if you need it:\n$cacheProvider->getPool(); // same as $pool\n```\n\nThis is a PSR-6 implementation of a hierarchical cache architecture.\n If you have a cache key like\n\n```\n|users|:uid|followers|:fid|likes\n```\n\nwhere `:uid` and `:fid` are arbitrary integers, you\n may flush all followers by flushing \n\n```\n|users|:uid|followers\n```\n\n```\n$user = 4711;\nfor ($i = 0; $i < 100; $i++) {\n $item = $pool->getItem(sprintf('|users|%d|followers|%d|likes', $user, $i));\n $item->set('<NAME>');\n $pool->save($item);\n}\n\n$pool->hasItem('|users|4711|followers|12|likes'); // True\n\n$pool->deleteItem('|users|4711|followers');\n\n$pool->hasItem('|users|4711|followers|12|likes'); // False\n```\n\nDate: 2015-01-01\nCategories: \nTags: \n\nIn the late fall 2015 the PHP-FIG released a specification for caching. It is called PSR-6. Like all the PSRs, it is just a recommendation for improve interoperability.\n\n## Basic concepts¶\n\nThe PSR-6 defines two interfaces; a CachePool and a CacheItem. You interact with the pool to get an item. You will never create an item yourself.\n\n```\n// Create a pool\n$pool = new ApcuCachePool();\n\n// Get an item (existing or new)\n$item = $pool->getItem('cache_key');\n\n// Set some values and store\n$item->set('value');\n$item->expiresAfter(60);\n$pool->save($item);\n\n// Verify existence\n$pool->hasItem('cache_key'); // True\n$item->isHit(); // True\n\n// Get stored values\n$myValue = $item->get();\necho $myValue; // \"value\"\n\n// Delete\n$pool->deleteItem('cache_key');\n$pool->hasItem('cache_key'); // False\n```\n\nNote: A cache item can not be transferred from one pool to another.\n\n## Cache keys¶\n\n According to PSR-6 the following characters could be used in a valid cache key; `A-Z` , `a-z` , `0-9` , `_` , and `.` . Some characters are forbidden like: `{}()/\\@:` . If\nyou use any of the forbidden characters you will get an exception. Other characters\n(like `-` ) are not forbidden nor valid. It is up to the implementation if they support\nthat character or not. \nWe recommend to always use valid characters in the cache key.\n\nTo make sure you do not use an invalid character by mistake, you should hash your keys.\n\n```\n$cacheKey = sha1($_SERVER['REQUEST_URI']);\n```\n\n## Choosing a pool¶\n\nWhat cache pool you want to use is up to you. As you can see in the table on the startpage, different pools have different features. Commonly used pools are:\n\n* Apcu\n * Array\n * Memcached\n * Redis\n\nThe steps to create a pool varies from pool to pool. Look at the pool's repository to see how to create it. Generally you create a connection to a cache implementation and then give it to the pool. See a Memcached example:\n\n```\n$client = new \\Memcached();\n$client->addServer('localhost', 11211);\n$pool = new MemcachedCachePool($client);\n```\n\n## Features¶\n\nThe PHP-Cache organization has built some more features on top of PSR-6. Look at our documentation for hierarchy or namespace for more information.\n\n The namespaced cache pool is a special cache of the hierarchical cache architecture where it only uses one level of hierarchy. Use this when you provide a third party library with a PSR-6 cache pool to make sure you never get key conflicts or unexpected calls to\n\n```\nCacheItemPoolInterface::clear()\n```\n\n```\n// Get a pool that supports hierarchy\n$client = new \\Redis();\n$client->connect('127.0.0.1', 6379);\n$pool = new RedisCachePool($client);\n\n// Decorate it with a NamespacedPool\n$namespacedPool = new NamespacedCachePool($pool, 'acme');\n\n$item = $namespacedPool->getItem('foo')->set('bar');\n$namespacedPool->save($item);\n\n$namespacedPool->hasItem('foo'); // True\n$pool->hasItem('foo'); // False\n```\n\n Internally the `NamespacedCachePool` will prepend all keys with `|acme|` . Which means that the `$pool` may access items in the namespace by doing the very same: \n\n```\n$pool->hasItem('|acme|foo'); // True\n```\n\n## Symfony integration¶\n\n If you are are using our bundles you may configure a service with as a namespaced pool by using the `pool_namespace` configuration option.\n\n This includes common resources that you are likely to use when implementing a cache pool. The `AbstractCachePool` and `CacheItem` live here. This includes common resources that you are likely to use when implementing a cache pool. The `AbstractCachePool` and `CacheItem` live here.\n\nThis bundle helps you configure and register PSR-6 cache services.\n\n## To Install¶\n\nRun the following in your project root, assuming you have composer set up for your project\n\n```\ncomposer require cache/adapter-bundle\n```\n\nAdd the bundle to app/AppKernel.php\n\n```\n$bundles = [\n // ...\n new Cache\\AdapterBundle\\CacheAdapterBundle(),\n];\n```\n\n## Configuration¶\n\nTo configure and register a cache pool adapter you need a factory service. You may also need to change some options to the factory. Below is the template which all adapter follow.\n\n```\ncache_adapter:\n providers:\n my_adapter:\n factory: 'service_id.to.my_adapter_factory'\n options: []\n aliases: ['alias.my_adapter']\n```\n\nThe factories that come with this bundle can be found in the table below.\n\nFactory service id | Options |\n | --- | --- |\ncache.factory.apc |\ncache.factory.apcu |\ncache.factory.array | |\n\nb'Factory service id\\n Options\\n \\n \\n cache.factory.apc\\n \\n \\n cache.factory.apcu\\n \\n \\n cache.factory.array\\n \\n pool_namespace\\n \\n \\n \\n cache.factory.chain\\n services, skip_on_failure\\n \\n \\n cache.factory.filesystem\\n \\n flysystem_service\\n \\n \\n \\n cache.factory.memcached\\n host, port, persistent_id, pool_namespace, redundant_servers\\n \\n \\n cache.factory.memcache\\n host, port, redundant_servers\\n \\n \\n cache.factory.mongodb\\n host, port, namespace, dsn\\n \\n \\n cache.factory.predis\\n host, port, schema, pool_namespace\\n \\n \\n cache.factory.redis\\n host, port, pool_namespace, dsn\\n \\n \\n cache.factory.void\\n \\n \\n cache.factory.namespaced\\n service, namespace\\n \\n \\n cache.factory.prefixed\\n service, prefix\\n \\n \\n cache.factory.doctrine_couchbase\\n host, user, password, bucket\\n \\n \\n cache.factory.doctrine_filesystem\\n directory, extension, umask\\n \\n \\n cache.factory.doctrine_memcached\\n host, port\\n \\n \\n cache.factory.doctrine_memcache\\n host, port\\n \\n \\n cache.factory.doctrine_mongodb\\n host, collection\\n \\n \\n cache.factory.doctrine_predis\\n host, port, schema\\n \\n \\n cache.factory.doctrine_redis\\n host, port\\n \\n \\n cache.factory.doctrine_riak\\n host, port, type\\n \\n \\n cache.factory.doctrine_sqlite3\\n file_path, table\\n \\n \\n cache.factory.doctrine_wincache\\n \\n \\n cache.factory.doctrine_xcache\\n \\n \\n cache.factory.doctrine_zenddata'\n\n### Example configuration¶\n\n```\ncache_adapter:\n providers:\n my_redis:\n factory: 'cache.factory.redis'\n options: \n host: 10.0.0.15 # Using custom host\n port: 6379\n my_memcached:\n factory: 'cache.factory.memcached'\n aliases: ['alias.my_memcached']\n options:\n host: 10.0.0.16\n port: 11218\n redundant_servers:\n - { host: 10.0.0.17, port: 11211 }\n - { host: 10.0.0.18 } # port 11218 will be used\n my_file_system:\n factory: 'cache.factory.filesystem'\n options:\n flysystem_service: 'oneup_flysystem.local_filesystem'\n my_apc:\n factory: 'cache.factory.apc' \n my_chain:\n factory: 'cache.factory.chain'\n options:\n services:\n - '@cache.provider.my_apc'\n - '@cache.provider.my_redis'\n```\n\n### Usage¶\n\n When using the example configuration above, you will get four services with the ids\n\n```\ncache.provider.my_redis\n```\n\n```\ncache.provider.my_memcached\n```\n\n```\ncache.provider.my_file_system\n```\n\nand \n\n```\ncache.provider.my_apc\n```\n\n. \nUse the new service as any PSR-6 cache.\n\n```\n/** @var CacheItemPoolInterface $pool */\n$pool = $this->container->get('cache.provider.my_memcached');\n\n/** or */\n$pool = $this->container->get('alias.my_memcached');\n\n/** @var CacheItemInterface $item */\n$item = $pool->getItem('cache-key');\n$item->set('foobar');\n$item->expiresAfter(3600);\n$pool->save($item);\n```\n\n The first configured adapter or a adapter named `default` will be aliased to service id `cache` . \n\n```\n// This will be an alias for 'cache.provider.my_redis'\n$cache = $this->container->get('cache');\n```"}}},{"rowIdx":59172,"cells":{"project":{"kind":"string","value":"spaczz"},"source":{"kind":"string","value":"readthedoc"},"language":{"kind":"string","value":"Python"},"content":{"kind":"string","value":"spaczz documentation\n\n[spaczz](#)\n\n---\n\nspaczz: Fuzzy matching and more for spaCy[¶](#spaczz-fuzzy-matching-and-more-for-spacy)\n===\n\nReference[¶](#reference)\n---\n\n* [spaczz.matcher](#module-spaczz.matcher)\n* [spaczz.pipeline](#module-spaczz.pipeline)\n* [spaczz.registry](#module-spaczz.registry)\n* [spaczz.customattrs](#module-spaczz.customattrs)\n* [spaczz.customtypes](#module-spaczz.customtypes)\n* [spaczz.exceptions](#module-spaczz.exceptions)\n\n### spaczz.matcher[¶](#module-spaczz.matcher)\n\nModule for matchers.\n\n*class* spaczz.matcher.FuzzyMatcher(*vocab*, ***defaults*)[¶](#spaczz.matcher.FuzzyMatcher)\nspaCy-like matcher for finding fuzzy phrase matches in Doc objects.\n\nFuzzy matches patterns against the Doc it is called on.\nAccepts labeled patterns in the form of Doc objects with optional,\nper-pattern match settings.\n\nname[¶](#spaczz.matcher.FuzzyMatcher.name)\nClass attribute - the name of the matcher.\n\nType str\n\ndefaults[¶](#spaczz.matcher.FuzzyMatcher.defaults)\nKeyword arguments to be used as default match settings.\nPer-pattern match settings take precedence over defaults.\n\nType dict[str, bool|int|str|Literal[‘default’, ‘min’, ‘max’]]\n\nMatch Settings\n* **ignore_case** (*bool*) – Whether to lower-case text before matching.\nDefault is True.\n* **min_r** (*int*) – Minimum match ratio required.\n* **thresh** (*int*) – If this ratio is exceeded in initial scan,\nand flex > 0, no optimization will be attempted.\nIf flex == 0, thresh has no effect. Default is 100.\n* **fuzzy_func** (*str*) – Key name of fuzzy matching function to use.\nAll rapidfuzz matching functions with default settings are available.\nAdditional fuzzy matching functions can be registered by users.\nIncluded functions are:\n\n\t+ “simple” = ratio\n\t+ “partial” = partial_ratio\n\t+ “token” = token_ratio\n\t+ “token_set” = token_set_ratio\n\t+ “token_sort” = token_sort_ratio\n\t+ “partial_token” = partial_token_ratio\n\t+ “partial_token_set” = partial_token_set_ratio\n\t+ “partial_token_sort” = partial_token_sort_ratio\n\t+ “weighted” = WRatio\n\t+ “quick” = QRatio\n\t+ “partial_alignment” = partial_ratio_alignment\n\t(Requires rapidfuzz>=2.0.3)Default is “simple”.\n* **flex** (*int|Literal[‘default’, ‘min’, ‘max’]*) – Number of tokens to move match boundaries left and right during optimization. Can be an int with a max of len(pattern) and a min of 0, (will warn and change if higher or lower).\n“max”, “min”, or “default” are also valid.\nDefault is “default”: len(pattern) // 2.\n* **min_r1** (*int|None*) – Optional granular control over the minimum match ratio required for selection during the initial scan.\nIf flex == 0, min_r1 will be overwritten by min_r2.\nIf flex > 0, min_r1 must be lower than min_r2 and “low” in general because match boundaries are not flexed initially.\nDefault is None, which will result in min_r1 being set to round(min_r / 1.5).\n* **min_r2** (*int|None*) – Optional granular control over the minimum match ratio required for selection during match optimization.\nNeeds to be higher than min_r1 and “high” in general to ensure only quality matches are returned. Default is None, which will result in min_r2 being set to min_r.\n\n__call__(*doc*)[¶](#spaczz.matcher.FuzzyMatcher.__call__)\nFinds matches in doc given the matchers patterns.\n\nParameters\n**doc** (`Doc`) – The Doc object to match over.\n\nReturn type\n`List`[`Tuple`[`str`, `int`, `int`, `int`, `str`]]\n\nReturns A list of MatchResult tuples,\n(label, start index, end index, match ratio, pattern).\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import FuzzyMatcher\n>>> nlp = spacy.blank(\"en\")\n>>> matcher = FuzzyMatcher(nlp.vocab)\n>>> doc = nlp(\"<NAME> was the director of Alien.\")\n>>> matcher.add(\"NAME\", [nlp.make_doc(\"<NAME>\")])\n>>> matcher(doc)\n[('NAME', 0, 2, 96, '<NAME>')]\n```\n__contains__(*label*)[¶](#spaczz.matcher.FuzzyMatcher.__contains__)\nWhether the matcher contains patterns for a label.\n\nReturn type\n`bool`\n\n__len__()[¶](#spaczz.matcher.FuzzyMatcher.__len__)\nThe number of labels added to the matcher.\n\nReturn type\n`int`\n\n__reduce__()[¶](#spaczz.matcher.FuzzyMatcher.__reduce__)\nInterface for pickling the matcher.\n\nReturn type\n`Tuple`[`Any`, `Any`]\n\nadd(*label*, *patterns*, *kwargs=None*, *on_match=None*)[¶](#spaczz.matcher.FuzzyMatcher.add)\nAdd a rule to the matcher, consisting of a label and one or more patterns.\n\nPatterns must be a list of Doc objects and if kwargs is not None,\nkwargs must be a list of dicts.\n\nParameters\n* **label** (`str`) – Name of the rule added to the matcher.\n* **patterns** (`List`[`Doc`]) – Doc objects that will be matched against the Doc object the matcher is called on.\n* **kwargs** (`Optional`[`List`[`Dict`[`str`, `Any`]]]) – Optional settings to modify the matching behavior.\nIf supplying kwargs, one per pattern should be included.\nEmpty dicts will use the matcher instances default settings.\nDefault is None.\n* **on_match** (`Optional`[`Callable`[[`TypeVar`(`PMT`, bound= `PhraseMatcher`), `Doc`, `int`, `List`[`Tuple`[`str`, `int`, `int`, `int`, `str`]]], `None`]]) – Optional callback function to modify the Doc object the matcher is called on after matching.\nDefault is None.\n\nWarning\n\nKwargsWarning:* If there are more patterns than kwargs default matching settings will be used for extra patterns.\n* If there are more kwargs dicts than patterns,\nthe extra kwargs will be ignored.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import FuzzyMatcher\n>>> nlp = spacy.blank(\"en\")\n>>> matcher = FuzzyMatcher(nlp.vocab)\n>>> matcher.add(\"SOUND\", [nlp.make_doc(\"mooo\")])\n>>> \"SOUND\" in matcher True\n```\nReturn type\n`None`\n\n*property* labels*: Tuple[str, ...]*[¶](#spaczz.matcher.FuzzyMatcher.labels)\nAll labels present in the matcher.\n\nReturn type\n`Tuple`[`str`, `...`]\n\nReturns The unique labels as a tuple of strings.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import FuzzyMatcher\n>>> nlp = spacy.blank(\"en\")\n>>> matcher = FuzzyMatcher(nlp.vocab)\n>>> matcher.add(\"AUTHOR\", [nlp.make_doc(\"Kerouac\")])\n>>> matcher.labels\n('AUTHOR',)\n```\n*property* patterns*: List[Dict[str, Any]]*[¶](#spaczz.matcher.FuzzyMatcher.patterns)\nGet all patterns and match settings that were added to the matcher.\n\nReturn type\n`List`[`Dict`[`str`, `Any`]]\n\nReturns The patterns and their respective match settings as a list of dicts.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import FuzzyMatcher\n>>> nlp = spacy.blank(\"en\")\n>>> matcher = FuzzyMatcher(nlp.vocab)\n>>> matcher.add(\"AUTHOR\", [nlp.make_doc(\"Kerouac\")],\n [{\"ignore_case\": False}])\n>>> matcher.patterns == [\n {\n \"label\": \"AUTHOR\",\n \"pattern\": \"Kerouac\",\n \"type\": \"fuzzy\",\n \"kwargs\": {\"ignore_case\": False}\n },\n ]\nTrue\n```\nremove(*label*)[¶](#spaczz.matcher.FuzzyMatcher.remove)\nRemove a label and its respective patterns from the matcher.\n\nParameters\n**label** (`str`) – Name of the rule added to the matcher.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import FuzzyMatcher\n>>> nlp = spacy.blank(\"en\")\n>>> matcher = FuzzyMatcher(nlp.vocab)\n>>> matcher.add(\"SOUND\", [nlp.make_doc(\"mooo\")])\n>>> matcher.remove(\"SOUND\")\n>>> \"SOUND\" in matcher False\n```\nReturn type\n`None`\n\n*property* type*: Literal['fuzzy', 'regex', 'token', 'similarity', 'phrase']*[¶](#spaczz.matcher.FuzzyMatcher.type)\nGetter for the matchers SpaczzType.\n\nReturn type\n`Literal`[‘fuzzy’, ‘regex’, ‘token’, ‘similarity’, ‘phrase’]\n\n*property* vocab*: spacy.vocab.Vocab*[¶](#spaczz.matcher.FuzzyMatcher.vocab)\nGetter for the matchers Vocab.\n\nReturn type\n`Vocab`\n\n*class* spaczz.matcher.RegexMatcher(*vocab*, ***defaults*)[¶](#spaczz.matcher.RegexMatcher)\nspaCy-like matcher for finding regex phrase matches in Doc objects.\n\nRegex matches patterns against the Doc it is called on.\nAccepts labeled patterns in the form of strings with optional,\nper-pattern match settings.\n\nTo utilize regex flags, use inline flags.\n\nname[¶](#spaczz.matcher.RegexMatcher.name)\nClass attribute - the name of the matcher.\n\nType str\n\ndefaults[¶](#spaczz.matcher.RegexMatcher.defaults)\nKeyword arguments to be used as default match settings.\nPer-pattern match settings take precedence over defaults.\n\nType dict[str, bool|int|str]\n\nMatch Settings\n* **ignore_case** (*bool*) – Whether to lower-case text before matching.\nDefault is True.\n* **min_r** (*int*) – Minimum match ratio required.\n* **fuzzy_weights** (*str*) – Name of weighting method for regex insertion, deletion, and substituion counts. Additional weighting methods can be registered by users. Included weighting methods are:\n\n\t+ “indel” = (1, 1, 2)\n\t+ “lev” = (1, 1, 1)Default is “indel”.\n* **partial** – (bool): Whether partial matches should be extended to Token or Span boundaries in doc or not.\nFor example, the regex only matches part of a Token or Span in doc. Default is True.\n* **predef** (*string*) – Whether the regex string should be interpreted as a key to a predefined regex pattern or not. Additional predefined regex patterns can be registered by users. The included predefined regex patterns are:\n\n\t+ “dates”\n\t+ “times”\n\t+ “phones”\n\t+ “phones_with_exts”\n\t+ “links”\n\t+ “emails”\n\t+ “ips”\n\t+ “ipv6s”\n\t+ “prices”\n\t+ “hex_colors”\n\t+ “credit_cards”\n\t+ “btc_addresses”\n\t+ “street_addresses”\n\t+ “zip_codes”\n\t+ “po_boxes”\n\t+ “ssn_numbers”Default is False.\n\n__call__(*doc*)[¶](#spaczz.matcher.RegexMatcher.__call__)\nFinds matches in doc given the matchers patterns.\n\nParameters\n**doc** (`Doc`) – The Doc object to match over.\n\nReturn type\n`List`[`Tuple`[`str`, `int`, `int`, `int`, `str`]]\n\nReturns A list of MatchResult tuples,\n(label, start index, end index, match ratio, pattern).\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import RegexMatcher\n>>> nlp = spacy.blank(\"en\")\n>>> matcher = RegexMatcher(nlp.vocab)\n>>> doc = nlp.make_doc(\"I live in the united states, or the US\")\n>>> matcher.add(\"GPE\", [\"[Uu](nited|\\.?) ?[Ss](tates|\\.?)\"])\n>>> matcher(doc)[0]\n('GPE', 4, 6, 100, '[Uu](nited|\\\\.?) ?[Ss](tates|\\\\.?)')\n```\n__contains__(*label*)[¶](#spaczz.matcher.RegexMatcher.__contains__)\nWhether the matcher contains patterns for a label.\n\nReturn type\n`bool`\n\n__len__()[¶](#spaczz.matcher.RegexMatcher.__len__)\nThe number of labels added to the matcher.\n\nReturn type\n`int`\n\n__reduce__()[¶](#spaczz.matcher.RegexMatcher.__reduce__)\nInterface for pickling the matcher.\n\nReturn type\n`Tuple`[`Any`, `Any`]\n\nadd(*label*, *patterns*, *kwargs=None*, *on_match=None*)[¶](#spaczz.matcher.RegexMatcher.add)\nAdd a rule to the matcher, consisting of a label and one or more patterns.\n\nPatterns must be a list of Doc objects and if kwargs is not None,\nkwargs must be a list of dicts.\n\nParameters\n* **label** (`str`) – Name of the rule added to the matcher.\n* **patterns** (`List`[`str`]) – Doc objects that will be matched against the Doc object the matcher is called on.\n* **kwargs** (`Optional`[`List`[`Dict`[`str`, `Any`]]]) – Optional settings to modify the matching behavior.\nIf supplying kwargs, one per pattern should be included.\nEmpty dicts will use the matcher instances default settings.\nDefault is None.\n* **on_match** (`Optional`[`Callable`[[[`RegexMatcher`](#spaczz.matcher.RegexMatcher), `Doc`, `int`, `List`[`Tuple`[`str`, `int`, `int`, `int`, `str`]]], `None`]]) – Optional callback function to modify the Doc object the matcher is called on after matching.\nDefault is None.\n\nRaises\n* **TypeError** – If patterns is not a list of strings.\n* **TypeError** – If kwargs is not a list of dictionaries.\n\nWarning\n\nKwargsWarning:* If there are more patterns than kwargs default matching settings will be used for extra patterns.\n* If there are more kwargs dicts than patterns,\nthe extra kwargs will be ignored.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import RegexMatcher\n>>> nlp = spacy.blank(\"en\")\n>>> matcher = RegexMatcher(nlp.vocab)\n>>> matcher.add(\"GPE\", [\"[Uu](nited|\\.?) ?[Ss](tates|\\.?)\"])\n>>> \"GPE\" in matcher True\n```\nReturn type\n`None`\n\n*property* labels*: Tuple[str, ...]*[¶](#spaczz.matcher.RegexMatcher.labels)\nAll labels present in the matcher.\n\nReturn type\n`Tuple`[`str`, `...`]\n\nReturns The unique labels as a tuple of strings.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import RegexMatcher\n>>> nlp = spacy.blank(\"en\")\n>>> matcher = RegexMatcher(nlp.vocab)\n>>> matcher.add(\"ZIP\", [\"zip_codes\"], [{\"predef\": True}])\n>>> matcher.labels\n('ZIP',)\n```\n*property* patterns*: List[Dict[str, Any]]*[¶](#spaczz.matcher.RegexMatcher.patterns)\nGet all patterns and match settings that were added to the matcher.\n\nReturn type\n`List`[`Dict`[`str`, `Any`]]\n\nReturns The patterns and their respective match settings as a list of dicts.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import RegexMatcher\n>>> nlp = spacy.blank(\"en\")\n>>> matcher = RegexMatcher(nlp.vocab)\n>>> matcher.add(\"ZIP\", [\"zip_codes\"], [{\"predef\": True}])\n>>> matcher.patterns == [\n {\n \"label\": \"ZIP\",\n \"pattern\": \"zip_codes\",\n \"type\": \"regex\",\n \"kwargs\": {\"predef\": True},\n }\n ]\nTrue\n```\nremove(*label*)[¶](#spaczz.matcher.RegexMatcher.remove)\nRemove a label and its respective patterns from the matcher.\n\nParameters\n**label** (`str`) – Name of the rule added to the matcher.\n\nRaises\n**ValueError** – If label does not exist in the matcher.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import RegexMatcher\n>>> nlp = spacy.blank(\"en\")\n>>> matcher = RegexMatcher(nlp.vocab)\n>>> matcher.add(\"GPE\", [\"[Uu](nited|\\.?) ?[Ss](tates|\\.?)\"])\n>>> matcher.remove(\"GPE\")\n>>> \"GPE\" in matcher False\n```\nReturn type\n`None`\n\n*property* type*: Literal['fuzzy', 'regex', 'token', 'similarity', 'phrase']*[¶](#spaczz.matcher.RegexMatcher.type)\nGetter for the matchers SpaczzType.\n\nReturn type\n`Literal`[‘fuzzy’, ‘regex’, ‘token’, ‘similarity’, ‘phrase’]\n\n*property* vocab*: spacy.vocab.Vocab*[¶](#spaczz.matcher.RegexMatcher.vocab)\nGetter for the matchers Vocab.\n\nReturn type\n`Vocab`\n\n*class* spaczz.matcher.SimilarityMatcher(*vocab*, ***defaults*)[¶](#spaczz.matcher.SimilarityMatcher)\nspaCy-like matcher for finding phrase similarity matches in Doc objects.\n\nSimilarity matches patterns against the Doc it is called on.\nAccepts labeled patterns in the form of Doc objects with optional,\nper-pattern match settings.\n\nSimilarity matching uses spaCy word vectors if available,\ntherefore spaCy vocabs without word vectors may not produce useful results. The spaCy medium and large English models provide word vectors that will work for this purpose.\n\nSearching over/with Doc objects that do not have vectors will always return a similarity score of 0.\n\nWarnings from spaCy about the above two scenarios are suppressed for convenience. However, spaczz will still warn about the former.\n\nname[¶](#spaczz.matcher.SimilarityMatcher.name)\nClass attribute - the name of the matcher.\n\nType str\n\ndefaults[¶](#spaczz.matcher.SimilarityMatcher.defaults)\nKeyword arguments to be used as default match settings.\nPer-pattern match settings take precedence over defaults.\n\nType dict[str, bool|int|str|Literal[‘default’, ‘min’, ‘max’]]\n\nMatch Settings\n* **ignore_case** (*bool*) – Whether to lower-case text before fuzzy matching.\nDefault is True.\n* **min_r** (*int*) – Minimum match ratio required.\n* **thresh** (*int*) – If this ratio is exceeded in initial scan,\nand flex > 0, no optimization will be attempted.\nIf flex == 0, thresh has no effect. Default is 100.\n* **flex** (*int|Literal[‘default’, ‘min’, ‘max’]*) – Number of tokens to move match boundaries left and right during optimization. Can be an int with a max of len(pattern) and a min of 0, (will warn and change if higher or lower).\n“max”, “min”, or “default” are also valid.\nDefault is “default”: len(pattern) // 2.\n* **min_r1** (*int|None*) – Optional granular control over the minimum match ratio required for selection during the initial scan.\nIf flex == 0, min_r1 will be overwritten by min_r2.\nIf flex > 0, min_r1 must be lower than min_r2 and “low” in general because match boundaries are not flexed initially.\nDefault is None, which will result in min_r1 being set to round(min_r / 1.5).\n* **min_r2** (*int|None*) – Optional granular control over the minimum match ratio required for selection during match optimization.\nNeeds to be higher than min_r1 and “high” in general to ensure only quality matches are returned. Default is None, which will result in min_r2 being set to min_r.\n\nWarning\n\nMissingVectorsWarning:If vocab does not contain any word vectors.\n\n__call__(*doc*)[¶](#spaczz.matcher.SimilarityMatcher.__call__)\nFinds matches in doc given the matchers patterns.\n\nParameters\n**doc** (`Doc`) – The Doc object to match over.\n\nReturn type\n`List`[`Tuple`[`str`, `int`, `int`, `int`, `str`]]\n\nReturns A list of MatchResult tuples,\n(label, start index, end index, match ratio, pattern).\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import SimilarityMatcher\n>>> nlp = spacy.load(\"en_core_web_md\")\n>>> matcher = SimilarityMatcher(nlp.vocab)\n>>> doc = nlp(\"I like apples.\")\n>>> matcher.add(\"FRUIT\", [nlp(\"fruit\")], [{'min_r': 60}])\n>>> matcher(doc)\n[('FRUIT', 2, 3, 70, 'fruit')]\n```\n__contains__(*label*)[¶](#spaczz.matcher.SimilarityMatcher.__contains__)\nWhether the matcher contains patterns for a label.\n\nReturn type\n`bool`\n\n__len__()[¶](#spaczz.matcher.SimilarityMatcher.__len__)\nThe number of labels added to the matcher.\n\nReturn type\n`int`\n\n__reduce__()[¶](#spaczz.matcher.SimilarityMatcher.__reduce__)\nInterface for pickling the matcher.\n\nReturn type\n`Tuple`[`Any`, `Any`]\n\nadd(*label*, *patterns*, *kwargs=None*, *on_match=None*)[¶](#spaczz.matcher.SimilarityMatcher.add)\nAdd a rule to the matcher, consisting of a label and one or more patterns.\n\nPatterns must be a list of Doc objects and if kwargs is not None,\nkwargs must be a list of dicts.\n\nParameters\n* **label** (`str`) – Name of the rule added to the matcher.\n* **patterns** (`List`[`Doc`]) – Doc objects that will be matched against the Doc object the matcher is called on.\n* **kwargs** (`Optional`[`List`[`Dict`[`str`, `Any`]]]) – Optional settings to modify the matching behavior.\nIf supplying kwargs, one per pattern should be included.\nEmpty dicts will use the matcher instances default settings.\nDefault is None.\n* **on_match** (`Optional`[`Callable`[[`TypeVar`(`PMT`, bound= `PhraseMatcher`), `Doc`, `int`, `List`[`Tuple`[`str`, `int`, `int`, `int`, `str`]]], `None`]]) – Optional callback function to modify the Doc object the matcher is called on after matching.\nDefault is None.\n\nWarning\n\nKwargsWarning:* If there are more patterns than kwargs default matching settings will be used for extra patterns.\n* If there are more kwargs dicts than patterns,\nthe extra kwargs will be ignored.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import SimilarityMatcher\n>>> nlp = spacy.load(\"en_core_web_md\")\n>>> matcher = SimilarityMatcher(nlp.vocab)\n>>> matcher.add(\"SOUND\", [nlp(\"mooo\")])\n>>> \"SOUND\" in matcher True\n```\nReturn type\n`None`\n\n*property* labels*: Tuple[str, ...]*[¶](#spaczz.matcher.SimilarityMatcher.labels)\nAll labels present in the matcher.\n\nReturn type\n`Tuple`[`str`, `...`]\n\nReturns The unique labels as a tuple of strings.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import SimilarityMatcher\n>>> nlp = spacy.load(\"en_core_web_md\")\n>>> matcher = SimilarityMatcher(nlp.vocab)\n>>> matcher.add(\"AUTHOR\", [nlp(\"Kerouac\")])\n>>> matcher.labels\n('AUTHOR',)\n```\n*property* patterns*: List[Dict[str, Any]]*[¶](#spaczz.matcher.SimilarityMatcher.patterns)\nGet all patterns and match settings that were added to the matcher.\n\nReturn type\n`List`[`Dict`[`str`, `Any`]]\n\nReturns The patterns and their respective match settings as a list of dicts.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import SimilarityMatcher\n>>> nlp = spacy.load(\"en_core_web_md\")\n>>> matcher = SimilarityMatcher(nlp.vocab)\n>>> matcher.add(\"AUTHOR\", [nlp(\"Kerouac\")],\n [{\"ignore_case\": False}])\n>>> matcher.patterns == [\n {\n \"label\": \"AUTHOR\",\n \"pattern\": \"Kerouac\",\n \"type\": \"similarity\",\n \"kwargs\": {\"ignore_case\": False}\n },\n ]\nTrue\n```\nremove(*label*)[¶](#spaczz.matcher.SimilarityMatcher.remove)\nRemove a label and its respective patterns from the matcher.\n\nParameters\n**label** (`str`) – Name of the rule added to the matcher.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import SimilarityMatcher\n>>> nlp = spacy.load(\"en_core_web_md\")\n>>> matcher = SimilarityMatcher(nlp.vocab)\n>>> matcher.add(\"SOUND\", [nlp(\"mooo\")])\n>>> matcher.remove(\"SOUND\")\n>>> \"SOUND\" in matcher False\n```\nReturn type\n`None`\n\n*property* type*: Literal['fuzzy', 'regex', 'token', 'similarity', 'phrase']*[¶](#spaczz.matcher.SimilarityMatcher.type)\nGetter for the matchers SpaczzType.\n\nReturn type\n`Literal`[‘fuzzy’, ‘regex’, ‘token’, ‘similarity’, ‘phrase’]\n\n*property* vocab*: spacy.vocab.Vocab*[¶](#spaczz.matcher.SimilarityMatcher.vocab)\nGetter for the matchers Vocab.\n\nReturn type\n`Vocab`\n\n*class* spaczz.matcher.TokenMatcher(*vocab*, ***defaults*)[¶](#spaczz.matcher.TokenMatcher)\nspaCy-like matcher for finding fuzzy token matches in Doc objects.\n\nFuzzy matches added patterns against the Doc object it is called on.\nAccepts labeled patterns in the form of lists of dictionaries where each list describes an individual pattern and each dictionary describes an individual token.\n\nUses extended spaCy token matching patterns.\n“FUZZY” and “FREGEX” are the two additional spaCy token pattern options.\n\nFor example:\n```\n[\n {\"TEXT\": {\"FREGEX\": \"(database){e<=1}\"}},\n {\"LOWER\": {\"FUZZY\": \"access\", \"MIN_R\": 85, \"FUZZY_FUNC\": \"partial\"}},\n]\n```\nMake sure to use uppercase dictionary keys in patterns.\n\nname[¶](#spaczz.matcher.TokenMatcher.name)\nClass attribute - the name of the matcher.\n\nType str\n\ndefaults[¶](#spaczz.matcher.TokenMatcher.defaults)\nKeyword arguments to be used as default match settings.\nPer-pattern match settings take precedence over defaults.\n\nType dict[str, bool|int|str]\n\nMatch Settings\n* **ignore_case** (*bool*) – Whether to lower-case text before matching.\nCan only be set at the pattern level. For “FUZZY” and “FREGEX” patterns.\nDefault is True.\n* **min_r** (*int*) – Minimum match ratio required. For “FUZZY” and “FREGEX” patterns.\n* **fuzzy_func** (*str*) – Key name of fuzzy matching function to use.\nCan only be set at the pattern level. For “FUZZY” patterns only.\nAll rapidfuzz matching functions with default settings are available,\nhowever any token-based functions provide no utility at the individual token level. Additional fuzzy matching functions can be registered by users.\nIncluded, and useful, functions are:\n\n\t+ “simple” = ratio\n\t+ “partial” = partial_ratio\n\t+ “quick” = QRatio\n\t+ “partial_alignment” = partial_ratio_alignment(Requires rapidfuzz>=2.0.3)Default is “simple”.\n* **fuzzy_weights** – Name of weighting method for regex insertion, deletion, and substituion counts. Can only be set at the pattern level. For “FREGEX”\npatterns only. Included weighting methods are:\n\n\t+ “indel” = (1, 1, 2)\n\t+ “lev” = (1, 1, 1)Default is “indel”.\n* **predef** – Whether regex should be interpreted as a key to a predefined regex pattern or not. Can only be set at the pattern level.\nFor “FREGEX” patterns only. Default is False.\n\n__call__(*doc*)[¶](#spaczz.matcher.TokenMatcher.__call__)\nFinds matches in doc given the matchers patterns.\n\nParameters\n**doc** (`Doc`) – The Doc object to match over.\n\nReturn type\n`List`[`Tuple`[`str`, `int`, `int`, `int`, `str`]]\n\nReturns A list of MatchResult tuples,\n(label, start index, end index, match ratio, pattern).\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import TokenMatcher\n>>> nlp = spacy.blank(\"en\")\n>>> matcher = TokenMatcher(nlp.vocab)\n>>> doc = nlp(\"<NAME> was the director of Alien.\")\n>>> matcher.add(\"NAME\", [\n [{\"TEXT\": {\"FUZZY\": \"Ridley\"}},\n {\"TEXT\": {\"FUZZY\": \"Scott\"}}]\n ])\n>>> matcher(doc)[0][:4]\n('NAME', 0, 2, 90)\n```\n__contains__(*label*)[¶](#spaczz.matcher.TokenMatcher.__contains__)\nWhether the matcher contains patterns for a label.\n\nReturn type\n`bool`\n\n__len__()[¶](#spaczz.matcher.TokenMatcher.__len__)\nThe number of labels added to the matcher.\n\nReturn type\n`int`\n\n__reduce__()[¶](#spaczz.matcher.TokenMatcher.__reduce__)\nInterface for pickling the matcher.\n\nReturn type\n`Tuple`[`Any`, `Any`]\n\nadd(*label*, *patterns*, *on_match=None*)[¶](#spaczz.matcher.TokenMatcher.add)\nAdd a rule to the matcher, consisting of a label and one or more patterns.\n\nPatterns must be a list of lists of dicts where each list of dicts represent an individual pattern and each dictionary represents an individual token.\n\nUses extended spaCy token matching patterns.\n“FUZZY” and “FREGEX” are the two additional spaCy token pattern options.\n\nFor example:\n```\n[\n {\"TEXT\": {\"FREGEX\": \"(database){e<=1}\"}},\n {\"LOWER\": {\"FUZZY\": \"access\", \"MIN_R\": 85, \"FUZZY_FUNC\": \"partial\"}},\n]\n```\nMake sure to use uppercase dictionary keys in patterns.\n\nParameters\n* **label** (`str`) – Name of the rule added to the matcher.\n* **patterns** (`List`[`List`[`Dict`[`str`, `Any`]]]) – List of lists of dicts that will be matched against the Doc object the matcher is called on.\n* **on_match** (`Optional`[`Callable`[[[`TokenMatcher`](#spaczz.matcher.TokenMatcher), `Doc`, `int`, `List`[`Tuple`[`str`, `int`, `int`, `int`, `str`]]], `None`]]) – Optional callback function to modify the Doc object the matcher is called on after matching.\nDefault is None.\n\nRaises\n* **TypeError** – If patterns is not a list of Doc objects.\n* **ValueError** – Patterns cannot have zero tokens.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import TokenMatcher\n>>> nlp = spacy.blank(\"en\")\n>>> matcher = TokenMatcher(nlp.vocab)\n>>> matcher.add(\"AUTHOR\", [[{\"TEXT\": {\"FUZZY\": \"Kerouac\"}}]])\n>>> \"AUTHOR\" in matcher True\n```\nReturn type\n`None`\n\n*property* labels*: Tuple[str, ...]*[¶](#spaczz.matcher.TokenMatcher.labels)\nAll labels present in the matcher.\n\nReturn type\n`Tuple`[`str`, `...`]\n\nReturns The unique labels as a tuple of strings.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import TokenMatcher\n>>> nlp = spacy.blank(\"en\")\n>>> matcher = TokenMatcher(nlp.vocab)\n>>> matcher.add(\"AUTHOR\", [[{\"TEXT\": {\"FUZZY\": \"Kerouac\"}}]])\n>>> matcher.labels\n('AUTHOR',)\n```\n*property* patterns*: List[Dict[str, Any]]*[¶](#spaczz.matcher.TokenMatcher.patterns)\nGet all patterns and match settings that were added to the matcher.\n\nReturn type\n`List`[`Dict`[`str`, `Any`]]\n\nReturns The patterns and their respective match settings as a list of dicts.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import TokenMatcher\n>>> nlp = spacy.blank(\"en\")\n>>> matcher = TokenMatcher(nlp.vocab)\n>>> matcher.add(\"AUTHOR\", [[{\"TEXT\": {\"FUZZY\": \"Kerouac\"}}]])\n>>> matcher.patterns == [\n {\n \"label\": \"AUTHOR\",\n \"pattern\": [{\"TEXT\": {\"FUZZY\": \"Kerouac\"}}],\n \"type\": \"token\",\n },\n ]\nTrue\n```\nremove(*label*)[¶](#spaczz.matcher.TokenMatcher.remove)\nRemove a label and its respective patterns from the matcher.\n\nParameters\n**label** (`str`) – Name of the rule added to the matcher.\n\nRaises\n**ValueError** – If label does not exist in the matcher.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.matcher import TokenMatcher\n>>> nlp = spacy.blank(\"en\")\n>>> matcher = TokenMatcher(nlp.vocab)\n>>> matcher.add(\"AUTHOR\", [[{\"TEXT\": {\"FUZZY\": \"Kerouac\"}}]])\n>>> matcher.remove(\"AUTHOR\")\n>>> \"AUTHOR\" in matcher False\n```\nReturn type\n`None`\n\n*property* type*: Literal['fuzzy', 'regex', 'token', 'similarity', 'phrase']*[¶](#spaczz.matcher.TokenMatcher.type)\nGetter for the matchers SpaczzType.\n\nReturn type\n`Literal`[‘fuzzy’, ‘regex’, ‘token’, ‘similarity’, ‘phrase’]\n\n*property* vocab*: spacy.vocab.Vocab*[¶](#spaczz.matcher.TokenMatcher.vocab)\nGetter for the matchers Vocab.\n\nReturn type\n`Vocab`\n\n### spaczz.pipeline[¶](#module-spaczz.pipeline)\n\nModule for pipeline components.\n\n*class* spaczz.pipeline.SpaczzRuler(*nlp*, *name='spaczz_ruler'*, ***, *overwrite_ents=False*, *ent_id_sep='||'*, *fuzzy_defaults={}*, *regex_defaults={}*, *token_defaults={}*, *patterns=None*, *scorer=<function spaczz_ruler_scorer>*)[¶](#spaczz.pipeline.SpaczzRuler)\nThe SpaczzRuler adds fuzzy matches to spaCy Doc.ents.\n\nIt can be combined with other spaCy NER components like the statistical EntityRecognizer, and/or the EntityRuler it is inspired by, to boost accuracy.\nAfter initialization, the component is typically added to the pipeline using nlp.add_pipe.\n\nnlp[¶](#spaczz.pipeline.SpaczzRuler.nlp)\nThe shared Language object that passes its Vocab to the matchers (not currently used by spaczz matchers) and processes fuzzy patterns.\n\nType Language\n\nname[¶](#spaczz.pipeline.SpaczzRuler.name)\nInstance name of the current pipeline component. Typically passed in automatically from the factory when the component is added. Used to disable the current entity ruler while creating phrase patterns with the nlp object.\n\nType str\n\noverwrite_ents[¶](#spaczz.pipeline.SpaczzRuler.overwrite_ents)\nIf existing entities are present, e.g. entities added by the model, overwrite them by matches if necessary.\n\nType bool\n\nent_id_sep[¶](#spaczz.pipeline.SpaczzRuler.ent_id_sep)\nSeparator used internally for entity IDs.\n\nType str\n\nscorer[¶](#spaczz.pipeline.SpaczzRuler.scorer)\nThe scoring method for the ruler.\n\nType Optional[Callable]\n\nfuzzy_matcher[¶](#spaczz.pipeline.SpaczzRuler.fuzzy_matcher)\nThe FuzzyMatcher instance the spaczz ruler will use for fuzzy phrase matching.\n\nType\n[FuzzyMatcher](index.html#spaczz.matcher.FuzzyMatcher)\n\nregex_matcher[¶](#spaczz.pipeline.SpaczzRuler.regex_matcher)\nThe RegexMatcher instance the spaczz ruler will use for regex phrase matching.\n\nType\n[RegexMatcher](index.html#spaczz.matcher.RegexMatcher)\n\ntoken_matcher[¶](#spaczz.pipeline.SpaczzRuler.token_matcher)\nThe TokenMatcher instance the spaczz ruler will use for fuzzy token matching.\n\nType\n[TokenMatcher](index.html#spaczz.matcher.TokenMatcher)\n\ndefaults[¶](#spaczz.pipeline.SpaczzRuler.defaults)\nDefault match settings for their respective matchers.\n\nType Dict[str, Any]\n\n__call__(*doc*)[¶](#spaczz.pipeline.SpaczzRuler.__call__)\nFind matches in document and add them as entities.\n\nParameters\n**doc** (`Doc`) – The Doc object in the pipeline.\n\nReturn type\n`Doc`\n\nReturns The Doc with added entities, if available.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.pipeline import SpaczzRuler\n>>> nlp = spacy.blank(\"en\")\n>>> ruler = SpaczzRuler(nlp)\n>>> doc = nlp.make_doc(\"My name is <NAME>\")\n>>> ruler.add_patterns([{\"label\": \"NAME\", \"pattern\": \"<NAME>\",\n \"type\": \"fuzzy\", \"kwargs\": {\"fuzzy_func\": \"token_sort\"}}])\n>>> doc = ruler(doc)\n>>> \"<NAME>\" in [ent.text for ent in doc.ents]\nTrue\n```\n__contains__(*label*)[¶](#spaczz.pipeline.SpaczzRuler.__contains__)\nWhether a label is present in the patterns.\n\nReturn type\n`bool`\n\n__len__()[¶](#spaczz.pipeline.SpaczzRuler.__len__)\nThe number of all patterns added to the ruler.\n\nReturn type\n`int`\n\nadd_patterns(*patterns*)[¶](#spaczz.pipeline.SpaczzRuler.add_patterns)\nAdd patterns to the ruler.\n\nA pattern must be a spaczz pattern:\n{label (str), pattern (str or list), type (str),\noptional kwargs (dict[str, Any]), and optional id (str)}.\n\nFor example, a fuzzy phrase pattern:\n```\n{\n 'label': 'ORG',\n 'pattern': 'Apple',\n 'kwargs': {'min_r2': 90},\n 'type': 'fuzzy',\n}\n```\nOr, a token pattern:\n```\n{\n 'label': 'ORG',\n 'pattern': [{'TEXT': {'FUZZY': 'Apple'}}],\n 'type': 'token',\n}\n```\nTo utilize regex flags, use inline flags.\n\nParameters\n**patterns** (`List`[`Dict`[`str`, `Union`[`str`, `Dict`[`str`, `Any`], `List`[`Dict`[`str`, `Any`]]]]]) – The spaczz patterns to add.\n\nRaises\n* **TypeError** – If patterns is not a list of dicts.\n* **ValueError** – If one or more patterns do not conform\n the spaczz pattern structure.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.pipeline import SpaczzRuler\n>>> nlp = spacy.blank(\"en\")\n>>> ruler = SpaczzRuler(nlp)\n>>> ruler.add_patterns([{\"label\": \"AUTHOR\", \"pattern\": \"Kerouac\",\n \"type\": \"fuzzy\"}])\n>>> \"AUTHOR\" in ruler.labels True\n```\nReturn type\n`None`\n\nclear()[¶](#spaczz.pipeline.SpaczzRuler.clear)\nReset all patterns.\n\nReturn type\n`None`\n\n*property* ent_ids*: Tuple[Optional[str], ...]*[¶](#spaczz.pipeline.SpaczzRuler.ent_ids)\nAll entity ids present in the match patterns id properties.\n\nReturn type\n`Tuple`[`Optional`[`str`], `...`]\n\nReturns The unique string entity ids as a tuple.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.pipeline import SpaczzRuler\n>>> nlp = spacy.blank(\"en\")\n>>> ruler = SpaczzRuler(nlp)\n>>> ruler.add_patterns([{\"label\": \"AUTHOR\", \"pattern\": \"Kerouac\",\n \"type\": \"fuzzy\", \"id\": \"BEAT\"}])\n>>> ruler.ent_ids\n('BEAT',)\n```\nfrom_bytes(*patterns_bytes*, ***, *exclude=[]*)[¶](#spaczz.pipeline.SpaczzRuler.from_bytes)\nLoad the spaczz ruler from a bytestring.\n\nParameters\n* **patterns_bytes** (`bytes`) – The bytestring to load.\n* **exclude** (`Iterable`[`str`]) – For spaCy consistency.\n\nReturn type\n[`SpaczzRuler`](#spaczz.pipeline.SpaczzRuler)\n\nReturns The loaded spaczz ruler.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.pipeline import SpaczzRuler\n>>> nlp = spacy.blank(\"en\")\n>>> ruler = SpaczzRuler(nlp)\n>>> ruler.add_patterns([{\"label\": \"AUTHOR\", \"pattern\": \"Kerouac\",\n \"type\": \"fuzzy\"}])\n>>> ruler_bytes = ruler.to_bytes()\n>>> new_ruler = SpaczzRuler(nlp)\n>>> new_ruler = new_ruler.from_bytes(ruler_bytes)\n>>> \"AUTHOR\" in new_ruler True\n```\nfrom_disk(*path*, ***, *exclude=[]*)[¶](#spaczz.pipeline.SpaczzRuler.from_disk)\nLoad the spaczz ruler from a file.\n\nExpects a file containing newline-delimited JSON (JSONL)\nwith one entry per line.\n\nParameters\n* **path** (`Union`[`str`, `Path`]) – The JSONL file to load.\n* **exclude** (`Iterable`[`str`]) – For spaCy consistency.\n\nReturn type\n[`SpaczzRuler`](#spaczz.pipeline.SpaczzRuler)\n\nReturns The loaded spaczz ruler.\n\nRaises\n**ValueError** – If path does not exist or cannot be accessed.\n\nExample\n```\n>>> import os\n>>> import tempfile\n>>> import spacy\n>>> from spaczz.pipeline import SpaczzRuler\n>>> nlp = spacy.blank(\"en\")\n>>> ruler = SpaczzRuler(nlp)\n>>> ruler.add_patterns([{\"label\": \"AUTHOR\", \"pattern\": \"Kerouac\",\n \"type\": \"fuzzy\"}])\n>>> with tempfile.TemporaryDirectory() as tmpdir:\n>>> ruler.to_disk(f\"{tmpdir}/ruler\")\n>>> new_ruler = SpaczzRuler(nlp)\n>>> new_ruler = new_ruler.from_disk(f\"{tmpdir}/ruler\")\n>>> \"AUTHOR\" in new_ruler True\n```\ninitialize(*get_examples*, ***, *nlp=None*, *patterns=None*)[¶](#spaczz.pipeline.SpaczzRuler.initialize)\nInitialize the pipe for training.\n\nParameters\n* **get_examples** (`Callable`[[], `Iterable`[`Example`]]) – Function that returns a representative sample of gold-standard Example objects.\n* **nlp** (`Optional`[`Language`]) – The current nlp object the component is part of.\n* **patterns** (`Optional`[`Sequence`[`Dict`[`str`, `Union`[`str`, `Dict`[`str`, `Any`], `List`[`Dict`[`str`, `Any`]]]]]]) – The list of patterns.\n\nReturn type\n`None`\n\n*property* labels*: Tuple[str, ...]*[¶](#spaczz.pipeline.SpaczzRuler.labels)\nAll labels present in the ruler.\n\nReturn type\n`Tuple`[`str`, `...`]\n\nReturns The unique string labels as a tuple.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.pipeline import SpaczzRuler\n>>> nlp = spacy.blank(\"en\")\n>>> ruler = SpaczzRuler(nlp)\n>>> ruler.add_patterns([{\"label\": \"AUTHOR\", \"pattern\": \"Kerouac\",\n \"type\": \"fuzzy\"}])\n>>> ruler.labels\n('AUTHOR',)\n```\nmatch(*doc*)[¶](#spaczz.pipeline.SpaczzRuler.match)\nUsed in call to find matches in doc.\n\nReturn type\n`List`[`Tuple`[`str`, `int`, `int`, `int`, `str`, `Literal`[‘fuzzy’, ‘regex’, ‘token’, ‘similarity’, ‘phrase’]]]\n\n*property* patterns*: List[Dict[str, Union[str, Dict[str, Any], List[Dict[str, Any]]]]]*[¶](#spaczz.pipeline.SpaczzRuler.patterns)\nGet all patterns and kwargs that were added to the ruler.\n\nReturn type\n`List`[`Dict`[`str`, `Union`[`str`, `Dict`[`str`, `Any`], `List`[`Dict`[`str`, `Any`]]]]]\n\nReturns The original patterns and kwargs, one dictionary for each combination.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.pipeline import SpaczzRuler\n>>> nlp = spacy.blank(\"en\")\n>>> ruler = SpaczzRuler(nlp)\n>>> ruler.add_patterns([{\"label\": \"STREET\", \"pattern\": \"street_addresses\",\n \"type\": \"regex\", \"kwargs\": {\"predef\": True}}])\n>>> ruler.patterns == [\n {\n \"label\": \"STREET\",\n \"pattern\": \"street_addresses\",\n \"type\": \"regex\",\n \"kwargs\": {\"predef\": True},\n },\n ]\nTrue\n```\nremove(*ent_id*)[¶](#spaczz.pipeline.SpaczzRuler.remove)\nRemove patterns by their ent_id.\n\nReturn type\n`None`\n\nscore(*examples*, ***kwargs*)[¶](#spaczz.pipeline.SpaczzRuler.score)\nPipeline scoring for spaCy >= 3.0, < 3.2 compatibility.\n\nReturn type\n`Any`\n\nset_annotations(*doc*, *matches*)[¶](#spaczz.pipeline.SpaczzRuler.set_annotations)\nModify the document in place.\n\nReturn type\n`None`\n\nto_bytes(***, *exclude=[]*)[¶](#spaczz.pipeline.SpaczzRuler.to_bytes)\nSerialize the spaczz ruler patterns to a bytestring.\n\nParameters\n**exclude** (`Iterable`[`str`]) – For spaCy consistency.\n\nReturn type\n`bytes`\n\nReturns The serialized patterns.\n\nExample\n```\n>>> import spacy\n>>> from spaczz.pipeline import SpaczzRuler\n>>> nlp = spacy.blank(\"en\")\n>>> ruler = SpaczzRuler(nlp)\n>>> ruler.add_patterns([{\"label\": \"AUTHOR\", \"pattern\": \"Kerouac\",\n \"type\": \"fuzzy\"}])\n>>> ruler_bytes = ruler.to_bytes()\n>>> isinstance(ruler_bytes, bytes)\nTrue\n```\nto_disk(*path*, ***, *exclude=[]*)[¶](#spaczz.pipeline.SpaczzRuler.to_disk)\nSave the spaczz ruler patterns to a directory.\n\nThe patterns will be saved as newline-delimited JSON (JSONL).\n\nParameters\n* **path** (`Union`[`str`, `Path`]) – The JSONL file to save.\n* **exclude** (`Iterable`[`str`]) – For spaCy consistency.\n\nExample\n```\n>>> import os\n>>> import tempfile\n>>> import spacy\n>>> from spaczz.pipeline import SpaczzRuler\n>>> nlp = spacy.blank(\"en\")\n>>> ruler = SpaczzRuler(nlp)\n>>> ruler.add_patterns([{\"label\": \"AUTHOR\", \"pattern\": \"Kerouac\",\n \"type\": \"fuzzy\"}])\n>>> with tempfile.TemporaryDirectory() as tmpdir:\n>>> ruler.to_disk(f\"{tmpdir}/ruler\")\n>>> isdir = os.path.isdir(f\"{tmpdir}/ruler\")\n>>> isdir True\n```\nReturn type\n`None`\n\n### spaczz.registry[¶](#module-spaczz.registry)\n\nFunction and object registries.\n\nspaczz.registry.get_fuzzy_func(*name*)[¶](#spaczz.registry.get_fuzzy_func)\nGet the registered function for a given name.\n\nname (str): The name.\nRETURNS (Any): The registered function.\n\nReturn type\n`Any`\n\nspaczz.registry.get_re_pattern(*name*)[¶](#spaczz.registry.get_re_pattern)\nGet the registered function for a given name.\n\nname (str): The name.\nRETURNS (Any): The registered function.\n\nReturn type\n`Any`\n\nspaczz.registry.get_re_weights(*name*)[¶](#spaczz.registry.get_re_weights)\nGet the registered function for a given name.\n\nname (str): The name.\nRETURNS (Any): The registered function.\n\nReturn type\n`Any`\n\n### spaczz.customattrs[¶](#module-spaczz.customattrs)\n\nCustom spaCy attributes for spaczz.\n\n*class* spaczz.customattrs.SpaczzAttrs[¶](#spaczz.customattrs.SpaczzAttrs)\nAdds spaczz custom attributes to spaCy.\n\n*static* get_doc_types(*doc*)[¶](#spaczz.customattrs.SpaczzAttrs.get_doc_types)\nGetter for spaczz_types Doc attribute.\n\nReturn type\n`Set`[`Literal`[‘fuzzy’, ‘regex’, ‘token’, ‘similarity’, ‘phrase’]]\n\n*classmethod* get_pattern(*span*)[¶](#spaczz.customattrs.SpaczzAttrs.get_pattern)\nGetter for spaczz_pattern Span attribute.\n\nReturn type\n`Optional`[`str`]\n\n*classmethod* get_ratio(*span*)[¶](#spaczz.customattrs.SpaczzAttrs.get_ratio)\nGetter for spaczz_ratio Span attribute.\n\nReturn type\n`Optional`[`int`]\n\n*static* get_spaczz_doc(*doc*)[¶](#spaczz.customattrs.SpaczzAttrs.get_spaczz_doc)\nGetter for spaczz_doc Doc attribute.\n\nReturn type\n`bool`\n\n*static* get_spaczz_ent(*span*)[¶](#spaczz.customattrs.SpaczzAttrs.get_spaczz_ent)\nGetter for spaczz_ent Span attribute.\n\nReturn type\n`bool`\n\n*classmethod* get_span_type(*span*)[¶](#spaczz.customattrs.SpaczzAttrs.get_span_type)\nGetter for spaczz_type Span attribute.\n\nReturn type\n`Optional`[`Literal`[‘fuzzy’, ‘regex’, ‘token’, ‘similarity’, ‘phrase’]]\n\n*static* get_span_types(*span*)[¶](#spaczz.customattrs.SpaczzAttrs.get_span_types)\nGetter for spaczz_types Span attribute.\n\nReturn type\n`Set`[`Literal`[‘fuzzy’, ‘regex’, ‘token’, ‘similarity’, ‘phrase’]]\n\n*classmethod* initialize()[¶](#spaczz.customattrs.SpaczzAttrs.initialize)\nInitializes and registers custom attributes.\n\nReturn type\n`None`\n\n### spaczz.customtypes[¶](#module-spaczz.customtypes)\n\nCustom spaczz types.\n\n### spaczz.exceptions[¶](#module-spaczz.exceptions)\n\nModule for custom exceptions and warnings.\n\n*exception* spaczz.exceptions.AttrOverwriteWarning[¶](#spaczz.exceptions.AttrOverwriteWarning)\nIt warns if custom attributes are being overwritten.\n\n*exception* spaczz.exceptions.FlexWarning[¶](#spaczz.exceptions.FlexWarning)\nIt warns if flex value is changed if too large.\n\n*exception* spaczz.exceptions.KwargsWarning[¶](#spaczz.exceptions.KwargsWarning)\nIt warns if there are more kwargs than patterns or vice versa.\n\n*exception* spaczz.exceptions.MissingVectorsWarning[¶](#spaczz.exceptions.MissingVectorsWarning)\nIt warns if the spaCy Vocab does not have word vectors.\n\n*exception* spaczz.exceptions.PatternTypeWarning[¶](#spaczz.exceptions.PatternTypeWarning)\nIt warns if the spaczz pattern does not have a valid pattern type.\n\n*exception* spaczz.exceptions.RatioWarning[¶](#spaczz.exceptions.RatioWarning)\nIt warns if match ratio values are incompatible with each other.\n\n*exception* spaczz.exceptions.RegexParseError[¶](#spaczz.exceptions.RegexParseError)\nGeneral error for errors that may happen during regex compilation.\n\nLicense[¶](#license)\n---\n\nMIT License\n\nCopyright (c) 2020 <NAME> (gandersen101)\n\nPermission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the “Software”), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:\n\nThe above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.\n\nTHE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\nFITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\nOUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.\n\nSpaczz provides fuzzy matching and multi-token regex matching functionality to [spaCy](https://spacy.io/).\nSpaczz’s components have similar APIs to their spaCy counterparts and spaczz pipeline components can integrate into spaCy pipelines where they can be saved/loaded as models.\n\nWhile this website will eventually be the home for definitive spaczz documentation, for now it is kind of a placeholder.\nPlease visit [spaczz’s GitHub page](https://github.com/gandersen101/spaczz) for now to see usage documentation and more.\n\nInstallation[¶](#installation)\n---\n\nSpaczz can be installed using pip.\n```\npip install spaczz\n```"}}},{"rowIdx":59173,"cells":{"project":{"kind":"string","value":"time"},"source":{"kind":"string","value":"cran"},"language":{"kind":"string","value":"R"},"content":{"kind":"string","value":"Package ‘time.slots’\n October 14, 2022\nTitle Display Data in a Weekly Calendar View\nVersion 0.2.0\nDescription Generate weekly timetables as a ggplot2 layer. Add\n informative timeslots with elements such as title, key-value pairs, or\n colour to reveal trends.\nLicense MIT + file LICENSE\nURL https://bitbucket.org/annalectnz/time.slots\nBugReports https://bitbucket.org/annalectnz/time.slots/issues\nImports dplyr, ggfittext, ggplot2, lubridate, scales\nSuggests testthat (>= 3.0.0), tidyr\nConfig/testthat/edition 3\nEncoding UTF-8\nRoxygenNote 7.1.2\nNeedsCompilation no\nAuthor <NAME> [aut, cre],\n <NAME> [ctb]\nMaintainer <NAME> <<EMAIL>>\nRepository CRAN\nDate/Publication 2022-04-22 09:20:02 UTC\nR topics documented:\ngeom_weekly_timeslot... 2\nSlotsDetail... 2\n geom_weekly_timeslots Construct geom_weekly_timeslots function\nDescription\n Construct geom_weekly_timeslots function\nUsage\n geom_weekly_timeslots(mapping = NULL, data = NULL, ...)\nArguments\n mapping Set of aesthetic mappings created by aes() or aes_(). If specified and inherit.aes\n = TRUE (the default), it is combined with the default mapping at the top level of\n the plot. You must supply mapping if there is no plot mapping.\n data (SlotsDetails) A SlotsDetails data table.\n ... Other arguments passed on to layer(). These are often aesthetics, used to set\n an aesthetic to a fixed value, like colour = \"red\" or size = 3. They may also\n be parameters to the paired geom/stat.\nValue\n (ggplot) A ggplot object.\nExamples\n geom_weekly_timeslots(data = SlotsDetails())\n SlotsDetails Construct SlotsDetails value object\nDescription\n Construct SlotsDetails value object\nUsage\n SlotsDetails(\n start_datetime = NA,\n end_datetime = NA,\n title = NA_character_,\n body = list(NA_character_)\n )\nArguments\n start_datetime (POSIXct) Is the start of the time slot.For example \"2022-03-16 10:00:00\".\n end_datetime (POSIXct) Is the end of the time slot.For example \"2022-03-16 11:30:00\".\n title (character) Is the title of the time slot. For example \"News 1\".\n body (list) Is a nested list with name value pairs of the time slot details. For example\n list(cost = \"$1200\", TARP = 5).\nValue\n (data.frame) Is a long dataframe of time slots.\nExamples\n SlotsDetails()"}}},{"rowIdx":59174,"cells":{"project":{"kind":"string","value":"owls"},"source":{"kind":"string","value":"readthedoc"},"language":{"kind":"string","value":"Unknown"},"content":{"kind":"string","value":"# AD Leo\n\nDate: 2021-06-02\nCategories: \nTags: \n\nAD Leo# Search exo.mast Search SIMBAD Date S err 2/6/2021 8:39:32 8.660 0.361 1/12/2022 10:42:27 9.671 0.404 TESS Light Curve#\n\n# EV Lac\n\nDate: 2021-11-09\nCategories: \nTags: \n\nEV Lac# Search exo.mast Search SIMBAD Date S err 9/11/2021 9:26:27 9.614 0.405 9/13/2022 9:34:17 11.345 0.490 TESS Light Curve#\n\n# Gl 436\n\nGl 436# Search exo.mast Search SIMBAD Date S err 6/10/20 4:36 AM 0.691 0.031 1/6/21 12:21 PM 0.755 0.034 2/11/22 10:39 AM 0.728 0.032 TESS Light Curve#\n\n# HAT-P-11\n\n# HAT-P-11#\n\nDate | S | err |\n | --- | --- | --- |\n9/11/21 7:42 AM | 0.507 | 0.021 |\n9/13/22 7:32 AM | 0.499 | 0.021 |\n\nHAT-P-11 |\n | --- |\nst_teff | 4780.0 |\nst_spectype | K4 |\nst_rad | 0.68 |\nst_mass | 0.81 |\nst_rotp | 29.2 |\nsy_bmag | 10.556 |\nsy_vmag | 9.46 |\nsy_gaiamag | 9.15098 |\n\n# HAT-P-13\n\n# HAT-P-13#\n\nDate | S | err |\n | --- | --- | --- |\n1/6/21 8:31 AM | 0.141 | 0.006 |\n1/25/22 8:11 AM | 0.144 | 0.006 |\n\nHAT-P-13 |\n | --- |\nst_teff | 5653.0 |\nst_spectype | G4 |\nst_rad | 1.76 |\nst_mass | 1.32 |\nst_rotp | NaN |\nsy_bmag | 11.147 |\nsy_vmag | 10.421 |\nsy_gaiamag | 10.4168 |\n\n# HAT-P-14\n\n# HAT-P-14#\n\n# HAT-P-16\n\n# HAT-P-16#\n\n# HAT-P-17\n\n# HAT-P-17#\n\n# HAT-P-33\n\n# HAT-P-33#\n\n# HD 106315\n\n# HD 106315#\n\n# HD 129333\n\nHD 129333# Search exo.mast Search SIMBAD Date S err 3/21/21 6:48 0.572 0.023 4/12/22 7:51 0.576 0.023 4/16/22 6:47 0.633 0.026 TESS Light Curve#\n\n# HD 17925\n\nHD 17925# Search exo.mast Search SIMBAD Date S err 9/22/21 8:05 0.633 0.027 1/29/23 2:43 0.712 0.029 1/29/23 2:47 0.715 0.029 TESS Light Curve#\n\n# HD 190007\n\nHD 190007# Search exo.mast Search SIMBAD Date S err 8/20/21 8:03 0.678 0.029 9/18/22 5:03 0.718 0.030\n\n# HD 190360\n\nHD 190360# Search exo.mast Search SIMBAD Date S err 8/20/21 8:46 0.154 0.006 9/13/22 8:00 0.160 0.007\n\n# HD 219834 A\n\nHD 219834 A# Search exo.mast Search SIMBAD Date S err 11/13/21 3:00 0.16 0.007 9/18/22 8:33 0.16 0.006 TESS Light Curve#\n\n# HD 22072\n\nHD 22072# Search exo.mast Search SIMBAD Date S err 8/29/21 11:30 0.137 0.006 11/21/21 9:53 0.213 0.013 11/21/21 9:57 0.224 0.014 TESS Light Curve#\n\n# HD 3167\n\nHD 3167# Search exo.mast Search SIMBAD Date S err 8/2/20 8:50 0.190 0.008 11/13/21 5:30 0.198 0.008 11/7/22 5:17 0.193 0.008\n\n# HD 32147\n\nHD 32147# Search exo.mast Search SIMBAD Date S err 3/21/21 2:12 0.350 0.015 8/29/21 11:44 0.343 0.014 1/16/22 5:24 0.351 0.015 TESS Light Curve#\n\n# HD 3443\n\nHD 3443# Search exo.mast Search SIMBAD Date S err 9/22/21 6:42 0.191 0.008 1/29/23 1:24 0.190 0.008 1/29/23 1:28 0.191 0.008 TESS Light Curve#\n\n# HD 80653\n\nHD 80653# Search exo.mast Search SIMBAD Date S err 1/6/21 9:38 0.157 0.007 2/13/22 5:27 0.164 0.007 TESS Light Curve#\n\n# HD 97658\n\nHD 97658# Search exo.mast Search SIMBAD Date S err 6/10/20 4:14 0.230 0.009 1/6/21 12:42 0.220 0.009 2/16/22 7:55 0.237 0.010 TESS Light Curve#\n\n# K2-100\n\n# K2-100#\n\nDate | S | err |\n | --- | --- | --- |\n5/23/21 2:54 | 0.353 | 0.017 |\n11/20/21 9:27 | 0.318 | 0.013 |\n\nK2-100 |\n | --- |\nst_teff | 5945.0 |\nst_spectype | G0 V |\nst_rad | 1.24 |\nst_mass | 1.15 |\nst_rotp | 4.3 |\nsy_bmag | 11.234 |\nsy_vmag | 10.651 |\nsy_gaiamag | 10.4053 |\n\n# K2-105\n\n# K2-105#\n\nDate | S | err |\n | --- | --- | --- |\n10/31/20 10:09 | 0.265 | 0.013 |\n1/12/22 9:12 | 0.260 | 0.013 |\n\nK2-105 |\n | --- |\nst_teff | 5636.0 |\nst_spectype | K1 |\nst_rad | 0.97 |\nst_mass | 1.05 |\nst_rotp | NaN |\nsy_bmag | 12.681 |\nsy_vmag | 11.752 |\nsy_gaiamag | 11.7124 |\n\n# K2-111\n\n# K2-111#\n\nDate | S | err |\n | --- | --- | --- |\n10/2/20 10:37 | 0.154 | 0.007 |\n1/16/22 4:20 | 0.163 | 0.007 |\n\nK2-111 |\n | --- |\nst_teff | 5775.0 |\nst_spectype | G2 |\nst_rad | 1.25 |\nst_mass | 0.84 |\nst_rotp | 29.2 |\nsy_bmag | 11.796 |\nsy_vmag | 11.13 |\nsy_gaiamag | 10.9294 |\n\n# K2-131\n\n# K2-131#\n\nDate | S | err |\n | --- | --- | --- |\n2/6/21 11:07 | 0.493 | 0.026 |\n2/11/22 11:12 | 0.523 | 0.026 |\n\nK2-131 |\n | --- |\nst_teff | 5200.0 |\nst_spectype |\nst_rad | 0.81 |\nst_mass | 0.84 |\nst_rotp | 9.37 |\nsy_bmag | 13.054 |\nsy_vmag | 12.129 |\nsy_gaiamag | 11.8999 |\n\n# K2-136\n\n# K2-136#\n\nDate | S | err |\n | --- | --- | --- |\n10/2/20 11:10 | 1.345 | 0.058 |\n9/22/21 10:52 | 1.236 | 0.056 |\n9/22/21 12:01 | 1.195 | 0.055 |\n\nK2-136 |\n | --- |\nst_teff | 4499.0 |\nst_spectype | K5.5 |\nst_rad | 0.66 |\nst_mass | 0.74 |\nst_rotp | 15.0 |\nsy_bmag | 12.479 |\nsy_vmag | 11.101 |\nsy_gaiamag | 10.8545 |\n\n# K2-162\n\n# K2-162#\n\n# K2-182\n\n# K2-182#\n\n# K2-184\n\n# K2-184#\n\n# K2-222\n\n# K2-222#\n\n# K2-229\n\n# K2-229#\n\nDate | S | err |\n | --- | --- | --- |\n2/20/21 9:19 | 0.422 | 0.018 |\n2/16/22 9:55 | 0.408 | 0.018 |\n\nK2-229 |\n | --- |\nst_teff | 5315.0 |\nst_spectype | G9 |\nst_rad | 0.79 |\nst_mass | 0.87 |\nst_rotp | 18.1 |\nsy_bmag | 12.049 |\nsy_vmag | 10.985 |\nsy_gaiamag | 10.7861 |\n\n# K2-232\n\n# K2-232#\n\n# KELT-18\n\n# KELT-18#\n\nDate | S | err |\n | --- | --- | --- |\n9/29/20 2:30 | 0.249 | 0.012 |\n4/12/22 7:30 | 0.247 | 0.010 |\n\nKELT-18 |\n | --- |\nst_teff | 6670.0 |\nst_spectype | F4 V |\nst_rad | 1.91 |\nst_mass | 1.52 |\nst_rotp | NaN |\nsy_bmag | 10.62 |\nsy_vmag | 10.16 |\nsy_gaiamag | 10.1036 |\n\n# Kepler-410A\n\nKepler-410A# Search exo.mast Search SIMBAD Date S err 9/29/20 3:42 0.142 0.006 11/13/21 1:29 0.165 0.007 TESS Light Curve#\n\n# Kepler-96\n\n# Kepler-96#\n\n# Qatar-6\n\n# Qatar-6#\n\n# TOI 1180\n\nTOI 1180# Search exo.mast Search SIMBAD Date S err 3/3/21 10:46 0.436 0.022 2/16/22 10:52 0.336 0.019 TESS Light Curve#\n\n# TOI 1260\n\n# TOI 1260#\n\n# TOI 1280\n\nTOI 1280# Search exo.mast Search SIMBAD Date S err 8/20/21 8:26 0.134 0.006 11/7/22 2:26 0.510 0.023 TESS Light Curve#\n\n# TOI 1411\n\nTOI 1411# Search exo.mast Search SIMBAD Date S err 3/3/21 11:11 1.06 0.049 3/20/22 10:26 0.95 0.040 TESS Light Curve#\n\n# TOI 1416\n\nTOI 1416# Search exo.mast Search SIMBAD Date S err 2/20/21 10:47 0.304 0.012 3/20/22 9:53 0.284 0.012 4/16/22 8:44 0.274 0.012 TESS Light Curve#\n\n# TOI 1468\n\nTOI 1468# Search exo.mast Search SIMBAD Date S err 8/29/21 10:01 1.130 0.067 10/11/21 8:10 1.409 0.077 11/7/22 6:23 0.844 0.044 TESS Light Curve#\n\n# TOI 1730\n\nTOI 1730# Search exo.mast Search SIMBAD Date S err 2/20/21 8:18 0.921 0.056 10/11/21 11:07 1.359 0.063 TESS Light Curve#\n\n# TOI 2093\n\nDate: 2021-09-11\nCategories: \nTags: \n\nTOI 2093# Search exo.mast Search SIMBAD Date S err 2021-09-11 8:14:43 0.545 0.034 2022-09-13 8:19:02 0.497 0.031 TESS Light Curve#\n\n# TOI 2318\n\nDate: 2022-05-22\nCategories: \nTags: \n\nTOI 2318# Search exo.mast Search SIMBAD Date S err 2021-06-21 5:39:01 0.442 0.019 2022-05-22 10:27:18 0.475 0.026 2022-07-24 7:18:24 0.435 0.019 TESS Light Curve#\n\n# TOI 2431\n\nDate: 2022-01-17\nCategories: \nTags: \n\nTOI 2431# Search exo.mast Search SIMBAD Date S err 2022-01-17 1:32:10 1.275 0.055 2023-01-29 1:45:49 1.240 0.054 TESS Light Curve#\n\n# TOI 2443\n\nDate: 2021-09-22\nCategories: \nTags: \n\nTOI 2443# Search exo.mast Search SIMBAD Date S err 2021-09-22 7:55:35 0.436 0.022 2022-01-17 2:02:10 0.467 0.020 2023-01-29 2:20:06 0.476 0.020 TESS Light Curve#\n\n# TOI 969\n\nDate: 2021-10-29\nCategories: \nTags: \n\nTOI 969# Search exo.mast Search SIMBAD Date S err 2021-03-21 4:05:20 1.213 0.057 2021-10-29 11:29:19 1.128 0.052 2022-02-13 3:53:56 1.342 0.060 TESS Light Curve#\n\n# WASP-106\n\n# WASP-106#\n\nDate | S | err |\n | --- | --- | --- |\n2021-01-06 11:16:29 | 0.211 | 0.010 |\n2022-02-11 10:06:34 | 0.156 | 0.007 |\n\nWASP-106 |\n | --- |\nst_teff | 6055.0 |\nst_spectype | F9 |\nst_rad | 1.31 |\nst_mass | 0.93 |\nst_rotp | NaN |\nsy_bmag | 12.04 |\nsy_vmag | 11.21 |\nsy_gaiamag | 11.3659 |\n\n# WASP-107\n\nDate: 2021-02-06\nCategories: \nTags: \n\n# WASP-107#\n\nDate | S | err |\n | --- | --- | --- |\n2021-02-06 11:35:14 | 0.828 | 0.050 |\n2021-02-20 9:51:43 | 0.889 | 0.044 |\n2022-03-20 6:43:12 | 0.877 | 0.042 |\n\nWASP-107 |\n | --- |\nst_teff | 4425.0 |\nst_spectype | K6 |\nst_rad | 0.67 |\nst_mass | 0.68 |\nst_rotp | 17.0 |\nsy_bmag | 14.913 |\nsy_vmag | 11.592 |\nsy_gaiamag | 11.174 |\n\n# WASP-11\n\n# WASP-11#\n\nDate | S | err |\n | --- | --- | --- |\n2021-01-06 7:40:40 | 0.383 | 0.023 |\n2021-10-29 7:00:16 | 0.290 | 0.013 |\n\nWASP-11 |\n | --- |\nst_teff | 4800.0 |\nst_spectype | K3 V |\nst_rad | 0.89 |\nst_mass | 1.42 |\nst_rotp | NaN |\nsy_bmag | 12.575 |\nsy_vmag | 11.567 |\nsy_gaiamag | 11.5535 |\n\n# WASP-52\n\nDate: 2020-10-02\nCategories: \nTags: \n\n# WASP-52#\n\nDate | S | err |\n | --- | --- | --- |\n2020-10-02 7:28:51 | 0.497 | 0.023 |\n2022-09-18 7:34:26 | 0.574 | 0.027 |\n\nWASP-52 |\n | --- |\nst_teff | 5000.0 |\nst_spectype | K2 V |\nst_rad | 0.79 |\nst_mass | 0.87 |\nst_rotp | NaN |\nsy_bmag | 13.045 |\nsy_vmag | 12.192 |\nsy_gaiamag | 11.9537 |\n\n# WASP-65\n\nDate: 2021-01-05\nCategories: \nTags: \n\n# WASP-65#\n\nDate | S | err |\n | --- | --- | --- |\n2021-01-05 9:56:30 | 0.238 | 0.011 |\n2022-01-25 8:44:26 | 0.211 | 0.009 |\n\nWASP-65 |\n | --- |\nst_teff | 5600.0 |\nst_spectype | G6 |\nst_rad | 1.01 |\nst_mass | 0.93 |\nst_rotp | NaN |\nsy_bmag | 12.386 |\nsy_vmag | 11.869 |\nsy_gaiamag | 11.7509 |\n\n# WASP-84\n\nDate: 2020-10-31\nCategories: \nTags: \n\n# WASP-84#\n\nDate | S | err |\n | --- | --- | --- |\n2020-10-31 11:16:38 | 0.473 | 0.020 |\n2021-11-20 10:32:31 | 0.520 | 0.022 |\n2022-02-27 5:05:18 | 0.497 | 0.021 |\n\nWASP-84 |\n | --- |\nst_teff | 5314.0 |\nst_spectype | K0 |\nst_rad | 0.75 |\nst_mass | 0.84 |\nst_rotp | 14.36 |\nsy_bmag | 11.64 |\nsy_vmag | 10.825 |\nsy_gaiamag | 10.5745 |\n\n# XO-1\n\nDate: 2020-06-10\nCategories: \nTags: \n\n# XO-1#"}}},{"rowIdx":59175,"cells":{"project":{"kind":"string","value":"github.com/go-vgo/robotgo"},"source":{"kind":"string","value":"go"},"language":{"kind":"string","value":"Go"},"content":{"kind":"string","value":"README\n [¶](#section-readme)\n---\n\n### Robotgo\n\n[](https://github.com/go-vgo/robotgo/commits/master)\n[](https://circleci.com/gh/go-vgo/robotgo)\n[](https://travis-ci.org/go-vgo/robotgo)\n\n[](https://goreportcard.com/report/github.com/go-vgo/robotgo)\n[](https://godoc.org/github.com/go-vgo/robotgo)\n[](https://github.com/go-vgo/robotgo/releases/latest)\n[](https://gitter.im/go-vgo/robotgo?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge)\n\nDocumentation\n [¶](#section-documentation)\n---\n\n[Rendered for](https://go.dev/about#build-context)\n\nlinux/amd64 windows/amd64 darwin/amd64 js/wasm\n\n \n### Overview [¶](#pkg-overview)\n\nPackage robotgo Go native cross-platform system automation.\n\nPlease make sure Golang, GCC is installed correctly before installing RobotGo;\n\nSee Requirements:\n```\nhttps://github.com/go-vgo/robotgo#requirements\n```\nInstallation:\nWith Go module support (Go 1.11+), just import:\n```go\n```\nimport \"github.com/go-vgo/robotgo\"\n```\n```\nOtherwise, to install the robotgo package, run the command:\n```\ngo get -u github.com/go-vgo/robotgo\n```\n### Index [¶](#pkg-index)\n\n* [Constants](#pkg-constants)\n* [Variables](#pkg-variables)\n* [func ActiveName(name string) error](#ActiveName)\n* [func ActivePID(pid int32, args ...int) error](#ActivePID)\n* [func ActivePIDC(pid int32, args ...int) error](#ActivePIDC)\n* [func AddEvent(key string) bool](#AddEvent)\n* [func AddEvents(key string, arr ...string) bool](#AddEvents)\n* [func AddMouse(btn string, x ...int16) bool](#AddMouse)\n* [func AddMousePos(x, y int16) bool](#AddMousePos)\n* [func BitmapClick(bitmap C.MMBitmapRef, args ...interface{})](#BitmapClick)\n* [func BitmapFromStr(str string) C.MMBitmapRef](#BitmapFromStr)\n* [func BitmapStr(str string) C.MMBitmapRef](#BitmapStr)deprecated\n* [func ByteToImg(b []byte) (image.Image, error)](#ByteToImg)\n* [func CaptureImg(args ...int) image.Image](#CaptureImg)\n* [func CaptureScreen(args ...int) C.MMBitmapRef](#CaptureScreen)\n* [func CharCodeAt(s string, n int) rune](#CharCodeAt)\n* [func CheckMouse(btn string) C.MMMouseButton](#CheckMouse)\n* [func Click(args ...interface{})](#Click)\n* [func CloseWindow(args ...int32)](#CloseWindow)\n* [func Convert(opath, spath string, args ...int) string](#Convert)\n* [func CopyBitPB(bitmap C.MMBitmapRef) bool](#CopyBitPB)\n* [func CopyBitpb(bitmap C.MMBitmapRef) bool](#CopyBitpb)deprecated\n* [func CountBitmap(bitmap, sbit C.MMBitmapRef, args ...float32) int](#CountBitmap)\n* [func CountColor(color CHex, args ...interface{}) int](#CountColor)\n* [func CountColorCS(color CHex, x, y, w, h int, args ...float64) int](#CountColorCS)\n* [func DecodeImg(path string) (image.Image, string, error)](#DecodeImg)\n* [func DeepCopyBit(bitmap C.MMBitmapRef) C.MMBitmapRef](#DeepCopyBit)\n* [func Drag(x, y int, args ...string)](#Drag)deprecated\n* [func DragMouse(x, y int, args ...interface{})](#DragMouse)deprecated\n* [func DragSmooth(x, y int, args ...interface{})](#DragSmooth)\n* [func End()](#End)deprecated\n* [func EventEnd()](#EventEnd)\n* [func EventHook(when uint8, keysPressed []string, callback func(hook.Event))](#EventHook)\n* [func EventProcess(events chan hook.Event) chan bool](#EventProcess)\n* [func EventStart() chan hook.Event](#EventStart)\n* [func FindBitmap(bit C.MMBitmapRef, args ...interface{}) (int, int)](#FindBitmap)\n* [func FindCBitmap(bmp CBitmap, args ...interface{}) (int, int)](#FindCBitmap)\n* [func FindColor(color CHex, args ...interface{}) (int, int)](#FindColor)\n* [func FindColorCS(color CHex, x, y, w, h int, args ...float64) (int, int)](#FindColorCS)\n* [func FindIds(name string) ([]int32, error)](#FindIds)\n* [func FindName(pid int32) (string, error)](#FindName)\n* [func FindNames() ([]string, error)](#FindNames)\n* [func FindPath(pid int32) (string, error)](#FindPath)\n* [func FindPic(path string, args ...interface{}) (int, int)](#FindPic)\n* [func FreeBitmap(bitmap C.MMBitmapRef)](#FreeBitmap)\n* [func FreeBitmapArr(bit ...C.MMBitmapRef)](#FreeBitmapArr)\n* [func FreeMMPointArr(pointArray C.MMPointArrayRef)](#FreeMMPointArr)\n* [func GetActive() C.MData](#GetActive)\n* [func GetBHandle() int](#GetBHandle)deprecated\n* [func GetBounds(pid int32, args ...int) (int, int, int, int)](#GetBounds)\n* [func GetColor(bitmap C.MMBitmapRef, x, y int) C.MMRGBHex](#GetColor)\n* [func GetColors(bitmap C.MMBitmapRef, x, y int) string](#GetColors)\n* [func GetHandPid(pid int32, args ...int32) C.MData](#GetHandPid)\n* [func GetHandle() int](#GetHandle)\n* [func GetImgSize(imgPath string) (int, int)](#GetImgSize)\n* [func GetMouseColor() string](#GetMouseColor)\n* [func GetMousePos() (int, int)](#GetMousePos)\n* [func GetPID() int32](#GetPID)\n* [func GetPixelColor(x, y int) string](#GetPixelColor)\n* [func GetPortion(bit C.MMBitmapRef, x, y, w, h int) C.MMBitmapRef](#GetPortion)\n* [func GetPxColor(x, y int) C.MMRGBHex](#GetPxColor)\n* [func GetScaleSize() (int, int)](#GetScaleSize)\n* [func GetScreenSize() (int, int)](#GetScreenSize)\n* [func GetText(imgPath string, args ...string) (string, error)](#GetText)\n* [func GetTitle(args ...int32) string](#GetTitle)\n* [func GetVersion() string](#GetVersion)\n* [func GetXDisplayName() string](#GetXDisplayName)\n* [func GetXId(xu *xgbutil.XUtil, pid int32) (xproto.Window, error)](#GetXId)\n* [func GetXidFromPid(xu *xgbutil.XUtil, pid int32) (xproto.Window, error)](#GetXidFromPid)\n* [func GoString(char *C.char) string](#GoString)\n* [func Height(img image.Image) int](#Height)\n* [func HexToRgb(hex uint32) *C.uint8_t](#HexToRgb)\n* [func ImgSize(path string) (int, int, error)](#ImgSize)\n* [func Is64Bit() bool](#Is64Bit)\n* [func IsValid() bool](#IsValid)\n* [func KeyDown(key string)](#KeyDown)\n* [func KeyPress(key string)](#KeyPress)\n* [func KeyTap(tapKey string, args ...interface{}) string](#KeyTap)\n* [func KeyToggle(key string, args ...string) string](#KeyToggle)\n* [func KeyUp(key string)](#KeyUp)\n* [func Kill(pid int32) error](#Kill)\n* [func MaxWindow(pid int32, args ...interface{})](#MaxWindow)\n* [func MicroSleep(tm float64)](#MicroSleep)\n* [func MilliSleep(tm int)](#MilliSleep)\n* [func MinWindow(pid int32, args ...interface{})](#MinWindow)\n* [func MouseClick(args ...interface{})](#MouseClick)deprecated\n* [func MouseToggle(togKey string, args ...interface{}) int](#MouseToggle)deprecated\n* [func Move(x, y int)](#Move)\n* [func MoveArgs(x, y int) (int, int)](#MoveArgs)\n* [func MoveClick(x, y int, args ...interface{})](#MoveClick)\n* [func MoveMouse(x, y int)](#MoveMouse)deprecated\n* [func MoveMouseSmooth(x, y int, args ...interface{}) bool](#MoveMouseSmooth)deprecated\n* [func MoveRelative(x, y int)](#MoveRelative)\n* [func MoveSmooth(x, y int, args ...interface{}) bool](#MoveSmooth)\n* [func MoveSmoothRelative(x, y int, args ...interface{})](#MoveSmoothRelative)\n* [func MovesClick(x, y int, args ...interface{})](#MovesClick)\n* [func Mul(x int) int](#Mul)deprecated\n* [func OpenBitmap(gpath string, args ...int) C.MMBitmapRef](#OpenBitmap)\n* [func OpenImg(path string) ([]byte, error)](#OpenImg)\n* [func PadHex(hex C.MMRGBHex) string](#PadHex)\n* [func PasteStr(str string) string](#PasteStr)\n* [func PidExists(pid int32) (bool, error)](#PidExists)\n* [func Pids() ([]int32, error)](#Pids)\n* [func PointInBounds(bitmap C.MMBitmapRef, x, y int) bool](#PointInBounds)\n* [func Read(path string) (image.Image, error)](#Read)\n* [func ReadAll() (string, error)](#ReadAll)\n* [func ReadBitmap(bitmap C.MMBitmapRef) bool](#ReadBitmap)\n* [func RgbToHex(r, g, b uint8) C.uint32_t](#RgbToHex)\n* [func Run(path string) ([]byte, error)](#Run)\n* [func Save(img image.Image, path string) error](#Save)\n* [func SaveBitmap(bitmap C.MMBitmapRef, gpath string, args ...int) string](#SaveBitmap)\n* [func SaveCapture(spath string, args ...int) string](#SaveCapture)\n* [func SaveImg(b []byte, path string) error](#SaveImg)\n* [func SaveJpeg(img image.Image, path string) error](#SaveJpeg)\n* [func SavePng(img image.Image, path string) error](#SavePng)\n* [func Scale() int](#Scale)deprecated\n* [func Scale0() int](#Scale0)deprecated\n* [func ScaleF() float64](#ScaleF)\n* [func ScaleX() int](#ScaleX)deprecated\n* [func ScaleY() int](#ScaleY)deprecated\n* [func Scaled(x int) int](#Scaled)\n* [func Scaled0(x int, f float64) int](#Scaled0)\n* [func Scroll(x, y int, args ...int)](#Scroll)\n* [func ScrollMouse(x int, direction string)](#ScrollMouse)deprecated\n* [func ScrollRelative(x, y int, args ...int)](#ScrollRelative)\n* [func ScrollSmooth(to int, args ...int)](#ScrollSmooth)\n* [func SetActive(win C.MData)](#SetActive)\n* [func SetDelay(d ...int)](#SetDelay)\n* [func SetHandle(hwnd int)](#SetHandle)\n* [func SetHandlePid(pid int32, args ...int32)](#SetHandlePid)\n* [func SetKeyDelay(delay int)](#SetKeyDelay)\n* [func SetKeyboardDelay(delay int)](#SetKeyboardDelay)deprecated\n* [func SetMouseDelay(delay int)](#SetMouseDelay)\n* [func SetXDisplayName(name string) string](#SetXDisplayName)\n* [func ShowAlert(title, msg string, args ...string) bool](#ShowAlert)\n* [func Sleep(tm int)](#Sleep)\n* [func Start() chan hook.Event](#Start)deprecated\n* [func StopEvent()](#StopEvent)\n* [func StrToImg(data string) (image.Image, error)](#StrToImg)\n* [func SysScale() float64](#SysScale)\n* [func ToBitmapBytes(bit C.MMBitmapRef) []byte](#ToBitmapBytes)\n* [func ToByteImg(img image.Image, fm ...string) []byte](#ToByteImg)\n* [func ToCBitmap(bit Bitmap) C.MMBitmapRef](#ToCBitmap)\n* [func ToImage(bit C.MMBitmapRef) image.Image](#ToImage)\n* [func ToMMBitmapRef(bit CBitmap) C.MMBitmapRef](#ToMMBitmapRef)\n* [func ToMMRGBHex(hex CHex) C.MMRGBHex](#ToMMRGBHex)\n* [func ToRGBA(bit C.MMBitmapRef) *image.RGBA](#ToRGBA)\n* [func ToRGBAGo(bmp1 Bitmap) *image.RGBA](#ToRGBAGo)\n* [func ToStringImg(img image.Image, fm ...string) string](#ToStringImg)\n* [func ToUC(text string) []string](#ToUC)\n* [func ToUint8p(dst []uint8) *uint8](#ToUint8p)\n* [func TocharBitmap(bit C.MMBitmapRef) *C.char](#TocharBitmap)\n* [func Toggle(key ...string) int](#Toggle)\n* [func TostringBitmap(bit C.MMBitmapRef) string](#TostringBitmap)\n* [func Try(fun func(), handler func(interface{}))](#Try)\n* [func TypeStr(str string, args ...float64)](#TypeStr)\n* [func TypeStrDelay(str string, delay int)](#TypeStrDelay)\n* [func TypeString(str string, delay ...int)](#TypeString)deprecated\n* [func TypeStringDelayed(str string, delay int)](#TypeStringDelayed)deprecated\n* [func U32ToHex(hex C.uint32_t) C.MMRGBHex](#U32ToHex)\n* [func U8ToHex(hex *C.uint8_t) C.MMRGBHex](#U8ToHex)\n* [func UnicodeType(str uint32)](#UnicodeType)\n* [func Width(img image.Image) int](#Width)\n* [func WriteAll(text string) error](#WriteAll)\n* [type Bitmap](#Bitmap)\n* + [func GoCaptureScreen(args ...int) Bitmap](#GoCaptureScreen)\n\t+ [func ImgToBitmap(m image.Image) (bit Bitmap)](#ImgToBitmap)\n\t+ [func RGBAToBitmap(r1 *image.RGBA) (bit Bitmap)](#RGBAToBitmap)\n\t+ [func ToBitmap(bit C.MMBitmapRef) Bitmap](#ToBitmap)\n* [type CBitmap](#CBitmap)\n* [type CHex](#CHex)\n* + [func UintToHex(u uint32) CHex](#UintToHex)\n* [type Map](#Map)\n* [type Nps](#Nps)\n* + [func Process() ([]Nps, error)](#Process)\n* [type Point](#Point)\n* + [func FindAllBitmap(bit C.MMBitmapRef, args ...interface{}) (posArr []Point)](#FindAllBitmap)\n\t+ [func FindAllColor(color CHex, args ...interface{}) (posArr []Point)](#FindAllColor)\n\t+ [func FindEveryBitmap(bit C.MMBitmapRef, args ...interface{}) []Point](#FindEveryBitmap)deprecated\n\t+ [func FindEveryColor(color CHex, args ...interface{}) []Point](#FindEveryColor)deprecated\n* [type Rect](#Rect)\n* + [func GetScreenRect(displayId ...int) Rect](#GetScreenRect)\n* [type Size](#Size)\n\n### Constants [¶](#pkg-constants)\n\n```\nconst (\n // Version get the robotgo version\n\tVersion = \"v0.100.0.1189, MT. Baker!\"\n)\n```\n### Variables [¶](#pkg-variables)\n\n```\nvar (\n // MouseSleep set the mouse default millisecond sleep time\n\tMouseSleep = 0\n // KeySleep set the key default millisecond sleep time\n\tKeySleep = 0\n)\n```\n\n```\nvar Keycode = [keycode](/github.com/vcaesar/keycode).[Keycode](/github.com/vcaesar/keycode#Keycode)\n```\nKeycode robotgo hook key's code map\n\n```\nvar MouseMap = [keycode](/github.com/vcaesar/keycode).[MouseMap](/github.com/vcaesar/keycode#MouseMap)\n```\nMouseMap robotgo hook mouse's code map\n\n```\nvar Special = [keycode](/github.com/vcaesar/keycode).[Special](/github.com/vcaesar/keycode#Special)\n```\nSpecial is the special key map\n\n### Functions [¶](#pkg-functions)\n\n#### \nfunc [ActiveName](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1391) [¶](#ActiveName)\n```\nfunc ActiveName(name [string](/builtin#string)) [error](/builtin#error)\n```\nActiveName active the window by name\n\nExamples:\n```\nrobotgo.ActiveName(\"chrome\")\n```\n#### \nfunc [ActivePID](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo_unix.go#L85) [¶](#ActivePID)\n```\nfunc ActivePID(pid [int32](/builtin#int32), args ...[int](/builtin#int)) [error](/builtin#error)\n```\nActivePID active the window by PID,\n\nIf args[0] > 0 on the Windows platform via a window handle to active,\nIf args[0] > 0 on the unix platform via a xid to active\n\n#### \nfunc [ActivePIDC](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo_unix.go#L63) [¶](#ActivePIDC)\n```\nfunc ActivePIDC(pid [int32](/builtin#int32), args ...[int](/builtin#int)) [error](/builtin#error)\n```\nActivePIDC active the window by PID,\nIf args[0] > 0 on the unix platform via a xid to active\n\n#### \nfunc [AddEvent](https://github.com/go-vgo/robotgo/blob/v0.100.10/hook.go#L78) [¶](#AddEvent)\n```\nfunc AddEvent(key [string](/builtin#string)) [bool](/builtin#bool)\n```\nAddEvent add event listener,\n\nparameters for the string type,\nthe keyboard corresponding key parameters,\n\nmouse arguments: mleft, center, mright, wheelDown, wheelUp,\nwheelLeft, wheelRight.\n\nUse \"robotgo.AddEvents()\" or \"gohook\" add asynchronous event listener\n\n#### \nfunc [AddEvents](https://github.com/go-vgo/robotgo/blob/v0.100.10/hook.go#L106) [¶](#AddEvents)\n```\nfunc AddEvents(key [string](/builtin#string), arr ...[string](/builtin#string)) [bool](/builtin#bool)\n```\nAddEvents add global event hook\n\nrobotgo.AddEvents(\"q\")\nrobotgo.AddEvents(\"q\", \"ctrl\")\nrobotgo.AddEvents(\"q\", \"ctrl\", \"shift\")\n\n#### \nfunc [AddMouse](https://github.com/go-vgo/robotgo/blob/v0.100.10/hook.go#L157) [¶](#AddMouse)\n```\nfunc AddMouse(btn [string](/builtin#string), x ...[int16](/builtin#int16)) [bool](/builtin#bool)\n```\nAddMouse add mouse event hook\n\nmouse arguments: left, center, right, wheelDown, wheelUp,\nwheelLeft, wheelRight.\n\nrobotgo.AddMouse(\"left\")\nrobotgo.AddMouse(\"left\", 100, 100)\n\n#### \nfunc [AddMousePos](https://github.com/go-vgo/robotgo/blob/v0.100.10/hook.go#L183) [¶](#AddMousePos)\n```\nfunc AddMousePos(x, y [int16](/builtin#int16)) [bool](/builtin#bool)\n```\nAddMousePos add listen mouse event pos hook\n\n#### \nfunc [BitmapClick](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L260) [¶](#BitmapClick)\n```\nfunc BitmapClick(bitmap [C](/C).[MMBitmapRef](/C#MMBitmapRef), args ...interface{})\n```\nBitmapClick find the bitmap and click\n\n#### \nfunc [BitmapFromStr](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L300) [¶](#BitmapFromStr)\n```\nfunc BitmapFromStr(str [string](/builtin#string)) [C](/C).[MMBitmapRef](/C#MMBitmapRef)\n```\nBitmapFromStr read bitmap from the string\n\n#### \n\n func [BitmapStr](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L295)\ndeprecated\n```\nfunc BitmapStr(str [string](/builtin#string)) [C](/C).[MMBitmapRef](/C#MMBitmapRef)\n```\nDeprecated: use the BitmapFromStr(),\n\nBitmapStr bitmap from string\n\n#### \nfunc [ByteToImg](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L72) [¶](#ByteToImg)\n\nadded in v0.93.2\n```\nfunc ByteToImg(b [][byte](/builtin#byte)) ([image](/image).[Image](/image#Image), [error](/builtin#error))\n```\nByteToImg convert []byte to image.Image\n\n#### \nfunc [CaptureImg](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L310) [¶](#CaptureImg)\n\nadded in v0.100.1\n```\nfunc CaptureImg(args ...[int](/builtin#int)) [image](/image).[Image](/image#Image)\n```\nCaptureImg capture the screen and return image.Image\n\n#### \nfunc [CaptureScreen](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L279) [¶](#CaptureScreen)\n```\nfunc CaptureScreen(args ...[int](/builtin#int)) [C](/C).[MMBitmapRef](/C#MMBitmapRef)\n```\nCaptureScreen capture the screen return bitmap(c struct),\nuse `defer robotgo.FreeBitmap(bitmap)` to free the bitmap\n\nrobotgo.CaptureScreen(x, y, w, h int)\n\n#### \nfunc [CharCodeAt](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L989) [¶](#CharCodeAt)\n```\nfunc CharCodeAt(s [string](/builtin#string), n [int](/builtin#int)) [rune](/builtin#rune)\n```\nCharCodeAt char code at utf-8\n\n#### \nfunc [CheckMouse](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L432) [¶](#CheckMouse)\n```\nfunc CheckMouse(btn [string](/builtin#string)) [C](/C).[MMMouseButton](/C#MMMouseButton)\n```\nCheckMouse check the mouse button\n\n#### \nfunc [Click](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L610) [¶](#Click)\n```\nfunc Click(args ...interface{})\n```\nClick click the mouse button\n\nrobotgo.Click(button string, double bool)\n\nExamples:\n```\nrobotgo.Click() // default is left button robotgo.Click(\"right\")\nrobotgo.Click(\"wheelLeft\")\n```\n#### \nfunc [CloseWindow](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1262) [¶](#CloseWindow)\n```\nfunc CloseWindow(args ...[int32](/builtin#int32))\n```\nCloseWindow close the window\n\n#### \nfunc [Convert](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L339) [¶](#Convert)\n```\nfunc Convert(opath, spath [string](/builtin#string), args ...[int](/builtin#int)) [string](/builtin#string)\n```\nConvert convert the bitmap\n\nrobotgo.Convert(opath, spath string, type int)\n\n#### \nfunc [CopyBitPB](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L359) [¶](#CopyBitPB)\n```\nfunc CopyBitPB(bitmap [C](/C).[MMBitmapRef](/C#MMBitmapRef)) [bool](/builtin#bool)\n```\nCopyBitPB copy bitmap to pasteboard\n\n#### \n\n func [CopyBitpb](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L369)\ndeprecated\n```\nfunc CopyBitpb(bitmap [C](/C).[MMBitmapRef](/C#MMBitmapRef)) [bool](/builtin#bool)\n```\nDeprecated: CopyBitpb copy bitmap to pasteboard, Wno-deprecated\n\nThis function will be removed in version v1.0.0\n\n#### \nfunc [CountBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L249) [¶](#CountBitmap)\n```\nfunc CountBitmap(bitmap, sbit [C](/C).[MMBitmapRef](/C#MMBitmapRef), args ...[float32](/builtin#float32)) [int](/builtin#int)\n```\nCountBitmap count of the bitmap\n\n#### \nfunc [CountColor](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L501) [¶](#CountColor)\n```\nfunc CountColor(color [CHex](#CHex), args ...interface{}) [int](/builtin#int)\n```\nCountColor count bitmap color\n\n#### \nfunc [CountColorCS](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L526) [¶](#CountColorCS)\n```\nfunc CountColorCS(color [CHex](#CHex), x, y, w, h [int](/builtin#int), args ...[float64](/builtin#float64)) [int](/builtin#int)\n```\nCountColorCS count bitmap color by CaptureScreen\n\n#### \nfunc [DecodeImg](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L22) [¶](#DecodeImg)\n```\nfunc DecodeImg(path [string](/builtin#string)) ([image](/image).[Image](/image#Image), [string](/builtin#string), [error](/builtin#error))\n```\nDecodeImg decode the image to image.Image and return\n\n#### \nfunc [DeepCopyBit](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L375) [¶](#DeepCopyBit)\n```\nfunc DeepCopyBit(bitmap [C](/C).[MMBitmapRef](/C#MMBitmapRef)) [C](/C).[MMBitmapRef](/C#MMBitmapRef)\n```\nDeepCopyBit deep copy bitmap\n\n#### \n\n func [Drag](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L491)\ndeprecated\n```\nfunc Drag(x, y [int](/builtin#int), args ...[string](/builtin#string))\n```\nDeprecated: use the DragSmooth(),\n\nDrag drag the mouse to (x, y),\nIt's not valid now, use the DragSmooth()\n\n#### \n\n func [DragMouse](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L479)\ndeprecated\n```\nfunc DragMouse(x, y [int](/builtin#int), args ...interface{})\n```\nDeprecated: use the DragSmooth(),\n\nDragMouse drag the mouse to (x, y),\nIt's same with the DragSmooth() now\n\n#### \nfunc [DragSmooth](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L508) [¶](#DragSmooth)\n```\nfunc DragSmooth(x, y [int](/builtin#int), args ...interface{})\n```\nDragSmooth drag the mouse like smooth to (x, y)\n\nExamples:\n```\nrobotgo.DragSmooth(10, 10)\n```\n#### \n\n func [End](https://github.com/go-vgo/robotgo/blob/v0.100.10/hook.go#L50)\ndeprecated\n```\nfunc End()\n```\nDeprecated: use the EventEnd(),\n\nEnd removes global event hook\n\n#### \nfunc [EventEnd](https://github.com/go-vgo/robotgo/blob/v0.100.10/hook.go#L35) [¶](#EventEnd)\n```\nfunc EventEnd()\n```\nEventEnd removes global event hook\n\n#### \nfunc [EventHook](https://github.com/go-vgo/robotgo/blob/v0.100.10/hook.go#L65) [¶](#EventHook)\n```\nfunc EventHook(when [uint8](/builtin#uint8), keysPressed [][string](/builtin#string), callback func([hook](/github.com/robotn/gohook).[Event](/github.com/robotn/gohook#Event)))\n```\nEventHook register gohook event\n\n#### \nfunc [EventProcess](https://github.com/go-vgo/robotgo/blob/v0.100.10/hook.go#L60) [¶](#EventProcess)\n```\nfunc EventProcess(events chan [hook](/github.com/robotn/gohook).[Event](/github.com/robotn/gohook#Event)) chan [bool](/builtin#bool)\n```\nEventProcess return go hook process\n\n#### \nfunc [EventStart](https://github.com/go-vgo/robotgo/blob/v0.100.10/hook.go#L30) [¶](#EventStart)\n```\nfunc EventStart() chan [hook](/github.com/robotn/gohook).[Event](/github.com/robotn/gohook#Event)\n```\nEventStart start global event hook return event channel\n\n#### \nfunc [FindBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L123) [¶](#FindBitmap)\n```\nfunc FindBitmap(bit [C](/C).[MMBitmapRef](/C#MMBitmapRef), args ...interface{}) ([int](/builtin#int), [int](/builtin#int))\n```\nFindBitmap find the bitmap's pos\n```\nrobotgo.FindBitmap(bitmap, source_bitamp C.MMBitmapRef, tolerance float64)\n\n|tolerance| should be in the range 0.0f - 1.0f, denoting how closely the colors in the bitmaps need to match, with 0 being exact and 1 being any.\n```\nThis method only automatically free the internal bitmap,\nuse `defer robotgo.FreeBitmap(bit)` to free the bitmap\n\n#### \nfunc [FindCBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L110) [¶](#FindCBitmap)\n```\nfunc FindCBitmap(bmp [CBitmap](#CBitmap), args ...interface{}) ([int](/builtin#int), [int](/builtin#int))\n```\nFindCBitmap find bitmap's pos by CBitmap\n\n#### \nfunc [FindColor](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L397) [¶](#FindColor)\n```\nfunc FindColor(color [CHex](#CHex), args ...interface{}) ([int](/builtin#int), [int](/builtin#int))\n```\nFindColor find bitmap color\n\nrobotgo.FindColor(color CHex, bitmap C.MMBitmapRef, tolerance float)\n\n#### \nfunc [FindColorCS](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L425) [¶](#FindColorCS)\n```\nfunc FindColorCS(color [CHex](#CHex), x, y, w, h [int](/builtin#int), args ...[float64](/builtin#float64)) ([int](/builtin#int), [int](/builtin#int))\n```\nFindColorCS findcolor by CaptureScreen\n\n#### \nfunc [FindIds](https://github.com/go-vgo/robotgo/blob/v0.100.10/ps.go#L60) [¶](#FindIds)\n```\nfunc FindIds(name [string](/builtin#string)) ([][int32](/builtin#int32), [error](/builtin#error))\n```\nFindIds finds the all processes named with a subset of \"name\" (case insensitive),\nreturn matched IDs.\n\n#### \nfunc [FindName](https://github.com/go-vgo/robotgo/blob/v0.100.10/ps.go#L48) [¶](#FindName)\n```\nfunc FindName(pid [int32](/builtin#int32)) ([string](/builtin#string), [error](/builtin#error))\n```\nFindName find the process name by the process id\n\n#### \nfunc [FindNames](https://github.com/go-vgo/robotgo/blob/v0.100.10/ps.go#L53) [¶](#FindNames)\n```\nfunc FindNames() ([][string](/builtin#string), [error](/builtin#error))\n```\nFindNames find the all process name\n\n#### \nfunc [FindPath](https://github.com/go-vgo/robotgo/blob/v0.100.10/ps.go#L65) [¶](#FindPath)\n```\nfunc FindPath(pid [int32](/builtin#int32)) ([string](/builtin#string), [error](/builtin#error))\n```\nFindPath find the process path by the process pid\n\n#### \nfunc [FindPic](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L154) [¶](#FindPic)\n```\nfunc FindPic(path [string](/builtin#string), args ...interface{}) ([int](/builtin#int), [int](/builtin#int))\n```\nFindPic finding the image by path\n```\nrobotgo.FindPic(path string, source_bitamp C.MMBitmapRef, tolerance float64)\n```\nThis method only automatically free the internal bitmap,\nuse `defer robotgo.FreeBitmap(bit)` to free the bitmap\n\n#### \nfunc [FreeBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L318) [¶](#FreeBitmap)\n```\nfunc FreeBitmap(bitmap [C](/C).[MMBitmapRef](/C#MMBitmapRef))\n```\nFreeBitmap free and dealloc the C bitmap\n\n#### \nfunc [FreeBitmapArr](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L53) [¶](#FreeBitmapArr)\n\nadded in v0.100.1\n```\nfunc FreeBitmapArr(bit ...[C](/C).[MMBitmapRef](/C#MMBitmapRef))\n```\nFreeBitmapArr free and dealloc the C bitmap array\n\n#### \nfunc [FreeMMPointArr](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L182) [¶](#FreeMMPointArr)\n\nadded in v0.93.0\n```\nfunc FreeMMPointArr(pointArray [C](/C).[MMPointArrayRef](/C#MMPointArrayRef))\n```\nFreeMMPointArr free MMPoint array\n\n#### \nfunc [GetActive](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1221) [¶](#GetActive)\n```\nfunc GetActive() [C](/C).[MData](/C#MData)\n```\nGetActive get the active window\n\n#### \n\n func [GetBHandle](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1318)\ndeprecated\n```\nfunc GetBHandle() [int](/builtin#int)\n```\n* [GetBHandle get the window handle, Wno-deprecated](#hdr-GetBHandle_get_the_window_handle__Wno_deprecated)\n\nDeprecated: use the GetHandle(),\n\n#### GetBHandle get the window handle, Wno-deprecated [¶](#hdr-GetBHandle_get_the_window_handle__Wno_deprecated)\n\nThis function will be removed in version v1.0.0\n\n#### \nfunc [GetBounds](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo_unix.go#L28) [¶](#GetBounds)\n```\nfunc GetBounds(pid [int32](/builtin#int32), args ...[int](/builtin#int)) ([int](/builtin#int), [int](/builtin#int), [int](/builtin#int), [int](/builtin#int))\n```\nGetBounds get the window bounds\n\n#### \nfunc [GetColor](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L381) [¶](#GetColor)\n```\nfunc GetColor(bitmap [C](/C).[MMBitmapRef](/C#MMBitmapRef), x, y [int](/builtin#int)) [C](/C).[MMRGBHex](/C#MMRGBHex)\n```\nGetColor get the bitmap color\n\n#### \nfunc [GetColors](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L388) [¶](#GetColors)\n```\nfunc GetColors(bitmap [C](/C).[MMBitmapRef](/C#MMBitmapRef), x, y [int](/builtin#int)) [string](/builtin#string)\n```\nGetColors get bitmap color retrun string\n\n#### \nfunc [GetHandPid](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1296) [¶](#GetHandPid)\n```\nfunc GetHandPid(pid [int32](/builtin#int32), args ...[int32](/builtin#int32)) [C](/C).[MData](/C#MData)\n```\nGetHandPid get handle mdata by pid\n\n#### \nfunc [GetHandle](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1306) [¶](#GetHandle)\n```\nfunc GetHandle() [int](/builtin#int)\n```\nGetHandle get the window handle\n\n#### \nfunc [GetImgSize](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L541) [¶](#GetImgSize)\n```\nfunc GetImgSize(imgPath [string](/builtin#string)) ([int](/builtin#int), [int](/builtin#int))\n```\nGetImgSize get the image size\n\n#### \nfunc [GetMouseColor](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L218) [¶](#GetMouseColor)\n```\nfunc GetMouseColor() [string](/builtin#string)\n```\nGetMouseColor get the mouse pos's color\n\n#### \nfunc [GetMousePos](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L585) [¶](#GetMousePos)\n```\nfunc GetMousePos() ([int](/builtin#int), [int](/builtin#int))\n```\nGetMousePos get the mouse's portion return x, y\n\n#### \nfunc [GetPID](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1355) [¶](#GetPID)\n```\nfunc GetPID() [int32](/builtin#int32)\n```\nGetPID get the process id return int32\n\n#### \nfunc [GetPixelColor](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L213) [¶](#GetPixelColor)\n```\nfunc GetPixelColor(x, y [int](/builtin#int)) [string](/builtin#string)\n```\nGetPixelColor get the pixel color return string\n\n#### \nfunc [GetPortion](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L325) [¶](#GetPortion)\n```\nfunc GetPortion(bit [C](/C).[MMBitmapRef](/C#MMBitmapRef), x, y, w, h [int](/builtin#int)) [C](/C).[MMBitmapRef](/C#MMBitmapRef)\n```\nGetPortion get bitmap portion\n\n#### \nfunc [GetPxColor](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L204) [¶](#GetPxColor)\n```\nfunc GetPxColor(x, y [int](/builtin#int)) [C](/C).[MMRGBHex](/C#MMRGBHex)\n```\nGetPxColor get the pixel color return C.MMRGBHex\n\n#### \nfunc [GetScaleSize](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L269) [¶](#GetScaleSize)\n```\nfunc GetScaleSize() ([int](/builtin#int), [int](/builtin#int))\n```\nGetScaleSize get the screen scale size\n\n#### \nfunc [GetScreenSize](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L241) [¶](#GetScreenSize)\n```\nfunc GetScreenSize() ([int](/builtin#int), [int](/builtin#int))\n```\nGetScreenSize get the screen size\n\n#### \nfunc [GetText](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L165) [¶](#GetText)\n```\nfunc GetText(imgPath [string](/builtin#string), args ...[string](/builtin#string)) ([string](/builtin#string), [error](/builtin#error))\n```\nGetText get the image text by tesseract ocr\n\nrobotgo.GetText(imgPath, lang string)\n\n#### \nfunc [GetTitle](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1340) [¶](#GetTitle)\n```\nfunc GetTitle(args ...[int32](/builtin#int32)) [string](/builtin#string)\n```\nGetTitle get the window title return string\n\nExamples:\n```\nfmt.Println(robotgo.GetTitle())\n\nids, _ := robotgo.FindIds()\nrobotgo.GetTitle(ids[0])\n```\n#### \nfunc [GetVersion](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L77) [¶](#GetVersion)\n```\nfunc GetVersion() [string](/builtin#string)\n```\nGetVersion get the robotgo version\n\n#### \nfunc [GetXDisplayName](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L360) [¶](#GetXDisplayName)\n```\nfunc GetXDisplayName() [string](/builtin#string)\n```\nGetXDisplayName get XDisplay name (Linux)\n\n#### \nfunc [GetXId](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo_unix.go#L118) [¶](#GetXId)\n```\nfunc GetXId(xu *[xgbutil](/github.com/robotn/xgbutil).[XUtil](/github.com/robotn/xgbutil#XUtil), pid [int32](/builtin#int32)) ([xproto](/github.com/robotn/xgb/xproto).[Window](/github.com/robotn/xgb/xproto#Window), [error](/builtin#error))\n```\nGetXId get the xid return window and error\n\n#### \nfunc [GetXidFromPid](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo_unix.go#L133) [¶](#GetXidFromPid)\n```\nfunc GetXidFromPid(xu *[xgbutil](/github.com/robotn/xgbutil).[XUtil](/github.com/robotn/xgbutil#XUtil), pid [int32](/builtin#int32)) ([xproto](/github.com/robotn/xgb/xproto).[Window](/github.com/robotn/xgb/xproto#Window), [error](/builtin#error))\n```\nGetXidFromPid get the xide from pid\n\n#### \nfunc [GoString](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L150) [¶](#GoString)\n```\nfunc GoString(char *[C](/C).[char](/C#char)) [string](/builtin#string)\n```\nGoString trans C.char to string\n\n#### \nfunc [Height](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L87) [¶](#Height)\n\nadded in v0.100.2\n```\nfunc Height(img [image](/image).[Image](/image#Image)) [int](/builtin#int)\n```\nHeight return the image.Image height\n\n#### \nfunc [HexToRgb](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L194) [¶](#HexToRgb)\n```\nfunc HexToRgb(hex [uint32](/builtin#uint32)) *[C](/C).[uint8_t](/C#uint8_t)\n```\nHexToRgb trans hex to rgb\n\n#### \nfunc [ImgSize](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L77) [¶](#ImgSize)\n\nadded in v0.100.2\n```\nfunc ImgSize(path [string](/builtin#string)) ([int](/builtin#int), [int](/builtin#int), [error](/builtin#error))\n```\nImgSize get the file image size\n\n#### \nfunc [Is64Bit](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1367) [¶](#Is64Bit)\n```\nfunc Is64Bit() [bool](/builtin#bool)\n```\nIs64Bit determine whether the sys is 64bit\n\n#### \nfunc [IsValid](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1208) [¶](#IsValid)\n```\nfunc IsValid() [bool](/builtin#bool)\n```\nIsValid valid the window\n\n#### \nfunc [KeyDown](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L969) [¶](#KeyDown)\n\nadded in v0.100.0\n```\nfunc KeyDown(key [string](/builtin#string))\n```\nKeyDown press down a key\n\n#### \nfunc [KeyPress](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L962) [¶](#KeyPress)\n\nadded in v0.100.0\n```\nfunc KeyPress(key [string](/builtin#string))\n```\nKeyPress press key string\n\n#### \nfunc [KeyTap](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L803) [¶](#KeyTap)\n```\nfunc KeyTap(tapKey [string](/builtin#string), args ...interface{}) [string](/builtin#string)\n```\nKeyTap tap the keyboard code;\n\nSee keys:\n```\nhttps://github.com/go-vgo/robotgo/blob/master/docs/keys.md\n```\nExamples:\n```\nrobotgo.KeySleep = 100 // 100 millisecond robotgo.KeyTap(\"a\")\nrobotgo.KeyTap(\"i\", \"alt\", \"command\")\n\narr := []string{\"alt\", \"command\"}\nrobotgo.KeyTap(\"i\", arr)\n```\n#### \nfunc [KeyToggle](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L902) [¶](#KeyToggle)\n```\nfunc KeyToggle(key [string](/builtin#string), args ...[string](/builtin#string)) [string](/builtin#string)\n```\nKeyToggle toggle the keyboard, if there not have args default is \"down\"\n\nSee keys:\n```\nhttps://github.com/go-vgo/robotgo/blob/master/docs/keys.md\n```\nExamples:\n```\nrobotgo.KeyToggle(\"a\")\nrobotgo.KeyToggle(\"a\", \"up\")\n\nrobotgo.KeyToggle(\"a\", \"up\", \"alt\", \"cmd\")\n```\n#### \nfunc [KeyUp](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L974) [¶](#KeyUp)\n\nadded in v0.100.0\n```\nfunc KeyUp(key [string](/builtin#string))\n```\nKeyUp press up a key\n\n#### \nfunc [Kill](https://github.com/go-vgo/robotgo/blob/v0.100.10/ps.go#L75) [¶](#Kill)\n```\nfunc Kill(pid [int32](/builtin#int32)) [error](/builtin#error)\n```\nKill kill the process by PID\n\n#### \nfunc [MaxWindow](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1245) [¶](#MaxWindow)\n```\nfunc MaxWindow(pid [int32](/builtin#int32), args ...interface{})\n```\nMaxWindow set the window max\n\n#### \nfunc [MicroSleep](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L145) [¶](#MicroSleep)\n```\nfunc MicroSleep(tm [float64](/builtin#float64))\n```\nMicroSleep time C.microsleep(tm)\n\n#### \nfunc [MilliSleep](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L135) [¶](#MilliSleep)\n```\nfunc MilliSleep(tm [int](/builtin#int))\n```\nMilliSleep sleep tm milli second\n\n#### \nfunc [MinWindow](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1228) [¶](#MinWindow)\n```\nfunc MinWindow(pid [int32](/builtin#int32), args ...interface{})\n```\nMinWindow set the window min\n\n#### \n\n func [MouseClick](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L598)\ndeprecated\n```\nfunc MouseClick(args ...interface{})\n```\n* [MouseClick click the mouse](#hdr-MouseClick_click_the_mouse)\n\nDeprecated: use the Click(),\n\n#### MouseClick click the mouse [¶](#hdr-MouseClick_click_the_mouse)\n\nrobotgo.MouseClick(button string, double bool)\n\n#### \n\n func [MouseToggle](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L680)\ndeprecated\n```\nfunc MouseToggle(togKey [string](/builtin#string), args ...interface{}) [int](/builtin#int)\n```\n* [MouseToggle toggle the mouse](#hdr-MouseToggle_toggle_the_mouse)\n\nDeprecated: use the Toggle(),\n\n#### MouseToggle toggle the mouse [¶](#hdr-MouseToggle_toggle_the_mouse)\n\nExamples:\n```\nrobotgo.MouseToggle(\"down\", \"right\")\nrobotgo.MouseToggle(\"up\", \"right\")\n```\n#### \nfunc [Move](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L462) [¶](#Move)\n```\nfunc Move(x, y [int](/builtin#int))\n```\nMove move the mouse to (x, y)\n\nExamples:\n```\nrobotgo.MouseSleep = 100 // 100 millisecond robotgo.Move(10, 10)\n```\n#### \nfunc [MoveArgs](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L565) [¶](#MoveArgs)\n```\nfunc MoveArgs(x, y [int](/builtin#int)) ([int](/builtin#int), [int](/builtin#int))\n```\nMoveArgs get the mouse relative args\n\n#### \nfunc [MoveClick](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L635) [¶](#MoveClick)\n```\nfunc MoveClick(x, y [int](/builtin#int), args ...interface{})\n```\nMoveClick move and click the mouse\n\nrobotgo.MoveClick(x, y int, button string, double bool)\n\nExamples:\n```\nrobotgo.MouseSleep = 100 robotgo.MoveClick(10, 10)\n```\n#### \n\n func [MoveMouse](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L453)\ndeprecated\n```\nfunc MoveMouse(x, y [int](/builtin#int))\n```\nDeprecated: use the Move(),\n\nMoveMouse move the mouse\n\n#### \n\n func [MoveMouseSmooth](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L519)\ndeprecated\n```\nfunc MoveMouseSmooth(x, y [int](/builtin#int), args ...interface{}) [bool](/builtin#bool)\n```\nDeprecated: use the MoveSmooth(),\n\nMoveMouseSmooth move the mouse smooth,\nmoves mouse to x, y human like, with the mouse button up.\n\n#### \nfunc [MoveRelative](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L574) [¶](#MoveRelative)\n```\nfunc MoveRelative(x, y [int](/builtin#int))\n```\nMoveRelative move mouse with relative\n\n#### \nfunc [MoveSmooth](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L531) [¶](#MoveSmooth)\n```\nfunc MoveSmooth(x, y [int](/builtin#int), args ...interface{}) [bool](/builtin#bool)\n```\nMoveSmooth move the mouse smooth,\nmoves mouse to x, y human like, with the mouse button up.\n\nrobotgo.MoveSmooth(x, y int, low, high float64, mouseDelay int)\n\nExamples:\n```\nrobotgo.MoveSmooth(10, 10)\nrobotgo.MoveSmooth(10, 10, 1.0, 2.0)\n```\n#### \nfunc [MoveSmoothRelative](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L579) [¶](#MoveSmoothRelative)\n```\nfunc MoveSmoothRelative(x, y [int](/builtin#int), args ...interface{})\n```\nMoveSmoothRelative move mouse smooth with relative\n\n#### \nfunc [MovesClick](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L644) [¶](#MovesClick)\n```\nfunc MovesClick(x, y [int](/builtin#int), args ...interface{})\n```\nMovesClick move smooth and click the mouse\n\nuse the `robotgo.MouseSleep = 100`\n\n#### \n\n func [Mul](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L416)\ndeprecated\n```\nfunc Mul(x [int](/builtin#int)) [int](/builtin#int)\n```\nDeprecated: use the ScaledF(),\n\nMul mul the scale, drop\n\n#### \nfunc [OpenBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L278) [¶](#OpenBitmap)\n```\nfunc OpenBitmap(gpath [string](/builtin#string), args ...[int](/builtin#int)) [C](/C).[MMBitmapRef](/C#MMBitmapRef)\n```\nOpenBitmap open the bitmap return C.MMBitmapRef\n\nrobotgo.OpenBitmap(path string, type int)\n\n#### \nfunc [OpenImg](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L27) [¶](#OpenImg)\n```\nfunc OpenImg(path [string](/builtin#string)) ([][byte](/builtin#byte), [error](/builtin#error))\n```\nOpenImg open the image return []byte\n\n#### \nfunc [PadHex](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L185) [¶](#PadHex)\n```\nfunc PadHex(hex [C](/C).[MMRGBHex](/C#MMRGBHex)) [string](/builtin#string)\n```\nPadHex trans C.MMRGBHex to string\n\n#### \nfunc [PasteStr](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1083) [¶](#PasteStr)\n```\nfunc PasteStr(str [string](/builtin#string)) [string](/builtin#string)\n```\nPasteStr paste a string, support UTF-8,\nwrite the string to clipboard and tap `cmd + v`\n\n#### \nfunc [PidExists](https://github.com/go-vgo/robotgo/blob/v0.100.10/ps.go#L27) [¶](#PidExists)\n```\nfunc PidExists(pid [int32](/builtin#int32)) ([bool](/builtin#bool), [error](/builtin#error))\n```\nPidExists determine whether the process exists\n\n#### \nfunc [Pids](https://github.com/go-vgo/robotgo/blob/v0.100.10/ps.go#L22) [¶](#Pids)\n```\nfunc Pids() ([][int32](/builtin#int32), [error](/builtin#error))\n```\nPids get the all process id\n\n#### \nfunc [PointInBounds](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L266) [¶](#PointInBounds)\n```\nfunc PointInBounds(bitmap [C](/C).[MMBitmapRef](/C#MMBitmapRef), x, y [int](/builtin#int)) [bool](/builtin#bool)\n```\nPointInBounds bitmap point in bounds\n\n#### \nfunc [Read](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L32) [¶](#Read)\n\nadded in v0.100.0\n```\nfunc Read(path [string](/builtin#string)) ([image](/image).[Image](/image#Image), [error](/builtin#error))\n```\nRead read the file return image.Image\n\n#### \nfunc [ReadAll](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L979) [¶](#ReadAll)\n```\nfunc ReadAll() ([string](/builtin#string), [error](/builtin#error))\n```\nReadAll read string from clipboard\n\n#### \nfunc [ReadBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L352) [¶](#ReadBitmap)\n```\nfunc ReadBitmap(bitmap [C](/C).[MMBitmapRef](/C#MMBitmapRef)) [bool](/builtin#bool)\n```\nReadBitmap returns false and sets error if |bitmap| is NULL\n\n#### \nfunc [RgbToHex](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L199) [¶](#RgbToHex)\n```\nfunc RgbToHex(r, g, b [uint8](/builtin#uint8)) [C](/C).[uint32_t](/C#uint32_t)\n```\nRgbToHex trans rgb to hex\n\n#### \nfunc [Run](https://github.com/go-vgo/robotgo/blob/v0.100.10/ps.go#L70) [¶](#Run)\n\nadded in v0.100.9\n```\nfunc Run(path [string](/builtin#string)) ([][byte](/builtin#byte), [error](/builtin#error))\n```\nRun run a cmd shell\n\n#### \nfunc [Save](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L37) [¶](#Save)\n\nadded in v0.100.0\n```\nfunc Save(img [image](/image).[Image](/image#Image), path [string](/builtin#string)) [error](/builtin#error)\n```\nSave create a image file with the image.Image\n\n#### \nfunc [SaveBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L311) [¶](#SaveBitmap)\n```\nfunc SaveBitmap(bitmap [C](/C).[MMBitmapRef](/C#MMBitmapRef), gpath [string](/builtin#string), args ...[int](/builtin#int)) [string](/builtin#string)\n```\nSaveBitmap save the bitmap to image\n\nrobotgo.SaveBimap(bitmap C.MMBitmapRef, path string, type int)\n\n#### \nfunc [SaveCapture](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L44) [¶](#SaveCapture)\n```\nfunc SaveCapture(spath [string](/builtin#string), args ...[int](/builtin#int)) [string](/builtin#string)\n```\nSaveCapture capture screen and save\n\n#### \nfunc [SaveImg](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L42) [¶](#SaveImg)\n```\nfunc SaveImg(b [][byte](/builtin#byte), path [string](/builtin#string)) [error](/builtin#error)\n```\nSaveImg save the image by []byte\n\n#### \nfunc [SaveJpeg](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L52) [¶](#SaveJpeg)\n\nadded in v0.100.0\n```\nfunc SaveJpeg(img [image](/image).[Image](/image#Image), path [string](/builtin#string)) [error](/builtin#error)\n```\nSaveJpeg save the image by image.Image\n\n#### \nfunc [SavePng](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L47) [¶](#SavePng)\n\nadded in v0.92.0\n```\nfunc SavePng(img [image](/image).[Image](/image#Image), path [string](/builtin#string)) [error](/builtin#error)\n```\nSavePng save the image by image.Image\n\n#### \n\n func [Scale](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L385)\ndeprecated\n```\nfunc Scale() [int](/builtin#int)\n```\nDeprecated: use the ScaledF(),\n\nScale get the screen scale (only windows old), drop\n\n#### \n\n func [Scale0](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L409)\ndeprecated\n\nadded in v0.100.1\n```\nfunc Scale0() [int](/builtin#int)\n```\nDeprecated: use the ScaledF(),\n\nScale0 return ScaleX() / 0.96, drop\n\n#### \nfunc [ScaleF](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo_mac.go#L17) [¶](#ScaleF)\n\nadded in v0.100.8\n```\nfunc ScaleF() [float64](/builtin#float64)\n```\nScaleF get the system scale val\n\n#### \n\n func [ScaleX](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L371)\ndeprecated\n```\nfunc ScaleX() [int](/builtin#int)\n```\nDeprecated: use the ScaledF(),\n\nScaleX get the primary display horizontal DPI scale factor, drop\n\n#### \n\n func [ScaleY](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L378)\ndeprecated\n```\nfunc ScaleY() [int](/builtin#int)\n```\nDeprecated: use the ScaledF(),\n\nScaleY get primary display vertical DPI scale factor, drop\n\n#### \nfunc [Scaled](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L230) [¶](#Scaled)\n```\nfunc Scaled(x [int](/builtin#int)) [int](/builtin#int)\n```\nScaled get the screen scaled size\n\n#### \nfunc [Scaled0](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L236) [¶](#Scaled0)\n\nadded in v0.100.8\n```\nfunc Scaled0(x [int](/builtin#int), f [float64](/builtin#float64)) [int](/builtin#int)\n```\nScaled0 return int(x * f)\n\n#### \nfunc [Scroll](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L717) [¶](#Scroll)\n```\nfunc Scroll(x, y [int](/builtin#int), args ...[int](/builtin#int))\n```\nScroll scroll the mouse to (x, y)\n\nrobotgo.Scroll(x, y, msDelay int)\n\nExamples:\n```\nrobotgo.Scroll(10, 10)\n```\n#### \n\n func [ScrollMouse](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L702)\ndeprecated\n```\nfunc ScrollMouse(x [int](/builtin#int), direction [string](/builtin#string))\n```\nDeprecated: use the Scroll(),\n\nScrollMouse scroll the mouse to (x, \"up\")\n\nExamples:\n```\nrobotgo.ScrollMouse(10, \"down\")\nrobotgo.ScrollMouse(10, \"up\")\n```\n#### \nfunc [ScrollRelative](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L769) [¶](#ScrollRelative)\n\nadded in v0.100.1\n```\nfunc ScrollRelative(x, y [int](/builtin#int), args ...[int](/builtin#int))\n```\nScrollRelative scroll mouse with relative\n\nExamples:\n```\nrobotgo.ScrollRelative(10, 10)\n```\n#### \nfunc [ScrollSmooth](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L739) [¶](#ScrollSmooth)\n\nadded in v0.100.7\n```\nfunc ScrollSmooth(to [int](/builtin#int), args ...[int](/builtin#int))\n```\nScrollSmooth scroll the mouse smooth,\ndefault scroll 5 times and sleep 100 millisecond\n\nrobotgo.ScrollSmooth(toy, num, sleep, tox)\n\nExamples:\n```\nrobotgo.ScrollSmooth(-10)\nrobotgo.ScrollSmooth(-10, 6, 200, -10)\n```\n#### \nfunc [SetActive](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1216) [¶](#SetActive)\n```\nfunc SetActive(win [C](/C).[MData](/C#MData))\n```\nSetActive set the window active\n\n#### \nfunc [SetDelay](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1147) [¶](#SetDelay)\n```\nfunc SetDelay(d ...[int](/builtin#int))\n```\nSetDelay set the key and mouse delay\n\n#### \nfunc [SetHandle](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1280) [¶](#SetHandle)\n```\nfunc SetHandle(hwnd [int](/builtin#int))\n```\nSetHandle set the window handle\n\n#### \nfunc [SetHandlePid](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1286) [¶](#SetHandlePid)\n```\nfunc SetHandlePid(pid [int32](/builtin#int32), args ...[int32](/builtin#int32))\n```\nSetHandlePid set the window handle by pid\n\n#### \nfunc [SetKeyDelay](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1132) [¶](#SetKeyDelay)\n```\nfunc SetKeyDelay(delay [int](/builtin#int))\n```\nSetKeyDelay set keyboard delay\n\n#### \n\n func [SetKeyboardDelay](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1141)\ndeprecated\n```\nfunc SetKeyboardDelay(delay [int](/builtin#int))\n```\nDeprecated: use the SetKeyDelay(),\n\nSetKeyboardDelay set keyboard delay, Wno-deprecated,\n\nThis function will be removed in version v1.0.0\n\n#### \nfunc [SetMouseDelay](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L775) [¶](#SetMouseDelay)\n```\nfunc SetMouseDelay(delay [int](/builtin#int))\n```\nSetMouseDelay set mouse delay\n\n#### \nfunc [SetXDisplayName](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L349) [¶](#SetXDisplayName)\n```\nfunc SetXDisplayName(name [string](/builtin#string)) [string](/builtin#string)\n```\nSetXDisplayName set XDisplay name (Linux)\n\n#### \nfunc [ShowAlert](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1173) [¶](#ShowAlert)\n```\nfunc ShowAlert(title, msg [string](/builtin#string), args ...[string](/builtin#string)) [bool](/builtin#bool)\n```\nShowAlert show a alert window Displays alert with the attributes.\nIf cancel button is not given, only the default button is displayed\n\nExamples:\n```\nrobotgo.ShowAlert(\"hi\", \"window\", \"ok\", \"cancel\")\n```\n#### \nfunc [Sleep](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L140) [¶](#Sleep)\n```\nfunc Sleep(tm [int](/builtin#int))\n```\nSleep time.Sleep tm second\n\n#### \n\n func [Start](https://github.com/go-vgo/robotgo/blob/v0.100.10/hook.go#L43)\ndeprecated\n```\nfunc Start() chan [hook](/github.com/robotn/gohook).[Event](/github.com/robotn/gohook#Event)\n```\nDeprecated: use the EventStart(),\n\nStart start global event hook return event channel\n\n#### \nfunc [StopEvent](https://github.com/go-vgo/robotgo/blob/v0.100.10/hook.go#L55) [¶](#StopEvent)\n```\nfunc StopEvent()\n```\nStopEvent stop event listener, use by AddEvent()\n\n#### \nfunc [StrToImg](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L67) [¶](#StrToImg)\n\nadded in v0.94.0\n```\nfunc StrToImg(data [string](/builtin#string)) ([image](/image).[Image](/image#Image), [error](/builtin#error))\n```\nStrToImg convert base64 string to image.Image\n\n#### \nfunc [SysScale](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L224) [¶](#SysScale)\n```\nfunc SysScale() [float64](/builtin#float64)\n```\nSysScale get the sys scale\n\n#### \nfunc [ToBitmapBytes](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L79) [¶](#ToBitmapBytes)\n```\nfunc ToBitmapBytes(bit [C](/C).[MMBitmapRef](/C#MMBitmapRef)) [][byte](/builtin#byte)\n```\nToBitmapBytes saves Bitmap to bitmap format in bytes\n\n#### \nfunc [ToByteImg](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L57) [¶](#ToByteImg)\n\nadded in v0.94.0\n```\nfunc ToByteImg(img [image](/image).[Image](/image#Image), fm ...[string](/builtin#string)) [][byte](/builtin#byte)\n```\nToByteImg convert image.Image to []byte\n\n#### \nfunc [ToCBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L60) [¶](#ToCBitmap)\n```\nfunc ToCBitmap(bit [Bitmap](#Bitmap)) [C](/C).[MMBitmapRef](/C#MMBitmapRef)\n```\nToCBitmap trans Bitmap to C.MMBitmapRef\n\n#### \nfunc [ToImage](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L338) [¶](#ToImage)\n\nadded in v0.92.0\n```\nfunc ToImage(bit [C](/C).[MMBitmapRef](/C#MMBitmapRef)) [image](/image).[Image](/image#Image)\n```\nToImage convert C.MMBitmapRef to standard image.Image\n\n#### \nfunc [ToMMBitmapRef](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L74) [¶](#ToMMBitmapRef)\n```\nfunc ToMMBitmapRef(bit [CBitmap](#CBitmap)) [C](/C).[MMBitmapRef](/C#MMBitmapRef)\n```\nToMMBitmapRef trans CBitmap to C.MMBitmapRef\n\n#### \nfunc [ToMMRGBHex](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L164) [¶](#ToMMRGBHex)\n```\nfunc ToMMRGBHex(hex [CHex](#CHex)) [C](/C).[MMRGBHex](/C#MMRGBHex)\n```\nToMMRGBHex trans CHex to C.MMRGBHex\n\n#### \nfunc [ToRGBA](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L343) [¶](#ToRGBA)\n\nadded in v0.99.9\n```\nfunc ToRGBA(bit [C](/C).[MMBitmapRef](/C#MMBitmapRef)) *[image](/image).[RGBA](/image#RGBA)\n```\nToRGBA convert C.MMBitmapRef to standard image.RGBA\n\n#### \nfunc [ToRGBAGo](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L137) [¶](#ToRGBAGo)\n\nadded in v0.99.9\n```\nfunc ToRGBAGo(bmp1 [Bitmap](#Bitmap)) *[image](/image).[RGBA](/image#RGBA)\n```\nToRGBAGo convert Bitmap to standard image.RGBA\n\n#### \nfunc [ToStringImg](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L62) [¶](#ToStringImg)\n\nadded in v0.94.0\n```\nfunc ToStringImg(img [image](/image).[Image](/image#Image), fm ...[string](/builtin#string)) [string](/builtin#string)\n```\nToStringImg convert image.Image to string\n\n#### \nfunc [ToUC](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1008) [¶](#ToUC)\n\nadded in v0.99.1\n```\nfunc ToUC(text [string](/builtin#string)) [][string](/builtin#string)\n```\nToUC trans string to unicode []string\n\n#### \nfunc [ToUint8p](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L123) [¶](#ToUint8p)\n\nadded in v0.99.9\n```\nfunc ToUint8p(dst [][uint8](/builtin#uint8)) *[uint8](/builtin#uint8)\n```\nToUint8p convert the []uint8 to uint8 pointer\n\n#### \nfunc [TocharBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L98) [¶](#TocharBitmap)\n```\nfunc TocharBitmap(bit [C](/C).[MMBitmapRef](/C#MMBitmapRef)) *[C](/C).[char](/C#char)\n```\nTocharBitmap tostring bitmap to C.char\n\n#### \nfunc [Toggle](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L657) [¶](#Toggle)\n\nadded in v0.100.7\n```\nfunc Toggle(key ...[string](/builtin#string)) [int](/builtin#int)\n```\nToggle toggle the mouse, support button:\n```\n\t\"left\", \"center\", \"right\",\n \"wheelDown\", \"wheelUp\", \"wheelLeft\", \"wheelRight\"\n```\nExamples:\n```\nrobotgo.Toggle(\"left\") // default is down robotgo.Toggle(\"left\", \"up\")\n```\n#### \nfunc [TostringBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L92) [¶](#TostringBitmap)\n```\nfunc TostringBitmap(bit [C](/C).[MMBitmapRef](/C#MMBitmapRef)) [string](/builtin#string)\n```\nTostringBitmap tostring bitmap to string\n\n#### \nfunc [Try](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L125) [¶](#Try)\n```\nfunc Try(fun func(), handler func(interface{}))\n```\nTry handler(err)\n\n#### \nfunc [TypeStr](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1042) [¶](#TypeStr)\n```\nfunc TypeStr(str [string](/builtin#string), args ...[float64](/builtin#float64))\n```\nTypeStr send a string, support UTF-8\n\nrobotgo.TypeStr(string: The string to send, float64: microsleep time, x11 option)\n\nExamples:\n```\nrobotgo.TypeStr(\"abc@123, hi, こんにちは\")\n```\n#### \nfunc [TypeStrDelay](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1116) [¶](#TypeStrDelay)\n```\nfunc TypeStrDelay(str [string](/builtin#string), delay [int](/builtin#int))\n```\nTypeStrDelay type string delayed\n\n#### \n\n func [TypeString](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1102)\ndeprecated\n```\nfunc TypeString(str [string](/builtin#string), delay ...[int](/builtin#int))\n```\nDeprecated: use the TypeStr(),\n\nTypeString send a string, support unicode(no linux support)\nTypeStr(string: The string to send), Wno-deprecated\n\nThis function will be removed in version v1.0.0\n\n#### \n\n func [TypeStringDelayed](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1126)\ndeprecated\n```\nfunc TypeStringDelayed(str [string](/builtin#string), delay [int](/builtin#int))\n```\n* [TypeStringDelayed type string delayed, Wno-deprecated](#hdr-TypeStringDelayed_type_string_delayed__Wno_deprecated)\n\nDeprecated: use the TypeStr(),\n\n#### TypeStringDelayed type string delayed, Wno-deprecated [¶](#hdr-TypeStringDelayed_type_string_delayed__Wno_deprecated)\n\nThis function will be removed in version v1.0.0\n\n#### \nfunc [U32ToHex](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L175) [¶](#U32ToHex)\n```\nfunc U32ToHex(hex [C](/C).[uint32_t](/C#uint32_t)) [C](/C).[MMRGBHex](/C#MMRGBHex)\n```\nU32ToHex trans C.uint32_t to C.MMRGBHex\n\n#### \nfunc [U8ToHex](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L180) [¶](#U8ToHex)\n```\nfunc U8ToHex(hex *[C](/C).[uint8_t](/C#uint8_t)) [C](/C).[MMRGBHex](/C#MMRGBHex)\n```\nU8ToHex trans *C.uint8_t to C.MMRGBHex\n\n#### \nfunc [UnicodeType](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L1002) [¶](#UnicodeType)\n```\nfunc UnicodeType(str [uint32](/builtin#uint32))\n```\nUnicodeType tap uint32 unicode\n\n#### \nfunc [Width](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L82) [¶](#Width)\n\nadded in v0.100.2\n```\nfunc Width(img [image](/image).[Image](/image#Image)) [int](/builtin#int)\n```\nWidth return the image.Image width\n\n#### \nfunc [WriteAll](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L984) [¶](#WriteAll)\n```\nfunc WriteAll(text [string](/builtin#string)) [error](/builtin#error)\n```\nWriteAll write string to clipboard\n\n### Types [¶](#pkg-types)\n\n#### \ntype [Bitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L98) [¶](#Bitmap)\n```\ntype Bitmap struct {\n ImgBuf *[uint8](/builtin#uint8)\n Width, Height [int](/builtin#int)\n\n Bytewidth [int](/builtin#int)\n BitsPixel [uint8](/builtin#uint8)\n BytesPerPixel [uint8](/builtin#uint8)\n}\n```\nBitmap is Bitmap struct\n\n#### \nfunc [GoCaptureScreen](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L302) [¶](#GoCaptureScreen)\n```\nfunc GoCaptureScreen(args ...[int](/builtin#int)) [Bitmap](#Bitmap)\n```\nGoCaptureScreen capture the screen and return bitmap(go struct)\n\n#### \nfunc [ImgToBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L107) [¶](#ImgToBitmap)\n\nadded in v0.99.9\n```\nfunc ImgToBitmap(m [image](/image).[Image](/image#Image)) (bit [Bitmap](#Bitmap))\n```\nImgToBitmap convert the standard image.Image to Bitmap\n\n#### \nfunc [RGBAToBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/img.go#L92) [¶](#RGBAToBitmap)\n\nadded in v0.99.9\n```\nfunc RGBAToBitmap(r1 *[image](/image).[RGBA](/image#RGBA)) (bit [Bitmap](#Bitmap))\n```\nRGBAToBitmap convert the standard image.RGBA to Bitmap\n\n#### \nfunc [ToBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L324) [¶](#ToBitmap)\n```\nfunc ToBitmap(bit [C](/C).[MMBitmapRef](/C#MMBitmapRef)) [Bitmap](#Bitmap)\n```\nToBitmap trans C.MMBitmapRef to Bitmap\n\n#### \ntype [CBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L94) [¶](#CBitmap)\n```\ntype CBitmap [C](/C).[MMBitmapRef](/C#MMBitmapRef)\n```\nCBitmap define CBitmap as C.MMBitmapRef type\n\n#### \ntype [CHex](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L92) [¶](#CHex)\n```\ntype CHex [C](/C).[MMRGBHex](/C#MMRGBHex)\n```\nCHex define CHex as c rgb Hex type (C.MMRGBHex)\n\n#### \nfunc [UintToHex](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L169) [¶](#UintToHex)\n```\nfunc UintToHex(u [uint32](/builtin#uint32)) [CHex](#CHex)\n```\nUintToHex trans uint32 to robotgo.CHex\n\n#### \ntype [Map](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L90) [¶](#Map)\n```\ntype Map map[[string](/builtin#string)]interface{}\n```\nMap a map[string]interface{}\n\n#### \ntype [Nps](https://github.com/go-vgo/robotgo/blob/v0.100.10/ps.go#L16) [¶](#Nps)\n```\ntype Nps struct {\n Pid [int32](/builtin#int32)\n Name [string](/builtin#string)\n}\n```\nNps process struct\n\n#### \nfunc [Process](https://github.com/go-vgo/robotgo/blob/v0.100.10/ps.go#L32) [¶](#Process)\n```\nfunc Process() ([][Nps](#Nps), [error](/builtin#error))\n```\nProcess get the all process struct\n\n#### \ntype [Point](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L108) [¶](#Point)\n\nadded in v0.100.3\n```\ntype Point struct {\n X [int](/builtin#int)\n Y [int](/builtin#int)\n}\n```\nPoint is point struct\n\n#### \nfunc [FindAllBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L194) [¶](#FindAllBitmap)\n\nadded in v0.100.9\n```\nfunc FindAllBitmap(bit [C](/C).[MMBitmapRef](/C#MMBitmapRef), args ...interface{}) (posArr [][Point](#Point))\n```\nFindAllBitmap find the all bitmap\n\n#### \nfunc [FindAllColor](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L447) [¶](#FindAllColor)\n\nadded in v0.100.9\n```\nfunc FindAllColor(color [CHex](#CHex), args ...interface{}) (posArr [][Point](#Point))\n```\nFindAllColor find the all color\n\n#### \n\n func [FindEveryBitmap](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L189)\ndeprecated\n```\nfunc FindEveryBitmap(bit [C](/C).[MMBitmapRef](/C#MMBitmapRef), args ...interface{}) [][Point](#Point)\n```\nDeprecated: use the FindAllBitmap(),\n\nFindEveryBitmap find the every bitmap, same with the FindAllBitmap()\n\n#### \n\n func [FindEveryColor](https://github.com/go-vgo/robotgo/blob/v0.100.10/bitmap.go#L442)\ndeprecated\n\nadded in v0.93.0\n```\nfunc FindEveryColor(color [CHex](#CHex), args ...interface{}) [][Point](#Point)\n```\nDeprecated: use the FindAllColor(),\n\nFindEveryColor find the every color, same with the FindAllColor()\n\n#### \ntype [Rect](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L119) [¶](#Rect)\n\nadded in v0.100.3\n```\ntype Rect struct {\n [Point](#Point)\n [Size](#Size)\n}\n```\nRect is rect structure\n\n#### \nfunc [GetScreenRect](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L248) [¶](#GetScreenRect)\n\nadded in v0.100.3\n```\nfunc GetScreenRect(displayId ...[int](/builtin#int)) [Rect](#Rect)\n```\nGetScreenRect get the screen rect (x, y, w, h)\n\n#### \ntype [Size](https://github.com/go-vgo/robotgo/blob/v0.100.10/robotgo.go#L114) [¶](#Size)\n\nadded in v0.100.3\n```\ntype Size struct {\n W, H [int](/builtin#int)\n}\n```\nSize is size structure"}}},{"rowIdx":59176,"cells":{"project":{"kind":"string","value":"ReFrame-HPC"},"source":{"kind":"string","value":"readthedoc"},"language":{"kind":"string","value":"Python"},"content":{"kind":"string","value":"ReFrame 4.1.1 documentation\n\n[ReFrame](#)\n\n**ReFrame 4.1.1**\n\n---\n\nWelcome to ReFrame[¶](#welcome-to-reframe)\n===\n\nReFrame is a powerful framework for writing system regression tests and benchmarks, specifically targeted to HPC systems.\nThe goal of the framework is to abstract away the complexity of the interactions with the system, separating the logic of a test from the low-level details, which pertain to the system configuration and setup.\nThis allows users to write portable tests in a declarative way that describes only the test’s functionality.\n\nTests in ReFrame are simple Python classes that specify the basic variables and parameters of the test.\nReFrame offers an intuitive and very powerful syntax that allows users to create test libraries, test factories, as well as complete test workflows using other tests as fixtures.\nReFrame will load the tests and send them down a well-defined pipeline that will execute them in parallel.\nThe stages of this pipeline take care of all the system interaction details, such as programming environment switching, compilation, job submission, job status query, sanity checking and performance assessment.\n\nReFrame also offers a high-level and flexible abstraction for writing sanity and performance checks for your regression tests, without having to care about the details of parsing output files, searching for patterns and testing against reference values for different systems.\n\nFinally, ReFrame offers a powerful and efficient runtime for running and managing the execution of tests, as well as integration with common logging facilities, where ReFrame can send live data from currently running performance tests.\n\nPublications[¶](#publications)\n---\n\n* Slides [[pdf](https://drive.google.com/file/d/1vmaWyRHgtq3DrYhSCVBzR8U5ErKbxGNf/view?usp=sharing)] @ [7th EasyBuild User Meeting 2022](https://easybuild.io/eum22/).\n* Slides [[pdf](https://drive.google.com/file/d/1kNZu1QNBDDsbKarzwNWYjTGKgOukg-96/view?usp=sharing)] @ [6th EasyBuild User Meeting 2021](https://easybuild.io/eum21/).\n* Slides [[pdf](https://drive.google.com/open?id=1W7R5lfRkXvBpVDSZ7dVBadk_d3K4dFrS)] @ [5th EasyBuild User Meeting 2020](https://github.com/easybuilders/easybuild/wiki/5th-EasyBuild-User-Meeting).\n* Slides [[pdf](https://drive.google.com/open?id=1Z3faPh9OSSXvlLHL07co3MRRn443dYsY)] @ [HPC System Testing BoF](https://sc19.supercomputing.org/session/?sess=sess324), SC’19.\n* Slides [[pdf](https://drive.google.com/open?id=1JOFqY3ejbR1X5kTn_IZyp1GlCd2ZZS58)] @ [HUST 2019](https://sc19.supercomputing.org/session/?sess=sess116), SC’19.\n* Slides [[pdf](https://drive.google.com/open?id=1iwg1I48LVaWhhZCZIYPJSi3hdFLRcuhi)] @ [HPC Knowledge Meeting ‘19](https://hpckp.org/).\n* Slides [[pdf](https://fosdem.org/2019/schedule/event/reframe/attachments/slides/3226/export/events/attachments/reframe/slides/3226/FOSDEM_2019.pdf)] & [Talk](https://fosdem.org/2019/schedule/event/reframe/) @ [FOSDEM’19](https://fosdem.org/2019/).\n* Slides [[pdf](https://indico.cism.ucl.ac.be/event/4/contributions/24/attachments/30/62/ReFrame_EUM_2019.pdf)] @ [4th EasyBuild User Meeting](https://github.com/easybuilders/easybuild/wiki/4th-EasyBuild-User-Meeting).\n* Slides [[pdf](https://drive.google.com/open?id=1bSykDrl1e2gPflf4jFJ8kfe_SZAtrJ_Q)] @ [HUST 2018](https://sc18.supercomputing.org/), SC’18.\n* Slides [[pdf](https://github.com/eth-cscs/UserLabDay/blob/master/2018/slides/ci_and_regression/ReFrame_CI.pdf)] @ [CSCS User Lab Day 2018](https://github.com/eth-cscs/UserLabDay).\n* Slides [[pdf](https://drive.google.com/open?id=1sZhibvUlGlT670aOHPdMlWFffWptYzLX)] @ [HPC Advisory Council 2018](http://www.hpcadvisorycouncil.com/events/2018/swiss-workshop/).\n* Slides [[pdf](https://drive.google.com/open?id=1EyJ-siupkgLeVT54A4WlFpQtrJaU0xOy)] @ [SC17](https://sc17.supercomputing.org/).\n* Slides [[pdf](https://drive.google.com/open?id=18VrCy0MTplGo67uxVbzYZicQChor9VSY)] @ [CUG 2017](https://cug.org/cug-2017/).\n\nWebinars[¶](#webinars)\n---\n\n* “ReFrame – Efficient System and Application Performance Testing,” CSCS Webinar, Aug. 29, 2022 [[slides](https://drive.google.com/file/d/1nOS_daleR79ZB1IaToVdW5mDpJQYRcY2/view?usp=sharing)] [[recording](https://youtu.be/NDxlKATEcQk)] [[demo run](https://asciinema.org/a/517693)].\n\n### Getting Started[¶](#getting-started)\n\n#### Requirements[¶](#requirements)\n\n* Python 3.6 or higher.\nPython 2 is not supported.\n* The required Python packages are the following:\n```\narchspec==0.2.0 argcomplete==2.0.0 coverage==6.2; python_version < '3.7'\ncoverage==7.1.0; python_version >= '3.7'\nimportlib_metadata==4.0.1; python_version < '3.8'\njsonschema==3.2.0 lxml==4.9.2 pytest==7.0.1 pytest-forked==1.4.0 pytest-parallel==0.1.1 pytest-rerunfailures==10.3; python_version == '3.6'\npytest-rerunfailures==11.0; python_version >= '3.7'\nPyYAML==6.0 requests==2.27.1; python_version == '3.6'\nrequests==2.28.1; python_version >= '3.7'\nsemver==2.13.0 setuptools==59.6.0; python_version == '3.6'\nsetuptools==67.1.0; python_version >= '3.7'\nwcwidth==0.2.6\n```\nNote\n\nChanged in version 3.0: Support for Python 3.5 has been dropped.\n\n#### Getting the Framework[¶](#getting-the-framework)\n\nStable ReFrame releases are available through different channels.\n\n##### Spack[¶](#spack)\n\nReFrame is available as a [Spack](https://spack.io/) package:\n```\nspack install reframe\n```\nThere are the following variants available:\n\n* `+docs`: This will install the man pages of ReFrame.\n* `+gelf`: This will install the bindings for handling [Graylog](https://docs.graylog.org/) log messages.\n\n##### EasyBuild[¶](#easybuild)\n\nReFrame is available as an [EasyBuild](https://easybuild.readthedocs.io/en/latest/) package:\n```\neb ReFrame-VERSION.eb -r\n```\nThis will install the man pages as well as the [Graylog](https://docs.graylog.org/) bindings.\n\n##### PyPI[¶](#pypi)\n\nReFrame is available as a [PyPI](https://pypi.org/project/ReFrame-HPC/) package:\n```\npip install reframe-hpc\n```\nThis is a bare installation of the framework.\nIt will not install the documentation, the tutorial examples or the bindings for handling [Graylog](https://docs.graylog.org/) log messages.\n\n##### Github[¶](#github)\n\nAny ReFrame version can be very easily installed directly from Github:\n```\npushd /path/to/install/prefix git clone -q --depth 1 --branch VERSION_TAG https://github.com/reframe-hpc/reframe.git pushd reframe && ./bootstrap.sh && popd export PATH=$(pwd)/bin:$PATH popd\n```\nThe `VERSION_TAG` is the version number prefixed by `v`, e.g., `v3.5.0`.\nThe `./bootstrap.sh` script will fetch ReFrame’s requirements under its installation prefix.\nIt will not set the `PYTHONPATH`, so it will not affect the user’s Python installation.\nThe `./bootstrap.sh` has two additional variant options:\n\n* `+docs`: This will also build the documentation.\n* `+pygelf`: This will install the bindings for handling [Graylog](https://docs.graylog.org/) log messages.\n\nNote\n\nNew in version 3.1: The bootstrap script for ReFrame was added.\nFor previous ReFrame versions you should install its requirements using `pip install -r requirements.txt` in a Python virtual environment.\n\n#### Enabling auto-completion[¶](#enabling-auto-completion)\n\nNew in version 3.4.1.\n\nYou can enable auto-completion for ReFrame by sourcing in your shell the corresponding script in `<install_prefix>/share/completions/reframe.<shell>`.\nAuto-completion is supported for Bash, Tcsh and Fish shells.\n\nNote\n\nChanged in version 3.4.2: The shell completion scripts have been moved under `share/completions/`.\n\n#### Where to Go from Here[¶](#where-to-go-from-here)\n\nIf you are new to ReFrame, the place to start is the first tutorial [Tutorial 1: Getting Started with ReFrame](index.html#document-tutorial_basics), which will guide you step-by-step in both writing your first tests and in configuring ReFrame.\nThe rest of the tutorials explore additional capabilities of the framework and cover several topics that you will likely come across when writing your own tests.\n\nThe [Configuring ReFrame for Your Site](index.html#document-configure) page provides more details on how a configuration file is structured and the [Advanced Topics](index.html#document-topics) explain some more advanced concepts as well as some implementation details.\nThe [ReFrame Manuals](index.html#document-manuals) provide complete reference guides for the command line interface, the configuration parameters and the programming APIs for writing tests.\n\nFinally, if you are not new to ReFrame and you have been using the 3.x versions, you should read the [What’s New in ReFrame 4.0](index.html#document-whats_new_40) page, which explains what are the key new features of ReFrame 4.0 as well as all the breaking changes.\n\n### What’s New in ReFrame 4.0[¶](#what-s-new-in-reframe-4-0)\n\nReFrame 4.0 introduces some important new features and removes all features, configuration options and interfaces that were deprecated in the 3.x versions.\nIt also introduces a couple of new deprecations.\n\nReFrame 4.0 maintains backward compatibility as much as possible.\nExisting 3.x configurations and 3.x tests are expected to run out-of-the-box, despite any warnings issued.\nThe framework’s behavior with respect to performance logging has also changed, but configuration options are offered so that users can switch to the old behavior.\n\nThis page summarizes the key changes in ReFrame 4.0 and what users should pay attention to.\n\nFor a complete list of changes, please refer to the [Release Notes](https://github.com/reframe-hpc/reframe/releases/tag/v4.0.0).\n\n#### New Features and Enchancements[¶](#new-features-and-enchancements)\n\n##### Chaining Configuration Files[¶](#chaining-configuration-files)\n\nThere is no need anymore to keep a huge configuration file with all your system and environment definitions and it is no more required to carry on the generic system configuration as well as any part of the builtin configuration.\nReFrame 4.0 allows you to split your configuration in multiple files.\nThis allows you to create minimal configuration files that contain only the necessary parts.\nFor example, if you want to define a general configuration parameter, you don’t need to copy the builtin configuration file and add it, but you simply add it in a single `general` section.\nThis can also be very useful if you maintain a ReFrame installation used by others, as you can update your settings (systems, environments and other options) and any of your users’ custom configuration will automatically inherit your settings if it is properly chained.\nTo assist with system-wide installation the `RFM_CONFIG_PATH` environment variable is introduced that allows you to specify a path where ReFrame will look for configuration files to load.\n\nNow that systems and environments definitions can be distributed over multiple configuration files, it can become easy to accidentally redefine a system or environment without a notice.\nFor this reason, ReFrame warns you if a system or an environment is redefined in the same scope.\nSince in the past all configuration files where extended copies of the builtin configuration, you will get warnings that the `generic` system and the `builtin` environment are redefined, as ReFrame finds them in the builtin configuration, which is always loaded.\nYou can safely ignore these warnings and use the definitions in your configuration file.\nIf you want to eliminate them, though, you should remove the conflicting definitions from your configuration file.\n\nAlthough ReFrame will not warn you for redefining other configuration sections, you are also advised to tidy up your configuration file and remove any parts that were copied unchanged from the builtin configuration.\n\nFor more information on how ReFrame 4.0 builds and loads its configuration, please refer to the documentation of the [`--config-file`](index.html#cmdoption-C) option, as well as the [Building the Final Configuration](index.html#building-the-final-config) section.\n\n##### Performance Reporting and Logging[¶](#performance-reporting-and-logging)\n\nReFrame 4.0 improves on how performance values are logged and reported.\nThis is a breaking change, but you can easily revert to the old behavior.\n\nReFrame now logs performance *after* the test has finished and not during the performance stage.\nYou can now log the result of the test by including `%(check_result)s` in your log handler format string.\nHowever, now, by default, ReFrame will log all the performance variables in a *single* record;\nin the past, a new record was logged for each performance variable.\nAlso, the `%(check_perf_*)s` format placeholders are valid only in the [`format_perfvars`](index.html#config.logging.handlers_perflog.format_perfvars) configuration parameter and will be used to format the performance values if the `%(check_perfvalues)s` placeholder is present in the handler’s [`format`](index.html#config.logging.handlers_perflog.format) parameter.\nThis change in behavior will likely break your log processing, especially if you are using the `graylog` or `httpjson` handlers or any handler that sends the full record to a log server.\nYou can revert to the old behavior by setting the [`perflog_compat`](index.html#config.logging.perflog_compat) configuration parameter.\nThis will send a separate record for each performance variable that will include all the individual `%(check_perf_*)s` attributes.\nFor more information, check the documentation of the [`format_perfvars`](index.html#config.logging.handlers_perflog.format_perfvars) configuration parameter.\n\nThe behavior of the `filelog` is also substantially improved.\nThe log file is printed by default in CSV format and a header is always printed at the beginning of each log file.\nIf the log format changes or the performance variables logged by the test change, a new log file will be created with an adapted header.\nThis way, every log file is consistent with the data in contains.\nFor more information, please refer to the `filelog` handler [documentation](index.html#filelog-handler).\n\nWhen you run a performance test, ReFrame will now print immediately after the test has finished a short summary of its performance.\nYou can suppress this output by setting the log level at which this information is printed to `verbose` by setting the [`perf_info_level`](index.html#config.general.perf_info_level) general configuration parameter.\n\nFinally, the performance report printed at the end of the run using the [`--performance-report`](index.html#cmdoption-performance-report) is revised providing more information in more compact form.\n\n##### New Test Naming Scheme[¶](#new-test-naming-scheme)\n\nReFrame 4.0 makes default the new test naming scheme introduced in 3.10.0 and drops support of the old naming scheme.\nThe new naming scheme does not affect normal tests, but it changes how parameterized tests and fixtures are named.\nEach test is now also associated with a unique hash code.\nFor parameterized tests and fixtures this hash code is appended to the test’s or fixture’s base name when creating any test-specific directories and files, such as the test stage and output directories.\nThe [`-n`](index.html#cmdoption-n) option can match a test either by its display name (the default), or by its unique internal name or by its unique hash code.\nCheck the documentation of the [`-n`](index.html#cmdoption-n) for more information.\nFor the details of the new naming scheme, please refer to the [Test Naming Scheme](index.html#test-naming-scheme) section.\n\nNote that any tests that used the old naming scheme to depend on parameterized tests will break with this change.\nCheck the tutorial [Depending on Parameterized Tests](index.html#param-deps) on how to create dependencies on parameterized tests in a portable way.\n\n##### Custom parallel launchers[¶](#custom-parallel-launchers)\n\nBy relaxing the configuration schema, users can now define custom parallel launchers inside their Python configuration file.\nCheck the tutorial [Adding a custom launcher to a partition](index.html#custom-launchers) to find out how this can be achieved.\n\n##### Unique run reports[¶](#unique-run-reports)\n\nReFrame now generates a unique report for each run inside the `$HOME/.reframe/reports` directory.\nIf you want to revert to the old behavior, where a single file was generated and was overwritten in every run, you should set the [`report_file`](index.html#config.general.report_file) configuration option or the [`RFM_REPORT_FILE`](index.html#envvar-RFM_REPORT_FILE) environment variable.\n\n##### New Backends[¶](#new-backends)\n\nReFrame 4.0 adds support for the [Apptainer](https://apptainer.org/) container platform and the [Flux framework](http://flux-framework.org/).\n\n#### Dropped Features and Deprecations[¶](#dropped-features-and-deprecations)\n\nReFrame 4.0 drops support for all the deprecated features and behaviors of ReFrame 3.x versions.\nMore specifically, the following deprecated features are dropped:\n\n* The `@parameterized_test` decorator is dropped in favor of the [`parameter`](index.html#reframe.core.builtins.parameter) builtin.\n* The [`name`](index.html#reframe.core.pipeline.RegressionTest.name) of the test is now read-only.\n* The decorators `@final`, [`@require_deps`](index.html#reframe.core.builtins.require_deps), [`@run_after`](index.html#reframe.core.builtins.run_after) and [`@run_before`](index.html#reframe.core.builtins.run_before) are no more accesible via the `reframe` module.\nThey are directly available in the [`RegressionTest`](index.html#reframe.core.pipeline.RegressionTest) namespace without the need of importing anything.\n* The `@reframe.utility.sanity.sanity_function` decorator is dropped in favor of the [`@deferrable`](index.html#reframe.core.builtins.deferrable) builtin.\n* The `commands` attribute of the [`ContainerPlatform`](index.html#reframe.core.containers.ContainerPlatform) is dropped in favor of the [`command`](index.html#reframe.core.containers.ContainerPlatform.command) attribute.\n* The `launcher` attribute of the [`System`](index.html#reframe.core.systems.System) is dropped in favor of the `launcher_type` attribute.\n* The `@required_version` decorator is dropped in favor of the `require_version` builtin.\nAlso, automatically converting version strings that do not comply with the semantic versioning scheme is no more supported.\n* The `DEPEND_EXACT`, `DEPEND_BY_ENV` and `DEPEND_FULLY` integer constants that were passed as the `how` argument of the [`depends_on()`](index.html#reframe.core.pipeline.RegressionTest.depends_on) method are no more supported and a callable should be used instead.\nThe `subdeps` argument is also dropped.\n* The low-level `poll()` and `wait()` `RegressionTest` methods are dropped in favor of the [`run_complete()`](index.html#reframe.core.pipeline.RegressionTest.run_complete) and [`run_wait()`](index.html#reframe.core.pipeline.RegressionTest.run_wait), respectively.\n* The `schedulers` configuration section is dropped in favor of the partition-specific [`sched_options`](index.html#config.systems.partitions.sched_options).\nUsers should move any options set in the old section to the corresponding partition options.\n* The `--ignore-check-conflicts` command line option and the corresponding `RFM_IGNORE_CHECK_CONFLICTS` environment variable are dropped.\n* The `--force-local` and `--strict` command line options are removed.\nPlease use instead `-S local=1` and `-S strict_check=1`, respectively.\n* The `RFM_GRAYLOG_SERVER` environment variable is dropped in favor of the [`RFM_GRAYLOG_ADDRESS`](index.html#envvar-RFM_GRAYLOG_ADDRESS).\n\n##### New Deprecations[¶](#new-deprecations)\n\n* All occurrences of the `variables` name are deprecated in favor of `env_vars`.\nThis includes the [`variables`](index.html#reframe.core.pipeline.RegressionTest.variables) test attribute and the homonym systems, partitions and environments configuration parameters as well as the [`variables`](index.html#reframe.core.environments.Environment.variables) of the [`Environment`](index.html#reframe.core.environments.Environment) base class.\n* Although [`perf_patterns`](index.html#reframe.core.pipeline.RegressionTest.perf_patterns) attribute is not deprecated, users are recommended to migrate to using the new [`@performance_function`](index.html#reframe.core.builtins.performance_function) builtin.\nPlease refer to [Writing A Performance Test](index.html#perftest-basics) tutorial for a starting point.\n\n### ReFrame Tutorials[¶](#reframe-tutorials)\n\n#### Tutorial 1: Getting Started with ReFrame[¶](#tutorial-1-getting-started-with-reframe)\n\nNew in version 3.1.\n\nThis tutorial will give you a first overview of ReFrame and will acquaint you with its basic concepts.\nWe will start with a simple “Hello, World!” test running with the default configuration and we will expand the example along the way.\nWe will also explore performance tests and port our tests to an HPC cluster.\nThe examples of this tutorial can be found under `tutorials/basics/`.\n\n##### Getting Ready[¶](#getting-ready)\n\nAll you need to start off with this tutorial is to have [installed](started.html#getting-the-framework) ReFrame.\nIf you haven’t done so yet, all you need is Python 3.6 and above and to follow the steps below:\n```\ngit clone https://github.com/reframe-hpc/reframe.git cd reframe\n./bootstrap.sh\n./bin/reframe -V\n```\nWe’re now good to go!\n\n##### The “Hello, World!” test[¶](#the-hello-world-test)\n\nAs simple as it may sound, a series of “naive” “Hello, World!” tests can reveal lots of regressions in the programming environment of HPC clusters, but the bare minimum of those also serves perfectly the purpose of starting this tutorial.\nHere is its C version:\n```\ncat tutorials/basics/hello/src/hello.c\n```\n```\n#include <stdio.hint main()\n{\n printf(\"Hello, World!\\n\");\n return 0;\n}\n```\nAnd here is the ReFrame version of it:\n```\ncat tutorials/basics/hello/hello1.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class HelloTest(rfm.RegressionTest):\n valid_systems = ['*']\n valid_prog_environs = ['*']\n sourcepath = 'hello.c'\n\n @sanity_function\n def assert_hello(self):\n return sn.assert_found(r'Hello, World\\!', self.stdout)\n```\nRegression tests in ReFrame are specially decorated classes that ultimately derive from [`RegressionTest`](index.html#reframe.core.pipeline.RegressionTest).\nThe [`@simple_test`](index.html#reframe.core.decorators.simple_test) decorator registers a test class with ReFrame and makes it available to the framework.\nThe test variables are essentially attributes of the test class and can be defined directly in the class body.\nEach test must always set the [`valid_systems`](index.html#reframe.core.pipeline.RegressionTest.valid_systems) and [`valid_prog_environs`](index.html#reframe.core.pipeline.RegressionTest.valid_prog_environs) attributes.\nThese define the systems and/or system partitions that this test is allowed to run on, as well as the programming environments that it is valid for.\nA programming environment is essentially a compiler toolchain.\nWe will see later on in the tutorial how a programming environment can be defined.\nThe generic configuration of ReFrame assumes a single programming environment named `builtin` which comprises a C compiler that can be invoked with `cc`.\nIn this particular test we set both these attributes to `['*']`, essentially allowing this test to run everywhere.\n\nA ReFrame test must either define an executable to execute or a source file (or source code) to be compiled.\nIn this example, it is enough to define the source file of our hello program.\nReFrame knows the executable that was produced and will use that to run the test.\n\nFinally, every regression test must always decorate a member function as the test’s `@sanity_function`.\nThis decorated function is converted into a [lazily evaluated](deferrables.html) expression that asserts the sanity of the test.\nIn this particular case, the specified sanity function checks that the executable has produced the desired phrase into the test’s standard output [`stdout`](index.html#reframe.core.pipeline.RegressionTest.stdout).\nNote that ReFrame does not determine the success of a test by its exit code.\nInstead, the assessment of success is responsibility of the test itself.\n\nBefore running the test let’s inspect the directory structure surrounding it:\n```\ntutorials/basics/hello\n├── hello1.py\n└── src\n └── hello.c\n```\nOur test is `hello1.py` and its resources, i.e., the `hello.c` source file, are located inside the `src/` subdirectory.\nIf not specified otherwise, the [`sourcepath`](index.html#reframe.core.pipeline.RegressionTest.sourcepath) attribute is always resolved relative to `src/`.\nThere is full flexibility in organizing the tests.\nMultiple tests may be defined in a single file or they may be split in multiple files.\nSimilarly, several tests may share the same resources directory or they can simply have their own.\n\nNow it’s time to run our first test:\n```\n./bin/reframe -c tutorials/basics/hello/hello1.py -r\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe -c tutorials/basics/hello/hello1.py -r'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>'\n check search path: '/home/user/Repositories/reframe/tutorials/basics/hello/hello1.py'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/var/<KEY>/rfm-tgqpdq_b.log'\n\n[===] Running 1 check(s)\n[===] Started on Sat Nov 12 19:00:44 2022 \n\n[---] start processing checks\n[ RUN ] HelloTest /2b3e4546 @generic:default+builtin\n[ OK ] (1/1) HelloTest /2b3e4546 @generic:default+builtin\n[---] all spawned checks have finished\n\n[ PASSED ] Ran 1/1 test case(s) from 1 check(s) (0 failure(s), 0 skipped)\n[===] Finished on Sat Nov 12 19:00:45 2022 \nRun report saved in '/home/user/.reframe/reports/run-report-319.json'\nLog file(s) saved in '/var/folders/<KEY>000gp/T/rfm-tgqpdq_b.log'\n```\nPerfect! We have verified that we have a functioning C compiler in our system.\n\nWhen ReFrame runs a test, it copies all its resources to a stage directory and performs all test-related operations (compilation, run, sanity checking etc.) from that directory.\nOn successful outcome of the test, the stage directory is removed by default, but interesting files are copied to an output directory for archiving and later inspection.\nThe prefixes of these directories are printed in the first section of the output.\nLet’s inspect what files ReFrame produced for this test:\n```\nls output/generic/default/builtin/HelloTest/\n```\n```\nrfm_HelloTest_build.err rfm_HelloTest_build.sh rfm_HelloTest_job.out rfm_HelloTest_build.out rfm_HelloTest_job.err rfm_HelloTest_job.sh\n```\nReFrame stores in the output directory of the test the build and run scripts it generated for building and running the code along with their standard output and error.\nAll these files are prefixed with `rfm_`.\n\nReFrame also generates a detailed JSON report for the whole regression testing session.\nBy default, this is stored inside the `${HOME}/.reframe/reports` directory and a new report file is generated every time ReFrame is run, but you can control this through the [`--report-file`](index.html#cmdoption-report-file) command-line option.\n\nHere are the contents of the report file for our first ReFrame run:\n```\ncat ~/.reframe/reports/run-report.json\n```\n```\n{\n \"session_info\": {\n \"cmdline\": \"./bin/reframe -c tutorials/basics/hello/hello1.py -r\",\n \"config_file\": \"<builtin>\",\n \"data_version\": \"2.0\",\n \"hostname\": \"host\",\n \"prefix_output\": \"/path/to/reframe/output\",\n \"prefix_stage\": \"/path/to/reframe/stage\",\n \"user\": \"user\",\n \"version\": \"3.10.0-dev.3+c22440c1\",\n \"workdir\": \"/path/to/reframe\",\n \"time_start\": \"2022-01-22T13:21:50+0100\",\n \"time_end\": \"2022-01-22T13:21:51+0100\",\n \"time_elapsed\": 0.8124568462371826,\n \"num_cases\": 1,\n \"num_failures\": 0\n },\n \"runs\": [\n {\n \"num_cases\": 1,\n \"num_failures\": 0,\n \"num_aborted\": 0,\n \"num_skipped\": 0,\n \"runid\": 0,\n \"testcases\": [\n {\n \"build_stderr\": \"rfm_HelloTest_build.err\",\n \"build_stdout\": \"rfm_HelloTest_build.out\",\n \"dependencies_actual\": [],\n \"dependencies_conceptual\": [],\n \"description\": \"HelloTest\",\n \"display_name\": \"HelloTest\",\n \"filename\": \"/path/to/reframe/tutorials/basics/hello/hello1.py\",\n \"environment\": \"builtin\",\n \"fail_phase\": null,\n \"fail_reason\": null,\n \"jobid\": \"43152\",\n \"job_stderr\": \"rfm_HelloTest_job.err\",\n \"job_stdout\": \"rfm_HelloTest_job.out\",\n \"maintainers\": [],\n \"name\": \"HelloTest\",\n \"nodelist\": [\n \"tresa.local\"\n ],\n \"outputdir\": \"/path/to/reframe/output/generic/default/builtin/HelloTest\",\n \"perfvars\": null,\n \"prefix\": \"/path/to/reframe/tutorials/basics/hello\",\n \"result\": \"success\",\n \"stagedir\": \"/path/to/reframe/stage/generic/default/builtin/HelloTest\",\n \"scheduler\": \"local\",\n \"system\": \"generic:default\",\n \"tags\": [],\n \"time_compile\": 0.27164483070373535,\n \"time_performance\": 0.00010180473327636719,\n \"time_run\": 0.3764667510986328,\n \"time_sanity\": 0.0006909370422363281,\n \"time_setup\": 0.007919073104858398,\n \"time_total\": 0.8006880283355713,\n \"unique_name\": \"HelloTest\"\n }\n ]\n }\n ],\n \"restored_cases\": []\n}\n```\n##### More of “Hello, World!”[¶](#more-of-hello-world)\n\nWe want to extend our test and run a C++ “Hello, World!” as well.\nWe could simply copy paste the `hello1.py` and change the source file extension to refer to the C++ source code.\nBut this duplication is something that we generally want to avoid.\nReFrame allows you to avoid this in several ways but the most compact is to define the new test as follows:\n```\ncat tutorials/basics/hello/hello2.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class HelloMultiLangTest(rfm.RegressionTest):\n lang = parameter(['c', 'cpp'])\n\n valid_systems = ['*']\n valid_prog_environs = ['*']\n\n @run_before('compile')\n def set_sourcepath(self):\n self.sourcepath = f'hello.{self.lang}'\n\n @sanity_function\n def assert_hello(self):\n return sn.assert_found(r'Hello, World\\!', self.stdout)\n```\nThis test extends the `hello1.py` test by defining the `lang` parameter with the `parameter()` built-in.\nThis parameter will cause as many instantiations as parameter values available, each one setting the `lang` attribute to one single value.\nHence, this example will create two test instances, one with `lang='c'` and another with `lang='cpp'`.\nThe parameter is available as an attribute of the test instance and, in this example, we use it to set the extension of the source file.\nHowever, at the class level, a test parameter holds all the possible values for itself, and this is only assigned a single value after the class is instantiated.\nTherefore, the variable `sourcepath`, which depends on this parameter, also needs to be set after the class instantiation.\nThe simplest way to do this would be to move the `sourcepath` assignment into the `__init__()` method as shown in the code snippet below, but this has some disadvantages when writing larger tests.\n```\ndef __init__(self):\n self.sourcepath = f'hello.{self.lang}'\n```\nFor example, when writing a base class for a test with a large amount of code into the `__init__()` method, the derived class may want to do a partial override of the code in this function.\nThis would force us to understand the full implementation of the base class’ `__init__()` despite that we may just be interested in overriding a small part of it.\nDoable, but not ideal.\nInstead, through pipeline hooks, ReFrame provides a mechanism to attach independent functions to execute at a given time before the data they set is required by the test.\nThis is exactly what we want to do here, and we know that the test sources are needed to compile the code.\nHence, we move the `sourcepath` assignment into a pre-compile hook.\n```\n @run_before('compile')\n def set_sourcepath(self):\n self.sourcepath = f'hello.{self.lang}'\n```\nThe use of hooks is covered in more detail later on, but for now, let’s just think of them as a way to defer the execution of a function to a given stage of the test’s pipeline.\nBy using hooks, any user could now derive from this class and attach other hooks (for example, adding some compiler flags) without having to worry about overriding the base method that sets the `sourcepath` variable.\n\nLet’s run the test now:\n```\n./bin/reframe -c tutorials/basics/hello/hello2.py -r\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe -c tutorials/basics/hello/hello2.py -r'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>'\n check search path: '/home/user/Repositories/reframe/tutorials/basics/hello/hello2.py'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/<KEY>'\n\n[===] Running 2 check(s)\n[===] Started on Sat Nov 12 19:00:45 2022 \n\n[---] start processing checks\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @generic:default+builtin\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @generic:default+builtin\n[ FAIL ] (1/2) HelloMultiLangTest %lang=cpp /71bf65a3 @generic:default+builtin\n==> test failed during 'compile': test staged in '/home/user/Repositories/reframe/stage/generic/default/builtin/HelloMultiLangTest_71bf65a3'\nrfm_job.out\n[ OK ] (2/2) HelloMultiLangTest %lang=c /7cfa870e @generic:default+builtin\n[---] all spawned checks have finished\n\n[ FAILED ] Ran 2/2 test case(s) from 2 check(s) (1 failure(s), 0 skipped)\n[===] Finished on Sat Nov 12 19:00:46 2022 \n\n===\nSUMMARY OF FAILURES\n---\nFAILURE INFO for HelloMultiLangTest_1 \n * Expanded name: HelloMultiLangTest %lang=cpp\n * Description: \n * System partition: generic:default\n * Environment: builtin\n * Stage directory: /home/user/Repositories/reframe/stage/generic/default/builtin/HelloMultiLangTest_71bf65a3\n * Node list: \n * Job type: local (id=None)\n * Dependencies (conceptual): []\n * Dependencies (actual): []\n * Maintainers: []\n * Failing phase: compile\n * Rerun with '-n /71bf65a3 -p builtin --system generic:default -r'\n * Reason: build system error: I do not know how to compile a C++ program\n---\nRun report saved in '/home/user/.reframe/reports/run-report-320.json'\nLog file(s) saved in <KEY>'\n```\nOops! The C++ test has failed.\nReFrame complains that it does not know how to compile a C++ program.\nRemember our discussion above that the default configuration of ReFrame defines a minimal programming environment named `builtin` which only knows of a `cc` compiler.\nWe will fix that in a moment, but before doing that it’s worth looking into the failure information provided for the test.\nFor each failed test, ReFrame will print a short summary with information about the system partition and the programming environment that the test failed for, its job or process id (if any), the nodes it was running on, its stage directory, the phase that failed etc.\n\nWhen a test fails its stage directory is kept intact, so that users can inspect the failure and try to reproduce it manually.\nIn this case, the stage directory contains only the “Hello, World” source files, since ReFrame could not produce a build script for the C++ test, as it doesn’t know to compile a C++ program for the moment.\n```\nls stage/generic/default/builtin/HelloMultiLangTest_cpp\n```\n```\nhello.c hello.cpp\n```\nLet’s go on and fix this failure by defining a new system and programming environments for the machine we are running on.\nFor this we need to create our own configuration file.\n```\nvi tutorials/config/tresa.py\n```\nHere is what we need to type:\n```\n# Copyright 2016-2022 Swiss National Supercomputing Centre (CSCS/ETH Zurich)\n# and other ReFrame Project Developers. See the top-level LICENSE file for\n# details.\n#\n# SPDX-License-Identifier: BSD-3-Clause\n\nsite_configuration = {\n 'systems': [\n {\n 'name': 'tresa',\n 'descr': 'My Mac',\n 'hostnames': ['tresa'],\n 'modules_system': 'nomod',\n 'partitions': [\n {\n 'name': 'default',\n 'scheduler': 'local',\n 'launcher': 'local',\n 'environs': ['gnu', 'clang'],\n }\n ]\n }\n ],\n 'environments': [\n {\n 'name': 'gnu',\n 'cc': 'gcc-12',\n 'cxx': 'g++-12',\n 'ftn': 'gfortran-12',\n 'target_systems': ['tresa']\n },\n {\n 'name': 'clang',\n 'cc': 'clang',\n 'cxx': 'clang++',\n 'ftn': '',\n 'target_systems': ['tresa']\n },\n ]\n}\n```\nWe define a system named `tresa` that has one partition named `default`.\nThis partition makes no use of any [workload manager](config_reference.html#config.systems.partitions.scheduler), but instead launches any jobs locally as OS processes.\nTwo programming environments are relevant for that partition, namely `gnu` and `clang`, which are defined in the section [`environments`](index.html#environments) of the configuration file.\nThe `gnu` programming environment provides GCC 12, whereas the `clang` one provides the Clang compiler from the system.\nNotice, how you can define the actual commands for invoking the C, C++ and Fortran compilers in each programming environment.\nAs soon as a programming environment defines the different compilers, ReFrame will automatically pick the right compiler based on the source file extension.\nIn addition to C, C++ and Fortran programs, ReFrame will recognize the `.cu` extension as well and will try to invoke the `nvcc` compiler for CUDA programs.\nNote also that we set the [`target_systems`](index.html#config.environments.target_systems) for each environment definition.\nThis restricts the definition of the environment being defined to the specified systems only.\nReFrame will always pick the definition that is a closest match for the current system.\nRestricting the environment definitions is generally a good practice if you plan to define multiple systems in multiple configuration files, as ReFrame would otherwise complain that an environment is redefined.\nOn the other hand, if you want to provide generic definitions of environments that are valid for multiple systems, you may skip that.\nThis is what the builtin configuration of ReFrame does for its generic `builtin` environment.\n\nFinally, the new system that we defined may be identified by the hostname `tresa` (see the [`hostnames`](index.html#config.systems.hostnames) systems configuration parameter) and it will not use any environment modules system (see the [`modules_system`](index.html#config.systems.modules_system) configuration parameter).\nThe [`hostnames`](index.html#config.systems.hostnames) attribute will help ReFrame to automatically pick the right configuration when running on it.\nNotice, how the `generic` system matches any hostname, so that it acts as a fallback system.\n\nNote\n\nMultiple systems may defined in a configuration file, in which case they are tried in order and the first match is picked.\nThis means that the systems whose `hostnames` patterns are more generic, they should go to the end of the list.\n\nThe [Configuring ReFrame for Your Site](index.html#document-configure) page describes the configuration file in more detail and the [Configuration Reference](index.html#document-config_reference) provides a complete reference guide of all the configuration options of ReFrame.\n\nLet’s now rerun our “Hello, World!” tests:\n```\n./bin/reframe -C tutorials/config/tresa.py -c tutorials/basics/hello/hello2.py -r\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe -C tutorials/config/tresa.py -c tutorials/basics/hello/hello2.py -r'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>', 'tutorials/config/tresa.py'\n check search path: '/home/user/Repositories/reframe/tutorials/basics/hello/hello2.py'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/<KEY>rfm-e3dlf19_.log'\n\n[===] Running 2 check(s)\n[===] Started on Sat Nov 12 19:00:46 2022 \n\n[---] start processing checks\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @tresa:default+gnu\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @tresa:default+clang\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @tresa:default+gnu\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @tresa:default+clang rfm_job.out\n[ OK ] (1/4) HelloMultiLangTest %lang=c /7cfa870e @tresa:default+gnu rfm_job.out\n[ OK ] (2/4) HelloMultiLangTest %lang=c /7cfa870e @tresa:default+clang rfm_job.out\n[ OK ] (3/4) HelloMultiLangTest %lang=cpp /71bf65a3 @tresa:default+gnu rfm_job.out\n[ OK ] (4/4) HelloMultiLangTest %lang=cpp /71bf65a3 @tresa:default+clang\n[---] all spawned checks have finished\n\n[ PASSED ] Ran 4/4 test case(s) from 2 check(s) (0 failure(s), 0 skipped)\n[===] Finished on Sat Nov 12 19:00:48 2022 \nRun report saved in '/home/user/.reframe/reports/run-report-321.json'\nLog file(s) saved in '/<KEY>rfm-e3dlf19_.log'\n```\nNotice how the same tests are now tried with both the `gnu` and `clang` programming environments, without having to touch them at all!\nThat’s one of the powerful features of ReFrame and we shall see later on, how easily we can port our tests to an HPC cluster with minimal changes.\nIn order to instruct ReFrame to use our configuration file, we use the [`-C`](index.html#cmdoption-C) command line option.\nSince we don’t want to type it throughout the tutorial, we could set the [`RFM_CONFIG_FILES`](index.html#envvar-RFM_CONFIG_FILES) environment variable, which takes a colon-separated list of configuration files that ReFrame will load.\nWe will take advantage of multiple configuration files later in the tutorial.\n```\nexport RFM_CONFIG_FILES=$(pwd)/tutorials/config/tresa.py\n```\nTip\n\nIf our configuration file was named `settings.py` and we did not intend to use multiple configuration files in the same directory, we could also set the [`RFM_CONFIG_PATH`](index.html#envvar-RFM_CONFIG_PATH) environment variable.\n\n##### A Multithreaded “Hello, World!”[¶](#a-multithreaded-hello-world)\n\nWe extend our C++ “Hello, World!” example to print the greetings from multiple threads:\n```\ncat tutorials/basics/hellomp/src/hello_threads.cpp\n```\n```\n#include <iomanip>\n#include <iostream>\n#include <mutex>\n#include <thread>\n#include <vector#ifdef SYNC_MESSAGES std::mutex hello_mutex;\n#endif\n\nvoid greetings(int tid)\n{\n#ifdef SYNC_MESSAGES\n const std::lock_guard<std::mutex> lock(hello_mutex);\n#endif\n std::cout << \"[\" << std::setw(2) << tid << \"] \" << \"Hello, World!\\n\";\n}\n\nint main(int argc, char *argv[])\n{\n int nr_threads = 1;\n if (argc > 1) {\n nr_threads = std::atoi(argv[1]);\n }\n\n if (nr_threads <= 0) {\n std::cerr << \"thread count must a be positive integer\\n\";\n return 1;\n }\n\n std::vector<std::thread> threads;\n for (auto i = 0; i < nr_threads; ++i) {\n threads.push_back(std::thread(greetings, i));\n }\n\n for (auto &t : threads) {\n t.join();\n }\n\n return 0;\n}\n```\nThis program takes as argument the number of threads it will create and it uses `std::thread`, which is a C++11 addition, meaning that we will need to pass `-std=c++11` to our compilers.\nHere is the corresponding ReFrame test, where the new concepts introduced are highlighted:\n```\ncat tutorials/basics/hellomp/hellomp1.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class HelloThreadedTest(rfm.RegressionTest):\n valid_systems = ['*']\n valid_prog_environs = ['*']\n sourcepath = 'hello_threads.cpp'\n build_system = 'SingleSource'\n executable_opts = ['16']\n\n @run_before('compile')\n def set_compilation_flags(self):\n self.build_system.cxxflags = ['-std=c++11', '-Wall']\n environ = self.current_environ.name\n if environ in {'clang', 'gnu'}:\n self.build_system.cxxflags += ['-pthread']\n\n @sanity_function\n def assert_hello(self):\n return sn.assert_found(r'Hello, World\\!', self.stdout)\n```\nReFrame delegates the compilation of a test to a [`build_system`](index.html#reframe.core.pipeline.RegressionTest.build_system), which is an abstraction of the steps needed to compile the test.\nBuild systems take also care of interactions with the programming environment if necessary.\nCompilation flags are a property of the build system.\nIf not explicitly specified, ReFrame will try to pick the correct build system (e.g., CMake, Autotools etc.) by inspecting the test resources, but in cases as the one presented here where we need to set the compilation flags, we need to specify a build system explicitly.\nIn this example, we instruct ReFrame to compile a single source file using the `-std=c++11 -pthread -Wall` compilation flags.\nHowever, the flag `-pthread` is only needed to compile applications using `std::thread` with the GCC and Clang compilers.\nHence, since this flag may not be valid for other compilers, we need to include it only in the tests that use either GCC or Clang.\nSimilarly to the `lang` parameter in the previous example, the information regarding which compiler is being used is only available after the class is instantiated (after completion of the `setup` pipeline stage), so we also defer the addition of this optional compiler flag with a pipeline hook.\nIn this case, we set the `set_compile_flags()` hook to run before the ReFrame pipeline stage `compile`.\n\nNote\n\nThe pipeline hooks, as well as the regression test pipeline itself, are covered in more detail later on in the tutorial.\n\nIn this example, the generated executable takes a single argument which sets the number of threads to be used.\nThe options passed to the test’s executable can be set through the [`executable_opts`](index.html#reframe.core.pipeline.RegressionTest.executable_opts) variable, which in this case is set to `'16'`.\n\nLet’s run the test now:\n```\n./bin/reframe -c tutorials/basics/hellomp/hellomp1.py -r\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe -c tutorials/basics/hellomp/hellomp1.py -r'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>', '/home/user/Repositories/reframe/tutorials/config/tresa.py'\n check search path: '/home/user/Repositories/reframe/tutorials/basics/hellomp/hellomp1.py'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/<KEY>80000gp/T/rfm-v56bz2uo.log'\n\n[===] Running 1 check(s)\n[===] Started on Sat Nov 12 19:00:48 2022 \n\n[---] start processing checks\n[ RUN ] HelloThreadedTest /a6fa300f @tresa:default+gnu\n[ RUN ] HelloThreadedTest /a6fa300f @tresa:default+clang\n[ OK ] (1/2) HelloThreadedTest /a6fa300f @tresa:default+gnu\n[ OK ] (2/2) HelloThreadedTest /a6fa300f @tresa:default+clang\n[---] all spawned checks have finished\n\n[ PASSED ] Ran 2/2 test case(s) from 1 check(s) (0 failure(s), 0 skipped)\n[===] Finished on Sat Nov 12 19:00:50 2022 \nRun report saved in '/home/user/.reframe/reports/run-report-322.json'\nLog file(s) saved in '/<KEY>rfm-v56bz2uo.log'\n```\nEverything looks fine, but let’s inspect the actual output of one of the tests:\n```\ncat output/catalina/default/clang/HelloThreadedTest/rfm_HelloThreadedTest_job.out\n```\n```\n[[[[ 8] Hello, World!\n1] Hello, World!\n5[[0[ 7] Hello, World!\n] ] Hello, World!\n[ Hello, World!\n6[] Hello, World!\n9] Hello, World!\n 2 ] Hello, World!\n4] [[10 3] Hello, World!\n] Hello, World!\n[Hello, World!\n11] Hello, World!\n[12] Hello, World!\n[13] Hello, World!\n[14] Hello, World!\n[15] Hello, World!\n```\nNot exactly what we were looking for!\nIn the following we write a more robust sanity check that can catch this havoc.\n\n###### More advanced sanity checking[¶](#more-advanced-sanity-checking)\n\nSo far, we have seen only a `grep`-like search for a string in the test’s [`stdout`](index.html#reframe.core.pipeline.RegressionTest.stdout), but ReFrame’s `@sanity_function` are much more capable than this.\nIn fact, one could practically do almost any operation in the output and process it as you would like before assessing the test’s sanity.\nIn the following, we extend the sanity checking of the above multithreaded “Hello, World!” to assert that all the threads produce a greetings line.\nSee the highlighted lines below in the modified version of the `@sanity_function`.\n```\ncat tutorials/basics/hellomp/hellomp2.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class HelloThreadedExtendedTest(rfm.RegressionTest):\n valid_systems = ['*']\n valid_prog_environs = ['*']\n sourcepath = 'hello_threads.cpp'\n build_system = 'SingleSource'\n executable_opts = ['16']\n\n @run_before('compile')\n def set_compilation_flags(self):\n self.build_system.cxxflags = ['-std=c++11', '-Wall']\n environ = self.current_environ.name\n if environ in {'clang', 'gnu'}:\n self.build_system.cxxflags += ['-pthread']\n\n @sanity_function\n def assert_num_messages(self):\n num_messages = sn.len(sn.findall(r'\\[\\s?\\d+\\] Hello, World\\!',\n self.stdout))\n return sn.assert_eq(num_messages, 16)\n```\nThis new `@sanity_function` counts all the pattern matches in the tests’s [`stdout`](index.html#reframe.core.pipeline.RegressionTest.stdout) and checks that this count matches the expected value.\nThe execution of the function `assert_num_messages()` is deferred to the `sanity` stage in the test’s pipeline, after the executable has run and the [`stdout`](index.html#reframe.core.pipeline.RegressionTest.stdout) file has been populated.\nIn this example, we have used the [`findall()`](index.html#reframe.utility.sanity.findall) utility function from the [`sanity`](index.html#module-reframe.utility.sanity) module to conveniently extract the pattern matches.\nThis module provides a broad range of utility functions that can be used to compose more complex sanity checks.\nHowever, note that the utility functions in this module are lazily evaluated expressions or deferred expressions which must be evaluated either implicitly or explicitly (see [Deferrable Functions Reference](index.html#document-deferrable_functions_reference)).\n\nLet’s run this version of the test now and see if it fails:\n```\n./bin/reframe -c tutorials/basics/hellomp/hellomp2.py -r\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe -c tutorials/basics/hellomp/hellomp2.py -r'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>', '/home/user/Repositories/reframe/tutorials/config/tresa.py'\n check search path: '/home/user/Repositories/reframe/tutorials/basics/hellomp/hellomp2.py'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/<KEY>80000gp/T/rfm-a2tt4eqp.log'\n\n[===] Running 1 check(s)\n[===] Started on Sat Nov 12 19:00:50 2022 \n\n[---] start processing checks\n[ RUN ] HelloThreadedExtendedTest /4733a67d @tresa:default+gnu\n[ RUN ] HelloThreadedExtendedTest /4733a67d @tresa:default+clang\n[ FAIL ] (1/2) HelloThreadedExtendedTest /4733a67d @tresa:default+gnu\n==> test failed during 'sanity': test staged in '/home/user/Repositories/reframe/stage/tresa/default/gnu/HelloThreadedExtendedTest'\n[ FAIL ] (2/2) HelloThreadedExtendedTest /4733a67d @tresa:default+clang\n==> test failed during 'sanity': test staged in '/home/user/Repositories/reframe/stage/tresa/default/clang/HelloThreadedExtendedTest'\n[---] all spawned checks have finished\n\n[ FAILED ] Ran 2/2 test case(s) from 1 check(s) (2 failure(s), 0 skipped)\n[===] Finished on Sat Nov 12 19:00:52 2022 \n\n===\nSUMMARY OF FAILURES\n---\nFAILURE INFO for HelloThreadedExtendedTest \n * Expanded name: HelloThreadedExtendedTest\n * Description: \n * System partition: tresa:default\n * Environment: gnu\n * Stage directory: /home/user/Repositories/reframe/stage/tresa/default/gnu/HelloThreadedExtendedTest\n * Node list: hostNone\n * Job type: local (id=59525)\n * Dependencies (conceptual): []\n * Dependencies (actual): []\n * Maintainers: []\n * Failing phase: sanity\n * Rerun with '-n /4733a67d -p gnu --system tresa:default -r'\n * Reason: sanity error: 13 != 16\n---\nFAILURE INFO for HelloThreadedExtendedTest \n * Expanded name: HelloThreadedExtendedTest\n * Description: \n * System partition: tresa:default\n * Environment: clang\n * Stage directory: /home/user/Repositories/reframe/stage/tresa/default/clang/HelloThreadedExtendedTest\n * Node list: hostNone\n * Job type: local (id=59528)\n * Dependencies (conceptual): []\n * Dependencies (actual): []\n * Maintainers: []\n * Failing phase: sanity\n * Rerun with '-n /4733a67d -p clang --system tresa:default -r'\n * Reason: sanity error: 11 != 16\n---\nRun report saved in '/home/user/.reframe/reports/run-report-323.json'\nLog file(s) saved in '/<KEY>'\n```\nAs expected, only some of lines are printed correctly which makes the test fail.\nTo fix this test, we need to compile with `-DSYNC_MESSAGES`, which will synchronize the printing of messages.\n```\ncat tutorials/basics/hellomp/hellomp3.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class HelloThreadedExtended2Test(rfm.RegressionTest):\n valid_systems = ['*']\n valid_prog_environs = ['*']\n sourcepath = 'hello_threads.cpp'\n build_system = 'SingleSource'\n executable_opts = ['16']\n\n @run_before('compile')\n def set_compilation_flags(self):\n self.build_system.cppflags = ['-DSYNC_MESSAGES']\n self.build_system.cxxflags = ['-std=c++11', '-Wall']\n environ = self.current_environ.name\n if environ in {'clang', 'gnu'}:\n self.build_system.cxxflags += ['-pthread']\n\n @sanity_function\n def assert_num_messages(self):\n num_messages = sn.len(sn.findall(r'\\[\\s?\\d+\\] Hello, World\\!',\n self.stdout))\n return sn.assert_eq(num_messages, 16)\n```\n##### Writing A Performance Test[¶](#writing-a-performance-test)\n\nAn important aspect of regression testing is checking for performance regressions.\nIn this example, we write a test that downloads the [STREAM](https://raw.githubusercontent.com/jeffhammond/STREAM/master/stream.c) benchmark, compiles it, runs it and records its performance.\nIn the test below, we highlight the lines that introduce new concepts.\n```\ncat tutorials/basics/stream/stream1.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class StreamTest(rfm.RegressionTest):\n valid_systems = ['*']\n valid_prog_environs = ['gnu']\n prebuild_cmds = [\n 'wget https://raw.githubusercontent.com/jeffhammond/STREAM/master/stream.c' # noqa: E501\n ]\n build_system = 'SingleSource'\n sourcepath = 'stream.c'\n env_vars = {\n 'OMP_NUM_THREADS': '4',\n 'OMP_PLACES': 'cores'\n }\n\n @run_before('compile')\n def set_compiler_flags(self):\n self.build_system.cppflags = ['-DSTREAM_ARRAY_SIZE=$((1 << 25))']\n self.build_system.cflags = ['-fopenmp', '-O3', '-Wall']\n\n @sanity_function\n def validate_solution(self):\n return sn.assert_found(r'Solution Validates', self.stdout)\n\n @performance_function('MB/s', perf_key='Copy')\n def extract_copy_perf(self):\n return sn.extractsingle(r'Copy:\\s+(\\S+)\\s+.*', self.stdout, 1, float)\n\n @performance_function('MB/s', perf_key='Scale')\n def extract_scale_perf(self):\n return sn.extractsingle(r'Scale:\\s+(\\S+)\\s+.*', self.stdout, 1, float)\n\n @performance_function('MB/s', perf_key='Add')\n def extract_add_perf(self):\n return sn.extractsingle(r'Add:\\s+(\\S+)\\s+.*', self.stdout, 1, float)\n\n @performance_function('MB/s', perf_key='Triad')\n def extract_triad_perf(self):\n return sn.extractsingle(r'Triad:\\s+(\\S+)\\s+.*', self.stdout, 1, float)\n```\nFirst of all, notice that we restrict the programming environments to `gnu` only, since this test requires OpenMP, which our installation of Clang does not have.\nThe next thing to notice is the [`prebuild_cmds`](index.html#reframe.core.pipeline.RegressionTest.prebuild_cmds) attribute, which provides a list of commands to be executed before the build step.\nThese commands will be executed from the test’s stage directory.\nIn this case, we just fetch the source code of the benchmark.\nFor running the benchmark, we need to set the OpenMP number of threads and pin them to the right CPUs through the `OMP_NUM_THREADS` and `OMP_PLACES` environment variables.\nYou can set environment variables in a ReFrame test through the [`env_vars`](index.html#reframe.core.pipeline.RegressionTest.env_vars) dictionary.\n\nWhat makes a ReFrame test a performance test is the definition of at least one [performance function](index.html#deferrable-performance-functions).\nSimilarly to a test’s [`@sanity_function`](index.html#reframe.core.builtins.sanity_function), a performance function is a member function decorated with the [`@performance_function`](index.html#reframe.core.builtins.performance_function) decorator that merely extracts or computes a performance metric from the test’s output and associates it with a unit.\nBy default, every performance function defined in the test is assigned to a *performance variable* with the function’s name.\nA performance variable is a named quantity representing a performance metric that ReFrame will report on, log and can also check against a reference value.\nThe performance variables of a test are stored in the [`perf_variables`](index.html#reframe.core.pipeline.RegressionTest.perf_variables) dictionary.\nThe keys are the names of the metrics, whereas the values are [performance functions](index.html#deferrable-performance-functions).\nThe [`@performance_function`](index.html#reframe.core.builtins.performance_function) decorator apart from turning an ordinary method into a “performance function”, it also creates an entry in the [`perf_variables`](index.html#reframe.core.pipeline.RegressionTest.perf_variables) dictionary.\nThe optional `perf_key` argument can be used to assign a different name to the newly created performance variable.\n\nIn this example, we extract four performance variables, namely the memory bandwidth values for each of the “Copy”, “Scale”, “Add” and “Triad” sub-benchmarks of STREAM, where each of the performance functions use the [`extractsingle()`](index.html#reframe.utility.sanity.extractsingle) utility function.\nFor each of the sub-benchmarks we extract the “Best Rate MB/s” column of the output (see below) and we convert that to a float.\n```\nFunction Best Rate MB/s Avg time Min time Max time Copy: 24939.4 0.021905 0.021527 0.022382 Scale: 16956.3 0.031957 0.031662 0.032379 Add: 18648.2 0.044277 0.043184 0.046349 Triad: 19133.4 0.042935 0.042089 0.044283\n```\nLet’s run the test now:\n```\n./bin/reframe -c tutorials/basics/stream/stream1.py -r --performance-report\n```\nThe [`--performance-report`](index.html#cmdoption-performance-report) will generate a short report at the end of the run for each performance test that has run.\nAdditionally, as soon as a performance test finishes, the obtained performance for each of the metrics is immediately reported.\nThis is especially useful if you run long suites of performance exploration tests and you do not want to wait until the end of the run to have an overview of the obtained performance.\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe -c tutorials/basics/stream/stream1.py -r --performance-report'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>', '/home/user/Repositories/reframe/tutorials/config/tresa.py'\n check search path: '/home/user/Repositories/reframe/tutorials/basics/stream/stream1.py'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/<KEY>'\n\n[===] Running 1 check(s)\n[===] Started on Sat Nov 12 19:00:53 2022 \n\n[---] start processing checks\n[ RUN ] StreamTest /cdf4820d @tresa:default+gnu\n[ OK ] (1/1) StreamTest /cdf4820d @tresa:default+gnu P: Copy: 24031.8 MB/s (r:0, l:None, u:None)\nP: Scale: 16297.9 MB/s (r:0, l:None, u:None)\nP: Add: 17843.8 MB/s (r:0, l:None, u:None)\nP: Triad: 18278.3 MB/s (r:0, l:None, u:None)\n[---] all spawned checks have finished\n\n[ PASSED ] Ran 1/1 test case(s) from 1 check(s) (0 failure(s), 0 skipped)\n[===] Finished on Sat Nov 12 19:00:56 2022 \n\n===\nPERFORMANCE REPORT\n---\n[StreamTest /cdf4820d @tresa:default:gnu]\n num_tasks: 1\n num_gpus_per_node: 0\n performance:\n - Copy: 24031.8 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 16297.9 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 17843.8 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 18278.3 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n---\nRun report saved in '/home/user/.reframe/reports/run-report-324.json'\nLog file(s) saved in '/<KEY>'\n```\n###### Setting explicitly the test’s performance variables[¶](#setting-explicitly-the-test-s-performance-variables)\n\nUsers are allowed to manipulate the test’s [`perf_variables`](index.html#reframe.core.pipeline.RegressionTest.perf_variables) dictionary directly.\nThis is useful to avoid code repetition or in cases that relying on decorated methods to populate the [`perf_variables`](index.html#reframe.core.pipeline.RegressionTest.perf_variables) is impractical, e.g., creating multiple performance variables in a loop.\n\nYou might have noticed that in our STREAM example above, all four performance functions are almost identical except for a small part of the regex pattern.\nIn the following example, we define a single performance function, `extract_bw()`, that can extract any of the requested bandwidth metrics, and we populate the [`perf_variables`](index.html#reframe.core.pipeline.RegressionTest.perf_variables) ourselves in a pre-performance hook:\n```\ncat tutorials/basics/stream/stream2.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class StreamAltTest(rfm.RegressionTest):\n valid_systems = ['*']\n valid_prog_environs = ['gnu']\n prebuild_cmds = [\n 'wget https://raw.githubusercontent.com/jeffhammond/STREAM/master/stream.c' # noqa: E501\n ]\n build_system = 'SingleSource'\n sourcepath = 'stream.c'\n env_vars = {\n 'OMP_NUM_THREADS': '4',\n 'OMP_PLACES': 'cores'\n }\n\n @run_before('compile')\n def set_compiler_flags(self):\n self.build_system.cppflags = ['-DSTREAM_ARRAY_SIZE=$((1 << 25))']\n self.build_system.cflags = ['-fopenmp', '-O3', '-Wall']\n\n @sanity_function\n def validate_solution(self):\n return sn.assert_found(r'Solution Validates', self.stdout)\n\n @performance_function('MB/s')\n def extract_bw(self, kind='Copy'):\n '''Generic performance extraction function.'''\n\n if kind not in ('Copy', 'Scale', 'Add', 'Triad'):\n raise ValueError(f'illegal value in argument kind ({kind!r})')\n\n return sn.extractsingle(rf'{kind}:\\s+(\\S+)\\s+.*',\n self.stdout, 1, float)\n\n @run_before('performance')\n def set_perf_variables(self):\n '''Build the dictionary with all the performance variables.'''\n\n self.perf_variables = {\n 'Copy': self.extract_bw(),\n 'Scale': self.extract_bw('Scale'),\n 'Add': self.extract_bw('Add'),\n 'Triad': self.extract_bw('Triad'),\n }\n```\nAs mentioned in the previous section the [`@performance_function`](index.html#reframe.core.builtins.performance_function) decorator performs two tasks:\n\n1. It converts a test method to *performance function*, i.e., a function that is suitable for extracting a performance metric.\n2. It updates the [`perf_variables`](index.html#reframe.core.pipeline.RegressionTest.perf_variables) dictionary with the newly created performance function.\n\nIn this example, we are only interested in the first functionality and that’s why we redefine completely the test’s [`perf_variables`](index.html#reframe.core.pipeline.RegressionTest.perf_variables) using the `extract_bw()` performance function.\nIf you are inheriting from a base test and you don’t want to override completely its performance variables, you could call instead `update()` on [`perf_variables`](index.html#reframe.core.pipeline.RegressionTest.perf_variables).\n\nFinally, you can convert any arbitrary function or [deferred expression](index.html#document-deferrable_functions_reference) into a performance function by calling the [`make_performance_function()`](index.html#reframe.utility.sanity.make_performance_function) utility as shown below:\n```\n@run_before('performance')\ndef set_perf_vars(self):\n self.perf_variables = {\n 'Copy': sn.make_performance_function(\n sn.extractsingle(r'Copy:\\s+(\\S+)\\s+.*',\n self.stdout, 1, float),\n 'MB/s'\n )\n }\n```\nNote that in this case, the newly created performance function is not assigned to a test’s performance variable and you will have to do this independently.\n\n###### Adding reference values[¶](#adding-reference-values)\n\nOn its current state, the above STREAM performance test will simply extract and report the performance variables regardless of the actual performance values.\nHowever, in some situations, it might be useful to check that the extracted performance values are within an expected range, and report a failure whenever a test performs below expectations.\nTo this end, ReFrame tests include the [`reference`](index.html#reframe.core.pipeline.RegressionTest.reference) variable, which enables setting references for each of the performance variables defined in a test and also set different references for different systems.\nIn the following example, we set the reference values for all the STREAM sub-benchmarks for the system we are currently running on.\n\nNote\n\nOptimizing STREAM benchmark performance is outside the scope of this tutorial.\n```\ncat tutorials/basics/stream/stream3.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class StreamWithRefTest(rfm.RegressionTest):\n valid_systems = ['*']\n valid_prog_environs = ['gnu']\n prebuild_cmds = [\n 'wget https://raw.githubusercontent.com/jeffhammond/STREAM/master/stream.c' # noqa: E501\n ]\n build_system = 'SingleSource'\n sourcepath = 'stream.c'\n env_vars = {\n 'OMP_NUM_THREADS': '4',\n 'OMP_PLACES': 'cores'\n }\n reference = {\n 'catalina': {\n 'Copy': (25200, -0.05, 0.05, 'MB/s'),\n 'Scale': (16800, -0.05, 0.05, 'MB/s'),\n 'Add': (18500, -0.05, 0.05, 'MB/s'),\n 'Triad': (18800, -0.05, 0.05, 'MB/s')\n }\n }\n\n @run_before('compile')\n def set_compiler_flags(self):\n self.build_system.cppflags = ['-DSTREAM_ARRAY_SIZE=$((1 << 25))']\n self.build_system.cflags = ['-fopenmp', '-O3', '-Wall']\n\n @sanity_function\n def validate_solution(self):\n return sn.assert_found(r'Solution Validates', self.stdout)\n\n @performance_function('MB/s')\n def extract_bw(self, kind='Copy'):\n '''Generic performance extraction function.'''\n\n if kind not in ('Copy', 'Scale', 'Add', 'Triad'):\n raise ValueError(f'illegal value in argument kind ({kind!r})')\n\n return sn.extractsingle(rf'{kind}:\\s+(\\S+)\\s+.*',\n self.stdout, 1, float)\n\n @run_before('performance')\n def set_perf_variables(self):\n '''Build the dictionary with all the performance variables.'''\n\n self.perf_variables = {\n 'Copy': self.extract_bw(),\n 'Scale': self.extract_bw('Scale'),\n 'Add': self.extract_bw('Add'),\n 'Triad': self.extract_bw('Triad'),\n }\n```\nThe performance reference tuple consists of the reference value, the lower and upper thresholds expressed as fractional numbers relative to the reference value, and the unit of measurement.\nIf any of the thresholds is not relevant, `None` may be used instead.\nAlso, the units in this [`reference`](index.html#reframe.core.pipeline.RegressionTest.reference) variable are entirely optional, since they were already provided through the `@performance_function` decorator.\n\nIf any obtained performance value is beyond its respective thresholds, the test will fail with a summary as shown below:\n```\n./bin/reframe -c tutorials/basics/stream/stream3.py -r --performance-report\n```\n```\nFAILURE INFO for StreamWithRefTest\n * Expanded name: StreamWithRefTest\n * Description:\n * System partition: catalina:default\n * Environment: gnu\n * Stage directory: /Users/user/Repositories/reframe/stage/catalina/default/gnu/StreamWithRefTest\n * Node list: tresa.localNone\n * Job type: local (id=4576)\n * Dependencies (conceptual): []\n * Dependencies (actual): []\n * Maintainers: []\n * Failing phase: performance\n * Rerun with '-n /f925207b -p gnu --system catalina:default -r'\n * Reason: performance error: failed to meet reference: Add=19585.3, expected 18500 (l=17575.0, u=19425.0)\n```\n###### Examining the performance logs[¶](#examining-the-performance-logs)\n\nReFrame has a powerful mechanism for logging its activities as well as performance data.\nIt supports different types of log channels and it can send data simultaneously in any number of them.\nFor example, performance data might be logged in files and at the same time being sent to Syslog or to a centralized log management server.\nBy default (i.e., starting off from the builtin configuration file), ReFrame sends performance data to files per test under the `perflogs/` directory:\n```\nperflogs\n└── catalina\n └── default\n ├── StreamTest.log\n └── StreamWithRefTest.log\n```\nReFrame creates a log file per test per system and per partition and appends to it every time the test is run on that system/partition combination.\nLet’s inspect the log file from our last test:\n```\ntail perflogs/catalina/default/StreamWithRefTest.log\n```\n```\njob_completion_time,version,display_name,system,partition,environ,jobid,result,Copy_value,Copy_unit,Copy_ref,Copy_lower,Copy_upper,Scale_value,Scale_unit,Scale_ref,Scale_lower,Scale_upper,Add_value,Add_unit,Add_ref,Add_lower,Add_upper,Triad_value,Triad_unit,Triad_ref,Triad_lower,Triad_upper 2022-10-18T21:41:25,4.0.0-dev.2+90fbd3ef,StreamWithRefTest,catalina,default,gnu,81351,pass,24235.6,MB/s,25200,-0.05,0.05,16044.2,MB/s,16800,-0.05,0.05,17733.7,MB/s,18500,-0.05,0.05,18232.0,MB/s,18800,-0.05,0.05 2022-10-18T21:41:31,4.0.0-dev.2+90fbd3ef,StreamWithRefTest,catalina,default,gnu,81377,fail,23615.4,MB/s,25200,-0.05,0.05,16394.5,MB/s,16800,-0.05,0.05,17841.3,MB/s,18500,-0.05,0.05,18284.1,MB/s,18800,-0.05,0.05 2022-10-18T21:46:06,4.0.0-dev.2+90fbd3ef,StreamWithRefTest,catalina,default,gnu,81480,fail,23736.4,MB/s,25200,-0.05,0.05,16242.8,MB/s,16800,-0.05,0.05,17699.1,MB/s,18500,-0.05,0.05,18077.3,MB/s,18800,-0.05,0.05\n```\nThe format of this file is controlled by [`handlers_perflog`](index.html#config.logging.handlers_perflog) logging configuration parameter and, by default, contains several information about the test.\nFor each test, all of its performance variables are logged along with their unit, the obtained value, the reference and the lower and upper threshold.\nThe default format is in CSV, so that it can be easily post-processed.\nFor this reason, a header is also printed to help identify the different fields.\n\nSince version 4.0, ReFrame is very cautious when generating this file: if a change is detected in the information that is being logged, ReFrame will not append to the file, but it will instead create a new one, saving the old file using the `.h<N>` suffix, where `N` is an integer that is increased every time a new file is being created due to such changes.\nExamples of changes in the logged information are when the log record format changes or a new performance metric is added, deleted or has its name changed.\nThis behavior guarantees that each log file is consistent and it will not break existing parsers.\n\nFor more information on configuring performance logging in ReFrame as well as logging in general, you may refer to the [Logging Configuration](index.html#logging-config-reference) reference.\n\n##### Porting The Tests to an HPC cluster[¶](#porting-the-tests-to-an-hpc-cluster)\n\nIt’s now time to port our tests to an HPC cluster.\nObviously, HPC clusters are much more complex than our laptop or PC.\nUsually there are many more compilers, the user environment is handled in a different way, and the way to launch the tests varies significantly, since you have to go through a workload manager in order to access the actual compute nodes.\nBesides that, there might be multiple types of compute nodes that we would like to run our tests on, but each type might be accessed in a different way.\nIt is already apparent that porting even an as simple as a “Hello, World” test to such a system is not that straightforward.\nAs we shall see in this section, ReFrame makes that pretty easy.\n\n###### Adapting the configuration[¶](#adapting-the-configuration)\n\nOur target system is the [Piz Daint](https://www.cscs.ch/computers/piz-daint/) supercomputer at CSCS, but you can adapt the process to your target HPC system.\nIn ReFrame, all the details of the various interactions of a test with the system environment are handled transparently and are set up in its configuration file.\nLet’s create a new configuration file for Piz Daint:\n```\nsite_configuration = {\n 'systems': [\n {\n 'name': 'daint',\n 'descr': 'Piz Daint Supercomputer',\n 'hostnames': ['daint'],\n 'modules_system': 'tmod32',\n 'partitions': [\n {\n 'name': 'login',\n 'descr': 'Login nodes',\n 'scheduler': 'local',\n 'launcher': 'local',\n 'environs': ['builtin', 'gnu', 'intel', 'nvidia', 'cray'],\n },\n {\n 'name': 'gpu',\n 'descr': 'Hybrid nodes',\n 'scheduler': 'slurm',\n 'launcher': 'srun',\n 'access': ['-C gpu', '-A csstaff'],\n 'environs': ['gnu', 'intel', 'nvidia', 'cray'],\n 'max_jobs': 100,\n },\n {\n 'name': 'mc',\n 'descr': 'Multicore nodes',\n 'scheduler': 'slurm',\n 'launcher': 'srun',\n 'access': ['-C mc', '-A csstaff'],\n 'environs': ['gnu', 'intel', 'nvidia', 'cray'],\n 'max_jobs': 100,\n 'resources': [\n {\n 'name': 'memory',\n 'options': ['--mem={size}']\n }\n ]\n }\n ]\n }\n ],\n 'environments': [\n {\n 'name': 'gnu',\n 'modules': ['PrgEnv-gnu'],\n 'cc': 'cc',\n 'cxx': 'CC',\n 'ftn': 'ftn',\n 'target_systems': ['daint']\n },\n {\n 'name': 'cray',\n 'modules': ['PrgEnv-cray'],\n 'cc': 'cc',\n 'cxx': 'CC',\n 'ftn': 'ftn',\n 'target_systems': ['daint']\n },\n {\n 'name': 'intel',\n 'modules': ['PrgEnv-intel'],\n 'cc': 'cc',\n 'cxx': 'CC',\n 'ftn': 'ftn',\n 'target_systems': ['daint']\n },\n {\n 'name': 'nvidia',\n 'modules': ['PrgEnv-nvidia'],\n 'cc': 'cc',\n 'cxx': 'CC',\n 'ftn': 'ftn',\n 'target_systems': ['daint']\n },\n {\n 'name': 'builtin',\n 'cc': 'cc',\n 'cxx': 'CC',\n 'ftn': 'ftn',\n 'target_systems': ['daint']\n }\n ] # end of environments\n}\n```\nFirst of all, we need to define a new system and set the list of hostnames that will help ReFrame identify it.\nWe also set the [`modules_system`](index.html#config.systems.modules_system) configuration parameter to instruct ReFrame that this system makes use of the [environment modules](http://modules.sourceforge.net/) for managing the user environment.\nThen we define the system partitions that we want to test.\nIn this case, we define three partitions:\n\n1. the login nodes,\n2. the multicore partition (2x Broadwell CPUs per node) and 3. the hybrid partition (1x Haswell CPU + 1x Pascal GPU).\n\nThe login nodes are pretty much similar to the `tresa:default` partition which corresponded to our laptop: tests will be launched and run locally.\nThe other two partitions are handled by [Slurm](https://slurm.schedmd.com/) and parallel jobs are launched using the [`srun`](https://slurm.schedmd.com/srun.html) command.\nAdditionally, in order to access the different types of nodes represented by those partitions, users have to specify either `-C mc` or `-C gpu` options along with their account.\nThis is what we do exactly with the [`access`](index.html#config.systems.partitions.access) partition configuration option.\n\nNote\n\nSystem partitions in ReFrame do not necessarily correspond to real job scheduler partitions.\n\nPiz Daint’s programming environment offers four compilers: Cray, GNU, Intel and NVIDIA.\nWe want to test all of them, so we include them in the [`environs`](index.html#config.systems.partitions.environs) lists.\nNotice that we do not include Clang in the list, since there is no such compiler on this particular system.\nOn the other hand, we include a different version of the `builtin` environment, which corresponds to the default login environment without loading any modules.\nIt is generally useful to define such an environment so as to use it for tests that are running simple utilities and don’t need to compile anything.\n\nBefore looking into the definition of the new environments for the four compilers, it is worth mentioning the [`max_jobs`](index.html#config.systems.partitions.max_jobs) parameter.\nThis parameter specifies the maximum number of ReFrame test jobs that can be simultaneously in flight.\nReFrame will try to keep concurrency close to this limit (but not exceeding it).\nBy default, this is set to `8`, so you are advised to set it to a higher number if you want to increase the throughput of completed tests.\n\nThe new environments are defined similarly to the ones we had for our local system, except that now we add also the [`modules`](index.html#config.environments.modules) parameter.\nThe [`modules`](index.html#config.environments.modules) parameter is a list of environment modules that needs to be loaded, in order to make available this compiler.\n\n###### Running the tests[¶](#running-the-tests)\n\nWe are now ready to run our tests on Piz Daint.\nWe will only do so with the final versions of the tests from the previous section, which we will select using [`-n`](index.html#cmdoption-n) option.\n```\nexport RFM_CONFIG_FILES=$(pwd)/tutorials/config/daint.py\n./bin/reframe -c tutorials/basics/ -R -n 'HelloMultiLangTest|HelloThreadedExtended2Test|StreamWithRefTest' --performance-report -r\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2\n command: './bin/reframe -c tutorials/basics/ -R -n HelloMultiLangTest|HelloThreadedExtended2Test|StreamWithRefTest --performance-report -r'\n launched by: user@host\n working directory: '/home/user/Devel/reframe'\n settings files: '<builtin>', '/home/user/Devel/reframe/tutorials/config/daint.py'\n check search path: (R) '/home/user/Devel/reframe/tutorials/basics'\n stage directory: '/home/user/Devel/reframe/stage'\n output directory: '/home/user/Devel/reframe/output'\n log files: '/tmp/rfm-nyqs7jb9.log'\n\n[===] Running 4 check(s)\n[===] Started on Tue Nov 15 18:20:32 2022\n\n[---] start processing checks\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @daint:login+builtin\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @daint:login+gnu\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @daint:login+intel\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @daint:login+nvidia\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @daint:login+cray\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @daint:gpu+gnu\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @daint:gpu+intel\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @daint:gpu+nvidia\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @daint:gpu+cray\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @daint:mc+gnu\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @daint:mc+intel\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @daint:mc+nvidia\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @daint:mc+cray\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @daint:login+builtin\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @daint:login+gnu\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @daint:login+intel\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @daint:login+nvidia\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @daint:login+cray\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @daint:gpu+gnu\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @daint:gpu+intel\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @daint:gpu+nvidia\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @daint:gpu+cray\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @daint:mc+gnu\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @daint:mc+intel\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @daint:mc+nvidia\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @daint:mc+cray\n[ RUN ] HelloThreadedExtended2Test /57223829 @daint:login+builtin\n[ RUN ] HelloThreadedExtended2Test /57223829 @daint:login+gnu\n[ RUN ] HelloThreadedExtended2Test /57223829 @daint:login+intel\n[ RUN ] HelloThreadedExtended2Test /57223829 @daint:login+nvidia\n[ RUN ] HelloThreadedExtended2Test /57223829 @daint:login+cray\n[ RUN ] HelloThreadedExtended2Test /57223829 @daint:gpu+gnu\n[ RUN ] HelloThreadedExtended2Test /57223829 @daint:gpu+intel\n[ RUN ] HelloThreadedExtended2Test /57223829 @daint:gpu+nvidia\n[ RUN ] HelloThreadedExtended2Test /57223829 @daint:gpu+cray\n[ RUN ] HelloThreadedExtended2Test /57223829 @daint:mc+gnu\n[ RUN ] HelloThreadedExtended2Test /57223829 @daint:mc+intel\n[ RUN ] HelloThreadedExtended2Test /57223829 @daint:mc+nvidia\n[ RUN ] HelloThreadedExtended2Test /57223829 @daint:mc+cray\n[ RUN ] StreamWithRefTest /f925207b @daint:login+gnu\n[ RUN ] StreamWithRefTest /f925207b @daint:gpu+gnu\n[ RUN ] StreamWithRefTest /f925207b @daint:mc+gnu\n[ OK ] ( 1/42) HelloMultiLangTest %lang=cpp /71bf65a3 @daint:login+builtin\n[ OK ] ( 2/42) HelloMultiLangTest %lang=cpp /71bf65a3 @daint:login+gnu\n[ OK ] ( 3/42) HelloMultiLangTest %lang=cpp /71bf65a3 @daint:login+intel\n[ OK ] ( 4/42) HelloMultiLangTest %lang=cpp /71bf65a3 @daint:login+nvidia\n[ OK ] ( 5/42) HelloMultiLangTest %lang=cpp /71bf65a3 @daint:login+cray\n[ OK ] ( 6/42) HelloMultiLangTest %lang=c /7cfa870e @daint:login+builtin\n[ OK ] ( 7/42) HelloMultiLangTest %lang=c /7cfa870e @daint:login+gnu\n[ OK ] ( 8/42) HelloMultiLangTest %lang=c /7cfa870e @daint:login+intel\n[ OK ] ( 9/42) HelloMultiLangTest %lang=c /7cfa870e @daint:login+nvidia\n[ OK ] (10/42) HelloMultiLangTest %lang=c /7cfa870e @daint:login+cray\n[ OK ] (11/42) HelloMultiLangTest %lang=cpp /71bf65a3 @daint:gpu+cray\n[ OK ] (12/42) HelloMultiLangTest %lang=cpp /71bf65a3 @daint:mc+nvidia\n[ OK ] (13/42) HelloMultiLangTest %lang=cpp /71bf65a3 @daint:mc+cray\n[ OK ] (14/42) HelloMultiLangTest %lang=c /7cfa870e @daint:mc+cray\n[ OK ] (15/42) HelloMultiLangTest %lang=cpp /71bf65a3 @daint:gpu+nvidia\n[ OK ] (16/42) HelloMultiLangTest %lang=c /7cfa870e @daint:gpu+intel\n[ OK ] (17/42) HelloMultiLangTest %lang=c /7cfa870e @daint:gpu+nvidia\n[ OK ] (18/42) HelloMultiLangTest %lang=c /7cfa870e @daint:mc+intel\n[ OK ] (19/42) HelloThreadedExtended2Test /57223829 @daint:login+builtin\n[ OK ] (20/42) HelloThreadedExtended2Test /57223829 @daint:login+gnu\n[ OK ] (21/42) HelloThreadedExtended2Test /57223829 @daint:login+intel\n[ OK ] (22/42) HelloMultiLangTest %lang=c /7cfa870e @daint:gpu+cray\n[ OK ] (23/42) HelloMultiLangTest %lang=c /7cfa870e @daint:mc+gnu\n[ OK ] (24/42) HelloThreadedExtended2Test /57223829 @daint:login+nvidia\n[ OK ] (25/42) HelloThreadedExtended2Test /57223829 @daint:login+cray\n[ OK ] (26/42) HelloMultiLangTest %lang=c /7cfa870e @daint:mc+nvidia\n[ OK ] (27/42) HelloMultiLangTest %lang=cpp /71bf65a3 @daint:gpu+gnu\n[ OK ] (28/42) HelloMultiLangTest %lang=cpp /71bf65a3 @daint:gpu+intel\n[ OK ] (29/42) HelloMultiLangTest %lang=cpp /71bf65a3 @daint:mc+gnu\n[ OK ] (30/42) HelloMultiLangTest %lang=cpp /71bf65a3 @daint:mc+intel\n[ OK ] (31/42) HelloMultiLangTest %lang=c /7cfa870e @daint:gpu+gnu\n[ OK ] (32/42) StreamWithRefTest /f925207b @daint:login+gnu P: Copy: 71061.6 MB/s (r:0, l:None, u:None)\nP: Scale: 44201.5 MB/s (r:0, l:None, u:None)\nP: Add: 48178.5 MB/s (r:0, l:None, u:None)\nP: Triad: 48063.3 MB/s (r:0, l:None, u:None)\n[ OK ] (33/42) HelloThreadedExtended2Test /57223829 @daint:mc+cray\n[ OK ] (34/42) HelloThreadedExtended2Test /57223829 @daint:mc+intel\n[ OK ] (35/42) HelloThreadedExtended2Test /57223829 @daint:mc+gnu\n[ OK ] (36/42) HelloThreadedExtended2Test /57223829 @daint:mc+nvidia\n[ OK ] (37/42) StreamWithRefTest /f925207b @daint:mc+gnu P: Copy: 52660.1 MB/s (r:0, l:None, u:None)\nP: Scale: 33117.6 MB/s (r:0, l:None, u:None)\nP: Add: 34876.7 MB/s (r:0, l:None, u:None)\nP: Triad: 35150.7 MB/s (r:0, l:None, u:None)\n[ OK ] (38/42) HelloThreadedExtended2Test /57223829 @daint:gpu+intel\n[ OK ] (39/42) HelloThreadedExtended2Test /57223829 @daint:gpu+cray\n[ OK ] (40/42) HelloThreadedExtended2Test /57223829 @daint:gpu+nvidia\n[ OK ] (41/42) HelloThreadedExtended2Test /57223829 @daint:gpu+gnu\n[ OK ] (42/42) StreamWithRefTest /f925207b @daint:gpu+gnu P: Copy: 49682.3 MB/s (r:0, l:None, u:None)\nP: Scale: 34452.3 MB/s (r:0, l:None, u:None)\nP: Add: 38030.7 MB/s (r:0, l:None, u:None)\nP: Triad: 38379.0 MB/s (r:0, l:None, u:None)\n[---] all spawned checks have finished\n\n[ PASSED ] Ran 42/42 test case(s) from 4 check(s) (0 failure(s), 0 skipped)\n[===] Finished on Tue Nov 15 18:22:48 2022\n\n===\nPERFORMANCE REPORT\n---\n[StreamWithRefTest /f925207b @daint:login:gnu]\n num_gpus_per_node: 0\n num_tasks: 1\n performance:\n - Copy: 71061.6 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 44201.5 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 48178.5 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 48063.3 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n[StreamWithRefTest /f925207b @daint:gpu:gnu]\n num_gpus_per_node: 0\n num_tasks: 1\n performance:\n - Copy: 49682.3 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 34452.3 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 38030.7 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 38379.0 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n[StreamWithRefTest /f925207b @daint:mc:gnu]\n num_gpus_per_node: 0\n num_tasks: 1\n performance:\n - Copy: 52660.1 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 33117.6 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 34876.7 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 35150.7 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n---\nRun report saved in '/home/user/.reframe/reports/run-report-1.json'\nLog file(s) saved in '/tmp/rfm-nyqs7jb9.log'\n```\nThere it is!\nWithout any change in our tests, we could simply run them in a HPC cluster with all of its intricacies.\nNotice how our original four tests expanded to more than 40 test cases on that particular HPC cluster!\nOne reason we could run immediately our tests on a new system was that we have not been restricting neither the valid system they can run nor the valid programming environments they can run with (except for the STREAM test).\nOtherwise we would have to add `daint` and its corresponding programming environments in `valid_systems` and `valid_prog_environs` lists respectively.\n\nTip\n\nA quick way to try a test on a new system, if it’s not generic, is to pass the [`--skip-system-check`](index.html#cmdoption-skip-system-check) and the [`--skip-prgenv-check`](index.html#cmdoption-skip-prgenv-check) command line options which will cause ReFrame to skip any test validity checks for systems or programming environments.\n\nAlthough the tests remain the same, ReFrame has generated completely different job scripts for each test depending on where it was going to run.\nLet’s check the job script generated for the `StreamWithRefTest`:\n```\ncat output/daint/gpu/gnu/StreamWithRefTest/rfm_StreamWithRefTest_job.sh\n```\n```\n#!/bin/bash\n#SBATCH --job-name=\"rfm_StreamWithRefTest_job\"\n#SBATCH --ntasks=1\n#SBATCH --output=rfm_StreamWithRefTest_job.out\n#SBATCH --error=rfm_StreamWithRefTest_job.err\n#SBATCH --time=0:10:0\n#SBATCH -A csstaff\n#SBATCH --constraint=gpu module unload PrgEnv-cray module load PrgEnv-gnu export OMP_NUM_THREADS=4 export OMP_PLACES=cores srun ./StreamWithRefTest\n```\nWhereas the exact same test running on our laptop was as simple as the following:\n```\n#!/bin/bash export OMP_NUM_THREADS=4 export OMP_PLACES=cores\n ./StreamWithRefTest\n```\nIn ReFrame, you don’t have to care about all the system interaction details, but rather about the logic of your tests as we shall see in the next section.\n\n###### Adapting a test to new systems and programming environments[¶](#adapting-a-test-to-new-systems-and-programming-environments)\n\nUnless a test is rather generic, you will need to make some adaptations for the system that you port it to.\nIn this case, we will adapt the STREAM benchmark so as to run it with multiple compiler and adjust its execution based on the target architecture of each partition.\nLet’s see and comment the changes:\n```\ncat tutorials/basics/stream/stream4.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class StreamMultiSysTest(rfm.RegressionTest):\n valid_systems = ['*']\n valid_prog_environs = ['cray', 'gnu', 'intel', 'nvidia']\n prebuild_cmds = [\n 'wget https://raw.githubusercontent.com/jeffhammond/STREAM/master/stream.c' # noqa: E501\n ]\n build_system = 'SingleSource'\n sourcepath = 'stream.c'\n env_vars = {\n 'OMP_NUM_THREADS': 4,\n 'OMP_PLACES': 'cores'\n }\n reference = {\n 'catalina': {\n 'Copy': (25200, -0.05, 0.05, 'MB/s'),\n 'Scale': (16800, -0.05, 0.05, 'MB/s'),\n 'Add': (18500, -0.05, 0.05, 'MB/s'),\n 'Triad': (18800, -0.05, 0.05, 'MB/s')\n }\n }\n\n # Flags per programming environment\n flags = variable(dict, value={\n 'cray': ['-fopenmp', '-O3', '-Wall'],\n 'gnu': ['-fopenmp', '-O3', '-Wall'],\n 'intel': ['-qopenmp', '-O3', '-Wall'],\n 'nvidia': ['-mp', '-O3']\n })\n\n # Number of cores for each system\n cores = variable(dict, value={\n 'catalina:default': 4,\n 'daint:gpu': 12,\n 'daint:mc': 36,\n 'daint:login': 10\n })\n\n @run_before('compile')\n def set_compiler_flags(self):\n self.build_system.cppflags = ['-DSTREAM_ARRAY_SIZE=$((1 << 25))']\n environ = self.current_environ.name\n self.build_system.cflags = self.flags.get(environ, [])\n\n @run_before('run')\n def set_num_threads(self):\n num_threads = self.cores.get(self.current_partition.fullname, 1)\n self.num_cpus_per_task = num_threads\n self.env_vars = {\n 'OMP_NUM_THREADS': num_threads,\n 'OMP_PLACES': 'cores'\n }\n\n @sanity_function\n def validate_solution(self):\n return sn.assert_found(r'Solution Validates', self.stdout)\n\n @performance_function('MB/s')\n def extract_bw(self, kind='Copy'):\n if kind not in {'Copy', 'Scale', 'Add', 'Triad'}:\n raise ValueError(f'illegal value in argument kind ({kind!r})')\n\n return sn.extractsingle(rf'{kind}:\\s+(\\S+)\\s+.*',\n self.stdout, 1, float)\n\n @run_before('performance')\n def set_perf_variables(self):\n self.perf_variables = {\n 'Copy': self.extract_bw(),\n 'Scale': self.extract_bw('Scale'),\n 'Add': self.extract_bw('Add'),\n 'Triad': self.extract_bw('Triad'),\n }\n```\nFirst of all, we need to add the new programming environments in the list of the supported ones.\nNow there is the problem that each compiler has its own flags for enabling OpenMP, so we need to differentiate the behavior of the test based on the programming environment.\nFor this reason, we define the flags for each compiler in a separate dictionary (`flags` variable) and we set them in the `set_compiler_flags()` pipeline hook.\nWe have first seen the pipeline hooks in the multithreaded “Hello, World!” example and now we explain them in more detail.\nWhen ReFrame loads a test file, it instantiates all the tests it finds in it.\nBased on the system ReFrame runs on and the supported environments of the tests, it will generate different test cases for each system partition and environment combination and it will finally send the test cases for execution.\nDuring its execution, a test case goes through the *regression test pipeline*, which is a series of well defined phases.\nUsers can attach arbitrary functions to run before or after any pipeline stage and this is exactly what the `set_compiler_flags()` function is.\nWe instruct ReFrame to run this function before the test enters the `compile` stage and set accordingly the compilation flags.\nThe system partition and the programming environment of the currently running test case are available to a ReFrame test through the [`current_partition`](index.html#reframe.core.pipeline.RegressionTest.current_partition) and [`current_environ`](index.html#reframe.core.pipeline.RegressionTest.current_environ) attributes respectively.\nThese attributes, however, are only set after the first stage (`setup`) of the pipeline is executed, so we can’t use them inside the test’s constructor.\n\nWe do exactly the same for setting the `OMP_NUM_THREADS` environment variables depending on the system partition we are running on, by attaching the `set_num_threads()` pipeline hook to the `run` phase of the test.\nIn that same hook we also set the [`num_cpus_per_task`](index.html#reframe.core.pipeline.RegressionTest.num_cpus_per_task) attribute of the test, so as to instruct the backend job scheduler to properly assign CPU cores to the test.\nIn ReFrame tests you can set a series of task allocation attributes that will be used by the backend schedulers to emit the right job submission script.\nThe section [Mapping of Test Attributes to Job Scheduler Backends](index.html#scheduler-options) of the [Test API Reference](index.html#document-regression_test_api) summarizes these attributes and the actual backend scheduler options that they correspond to.\n\nFor more information about the regression test pipeline and how ReFrame executes the tests in general, have a look at [How ReFrame Executes Tests](index.html#document-pipeline).\n\nNote\n\nReFrame tests are ordinary Python classes so you can define your own attributes as we do with `flags` and `cores` in this example.\n\nLet’s run our adapted test now:\n```\n./bin/reframe -c tutorials/basics/stream/stream4.py -r --performance-report\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2\n command: './bin/reframe -c tutorials/basics/stream/stream4.py -r --performance-report'\n launched by: user@host\n working directory: '/home/user/Devel/reframe'\n settings files: '<builtin>', '/home/user/Devel/reframe/tutorials/config/daint.py'\n check search path: '/home/user/Devel/reframe/tutorials/basics/stream/stream4.py'\n stage directory: '/home/user/Devel/reframe/stage'\n output directory: '/home/user/Devel/reframe/output'\n log files: '/tmp/rfm-yf6xjn_4.log'\n\n[===] Running 1 check(s)\n[===] Started on Tue Nov 15 18:22:48 2022\n\n[---] start processing checks\n[ RUN ] StreamMultiSysTest /eec1c676 @daint:login+gnu\n[ RUN ] StreamMultiSysTest /eec1c676 @daint:login+intel\n[ RUN ] StreamMultiSysTest /eec1c676 @daint:login+nvidia\n[ RUN ] StreamMultiSysTest /eec1c676 @daint:login+cray\n[ RUN ] StreamMultiSysTest /eec1c676 @daint:gpu+gnu\n[ RUN ] StreamMultiSysTest /eec1c676 @daint:gpu+intel\n[ RUN ] StreamMultiSysTest /eec1c676 @daint:gpu+nvidia\n[ RUN ] StreamMultiSysTest /eec1c676 @daint:gpu+cray\n[ RUN ] StreamMultiSysTest /eec1c676 @daint:mc+gnu\n[ RUN ] StreamMultiSysTest /eec1c676 @daint:mc+intel\n[ RUN ] StreamMultiSysTest /eec1c676 @daint:mc+nvidia\n[ RUN ] StreamMultiSysTest /eec1c676 @daint:mc+cray\n[ OK ] ( 1/12) StreamMultiSysTest /eec1c676 @daint:login+gnu P: Copy: 97772.6 MB/s (r:0, l:None, u:None)\nP: Scale: 69418.6 MB/s (r:0, l:None, u:None)\nP: Add: 71941.0 MB/s (r:0, l:None, u:None)\nP: Triad: 73679.7 MB/s (r:0, l:None, u:None)\n[ OK ] ( 2/12) StreamMultiSysTest /eec1c676 @daint:login+intel P: Copy: 85123.0 MB/s (r:0, l:None, u:None)\nP: Scale: 79701.7 MB/s (r:0, l:None, u:None)\nP: Add: 81632.7 MB/s (r:0, l:None, u:None)\nP: Triad: 44391.5 MB/s (r:0, l:None, u:None)\n[ OK ] ( 3/12) StreamMultiSysTest /eec1c676 @daint:login+nvidia P: Copy: 76641.4 MB/s (r:0, l:None, u:None)\nP: Scale: 59041.9 MB/s (r:0, l:None, u:None)\nP: Add: 64792.5 MB/s (r:0, l:None, u:None)\nP: Triad: 69441.4 MB/s (r:0, l:None, u:None)\n[ OK ] ( 4/12) StreamMultiSysTest /eec1c676 @daint:login+cray P: Copy: 35658.5 MB/s (r:0, l:None, u:None)\nP: Scale: 27732.2 MB/s (r:0, l:None, u:None)\nP: Add: 39037.7 MB/s (r:0, l:None, u:None)\nP: Triad: 45310.3 MB/s (r:0, l:None, u:None)\n[ OK ] ( 5/12) StreamMultiSysTest /eec1c676 @daint:gpu+gnu P: Copy: 42666.3 MB/s (r:0, l:None, u:None)\nP: Scale: 38491.0 MB/s (r:0, l:None, u:None)\nP: Add: 43686.4 MB/s (r:0, l:None, u:None)\nP: Triad: 43466.6 MB/s (r:0, l:None, u:None)\n[ OK ] ( 6/12) StreamMultiSysTest /eec1c676 @daint:gpu+intel P: Copy: 51726.7 MB/s (r:0, l:None, u:None)\nP: Scale: 54185.6 MB/s (r:0, l:None, u:None)\nP: Add: 57608.3 MB/s (r:0, l:None, u:None)\nP: Triad: 57390.7 MB/s (r:0, l:None, u:None)\n[ OK ] ( 7/12) StreamMultiSysTest /eec1c676 @daint:gpu+nvidia P: Copy: 51810.8 MB/s (r:0, l:None, u:None)\nP: Scale: 39653.4 MB/s (r:0, l:None, u:None)\nP: Add: 44008.0 MB/s (r:0, l:None, u:None)\nP: Triad: 44384.4 MB/s (r:0, l:None, u:None)\n[ OK ] ( 8/12) StreamMultiSysTest /eec1c676 @daint:gpu+cray P: Copy: 51101.8 MB/s (r:0, l:None, u:None)\nP: Scale: 38568.1 MB/s (r:0, l:None, u:None)\nP: Add: 43193.6 MB/s (r:0, l:None, u:None)\nP: Triad: 43142.9 MB/s (r:0, l:None, u:None)\n[ OK ] ( 9/12) StreamMultiSysTest /eec1c676 @daint:mc+gnu P: Copy: 48292.9 MB/s (r:0, l:None, u:None)\nP: Scale: 38499.5 MB/s (r:0, l:None, u:None)\nP: Add: 43555.7 MB/s (r:0, l:None, u:None)\nP: Triad: 43871.4 MB/s (r:0, l:None, u:None)\n[ OK ] (10/12) StreamMultiSysTest /eec1c676 @daint:mc+cray P: Copy: 46538.3 MB/s (r:0, l:None, u:None)\nP: Scale: 40133.3 MB/s (r:0, l:None, u:None)\nP: Add: 43363.9 MB/s (r:0, l:None, u:None)\nP: Triad: 43450.3 MB/s (r:0, l:None, u:None)\n[ OK ] (11/12) StreamMultiSysTest /eec1c676 @daint:mc+nvidia P: Copy: 46648.2 MB/s (r:0, l:None, u:None)\nP: Scale: 40384.5 MB/s (r:0, l:None, u:None)\nP: Add: 44001.1 MB/s (r:0, l:None, u:None)\nP: Triad: 44489.7 MB/s (r:0, l:None, u:None)\n[ OK ] (12/12) StreamMultiSysTest /eec1c676 @daint:mc+intel P: Copy: 51335.9 MB/s (r:0, l:None, u:None)\nP: Scale: 49490.3 MB/s (r:0, l:None, u:None)\nP: Add: 56859.9 MB/s (r:0, l:None, u:None)\nP: Triad: 56544.5 MB/s (r:0, l:None, u:None)\n[---] all spawned checks have finished\n\n[ PASSED ] Ran 12/12 test case(s) from 1 check(s) (0 failure(s), 0 skipped)\n[===] Finished on Tue Nov 15 18:24:00 2022\n\n===\nPERFORMANCE REPORT\n---\n[StreamMultiSysTest /eec1c676 @daint:login:gnu]\n num_cpus_per_task: 10\n num_gpus_per_node: 0\n num_tasks: 1\n performance:\n - Copy: 97772.6 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 69418.6 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 71941.0 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 73679.7 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n[StreamMultiSysTest /eec1c676 @daint:login:intel]\n num_cpus_per_task: 10\n num_gpus_per_node: 0\n num_tasks: 1\n performance:\n - Copy: 85123.0 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 79701.7 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 81632.7 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 44391.5 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n[StreamMultiSysTest /eec1c676 @daint:login:nvidia]\n num_cpus_per_task: 10\n num_gpus_per_node: 0\n num_tasks: 1\n performance:\n - Copy: 76641.4 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 59041.9 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 64792.5 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 69441.4 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n[StreamMultiSysTest /eec1c676 @daint:login:cray]\n num_cpus_per_task: 10\n num_gpus_per_node: 0\n num_tasks: 1\n performance:\n - Copy: 35658.5 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 27732.2 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 39037.7 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 45310.3 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n[StreamMultiSysTest /eec1c676 @daint:gpu:gnu]\n num_cpus_per_task: 12\n num_gpus_per_node: 0\n num_tasks: 1\n performance:\n - Copy: 42666.3 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 38491.0 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 43686.4 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 43466.6 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n[StreamMultiSysTest /eec1c676 @daint:gpu:intel]\n num_cpus_per_task: 12\n num_gpus_per_node: 0\n num_tasks: 1\n performance:\n - Copy: 51726.7 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 54185.6 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 57608.3 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 57390.7 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n[StreamMultiSysTest /eec1c676 @daint:gpu:nvidia]\n num_cpus_per_task: 12\n num_gpus_per_node: 0\n num_tasks: 1\n performance:\n - Copy: 51810.8 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 39653.4 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 44008.0 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 44384.4 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n[StreamMultiSysTest /eec1c676 @daint:gpu:cray]\n num_cpus_per_task: 12\n num_gpus_per_node: 0\n num_tasks: 1\n performance:\n - Copy: 51101.8 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 38568.1 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 43193.6 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 43142.9 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n[StreamMultiSysTest /eec1c676 @daint:mc:gnu]\n num_cpus_per_task: 36\n num_gpus_per_node: 0\n num_tasks: 1\n performance:\n - Copy: 48292.9 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 38499.5 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 43555.7 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 43871.4 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n[StreamMultiSysTest /eec1c676 @daint:mc:intel]\n num_cpus_per_task: 36\n num_gpus_per_node: 0\n num_tasks: 1\n performance:\n - Copy: 51335.9 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 49490.3 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 56859.9 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 56544.5 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n[StreamMultiSysTest /eec1c676 @daint:mc:nvidia]\n num_cpus_per_task: 36\n num_gpus_per_node: 0\n num_tasks: 1\n performance:\n - Copy: 46648.2 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 40384.5 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 44001.1 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 44489.7 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n[StreamMultiSysTest /eec1c676 @daint:mc:cray]\n num_cpus_per_task: 36\n num_gpus_per_node: 0\n num_tasks: 1\n performance:\n - Copy: 46538.3 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Scale: 40133.3 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Add: 43363.9 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n - Triad: 43450.3 MB/s (r: 0 MB/s l: -inf% u: +inf%)\n---\nRun report saved in '/home/user/.reframe/reports/run-report-2.json'\nLog file(s) saved in '/tmp/rfm-yf6xjn_4.log'\n```\nNotice the improved performance of the benchmark in all partitions and the differences in performance between the different compilers.\n\nThis concludes our introductory tutorial to ReFrame!\n\n#### Tutorial 2: Customizing Further a Regression Test[¶](#tutorial-2-customizing-further-a-regression-test)\n\nIn this tutorial we will present common patterns that can come up when writing regression tests with ReFrame.\nAll examples use either the configuration files presented in [Tutorial 1: Getting Started with ReFrame](index.html#document-tutorial_basics), which you can find in `tutorials/config/tresa.py` and `tutorials/config/daint.py` or build on top of them.\nWe also assume that the reader is already familiar with the concepts presented in the basic tutorial.\nFinally, to avoid specifying the tutorial configuration every time, make sure to export it here:\n```\nexport RFM_CONFIG_FILES=$(pwd)/tutorials/config/tresa.py:$(pwd)/tutorials/config/daint.py\n```\n##### Parameterizing a Regression Test[¶](#parameterizing-a-regression-test)\n\nWe have briefly looked into parameterized tests in [Tutorial 1: Getting Started with ReFrame](index.html#document-tutorial_basics) where we parameterized the “Hello, World!” test based on the programming language.\nTest parameterization in ReFrame is quite powerful since it allows you to create a multitude of similar tests automatically.\nIn this example, we will parameterize the last version of the STREAM test from the [Tutorial 1: Getting Started with ReFrame](index.html#document-tutorial_basics) by changing the array size, so as to check the bandwidth of the different cache levels.\nHere is the adapted code with the relevant parts highlighted (for simplicity, we are interested only in the “Triad” benchmark):\n```\ncat tutorials/advanced/parameterized/stream.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class StreamMultiSysTest(rfm.RegressionTest):\n num_bytes = parameter(1 << pow for pow in range(19, 30))\n array_size = variable(int)\n ntimes = variable(int)\n\n valid_systems = ['*']\n valid_prog_environs = ['cray', 'gnu', 'intel', 'nvidia']\n prebuild_cmds = [\n 'wget https://raw.githubusercontent.com/jeffhammond/STREAM/master/stream.c' # noqa: E501\n ]\n build_system = 'SingleSource'\n sourcepath = 'stream.c'\n env_vars = {\n 'OMP_NUM_THREADS': '4',\n 'OMP_PLACES': 'cores'\n }\n reference = {\n '*': {\n 'Triad': (0, None, None, 'MB/s'),\n }\n }\n\n # Flags per programming environment\n flags = variable(dict, value={\n 'cray': ['-fopenmp', '-O3', '-Wall'],\n 'gnu': ['-fopenmp', '-O3', '-Wall'],\n 'intel': ['-qopenmp', '-O3', '-Wall'],\n 'nvidia': ['-mp', '-O3']\n })\n\n # Number of cores for each system\n cores = variable(dict, value={\n 'catalina:default': 4,\n 'daint:gpu': 12,\n 'daint:mc': 36,\n 'daint:login': 10\n })\n\n @run_after('init')\n def setup_build(self):\n self.array_size = (self.num_bytes >> 3) // 3\n self.ntimes = 100*1024*1024 // self.array_size\n self.descr = (\n f'STREAM test (array size: {self.array_size}, '\n f'ntimes: {self.ntimes})'\n )\n\n @run_before('compile')\n def set_compiler_flags(self):\n self.build_system.cppflags = [f'-DSTREAM_ARRAY_SIZE={self.array_size}',\n f'-DNTIMES={self.ntimes}']\n environ = self.current_environ.name\n self.build_system.cflags = self.flags.get(environ, [])\n\n @run_before('run')\n def set_num_threads(self):\n num_threads = self.cores.get(self.current_partition.fullname, 1)\n self.num_cpus_per_task = num_threads\n self.env_vars = {\n 'OMP_NUM_THREADS': num_threads,\n 'OMP_PLACES': 'cores'\n }\n\n @sanity_function\n def validate_solution(self):\n return sn.assert_found(r'Solution Validates', self.stdout)\n\n @performance_function('MB/s', perf_key='Triad')\n def extract_triad_bw(self):\n return sn.extractsingle(r'Triad:\\s+(\\S+)\\s+.*', self.stdout, 1, float)\n```\nAny ordinary ReFrame test becomes a parameterized one if the user defines parameters inside the class body of the test.\nThis is done using the `parameter()` ReFrame built-in function, which accepts the list of parameter values.\nFor each parameter value ReFrame will instantiate a different regression test by assigning the corresponding value to an attribute named after the parameter.\nSo in this example, ReFrame will generate automatically 11 tests with different values for their `num_bytes` attribute.\nFrom this point on, you can adapt the test based on the parameter values, as we do in this case, where we compute the STREAM array sizes, as well as the number of iterations to be performed on each benchmark, and we also compile the code accordingly.\n\nLet’s try listing the generated tests:\n```\n./bin/reframe -c tutorials/advanced/parameterized/stream.py -l\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe -c tutorials/advanced/parameterized/stream.py -l'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>', '/home/user/Repositories/reframe/tutorials/config/tresa.py'\n check search path: '/home/user/Repositories/reframe/tutorials/advanced/parameterized/stream.py'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/<KEY>'\n\n[List of matched checks]\n- StreamMultiSysTest %num_bytes=536870912 /cf10843f\n- StreamMultiSysTest %num_bytes=268435456 /97fb363f\n- StreamMultiSysTest %num_bytes=134217728 /7b4d01d3\n- StreamMultiSysTest %num_bytes=67108864 /530b0154\n- StreamMultiSysTest %num_bytes=33554432 /7199fc93\n- StreamMultiSysTest %num_bytes=16777216 /9d1b9ea8\n- StreamMultiSysTest %num_bytes=8388608 /3f29039f\n- StreamMultiSysTest %num_bytes=4194304 /e30054cd\n- StreamMultiSysTest %num_bytes=2097152 /45efaec5\n- StreamMultiSysTest %num_bytes=1048576 /92327981\n- StreamMultiSysTest %num_bytes=524288 /eb104cd0 Found 11 check(s)\n\nLog file(s) saved in '/<KEY>rfm-lou71c2g.log'\n```\nReFrame generates 11 tests from the single parameterized test.\nWhen listing parameterized tests, ReFrame adds the list of parameters after the base test name using the notation `%<param>=<value>`.\nEach generated test gets also a unique name.\nFor more details on how the test names are generated for various types of tests, please refer to [Test Naming Scheme](index.html#test-naming-scheme).\n\nTest parameterization in ReFrame is very powerful since you can parameterize your tests on anything and you can create complex parameterization spaces.\nA common pattern is to parameterize a test on the environment module that loads a software in order to test different versions of it.\nFor this reason, ReFrame offers the [`find_modules()`](index.html#reframe.utility.find_modules) function, which allows you to parameterize a test on the available modules for a given programming environment and partition combination.\nThe following example will create a test for each `GROMACS` module found on the software stack associated with a system partition and programming environment (toolchain):\n```\nimport reframe as rfm import reframe.utility as util\n\n@rfm.simple_test class MyTest(rfm.RegressionTest):\n module_info = parameter(util.find_modules('GROMACS'))\n\n @run_after('init')\n def process_module_info(self):\n s, e, m = self.module_info\n self.valid_systems = [s]\n self.valid_prog_environs = [e]\n self.modules = [m]\n```\n##### More On Building Tests[¶](#more-on-building-tests)\n\nWe have already seen how ReFrame can compile a test with a single source file.\nHowever, ReFrame can also build tests that use Make or a configure-Make approach.\nWe are going to demonstrate this through a simple C++ program that computes a dot-product of two vectors and is being compiled through a Makefile.\nAdditionally, we can select the type of elements for the vectors at compilation time.\nHere is the C++ program:\n```\ncat tutorials/advanced/makefiles/src/dotprod.cpp\n```\n```\n#include <cassert>\n#include <iostream>\n#include <random>\n#include <vector#ifndef ELEM_TYPE\n#define ELEM_TYPE double\n#endif\n\nusing elem_t = ELEM_TYPE;\n\ntemplate<typename T>\nT dotprod(const std::vector<T> &x, const std::vector<T> &y)\n{\n assert(x.size() == y.size());\n T sum = 0;\n for (std::size_t i = 0; i < x.size(); ++i) {\n sum += x[i] * y[i];\n }\n\n return sum;\n}\n\ntemplate<typename T>\nstruct type_name {\n static constexpr const char *value = nullptr;\n};\n\ntemplate<>\nstruct type_name<float> {\n static constexpr const char *value = \"float\";\n};\n\ntemplate<>\nstruct type_name<double> {\n static constexpr const char *value = \"double\";\n};\n\nint main(int argc, char *argv[])\n{\n if (argc < 2) {\n std::cerr << argv[0] << \": too few arguments\\n\";\n std::cerr << \"Usage: \" << argv[0] << \" DIM\\n\";\n return 1;\n }\n\n std::size_t N = std::atoi(argv[1]);\n if (N < 0) {\n std::cerr << argv[0]\n << \": array dimension must a positive integer: \" << argv[1]\n << \"\\n\";\n return 1;\n }\n\n std::vector<elem_t> x(N), y(N);\n std::random_device seed;\n std::mt19937 rand(seed());\n std::uniform_real_distribution<> dist(-1, 1);\n for (std::size_t i = 0; i < N; ++i) {\n x[i] = dist(rand);\n y[i] = dist(rand);\n }\n\n std::cout << \"Result (\" << type_name<elem_t>::value << \"): \"\n << dotprod(x, y) << \"\\n\";\n return 0;\n}\n```\nThe directory structure for this test is the following:\n```\ntutorials/makefiles/\n├── maketest.py\n└── src\n ├── Makefile\n └── dotprod.cpp\n```\nLet’s have a look at the test itself:\n```\ncat tutorials/advanced/makefiles/maketest.py\n```\n```\n@rfm.simple_test class MakefileTest(rfm.RegressionTest):\n elem_type = parameter(['float', 'double'])\n\n descr = 'Test demonstrating use of Makefiles'\n valid_systems = ['*']\n valid_prog_environs = ['clang', 'gnu']\n executable = './dotprod'\n executable_opts = ['100000']\n build_system = 'Make'\n\n @run_before('compile')\n def set_compiler_flags(self):\n self.build_system.cppflags = [f'-DELEM_TYPE={self.elem_type}']\n\n @sanity_function\n def validate_test(self):\n return sn.assert_found(rf'Result \\({self.elem_type}\\):', self.stdout)\n```\nFirst, if you’re using any build system other than `SingleSource`, you must set the `executable` attribute of the test, because ReFrame cannot know what is the actual executable to be run.\nWe then set the build system to [`Make`](index.html#reframe.core.buildsystems.Make) and set the preprocessor flags as we would do with the `SingleSource` build system.\n\nLet’s inspect the build script generated by ReFrame:\n```\n./bin/reframe -c tutorials/advanced/makefiles/maketest.py -r cat output/catalina/default/clang/MakefileTest_float/rfm_MakefileTest_build.sh\n```\n```\n#!/bin/bash\n\n_onerror()\n{\n exitcode=$?\n echo \"-reframe: command \\`$BASH_COMMAND' failed (exit code: $exitcode)\"\n exit $exitcode\n}\n\ntrap _onerror ERR\n\nmake -j 1 CC=\"cc\" CXX=\"CC\" FC=\"ftn\" NVCC=\"nvcc\" CPPFLAGS=\"-DELEM_TYPE=float\"\n```\nThe compiler variables (`CC`, `CXX` etc.) are set based on the corresponding values specified in the [configuration](config_reference.html#environment-configuration) of the current environment.\nWe can instruct the build system to ignore the default values from the environment by setting its `flags_from_environ` attribute to false:\n```\nself.build_system.flags_from_environ = False\n```\nIn this case, `make` will be invoked as follows:\n```\nmake -j 1 CPPFLAGS=\"-DELEM_TYPE=float\"\n```\nNotice that the `-j 1` option is always generated.\nWe can increase the build concurrency by setting the [`max_concurrency`](index.html#reframe.core.buildsystems.Make.max_concurrency) attribute.\nFinally, we may even use a custom Makefile by setting the [`makefile`](index.html#reframe.core.buildsystems.Make.makefile) attribute:\n```\nself.build_system.max_concurrency = 4 self.build_system.makefile = 'Makefile_custom'\n```\nAs a final note, as with the `SingleSource` build system, it wouldn’t have been necessary to specify one in this test, if we wouldn’t have to set the CPPFLAGS.\nReFrame could automatically figure out the correct build system if [`sourcepath`](index.html#reframe.core.pipeline.RegressionTest.sourcepath) refers to a directory.\nReFrame will inspect the directory and it will first try to determine whether this is a CMake or Autotools-based project.\n\nMore details on ReFrame’s build systems can be found [here](regression_test_api.html#build-systems).\n\n###### Retrieving the source code from a Git repository[¶](#retrieving-the-source-code-from-a-git-repository)\n\nIt might be the case that a regression test needs to clone its source code from a remote repository.\nThis can be achieved in two ways with ReFrame.\nOne way is to set the `sourcesdir` attribute to `None` and explicitly clone a repository using the [`prebuild_cmds`](index.html#reframe.core.pipeline.RegressionTest.prebuild_cmds):\n```\nself.sourcesdir = None self.prebuild_cmds = ['git clone https://github.com/me/myrepo .']\n```\nAlternatively, we can retrieve specifically a Git repository by assigning its URL directly to the `sourcesdir` attribute:\n```\nself.sourcesdir = 'https://github.com/me/myrepo'\n```\nReFrame will attempt to clone this repository inside the stage directory by executing `git clone <repo> .` and will then proceed with the build procedure as usual.\n\nNote\n\nReFrame recognizes only URLs in the `sourcesdir` attribute and requires passwordless access to the repository.\nThis means that the SCP-style repository specification will not be accepted.\nYou will have to specify it as URL using the `ssh://` protocol (see [Git documentation page](https://git-scm.com/docs/git-clone#_git_urls)).\n\n###### Adding a configuration step before compiling the code[¶](#adding-a-configuration-step-before-compiling-the-code)\n\nIt is often the case that a configuration step is needed before compiling a code with `make`.\nTo address this kind of projects, ReFrame aims to offer specific abstractions for “configure-make” style of build systems.\nIt supports [CMake-based](https://cmake.org/) projects through the [`CMake`](index.html#reframe.core.buildsystems.CMake) build system, as well as [Autotools-based](https://www.gnu.org/software/automake/) projects through the [`Autotools`](index.html#reframe.core.buildsystems.Autotools) build system.\n\nFor other build systems, you can achieve the same effect using the [`Make`](index.html#reframe.core.buildsystems.Make) build system and the [`prebuild_cmds`](index.html#reframe.core.pipeline.RegressionTest.prebuild_cmds) for performing the configuration step.\nThe following code snippet will configure a code with `./custom_configure` before invoking `make`:\n```\nself.prebuild_cmds = ['./custom_configure -with-mylib']\nself.build_system = 'Make'\nself.build_system.cppflags = ['-DHAVE_FOO']\nself.build_system.flags_from_environ = False\n```\nThe generated build script will then have the following lines:\n```\n./custom_configure -with-mylib make -j 1 CPPFLAGS='-DHAVE_FOO'\n```\n##### Writing a Run-Only Regression Test[¶](#writing-a-run-only-regression-test)\n\nThere are cases when it is desirable to perform regression testing for an already built executable.\nIn the following test we use simply the `echo` Bash shell command to print a random integer between specific lower and upper bounds.\nHere is the full regression test:\n```\ncat tutorials/advanced/runonly/echorand.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class EchoRandTest(rfm.RunOnlyRegressionTest):\n descr = 'A simple test that echoes a random number'\n valid_systems = ['*']\n valid_prog_environs = ['*']\n lower = variable(int, value=90)\n upper = variable(int, value=100)\n executable = 'echo'\n executable_opts = [\n 'Random: ',\n f'$((RANDOM%({upper}+1-{lower})+{lower}))'\n ]\n\n @sanity_function\n def assert_solution(self):\n return sn.assert_bounded(\n sn.extractsingle(\n r'Random: (?P<number>\\S+)', self.stdout, 'number', float\n ),\n self.lower, self.upper\n )\n```\nThere is nothing special for this test compared to those presented so far except that it derives from the [`RunOnlyRegressionTest`](index.html#reframe.core.pipeline.RunOnlyRegressionTest) class.\nNote that setting the [`executable`](index.html#reframe.core.pipeline.RegressionTest.executable) in this type of test is always required.\nRun-only regression tests may also have resources, as for instance a pre-compiled executable or some input data.\nThese resources may reside under the `src/` directory or under any directory specified in the [`sourcesdir`](index.html#reframe.core.pipeline.RegressionTest.sourcesdir) attribute.\nThese resources will be copied to the stage directory at the beginning of the run phase.\n\n##### Writing a Compile-Only Regression Test[¶](#writing-a-compile-only-regression-test)\n\nReFrame provides the option to write compile-only tests which consist only of a compilation phase without a specified executable.\nThis kind of tests must derive from the [`CompileOnlyRegressionTest`](index.html#reframe.core.pipeline.CompileOnlyRegressionTest) class provided by the framework.\nThe following test is a compile-only version of the `MakefileTest` presented [previously](#more-on-building-tests) which checks that no warnings are issued by the compiler:\n```\ncat tutorials/advanced/makefiles/maketest.py\n```\n```\n@rfm.simple_test class MakeOnlyTest(rfm.CompileOnlyRegressionTest):\n elem_type = parameter(['float', 'double'])\n descr = 'Test demonstrating use of Makefiles'\n valid_systems = ['*']\n valid_prog_environs = ['clang', 'gnu']\n build_system = 'Make'\n\n @run_before('compile')\n def set_compiler_flags(self):\n self.build_system.cppflags = [f'-DELEM_TYPE={self.elem_type}']\n\n @sanity_function\n def validate_compilation(self):\n return sn.assert_not_found(r'warning', self.stdout)\n```\nWhat is worth noting here is that the standard output and standard error of the test, which are accessible through the [`stdout`](index.html#reframe.core.pipeline.RegressionTest.stdout) and [`stderr`](index.html#reframe.core.pipeline.RegressionTest.stderr) attributes, correspond now to the standard output and error of the compilation command.\nTherefore sanity checking can be done in exactly the same way as with a normal test.\n\n##### Grouping parameter packs[¶](#grouping-parameter-packs)\n\nNew in version 3.4.2.\n\nIn the dot product example shown above, we had two independent tests that defined the same `elem_type` parameter.\nAnd the two tests cannot have a parent-child relationship, since one of them is a run-only test and the other is a compile-only one.\nReFrame offers the [`RegressionMixin`](index.html#reframe.core.pipeline.RegressionMixin) class that allows you to group parameters and other [builtins](regression_test_api.html#builtins) that are meant to be reused over otherwise unrelated tests.\nIn the example below, we create an `ElemTypeParam` mixin that holds the definition of the `elem_type` parameter which is inherited by both the concrete test classes:\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\nclass ElemTypeParam(rfm.RegressionMixin):\n elem_type = parameter(['float', 'double'])\n\n@rfm.simple_test class MakefileTestAlt(rfm.RegressionTest, ElemTypeParam):\n descr = 'Test demonstrating use of Makefiles'\n valid_systems = ['*']\n valid_prog_environs = ['clang', 'gnu']\n executable = './dotprod'\n executable_opts = ['100000']\n build_system = 'Make'\n\n @run_before('compile')\n def set_compiler_flags(self):\n self.build_system.cppflags = [f'-DELEM_TYPE={self.elem_type}']\n\n @sanity_function\n def validate_test(self):\n return sn.assert_found(\n rf'Result \\({self.elem_type}\\):', self.stdout\n )\n\n@rfm.simple_test class MakeOnlyTestAlt(rfm.CompileOnlyRegressionTest, ElemTypeParam):\n descr = 'Test demonstrating use of Makefiles'\n valid_systems = ['*']\n valid_prog_environs = ['clang', 'gnu']\n build_system = 'Make'\n\n @run_before('compile')\n def set_compiler_flags(self):\n self.build_system.cppflags = [f'-DELEM_TYPE={self.elem_type}']\n\n @sanity_function\n def validate_build(self):\n return sn.assert_not_found(r'warning', self.stdout)\n```\nNotice how the parameters are expanded in each of the individual tests:\n```\n./bin/reframe -c tutorials/advanced/makefiles/maketest_mixin.py -l\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe -c tutorials/advanced/makefiles/maketest_mixin.py -l'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>', '/home/user/Repositories/reframe/tutorials/config/tresa.py'\n check search path: '/home/user/Repositories/reframe/tutorials/advanced/makefiles/maketest_mixin.py'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/<KEY>'\n\n[List of matched checks]\n- MakeOnlyTestAlt %elem_type=double /8b62380e\n- MakeOnlyTestAlt %elem_type=float /da39ec20\n- MakefileTestAlt %elem_type=double /89aac4a2\n- MakefileTestAlt %elem_type=float /a998ce67 Found 4 check(s)\n\nLog file(s) saved in '/<KEY>'\n```\n##### Applying a Sanity Function Iteratively[¶](#applying-a-sanity-function-iteratively)\n\nIt is often the case that a common sanity function has to be applied many times.\nThe following script prints 100 random integers between the limits given by the environment variables `LOWER` and `UPPER`.\n```\ncat tutorials/advanced/random/src/random_numbers.sh\n```\n```\nif [ -z $LOWER ]; then\n export LOWER=90 fi\n\nif [ -z $UPPER ]; then\n export UPPER=100 fi\n\nfor i in {1..100}; do\n echo Random: $((RANDOM%($UPPER+1-$LOWER)+$LOWER))\ndone\n```\nIn the corresponding regression test we want to check that all the random numbers generated lie between the two limits, which means that a common sanity check has to be applied to all the printed random numbers.\nHere is the corresponding regression test:\n```\ncat tutorials/advanced/random/randint.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class DeferredIterationTest(rfm.RunOnlyRegressionTest):\n descr = 'Apply a sanity function iteratively'\n valid_systems = ['*']\n valid_prog_environs = ['*']\n executable = './random_numbers.sh'\n\n @sanity_function\n def validate_test(self):\n numbers = sn.extractall(\n r'Random: (?P<number>\\S+)', self.stdout, 'number', float\n )\n return sn.all([\n sn.assert_eq(sn.count(numbers), 100),\n sn.all(sn.map(lambda x: sn.assert_bounded(x, 90, 100), numbers))\n ])\n```\nFirst, we extract all the generated random numbers from the output.\nWhat we want to do is to apply iteratively the [`assert_bounded()`](index.html#reframe.utility.sanity.assert_bounded) sanity function for each number.\nThe problem here is that we cannot simply iterate over the `numbers` list, because that would trigger prematurely the evaluation of the [`extractall()`](index.html#reframe.utility.sanity.extractall).\nWe want to defer also the iteration.\nThis can be achieved by using the [`map()`](index.html#reframe.utility.sanity.map) ReFrame sanity function, which is a replacement of Python’s built-in [`map()`](https://docs.python.org/3/library/functions.html#map) function and does exactly what we want: it applies a function on all the elements of an iterable and returns another iterable with the transformed elements.\nPassing the result of the [`map()`](https://docs.python.org/3/library/functions.html#map) function to the [`all()`](index.html#reframe.utility.sanity.all) sanity function ensures that all the elements lie between the desired bounds.\n\nThere is still a small complication that needs to be addressed.\nAs a direct replacement of the built-in [`all()`](https://docs.python.org/3/library/functions.html#all) function, ReFrame’s [`all()`](index.html#reframe.utility.sanity.all) sanity function returns `True` for empty iterables, which is not what we want.\nSo we must make sure that all 100 numbers are generated.\nThis is achieved by the `sn.assert_eq(sn.count(numbers), 100)` statement, which uses the [`count()`](index.html#reframe.utility.sanity.count) sanity function for counting the generated numbers.\nFinally, we need to combine these two conditions to a single deferred expression that will be returned by the test’s `@sanity_function`.\nWe accomplish this by using the [`all()`](index.html#reframe.utility.sanity.all) sanity function.\n\nFor more information about how exactly sanity functions work and how their execution is deferred, please refer to [Understanding the Mechanism of Deferrable Functions](index.html#document-deferrables).\n\nNote\n\nNew in version 2.13: ReFrame offers also the [`allx()`](index.html#reframe.utility.sanity.allx) sanity function which, conversely to the builtin [`all()`](https://docs.python.org/3/library/functions.html#all) function, will return `False` if its iterable argument is empty.\n\n##### Customizing the Test Job Script[¶](#customizing-the-test-job-script)\n\nIt is often the case that we need to run some commands before or after the parallel launch of our executable.\nThis can be easily achieved by using the [`prerun_cmds`](index.html#reframe.core.pipeline.RegressionTest.prerun_cmds) and [`postrun_cmds`](index.html#reframe.core.pipeline.RegressionTest.postrun_cmds) attributes of a ReFrame test.\n\nThe following example is a slightly modified version of the random numbers test presented [above](#applying-a-sanity-function-iteratively).\nThe lower and upper limits for the random numbers are now set inside a helper shell script in `limits.sh` located in the test’s resources, which we need to source before running our tests.\nAdditionally, we want also to print `FINISHED` after our executable has finished.\nHere is the modified test file:\n```\ncat tutorials/advanced/random/prepostrun.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class PrepostRunTest(rfm.RunOnlyRegressionTest):\n descr = 'Pre- and post-run demo test'\n valid_systems = ['*']\n valid_prog_environs = ['*']\n prerun_cmds = ['source limits.sh']\n postrun_cmds = ['echo FINISHED']\n executable = './random_numbers.sh'\n\n @sanity_function\n def validate_test(self):\n numbers = sn.extractall(\n r'Random: (?P<number>\\S+)', self.stdout, 'number', float\n )\n return sn.all([\n sn.assert_eq(sn.count(numbers), 100),\n sn.all(sn.map(lambda x: sn.assert_bounded(x, 90, 100), numbers)),\n sn.assert_found(r'FINISHED', self.stdout)\n ])\n```\nThe `prerun_cmds` and `postrun_cmds` are lists of commands to be emitted in the generated job script before and after the parallel launch of the executable.\nObviously, the working directory for these commands is that of the job script itself, which is the stage directory of the test.\nThe generated job script for this test looks like the following:\n```\n./bin/reframe -c tutorials/advanced/random/prepostrun.py -r cat output/catalina/default/gnu/PrepostRunTest/rfm_PrepostRunTest_job.sh\n```\n```\n#!/bin/bash source limits.sh\n ./random_numbers.sh echo FINISHED\n```\nGenerally, ReFrame generates the job shell scripts using the following pattern:\n```\n#!/bin/bash -l\n{job_scheduler_preamble}\n{prepare_cmds}\n{env_load_cmds}\n{prerun_cmds}\n{parallel_launcher} {executable} {executable_opts}\n{postrun_cmds}\n```\nThe `job_scheduler_preamble` contains the backend job scheduler directives that control the job allocation.\nThe `prepare_cmds` are commands that can be emitted before the test environment commands.\nThese can be specified with the [`prepare_cmds`](index.html#config.systems.partitions.prepare_cmds) partition configuration option.\nThe `env_load_cmds` are the necessary commands for setting up the environment of the test.\nThese include any modules or environment variables set at the [system partition level](config_reference.html#system-partition-configuration) or any [modules](regression_test_api.html#reframe.core.pipeline.RegressionTest.modules) or [environment variables](regression_test_api.html#reframe.core.pipeline.RegressionTest.variables) set at the test level.\nThen the commands specified in [`prerun_cmds`](index.html#reframe.core.pipeline.RegressionTest.prerun_cmds) follow, while those specified in the [`postrun_cmds`](index.html#reframe.core.pipeline.RegressionTest.postrun_cmds) come after the launch of the parallel job.\nThe parallel launch itself consists of three parts:\n\n1. The parallel launcher program (e.g., `srun`, `mpirun` etc.) with its options,\n2. the regression test executable as specified in the [`executable`](index.html#reframe.core.pipeline.RegressionTest.executable) attribute and 3. the options to be passed to the executable as specified in the [`executable_opts`](index.html#reframe.core.pipeline.RegressionTest.executable_opts) attribute.\n\n###### Adding job scheduler options per test[¶](#adding-job-scheduler-options-per-test)\n\nSometimes a test needs to pass additional job scheduler options to the automatically generated job script.\nThis is fairly easy to achieve with ReFrame.\nIn the following test we want to test whether the `--mem` option of Slurm works as expected.\nWe compiled and ran a program that consumes all the available memory of the node, but we want to restrict the available memory with the `--mem` option.\nHere is the test:\n```\ncat tutorials/advanced/jobopts/eatmemory.py\n```\n```\n@rfm.simple_test class MemoryLimitTest(rfm.RegressionTest):\n valid_systems = ['daint:gpu', 'daint:mc']\n valid_prog_environs = ['gnu']\n sourcepath = 'eatmemory.c'\n executable_opts = ['2000M']\n\n @run_before('run')\n def set_memory_limit(self):\n self.job.options = ['--mem=1000']\n\n @sanity_function\n def validate_test(self):\n return sn.assert_found(\n r'(exceeded memory limit)|(Out Of Memory)', self.stderr\n )\n```\nEach ReFrame test has an associated [run job descriptor](regression_test_api.html#reframe.core.pipeline.RegressionTest.job) which represents the scheduler job that will be used to run this test.\nThis object has an `options` attribute, which can be used to pass arbitrary options to the scheduler.\nThe job descriptor is initialized by the framework during the [setup](pipeline.html#the-regression-test-pipeline) pipeline phase.\nFor this reason, we cannot directly set the job options inside the test constructor and we have to use a pipeline hook that runs before running (i.e., submitting the test).\n\nLet’s run the test and inspect the generated job script:\n```\n./bin/reframe -c tutorials/advanced/jobopts/eatmemory.py -n MemoryLimitTest -r cat output/daint/gpu/gnu/MemoryLimitTest/rfm_MemoryLimitTest_job.sh\n```\n```\n#!/bin/bash\n#SBATCH --job-name=\"rfm_MemoryLimitTest_job\"\n#SBATCH --ntasks=1\n#SBATCH --output=rfm_MemoryLimitTest_job.out\n#SBATCH --error=rfm_MemoryLimitTest_job.err\n#SBATCH --time=0:10:0\n#SBATCH -A csstaff\n#SBATCH --constraint=gpu\n#SBATCH --mem=1000 module unload PrgEnv-cray module load PrgEnv-gnu srun ./MemoryLimitTest 2000M\n```\nThe job options specified inside a ReFrame test are always the last to be emitted in the job script preamble and do not affect the options that are passed implicitly through other test attributes or configuration options.\n\nThere is a small problem with this test though.\nWhat if we change the job scheduler in that partition or what if we want to port the test to a different system that does not use Slurm and another option is needed to achieve the same result.\nThe obvious answer is to adapt the test, but is there a more portable way?\nThe answer is yes and this can be achieved through so-called *extra resources*.\nReFrame gives you the possibility to associate scheduler options to a “resource” managed by the partition scheduler.\nYou can then use those resources transparently from within your test.\n\nTo achieve this in our case, we first need to define a `memory` resource in the configuration of both of our `daint` partitions:\n```\n {\n 'name': 'gpu',\n 'descr': 'Hybrid nodes',\n 'scheduler': 'slurm',\n 'launcher': 'srun',\n 'access': ['-C gpu', '-A csstaff'],\n 'environs': ['gnu', 'intel', 'nvidia', 'cray'],\n 'max_jobs': 100,\n 'resources': [\n {\n 'name': 'memory',\n 'options': ['--mem={size}']\n }\n ]\n },\n```\nNotice that we do not define the resource for all the partitions, but only for those that it makes sense.\nEach resource has a name and a set of scheduler options that will be passed to the scheduler when this resource will be requested by the test.\nThe options specification can contain placeholders, whose value will also be set from the test.\nLet’s see how we can rewrite the `MemoryLimitTest` using the `memory` resource instead of passing the `--mem` scheduler option explicitly.\n```\ncat tutorials/advanced/jobopts/eatmemory.py\n```\n```\n@rfm.simple_test class MemoryLimitWithResourcesTest(rfm.RegressionTest):\n valid_systems = ['daint:gpu', 'daint:mc']\n valid_prog_environs = ['gnu']\n sourcepath = 'eatmemory.c'\n executable_opts = ['2000M']\n extra_resources = {\n 'memory': {'size': '1000'}\n }\n\n @sanity_function\n def validate_test(self):\n return sn.assert_found(\n r'(exceeded memory limit)|(Out Of Memory)', self.stderr\n )\n```\nThe extra resources that the test needs to obtain through its scheduler are specified in the [`extra_resources`](index.html#reframe.core.pipeline.RegressionTest.extra_resources) attribute, which is a dictionary with the resource names as its keys and another dictionary assigning values to the resource placeholders as its values.\nAs you can see, this syntax is completely scheduler-agnostic.\nIf the requested resource is not defined for the current partition, it will be simply ignored.\n\nYou can now run and verify that the generated job script contains the `--mem` option:\n```\n./bin/reframe -c tutorials/advanced/jobopts/eatmemory.py -n MemoryLimitWithResourcesTest -r cat output/daint/gpu/gnu/MemoryLimitWithResourcesTest/rfm_MemoryLimitWithResourcesTest_job.sh\n```\n###### Modifying the parallel launcher command[¶](#modifying-the-parallel-launcher-command)\n\nAnother relatively common need is to modify the parallel launcher command.\nReFrame gives the ability to do that and we will see some examples in this section.\n\nThe most common case is to pass arguments to the launcher command that you cannot normally pass as job options.\nThe `--cpu-bind` of `srun` is such an example.\nInside a ReFrame test, you can access the parallel launcher through the [`launcher`](index.html#reframe.core.schedulers.Job.launcher) of the job descriptor.\nThis object handles all the details of how the parallel launch command will be emitted.\nIn the following test we run a CPU affinity test using [this](https://github.com/vkarak/affinity) utility and we will pin the threads using the `--cpu-bind` option:\n```\ncat tutorials/advanced/affinity/affinity.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class AffinityTest(rfm.RegressionTest):\n valid_systems = ['daint:gpu', 'daint:mc']\n valid_prog_environs = ['*']\n sourcesdir = 'https://github.com/vkarak/affinity.git'\n build_system = 'Make'\n executable = './affinity'\n\n @run_before('compile')\n def set_build_system_options(self):\n self.build_system.options = ['OPENMP=1']\n\n @run_before('run')\n def set_cpu_binding(self):\n self.job.launcher.options = ['--cpu-bind=cores']\n\n @sanity_function\n def validate_test(self):\n return sn.assert_found(r'CPU affinity', self.stdout)\n```\nThe approach is identical to the approach we took in the `MemoryLimitTest` test [above](#adding-job-scheduler-options-per-test), except that we now set the launcher options.\n\nNote\n\nThe sanity checking in a real affinity checking test would be much more complex than this.\n\nAnother scenario that might often arise when testing parallel debuggers is the need to wrap the launcher command with the debugger command.\nFor example, in order to debug a parallel program with [ARM DDT](https://www.arm.com/products/development-tools/server-and-hpc/forge/ddt), you would need to invoke the program like this: `ddt [OPTIONS] srun [OPTIONS]`.\nReFrame allows you to wrap the launcher command without the test needing to know which is the actual parallel launcher command for the current partition.\nThis can be achieved with the following pipeline hook:\n```\nimport reframe as rfm from reframe.core.launchers import LauncherWrapper\n\nclass DebuggerTest(rfm.RunOnlyRegressionTest):\n ...\n\n @run_before('run')\n def set_launcher(self):\n self.job.launcher = LauncherWrapper(self.job.launcher, 'ddt',\n ['--offline'])\n```\nThe [`LauncherWrapper`](index.html#reframe.core.launchers.LauncherWrapper) is a pseudo-launcher that wraps another one and allows you to prepend anything to it.\nIn this case the resulting parallel launch command, if the current partition uses native Slurm, will be `ddt --offline srun [OPTIONS]`.\n\n###### Replacing the parallel launcher[¶](#replacing-the-parallel-launcher)\n\nSometimes you might need to replace completely the partition’s launcher command, because the software you are testing might use its own parallel launcher.\nExamples are [ipyparallel](https://ipyparallel.readthedocs.io/en/latest/), the [GREASY](https://github.com/BSC-Support-Team/GREASY) high-throughput scheduler, as well as some visualization software.\nThe trick here is to replace the parallel launcher with the local one, which practically does not emit any launch command, and by now you should almost be able to do it all by yourself:\n```\nimport reframe as rfm from reframe.core.backends import getlauncher\n\nclass CustomLauncherTest(rfm.RunOnlyRegressionTest):\n ...\n executable = 'custom_scheduler'\n executable_opts = [...]\n\n @run_before('run')\n def replace_launcher(self):\n self.job.launcher = getlauncher('local')()\n```\nThe [`getlauncher()`](index.html#reframe.core.backends.getlauncher) function takes the [registered](config_reference.html#systems-.partitions-.launcher) name of a launcher and returns the class that implements it.\nYou then instantiate the launcher and assign to the [`launcher`](index.html#reframe.core.schedulers.Job.launcher) attribute of the job descriptor.\n\n###### Adding more parallel launch commands[¶](#adding-more-parallel-launch-commands)\n\nReFrame uses a parallel launcher by default for anything defined explicitly or implicitly in the [`executable`](index.html#reframe.core.pipeline.RegressionTest.executable) test attribute.\nBut what if we want to generate multiple parallel launch commands?\nOne straightforward solution is to hardcode the parallel launch command inside the [`prerun_cmds`](index.html#reframe.core.pipeline.RegressionTest.prerun_cmds) or [`postrun_cmds`](index.html#reframe.core.pipeline.RegressionTest.postrun_cmds), but this is not so portable.\nThe best way is to ask ReFrame to emit the parallel launch command for you.\nThe following is a simple test for demonstration purposes that runs the `hostname` command several times using a parallel launcher.\nIt resembles a scaling test, except that all happens inside a single ReFrame test, instead of launching multiple instances of a parameterized test.\n```\ncat tutorials/advanced/multilaunch/multilaunch.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class MultiLaunchTest(rfm.RunOnlyRegressionTest):\n valid_systems = ['daint:gpu', 'daint:mc']\n valid_prog_environs = ['builtin']\n executable = 'hostname'\n num_tasks = 4\n num_tasks_per_node = 1\n\n @run_before('run')\n def pre_launch(self):\n cmd = self.job.launcher.run_command(self.job)\n self.prerun_cmds = [\n f'{cmd} -n {n} {self.executable}'\n for n in range(1, self.num_tasks)\n ]\n\n @sanity_function\n def validate_test(self):\n return sn.assert_eq(\n sn.count(sn.extractall(r'^nid\\d+', self.stdout)), 10\n )\n```\nThe additional parallel launch commands are inserted in either the `prerun_cmds` or `postrun_cmds` lists.\nTo retrieve the actual parallel launch command for the current partition that the test is running on, you can use the `run_command()` method of the launcher object.\nLet’s see how the generated job script looks like:\n```\n./bin/reframe -c tutorials/advanced/multilaunch/multilaunch.py -r cat output/daint/gpu/builtin/MultiLaunchTest/rfm_MultiLaunchTest_job.sh\n```\n```\n#!/bin/bash\n#SBATCH --job-name=\"rfm_MultiLaunchTest_job\"\n#SBATCH --ntasks=4\n#SBATCH --ntasks-per-node=1\n#SBATCH --output=rfm_MultiLaunchTest_job.out\n#SBATCH --error=rfm_MultiLaunchTest_job.err\n#SBATCH --time=0:10:0\n#SBATCH -A csstaff\n#SBATCH --constraint=gpu srun -n 1 hostname srun -n 2 hostname srun -n 3 hostname srun hostname\n```\nThe first three `srun` commands are emitted through the `prerun_cmds` whereas the last one comes from the test’s `executable` attribute.\n\n###### Adding a custom launcher to a partition[¶](#adding-a-custom-launcher-to-a-partition)\n\nNew in version 4.0.0.\n\nAn alternative to the approaches above would be to define your own custom parallel launcher and register it with the framework.\nYou could then use it as the launcher of a system partition in the configuration and use it in multiple tests.\n\nEach [launcher](regression_test_api.html#reframe.core.launchers.JobLauncher) needs to implement the [`command()`](index.html#reframe.core.launchers.JobLauncher.command) method and can optionally change the default [`run_command()`](index.html#reframe.core.launchers.JobLauncher.run_command) method.\n\nAs an example of how easy it is to define a new parallel launcher backend, here is the actual implementation of the `mpirun` launcher:\n```\nfrom reframe.core.backends import register_launcher from reframe.core.launchers import JobLauncher\n\n@register_launcher('mpirun')\nclass MpirunLauncher(JobLauncher):\n def command(self, job):\n return ['mpirun', '-np', str(job.num_tasks)]\n```\nThe [`command()`](index.html#reframe.core.launchers.JobLauncher.command) returns a list of command tokens that will be combined with any user-supplied [options](regression_test_api.html#reframe.core.launchers.JobLauncher.options) by the [`run_command()`](index.html#reframe.core.launchers.JobLauncher.run_command) method to generate the actual launcher command line.\nNotice you can use the `job` argument to get job-specific information that will allow you to construct the correct launcher invocation.\n\nIf you use a Python-based configuration file, you can define your custom launcher directly inside your config as follows:\n```\nfrom reframe.core.backends import register_launcher from reframe.core.launchers import JobLaucher\n\n@register_launcher('slrun')\nclass MySmartLauncher(JobLauncher):\n def command(self, job):\n return ['slrun', ...]\n\nsite_configuration = {\n 'systems': [\n {\n 'name': 'my_system',\n 'partitions': [\n {\n 'name': 'my_partition',\n 'launcher': 'slrun'\n ...\n }\n ],\n ...\n },\n ...\n ],\n ...\n}\n```\n##### Flexible Regression Tests[¶](#flexible-regression-tests)\n\nNew in version 2.15.\n\nReFrame can automatically set the number of tasks of a particular test, if its [`num_tasks`](index.html#reframe.core.pipeline.RegressionTest.num_tasks) attribute is set to a negative value or zero.\nIn ReFrame’s terminology, such tests are called *flexible*.\nNegative values indicate the minimum number of tasks that are acceptable for this test (a value of `-4` indicates that at least `4` tasks are required).\nA zero value indicates the default minimum number of tasks which is equal to [`num_tasks_per_node`](index.html#reframe.core.pipeline.RegressionTest.num_tasks_per_node).\n\nBy default, ReFrame will spawn such a test on all the idle nodes of the current system partition, but this behavior can be adjusted with the [`--flex-alloc-nodes`](manpage.html#cmdoption-flex-alloc-nodes) command-line option.\nFlexible tests are very useful for diagnostics tests, e.g., tests for checking the health of a whole set nodes.\nIn this example, we demonstrate this feature through a simple test that runs `hostname`.\nThe test will verify that all the nodes print the expected host name:\n```\ncat tutorials/advanced/flexnodes/flextest.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class HostnameCheck(rfm.RunOnlyRegressionTest):\n valid_systems = ['daint:gpu', 'daint:mc']\n valid_prog_environs = ['cray']\n executable = 'hostname'\n num_tasks = 0\n num_tasks_per_node = 1\n\n @sanity_function\n def validate_test(self):\n return sn.assert_eq(\n self.num_tasks,\n sn.count(sn.findall(r'^nid\\d+$', self.stdout))\n )\n```\nThe first thing to notice in this test is that [`num_tasks`](index.html#reframe.core.pipeline.RegressionTest.num_tasks) is set to zero as default, which is a requirement for flexible tests.\nHowever, with flexible tests, this value is updated right after the job completes to the actual number of tasks that were used.\nConsequently, this allows the sanity function of the test to assert that the number host names printed matches [`num_tasks`](index.html#reframe.core.pipeline.RegressionTest.num_tasks).\n\nTip\n\nIf you want to run multiple flexible tests at once, it’s better to run them using the serial execution policy, because the first test might take all the available nodes and will cause the rest to fail immediately, since there will be no available nodes for them.\n\n##### Testing containerized applications[¶](#testing-containerized-applications)\n\nNew in version 2.20.\n\nReFrame can be used also to test applications that run inside a container.\nFirst, we need to enable the container platform support in ReFrame’s configuration and, specifically, at the partition configuration level:\n```\n {\n 'name': 'gpu',\n 'descr': 'Hybrid nodes',\n 'scheduler': 'slurm',\n 'launcher': 'srun',\n 'access': ['-C gpu', '-A csstaff'],\n 'environs': ['gnu', 'intel', 'nvidia', 'cray'],\n 'max_jobs': 100,\n 'container_platforms': [\n {\n 'type': 'Sarus',\n 'modules': ['sarus']\n },\n {\n 'type': 'Singularity',\n 'modules': ['singularity']\n }\n ]\n },\n```\nFor each partition, users can define a list of all supported container platforms using the [`container_platforms`](index.html#config.systems.partitions.container_platforms) configuration parameter.\nIn this case, we define the [Sarus](https://github.com/eth-cscs/sarus) platform for which we set the [`modules`](index.html#config.systems.partitions.container_platforms.modules) parameter in order to instruct ReFrame to load the `sarus` module, whenever it needs to run with this container platform.\nSimilarly, we add an entry for the [Singularity](https://sylabs.io) platform.\nOptionally, users are allowed to set the `default` attribute to [`True`](https://docs.python.org/3/library/constants.html#True) in order to mark a specific container platform as the default of that partition (see below on how this information is being used).\nIf no default container platform is specified explicitly, then always the first in the list will be considered as successful.\n\nThe following parameterized test, will create two tests, one for each of the supported container platforms:\n```\ncat tutorials/advanced/containers/container_test.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class ContainerTest(rfm.RunOnlyRegressionTest):\n platform = parameter(['Sarus', 'Singularity'])\n valid_systems = ['daint:gpu']\n valid_prog_environs = ['builtin']\n\n @run_before('run')\n def setup_container_platf(self):\n self.descr = f'Run commands inside a container using {self.platform}'\n image_prefix = 'docker://' if self.platform == 'Singularity' else ''\n self.container_platform = self.platform\n self.container_platform.image = f'{image_prefix}ubuntu:18.04'\n self.container_platform.command = (\n \"bash -c 'cat /etc/os-release | tee /rfm_workdir/release.txt'\"\n )\n\n @sanity_function\n def assert_release(self):\n os_release_pattern = r'18.04.\\d+ LTS \\(Bionic Beaver\\)'\n return sn.assert_found(os_release_pattern, 'release.txt')\n```\nA container-based test can be written as [`RunOnlyRegressionTest`](index.html#reframe.core.pipeline.RunOnlyRegressionTest) that sets the [`container_platform`](index.html#reframe.core.pipeline.RegressionTest.container_platform) attribute.\nThis attribute accepts a string that corresponds to the name of the container platform that will be used to run the container for this test.\nIt is not necessary to specify this attribute, in which case, the default container platform of the current partition will be used.\nYou can still differentiate your test based on the actual container platform that is being used by checking the `self.container_platform.name` variable.\n\nAs soon as the container platform to be used is determined, you need to specify the container image to use by setting the [`image`](index.html#reframe.core.containers.ContainerPlatform.image).\nIf the image is not specified, then the container logic is skipped and the test executes as if the [`container_platform`](index.html#reframe.core.pipeline.RegressionTest.container_platform) was never set.\n\nIn the `Singularity` test variant, we add the `docker://` prefix to the image name, in order to instruct `Singularity` to pull the image from [DockerHub](https://hub.docker.com/).\nThe default command that the container runs can be overwritten by setting the [`command`](index.html#reframe.core.containers.ContainerPlatform.command) attribute of the container platform.\n\nThe [`image`](index.html#reframe.core.containers.ContainerPlatform.image) is the only mandatory attribute for container-based checks.\nIt is important to note that the [`executable`](index.html#reframe.core.pipeline.RegressionTest.executable) and [`executable_opts`](index.html#reframe.core.pipeline.RegressionTest.executable_opts) attributes of the actual test are ignored if the containerized code path is taken, i.e., when [`image`](index.html#reframe.core.containers.ContainerPlatform.image) is not [`None`](https://docs.python.org/3/library/constants.html#None).\n\nReFrame will run the container according to the given platform as follows:\n```\n# Sarus sarus run --mount=type=bind,source=\"/path/to/test/stagedir\",destination=\"/rfm_workdir\" ubuntu:18.04 bash -c 'cat /etc/os-release | tee /rfm_workdir/release.txt'\n\n# Singularity singularity exec -B\"/path/to/test/stagedir:/rfm_workdir\" docker://ubuntu:18.04 bash -c 'cat /etc/os-release | tee /rfm_workdir/release.txt'\n```\nIn the `Sarus` case, ReFrame will prepend the following command in order to pull the container image before running the container:\n```\nsarus pull ubuntu:18.04\n```\nThis is the default behavior of ReFrame, which can be changed if pulling the image is not desired by setting the [`pull_image`](index.html#reframe.core.containers.ContainerPlatform.pull_image) attribute to `False`.\nBy default ReFrame will mount the stage directory of the test under `/rfm_workdir` inside the container.\nOnce the commands are executed, the container is stopped and ReFrame goes on with the sanity and performance checks.\nBesides the stage directory, additional mount points can be specified through the `mount_points` attribute:\n```\nself.container_platform.mount_points = [('/path/to/host/dir1', '/path/to/container/mount_point1'),\n ('/path/to/host/dir2', '/path/to/container/mount_point2')]\n```\nThe container filesystem is ephemeral, therefore, ReFrame mounts the stage directory under `/rfm_workdir` inside the container where the user can copy artifacts as needed.\nThese artifacts will therefore be available inside the stage directory after the container execution finishes.\nThis is very useful if the artifacts are needed for the sanity or performance checks.\nIf the copy is not performed by the default container command, the user can override this command by settings the [`command`](index.html#reframe.core.containers.ContainerPlatform.command) attribute such as to include the appropriate copy commands.\nIn the current test, the output of the `cat /etc/os-release` is available both in the standard output as well as in the `release.txt` file, since we have used the command:\n```\nbash -c 'cat /etc/os-release | tee /rfm_workdir/release.txt'\n```\nand `/rfm_workdir` corresponds to the stage directory on the host system.\nTherefore, the `release.txt` file can now be used in the subsequent sanity checks:\n```\n @sanity_function\n def assert_release(self):\n os_release_pattern = r'18.04.\\d+ LTS \\(Bionic Beaver\\)'\n return sn.assert_found(os_release_pattern, 'release.txt')\n```\nFor a complete list of the available attributes of a specific container platform, please have a look at the [Container Platforms](index.html#container-platforms) section of the [Test API Reference](index.html#document-regression_test_api) guide.\nOn how to configure ReFrame for running containerized tests, please have a look at the [Container Platform Configuration](index.html#container-platform-configuration) section of the [Configuration Reference](index.html#document-config_reference).\n\nChanged in version 3.12.0: There is no need any more to explicitly set the `container_platform` in the test.\nThis is automatically initialized from the default platform of the current partition.\n\n###### Combining containerized and native application tests[¶](#combining-containerized-and-native-application-tests)\n\nNew in version 3.12.0.\n\nIt is very easy in ReFrame to have a single run-only test to either test the native or the containerized version of an application.\nThis is possible, since the framework will only take the “containerized” code path only if the [`image`](index.html#reframe.core.containers.ContainerPlatform.image) attribute of the [`container_platform`](index.html#reframe.core.pipeline.RegressionTest.container_platform) is not [`None`](https://docs.python.org/3/library/constants.html#None).\nOtherwise, the *bare metal* version of the tested application will be run.\nThe following test uses exactly this trick to test a series of GROMACS images as well as the native one provided on the Piz Daint supercomputer.\nIt also extends the GROMACS benchmark tests that are provided with ReFrame’s test library (see [ReFrame Test Library (experimental)](index.html#document-hpctestlib)).\nFor simplicity, we are assuming a single system here (the hybrid partition of Piz Daint) and we set fixed values for the `num_cpus_per_task` as well as the `-ntomp` option of GROMACS (NB: in a real-world test we would use the auto-detected processor topology information to set these values; see [Auto-detecting processor information](index.html#proc-autodetection) for more information).\nWe also redefine and restrict the benchmark’s parameters `benchmark_info` and `nb_impl` to the values that are of interest for the demonstration of this test.\nFinally, we also reset the executable to use `gmx` instead of the `gmx_mpi` that is used from the library test.\n```\nimport reframe as rfm from hpctestlib.sciapps.gromacs.benchmarks import gromacs_check\n\ndef _hecbiosim_bench(params):\n for p in params:\n if p[0] == 'HECBioSim/hEGFRDimerSmallerPL':\n return [p]\n\n@rfm.simple_test class gromacs_containerized_test(gromacs_check):\n # Restrict library test parameters to only those relevant for this example\n benchmark_info = parameter(inherit_params=True,\n filter_params=_hecbiosim_bench,\n fmt=lambda x: x[0])\n nb_impl = parameter(['gpu'])\n\n # New parameter for testing the various images\n gromacs_image = parameter([\n None,\n 'nvcr.io/hpc/gromacs:2020',\n 'nvcr.io/hpc/gromacs:2020.2',\n 'nvcr.io/hpc/gromacs:2021',\n 'nvcr.io/hpc/gromacs:2021.3',\n 'nvcr.io/hpc/gromacs:2022.1'\n ])\n valid_systems = ['daint:gpu']\n valid_prog_environs = ['gnu']\n use_multithreading = False\n executable = 'gmx mdrun'\n executable_opts += ['-dlb yes', '-ntomp 12', '-npme -1', '-v']\n num_tasks = 1\n num_tasks_per_node = 1\n num_cpus_per_task = 12\n\n @run_after('init')\n def setup_container_run(self):\n exec_cmd = ' '.join([self.executable, *self.executable_opts])\n self.container_platform.image = self.gromacs_image\n self.container_platform.command = exec_cmd\n if self.gromacs_image is None:\n self.modules = ['daint-gpu', 'GROMACS']\n```\nAll this test does in addition to the library test it inherits from is to set the [`image`](index.html#reframe.core.containers.ContainerPlatform.image) and the [`command`](index.html#reframe.core.containers.ContainerPlatform.command) attributes of the [`container_platform`](index.html#reframe.core.pipeline.RegressionTest.container_platform).\nThe former is set from the `gromacs_image` test parameter whereas the latter from the test’s [`executable`](index.html#reframe.core.pipeline.RegressionTest.executable) and [`executable_opts`](index.html#reframe.core.pipeline.RegressionTest.executable_opts) attributes.\nRemember that these attributes are ignored if the framework takes the path of launching a container.\nFinally, if the image is [`None`](https://docs.python.org/3/library/constants.html#None) we handle the case of the native run, in which case we load the modules required to run GROMACS natively on the target system.\n\nIn the following, we run the GPU version of a single benchmark with a series of images from NVIDIA and natively:\n```\n$ ./bin/reframe -C tutorials/config/daint.py -c tutorials/advanced/containers/gromacs_test.py -r\n```\n```\n[===] Running 6 check(s)\n[===] Started on Fri Jun 17 16:20:16 2022\n\n[---] start processing checks\n[ RUN ] gromacs_containerized_test %benchmark_info=HECBioSim/hEGFRDimerSmallerPL %nb_impl=gpu %gromacs_image=nvcr.io/hpc/gromacs:2022.1 @daint:gpu+gnu\n[ RUN ] gromacs_containerized_test %benchmark_info=HECBioSim/hEGFRDimerSmallerPL %nb_impl=gpu %gromacs_image=nvcr.io/hpc/gromacs:2021.3 @daint:gpu+gnu\n[ RUN ] gromacs_containerized_test %benchmark_info=HECBioSim/hEGFRDimerSmallerPL %nb_impl=gpu %gromacs_image=nvcr.io/hpc/gromacs:2021 @daint:gpu+gnu\n[ RUN ] gromacs_containerized_test %benchmark_info=HECBioSim/hEGFRDimerSmallerPL %nb_impl=gpu %gromacs_image=nvcr.io/hpc/gromacs:2020.2 @daint:gpu+gnu\n[ RUN ] gromacs_containerized_test %benchmark_info=HECBioSim/hEGFRDimerSmallerPL %nb_impl=gpu %gromacs_image=nvcr.io/hpc/gromacs:2020 @daint:gpu+gnu\n[ RUN ] gromacs_containerized_test %benchmark_info=HECBioSim/hEGFRDimerSmallerPL %nb_impl=gpu %gromacs_image=None @daint:gpu+gnu\n[ OK ] (1/6) gromacs_containerized_test %benchmark_info=HECBioSim/hEGFRDimerSmallerPL %nb_impl=gpu %gromacs_image=nvcr.io/hpc/gromacs:2020.2 @daint:gpu+gnu\n[ OK ] (2/6) gromacs_containerized_test %benchmark_info=HECBioSim/hEGFRDimerSmallerPL %nb_impl=gpu %gromacs_image=nvcr.io/hpc/gromacs:2020 @daint:gpu+gnu\n[ OK ] (3/6) gromacs_containerized_test %benchmark_info=HECBioSim/hEGFRDimerSmallerPL %nb_impl=gpu %gromacs_image=None @daint:gpu+gnu\n[ OK ] (4/6) gromacs_containerized_test %benchmark_info=HECBioSim/hEGFRDimerSmallerPL %nb_impl=gpu %gromacs_image=nvcr.io/hpc/gromacs:2022.1 @daint:gpu+gnu\n[ OK ] (5/6) gromacs_containerized_test %benchmark_info=HECBioSim/hEGFRDimerSmallerPL %nb_impl=gpu %gromacs_image=nvcr.io/hpc/gromacs:2021 @daint:gpu+gnu\n[ OK ] (6/6) gromacs_containerized_test %benchmark_info=HECBioSim/hEGFRDimerSmallerPL %nb_impl=gpu %gromacs_image=nvcr.io/hpc/gromacs:2021.3 @daint:gpu+gnu\n[---] all spawned checks have finished\n\n[ PASSED ] Ran 6/6 test case(s) from 6 check(s) (0 failure(s), 0 skipped)\n[===] Finished on Fri Jun 17 16:23:47 2022\n```\nWe can also inspect the generated job scripts for the native and a containerized run:\n```\ncat output/daint/gpu/gnu/gromacs_containerized_test_0/rfm_gromacs_containerized_test_0_job.sh\n```\n```\n#!/bin/bash\n#SBATCH --job-name=\"rfm_gromacs_containerized_test_0_job\"\n#SBATCH --ntasks=1\n#SBATCH --ntasks-per-node=1\n#SBATCH --cpus-per-task=12\n#SBATCH --output=rfm_gromacs_containerized_test_0_job.out\n#SBATCH --error=rfm_gromacs_containerized_test_0_job.err\n#SBATCH -A csstaff\n#SBATCH --constraint=gpu\n#SBATCH --hint=nomultithread module unload PrgEnv-cray module load PrgEnv-gnu module load daint-gpu module load GROMACS curl -LJO https://github.com/victorusu/GROMACS_Benchmark_Suite/raw/1.0.0/HECBioSim/hEGFRDimerSmallerPL/benchmark.tpr srun gmx mdrun -dlb yes -ntomp 12 -npme -1 -v -nb gpu -s benchmark.tpr\n```\nAnd the containerized run:\n```\ncat output/daint/gpu/gnu/gromacs_containerized_test_1/rfm_gromacs_containerized_test_1_job.sh\n```\n```\n#!/bin/bash\n#SBATCH --job-name=\"rfm_gromacs_containerized_test_1_job\"\n#SBATCH --ntasks=1\n#SBATCH --ntasks-per-node=1\n#SBATCH --cpus-per-task=12\n#SBATCH --output=rfm_gromacs_containerized_test_1_job.out\n#SBATCH --error=rfm_gromacs_containerized_test_1_job.err\n#SBATCH -A csstaff\n#SBATCH --constraint=gpu\n#SBATCH --hint=nomultithread module unload PrgEnv-cray module load PrgEnv-gnu module load sarus curl -LJO https://github.com/victorusu/GROMACS_Benchmark_Suite/raw/1.0.0/HECBioSim/hEGFRDimerSmallerPL/benchmark.tpr sarus pull nvcr.io/hpc/gromacs:2020 srun sarus run --mount=type=bind,source=\"/users/user/Devel/reframe/stage/daint/gpu/gnu/gromacs_containerized_test_43\",destination=\"/rfm_workdir\" -w /rfm_workdir nvcr.io/hpc/gromacs:2020 gmx mdrun -dlb yes -ntomp 12 -npme -1 -v -nb gpu -s benchmark.tpr\n```\n##### Writing reusable tests[¶](#writing-reusable-tests)\n\nNew in version 3.5.0.\n\nSo far, all the examples shown above were tight to a particular system or configuration, which makes reusing these tests in other systems not straightforward.\nHowever, the introduction of the `parameter()` and `variable()` ReFrame built-ins solves this problem, eliminating the need to specify any of the test variables in the `__init__()` method and simplifying code reuse.\nHence, readers who are not familiar with these built-in functions are encouraged to read their basic use examples (see `parameter()` and `variable()`) before delving any deeper into this tutorial.\n\nIn essence, parameters and variables can be treated as simple class attributes, which allows us to leverage Python’s class inheritance and write more modular tests.\nFor simplicity, we illustrate this concept with the above `ContainerTest` example, where the goal here is to re-write this test as a library that users can simply import from and derive their tests without having to rewrite the bulk of the test.\nAlso, for illustrative purposes, we parameterize this library test on a few different image tags (the above example just used `ubuntu:18.04`) and throw the container commands into a separate bash script just to create some source files.\nThus, removing all the system and configuration specific variables, and moving as many assignments as possible into the class body, the system agnostic library test looks as follows:\n```\ncat tutorials/advanced/library/lib/__init__.py\n```\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\nclass ContainerBase(rfm.RunOnlyRegressionTest, pin_prefix=True):\n '''Test that asserts the ubuntu version of the image.'''\n\n # Derived tests must override this parameter\n platform = parameter()\n image_prefix = variable(str, value='')\n\n # Parametrize the test on two different versions of ubuntu.\n dist = parameter(['18.04', '20.04'])\n dist_name = variable(dict, value={\n '18.04': 'Bionic Beaver',\n '20.04': 'Focal Fossa',\n })\n\n @run_after('setup')\n def set_description(self):\n self.descr = (\n f'Run commands inside a container using ubuntu {self.dist}'\n )\n\n @run_before('run')\n def set_container_platform(self):\n self.container_platform = self.platform\n self.container_platform.image = (\n f'{self.image_prefix}ubuntu:{self.dist}'\n )\n self.container_platform.command = (\n \"bash -c /rfm_workdir/get_os_release.sh\"\n )\n\n @property\n def os_release_pattern(self):\n name = self.dist_name[self.dist]\n return rf'{self.dist}.\\d+ LTS \\({name}\\)'\n\n @sanity_function\n def assert_release(self):\n return sn.all([\n sn.assert_found(self.os_release_pattern, 'release.txt'),\n sn.assert_found(self.os_release_pattern, self.stdout)\n ])\n```\nNote that the class `ContainerBase` is not decorated since it does not specify the required variables `valid_systems` and `valid_prog_environs`, and it declares the `platform` parameter without any defined values assigned.\nHence, the user can simply derive from this test and specialize it to use the desired container platforms.\nSince the parameters are defined directly in the class body, the user is also free to override or extend any of the other parameters in a derived test.\nIn this example, we have parameterized the base test to run with the `ubuntu:18.04` and `ubuntu:20.04` images, but these values from `dist` (and also the `dist_name` variable) could be modified by the derived class if needed.\n\nOn the other hand, the rest of the test depends on the values from the test parameters, and a parameter is only assigned a specific value after the class has been instantiated.\nThus, the rest of the test is expressed as hooks, without the need to write anything in the `__init__()` method.\nIn fact, writing the test in this way permits having hooks that depend on undefined variables or parameters.\nThis is the case with the `set_container_platform()` hook, which depends on the undefined parameter `platform`.\nHence, the derived test **must** define all the required parameters and variables; otherwise ReFrame will notice that the test is not well defined and will raise an error accordingly.\n\nBefore moving ahead with the derived test, note that the `ContainerBase` class takes the additional argument `pin_prefix=True`, which locks the prefix of all derived tests to this base test.\nThis will allow the retrieval of the sources located in the library by any derived test, regardless of what their containing directory is.\n```\ncat tutorials/advanced/library/lib/src/get_os_release.sh\n```\n```\n#!/bin/bash cat /etc/os-release | tee /rfm_workdir/release.txt\n```\nNow from the user’s perspective, the only thing to do is to import the above base test and specify the required variables and parameters.\nFor consistency with the above example, we set the `platform` parameter to use Sarus and Singularity, and we configure the test to run on Piz Daint with the built-in programming environment.\nHence, the above `ContainerTest` is now reduced to the following:\n```\ncat tutorials/advanced/library/usr/container_test.py\n```\n```\nimport tutorials.advanced.library.lib as lib\n\n@rfm.simple_test class ContainerTest(lib.ContainerBase):\n platform = parameter(['Sarus', 'Singularity'])\n valid_systems = ['daint:gpu']\n valid_prog_environs = ['builtin']\n\n @run_after('setup')\n def set_image_prefix(self):\n if self.platform == 'Singularity':\n self.image_prefix = 'docker://'\n```\nIn a similar fashion, any other user could reuse the above `ContainerBase` class and write the test for their own system with a few lines of code.\n\n*Happy test sharing!*\n\n#### Tutorial 3: Using Dependencies in ReFrame Tests[¶](#tutorial-3-using-dependencies-in-reframe-tests)\n\nNew in version 2.21.\n\nA ReFrame test may define dependencies to other tests.\nAn example scenario is to test different runtime configurations of a benchmark that you need to compile, or run a scaling analysis of a code.\nIn such cases, you don’t want to download and rebuild your test for each runtime configuration.\nYou could have a test where only the sources are fetched, and which all build tests would depend on.\nAnd, similarly, all the runtime tests would depend on their corresponding build test.\nThis is the approach we take with the following example, that fetches, builds and runs several [OSU benchmarks](http://mvapich.cse.ohio-state.edu/benchmarks/).\nWe first create a basic run-only test, that fetches the benchmarks:\n```\ncat tutorials/deps/osu_benchmarks.py\n```\n```\n@rfm.simple_test class OSUDownloadTest(rfm.RunOnlyRegressionTest):\n descr = 'OSU benchmarks download sources'\n valid_systems = ['daint:login']\n valid_prog_environs = ['builtin']\n executable = 'wget'\n executable_opts = [\n 'http://mvapich.cse.ohio-state.edu/download/mvapich/osu-micro-benchmarks-5.6.2.tar.gz' # noqa: E501\n ]\n postrun_cmds = [\n 'tar xzf osu-micro-benchmarks-5.6.2.tar.gz'\n ]\n\n @sanity_function\n def validate_download(self):\n return sn.assert_true(os.path.exists('osu-micro-benchmarks-5.6.2'))\n```\nThis test doesn’t need any specific programming environment, so we simply pick the `builtin` environment in the `login` partition.\nThe build tests would then copy the benchmark code and build it for the different programming environments:\n```\n@rfm.simple_test class OSUBuildTest(rfm.CompileOnlyRegressionTest):\n descr = 'OSU benchmarks build test'\n valid_systems = ['daint:gpu']\n valid_prog_environs = ['gnu', 'nvidia', 'intel']\n build_system = 'Autotools'\n\n @run_after('init')\n def inject_dependencies(self):\n self.depends_on('OSUDownloadTest', udeps.fully)\n\n @require_deps\n def set_sourcedir(self, OSUDownloadTest):\n self.sourcesdir = os.path.join(\n OSUDownloadTest(part='login', environ='builtin').stagedir,\n 'osu-micro-benchmarks-5.6.2'\n )\n\n @run_before('compile')\n def set_build_system_attrs(self):\n self.build_system.max_concurrency = 8\n\n @sanity_function\n def validate_build(self):\n return sn.assert_not_found('error', self.stderr)\n```\nThe only new thing that comes in with the `OSUBuildTest` test is the following:\n```\n @run_after('init')\n def inject_dependencies(self):\n self.depends_on('OSUDownloadTest', udeps.fully)\n```\nHere we tell ReFrame that this test depends on a test named `OSUDownloadTest`.\nThis test may or may not be defined in the same test file; all ReFrame needs is the test name.\nThe [`depends_on()`](index.html#reframe.core.pipeline.RegressionTest.depends_on) function will create dependencies between the individual test cases of the `OSUBuildTest` and the `OSUDownloadTest`, such that all the test cases of `OSUBuildTest` will depend on the outcome of the `OSUDownloadTest`.\nThis behaviour can be changed, but it is covered in detail in [How Test Dependencies Work In ReFrame](index.html#document-dependencies).\nYou can create arbitrary test dependency graphs, but they need to be acyclic.\nIf ReFrame detects cyclic dependencies, it will refuse to execute the set of tests and will issue an error pointing out the cycle.\n\nA ReFrame test with dependencies will execute, i.e., enter its “setup” stage, only after *all* of its dependencies have succeeded.\nIf any of its dependencies fails, the current test will be marked as failure as well.\n\nThe next step for the `OSUBuildTest` is to set its `sourcesdir` to point to the source code that was fetched by the `OSUDownloadTest`.\nThis is achieved with the following specially decorated function:\n```\n @require_deps\n def set_sourcedir(self, OSUDownloadTest):\n self.sourcesdir = os.path.join(\n OSUDownloadTest(part='login', environ='builtin').stagedir,\n 'osu-micro-benchmarks-5.6.2'\n )\n```\nThe `@require_deps` decorator binds each argument of the decorated function to the corresponding target dependency.\nIn order for the binding to work correctly the function arguments must be named after the target dependencies.\nReferring to a dependency only by the test’s name is not enough, since a test might be associated with multiple programming environments.\nFor this reason, each dependency argument is actually bound to a function that accepts as argument the name of the target partition and target programming environment.\nIf no arguments are passed, the current programming environment is implied, such that `OSUDownloadTest()` is equivalent to `OSUDownloadTest(self.current_environ.name, self.current_partition.name)`.\nIn this case, since both the partition and environment of the target dependency do not much those of the current test, we need to specify both.\n\nThis call returns the actual test case of the dependency that has been executed.\nThis allows you to access any attribute from the target test, as we do in this example by accessing the target test’s stage directory, which we use to construct the sourcesdir of the test.\n\nFor the next test we need to use the OSU benchmark binaries that we just built, so as to run the MPI ping-pong benchmark.\nHere is the relevant part:\n```\nclass OSUBenchmarkTestBase(rfm.RunOnlyRegressionTest):\n '''Base class of OSU benchmarks runtime tests'''\n\n valid_systems = ['daint:gpu']\n valid_prog_environs = ['gnu', 'nvidia', 'intel']\n sourcesdir = None\n num_tasks = 2\n num_tasks_per_node = 1\n\n @run_after('init')\n def set_dependencies(self):\n self.depends_on('OSUBuildTest', udeps.by_env)\n\n @sanity_function\n def validate_test(self):\n return sn.assert_found(r'^8', self.stdout)\n\n@rfm.simple_test class OSULatencyTest(OSUBenchmarkTestBase):\n descr = 'OSU latency test'\n\n @require_deps\n def set_executable(self, OSUBuildTest):\n self.executable = os.path.join(\n OSUBuildTest().stagedir,\n 'mpi', 'pt2pt', 'osu_latency'\n )\n self.executable_opts = ['-x', '100', '-i', '1000']\n\n @performance_function('us')\n def latency(self):\n return sn.extractsingle(r'^8\\s+(\\S+)', self.stdout, 1, float)\n```\nFirst, since we will have multiple similar benchmarks, we move all the common functionality to the `OSUBenchmarkTestBase` base class.\nAgain nothing new here; we are going to use two nodes for the benchmark and we set [`sourcesdir`](index.html#reframe.core.pipeline.RegressionTest.sourcesdir) to `None`, since none of the benchmark tests will use any additional resources.\nAs done previously, we define the dependencies with the following:\n```\n @run_after('init')\n def set_dependencies(self):\n self.depends_on('OSUBuildTest', udeps.by_env)\n```\nHere we tell ReFrame that this test depends on a test named `OSUBuildTest` “by environment.”\nThis means that the test cases of this test will only depend on the test cases of the `OSUBuildTest` that use the same environment;\npartitions may be different.\n\nThe next step for the `OSULatencyTest` is to set its executable to point to the binary produced by the `OSUBuildTest`.\nThis is achieved with the following specially decorated function:\n```\n @require_deps\n def set_executable(self, OSUBuildTest):\n self.executable = os.path.join(\n OSUBuildTest().stagedir,\n 'mpi', 'pt2pt', 'osu_latency'\n )\n self.executable_opts = ['-x', '100', '-i', '1000']\n```\nThis concludes the presentation of the `OSULatencyTest` test. The `OSUBandwidthTest` is completely analogous.\n\nThe `OSUAllreduceTest` shown below is similar to the other two, except that it is parameterized.\nIt is essentially a scalability test that is running the `osu_allreduce` executable created by the `OSUBuildTest` for 2, 4, 8 and 16 nodes.\n```\n@rfm.simple_test class OSUAllreduceTest(OSUBenchmarkTestBase):\n mpi_tasks = parameter(1 << i for i in range(1, 5))\n descr = 'OSU Allreduce test'\n\n @run_after('init')\n def set_num_tasks(self):\n self.num_tasks = self.mpi_tasks\n\n @require_deps\n def set_executable(self, OSUBuildTest):\n self.executable = os.path.join(\n OSUBuildTest().stagedir,\n 'mpi', 'collective', 'osu_allreduce'\n )\n self.executable_opts = ['-m', '8', '-x', '1000', '-i', '20000']\n\n @performance_function('us')\n def latency(self):\n return sn.extractsingle(r'^8\\s+(\\S+)', self.stdout, 1, float)\n```\nThe full set of OSU example tests is shown below:\n```\n# Copyright 2016-2022 Swiss National Supercomputing Centre (CSCS/ETH Zurich)\n# ReFrame Project Developers. See the top-level LICENSE file for details.\n#\n# SPDX-License-Identifier: BSD-3-Clause\n\nimport os\n\nimport reframe as rfm import reframe.utility.sanity as sn import reframe.utility.udeps as udeps\n\n# rfmdocstart: osupingpong class OSUBenchmarkTestBase(rfm.RunOnlyRegressionTest):\n '''Base class of OSU benchmarks runtime tests'''\n\n valid_systems = ['daint:gpu']\n valid_prog_environs = ['gnu', 'nvidia', 'intel']\n sourcesdir = None\n num_tasks = 2\n num_tasks_per_node = 1\n\n @run_after('init')\n def set_dependencies(self):\n self.depends_on('OSUBuildTest', udeps.by_env)\n\n @sanity_function\n def validate_test(self):\n return sn.assert_found(r'^8', self.stdout)\n\n@rfm.simple_test class OSULatencyTest(OSUBenchmarkTestBase):\n descr = 'OSU latency test'\n\n @require_deps\n def set_executable(self, OSUBuildTest):\n self.executable = os.path.join(\n OSUBuildTest().stagedir,\n 'mpi', 'pt2pt', 'osu_latency'\n )\n self.executable_opts = ['-x', '100', '-i', '1000']\n\n @performance_function('us')\n def latency(self):\n return sn.extractsingle(r'^8\\s+(\\S+)', self.stdout, 1, float)\n# rfmdocend: osupingpong\n\n@rfm.simple_test class OSUBandwidthTest(OSUBenchmarkTestBase):\n descr = 'OSU bandwidth test'\n\n @require_deps\n def set_executable(self, OSUBuildTest):\n self.executable = os.path.join(\n OSUBuildTest().stagedir,\n 'mpi', 'pt2pt', 'osu_bw'\n )\n self.executable_opts = ['-x', '100', '-i', '1000']\n\n @performance_function('MB/s')\n def bandwidth(self):\n return sn.extractsingle(r'^4194304\\s+(\\S+)',\n self.stdout, 1, float)\n\n@rfm.simple_test class OSUAllreduceTest(OSUBenchmarkTestBase):\n mpi_tasks = parameter(1 << i for i in range(1, 5))\n descr = 'OSU Allreduce test'\n\n @run_after('init')\n def set_num_tasks(self):\n self.num_tasks = self.mpi_tasks\n\n @require_deps\n def set_executable(self, OSUBuildTest):\n self.executable = os.path.join(\n OSUBuildTest().stagedir,\n 'mpi', 'collective', 'osu_allreduce'\n )\n self.executable_opts = ['-m', '8', '-x', '1000', '-i', '20000']\n\n @performance_function('us')\n def latency(self):\n return sn.extractsingle(r'^8\\s+(\\S+)', self.stdout, 1, float)\n\n@rfm.simple_test class OSUBuildTest(rfm.CompileOnlyRegressionTest):\n descr = 'OSU benchmarks build test'\n valid_systems = ['daint:gpu']\n valid_prog_environs = ['gnu', 'nvidia', 'intel']\n build_system = 'Autotools'\n\n @run_after('init')\n def inject_dependencies(self):\n self.depends_on('OSUDownloadTest', udeps.fully)\n\n @require_deps\n def set_sourcedir(self, OSUDownloadTest):\n self.sourcesdir = os.path.join(\n OSUDownloadTest(part='login', environ='builtin').stagedir,\n 'osu-micro-benchmarks-5.6.2'\n )\n\n @run_before('compile')\n def set_build_system_attrs(self):\n self.build_system.max_concurrency = 8\n\n @sanity_function\n def validate_build(self):\n return sn.assert_not_found('error', self.stderr)\n\n@rfm.simple_test class OSUDownloadTest(rfm.RunOnlyRegressionTest):\n descr = 'OSU benchmarks download sources'\n valid_systems = ['daint:login']\n valid_prog_environs = ['builtin']\n executable = 'wget'\n executable_opts = [\n 'http://mvapich.cse.ohio-state.edu/download/mvapich/osu-micro-benchmarks-5.6.2.tar.gz' # noqa: E501\n ]\n postrun_cmds = [\n 'tar xzf osu-micro-benchmarks-5.6.2.tar.gz'\n ]\n\n @sanity_function\n def validate_download(self):\n return sn.assert_true(os.path.exists('osu-micro-benchmarks-5.6.2'))\n```\nNotice that the order in which dependencies are defined in a test file is irrelevant.\nIn this case, we define `OSUBuildTest` at the end.\nReFrame will make sure to properly sort the tests and execute them.\n\nHere is the output when running the OSU tests with the asynchronous execution policy:\n```\n./bin/reframe -c tutorials/deps/osu_benchmarks.py -r\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2\n command: './bin/reframe -c tutorials/deps/osu_benchmarks.py -r'\n launched by: user@host\n working directory: '/home/user/Devel/reframe'\n settings files: '<builtin>', '/home/user/Devel/reframe/tutorials/config/daint.py'\n check search path: '/home/user/Devel/reframe/tutorials/deps/osu_benchmarks.py'\n stage directory: '/home/user/Devel/reframe/stage'\n output directory: '/home/user/Devel/reframe/output'\n log files: '/tmp/rfm-r1a7v0w3.log'\n\n[===] Running 8 check(s)\n[===] Started on Tue Nov 15 18:24:00 2022\n\n[---] start processing checks\n[ RUN ] OSUDownloadTest /7de668df @daint:login+builtin\n[ OK ] ( 1/22) OSUDownloadTest /7de668df @daint:login+builtin\n[ RUN ] OSUBuildTest /19b4fb56 @daint:gpu+gnu\n[ RUN ] OSUBuildTest /19b4fb56 @daint:gpu+intel\n[ RUN ] OSUBuildTest /19b4fb56 @daint:gpu+nvidia\n[ OK ] ( 2/22) OSUBuildTest /19b4fb56 @daint:gpu+gnu\n[ RUN ] OSUAllreduceTest %mpi_tasks=16 /7f033d39 @daint:gpu+gnu\n[ RUN ] OSUAllreduceTest %mpi_tasks=8 /005fca19 @daint:gpu+gnu\n[ RUN ] OSUAllreduceTest %mpi_tasks=4 /84b85d90 @daint:gpu+gnu\n[ RUN ] OSUAllreduceTest %mpi_tasks=2 /9d550c4f @daint:gpu+gnu\n[ RUN ] OSUBandwidthTest /764cdb0b @daint:gpu+gnu\n[ RUN ] OSULatencyTest /14f35a43 @daint:gpu+gnu\n[ OK ] ( 3/22) OSUBuildTest /19b4fb56 @daint:gpu+intel\n[ RUN ] OSUAllreduceTest %mpi_tasks=16 /7f033d39 @daint:gpu+intel\n[ RUN ] OSUAllreduceTest %mpi_tasks=8 /005fca19 @daint:gpu+intel\n[ RUN ] OSUAllreduceTest %mpi_tasks=4 /84b85d90 @daint:gpu+intel\n[ OK ] ( 4/22) OSUBuildTest /19b4fb56 @daint:gpu+nvidia\n[ RUN ] OSUAllreduceTest %mpi_tasks=16 /7f033d39 @daint:gpu+nvidia\n[ RUN ] OSUAllreduceTest %mpi_tasks=8 /005fca19 @daint:gpu+nvidia\n[ RUN ] OSUAllreduceTest %mpi_tasks=4 /84b85d90 @daint:gpu+nvidia\n[ RUN ] OSUAllreduceTest %mpi_tasks=2 /9d550c4f @daint:gpu+intel\n[ RUN ] OSUAllreduceTest %mpi_tasks=2 /9d550c4f @daint:gpu+nvidia\n[ RUN ] OSUBandwidthTest /764cdb0b @daint:gpu+intel\n[ RUN ] OSUBandwidthTest /764cdb0b @daint:gpu+nvidia\n[ RUN ] OSULatencyTest /14f35a43 @daint:gpu+intel\n[ RUN ] OSULatencyTest /14f35a43 @daint:gpu+nvidia\n[ OK ] ( 5/22) OSUAllreduceTest %mpi_tasks=4 /84b85d90 @daint:gpu+gnu P: latency: 5.31 us (r:0, l:None, u:None)\n[ OK ] ( 6/22) OSUAllreduceTest %mpi_tasks=2 /9d550c4f @daint:gpu+intel P: latency: 10.07 us (r:0, l:None, u:None)\n[ OK ] ( 7/22) OSUAllreduceTest %mpi_tasks=2 /9d550c4f @daint:gpu+gnu P: latency: 1.67 us (r:0, l:None, u:None)\n[ OK ] ( 8/22) OSUAllreduceTest %mpi_tasks=4 /84b85d90 @daint:gpu+intel P: latency: 24.97 us (r:0, l:None, u:None)\n[ OK ] ( 9/22) OSUAllreduceTest %mpi_tasks=2 /9d550c4f @daint:gpu+nvidia P: latency: 8.92 us (r:0, l:None, u:None)\n[ OK ] (10/22) OSUAllreduceTest %mpi_tasks=16 /7f033d39 @daint:gpu+intel P: latency: 14.78 us (r:0, l:None, u:None)\n[ OK ] (11/22) OSULatencyTest /14f35a43 @daint:gpu+nvidia P: latency: 2.19 us (r:0, l:None, u:None)\n[ OK ] (12/22) OSULatencyTest /14f35a43 @daint:gpu+gnu P: latency: 1.76 us (r:0, l:None, u:None)\n[ OK ] (13/22) OSUAllreduceTest %mpi_tasks=16 /7f033d39 @daint:gpu+gnu P: latency: 19.54 us (r:0, l:None, u:None)\n[ OK ] (14/22) OSULatencyTest /14f35a43 @daint:gpu+intel P: latency: 4.4 us (r:0, l:None, u:None)\n[ OK ] (15/22) OSUAllreduceTest %mpi_tasks=4 /84b85d90 @daint:gpu+nvidia P: latency: 6.88 us (r:0, l:None, u:None)\n[ OK ] (16/22) OSUAllreduceTest %mpi_tasks=8 /005fca19 @daint:gpu+intel P: latency: 21.37 us (r:0, l:None, u:None)\n[ OK ] (17/22) OSUAllreduceTest %mpi_tasks=8 /005fca19 @daint:gpu+gnu P: latency: 10.15 us (r:0, l:None, u:None)\n[ OK ] (18/22) OSUAllreduceTest %mpi_tasks=8 /005fca19 @daint:gpu+nvidia P: latency: 52.87 us (r:0, l:None, u:None)\n[ OK ] (19/22) OSUAllreduceTest %mpi_tasks=16 /7f033d39 @daint:gpu+nvidia P: latency: 64.77 us (r:0, l:None, u:None)\n[ OK ] (20/22) OSUBandwidthTest /764cdb0b @daint:gpu+intel P: bandwidth: 9118.51 MB/s (r:0, l:None, u:None)\n[ OK ] (21/22) OSUBandwidthTest /764cdb0b @daint:gpu+nvidia P: bandwidth: 8476.18 MB/s (r:0, l:None, u:None)\n[ OK ] (22/22) OSUBandwidthTest /764cdb0b @daint:gpu+gnu P: bandwidth: 8326.06 MB/s (r:0, l:None, u:None)\n[---] all spawned checks have finished\n\n[ PASSED ] Ran 22/22 test case(s) from 8 check(s) (0 failure(s), 0 skipped)\n[===] Finished on Tue Nov 15 18:27:13 2022 Run report saved in '/home/user/.reframe/reports/run-report-3.json'\nLog file(s) saved in '/tmp/rfm-r1a7v0w3.log'\n```\nBefore starting running the tests, ReFrame topologically sorts them based on their dependencies and schedules them for running using the selected execution policy.\nWith the serial execution policy, ReFrame simply executes the tests to completion as they “arrive,” since the tests are already topologically sorted.\nIn the asynchronous execution policy, tests are spawned and not waited for.\nIf a test’s dependencies have not yet completed, it will not start its execution immediately.\n\nReFrame’s runtime takes care of properly cleaning up the resources of the tests respecting dependencies.\nNormally when an individual test finishes successfully, its stage directory is cleaned up.\nHowever, if other tests are depending on this one, this would be catastrophic, since most probably the dependent tests would need the outcome of this test.\nReFrame fixes that by not cleaning up the stage directory of a test until all its dependent tests have finished successfully.\n\nWhen selecting tests using the test filtering options, such as the [`-t`](index.html#cmdoption-0), [`-n`](index.html#cmdoption-n) etc., ReFrame will automatically select any dependencies of these tests as well.\nFor example, if we select only the `OSULatencyTest` for running, ReFrame will also select the `OSUBuildTest` and the `OSUDownloadTest`:\n```\n./bin/reframe -c tutorials/deps/osu_benchmarks.py -n OSULatencyTest -l\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2\n command: './bin/reframe -c tutorials/deps/osu_benchmarks.py -n OSULatencyTest -l'\n launched by: user@host\n working directory: '/home/user/Devel/reframe'\n settings files: '<builtin>', '/home/user/Devel/reframe/tutorials/config/daint.py'\n check search path: '/home/user/Devel/reframe/tutorials/deps/osu_benchmarks.py'\n stage directory: '/home/user/Devel/reframe/stage'\n output directory: '/home/user/Devel/reframe/output'\n log files: '/tmp/rfm-7id7z75s.log'\n\n[List of matched checks]\n- OSULatencyTest /14f35a43\n ^OSUBuildTest /19b4fb56\n ^OSUDownloadTest /7de668df Found 3 check(s)\n\nLog file(s) saved in '/tmp/rfm-7id7z75s.log'\n```\nFinally, when ReFrame cannot resolve a dependency of a test, it will issue a warning and skip completely all the test cases that recursively depend on this one.\nIn the following example, we restrict the run of the `OSULatencyTest` to the `daint:gpu` partition.\nThis is problematic, since its dependencies cannot run on this partition and, particularly, the `OSUDownloadTest`.\nAs a result, its immediate dependency `OSUBuildTest` will be skipped, which will eventually cause all combinations of the `OSULatencyTest` to be skipped.\n```\n./bin/reframe -c tutorials/deps/osu_benchmarks.py --system=daint:gpu -n OSULatencyTest -l\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2\n command: './bin/reframe -c tutorials/deps/osu_benchmarks.py -n OSULatencyTest --system=daint:gpu -l'\n launched by: user@host\n working directory: '/home/user/Devel/reframe'\n settings files: '<builtin>', '/home/user/Devel/reframe/tutorials/config/daint.py'\n check search path: '/home/user/Devel/reframe/tutorials/deps/osu_benchmarks.py'\n stage directory: '/home/user/Devel/reframe/stage'\n output directory: '/home/user/Devel/reframe/output'\n log files: '/tmp/rfm-12gjxnvc.log'\n\nWARNING: could not resolve dependency: ('OSUBuildTest', 'daint:gpu', 'gnu') -> 'OSUDownloadTest'\nWARNING: could not resolve dependency: ('OSUBuildTest', 'daint:gpu', 'intel') -> 'OSUDownloadTest'\nWARNING: could not resolve dependency: ('OSUBuildTest', 'daint:gpu', 'nvidia') -> 'OSUDownloadTest'\nWARNING: skipping all dependent test cases\n - ('OSUBuildTest', 'daint:gpu', 'intel')\n - ('OSUBandwidthTest', 'daint:gpu', 'intel')\n - ('OSUBuildTest', 'daint:gpu', 'nvidia')\n - ('OSULatencyTest', 'daint:gpu', 'intel')\n - ('OSUAllreduceTest_3', 'daint:gpu', 'nvidia')\n - ('OSUBuildTest', 'daint:gpu', 'gnu')\n - ('OSUAllreduceTest_1', 'daint:gpu', 'nvidia')\n - ('OSUAllreduceTest_0', 'daint:gpu', 'intel')\n - ('OSUAllreduceTest_2', 'daint:gpu', 'nvidia')\n - ('OSUBandwidthTest', 'daint:gpu', 'gnu')\n - ('OSULatencyTest', 'daint:gpu', 'gnu')\n - ('OSUAllreduceTest_2', 'daint:gpu', 'intel')\n - ('OSUAllreduceTest_3', 'daint:gpu', 'intel')\n - ('OSUAllreduceTest_1', 'daint:gpu', 'intel')\n - ('OSUAllreduceTest_0', 'daint:gpu', 'nvidia')\n - ('OSUBandwidthTest', 'daint:gpu', 'nvidia')\n - ('OSULatencyTest', 'daint:gpu', 'nvidia')\n - ('OSUAllreduceTest_2', 'daint:gpu', 'gnu')\n - ('OSUAllreduceTest_1', 'daint:gpu', 'gnu')\n - ('OSUAllreduceTest_3', 'daint:gpu', 'gnu')\n - ('OSUAllreduceTest_0', 'daint:gpu', 'gnu')\n\n[List of matched checks]\nFound 0 check(s)\n\nLog file(s) saved in '/tmp/rfm-12gjxnvc.log'\n```\n##### Listing Dependencies[¶](#listing-dependencies)\n\nAs shown in the listing of `OSULatencyTest` before, the full dependency chain of the test is listed along with the test.\nEach target dependency is printed in a new line prefixed by the `^` character and indented proportionally to its level.\nIf a target dependency appears in multiple paths, it will only be listed once.\n\nThe default test listing will list the dependencies at the test level or the *conceptual* dependencies.\nReFrame generates multiple test cases from each test depending on the target system configuration.\nWe have seen in the [Tutorial 1: Getting Started with ReFrame](index.html#document-tutorial_basics) already how the STREAM benchmark generated many more test cases when it was run in a HPC system with multiple partitions and programming environments.\nThese are the *actual* depedencies and form the actual test case graph that will be executed by the runtime.\nThe mapping of a test to its concrete test cases that will be executed on a system is called *test concretization*.\nYou can view the exact concretization of the selected tests with `--list=concretized` or simply `-lC`.\nHere is how the OSU benchmarks of this tutorial are concretized on the system `daint`:\n```\n./bin/reframe -c tutorials/deps/osu_benchmarks.py -lC\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2\n command: './bin/reframe -c tutorials/deps/osu_benchmarks.py -lC'\n launched by: user@host\n working directory: '/home/user/Devel/reframe'\n settings files: '<builtin>', '/home/user/Devel/reframe/tutorials/config/daint.py'\n check search path: '/home/user/Devel/reframe/tutorials/deps/osu_benchmarks.py'\n stage directory: '/home/user/Devel/reframe/stage'\n output directory: '/home/user/Devel/reframe/output'\n log files: '/tmp/rfm-fremrbwf.log'\n\n[List of matched checks]\n- OSUAllreduceTest %mpi_tasks=16 /7f033d39 @daint:gpu+gnu\n ^OSUBuildTest /19b4fb56 @daint:gpu+gnu\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSUAllreduceTest %mpi_tasks=16 /7f033d39 @daint:gpu+intel\n ^OSUBuildTest /19b4fb56 @daint:gpu+intel\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSUAllreduceTest %mpi_tasks=16 /7f033d39 @daint:gpu+nvidia\n ^OSUBuildTest /19b4fb56 @daint:gpu+nvidia\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSUAllreduceTest %mpi_tasks=8 /005fca19 @daint:gpu+gnu\n ^OSUBuildTest /19b4fb56 @daint:gpu+gnu\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSUAllreduceTest %mpi_tasks=8 /005fca19 @daint:gpu+intel\n ^OSUBuildTest /19b4fb56 @daint:gpu+intel\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSUAllreduceTest %mpi_tasks=8 /005fca19 @daint:gpu+nvidia\n ^OSUBuildTest /19b4fb56 @daint:gpu+nvidia\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSUAllreduceTest %mpi_tasks=4 /84b85d90 @daint:gpu+gnu\n ^OSUBuildTest /19b4fb56 @daint:gpu+gnu\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSUAllreduceTest %mpi_tasks=4 /84b85d90 @daint:gpu+intel\n ^OSUBuildTest /19b4fb56 @daint:gpu+intel\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSUAllreduceTest %mpi_tasks=4 /84b85d90 @daint:gpu+nvidia\n ^OSUBuildTest /19b4fb56 @daint:gpu+nvidia\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSUAllreduceTest %mpi_tasks=2 /9d550c4f @daint:gpu+gnu\n ^OSUBuildTest /19b4fb56 @daint:gpu+gnu\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSUAllreduceTest %mpi_tasks=2 /9d550c4f @daint:gpu+intel\n ^OSUBuildTest /19b4fb56 @daint:gpu+intel\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSUAllreduceTest %mpi_tasks=2 /9d550c4f @daint:gpu+nvidia\n ^OSUBuildTest /19b4fb56 @daint:gpu+nvidia\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSUBandwidthTest /764cdb0b @daint:gpu+gnu\n ^OSUBuildTest /19b4fb56 @daint:gpu+gnu\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSUBandwidthTest /764cdb0b @daint:gpu+intel\n ^OSUBuildTest /19b4fb56 @daint:gpu+intel\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSUBandwidthTest /764cdb0b @daint:gpu+nvidia\n ^OSUBuildTest /19b4fb56 @daint:gpu+nvidia\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSULatencyTest /14f35a43 @daint:gpu+gnu\n ^OSUBuildTest /19b4fb56 @daint:gpu+gnu\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSULatencyTest /14f35a43 @daint:gpu+intel\n ^OSUBuildTest /19b4fb56 @daint:gpu+intel\n ^OSUDownloadTest /7de668df @daint:login+builtin\n- OSULatencyTest /14f35a43 @daint:gpu+nvidia\n ^OSUBuildTest /19b4fb56 @daint:gpu+nvidia\n ^OSUDownloadTest /7de668df @daint:login+builtin Concretized 22 test case(s)\n\nLog file(s) saved in '/tmp/rfm-fremrbwf.log'\n```\nNotice how the various test cases of the run benchmarks depend on the corresponding test cases of the build tests.\n\nThe concretization of test cases changes if a specifc partition or programming environment is passed from the command line or, of course, if the test is run on a different system.\nIf we scope our programming environments to `gnu` and `builtin` only, ReFrame will generate 8 test cases only instead of 22:\n\nNote\n\nIf we do not select the `builtin` environment, we will end up with a dangling dependency as in the example above and ReFrame will skip all the dependent test cases.\n```\n./bin/reframe -c tutorials/deps/osu_benchmarks.py -n OSULatencyTest -L -p builtin -p gnu\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2\n command: './bin/reframe -c tutorials/deps/osu_benchmarks.py -n OSULatencyTest -L -p builtin -p gnu'\n launched by: user@host\n working directory: '/home/user/Devel/reframe'\n settings files: '<builtin>', '/home/user/Devel/reframe/tutorials/config/daint.py'\n check search path: '/home/user/Devel/reframe/tutorials/deps/osu_benchmarks.py'\n stage directory: '/home/user/Devel/reframe/stage'\n output directory: '/home/user/Devel/reframe/output'\n log files: '/tmp/rfm-7hnco47r.log'\n\n[List of matched checks]\n- OSULatencyTest /14f35a43 [variant: 0, file: '/home/user/Devel/reframe/tutorials/deps/osu_benchmarks.py']\n ^OSUBuildTest /19b4fb56 [variant: 0, file: '/home/user/Devel/reframe/tutorials/deps/osu_benchmarks.py']\n ^OSUDownloadTest /7de668df [variant: 0, file: '/home/user/Devel/reframe/tutorials/deps/osu_benchmarks.py']\nFound 3 check(s)\n\nLog file(s) saved in '/tmp/rfm-7hnco47r.log'\n```\nTo gain a deeper understanding on how test dependencies work in Reframe, please refer to [How Test Dependencies Work In ReFrame](index.html#document-dependencies).\n\n##### Depending on Parameterized Tests[¶](#depending-on-parameterized-tests)\n\nAs shown earlier in this section, tests define their dependencies by referencing the target tests by their unique name.\nThis is straightforward when referring to regular tests, where their name matches the class name, but it becomes cumbersome trying to refer to a parameterized tests, since no safe assumption should be made as of the variant number of the test or how the parameters are encoded in the name.\nIn order to safely and reliably refer to a parameterized test, you should use the [`get_variant_nums()`](index.html#reframe.core.pipeline.RegressionMixin.get_variant_nums) and [`variant_name()`](index.html#reframe.core.pipeline.RegressionMixin.variant_name) class methods as shown in the following example:\n```\n# Copyright 2016-2022 Swiss National Supercomputing Centre (CSCS/ETH Zurich)\n# ReFrame Project Developers. See the top-level LICENSE file for details.\n#\n# SPDX-License-Identifier: BSD-3-Clause\n\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class TestA(rfm.RunOnlyRegressionTest):\n z = parameter(range(10))\n executable = 'echo'\n valid_systems = ['*']\n valid_prog_environs = ['*']\n\n @run_after('init')\n def set_exec_opts(self):\n self.executable_opts = [str(self.z)]\n\n @sanity_function\n def validate(self):\n return sn.assert_eq(\n sn.extractsingle(r'\\d+', self.stdout, 0, int), self.z\n )\n\n@rfm.simple_test class TestB(rfm.RunOnlyRegressionTest):\n executable = 'echo'\n valid_systems = ['*']\n valid_prog_environs = ['*']\n sanity_patterns = sn.assert_true(1)\n\n @run_after('init')\n def setdeps(self):\n variants = TestA.get_variant_nums(z=lambda x: x > 5)\n for v in variants:\n self.depends_on(TestA.variant_name(v))\n```\nIn this example, `TestB` depends only on selected variants of `TestA`.\nThe `get_variant_nums()` method accepts a set of key-value pairs representing the target test parameters and selector functions and returns the list of the variant numbers that correspond to these variants.\nUsing the `variant_name()` subsequently, we can get the actual name of the variant.\n```\n./bin/reframe -c tutorials/deps/parameterized.py -l\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe -c tutorials/deps/parameterized.py -l'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>', '/home/user/Repositories/reframe/tutorials/config/tresa.py'\n check search path: '/home/user/Repositories/reframe/tutorials/deps/parameterized.py'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/<KEY>00gp/T/rfm-u9ryq5d3.log'\n\n[List of matched checks]\n- TestB /cc291487\n ^TestA %z=9 /034f091a\n ^TestA %z=8 /a093d19f\n ^TestA %z=7 /77b4b8e6\n ^TestA %z=6 /40ce4759\n- TestA %z=5 /aa0cffc9\n- TestA %z=4 /83cd5dec\n- TestA %z=3 /1c51609b\n- TestA %z=2 /707b752c\n- TestA %z=1 /c65657d5\n- TestA %z=0 /1b9f44df Found 11 check(s)\n\nLog file(s) saved in '/var/folders/h7/k7cgrdl13r996m4dmsvjq7v80000gp/T/rfm-u9ryq5d3.log'\n```\n#### Tutorial 4: Using Test Fixtures[¶](#tutorial-4-using-test-fixtures)\n\nNew in version 3.9.0.\n\nA fixture in ReFrame is a test that manages a resource of another test.\nFixtures can be chained to create essentially a graph of dependencies.\nSimilarly to test dependencies, the test that uses the fixture will not execute until its fixture has executed.\nIn this tutorial, we will rewrite the OSU benchmarks example presented in [Tutorial 3: Using Dependencies in ReFrame Tests](index.html#document-tutorial_deps) using fixtures.\nWe will cover only the basic concepts of fixtures that will allow you to start using them in your tests.\nFor the full documentation of the test fixtures, you should refer to the [Test API Reference](index.html#document-regression_test_api) documentation.\n\nThe full example of the OSU benchmarks using test fixtures is shown below with the relevant parts highlighted:\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\nclass fetch_osu_benchmarks(rfm.RunOnlyRegressionTest):\n descr = 'Fetch OSU benchmarks'\n version = variable(str, value='5.6.2')\n executable = 'wget'\n executable_opts = [\n f'http://mvapich.cse.ohio-state.edu/download/mvapich/osu-micro-benchmarks-{version}.tar.gz' # noqa: E501\n ]\n local = True\n\n @sanity_function\n def validate_download(self):\n return sn.assert_eq(self.job.exitcode, 0)\n\nclass build_osu_benchmarks(rfm.CompileOnlyRegressionTest):\n descr = 'Build OSU benchmarks'\n build_system = 'Autotools'\n build_prefix = variable(str)\n osu_benchmarks = fixture(fetch_osu_benchmarks, scope='session')\n\n @run_before('compile')\n def prepare_build(self):\n tarball = f'osu-micro-benchmarks-{self.osu_benchmarks.version}.tar.gz'\n self.build_prefix = tarball[:-7] # remove .tar.gz extension\n\n fullpath = os.path.join(self.osu_benchmarks.stagedir, tarball)\n self.prebuild_cmds = [\n f'cp {fullpath} {self.stagedir}',\n f'tar xzf {tarball}',\n f'cd {self.build_prefix}'\n ]\n self.build_system.max_concurrency = 8\n\n @sanity_function\n def validate_build(self):\n # If compilation fails, the test would fail in any case, so nothing to\n # further validate here.\n return True\n\nclass OSUBenchmarkTestBase(rfm.RunOnlyRegressionTest):\n '''Base class of OSU benchmarks runtime tests'''\n\n valid_systems = ['daint:gpu']\n valid_prog_environs = ['gnu', 'nvidia', 'intel']\n num_tasks = 2\n num_tasks_per_node = 1\n osu_binaries = fixture(build_osu_benchmarks, scope='environment')\n\n @sanity_function\n def validate_test(self):\n return sn.assert_found(r'^8', self.stdout)\n\n@rfm.simple_test class osu_latency_test(OSUBenchmarkTestBase):\n descr = 'OSU latency test'\n\n @run_before('run')\n def prepare_run(self):\n self.executable = os.path.join(\n self.osu_binaries.stagedir,\n self.osu_binaries.build_prefix,\n 'mpi', 'pt2pt', 'osu_latency'\n )\n self.executable_opts = ['-x', '100', '-i', '1000']\n\n @performance_function('us')\n def latency(self):\n return sn.extractsingle(r'^8\\s+(\\S+)', self.stdout, 1, float)\n\n@rfm.simple_test class osu_bandwidth_test(OSUBenchmarkTestBase):\n descr = 'OSU bandwidth test'\n\n @run_before('run')\n def prepare_run(self):\n self.executable = os.path.join(\n self.osu_binaries.stagedir,\n self.osu_binaries.build_prefix,\n 'mpi', 'pt2pt', 'osu_bw'\n )\n self.executable_opts = ['-x', '100', '-i', '1000']\n\n @performance_function('MB/s')\n def bandwidth(self):\n return sn.extractsingle(r'^4194304\\s+(\\S+)',\n self.stdout, 1, float)\n\n@rfm.simple_test class osu_allreduce_test(OSUBenchmarkTestBase):\n mpi_tasks = parameter(1 << i for i in range(1, 5))\n descr = 'OSU Allreduce test'\n\n @run_before('run')\n def set_executable(self):\n self.num_tasks = self.mpi_tasks\n self.executable = os.path.join(\n self.osu_binaries.stagedir,\n self.osu_binaries.build_prefix,\n 'mpi', 'collective', 'osu_allreduce'\n )\n self.executable_opts = ['-m', '8', '-x', '1000', '-i', '20000']\n\n @performance_function('us')\n def latency(self):\n return sn.extractsingle(r'^8\\s+(\\S+)', self.stdout, 1, float)\n```\nLet’s start from the leaf tests, i.e. the tests that execute the benchmarks (`osu_latency_test`, `osu_bandwidth_test` and `osu_allreduce_test`).\nAs in the dependencies example, all these tests derive from the `OSUBenchmarkTestBase`, where we define a fixture that will take care of generating the binaries of the tests:\n```\n osu_binaries = fixture(build_osu_benchmarks, scope='environment')\n```\nA test defines a fixture using the `fixture()` builtin and assigns it a name by assigning the return value of the builtin to a test variable, here `osu_binaries`.\nThis name will be used later to access the resource managed by the fixture.\n\nAs stated previously, a fixture is another full-fledged ReFrame test, here the `build_osu_benchmarks` which will take care of building the OSU benchmarks.\nEach fixture is associated with a scope.\nThis practically indicates at which level a fixture is shared with other tests.\nThere are four fixture scopes, which are listed below in decreasing order of generality:\n\n* `session`: A fixture with this scope will be executed once per ReFrame run session and will be shared across the whole run.\n* `partition`: A fixture with this scope will be executed once per partition and will be shared across all tests that run in that partition.\n* `environment`: A fixture with this scope will be executed once per partition and environment combination and will be shared across all tests that run with this partition and environment combination.\n* `test`: A fixture with this scope is private to the test and will be executed for each test case.\n\nIn this example, we need to build once the OSU benchmarks for each partition and environment combination, so we use the `environment` scope.\n\nAccessing the fixture is very straightforward.\nThe fixture’s result is accessible after the *setup* pipeline stage through the corresponding variable in the test that is defining it.\nSince a fixture is a standard ReFrame test, you can access any information of the test.\nThe individual benchmarks do exactly that:\n```\n @run_before('run')\n def prepare_run(self):\n self.executable = os.path.join(\n self.osu_binaries.stagedir,\n self.osu_binaries.build_prefix,\n 'mpi', 'pt2pt', 'osu_latency'\n )\n self.executable_opts = ['-x', '100', '-i', '1000']\n```\nHere we construct the final executable path by accessing the standard [`stagedir`](index.html#reframe.core.pipeline.RegressionTest.stagedir) attribute of the test as well as the custom-defined `build_prefix` variable of the `build_osu_benchmarks` fixture.\n\nLet’s inspect now the `build_osu_benchmarks` fixture:\n```\nclass build_osu_benchmarks(rfm.CompileOnlyRegressionTest):\n descr = 'Build OSU benchmarks'\n build_system = 'Autotools'\n build_prefix = variable(str)\n osu_benchmarks = fixture(fetch_osu_benchmarks, scope='session')\n\n @run_before('compile')\n def prepare_build(self):\n tarball = f'osu-micro-benchmarks-{self.osu_benchmarks.version}.tar.gz'\n self.build_prefix = tarball[:-7] # remove .tar.gz extension\n\n fullpath = os.path.join(self.osu_benchmarks.stagedir, tarball)\n self.prebuild_cmds = [\n f'cp {fullpath} {self.stagedir}',\n f'tar xzf {tarball}',\n f'cd {self.build_prefix}'\n ]\n self.build_system.max_concurrency = 8\n\n @sanity_function\n def validate_build(self):\n # If compilation fails, the test would fail in any case, so nothing to\n # further validate here.\n return True\n```\nIt is obvious that it is a normal ReFrame test except that it does not need to be decorated with the [`@simple_test`](index.html#reframe.core.decorators.simple_test) decorator.\nThis means that the test will only be executed if it is a fixture of another test.\nIf it was decorated, it would be executed both as a standalone test and as a fixture of another test.\nAnother detail is that this test does not define the [`valid_systems`](index.html#reframe.core.pipeline.RegressionTest.valid_systems) and [`valid_prog_environs`](index.html#reframe.core.pipeline.RegressionTest.valid_prog_environs) variables.\nFixtures inherit those variables from the test that owns them depending on the scope.\n\nSimilarly to `OSUBenchmarkTestBase`, this test uses a fixture that fetches the OSU benchmarks sources.\nWe could fetch the OSU benchmarks in this test, but we choose to separate the two primarily for demonstration purposes, but it would also make sense in cases that the data fetch is too slow.\n\nThe `osu_benchmarks` fixture is defined at session scope, since we only need to download the benchmarks once for the whole session:\n```\n osu_benchmarks = fixture(fetch_osu_benchmarks, scope='session')\n```\nThe rest of the test is very straightforward.\n\nLet’s inspect the last fixture, the `fetch_osu_benchmarks`:\n```\nclass fetch_osu_benchmarks(rfm.RunOnlyRegressionTest):\n descr = 'Fetch OSU benchmarks'\n version = variable(str, value='5.6.2')\n executable = 'wget'\n executable_opts = [\n f'http://mvapich.cse.ohio-state.edu/download/mvapich/osu-micro-benchmarks-{version}.tar.gz' # noqa: E501\n ]\n local = True\n\n @sanity_function\n def validate_download(self):\n return sn.assert_eq(self.job.exitcode, 0)\n```\nThere is nothing special to this test – it is just an ordinary test – except that we force it to execute locally by setting its [`local`](index.html#reframe.core.pipeline.RegressionTest.local) variable.\nThe reason for that is that a fixture at session scope can execute with any partition/environment combination, so ReFrame could have to spawn a job in case it has chosen a remote partition to launch this fixture on.\nFor this reason, we simply force it to execute locally regardless of the chosen partition.\n\nIt is now time to run the new tests, but let us first list them:\n```\nreframe -c tutorials/fixtures/osu_benchmarks.py -l\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2\n command: './bin/reframe -c tutorials/fixtures/osu_benchmarks.py -l'\n launched by: user@host\n working directory: '/home/user/Devel/reframe'\n settings files: '<builtin>', '/home/user/Devel/reframe/tutorials/config/daint.py'\n check search path: '/home/user/Devel/reframe/tutorials/fixtures/osu_benchmarks.py'\n stage directory: '/home/user/Devel/reframe/stage'\n output directory: '/home/user/Devel/reframe/output'\n log files: '/tmp/rfm-to7wa4gh.log'\n\n[List of matched checks]\n- osu_allreduce_test %mpi_tasks=16 /1fe48834\n ^build_osu_benchmarks ~daint:gpu+gnu /f3269d42\n ^fetch_osu_benchmarks ~daint /79cd6023\n ^build_osu_benchmarks ~daint:gpu+intel /4d450880\n ^fetch_osu_benchmarks ~daint /79cd6023\n ^build_osu_benchmarks ~daint:gpu+nvidia /e9b8d152\n ^fetch_osu_benchmarks ~daint /79cd6023\n- osu_allreduce_test %mpi_tasks=8 /ae01c137\n ^build_osu_benchmarks ~daint:gpu+gnu /f3269d42\n ^fetch_osu_benchmarks ~daint /79cd6023\n ^build_osu_benchmarks ~daint:gpu+intel /4d450880\n ^fetch_osu_benchmarks ~daint /79cd6023\n ^build_osu_benchmarks ~daint:gpu+nvidia /e9b8d152\n ^fetch_osu_benchmarks ~daint /79cd6023\n- osu_allreduce_test %mpi_tasks=4 /2129dc34\n ^build_osu_benchmarks ~daint:gpu+gnu /f3269d42\n ^fetch_osu_benchmarks ~daint /79cd6023\n ^build_osu_benchmarks ~daint:gpu+intel /4d450880\n ^fetch_osu_benchmarks ~daint /79cd6023\n ^build_osu_benchmarks ~daint:gpu+nvidia /e9b8d152\n ^fetch_osu_benchmarks ~daint /79cd6023\n- osu_allreduce_test %mpi_tasks=2 /9f29c081\n ^build_osu_benchmarks ~daint:gpu+gnu /f3269d42\n ^fetch_osu_benchmarks ~daint /79cd6023\n ^build_osu_benchmarks ~daint:gpu+intel /4d450880\n ^fetch_osu_benchmarks ~daint /79cd6023\n ^build_osu_benchmarks ~daint:gpu+nvidia /e9b8d152\n ^fetch_osu_benchmarks ~daint /79cd6023\n- osu_bandwidth_test /026711a1\n ^build_osu_benchmarks ~daint:gpu+gnu /f3269d42\n ^fetch_osu_benchmarks ~daint /79cd6023\n ^build_osu_benchmarks ~daint:gpu+intel /4d450880\n ^fetch_osu_benchmarks ~daint /79cd6023\n ^build_osu_benchmarks ~daint:gpu+nvidia /e9b8d152\n ^fetch_osu_benchmarks ~daint /79cd6023\n- osu_latency_test /d2c978ad\n ^build_osu_benchmarks ~daint:gpu+gnu /f3269d42\n ^fetch_osu_benchmarks ~daint /79cd6023\n ^build_osu_benchmarks ~daint:gpu+intel /4d450880\n ^fetch_osu_benchmarks ~daint /79cd6023\n ^build_osu_benchmarks ~daint:gpu+nvidia /e9b8d152\n ^fetch_osu_benchmarks ~daint /79cd6023 Found 6 check(s)\n\nLog file(s) saved in '/tmp/rfm-to7wa4gh.log'\n```\nNotice how the `build_osu_benchmarks` fixture is populated three times, once for each partition and environment combination, and the `fetch_osu_benchmarks` is generated only once.\nThe following figure shows visually the conceptual dependencies of the `osu_bandwidth_test`.\n\nExpanded fixtures and dependencies for the OSU benchmarks example.[¶](#id1)\n\nA *scope* part is added to the base name of the fixture, which in this figure is indicated with red color.\n\nUnder the hood, fixtures use the test dependency mechanism which is described in [How Test Dependencies Work In ReFrame](index.html#document-dependencies).\nThe dependencies listed by default and shown in the previous figure are conceptual.\nDepending on the available partitions and environments, tests and fixtures can be concretized differently.\nFixtures in particular are also more flexible in the way they can be concretized depending on their scope.\nThe following listing and figure show the concretization of the `osu_bandwidth_test`:\n```\nreframe -c tutorials/fixtures/osu_benchmarks.py -n osu_bandwidth_test -lC\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2\n command: './bin/reframe -c tutorials/fixtures/osu_benchmarks.py -n osu_bandwidth_test -lC'\n launched by: user@host\n working directory: '/home/user/Devel/reframe'\n settings files: '<builtin>', '/home/user/Devel/reframe/tutorials/config/daint.py'\n check search path: '/home/user/Devel/reframe/tutorials/fixtures/osu_benchmarks.py'\n stage directory: '/home/user/Devel/reframe/stage'\n output directory: '/home/user/Devel/reframe/output'\n log files: '/tmp/rfm-m1w2t4eh.log'\n\n[List of matched checks]\n- osu_bandwidth_test /026711a1 @daint:gpu+gnu\n ^build_osu_benchmarks ~daint:gpu+gnu /f3269d42 @daint:gpu+gnu\n ^fetch_osu_benchmarks ~daint /79cd6023 @daint:gpu+gnu\n- osu_bandwidth_test /026711a1 @daint:gpu+intel\n ^build_osu_benchmarks ~daint:gpu+intel /4d450880 @daint:gpu+intel\n ^fetch_osu_benchmarks ~daint /79cd6023 @daint:gpu+gnu\n- osu_bandwidth_test /026711a1 @daint:gpu+nvidia\n ^build_osu_benchmarks ~daint:gpu+nvidia /e9b8d152 @daint:gpu+nvidia\n ^fetch_osu_benchmarks ~daint /79cd6023 @daint:gpu+gnu Concretized 7 test case(s)\n\nLog file(s) saved in '/tmp/rfm-m1w2t4eh.log'\n```\nThe actual dependencies for the OSU benchmarks example using fixtures.[¶](#id2)\n\nThe first thing to notice here is how the individual test cases of `osu_bandwidth_test` depend only the specific fixtures for their scope:\nwhen `osu_bandwidth_test` runs on the `daint:gpu` partition using the `gnu` compiler it will only depend on the `build_osu_benchmarks~daint:gpu+gnu` fixture.\nThe second thing to notice is where the `fetch_osu_benchmarks~daint` fixture will run.\nSince this is a *session* fixture, ReFrame has arbitrarily chosen to run it on `daint:gpu` using the `gnu` environment.\nA session fixture can run on any combination of valid partitions and environments.\nThe following listing and figure show how the test dependency DAG is concretized when we scope the valid programming environments from the command line using `-p nvidia`.\n```\nreframe -c tutorials/fixtures/osu_benchmarks.py -n osu_bandwidth_test -lC -p nvidia\n```\n```\n[ReFrame Setup]\n version: 3.10.0-dev.3+605af31a\n command: './bin/reframe -c tutorials/fixtures/osu_benchmarks.py -n osu_bandwidth_test -lC -p nvidia'\n launched by: user@host\n working directory: '/home/user/Devel/reframe'\n settings file: '/home/user/Devel/reframe/tutorials/config/settings.py'\n check search path: '/home/user/Devel/reframe/tutorials/fixtures/osu_benchmarks.py'\n stage directory: '/home/user/Devel/reframe/stage'\n output directory: '/home/user/Devel/reframe/output'\n\n[List of matched checks]\n- osu_bandwidth_test @daint:gpu+nvidia\n ^build_osu_benchmarks ~daint:gpu+nvidia @daint:gpu+nvidia\n ^fetch_osu_benchmarks ~daint @daint:gpu+nvidia Concretized 3 test case(s)\n\nLog file(s) saved in '/tmp/rfm-dnfdagj8.log'\n```\nThe dependency graph concretized for the ‘nvidia’ environment only.[¶](#id3)\n\nNotice how the `fetch_osu_benchmarks~daint` fixture is selected to run in the only valid partition/environment combination.\nThis is an important difference compared to the same example written using raw dependencies in [How Test Dependencies Work In ReFrame](index.html#document-dependencies), in which case in order not to have unresolved dependencies, we would need to specify the valid programming environment of the test that fetches the sources.\nFixtures do not need that, since you can impose less strict constraints by setting their scope accordingly.\n\nFinally, let’s run all the benchmarks at once:\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2\n command: './bin/reframe -c tutorials/fixtures/osu_benchmarks.py -r'\n launched by: user@host\n working directory: '/home/user/Devel/reframe'\n settings files: '<builtin>', '/home/user/Devel/reframe/tutorials/config/daint.py'\n check search path: '/home/user/Devel/reframe/tutorials/fixtures/osu_benchmarks.py'\n stage directory: '/home/user/Devel/reframe/stage'\n output directory: '/home/user/Devel/reframe/output'\n log files: '/tmp/rfm-63lwmv4b.log'\n\n[===] Running 10 check(s)\n[===] Started on Tue Nov 15 18:27:17 2022\n\n[---] start processing checks\n[ RUN ] fetch_osu_benchmarks ~daint /79cd6023 @daint:gpu+gnu\n[ OK ] ( 1/22) fetch_osu_benchmarks ~daint /79cd6023 @daint:gpu+gnu\n[ RUN ] build_osu_benchmarks ~daint:gpu+gnu /f3269d42 @daint:gpu+gnu\n[ RUN ] build_osu_benchmarks ~daint:gpu+intel /4d450880 @daint:gpu+intel\n[ RUN ] build_osu_benchmarks ~daint:gpu+nvidia /e9b8d152 @daint:gpu+nvidia\n[ OK ] ( 2/22) build_osu_benchmarks ~daint:gpu+gnu /f3269d42 @daint:gpu+gnu\n[ RUN ] osu_allreduce_test %mpi_tasks=16 /1fe48834 @daint:gpu+gnu\n[ RUN ] osu_allreduce_test %mpi_tasks=8 /ae01c137 @daint:gpu+gnu\n[ RUN ] osu_allreduce_test %mpi_tasks=4 /2129dc34 @daint:gpu+gnu\n[ RUN ] osu_allreduce_test %mpi_tasks=2 /9f29c081 @daint:gpu+gnu\n[ RUN ] osu_bandwidth_test /026711a1 @daint:gpu+gnu\n[ RUN ] osu_latency_test /d2c978ad @daint:gpu+gnu\n[ OK ] ( 3/22) build_osu_benchmarks ~daint:gpu+intel /4d450880 @daint:gpu+intel\n[ OK ] ( 4/22) build_osu_benchmarks ~daint:gpu+nvidia /e9b8d152 @daint:gpu+nvidia\n[ RUN ] osu_allreduce_test %mpi_tasks=16 /1fe48834 @daint:gpu+intel\n[ RUN ] osu_allreduce_test %mpi_tasks=16 /1fe48834 @daint:gpu+nvidia\n[ RUN ] osu_allreduce_test %mpi_tasks=8 /ae01c137 @daint:gpu+intel\n[ RUN ] osu_allreduce_test %mpi_tasks=8 /ae01c137 @daint:gpu+nvidia\n[ RUN ] osu_allreduce_test %mpi_tasks=4 /2129dc34 @daint:gpu+intel\n[ RUN ] osu_allreduce_test %mpi_tasks=4 /2129dc34 @daint:gpu+nvidia\n[ RUN ] osu_allreduce_test %mpi_tasks=2 /9f29c081 @daint:gpu+intel\n[ RUN ] osu_allreduce_test %mpi_tasks=2 /9f29c081 @daint:gpu+nvidia\n[ RUN ] osu_bandwidth_test /026711a1 @daint:gpu+intel\n[ RUN ] osu_bandwidth_test /026711a1 @daint:gpu+nvidia\n[ RUN ] osu_latency_test /d2c978ad @daint:gpu+intel\n[ RUN ] osu_latency_test /d2c978ad @daint:gpu+nvidia\n[ OK ] ( 5/22) osu_allreduce_test %mpi_tasks=2 /9f29c081 @daint:gpu+gnu P: latency: 2.76 us (r:0, l:None, u:None)\n[ OK ] ( 6/22) osu_allreduce_test %mpi_tasks=2 /9f29c081 @daint:gpu+intel P: latency: 1.68 us (r:0, l:None, u:None)\n[ OK ] ( 7/22) osu_allreduce_test %mpi_tasks=4 /2129dc34 @daint:gpu+intel P: latency: 4.89 us (r:0, l:None, u:None)\n[ OK ] ( 8/22) osu_latency_test /d2c978ad @daint:gpu+intel P: latency: 1.54 us (r:0, l:None, u:None)\n[ OK ] ( 9/22) osu_latency_test /d2c978ad @daint:gpu+gnu P: latency: 1.17 us (r:0, l:None, u:None)\n[ OK ] (10/22) osu_allreduce_test %mpi_tasks=4 /2129dc34 @daint:gpu+gnu P: latency: 3.22 us (r:0, l:None, u:None)\n[ OK ] (11/22) osu_allreduce_test %mpi_tasks=16 /1fe48834 @daint:gpu+gnu P: latency: 13.84 us (r:0, l:None, u:None)\n[ OK ] (12/22) osu_allreduce_test %mpi_tasks=16 /1fe48834 @daint:gpu+nvidia P: latency: 30.77 us (r:0, l:None, u:None)\n[ OK ] (13/22) osu_allreduce_test %mpi_tasks=4 /2129dc34 @daint:gpu+nvidia P: latency: 5.74 us (r:0, l:None, u:None)\n[ OK ] (14/22) osu_allreduce_test %mpi_tasks=16 /1fe48834 @daint:gpu+intel P: latency: 14.77 us (r:0, l:None, u:None)\n[ OK ] (15/22) osu_allreduce_test %mpi_tasks=2 /9f29c081 @daint:gpu+nvidia P: latency: 4.5 us (r:0, l:None, u:None)\n[ OK ] (16/22) osu_allreduce_test %mpi_tasks=8 /ae01c137 @daint:gpu+nvidia P: latency: 33.93 us (r:0, l:None, u:None)\n[ OK ] (17/22) osu_allreduce_test %mpi_tasks=8 /ae01c137 @daint:gpu+intel P: latency: 20.9 us (r:0, l:None, u:None)\n[ OK ] (18/22) osu_latency_test /d2c978ad @daint:gpu+nvidia P: latency: 1.18 us (r:0, l:None, u:None)\n[ OK ] (19/22) osu_allreduce_test %mpi_tasks=8 /ae01c137 @daint:gpu+gnu P: latency: 10.14 us (r:0, l:None, u:None)\n[ OK ] (20/22) osu_bandwidth_test /026711a1 @daint:gpu+gnu P: bandwidth: 9785.43 MB/s (r:0, l:None, u:None)\n[ OK ] (21/22) osu_bandwidth_test /026711a1 @daint:gpu+intel P: bandwidth: 9841.26 MB/s (r:0, l:None, u:None)\n[ OK ] (22/22) osu_bandwidth_test /026711a1 @daint:gpu+nvidia P: bandwidth: 9824.01 MB/s (r:0, l:None, u:None)\n[---] all spawned checks have finished\n\n[ PASSED ] Ran 22/22 test case(s) from 10 check(s) (0 failure(s), 0 skipped)\n[===] Finished on Tue Nov 15 18:30:34 2022 Run report saved in '/home/user/.reframe/reports/run-report-4.json'\nLog file(s) saved in '/tmp/rfm-63lwmv4b.log'\n```\nTip\n\nA reasonable question is how to choose between fixtures and dependencies?\n\nThe rule of thumb is use fixtures if your test needs to use any resource of the target test and use dependencies if you simply want to impose an order of execution for your tests.\n\n#### Tutorial 5: Using Build Automation Tools As a Build System[¶](#tutorial-5-using-build-automation-tools-as-a-build-system)\n\nIn this tutorial we will present how to use [Easybuild](https://easybuild.io/) and [Spack](https://spack.io/) as a build system for a ReFrame test.\nThe example uses the configuration file presented in [Tutorial 1: Getting Started with ReFrame](index.html#document-tutorial_basics), which you can find in `tutorials/config/settings.py`.\nWe also assume that the reader is already familiar with the concepts presented in the basic tutorial and has a working knowledge of EasyBuild and Spack.\n\n##### Using EasyBuild to Build the Test Code[¶](#using-easybuild-to-build-the-test-code)\n\nNew in version 3.5.0.\n\nLet’s consider a simple ReFrame test that installs `bzip2-1.0.6` given the easyconfig [bzip2-1.0.6.eb](https://github.com/eth-cscs/production/blob/master/easybuild/easyconfigs/b/bzip2/bzip2-1.0.6.eb) and checks that the installed version is correct.\nThe following code block shows the check, highlighting the lines specific to this tutorial:\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class BZip2EBCheck(rfm.RegressionTest):\n descr = 'Demo test using EasyBuild to build the test code'\n valid_systems = ['*']\n valid_prog_environs = ['builtin']\n executable = 'bzip2'\n executable_opts = ['--help']\n build_system = 'EasyBuild'\n\n @run_before('compile')\n def setup_build_system(self):\n self.build_system.easyconfigs = ['bzip2-1.0.6.eb']\n self.build_system.options = ['-f']\n\n @run_before('run')\n def prepare_run(self):\n self.modules = self.build_system.generated_modules\n\n @sanity_function\n def assert_version(self):\n return sn.assert_found(r'Version 1.0.6', self.stderr)\n```\nThe test looks pretty standard except for the highlighted blocks.\nLet’s have a look first to the block in the `BZip2Check` class.\n\nThe first thing is to specify that the EasyBuild build system will be used.\nThis is done by setting [`build_system`](index.html#reframe.core.pipeline.RegressionTest.build_system) to `'EasyBuild'`.\nThen, the software to be installed is passed as a list to [`easyconfigs`](index.html#reframe.core.buildsystems.EasyBuild.easyconfigs).\nHere only one easyconfig is given, but more than one can be passed.\nFinally, through [`options`](index.html#reframe.core.buildsystems.EasyBuild.options), command line options can be passed to the `eb` executable.\nIn this test we pass `-f` to make sure that `bzip2` will be built even if the module already exists externally.\n\nFor this test, ReFrame generates the following command to build and install the easyconfig:\n```\nexport EASYBUILD_BUILDPATH={stagedir}/easybuild/build export EASYBUILD_INSTALLPATH={stagedir}/easybuild export EASYBUILD_PREFIX={stagedir}/easybuild export EASYBUILD_SOURCEPATH={stagedir}/easybuild eb bzip2-1.0.6.eb -f\n```\nReFrame will keep all the files generated by EasyBuild (sources, temporary files, installed software and the corresponding modules) under the test’s stage directory.\nFor this reason it sets the relevant EasyBuild environment variables.\n\nTip\n\nUsers may set the EasyBuild prefix to a different location by setting the [`prefix`](index.html#reframe.core.buildsystems.EasyBuild.prefix) attribute of the build system.\nThis allows you to have the built software installed upon successful completion of the build phase, but if the test fails in a later stage (sanity, performance), the installed software will not be cleaned up automatically.\n\nNote\n\nReFrame assumes that the `eb` executable is available on the system where the compilation is run (typically the local host where ReFrame is executed).\n\nNow that we know everything related to building and installing the code, we can move to the part dealing with running it.\nTo run the code, the generated modules need to be loaded in order to make the software available.\nThe modules can be accessed through `generated_modules`, however, they are available only after EasyBuild completes the installation.\nThis means that [`modules`](index.html#reframe.core.pipeline.RegressionTest.modules) can be set only after the build phase finishes.\nFor that, we can set [`modules`](index.html#reframe.core.pipeline.RegressionTest.modules) in a class method wrapped by the `run_before()` built-in, specifying the `run` phase.\nThis test will then run the following commands:\n```\nmodule load bzip/1.0.6 bzip2 --help\n```\n###### Packaging the installation[¶](#packaging-the-installation)\n\nThe EasyBuild build system offers a way of packaging the installation via EasyBuild’s packaging support.\nTo use this feature, [the FPM package manager](https://fpm.readthedocs.io/en/latest/) must be available.\nBy setting the dictionary `package_opts` in the test, ReFrame will pass `--package-{key}={val}` to the EasyBuild invocation.\nFor instance, the following can be set to package the installations as an rpm file:\n```\nself.keep_files = ['easybuild/packages']\nself.build_system.package_opts = {\n 'type': 'rpm',\n}\n```\nThe packages are generated by EasyBuild in the stage directory.\nTo retain them after the test succeeds, [`keep_files`](index.html#reframe.core.pipeline.RegressionTest.keep_files) needs to be set.\n\n##### Using Spack to Build the Test Code[¶](#using-spack-to-build-the-test-code)\n\nNew in version 3.6.1.\n\nThis example is the equivalent to the previous one, except that it uses Spack to build `bzip2`.\nHere is the test’s code:\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class BZip2SpackCheck(rfm.RegressionTest):\n descr = 'Demo test using Spack to build the test code'\n valid_systems = ['*']\n valid_prog_environs = ['builtin']\n executable = 'bzip2'\n executable_opts = ['--help']\n build_system = 'Spack'\n\n @run_before('compile')\n def setup_build_system(self):\n self.build_system.specs = ['bzip2@1.0.6']\n\n @sanity_function\n def assert_version(self):\n return sn.assert_found(r'Version 1.0.6', self.stderr)\n```\nWhen [`build_system`](index.html#reframe.core.pipeline.RegressionTest.build_system) is set to `'Spack'`, ReFrame will leverage Spack environments in order to build the test code.\nBy default, ReFrame will create a new Spack environment in the test’s stage directory and add the requested [`specs`](index.html#reframe.core.buildsystems.Spack.specs) to it.\n\nNote\n\nOptional spec attributes, such as `target` and `os`, should be specified in [`specs`](index.html#reframe.core.buildsystems.Spack.specs) and not as install options in [`install_opts`](index.html#reframe.core.buildsystems.Spack.install_opts).\n\nUsers may also specify an existing Spack environment by setting the [`environment`](index.html#reframe.core.buildsystems.Spack.environment) attribute.\nIn this case, ReFrame treats the environment as a *test resource* so it expects to find it under the test’s [`sourcesdir`](index.html#reframe.core.pipeline.RegressionTest.sourcesdir), which defaults to `'src'`.\n\nAs with every other test, ReFrame will copy the test’s resources to its stage directory before building it.\nReFrame will then activate the generated environment (either the one provided by the user or the one generated by ReFrame), add the given specs using the `spack add` command and, finally, install the packages in the environment.\nHere is what ReFrame generates as a build script for this example:\n```\nspack env create -d rfm_spack_env spack -e rfm_spack_env config add \"config:install_tree:root:opt/spack\"\nspack -e rfm_spack_env add bzip2@1.0.6 spack -e rfm_spack_env install\n```\nAs you might have noticed ReFrame expects that Spack is already installed on the system.\nThe packages specified in the environment and the tests will be installed in the test’s stage directory, where the environment is copied before building.\nHere is the stage directory structure:\n```\nstage/generic/default/builtin/BZip2SpackCheck/\n├── rfm_spack_env\n│ ├── spack\n│ │ └── opt\n│ │ └── spack\n│ │ ├── bin\n│ │ └── darwin-catalina-skylake\n│ ├── spack.lock\n│ └── spack.yaml\n├── rfm_BZip2SpackCheck_build.err\n├── rfm_BZip2SpackCheck_build.out\n├── rfm_BZip2SpackCheck_build.sh\n├── rfm_BZip2SpackCheck_job.err\n├── rfm_BZip2SpackCheck_job.out\n└── rfm_BZip2SpackCheck_job.sh\n```\nFinally, here is the generated run script that ReFrame uses to run the test, once its build has succeeded:\n```\n#!/bin/bash spack env create -d rfm_spack_env eval `spack -e rfm_spack_env load --sh bzip2@1.0.6`\nbzip2 --help\n```\nFrom this point on, sanity and performance checking are exactly identical to any other ReFrame test.\n\nTip\n\nWhile developing a test using Spack or EasyBuild as a build system, it can be useful to run ReFrame with the [`--keep-stage-files`](index.html#cmdoption-keep-stage-files) and [`--dont-restage`](index.html#cmdoption-dont-restage) options to prevent ReFrame from removing the test’s stage directory upon successful completion of the test.\nFor this particular type of test, these options will avoid having to rebuild the required package dependencies every time the test is retried.\n\n#### Tutorial 6: Tips and Tricks[¶](#tutorial-6-tips-and-tricks)\n\nNew in version 3.4.\n\nThis tutorial focuses on some less known aspects of ReFrame’s command line interface that can be helpful.\n\n##### Debugging[¶](#debugging)\n\nReFrame tests are Python classes inside Python source files, so the usual debugging techniques for Python apply, but the ReFrame frontend will filter some errors and stack traces by default in order to keep the output clean.\nGenerally, ReFrame will not print the full stack trace for user programming errors and will not block the test loading process.\nIf a test has errors and cannot be loaded, an error message will be printed and the loading of the remaining tests will continue.\nIn the following, we have inserted a small typo in the `hello2.py` tutorial example:\n```\n./bin/reframe -c tutorials/basics/hello -R -l\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe -c tutorials/basics/hello -R -l'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>', '/home/user/Repositories/reframe/tutorials/config/tresa.py'\n check search path: (R) '/home/user/Repositories/reframe/tutorials/basics/hello'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/<KEY>/rfm-ldo5um3v.log'\n\nWARNING: skipping test file '/home/user/Repositories/reframe/tutorials/basics/hello/hello2.py': name error: tutorials/basics/hello/hello2.py:13: name 'paramter' is not defined\n lang = paramter(['c', 'cpp'])\n (rerun with '-v' for more information)\n[List of matched checks]\n- HelloTest /2b3e4546 Found 1 check(s)\n\nLog file(s) saved in '/<KEY>0000gp/T/rfm-ldo5um3v.log'\n```\nNotice how ReFrame prints also the source code line that caused the error.\nThis is not always the case, however.\nReFrame cannot always track a user error back to its source and this is particularly true for the ReFrame-specific syntactic elements, such as the class [builtins](regression_test_api.html#builtins).\nIn such cases, ReFrame will just print the error message but not the source code context.\nIn the following example, we introduce a typo in the argument of the `@run_before` decorator:\n```\n./bin/reframe: skipping test file '/Users/user/Repositories/reframe/tutorials/basics/hello/hello2.py': reframe syntax error: invalid pipeline stage specified: 'compil' (rerun with '-v' for more information)\n[List of matched checks]\n- HelloTest (found in '/Users/user/Repositories/reframe/tutorials/basics/hello/hello1.py')\nFound 1 check(s)\n```\nAs suggested by the warning message, passing [`-v`](index.html#cmdoption-4) will give you the stack trace for each of the failing tests, as well as some more information about what is going on during the loading.\n```\n./bin/reframe -c tutorials/basics/hello -R -l -v\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe -c tutorials/basics/hello -R -l -v'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>', '/home/user/Repositories/reframe/tutorials/config/tresa.py'\n check search path: (R) '/home/user/Repositories/reframe/tutorials/basics/hello'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/var/folders/h7/k7cgrdl13r996m4dmsvjq7v80000gp/T/rfm-xs3l6jud.log'\n\nWARNING: skipping test file '/home/user/Repositories/reframe/tutorials/basics/hello/hello2.py': name error: tutorials/basics/hello/hello2.py:13: name 'paramter' is not defined\n lang = paramter(['c', 'cpp'])\n (rerun with '-v' for more information)\nTraceback (most recent call last):\n File \"/home/user/Repositories/reframe/reframe/frontend/loader.py\", line 205, in load_from_file\n util.import_module_from_file(filename, force)\n File \"/home/user/Repositories/reframe/reframe/utility/__init__.py\", line 109, in import_module_from_file\n return importlib.import_module(module_name)\n File \"/usr/local/Cellar/python@3.10/3.10.7/Frameworks/Python.framework/Versions/3.10/lib/python3.10/importlib/__init__.py\", line 126, in import_module\n return _bootstrap._gcd_import(name[level:], package, level)\n File \"<frozen importlib._bootstrap>\", line 1050, in _gcd_import\n File \"<frozen importlib._bootstrap>\", line 1027, in _find_and_load\n File \"<frozen importlib._bootstrap>\", line 1006, in _find_and_load_unlocked\n File \"<frozen importlib._bootstrap>\", line 688, in _load_unlocked\n File \"<frozen importlib._bootstrap_external>\", line 883, in exec_module\n File \"<frozen importlib._bootstrap>\", line 241, in _call_with_frames_removed\n File \"/home/user/Repositories/reframe/tutorials/basics/hello/hello2.py\", line 12, in <module>\n class HelloMultiLangTest(rfm.RegressionTest):\n File \"/home/user/Repositories/reframe/tutorials/basics/hello/hello2.py\", line 13, in HelloMultiLangTest\n lang = paramter(['c', 'cpp'])\nNameError: name 'paramter' is not defined\n\nLoaded 1 test(s)\nGenerated 2 test case(s)\nFiltering test cases(s) by name: 2 remaining Filtering test cases(s) by tags: 2 remaining Filtering test cases(s) by other attributes: 2 remaining Final number of test cases: 2\n[List of matched checks]\n- HelloTest /2b3e4546 Found 1 check(s)\n\nLog file(s) saved in '/<KEY>rf<KEY>.log'\n```\nTip\n\nThe [`-v`](index.html#cmdoption-4) option can be given multiple times to increase the verbosity level further.\n\n###### Debugging deferred expressions[¶](#debugging-deferred-expressions)\n\nAlthough deferred expressions that are used in sanity and performance functions behave similarly to normal Python expressions, you need to understand their [implicit evaluation rules](deferrable_functions_reference.html#implicit-evaluation-of-deferrable-functions).\nOne of the rules is that `str()` triggers the implicit evaluation, so trying to use the standard [`print()`](https://docs.python.org/3/library/functions.html#print) function with a deferred expression, you might get unexpected results if that expression is not yet to be evaluated.\nFor this reason, ReFrame offers a sanity function counterpart of [`print()`](https://docs.python.org/3/library/functions.html#print), which allows you to safely print deferred expressions.\n\nLet’s see that in practice, by printing the filename of the standard output for `HelloMultiLangTest` test.\nThe [`stdout`](index.html#reframe.core.pipeline.RegressionTest.stdout) is a deferred expression and it will get its value later on while the test executes.\nTrying to use the standard print here [`print()`](https://docs.python.org/3/library/functions.html#print) function here would be of little help, since it would simply give us [`None`](https://docs.python.org/3/library/constants.html#None), which is the value of `stdout` when the test is created.\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class HelloMultiLangTest(rfm.RegressionTest):\n lang = parameter(['c', 'cpp'])\n valid_systems = ['*']\n valid_prog_environs = ['*']\n\n @run_after('compile')\n def set_sourcepath(self):\n self.sourcepath = f'hello.{self.lang}'\n\n @sanity_function\n def validate_output(self):\n return sn.assert_found(r'Hello, World\\!', sn.print(self.stdout))\n```\nIf we run the test, we can see that the correct standard output filename will be printed after sanity:\n```\n./bin/reframe -C tutorials/config/settings.py -c tutorials/basics/hello/hello2.py -r\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe -C tutorials/config/tresa.py -c tutorials/basics/hello/hello2.py -r'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>', 'tutorials/config/tresa.py'\n check search path: '/home/user/Repositories/reframe/tutorials/basics/hello/hello2.py'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/<KEY>'\n\n[===] Running 2 check(s)\n[===] Started on Sat Nov 12 19:00:58 2022 \n\n[---] start processing checks\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @tresa:default+gnu\n[ RUN ] HelloMultiLangTest %lang=cpp /71bf65a3 @tresa:default+clang\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @tresa:default+gnu\n[ RUN ] HelloMultiLangTest %lang=c /7cfa870e @tresa:default+clang rfm_job.out rfm_job.out\n[ OK ] (1/4) HelloMultiLangTest %lang=c /7cfa870e @tresa:default+gnu rfm_job.out rfm_job.out\n[ OK ] (2/4) HelloMultiLangTest %lang=c /7cfa870e @tresa:default+clang rfm_job.out rfm_job.out\n[ OK ] (3/4) HelloMultiLangTest %lang=cpp /71bf65a3 @tresa:default+gnu rfm_job.out rfm_job.out\n[ OK ] (4/4) HelloMultiLangTest %lang=cpp /71bf65a3 @tresa:default+clang\n[---] all spawned checks have finished\n\n[ PASSED ] Ran 4/4 test case(s) from 2 check(s) (0 failure(s), 0 skipped)\n[===] Finished on Sat Nov 12 19:01:00 2022 \nRun report saved in '/home/user/.reframe/reports/run-report-325.json'\nLog file(s) saved in '/<KEY>'\n```\n###### Debugging sanity and performance patterns[¶](#debugging-sanity-and-performance-patterns)\n\nWhen creating a new test that requires a complex output parsing for either the `sanity` or `performance` pipeline stages, tuning the functions decorated by `@sanity_function` or `@performance_function` may involve some trial and error to debug the complex regular expressions required.\nFor lightweight tests which execute in a few seconds, this trial and error may not be an issue at all.\nHowever, when dealing with tests which take longer to run, this method can quickly become tedious and inefficient.\n\nTip\n\nWhen dealing with `make`-based projects which take a long time to compile, you can use the command line option [`--dont-restage`](index.html#cmdoption-dont-restage) in order to speed up the compile stage in subsequent runs.\n\nWhen a test fails, ReFrame will keep the test output in the stage directory after its execution, which means that one can load this output into a Python shell or another helper script without having to rerun the expensive test again.\nIf the test is not failing but the user still wants to experiment or modify the existing sanity or performance functions, the command line option [`--keep-stage-files`](index.html#cmdoption-keep-stage-files) can be used when running ReFrame to avoid deleting the stage directory.\nWith the executable’s output available in the stage directory, one can simply use the [re](https://docs.python.org/3/library/re.html) module to debug regular expressions as shown below.\n```\n>>> import re\n\n>>> # Read the test's output\n>>> with open(the_output_file, 'r') as f:\n... test_output = ''.join(f.readlines())\n...\n>>> # Evaluate the regular expression\n>>> re.find(the_regex_pattern, test_output)\n```\nAlternatively to using the [re](https://docs.python.org/3/library/re.html) module, one could use all the [`sanity`](index.html#module-reframe.utility.sanity) utility provided by ReFrame directly from the Python shell.\nIn order to do so, if ReFrame was installed manually using the `bootstrap.sh` script, one will have to make all the Python modules from the `external` directory accessible to the Python shell as shown below.\n```\n>>> import sys\n>>> import os\n\n>>> # Make the external modules available\n>>> sys.path = [os.path.abspath('external')] + sys.path\n\n>>> # Import ReFrame-provided sanity functions\n>>> import reframe.utility.sanity as sn\n\n>>> # Evaluate the regular expression\n>>> assert sn.evaluate(sn.assert_found(the_regex_pattern, the_output_file))\n```\n###### Debugging test loading[¶](#debugging-test-loading)\n\nIf you are new to ReFrame, you might wonder sometimes why your tests are not loading or why your tests are not running on the partition they were supposed to run.\nThis can be due to ReFrame picking the wrong configuration entry or that your test is not written properly (not decorated, no [`valid_systems`](index.html#reframe.core.pipeline.RegressionTest.valid_systems) etc.).\nIf you try to load a test file and list its tests by increasing twice the verbosity level, you will get enough output to help you debug such issues.\nLet’s try loading the `tutorials/basics/hello/hello2.py` file:\n```\n./bin/reframe -C tutorials/config/settings.py -c tutorials/basics/hello/hello2.py -l -vv\n```\n```\nLoading user configuration Loading the generic configuration Loading configuration file: ('tutorials/config/tresa.py',)\nDetecting system using method: 'hostname'\nUsing standard hostname...\nRetrieved hostname: 'host'\nLooking for a matching configuration entry Configuration found: picking system 'tresa'\nInitializing runtime Initializing system partition 'default'\nInitializing system 'tresa'\nInitializing modules system 'nomod'\ndetecting topology info for tresa:default\n> found topology file '/home/user/.reframe/topology/tresa-default/processor.json'; loading...\n> device auto-detection is not supported\n[ReFrame Environment]\n RFM_AUTODETECT_FQDN=<not set>\n RFM_AUTODETECT_METHOD=<not set>\n RFM_AUTODETECT_XTHOSTNAME=<not set>\n RFM_CHECK_SEARCH_PATH=<not set>\n RFM_CHECK_SEARCH_RECURSIVE=<not set>\n RFM_CLEAN_STAGEDIR=<not set>\n RFM_COLORIZE=n\n RFM_COMPRESS_REPORT=<not set>\n RFM_CONFIG_FILES=/home/user/Repositories/reframe/tutorials/config/tresa.py\n RFM_CONFIG_PATH=<not set>\n RFM_DUMP_PIPELINE_PROGRESS=<not set>\n RFM_GIT_TIMEOUT=<not set>\n RFM_HTTPJSON_URL=<not set>\n RFM_IGNORE_CHECK_CONFLICTS=<not set>\n RFM_IGNORE_REQNODENOTAVAIL=<not set>\n RFM_INSTALL_PREFIX=/home/user/Repositories/reframe\n RFM_KEEP_STAGE_FILES=<not set>\n RFM_MODULE_MAPPINGS=<not set>\n RFM_MODULE_MAP_FILE=<not set>\n RFM_NON_DEFAULT_CRAYPE=<not set>\n RFM_OUTPUT_DIR=<not set>\n RFM_PERFLOG_DIR=<not set>\n RFM_PIPELINE_TIMEOUT=<not set>\n RFM_PREFIX=<not set>\n RFM_PURGE_ENVIRONMENT=<not set>\n RFM_REMOTE_DETECT=<not set>\n RFM_REMOTE_WORKDIR=<not set>\n RFM_REPORT_FILE=<not set>\n RFM_REPORT_JUNIT=<not set>\n RFM_RESOLVE_MODULE_CONFLICTS=<not set>\n RFM_SAVE_LOG_FILES=<not set>\n RFM_STAGE_DIR=<not set>\n RFM_SYSLOG_ADDRESS=<not set>\n RFM_SYSTEM=<not set>\n RFM_TIMESTAMP_DIRS=<not set>\n RFM_TRAP_JOB_ERRORS=<not set>\n RFM_UNLOAD_MODULES=<not set>\n RFM_USER_MODULES=<not set>\n RFM_USE_LOGIN_SHELL=<not set>\n RFM_VERBOSE=<not set>\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe -C tutorials/config/tresa.py -c tutorials/basics/hello/hello2.py -l -vv'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>', 'tutorials/config/tresa.py'\n check search path: '/home/user/Repositories/reframe/tutorials/basics/hello/hello2.py'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/var/folders/<KEY>96m4dmsvjq7v80000gp/T/rfm-3gcehyof.log'\n\nLooking for tests in '/home/user/Repositories/reframe/tutorials/basics/hello/hello2.py'\nValidating '/home/user/Repositories/reframe/tutorials/basics/hello/hello2.py': OK\n > Loaded 2 test(s)\nLoaded 2 test(s)\nGenerated 4 test case(s)\nFiltering test cases(s) by name: 4 remaining Filtering test cases(s) by tags: 4 remaining Filtering test cases(s) by other attributes: 4 remaining Building and validating the full test DAG Full test DAG:\n ('HelloMultiLangTest_1', 'tresa:default', 'gnu') -> []\n ('HelloMultiLangTest_1', 'tresa:default', 'clang') -> []\n ('HelloMultiLangTest_0', 'tresa:default', 'gnu') -> []\n ('HelloMultiLangTest_0', 'tresa:default', 'clang') -> []\nFinal number of test cases: 4\n[List of matched checks]\n- HelloMultiLangTest %lang=cpp /71bf65a3\n- HelloMultiLangTest %lang=c /7cfa870e Found 2 check(s)\n\nLog file(s) saved in <KEY>'\n>>> profiler report [start] <<<\nmain: 0.053832 s\n test processing: 0.012268 s\n RegressionCheckLoader.load_all: 0.008720 s\n TestRegistry.instantiate_all: 0.003012 s\n generate_testcases: 0.000049 s\n main.<locals>._sort_testcases: 0.000012 s\n build_deps: 0.000072 s\n validate_deps: 0.000061 s\n toposort: 0.000091 s\n list_checks: 0.001080 s\n>>> profiler report [ end ] <<<\n```\nYou can see all the different phases ReFrame’s frontend goes through when loading a test.\nAfter loading the configuration, ReFrame will print out its relevant environment variables and will start examining the given files in order to find and load ReFrame tests.\nBefore attempting to load a file, it will validate it and check if it looks like a ReFrame test.\nIf it does, it will load that file by importing it.\nThis is where any ReFrame tests are instantiated and initialized (see `Loaded 2 test(s)`), as well as the actual test cases (combination of tests, system partitions and environments) are generated.\nThen the test cases are filtered based on the various [filtering command line options](manpage.html#test-filtering) as well as the programming environments that are defined for the currently selected system.\nFinally, the test case dependency graph is built and everything is ready for running (or listing).\n\nTry passing a specific system or partition with the [`--system`](index.html#cmdoption-system) option or modify the test (e.g., removing the decorator that registers it) and see how the logs change.\n\n##### Execution modes[¶](#execution-modes)\n\nReFrame allows you to create pre-defined ways of running it, which you can invoke from the command line.\nThese are called *execution modes* and are essentially named groups of command line options that will be passed to ReFrame whenever you request them.\nThese are defined in the configuration file and can be requested with the [`--mode`](index.html#cmdoption-mode) command-line option.\nThe following configuration defines an execution mode named `maintenance` and sets up ReFrame in a certain way (selects tests to run, sets up stage and output paths etc.)\n```\n 'modes': [\n {\n 'name': 'maintenance',\n 'options': [\n '--unload-module=reframe',\n '--exec-policy=async',\n '-S strict_check=1',\n '--output=/path/to/$USER/regression/maintenance',\n '--perflogdir=/path/to/$USER/regression/maintenance/logs',\n '--stage=$SCRATCH/regression/maintenance/stage',\n '--report-file=/path/to/$USER/regression/maintenance/reports/maint_report_{sessionid}.json',\n '-Jreservation=maintenance',\n '--save-log-files',\n '--tag=maintenance',\n '--timestamp=%F_%H-%M-%S'\n ]\n },\n]\n```\nThe execution modes come handy in situations that you have a standardized way of running ReFrame and you don’t want to create and maintain shell scripts around it.\nIn this example, you can simply run ReFrame with\n```\n./bin/reframe --mode=maintenance -r\n```\nand it will be equivalent to passing explicitly all the above options.\nYou can still pass any additional command line option and it will supersede or be combined (depending on the behaviour of the option) with those defined in the execution mode.\nIn this particular example, we could change just the reservation name by running\n```\n./bin/reframe --mode=maintenance -J reservation=maint -r\n```\nThere are two options that you can’t use inside execution modes and these are the [`-C`](index.html#cmdoption-C) and [`--system`](index.html#cmdoption-system).\nThe reason is that these option select the configuration file and the configuration entry to load.\n\n##### Manipulating ReFrame’s environment[¶](#manipulating-reframe-s-environment)\n\nReFrame runs the selected tests in the same environment as the one that it executes.\nIt does not unload any environment modules nor sets or unsets any environment variable.\nNonetheless, it gives you the opportunity to modify the environment that the tests execute.\nYou can either purge completely all environment modules by passing the [`--purge-env`](index.html#cmdoption-purge-env) option or ask ReFrame to load or unload some environment modules before starting running any tests by using the [`-m`](index.html#cmdoption-3) and [`-u`](index.html#cmdoption-u) options respectively.\nOf course you could manage the environment manually, but it’s more convenient if you do that directly through ReFrame’s command-line.\nIf you used an environment module to load ReFrame, e.g., `reframe`, you can use the [`-u`](index.html#cmdoption-u) to have ReFrame unload it before running any tests, so that the tests start in a clean environment:\n```\n./bin/reframe -u reframe [...]\n```\n##### Environment Modules Mappings[¶](#environment-modules-mappings)\n\nReFrame allows you to replace environment modules used in tests with other modules on the fly.\nThis is quite useful if you want to test a new version of a module or another combination of modules.\nAssume you have a test that loads a `gromacs` module:\n```\nclass GromacsTest(rfm.RunOnlyRegressionTest):\n ...\n modules = ['gromacs']\n```\nThis test would use the default version of the module in the system, but you might want to test another version, before making that new one the default.\nYou can ask ReFrame to temporarily replace the `gromacs` module with another one as follows:\n```\n./bin/reframe -n GromacsTest -M 'gromacs:gromacs/2020.5' -r\n```\nEvery time ReFrame tries to load the `gromacs` module, it will replace it with `gromacs/2020.5`.\nYou can specify multiple mappings at once or provide a file with mappings using the [`--module-mappings`](index.html#cmdoption-module-mappings) option.\nYou can also replace a single module with multiple modules.\n\nA very convenient feature of ReFrame in dealing with modules is that you do not have to care about module conflicts at all, regardless of the modules system backend.\nReFrame will take care of unloading any conflicting modules, if the underlying modules system cannot do that automatically.\nIn case of module mappings, it will also respect the module order of the replacement modules and will produce the correct series of “load” and “unload” commands needed by the modules system backend used.\n\n##### Retrying and Rerunning Tests[¶](#retrying-and-rerunning-tests)\n\nIf you are running ReFrame regularly as part of a continuous testing procedure you might not want it to generate alerts for transient failures.\nIf a ReFrame test fails, you might want to retry a couple of times before marking it as a failure.\nYou can achieve this with the [`--max-retries`](index.html#cmdoption-max-retries).\nReFrame will then retry the failing test cases a maximum number of times before reporting them as actual failures.\nThe failed test cases will not be retried immediately after they have failed, but rather at the end of the run session.\nThis is done to give more chances of success in case the failures have been transient.\n\nAnother interesting feature introduced in ReFrame 3.4 is the ability to restore a previous test session.\nWhenever it runs, ReFrame stores a detailed JSON report of the last run under `$HOME/.reframe` (see [`--report-file`](index.html#cmdoption-report-file)).\nUsing that file, ReFrame can restore a previous run session using the [`--restore-session`](index.html#cmdoption-restore-session).\nThis option is useful when you combine it with the various test filtering options.\nFor example, you might want to rerun only the failed tests or just a specific test in a dependency chain.\nLet’s see an artificial example that uses the following test dependency graph.\n\nComplex test dependency graph. Nodes in red are set to fail.[¶](#id2)\n\nTests `T2` and `T8` are set to fail.\nLet’s run the whole test DAG:\n```\n./bin/reframe -c unittests/resources/checks_unlisted/deps_complex.py -r\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe -c unittests/resources/checks_unlisted/deps_complex.py -r'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>', '/home/user/Repositories/reframe/tutorials/config/tresa.py'\n check search path: '/home/user/Repositories/reframe/unittests/resources/checks_unlisted/deps_complex.py'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/<KEY>08n_el.log'\n\n[===] Running 10 check(s)\n[===] Started on Sat Nov 12 19:01:00 2022 \n\n[---] start processing checks\n[ RUN ] T0 /c9c2be9f @tresa:default+gnu\n[ RUN ] T0 /c9c2be9f @tresa:default+clang\n[ OK ] ( 1/20) T0 /c9c2be9f @tresa:default+gnu\n[ OK ] ( 2/20) T0 /c9c2be9f @tresa:default+clang\n[ RUN ] T4 /11ee5e9a @tresa:default+gnu\n[ RUN ] T4 /11ee5e9a @tresa:default+clang\n[ OK ] ( 3/20) T4 /11ee5e9a @tresa:default+gnu\n[ OK ] ( 4/20) T4 /11ee5e9a @tresa:default+clang\n[ RUN ] T5 /020d01e5 @tresa:default+gnu\n[ RUN ] T5 /020d01e5 @tresa:default+clang\n[ OK ] ( 5/20) T5 /020d01e5 @tresa:default+gnu\n[ OK ] ( 6/20) T5 /020d01e5 @tresa:default+clang\n[ RUN ] T1 /1f93603d @tresa:default+gnu\n[ RUN ] T1 /1f93603d @tresa:default+clang\n[ OK ] ( 7/20) T1 /1f93603d @tresa:default+gnu\n[ OK ] ( 8/20) T1 /1f93603d @tresa:default+clang\n[ RUN ] T8 /605fc1d6 @tresa:default+gnu\n[ FAIL ] ( 9/20) T8 /605fc1d6 @tresa:default+gnu\n==> test failed during 'setup': test staged in '/home/user/Repositories/reframe/stage/tresa/default/gnu/T8'\n[ RUN ] T8 /605fc1d6 @tresa:default+clang\n[ FAIL ] (10/20) T8 /605fc1d6 @tresa:default+clang\n==> test failed during 'setup': test staged in '/home/user/Repositories/reframe/stage/tresa/default/clang/T8'\n[ FAIL ] (11/20) T9 /78a78a4e @tresa:default+gnu\n==> test failed during 'startup': test staged in None\n[ FAIL ] (12/20) T9 /78a78a4e @tresa:default+clang\n==> test failed during 'startup': test staged in None\n[ RUN ] T6 /6dbdaf93 @tresa:default+gnu\n[ RUN ] T6 /6dbdaf93 @tresa:default+clang\n[ OK ] (13/20) T6 /6dbdaf93 @tresa:default+gnu\n[ OK ] (14/20) T6 /6dbdaf93 @tresa:default+clang\n[ RUN ] T2 /0f617ba9 @tresa:default+gnu\n[ RUN ] T2 /0f617ba9 @tresa:default+clang\n[ RUN ] T3 /5dd67f7f @tresa:default+gnu\n[ RUN ] T3 /5dd67f7f @tresa:default+clang\n[ FAIL ] (15/20) T2 /0f617ba9 @tresa:default+gnu\n==> test failed during 'sanity': test staged in '/home/user/Repositories/reframe/stage/tresa/default/gnu/T2'\n[ FAIL ] (16/20) T2 /0f617ba9 @tresa:default+clang\n==> test failed during 'sanity': test staged in '/home/user/Repositories/reframe/stage/tresa/default/clang/T2'\n[ FAIL ] (17/20) T7 /f005e93d @tresa:default+gnu\n==> test failed during 'startup': test staged in None\n[ FAIL ] (18/20) T7 /f005e93d @tresa:default+clang\n==> test failed during 'startup': test staged in None\n[ OK ] (19/20) T3 /5dd67f7f @tresa:default+gnu\n[ OK ] (20/20) T3 /5dd67f7f @tresa:default+clang\n[---] all spawned checks have finished\n\n[ FAILED ] Ran 20/20 test case(s) from 10 check(s) (8 failure(s), 0 skipped)\n[===] Finished on Sat Nov 12 19:01:03 2022 \n\n===\nSUMMARY OF FAILURES\n---\nFAILURE INFO for T8 \n * Expanded name: T8\n * Description: \n * System partition: tresa:default\n * Environment: gnu\n * Stage directory: /home/user/Repositories/reframe/stage/tresa/default/gnu/T8\n * Node list: \n * Job type: local (id=None)\n * Dependencies (conceptual): ['T1']\n * Dependencies (actual): [('T1', 'tresa:default', 'gnu')]\n * Maintainers: []\n * Failing phase: setup\n * Rerun with '-n /605fc1d6 -p gnu --system tresa:default -r'\n * Reason: exception Traceback (most recent call last):\n File \"/home/user/Repositories/reframe/reframe/frontend/executors/__init__.py\", line 303, in _safe_call\n return fn(*args, **kwargs)\n File \"/home/user/Repositories/reframe/reframe/core/hooks.py\", line 101, in _fn\n getattr(obj, h.__name__)()\n File \"/home/user/Repositories/reframe/reframe/core/hooks.py\", line 32, in _fn\n func(*args, **kwargs)\n File \"/home/user/Repositories/reframe/unittests/resources/checks_unlisted/deps_complex.py\", line 180, in fail\n raise Exception Exception\n\n---\nFAILURE INFO for T8 \n * Expanded name: T8\n * Description: \n * System partition: tresa:default\n * Environment: clang\n * Stage directory: /home/user/Repositories/reframe/stage/tresa/default/clang/T8\n * Node list: \n * Job type: local (id=None)\n * Dependencies (conceptual): ['T1']\n * Dependencies (actual): [('T1', 'tresa:default', 'clang')]\n * Maintainers: []\n * Failing phase: setup\n * Rerun with '-n /605fc1d6 -p clang --system tresa:default -r'\n * Reason: exception Traceback (most recent call last):\n File \"/home/user/Repositories/reframe/reframe/frontend/executors/__init__.py\", line 303, in _safe_call\n return fn(*args, **kwargs)\n File \"/home/user/Repositories/reframe/reframe/core/hooks.py\", line 101, in _fn\n getattr(obj, h.__name__)()\n File \"/home/user/Repositories/reframe/reframe/core/hooks.py\", line 32, in _fn\n func(*args, **kwargs)\n File \"/home/user/Repositories/reframe/unittests/resources/checks_unlisted/deps_complex.py\", line 180, in fail\n raise Exception Exception\n\n---\nFAILURE INFO for T9 \n * Expanded name: T9\n * Description: \n * System partition: tresa:default\n * Environment: gnu\n * Stage directory: None\n * Node list: \n * Job type: local (id=None)\n * Dependencies (conceptual): ['T8']\n * Dependencies (actual): [('T8', 'tresa:default', 'gnu')]\n * Maintainers: []\n * Failing phase: startup\n * Rerun with '-n /78a78a4e -p gnu --system tresa:default -r'\n * Reason: task dependency error: dependencies failed\n---\nFAILURE INFO for T9 \n * Expanded name: T9\n * Description: \n * System partition: tresa:default\n * Environment: clang\n * Stage directory: None\n * Node list: \n * Job type: local (id=None)\n * Dependencies (conceptual): ['T8']\n * Dependencies (actual): [('T8', 'tresa:default', 'clang')]\n * Maintainers: []\n * Failing phase: startup\n * Rerun with '-n /78a78a4e -p clang --system tresa:default -r'\n * Reason: task dependency error: dependencies failed\n---\nFAILURE INFO for T2 \n * Expanded name: T2\n * Description: \n * System partition: tresa:default\n * Environment: gnu\n * Stage directory: /home/user/Repositories/reframe/stage/tresa/default/gnu/T2\n * Node list: hostNone\n * Job type: local (id=59611)\n * Dependencies (conceptual): ['T6']\n * Dependencies (actual): [('T6', 'tresa:default', 'gnu')]\n * Maintainers: []\n * Failing phase: sanity\n * Rerun with '-n /0f617ba9 -p gnu --system tresa:default -r'\n * Reason: sanity error: 31 != 30\n---\nFAILURE INFO for T2 \n * Expanded name: T2\n * Description: \n * System partition: tresa:default\n * Environment: clang\n * Stage directory: /home/user/Repositories/reframe/stage/tresa/default/clang/T2\n * Node list: hostNone\n * Job type: local (id=59612)\n * Dependencies (conceptual): ['T6']\n * Dependencies (actual): [('T6', 'tresa:default', 'clang')]\n * Maintainers: []\n * Failing phase: sanity\n * Rerun with '-n /0f617ba9 -p clang --system tresa:default -r'\n * Reason: sanity error: 31 != 30\n---\nFAILURE INFO for T7 \n * Expanded name: T7\n * Description: \n * System partition: tresa:default\n * Environment: gnu\n * Stage directory: None\n * Node list: \n * Job type: local (id=None)\n * Dependencies (conceptual): ['T2']\n * Dependencies (actual): [('T2', 'tresa:default', 'gnu')]\n * Maintainers: []\n * Failing phase: startup\n * Rerun with '-n /f005e93d -p gnu --system tresa:default -r'\n * Reason: task dependency error: dependencies failed\n---\nFAILURE INFO for T7 \n * Expanded name: T7\n * Description: \n * System partition: tresa:default\n * Environment: clang\n * Stage directory: None\n * Node list: \n * Job type: local (id=None)\n * Dependencies (conceptual): ['T2']\n * Dependencies (actual): [('T2', 'tresa:default', 'clang')]\n * Maintainers: []\n * Failing phase: startup\n * Rerun with '-n /f005e93d -p clang --system tresa:default -r'\n * Reason: task dependency error: dependencies failed\n---\nRun report saved in '/home/user/.reframe/reports/run-report-326.json'\nLog file(s) saved in '/<KEY>rfm-_008n_el.log'\n```\nYou can restore the run session and run only the failed test cases as follows:\n```\n./bin/reframe --restore-session --failed -r\n```\nOf course, as expected, the run will fail again, since these tests were designed to fail.\n\nInstead of running the failed test cases of a previous run, you might simply want to rerun a specific test.\nThis has little meaning if you don’t use dependencies, because it would be equivalent to running it separately using the [`-n`](index.html#cmdoption-n) option.\nHowever, if a test was part of a dependency chain, using [`--restore-session`](index.html#cmdoption-restore-session) will not rerun its dependencies, but it will rather restore them.\nThis is useful in cases where the test that we want to rerun depends on time-consuming tests.\nThere is a little tweak, though, for this to work:\nyou need to have run with [`--keep-stage-files`](index.html#cmdoption-keep-stage-files) in order to keep the stage directory even for tests that have passed.\nThis is due to two reasons:\n(a) if a test needs resources from its parents, it will look into their stage directories and\n(b) ReFrame stores the state of a finished test case inside its stage directory and it will need that state information in order to restore a test case.\n\nLet’s try to rerun the `T6` test from the previous test dependency chain:\n```\n./bin/reframe -c unittests/resources/checks_unlisted/deps_complex.py --keep-stage-files -r\n```\n```\n./bin/reframe --restore-session --keep-stage-files -n T6 -r\n```\nNotice how only the `T6` test was rerun and none of its dependencies, since they were simply restored:\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe --restore-session --keep-stage-files -n T6 -r'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>', '/home/user/Repositories/reframe/tutorials/config/tresa.py'\n check search path: '/home/user/Repositories/reframe/unittests/resources/checks_unlisted/deps_complex.py'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/<KEY>'\n\n[===] Running 1 check(s)\n[===] Started on Sat Nov 12 19:01:06 2022 \n\n[---] start processing checks\n[ RUN ] T6 /6dbdaf93 @tresa:default+gnu\n[ RUN ] T6 /6dbdaf93 @tresa:default+clang\n[ OK ] (1/2) T6 /6dbdaf93 @tresa:default+gnu\n[ OK ] (2/2) T6 /6dbdaf93 @tresa:default+clang\n[---] all spawned checks have finished\n\n[ PASSED ] Ran 2/2 test case(s) from 1 check(s) (0 failure(s), 0 skipped)\n[===] Finished on Sat Nov 12 19:01:07 2022 \nRun report saved in '/home/user/.reframe/reports/run-report-328.json'\nLog file(s) saved in '/<KEY>'\n```\nIf we tried to run `T6` without restoring the session, we would have to rerun also the whole dependency chain, i.e., also `T5`, `T1`, `T4` and `T0`.\n```\n./bin/reframe -c unittests/resources/checks_unlisted/deps_complex.py -n T6 -r\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.2+5ea6b7a6\n command: './bin/reframe -c unittests/resources/checks_unlisted/deps_complex.py -n T6 -r'\n launched by: user@host\n working directory: '/home/user/Repositories/reframe'\n settings files: '<builtin>', '/home/user/Repositories/reframe/tutorials/config/tresa.py'\n check search path: '/home/user/Repositories/reframe/unittests/resources/checks_unlisted/deps_complex.py'\n stage directory: '/home/user/Repositories/reframe/stage'\n output directory: '/home/user/Repositories/reframe/output'\n log files: '/<KEY>.log'\n\n[===] Running 5 check(s)\n[===] Started on Sat Nov 12 19:01:07 2022 \n\n[---] start processing checks\n[ RUN ] T0 /c9c2be9f @tresa:default+gnu\n[ RUN ] T0 /c9c2be9f @tresa:default+clang\n[ OK ] ( 1/10) T0 /c9c2be9f @tresa:default+gnu\n[ OK ] ( 2/10) T0 /c9c2be9f @tresa:default+clang\n[ RUN ] T4 /11ee5e9a @tresa:default+gnu\n[ RUN ] T4 /11ee5e9a @tresa:default+clang\n[ OK ] ( 3/10) T4 /11ee5e9a @tresa:default+gnu\n[ OK ] ( 4/10) T4 /11ee5e9a @tresa:default+clang\n[ RUN ] T5 /020d01e5 @tresa:default+gnu\n[ RUN ] T5 /020d01e5 @tresa:default+clang\n[ OK ] ( 5/10) T5 /020d01e5 @tresa:default+gnu\n[ OK ] ( 6/10) T5 /020d01e5 @tresa:default+clang\n[ RUN ] T1 /1f93603d @tresa:default+gnu\n[ RUN ] T1 /1f93603d @tresa:default+clang\n[ OK ] ( 7/10) T1 /1f93603d @tresa:default+gnu\n[ OK ] ( 8/10) T1 /1f93603d @tresa:default+clang\n[ RUN ] T6 /6dbdaf93 @tresa:default+gnu\n[ RUN ] T6 /6dbdaf93 @tresa:default+clang\n[ OK ] ( 9/10) T6 /6dbdaf93 @tresa:default+gnu\n[ OK ] (10/10) T6 /6dbdaf93 @tresa:default+clang\n[---] all spawned checks have finished\n\n[ PASSED ] Ran 10/10 test case(s) from 5 check(s) (0 failure(s), 0 skipped)\n[===] Finished on Sat Nov 12 19:01:08 2022 \nRun report saved in '/home/user/.reframe/reports/run-report-329.json'\nLog file(s) saved in '/<KEY>rf<KEY>log'\n```\n##### Implementing test workarounds efficiently[¶](#implementing-test-workarounds-efficiently)\n\nNew in version 3.2.\n\nSometimes you may need to add a quick workaround in a test, because something in a system or an environment broken.\nThe best way to implement this is through hooks, because you can easily disable any hook from the command-line and you don’t need to update the test every time you want to check if the system is fixed and the workaround is not needed anymore.\n\nLet’s use one example from the [previous tutorial](tutorial_basics.html) and let’s assume that there is something wrong with one of the environments and a special macro needs to be defined in order for the compilation to succeed.\nInstead of adding another flag in the `set_compilation_flags()` hook, it is better to add another hook containing just the workaround as shown below:\n```\nimport reframe as rfm import reframe.utility.sanity as sn\n\n@rfm.simple_test class HelloThreadedExtended2Test(rfm.RegressionTest):\n valid_systems = ['*']\n valid_prog_environs = ['*']\n sourcepath = 'hello_threads.cpp'\n build_system = 'SingleSource'\n executable_opts = ['16']\n\n @run_before('compile')\n def set_compilation_flags(self):\n self.build_system.cppflags = ['-DSYNC_MESSAGES']\n self.build_system.cxxflags = ['-std=c++11', '-Wall']\n environ = self.current_environ.name\n if environ in {'clang', 'gnu'}:\n self.build_system.cxxflags += ['-pthread']\n\n @sanity_function\n def assert_num_messages(self):\n num_messages = sn.len(sn.findall(r'\\[\\s?\\d+\\] Hello, World\\!',\n self.stdout))\n return sn.assert_eq(num_messages, 16)\n\n @run_before('compile')\n def fooenv_workaround(self):\n ce = self.current_environ.name\n if ce == 'foo':\n self.build_system.cppflags += [\n '-D__GCC_ATOMIC_TEST_AND_SET_TRUEVAL'\n ]\n```\nThis way the test will start passing again allowing us to catch any new issues while waiting for the original issue to be fixed.\nThen we can run the test anytime using `--disable-hook=fooenv_workaround` to check if the workaround is not needed anymore.\n\n##### Integrating into a CI pipeline[¶](#integrating-into-a-ci-pipeline)\n\nNew in version 3.4.1.\n\nInstead of running your tests, you can ask ReFrame to generate a [child pipeline](https://docs.gitlab.com/ee/ci/parent_child_pipelines.html) specification for the Gitlab CI.\nThis will spawn a CI job for each ReFrame test respecting test dependencies.\nYou could run your tests in a single job of your Gitlab pipeline, but you would not take advantage of the parallelism across different CI jobs.\nHaving a separate CI job per test makes it also easier to spot the failing tests.\n\nAs soon as you have set up a [runner](https://docs.gitlab.com/ee/ci/quick_start/) for your repository, it is fairly straightforward to use ReFrame to automatically generate the necessary CI steps.\nThe following is an example of `.gitlab-ci.yml` file that does exactly that:\n```\nstages:\n - generate\n - test\n\ngenerate-pipeline:\n stage: generate\n script:\n - reframe --ci-generate=${CI_PROJECT_DIR}/pipeline.yml -c ${CI_PROJECT_DIR}/path/to/tests\n artifacts:\n paths:\n - ${CI_PROJECT_DIR}/pipeline.yml\n\ntest-jobs:\n stage: test\n trigger:\n include:\n - artifact: pipeline.yml\n job: generate-pipeline\n strategy: depend\n```\nIt defines two stages.\nThe first one, called `generate`, will call ReFrame to generate the pipeline specification for the desired tests.\nAll the usual [test selection options](manpage.html#test-filtering) can be used to select specific tests.\nReFrame will process them as usual, but instead of running the selected tests, it will generate the correct steps for running each test individually as a Gitlab job in a child pipeline.\nThe generated ReFrame command that will run each individual test reuses the [`-C`](index.html#cmdoption-C), [`-R`](index.html#cmdoption-R), [`-v`](index.html#cmdoption-4) and [`--mode`](index.html#cmdoption-mode) options passed to the initial invocation of ReFrame that was used to generate the pipeline.\nUsers can define CI-specific execution modes in their configuration in order to pass arbitrary options to the ReFrame invocation in the child pipeline.\n\nFinally, we pass the generated CI pipeline file to second phase as an artifact and we are done!\nIf `image` keyword is defined in `.gitlab-ci.yml`, the emitted pipeline will use the same image as the one defined in the parent pipeline.\nBesides, each job in the generated pipeline will output a separate junit report which can be used to create GitLab badges.\n\nThe following figure shows one part of the automatically generated pipeline for the test graph depicted [above](#fig-deps-complex).\n\nSnapshot of a Gitlab pipeline generated automatically by ReFrame.[¶](#id3)\n\nNote\n\nThe ReFrame executable must be available in the Gitlab runner that will run the CI jobs.\n\n#### Tutorial 7: The Flux Framework Scheduler[¶](#tutorial-7-the-flux-framework-scheduler)\n\nThis is a tutorial that will show how to use refame with [Flux Framework](https://github.com/flux-framework/). First, build the container here from the root of reframe.\n```\n$ docker build -f tutorials/flux/Dockerfile -t flux-reframe .\n```\nThen shell inside, optionally binding the present working directory if you want to develop.\n```\n$ docker run -it -v $PWD:/code flux-reframe\n$ docker run -it flux-reframe\n```\nNote that if you build the local repository, you’ll need to bootstrap and install again, as we have over-written the bin!\n```\n./bootstrap.sh\n```\nAnd then reframe will again be in the local `bin` directory:\n```\n# which reframe\n/code/bin/reframe\n```\nThen we can run ReFrame with the custom config <config.py>\nfor flux.\n```\n# What tests are under tutorials/flux?\n$ cd tutorials/flux\n$ reframe -c . -C settings.py -l\n```\n```\n[ReFrame Setup]\n version: 4.0.0-dev.1\n command: '/code/bin/reframe -c tutorials/flux -C tutorials/flux/settings.py -l'\n launched by: root@b1f6650222bc\n working directory: '/code'\n settings file: 'tutorials/flux/settings.py'\n check search path: '/code/tutorials/flux'\n stage directory: '/code/stage'\n output directory: '/code/output'\n\n[List of matched checks]\n- EchoRandTest /66b93401 Found 1 check(s)\n\nLog file(s) saved in '/tmp/rfm-ilqg7fqg.log'\n```\nThis also works\n```\n$ reframe -c tutorials/flux -C tutorials/flux/settings.py -l\n```\nAnd then to run tests, just replace `-l` (for list) with `-r` or\n`--run` (for run):\n```\n$ reframe -c tutorials/flux -C tutorials/flux/settings.py --run\n```\n```\nroot@b1f6650222bc:/code# reframe -c tutorials/flux -C tutorials/flux/settings.py --run\n[ReFrame Setup]\n version: 4.0.0-dev.1\n command: '/code/bin/reframe -c tutorials/flux -C tutorials/flux/settings.py --run'\n launched by: root@b1f6650222bc\n working directory: '/code'\n settings file: 'tutorials/flux/settings.py'\n check search path: '/code/tutorials/flux'\n stage directory: '/code/stage'\n output directory: '/code/output'\n\n[===] Running 1 check(s)\n[===] Started on Fri Sep 16 20:47:15 2022\n\n[---] start processing checks\n[ RUN ] EchoRandTest /66b93401 @generic:default+builtin\n[ OK ] (1/1) EchoRandTest /66b93401 @generic:default+builtin\n[---] all spawned checks have finished\n\n[ PASSED ] Ran 1/1 test case(s) from 1 check(s) (0 failure(s), 0 skipped)\n[===] Finished on Fri Sep 16 20:47:15 2022 Run report saved in '/root/.reframe/reports/run-report.json'\nLog file(s) saved in '/tmp/rfm-0avso9nb.log'\n```\nFor advanced users or developers, here is how to run tests within the container:\n\n##### Testing[¶](#testing)\n```\n./test_reframe.py --rfm-user-config=tutorials/flux/settings.py unittests/test_schedulers.py -xs\n```\n#### Online Tutorials[¶](#online-tutorials)\n\n* Tutorial at 6th EasyBuild User Meeting 2021 [[YouTube](https://youtube.com/playlist?list=PLhnGtSmEGEQjySVEPTUSLpewpOWwX5mjb)]\n\n### Configuring ReFrame for Your Site[¶](#configuring-reframe-for-your-site)\n\nReFrame comes pre-configured with a minimal generic configuration that will allow you to run ReFrame on any system.\nThis will allow you to run simple local tests using the default compiler of the system.\nOf course, ReFrame is much more powerful than that.\nThis section will guide you through configuring ReFrame for your site.\n\nReFrame’s configuration can be either in JSON or in Python format and can be split into multiple files.\nThe Python format is useful in cases that you want to generate configuration parameters on-the-fly, since ReFrame will import that Python file and the load the resulting configuration.\nIn the following we will use a single Python-based configuration file also for historical reasons, since it was the only way to configure ReFrame in versions prior to 3.0.\n\nChanged in version 4.0.0: The configuration can now be split into multiple files.\n\n#### Loading the configuration[¶](#loading-the-configuration)\n\nReFrame builds its final configuration gradually by combining multiple configuration files.\nEach one can have different parts of the configuration, for example different systems, different environments, different general options or different logging handlers.\nThis technique allows users to avoid having a single huge configuration file.\n\nThe first configuration file loaded in this chain is always the generic builtin configuration located under `${RFM_INSTALL_PREFIX}/reframe/core/settings.py`.\nThis contains everything that ReFrame needs to run on a generic system, as well as basic settings for logging, so subsequent configuration files may skip defining some configuration sections altogether, if they are not relevant.\n\nReFrame continues on looking for configuration files in the directories defined in [`RFM_CONFIG_PATH`](index.html#envvar-RFM_CONFIG_PATH).\nFor each directory, will look within it for a `settings.py` or `settings.json` file (in that order), and if it finds one, it will load it.\n\nFinally, ReFrame processes the [`--config-file`](index.html#cmdoption-C) option or the [`RFM_CONFIG_FILES`](index.html#envvar-RFM_CONFIG_FILES) environment variable to load any specific configuration files passed from the command line.\n\n#### Anatomy of the Configuration File[¶](#anatomy-of-the-configuration-file)\n\nThe whole configuration of ReFrame is a single JSON object whose properties are responsible for configuring the basic aspects of the framework.\nWe’ll refer to these top-level properties as *sections*.\nThese sections contain other objects which further define in detail the framework’s behavior.\nIf you are using a Python file to configure ReFrame, this big JSON configuration object is stored in a special variable called `site_configuration`.\n\nWe will explore the basic configuration of ReFrame by looking into the configuration file of the tutorials, which permits ReFrame to run on the Piz Daint supercomputer and a local computer.\nFor the complete listing and description of all configuration options, you should refer to the [Configuration Reference](index.html#document-config_reference).\n```\nsite_configuration = {\n 'systems': [\n {\n 'name': 'daint',\n 'descr': 'Piz Daint Supercomputer',\n 'hostnames': ['daint'],\n 'modules_system': 'tmod32',\n 'partitions': [\n {\n 'name': 'login',\n 'descr': 'Login nodes',\n 'scheduler': 'local',\n 'launcher': 'local',\n 'environs': ['builtin', 'gnu', 'intel', 'nvidia', 'cray'],\n },\n {\n 'name': 'gpu',\n 'descr': 'Hybrid nodes',\n 'scheduler': 'slurm',\n 'launcher': 'srun',\n 'access': ['-C gpu', '-A csstaff'],\n 'environs': ['gnu', 'intel', 'nvidia', 'cray'],\n 'max_jobs': 100,\n },\n {\n 'name': 'mc',\n 'descr': 'Multicore nodes',\n 'scheduler': 'slurm',\n 'launcher': 'srun',\n 'access': ['-C mc', '-A csstaff'],\n 'environs': ['gnu', 'intel', 'nvidia', 'cray'],\n 'max_jobs': 100,\n 'resources': [\n {\n 'name': 'memory',\n 'options': ['--mem={size}']\n }\n ]\n }\n ]\n }\n ],\n 'environments': [\n {\n 'name': 'gnu',\n 'modules': ['PrgEnv-gnu'],\n 'cc': 'cc',\n 'cxx': 'CC',\n 'ftn': 'ftn',\n 'target_systems': ['daint']\n },\n {\n 'name': 'cray',\n 'modules': ['PrgEnv-cray'],\n 'cc': 'cc',\n 'cxx': 'CC',\n 'ftn': 'ftn',\n 'target_systems': ['daint']\n },\n {\n 'name': 'intel',\n 'modules': ['PrgEnv-intel'],\n 'cc': 'cc',\n 'cxx': 'CC',\n 'ftn': 'ftn',\n 'target_systems': ['daint']\n },\n {\n 'name': 'nvidia',\n 'modules': ['PrgEnv-nvidia'],\n 'cc': 'cc',\n 'cxx': 'CC',\n 'ftn': 'ftn',\n 'target_systems': ['daint']\n },\n {\n 'name': 'builtin',\n 'cc': 'cc',\n 'cxx': 'CC',\n 'ftn': 'ftn',\n 'target_systems': ['daint']\n }\n ] # end of environments\n}\n```\nThere are three required sections that the final ReFrame configuration must have: `systems`, `environments` and `logging`, but in most cases you will define only the first two, as ReFrame’s builtin configuration already defines a reasonable logging configuration. We will first cover these sections and then move on to the optional ones.\n\nTip\n\nThese configuration sections may not all be defined in the same configuration file, but can reside in any configuration file that is being loaded.\nThis is the case of the example configuration shown above, where the `logging` section is “missing” as it’s defined in ReFrame’s builtin configuration.\n\n##### Systems Configuration[¶](#systems-configuration)\n\nReFrame allows you to configure multiple systems in the same configuration file.\nEach system is a different object inside the `systems` section.\nIn our example we define only Piz Daint:\n```\n 'systems': [\n {\n 'name': 'daint',\n 'descr': 'Piz Daint Supercomputer',\n 'hostnames': ['daint'],\n 'modules_system': 'tmod32',\n 'partitions': [\n {\n 'name': 'login',\n 'descr': 'Login nodes',\n 'scheduler': 'local',\n 'launcher': 'local',\n 'environs': ['builtin', 'gnu', 'intel', 'nvidia', 'cray'],\n },\n {\n 'name': 'gpu',\n 'descr': 'Hybrid nodes',\n 'scheduler': 'slurm',\n 'launcher': 'srun',\n 'access': ['-C gpu', '-A csstaff'],\n 'environs': ['gnu', 'intel', 'nvidia', 'cray'],\n 'max_jobs': 100,\n },\n {\n 'name': 'mc',\n 'descr': 'Multicore nodes',\n 'scheduler': 'slurm',\n 'launcher': 'srun',\n 'access': ['-C mc', '-A csstaff'],\n 'environs': ['gnu', 'intel', 'nvidia', 'cray'],\n 'max_jobs': 100,\n 'resources': [\n {\n 'name': 'memory',\n 'options': ['--mem={size}']\n }\n ]\n }\n ]\n }\n ],\n```\nEach system is associated with a set of properties, which in this case are the following:\n\n* `name`: The name of the system.\nThis should be an alphanumeric string (dashes `-` are allowed) and it will be used to refer to this system in other contexts.\n* `descr`: A detailed description of the system.\n* `hostnames`: This is a list of hostname patterns following the [Python Regular Expression Syntax](https://docs.python.org/3/library/re.html#regular-expression-syntax), which will be used by ReFrame when it tries to automatically select a configuration entry for the current system.\n* `modules_system`: This refers to the modules management backend which should be used for loading environment modules on this system.\nMultiple backends are supported, as well as the special `nomod` backend which implements the different modules system operations as no-ops.\nFor the complete list of the supported modules systems, see [here](config_reference.html#.systems[].modules_system).\n* `partitions`: The list of partitions that are defined for this system.\nEach partition is defined as a separate object.\nWe devote the rest of this section in system partitions, since they are an essential part of ReFrame’s configuration.\n\nA system partition in ReFrame is not bound to a real scheduler partition.\nIt is a virtual partition or separation of the system.\nIn the example shown here, we define three partitions that none of them corresponds to a scheduler partition.\nThe `login` partition refers to the login nodes of the system, whereas the `gpu` and `mc` partitions refer to two different set of nodes in the same cluster that are effectively separated using Slurm constraints.\nLet’s pick the `gpu` partition and look into it in more detail:\n```\n 'name': 'gpu',\n 'descr': 'Hybrid nodes',\n 'scheduler': 'slurm',\n 'launcher': 'srun',\n 'access': ['-C gpu', '-A csstaff'],\n 'environs': ['gnu', 'intel', 'nvidia', 'cray'],\n 'max_jobs': 100,\n```\nThe basic properties of a partition are the following:\n\n* `name`: The name of the partition.\nThis should be an alphanumeric string (dashes `-` are allowed) and it will be used to refer to this partition in other contexts.\n* `descr`: A detailed description of the system partition.\n* `scheduler`: The workload manager (job scheduler) used in this partition for launching parallel jobs.\nIn this particular example, the [Slurm](https://slurm.schedmd.com/) scheduler is used.\nFor a complete list of the supported job schedulers, see [here](config_reference.html#.systems[].partitions[].scheduler).\n* `launcher`: The parallel job launcher used in this partition.\nIn this case, the `srun` command will be used.\nFor a complete list of the supported parallel job launchers, see [here](config_reference.html#.systems[].partitions[].launcher).\n* `access`: A list of scheduler options that will be passed to the generated job script for gaining access to that logical partition.\nNotice how in this case, the nodes are selected through a constraint and not an actual scheduler partition.\n* `environs`: The list of environments that ReFrame will use to run regression tests on this partition.\nThese are just symbolic names that refer to environments defined in the `environments` section described below.\n* `max_jobs`: The maximum number of concurrent regression tests that may be active (i.e., not completed) on this partition.\nThis option is relevant only when ReFrame executes with the [asynchronous execution policy](pipeline.html#execution-policies).\n\nFor more partition configuration options, have a look [here](config_reference.html#system-partition-configuration).\n\n##### Environments Configuration[¶](#environments-configuration)\n\nWe have seen already environments to be referred to by the `environs` property of a partition.\nAn environment in ReFrame is simply a collection of environment modules, environment variables and compiler and compiler flags definitions.\nNone of these attributes is required.\nAn environment can simply by empty, in which case it refers to the actual environment that ReFrame runs in.\nIn fact, this is what the generic fallback configuration of ReFrame does.\n\nEnvironments in ReFrame are configured under the `environments` section of the documentation.\nFor each environment referenced inside a partition, a definition of it must be present in this section.\nIn our example, we define environments for all the basic compilers as well as a default built-in one, which is used with the generic system configuration.\nIn certain contexts, it is useful to see a ReFrame environment as a wrapper of a programming toolchain (MPI + compiler combination):\n```\n 'environments': [\n {\n 'name': 'gnu',\n 'modules': ['PrgEnv-gnu'],\n 'cc': 'cc',\n 'cxx': 'CC',\n 'ftn': 'ftn',\n 'target_systems': ['daint']\n },\n {\n 'name': 'cray',\n 'modules': ['PrgEnv-cray'],\n 'cc': 'cc',\n 'cxx': 'CC',\n 'ftn': 'ftn',\n 'target_systems': ['daint']\n },\n {\n 'name': 'intel',\n 'modules': ['PrgEnv-intel'],\n 'cc': 'cc',\n 'cxx': 'CC',\n 'ftn': 'ftn',\n 'target_systems': ['daint']\n },\n {\n 'name': 'nvidia',\n 'modules': ['PrgEnv-nvidia'],\n 'cc': 'cc',\n 'cxx': 'CC',\n 'ftn': 'ftn',\n 'target_systems': ['daint']\n },\n {\n 'name': 'builtin',\n 'cc': 'cc',\n 'cxx': 'CC',\n 'ftn': 'ftn',\n 'target_systems': ['daint']\n }\n ] # end of environments\n```\nEach environment is associated with a name.\nThis name will be used to reference this environment in different contexts, as for example in the `environs` property of the system partitions.\nA programming environment in ReFrame is essentially a collection of environment modules, environment variables and compiler definitions.\n\nAn important feature in ReFrame’s configuration, is that you can define section objects differently for different systems or system partitions by using the `target_systems` property.\nNotice, for example, how the `gnu` environment is defined differently for the system `daint` compared to the generic definition.\nThe `target_systems` property is a list of systems or system/partition combinations where this definition of the environment is in effect.\nThis means that `gnu` will be defined this way only for regression tests running on `daint`.\nFor all the other systems, it will be defined using the first definition.\n\n##### Logging configuration[¶](#logging-configuration)\n\nReFrame has a powerful logging mechanism that gives fine grained control over what information is being logged, where it is being logged and how this information is formatted.\nAdditionally, it allows for logging performance data from performance tests into different channels.\nLet’s see how logging is defined in the builtin configuration:\n```\n 'logging': [\n {\n 'handlers$': [\n {\n 'type': 'stream',\n 'name': 'stdout',\n 'level': 'info',\n 'format': '%(message)s'\n },\n ],\n 'handlers': [\n {\n 'type': 'file',\n 'level': 'debug2',\n 'format': '[%(asctime)s] %(levelname)s: %(check_info)s: %(message)s', # noqa: E501\n 'append': False\n }\n ],\n 'handlers_perflog': [\n {\n 'type': 'filelog',\n 'prefix': '%(check_system)s/%(check_partition)s',\n 'level': 'info',\n 'format': (\n '%(check_job_completion_time)s,%(version)s,'\n '%(check_display_name)s,%(check_system)s,'\n '%(check_partition)s,%(check_environ)s,'\n '%(check_jobid)s,%(check_result)s,%(check_perfvalues)s'\n ),\n 'format_perfvars': (\n '%(check_perf_value)s,%(check_perf_unit)s,'\n '%(check_perf_ref)s,%(check_perf_lower)s,'\n '%(check_perf_upper)s,'\n ),\n 'append': True\n }\n ]\n }\n ] # end of logging\n```\nLogging is configured under the `logging` section of the configuration, which is a list of logger objects.\nUnless you want to configure logging differently for different systems, a single logger object is enough.\nEach logger object is associated with a [logging level](config_reference.html#.logging[].level) stored in the `level` property and has a set of logging handlers that are actually responsible for handling the actual logging records.\nReFrame’s output is performed through its logging mechanism and that’s why there is the special `handlers$` property.\nThe handler defined in this property, in the builtin configuration shown here, defines how exactly the output of ReFrame will be printed.\nYou will not have to override this in your configuration files, unless you really need to change how ReFrame’s output look like.\n\nAs a user you might need to override the `handlers` property to define different sinks for ReFrame logs and/or output using different verbosity levels.\nNote that you can use multiple handlers at the same time.\nAll handler objects share a set of common properties.\nThese are the following:\n\n* `type`: This is the type of the handler, which determines its functionality.\nDepending on the handler type, handler-specific properties may be allowed or required.\nFor a complete list of available log handler types, see [here](config_reference.html#.logging[].handlers[].type).\n* `level`: The cut-off level for messages reaching this handler.\nAny message with a lower level number will be filtered out.\n* `format`: A format string for formatting the emitted log record.\nReFrame uses the format specifiers from [Python Logging](https://docs.python.org/3/library/logging.html?highlight=logging#logrecord-attributes), but also defines its owns specifiers.\n* `datefmt`: A time format string for formatting timestamps.\nThere are two log record fields that are considered timestamps: (a) `asctime` and (b) `check_job_completion_time`.\nReFrame follows the time formatting syntax of Python’s [time.strftime()](https://docs.python.org/3/library/time.html#time.strftime) with a small tweak allowing full RFC3339 compliance when formatting time zone differences.\n\nWe will not go into the details of the individual handlers here.\nIn this particular example we use three handlers of two distinct types:\n\n1. A file handler to print debug messages in the `reframe.log` file using a more extensive message format that contains a timestamp, the level name etc.\n2. A stream handler to print any informational messages (and warnings and errors) from ReFrame to the standard output.\nThis handles essentially the actual output of ReFrame.\n3. A file handler to print the framework’s output in the `reframe.out` file.\n\nIt might initially seem confusing the fact that there are two `level` properties: one at the logger level and one at the handler level.\nLogging in ReFrame works hierarchically.\nWhen a message is logged, a log record is created, which contains metadata about the message being logged (log level, timestamp, ReFrame runtime information etc.).\nThis log record first goes into ReFrame’s internal logger, where the record’s level is checked against the logger’s level (here `debug`).\nIf the log record’s level exceeds the log level threshold from the logger, it is forwarded to the logger’s handlers.\nThen each handler filters the log record differently and takes care of formatting the log record’s message appropriately.\nYou can view logger’s log level as a general cut off.\nFor example, if we have set it to `warning`, no debug or informational messages would ever be printed.\n\nFinally, there is a special set of handlers for handling performance log messages.\nPerformance log messages are generated *only* for [performance tests](tutorial_basics.html#writing-a-performance-test), i.e., tests defining the [`perf_variables`](index.html#reframe.core.pipeline.RegressionTest.perf_variables) or the [`perf_patterns`](index.html#reframe.core.pipeline.RegressionTest.perf_patterns) attributes.\nThe performance log handlers are stored in the `handlers_perflog` property.\nThe `filelog` handler used in this example will create a file per test and per system/partition combination (`./<system>/<partition>/<testname>.log`) and will append to it the obtained performance data every time a performance test is run.\nNotice how the message to be logged is structured in the `format` and `format_perfvars` properties, such that it can be easily parsed from post processing tools.\nApart from file logging, ReFrame offers more advanced performance logging capabilities through Syslog, Graylog and HTTP.\n\nFor a complete reference of logging configuration parameters, please refer to the [Configuration Reference](index.html#document-config_reference).\n\n##### General configuration options[¶](#general-configuration-options)\n\nGeneral configuration options of the framework go under the `general` section of the configuration file.\nThis section is optional and, in fact, we do not define it for our tutorial configuration file.\nHowever, there are several options that can go into this section, but the reader is referred to the [Configuration Reference](index.html#document-config_reference) for the complete list.\n\n##### Other configuration options[¶](#other-configuration-options)\n\nThere is finally one additional optional configuration section that is not discussed here:\n\nThe `modes` section defines different execution modes for the framework.\nExecution modes are discussed in the [How ReFrame Executes Tests](index.html#document-pipeline) page.\n\n#### Building the Final Configuration[¶](#building-the-final-configuration)\n\nNew in version 4.0.0.\n\nAs mentioned above ReFrame can build its final configuration incrementally from a series of user-specified configuration files starting from the basic builtin configuration.\nWe discussed briefly at the beginning of this page how ReFrame locates and loads these configuration files and the documentation of the [`-C`](index.html#cmdoption-C) option provides more detailed information.\nBut how are these configuration files actually combined?\nThis is what we will discuss in this section.\n\nConfiguration objects in the top-level configuration sections can be split in two categories: *named* and *unnamed*.\nNamed objects are the systems, the environments and the modes and the rest are unnamed.\nThe named object have a `name` property.\nWhen ReFrame builds its final configuration, named objects from newer configuration files are either appended or prepended in their respective sections, but unnamed objects are merged based on their `target_systems`.\nMore specifically, new systems are *prepended* in the list of the already defined, whereas environments and modes are *appended*.\nThe reason for that is that systems are tried from the beginning of the list until a match is found.\nSee [Picking the Right System Configuration](#pick-system-config) for more information on how ReFrame picks the right system.\nIf a system is redefined, ReFrame will warn about it, but it will still use the new definition.\nThis is done for backward compatibility with the old configuration mechanism, where users had to redefine also the builtin systems and environments in their configuration.\nSimilarly, if an environment or a mode is redefined, ReFrame will issue a warning, but only if the redefinition is at the same scope as the conflicting one.\nAgain this is done for backward compatibility.\n\nGiven the Piz Daint configuration shown in this section and the ReFrame’s builtin configuration, ReFrame will build internally the following configuration:\n```\nsite_configuration = {\n 'systems': [\n {\n # from the Daint config\n 'name': 'daint',\n ...\n },\n {\n # from the builtin config\n 'name': 'generic',\n ...\n }\n ],\n 'environments': [\n {\n # from the builtin config\n 'name': 'builtin'\n ...\n },\n {\n # from the Daint config\n 'name': 'gnu',\n ...\n }\n ],\n 'logging': [\n # from the builtin config\n ]\n}\n```\nYou might wonder why would I need to define multiple objects in sections such as `logging` or `general`.\nAs mentioned above, ReFrame merges them if they refer to the same target systems, but if they don’t they can serve as scopes for the configuration parameters they define.\nImagine the following `general` section:\n```\n'general': [\n {\n 'git_timeout': 5\n },\n {\n 'git_timeout': 10,\n 'target_systems': ['daint']\n },\n {\n 'git_timeout': 20,\n 'target_systems': ['tresa']\n }\n]\n```\nThis means that the default value for `git_timeout` is 5 seconds for any system, but it is 10 for `daint` and 20 for `tresa`.\nThe nice thing is that you can spread that in multiple configuration files and ReFrame will combine them internally in a single one with the various configuration options indexed by their scope.\n\n#### Picking the Right System Configuration[¶](#picking-the-right-system-configuration)\n\nAs discussed previously, ReFrame’s configuration file can store the configurations for multiple systems.\nWhen launched, ReFrame will pick the first matching configuration and load it.\n\nReFrame uses an auto-detection mechanism to get information about the host it is running on and uses that information to pick the right system configuration.\nCurrently, only one auto-detection method is supported that retrieves the hostname.\nBased on this, ReFrame goes through all the systems in its configuration and tries to match the hostname against any of the patterns defined in each system’s `hostnames` property.\nThe detection process stops at the first match found, and that system’s configuration is selected.\n\nThe auto-detection process can be controlled through the [`RFM_AUTODETECT_METHOD`](index.html#envvar-RFM_AUTODETECT_METHOD), [`RFM_AUTODETECT_FQDN`](index.html#envvar-RFM_AUTODETECT_FQDN) and [`RFM_AUTODETECT_XTHOSTNAME`](index.html#envvar-RFM_AUTODETECT_XTHOSTNAME) environment variables.\n\nAs soon as a system configuration is selected, all configuration objects that have a `target_systems` property are resolved against the selected system, and any configuration object that is not applicable is dropped.\nSo, internally, ReFrame keeps an *instantiation* of the site configuration for the selected system only.\nTo better understand this, let’s assume that we have the following `environments` defined:\n```\n'environments': [\n {\n 'name': 'cray',\n 'modules': ['cray']\n },\n {\n 'name': 'gnu',\n 'modules': ['gnu']\n },\n {\n 'name': 'gnu',\n 'modules': ['gnu', 'openmpi'],\n 'cc': 'mpicc',\n 'cxx': 'mpicxx',\n 'ftn': 'mpif90',\n 'target_systems': ['foo']\n }\n],\n```\nIf the selected system is `foo`, then ReFrame will use the second definition of `gnu` which is specific to the `foo` system.\n\nYou can override completely the system auto-selection process by specifying a system or system/partition combination with the `--system` option, e.g., `--system=daint` or `--system=daint:gpu`.\n\n#### Querying Configuration Options[¶](#querying-configuration-options)\n\nReFrame offers the powerful `--show-config` command-line option that allows you to query any configuration parameter of the framework and see how it is set for the selected system.\nUsing no arguments or passing `all` to this option, the whole configuration for the currently selected system will be printed in JSON format, which you can then pipe to a JSON command line editor, such as [jq](https://stedolan.github.io/jq/), and either get a colored output or even generate a completely new ReFrame configuration!\n\nPassing specific configuration keys in this option, you can query specific parts of the configuration.\nLet’s see some concrete examples:\n\n* Query the current system’s partitions:\n```\n./bin/reframe -C tutorials/config/settings.py --system=daint --show-config=systems/0/partitions\n```\n```\n[\n {\n \"name\": \"login\",\n \"descr\": \"Login nodes\",\n \"scheduler\": \"local\",\n \"launcher\": \"local\",\n \"environs\": [\n \"gnu\",\n \"intel\",\n \"nvidia\",\n \"cray\"\n ],\n \"max_jobs\": 10\n },\n {\n \"name\": \"gpu\",\n \"descr\": \"Hybrid nodes\",\n \"scheduler\": \"slurm\",\n \"launcher\": \"srun\",\n \"access\": [\n \"-C gpu\",\n \"-A csstaff\"\n ],\n \"environs\": [\n \"gnu\",\n \"intel\",\n \"nvidia\",\n \"cray\"\n ],\n \"max_jobs\": 100\n },\n {\n \"name\": \"mc\",\n \"descr\": \"Multicore nodes\",\n \"scheduler\": \"slurm\",\n \"launcher\": \"srun\",\n \"access\": [\n \"-C mc\",\n \"-A csstaff\"\n ],\n \"environs\": [\n \"gnu\",\n \"intel\",\n \"nvidia\",\n \"cray\"\n ],\n \"max_jobs\": 100\n }\n]\n```\nCheck how the output changes if we explicitly set system to `daint:login`:\n```\n./bin/reframe -C tutorials/config/settings.py --system=daint:login --show-config=systems/0/partitions\n```\n```\n[\n {\n \"name\": \"login\",\n \"descr\": \"Login nodes\",\n \"scheduler\": \"local\",\n \"launcher\": \"local\",\n \"environs\": [\n \"gnu\",\n \"intel\",\n \"nvidia\",\n \"cray\"\n ],\n \"max_jobs\": 10\n }\n]\n```\nReFrame will internally represent system `daint` as having a single partition only.\nNotice also how you can use indexes to objects elements inside a list.\n* Query an environment configuration:\n```\n./bin/reframe -C tutorials/config/settings.py --system=daint --show-config=environments/@gnu\n```\n```\n{\n \"name\": \"gnu\",\n \"modules\": [\n \"PrgEnv-gnu\"\n ],\n \"cc\": \"cc\",\n \"cxx\": \"CC\",\n \"ftn\": \"ftn\",\n \"target_systems\": [\n \"daint\"\n ]\n}\n```\nIf an object has a `name` property you can address it by name using the `@name` syntax, instead of its index.\n* Query an environment’s compiler:\n```\n./bin/reframe -C tutorials/config/settings.py --system=daint --show-config=environments/@gnu/cxx\n```\n```\n\"CC\"\n```\nIf you explicitly query a configuration value which is not defined in the configuration file, ReFrame will print its default value.\n\n#### Auto-detecting processor information[¶](#auto-detecting-processor-information)\n\nNew in version 3.7.0.\n\nReFrame is able to detect the processor topology of both local and remote partitions automatically.\nThe processor and device information are made available to the tests through the corresponding attributes of the [`current_partition`](index.html#reframe.core.pipeline.RegressionTest.current_partition) allowing a test to modify its behavior accordingly.\nCurrently, ReFrame supports auto-detection of the local or remote processor information only.\nIt does not support auto-detection of devices, in which cases users should explicitly specify this information using the [`devices`](config_reference.html#.systems[].partitions[].devices) configuration option.\nThe processor information auto-detection works as follows:\n\n1. If the [`processor`](config_reference.html#.systems[].partitions[].processor) configuration option is defined, then no auto-detection is attempted.\n2. If the [`processor`](config_reference.html#.systems[].partitions[].processor) configuration option is not defined, ReFrame will look for a processor configuration metadata file in `~/.reframe/topology/{system}-{part}/processor.json`.\nIf the file is found, the topology information is loaded from there.\nThese files are generated automatically by ReFrame from previous runs.\n3. If the corresponding metadata files are not found, the processor information will be auto-detected.\nIf the system partition is local (i.e., `local` scheduler + `local` launcher), the processor information is auto-detected unconditionally and stored in the corresponding metadata file for this partition.\nIf the partition is remote, ReFrame will not try to auto-detect it unless the [`RFM_REMOTE_DETECT`](index.html#envvar-RFM_REMOTE_DETECT) or the [`remote_detect`](config_reference.html#.general[].remote_detect) configuration option is set.\nIn that case, the steps to auto-detect the remote processor information are the following:\n\n> \t1. ReFrame creates a fresh clone of itself in a temporary directory created under `.` by default.\n> \tThis temporary directory prefix can be changed by setting the [`RFM_REMOTE_WORKDIR`](index.html#envvar-RFM_REMOTE_WORKDIR) environment variable.\n> \t2. ReFrame changes to that directory and launches a job that will first bootstrap the fresh clone and then run that clone with `{launcher} ./bin/reframe --detect-host-topology=topo.json`.\n> \tThe [`--detect-host-topology`](index.html#cmdoption-detect-host-topology) option causes ReFrame to detect the topology of the current host,\n> \twhich in this case would be the remote compute nodes.\nIn case of errors during auto-detection, ReFrame will simply issue a warning and continue.\n\n### Advanced Topics[¶](#advanced-topics)\n\n#### How ReFrame Executes Tests[¶](#how-reframe-executes-tests)\n\nA ReFrame test will be normally tried for different programming environments and different partitions within the same ReFrame run.\nThese can be defined in the test’s class body, in a post-init hook or in its `__init__()` method, but it is not this original test object that is scheduled for execution.\nThe following figure explains in more detail the process:\n\nHow ReFrame loads and schedules tests for execution.[¶](#id1)\n\nWhen ReFrame loads a test from the disk it unconditionally constructs it executing its `__init__()` method.\nThe practical implication of this is that your test will be instantiated even if it will not run on the current system.\nAfter all the tests are loaded, they are filtered based on the current system and any other criteria (such as programming environment, test attributes etc.) specified by the user (see [Test Filtering](manpage.html#test-filtering) for more details).\nAfter the tests are filtered, ReFrame creates the actual test cases to be run. A test case is essentially a tuple consisting of the test, the system partition and the programming environment to try.\nThe test that goes into a test case is essentially a clone of the original test that was instantiated upon loading.\nThis ensures that the test case’s state is not shared and may not be reused in any case.\nFinally, the generated test cases are passed to a runner that is responsible for scheduling them for execution based on the selected execution policy.\n\n##### The Regression Test Pipeline[¶](#the-regression-test-pipeline)\n\nEach ReFrame test case goes through a pipeline with clearly defined stages.\nReFrame tests can customize their operation as they execute by attaching hooks to the pipeline stages.\nThe following figure shows the different pipeline stages.\n\nThe regression test pipeline.[¶](#id2)\n\nAll tests will go through every stage one after the other.\nHowever, some types of tests implement some stages as no-ops, whereas the sanity or performance check phases may be skipped on demand (see [`--skip-sanity-check`](index.html#cmdoption-skip-sanity-check) and [`--skip-performance-check`](index.html#cmdoption-skip-performance-check) options).\nIn the following we describe in more detail what happens in every stage.\n\n###### The Setup Phase[¶](#the-setup-phase)\n\nDuring this phase the test will be set up for the currently selected system partition and programming environment.\nThe `current_partition` and `current_environ` test attributes will be set and the paths associated to this test case (stage, output and performance log directories) will be created.\nA [job descriptor](regression_test_api.html#reframe.core.pipeline.RegressionTest.job) will also be created for the test case containing information about the job to be submitted later in the pipeline.\n\n###### The Compile Phase[¶](#the-compile-phase)\n\nDuring this phase a job script for the compilation of the test will be created and it will be submitted for execution.\nThe source code associated with the test is compiled using the current programming environment.\nIf the test is [“run-only,”](regression_test_api.html#reframe.core.pipeline.RunOnlyRegressionTest) this phase is a no-op.\n\nBefore building the test, all the [resources](regression_test_api.html#reframe.core.pipeline.RegressionTest.sourcesdir) associated with it are copied to the test case’s stage directory.\nReFrame then temporarily switches to that directory and builds the test.\n\n###### The Run Phase[¶](#the-run-phase)\n\nDuring this phase a job script associated with the test case will be created and it will be submitted for execution.\nIf the test is [“run-only,”](regression_test_api.html#reframe.core.pipeline.RunOnlyRegressionTest) its [resources](regression_test_api.html#reframe.core.pipeline.RegressionTest.sourcesdir) will be first copied to the test case’s stage directory.\nReFrame will temporarily switch to that directory and spawn the test’s job from there.\nThis phase is executed asynchronously (either a batch job is spawned or a local process is started) and it is up to the selected [execution policy](#execution-policies) to block or not until the associated job finishes.\n\n###### The Sanity Phase[¶](#the-sanity-phase)\n\nDuring this phase, the sanity of the test’s output is checked.\nReFrame makes no assumption as of what a successful test is; it does not even look into its exit code.\nThis is entirely up to the test to define.\nReFrame provides a flexible and expressive way for specifying complex patterns and operations to be performed on the test’s output in order to determine the outcome of the test.\n\n###### The Performance Phase[¶](#the-performance-phase)\n\nDuring this phase, the performance metrics reported by the test (if it is performance test) are collected, logged and compared to their reference values.\nThe mechanism for extracting performance metrics from the test’s output is the same used by the sanity checking phase for extracting patterns from the test’s output.\n\n###### The Cleanup Phase[¶](#the-cleanup-phase)\n\nDuring this final stage of the pipeline, the test’s resources are cleaned up.\nMore specifically, if the test has finished successfully, all interesting test files (build/job scripts, build/job script output and any user-specified files) are copied to ReFrame’s output directory and the stage directory of the test is deleted.\n\nNote\n\nThis phase might be deferred in case a test has dependents (see [Cleaning up stage files](index.html#cleaning-up-stage-files) for more details).\n\n##### Execution Policies[¶](#execution-policies)\n\nAll regression tests in ReFrame will execute the pipeline stages described above.\nHowever, how exactly this pipeline will be executed is responsibility of the test execution policy.\nThere are two execution policies in ReFrame: the serial and the asynchronous execution policy.\n\nIn the serial execution policy, a new test gets into the pipeline after the previous one has exited.\nAs the figure below shows, this can lead to long idling times in the build and run phases, since the execution blocks until the associated test job finishes.\n\nThe serial execution policy.[¶](#id3)\n\nIn the asynchronous execution policy, multiple tests can be simultaneously on-the-fly.\nWhen a test enters the build or run phase, ReFrame does not block, but continues by picking the next test case to run.\nThis continues until no more test cases are left for execution or until a maximum concurrency limit is reached.\nAt the end, ReFrame enters a busy-wait loop monitoring the spawned test cases.\nAs soon as test case finishes, it resumes its pipeline and runs it to completion.\nThe following figure shows how the asynchronous execution policy works.\n\nThe asynchronous execution policy.[¶](#id4)\n\nReFrame tries to keep concurrency high by maintaining as many test cases as possible simultaneously active.\nWhen the [concurrency limit](config_reference.html#.systems[].partitions[].max_jobs) is reached, ReFrame will first try to free up execution slots by checking if any of the spawned jobs have finished, and it will fill that slots first before throttling execution.\n\nReFrame uses polling to check the status of the spawned jobs, but it does so in a dynamic way, in order to ensure both responsiveness and avoid overloading the system job scheduler with excessive polling.\n\nReFrame’s runtime internally encapsulates each test in a task, which is scheduled for execution.\nThis task can be in different states and is responsible for executing the test’s pipeline.\nThe following state diagram shows how test tasks are scheduled, as well as when the various test pipeline stages are executed.\n\nState diagram of the execution of test tasks with annotations for the execution of the actual pipeline stages.[¶](#id5)\n\nThere are a number of things to notice in this diagram:\n\n* If a test encounters an exception it is marked as a failure.\nEven normal failures, such as dependency failures and sanity or performance failures are also exceptions raised explicitly by the framework during a pipeline stage.\n* The pipeline stages that are executed asynchronously, namely the `compile` and `run` stages, are split in sub-stages for submitting the corresponding job and for checking or waiting its completion.\nThis is why in ReFrame error messages you may see `compile_complete` or `run_complete` being reported as the failing stage.\n* The execution of a test may be stalled if there are not enough execution slots available for submitting compile or run jobs on the target partition.\n* Although a test is officially marked as “completed” only when its cleanup phase is executed, it is reported as success or failure as soon as it is “retired,” i.e., as soon as its performance stage has passed successfully.\n* For successful tests, the `cleanup` stage is executed *after* the test is reported as a “success,” since a test may not clean up its resources until all of its immediate dependencies finish also successfully.\nIf the `cleanup` phase fails, the test is not marked as a failure, but this condition is marked as an error.\n\nChanged in version 3.10.0: The `compile` stage is now also executed asynchronously.\n\n###### Where each pipeline stage is executed?[¶](#where-each-pipeline-stage-is-executed)\n\nThere are two executions contexts where a pipeline stage can be executed: the ReFrame execution context and the partition execution context.\nThe *ReFrame execution context* is where ReFrame executes.\nThis is always the local host.\nThe *partition execution context* can either be local or remote depending on how the partition is configured.\nThe following table show in which context each pipeline stage executes:\n\n| Pipeline Stage | Execution Context |\n| --- | --- |\n| *Setup* | ReFrame |\n| *Compile* | ReFrame if [`build_locally`](index.html#reframe.core.pipeline.RegressionTest.build_locally) or [`local`](index.html#reframe.core.pipeline.RegressionTest.local) is [`True`](https://docs.python.org/3/library/constants.html#True), partition otherwise. |\n| *Run* | ReFrame if [`local`](index.html#reframe.core.pipeline.RegressionTest.local) is [`True`](https://docs.python.org/3/library/constants.html#True), partition otherwise. |\n| *Sanity* | ReFrame |\n| *Performance* | ReFrame |\n| *Cleanup* | ReFrame |\n\nIt should be noted that even if the partition execution context is local, it is treated differently from the ReFrame execution context.\nFor example, a test executing in the ReFrame context will not respect the [`max_jobs`](index.html#config.systems.partitions.max_jobs) partition configuration option, even if the partition is local.\nTo control the concurrency of the ReFrame execution context, users should set the [`max_local_jobs`](index.html#config.systems.max_local_jobs) option instead.\n\nChanged in version 3.10.0: Execution contexts were formalized.\n\n###### Tweaking the throughput and interactivity of test jobs in the asynchronous execution policy[¶](#tweaking-the-throughput-and-interactivity-of-test-jobs-in-the-asynchronous-execution-policy)\n\nReFrame’s asynchronous execution policy will iteratively cycle through all the in-flight tests and will try to advance the state (see state diagram above) of as many as possible within a given time slot.\nThe duration of this time slot is controlled by the [`pipeline_timeout`](index.html#config.general.pipeline_timeout) configuration option or the [`RFM_PIPELINE_TIMEOUT`](index.html#envvar-RFM_PIPELINE_TIMEOUT) environment variable.\nIf this timeout expires and at least one test has progressed, ReFrame will stop processing new tests in this time slot.\nIn the next time slot, it will try to further advance tests that have already started and if there is enough time left, it will also start new tests.\nEssentially, a small timeout value gives preference to tests that have already started, thus pushing them quicker down their pipeline, whereas higher values give preference to overall test throughput, as more tests will be running concurrently.\nThe default timeout is 10 seconds in order to balance interactivity and overall throughput.\n\nThere are cases when some tests take too long to proceed (e.g., due to copying of large files) and as a result they are blocking more tests from starting their pipeline.\nIn these cases, a higher timeout value will help to increase the test concurrency and therefore the overall throughput.\n\n##### Timing the Test Pipeline[¶](#timing-the-test-pipeline)\n\nNew in version 3.0.\n\nReFrame keeps track of the time a test spends in every pipeline stage and reports that after each test finishes.\nHowever, it does so from its own perspective and not from that of the scheduler backend used.\nThis has some practical implications:\nAs soon as a test enters the “run” phase, ReFrame’s timer for that phase starts ticking regardless if the associated job is pending.\nSimilarly, the “run” phase ends as soon as ReFrame realizes it.\nThis will happen after the associated job has finished.\nFor this reason, the time spent in the pipeline’s “run” phase should *not* be interpreted as the actual runtime of the test, especially if a non-local scheduler backend is used.\n\nFinally, the execution time of the “cleanup” phase is not reported when a test finishes, since it may be deferred in case that there exist tests that depend on that one.\nSee [How Test Dependencies Work In ReFrame](index.html#document-dependencies) for more information on how ReFrame treats tests with dependencies.\n\n#### How Test Dependencies Work In ReFrame[¶](#how-test-dependencies-work-in-reframe)\n\nDependencies in ReFrame are defined at the test level using the `depends_on()` function, but are projected to the [test cases](pipeline.html) space.\nWe will see the rules of that projection in a while.\nThe dependency graph construction and the subsequent dependency analysis happen also at the level of the test cases.\n\nLet’s assume that test `T1` depends on `T0`.\nThis can be expressed inside `T1` using the `depends_on()` method:\n```\n@rfm.simple_test class T0(rfm.RegressionTest):\n ...\n valid_systems = ['P0', 'P1']\n valid_prog_environs = ['E0', 'E1']\n\n@rfm.simple_test class T1(rfm.RegressionTest):\n ...\n valid_systems = ['P0', 'P1']\n valid_prog_environs = ['E0', 'E1']\n\n @run_after('init')\n def setup_deps(self):\n self.depends_on('T0')\n```\nConceptually, this dependency can be viewed at the test level as follows:\n\nSimple test dependency presented conceptually.[¶](#id2)\n\nFor most of the cases, this is sufficient to reason about test dependencies.\nIn reality, as mentioned above, dependencies are handled at the level of test cases.\nIf not specified differently, test cases on different partitions or programming environments are independent.\nThis is the default behavior of the `depends_on()` function.\nThe following image shows the actual test case dependencies of the two tests above:\n\nTest case dependencies partitioned by case (default).[¶](#id3)\n\nThis means that test cases of `T1` may start executing before all test cases of `T0` have finished.\nYou can impose a stricter dependency between tests, such that `T1` does not start execution unless all test cases of `T0` have finished.\nYou can achieve this as follows:\n```\nimport reframe.utility.udeps as udeps\n\n@rfm.simple_test class T1(rfm.RegressionTest):\n ...\n @run_after('init')\n def setup_deps(self):\n self.depends_on('T0', how=udeps.fully)\n```\nThis will create a fully connected graph between the test cases of the two tests as it is shown in the following figure:\n\nFully dependent test cases.[¶](#id4)\n\nThere are more options that the test case subgraph can be split than the two extremes we presented so far.\nThe following figures show the different splittings.\n\n##### Split by partition[¶](#split-by-partition)\n\nThe test cases are split in fully connected components per partition.\nTest cases from different partitions are independent.\n\nTest case dependencies partitioned by partition.[¶](#id5)\n\n##### Split by environment[¶](#split-by-environment)\n\nThe test cases are split in fully connected components per environment.\nTest cases from different environments are independent.\n\nTest case dependencies partitioned by environment.[¶](#id6)\n\n##### Split by exclusive partition[¶](#split-by-exclusive-partition)\n\nThe test cases are split in fully connected components that do not contain the same partition.\nTest cases from the same partition are independent.\n\nTest case dependencies partitioned by exclusive partition.[¶](#id7)\n\n##### Split by exclusive environment[¶](#split-by-exclusive-environment)\n\nThe test cases are split in fully connected components that do not contain the same environment.\nTest cases from the same environment are independent.\n\nTest case dependencies partitioned by exclusive environment.[¶](#id8)\n\n##### Split by exclusive case[¶](#split-by-exclusive-case)\n\nThe test cases are split in fully connected components that do not contain the same environment and the same partition.\nTest cases from the same environment and the same partition are independent.\n\nTest case dependencies partitioned by exclusive case.[¶](#id9)\n\n##### Custom splits[¶](#custom-splits)\n\nUsers may define custom dependency patterns by supplying their own `how` function.\nThe `how` argument accepts a `callable` which takes as arguments the source and destination of a possible edge in the test case subgraph.\nIf the callable returns `True`, then ReFrame will place an edge (i.e., a dependency) otherwise not.\nThe following code will create dependencies only if the source partition is `P0` and the destination environment is `E1`:\n```\ndef myway(src, dst):\n psrc, esrc = src\n pdst, edst = dst\n return psrc == 'P0' and edst == 'E1'\n\n@rfm.simple_test class T1(rfm.RegressionTest):\n ...\n @run_after('init')\n def setup_deps(self):\n self.depends_on('T0', how=myway)\n```\nThis corresponds to the following test case dependency subgraph:\n\nCustom test case dependencies.[¶](#id10)\n\nNotice how all the rest test cases are completely independent.\n\n##### Cyclic dependencies[¶](#cyclic-dependencies)\n\nObviously, cyclic dependencies between test cases are not allowed.\nCyclic dependencies between tests are not allowed either, even if the test case dependency graph is acyclic.\nFor example, the following dependency set up is invalid:\n\nThe test case dependencies here, clearly, do not form a cycle, but the edge from `(T0, P0, E0)` to `(T1, P0, E1)` introduces a dependency from `T0` to `T1` forming a cycle at the test level.\nIf you end up requiring such type of dependency in your tests, you might have to reconsider how you organize your tests.\n\nNote\n\nTechnically, the framework could easily support such types of dependencies, but ReFrame’s output would have to change substantially.\n\n##### Resolving dependencies[¶](#resolving-dependencies)\n\nAs shown in the [Tutorial 3: Using Dependencies in ReFrame Tests](index.html#document-tutorial_deps), test dependencies would be of limited usage if you were not able to use the results or information of the target tests.\nLet’s reiterate over the `set_executable()` function of the `OSULatencyTest` that we presented previously:\n```\n @require_deps\n def set_executable(self, OSUBuildTest):\n self.executable = os.path.join(\n OSUBuildTest().stagedir,\n 'mpi', 'pt2pt', 'osu_latency'\n )\n self.executable_opts = ['-x', '100', '-i', '1000']\n```\nThe `@require_deps` decorator does some magic – we will unravel this shortly – with the function arguments of the `set_executable()` function and binds them to the target test dependencies by their name.\nHowever, as discussed in this section, dependencies are defined at test case level, so the `OSUBuildTest` function argument is bound to a special function that allows you to retrieve an actual test case of the target dependency.\nThis is why you need to “call” `OSUBuildTest` in order to retrieve the desired test case.\nWhen no arguments are passed, this will retrieve the test case corresponding to the current partition and the current programming environment.\nWe could always retrieve the `PrgEnv-gnu` case by writing `OSUBuildTest('PrgEnv-gnu')`.\nIf a dependency cannot be resolved, because it is invalid, a runtime error will be thrown with an appropriate message.\n\nThe low-level method for retrieving a dependency is the [`getdep()`](index.html#reframe.core.pipeline.RegressionTest.getdep) method of the `RegressionTest`.\nIn fact, you can rewrite `set_executable()` function as follows:\n```\n@run_after('setup')\ndef set_executable(self):\n target = self.getdep('OSUBuildTest')\n self.executable = os.path.join(\n target.stagedir,\n 'osu-micro-benchmarks-5.6.2', 'mpi', 'pt2pt', 'osu_latency'\n )\n self.executable_opts = ['-x', '100', '-i', '1000']\n```\nNow it’s easier to understand what the `@require_deps` decorator does behind the scenes.\nIt binds the function arguments to a partial realization of the `getdep()` function and attaches the decorated function as an after-setup hook.\nIn fact, any `@require_deps`-decorated function will be invoked before any other after-setup hook.\n\n##### Cleaning up stage files[¶](#cleaning-up-stage-files)\n\nIn principle, the output of a test might be needed by its dependent tests.\nAs a result, the stage directory of the test will only be cleaned up after all of its *immediate* dependent tests have finished successfully.\nIf any of its children has failed, the cleanup phase will be skipped, such that all the test’s files will remain in the stage directory.\nThis allows users to reproduce manually the error of a failed test with dependencies, since all the needed resources of the failing test are left in their original location.\n\n#### Understanding the Mechanism of Deferrable Functions[¶](#understanding-the-mechanism-of-deferrable-functions)\n\nThis section describes the mechanism behind deferrable functions, which in ReFrame, they are used for sanity and performance checking.\nGenerally, writing a new sanity function in a [`RegressionTest`](index.html#reframe.core.pipeline.RegressionTest) is as straightforward as decorating a simple member function with the built-in `sanity_function()` decorator.\nBehind the scenes, this decorator will convert the Python function into a deferrable function and schedule its evaluation for the sanity stage of the test.\nHowever, when dealing with more complex scenarios such as a deferrable function taking as an argument the results from other deferrable functions, it is crucial to understand how a deferrable function differs from a regular Python function, and when is it actually evaluated.\n\n##### What Is a Deferrable Function?[¶](#what-is-a-deferrable-function)\n\nA deferrable function is a function whose a evaluation is deferred to a later point in time.\nYou can define any function as deferrable by wrapping it with the `deferrable()` when decorating a member function of a class derived from [`RegressionMixin`](index.html#reframe.core.pipeline.RegressionMixin), or alternatively, the [`reframe.utility.sanity.deferrable()`](index.html#reframe.utility.sanity.deferrable) decorator can be used for any other function.\nThe example below demonstrates a simple scenario:\n```\nimport reframe.utility.sanity as sn\n\n@sn.deferrable def foo():\n print('hello')\n```\nIf you try to call `foo()`, its code will not execute:\n```\n>>> foo()\n<reframe.core.deferrable._DeferredExpression object at 0x2b70fff23550>\n```\nInstead, a special object is returned that represents the function whose execution is deferred.\nNotice the more general *deferred expression* name of this object. We shall see later on why this name is used.\n\nIn order to explicitly trigger the execution of `foo()`, you have to call [`evaluate`](index.html#reframe.utility.sanity.evaluate) on it:\n```\n>>> from reframe.utility.sanity import evaluate\n>>> evaluate(foo())\nhello\n```\nIf the argument passed to [`evaluate`](index.html#reframe.utility.sanity.evaluate) is not a deferred expression, it will be simply returned as is.\n\nDeferrable functions may also be combined as we do with normal functions. Let’s extend our example with `foo()` accepting an argument and printing it:\n```\nimport reframe.utility.sanity as sn\n\n@sn.deferrable def foo(arg):\n print(arg)\n\n@sn.deferrable def greetings():\n return 'hello'\n```\nIf we now do `foo(greetings())`, again nothing will be evaluated:\n```\n>>> foo(greetings())\n<reframe.core.deferrable._DeferredExpression object at 0x2b7100e9e978>\n```\nIf we trigger the evaluation of `foo()` as before, we will get expected result:\n```\n>>> evaluate(foo(greetings()))\nhello\n```\nNotice how the evaluation mechanism goes down the function call graph and returns the expected result.\nAn alternative way to evaluate this expression would be the following:\n```\n>>> x = foo(greetings())\n>>> x.evaluate()\nhello\n```\nAs you may have noticed, you can assign a deferred function to a variable and evaluate it later.\nYou may also do `evaluate(x)`, which is equivalent to `x.evaluate()`.\n\nTo demonstrate more clearly how the deferred evaluation of a function works, let’s consider the following `size3()` deferrable function that simply checks whether an `iterable` passed as argument has three elements inside it:\n```\n@sn.deferrable def size3(iterable):\n return len(iterable) == 3\n```\nNow let’s assume the following example:\n```\n>>> l = [1, 2]\n>>> x = size3(l)\n>>> evaluate(x)\nFalse\n>>> l += [3]\n>>> evaluate(x)\nTrue\n```\nWe first call `size3()` and store its result in `x`.\nAs expected when we evaluate `x`, `False` is returned, since at the time of the evaluation our list has two elements.\nWe later append an element to our list and reevaluate `x` and we get `True`, since at this point the list has three elements.\n\nNote\n\nDeferred functions and expressions may be stored and (re)evaluated at any later point in the program.\n\nAn important thing to point out here is that deferrable functions *capture* their arguments at the point they are called.\nIf you change the binding of a variable name (either explicitly or implicitly by applying an operator to an immutable object), this change will not be reflected when you evaluate the deferred function.\nThe function instead will operate on its captured arguments.\nWe will demonstrate this by replacing the list in the above example with a tuple:\n```\n>>> l = (1, 2)\n>>> x = size3(l)\n>>> l += (3,)\n>>> l\n(1, 2, 3)\n>>> evaluate(x)\nFalse\n```\nWhy this is happening?\nThis is because tuples are immutable so when we are doing `l += (3,)` to append to our tuple, Python constructs a new tuple and rebinds `l` to the newly created tuple that has three elements.\nHowever, when we called our deferrable function, `l` was pointing to a different tuple object, and that was the actual tuple argument that our deferrable function has captured.\n\nThe following augmented example demonstrates this:\n```\n>>> l = (1, 2)\n>>> x = size3(l)\n>>> l += (3,)\n>>> l\n(1, 2, 3)\n>>> evaluate(x)\nFalse\n>>> l = (1, 2)\n>>> id(l)\n47764346657160\n>>> x = size3(l)\n>>> l += (3,)\n>>> id(l)\n47764330582232\n>>> l\n(1, 2, 3)\n>>> evaluate(x)\nFalse\n```\nNotice the different IDs of `l` before and after the `+=` operation.\nThis a key trait of deferrable functions and expressions that you should be aware of.\n\n##### Deferred expressions[¶](#deferred-expressions)\n\nYou might be still wondering why the internal name of a deferred function refers to the more general term deferred expression.\nHere is why:\n```\n>>> @sn.deferrable\n... def size(iterable):\n... return len(iterable)\n...\n>>> l = [1, 2]\n>>> x = 2*(size(l) + 3)\n>>> x\n<reframe.core.deferrable._DeferredExpression object at 0x2b1288f4e940>\n>>> evaluate(x)\n10\n```\nAs you can see, you can use the result of a deferred function inside arithmetic operations.\nThe result will be another deferred expression that you can evaluate later.\nYou can practically use any Python builtin operator or builtin function with a deferred expression and the result will be another deferred expression.\nThis is quite a powerful mechanism, since with the standard syntax you can create arbitrary expressions that may be evaluated later in your program.\n\nThere are some exceptions to this rule, though.\nThe logical [`and`](https://docs.python.org/3/reference/expressions.html#and), [`or`](https://docs.python.org/3/reference/expressions.html#or) and [`not`](https://docs.python.org/3/reference/expressions.html#not) operators as well as the [`in`](https://docs.python.org/3/reference/expressions.html#in) operator cannot be deferred automatically.\nThese operators try to take the truthy value of their arguments by calling `bool` on them.\nAs we shall see later, applying the `bool` function on a deferred expression causes its immediate evaluation and returns the result.\nIf you want to defer the execution of such operators, you should use the corresponding [`and_`](index.html#reframe.utility.sanity.and_), [`or_`](index.html#reframe.utility.sanity.or_), [`not_`](index.html#reframe.utility.sanity.not_) and [`contains`](index.html#reframe.utility.sanity.contains) functions in [`reframe.utility.sanity`](index.html#module-reframe.utility.sanity), which basically wrap the expression in a deferrable function.\n\nIn summary deferrable functions have the following characteristics:\n\n* You can make any function deferrable by wrapping it with the [`deferrable()`](index.html#reframe.utility.sanity.deferrable) decorator.\n* When you call a deferrable function, its body is not executed but its arguments are *captured* and an object representing the deferred function is returned.\n* You can execute the body of a deferrable function at any later point by calling [`evaluate`](index.html#reframe.utility.sanity.evaluate) on the deferred expression object that it has been returned by the call to the deferred function.\n* Deferred functions can accept other deferred expressions as arguments and may also return a deferred expression.\n* When you evaluate a deferrable function, any other deferrable function down the call tree will also be evaluated.\n* You can include a call to a deferrable function in any Python expression and the result will be another deferred expression.\n\n##### How a Deferred Expression Is Evaluated?[¶](#how-a-deferred-expression-is-evaluated)\n\nAs discussed before, you can create a new deferred expression by calling a function whose definition is decorated by the `@deferrable` decorator or by including an already deferred expression in any sort of arithmetic operation.\nWhen you call [`evaluate`](index.html#reframe.utility.sanity.evaluate) on a deferred expression, you trigger the evaluation of the whole subexpression tree.\nHere is how the evaluation process evolves:\n\nA deferred expression object is merely a placeholder of the target function and its arguments at the moment you call it.\nDeferred expressions leverage also the Python’s data model so as to capture all the binary and unary operators supported by the language.\nWhen you call `evaluate()` on a deferred expression object, the stored function will be called passing it the captured arguments.\nIf any of the arguments is a deferred expression, it will be evaluated too.\nIf the return value of the deferred expression is also a deferred expression, it will be evaluated as well.\n\nThis last property lets you call other deferrable functions from inside a deferrable function.\nHere is an example where we define two deferrable variations of the builtins `sum` and [`len`](https://docs.python.org/3/library/functions.html#len) and another deferrable function `avg()` that computes the average value of the elements of an iterable by calling our deferred builtin alternatives.\n```\n@sn.deferrable def dsum(iterable):\n return sum(iterable)\n\n@sn.deferrable def dlen(iterable):\n return len(iterable)\n\n@sn.deferrable def avg(iterable):\n return dsum(iterable) / dlen(iterable)\n```\nIf you try to evaluate `avg()` with a list, you will get the expected result:\n```\n>>> avg([1, 2, 3, 4])\n<reframe.core.deferrable._DeferredExpression object at 0x2b1288f54b70>\n>>> evaluate(avg([1, 2, 3, 4]))\n2.5\n```\nThe return value of `evaluate(avg())` would normally be a deferred expression representing the division of the results of the other two deferrable functions.\nHowever, the evaluation mechanism detects that the return value is a deferred expression and it automatically triggers its evaluation, yielding the expected result.\nThe following figure shows how the evaluation evolves for this particular example:\n\nSequence diagram of the evaluation of the deferrable `avg()` function.[¶](#id1)\n\n##### Implicit evaluation of a deferred expression[¶](#implicit-evaluation-of-a-deferred-expression)\n\nAlthough you can trigger the evaluation of a deferred expression at any time by calling `evaluate`, there are some cases where the evaluation is triggered implicitly:\n\n* When you try to get the truthy value of a deferred expression by calling `bool` on it.\nThis happens for example when you include a deferred expression in an [`if`](https://docs.python.org/3/reference/compound_stmts.html#if) statement or as an argument to the [`and`](https://docs.python.org/3/reference/expressions.html#and), [`or`](https://docs.python.org/3/reference/expressions.html#or), [`not`](https://docs.python.org/3/reference/expressions.html#not) and [`in`](https://docs.python.org/3/reference/expressions.html#in) (`__contains__`) operators.\nThe following example demonstrates this behavior:\n```\n>>> if avg([1, 2, 3, 4]) > 2:\n... print('hello')\n...\nhello\n```\nThe expression `avg([1, 2, 3, 4]) > 2` is a deferred expression, but its evaluation is triggered from the Python interpreter by calling the `bool()` method on it, in order to evaluate the [`if`](https://docs.python.org/3/reference/compound_stmts.html#if) statement.\nA similar example is the following that demonstrates the behaviour of the [`in`](https://docs.python.org/3/reference/expressions.html#in) operator:\n```\n>>> from reframe.utility.sanity import defer\n>>> l = defer([1, 2, 3])\n>>> l\n<reframe.core.deferrable._DeferredExpression object at 0x2b1288f54cf8>\n>>> evaluate(l)\n[1, 2, 3]\n>>> 4 in l False\n>>> 3 in l True\n```\nThe [`defer`](index.html#reframe.utility.sanity.defer) is simply a deferrable version of the identity function (a function that simply returns its argument).\nAs expected, `l` is a deferred expression that evaluates to the `[1, 2, 3]` list. When we apply the [`in`](https://docs.python.org/3/reference/expressions.html#in) operator, the deferred expression is immediately evaluated.\n\nNote\n\nPython expands this expression into `bool(l.__contains__(3))`.\nAlthough `__contains__` is also defined as a deferrable function in `_DeferredExpression`, its evaluation is triggered by the `bool` builtin.\n* When you try to iterate over a deferred expression by calling the [`iter`](https://docs.python.org/3/library/functions.html#iter) function on it.\nThis call happens implicitly by the Python interpreter when you try to iterate over a container.\nHere is an example:\n```\n>>> @sn.deferrable\n... def getlist(iterable):\n... ret = list(iterable)\n... ret += [1, 2, 3]\n... return ret\n>>> getlist([1, 2, 3])\n<reframe.core.deferrable._DeferredExpression object at 0x2b1288f54dd8>\n>>> for x in getlist([1, 2, 3]):\n... print(x)\n...\n1 2\n3 1\n2 3\n```\nSimply calling `getlist()` will not execute anything and a deferred expression object will be returned.\nHowever, when you try to iterate over the result of this call, then the deferred expression will be evaluated immediately.\n* When you try to call `str` on a deferred expression.\nThis will be called by the Python interpreter every time you try to print this expression.\nHere is an example with the `getlist` deferrable function:\n```\n>>> print(getlist([1, 2, 3]))\n[1, 2, 3, 1, 2, 3]\n```\n##### How to Write a Deferrable Function?[¶](#how-to-write-a-deferrable-function)\n\nThe answer is simple:\nlike you would with any other normal function!\nWe’ve done that already in all the examples we’ve shown in this documentation.\nA question that somehow naturally comes up here is whether you can call a deferrable function from within a deferrable function, since this doesn’t make a lot of sense:\nafter all, your function will be deferred anyway.\n\nThe answer is, yes.\nYou can call other deferrable functions from within a deferrable function.\nThanks to the implicit evaluation rules as well as the fact that the return value of a deferrable function is also evaluated if it is a deferred expression, you can write a deferrable function without caring much about whether the functions you call are themselves deferrable or not.\nHowever, you should be aware of passing mutable objects to deferrable functions.\nIf these objects happen to change between the actual call and the implicit evaluation of the deferrable function, you might run into surprises.\nIn any case, if you want the immediate evaluation of a deferrable function or expression, you can always do that by calling [`evaluate`](index.html#reframe.utility.sanity.evaluate) on it.\n\nThe following example demonstrates two different ways writing a deferrable function that checks the average of the elements of an iterable:\n```\nimport reframe.utility.sanity as sn\n\n@sn.deferrable def check_avg_with_deferrables(iterable):\n avg = sn.sum(iterable) / sn.len(iterable)\n return -1 if avg > 2 else 1\n\n@sn.deferrable def check_avg_without_deferrables(iterable):\n avg = sum(iterable) / len(iterable)\n return -1 if avg > 2 else 1\n```\n```\n>>> evaluate(check_avg_with_deferrables([1, 2, 3, 4]))\n-1\n>>> evaluate(check_avg_without_deferrables([1, 2, 3, 4]))\n-1\n```\nThe first version uses the [`sum`](index.html#reframe.utility.sanity.sum) and [`len`](index.html#reframe.utility.sanity.len) functions from [`reframe.utility.sanity`](index.html#module-reframe.utility.sanity), which are deferrable versions of the corresponding builtins.\nThe second version uses directly the builtin [`sum`](https://docs.python.org/3/library/functions.html#sum) and [`len`](https://docs.python.org/3/library/functions.html#len) functions.\nAs you can see, both of them behave in exactly the same way.\nIn the version with the deferrables, `avg` is a deferred expression but it is evaluated by the [`if`](https://docs.python.org/3/reference/compound_stmts.html#if) statement before returning.\n\nGenerally, inside a sanity function, it is a preferable to use the non-deferrable version of a function, if that exists, since you avoid the extra overhead and bookkeeping of the deferring mechanism.\n\n##### Ready to Go Deferrable Functions[¶](#ready-to-go-deferrable-functions)\n\nNormally, you will not have to implement your own deferrable functions, since ReFrame provides already a variety of them.\nYou can find the complete list of provided sanity functions in [Deferrable Functions Reference](index.html#deferrable-functions).\n\n##### Deferrable functions vs Generators[¶](#deferrable-functions-vs-generators)\n\nPython allows you to create functions that will be evaluated lazily.\nThese are called [generator functions](https://wiki.python.org/moin/Generators).\nTheir key characteristic is that instead of using the [`return`](https://docs.python.org/3/reference/simple_stmts.html#return) keyword to return values, they use the [`yield`](https://docs.python.org/3/reference/simple_stmts.html#yield) keyword.\nI’m not going to go into the details of the generators, since there is plenty of documentation out there, so I will focus on the similarities and differences with our deferrable functions.\n\n###### Similarities[¶](#similarities)\n\n* Both generators and our deferrables return an object representing the deferred expression when you call them.\n* Both generators and deferrables may be evaluated explicitly or implicitly when they appear in certain expressions.\n* When you try to iterate over a generator or a deferrable, you trigger its evaluation.\n\n###### Differences[¶](#differences)\n\n* You can include deferrables in any arithmetic expression and the result will be another deferrable expression.\nThis is not true with generator functions, which will raise a [`TypeError`](https://docs.python.org/3/library/exceptions.html#TypeError) in such cases or they will always evaluate to `False` if you include them in boolean expressions Here is an example demonstrating this:\n```\n>>> @sn.deferrable\n... def dsize(iterable):\n... print(len(iterable))\n... return len(iterable)\n...\n>>> def gsize(iterable):\n... print(len(iterable))\n... yield len(iterable)\n...\n>>> l = [1, 2]\n>>> dsize(l)\n<reframe.core.deferrable._DeferredExpression object at 0x2abc630abb38>\n>>> gsize(l)\n<generator object gsize at 0x2abc62a4bf10>\n>>> expr = gsize(l) == 2\n>>> expr False\n>>> expr = gsize(l) + 2 Traceback (most recent call last):\n File \"<stdin>\", line 1, in <module>\nTypeError: unsupported operand type(s) for +: 'generator' and 'int'\n>>> expr = dsize(l) == 2\n>>> expr\n<reframe.core.deferrable._DeferredExpression object at 0x2abc630abba8>\n>>> expr = dsize(l) + 2\n>>> expr\n<reframe.core.deferrable._DeferredExpression object at 0x2abc630abc18>\n```\nNotice that you cannot include generators in expressions, whereas you can generate arbitrary expressions with deferrables.\n\n* Generators are iterator objects, while deferred expressions are not.\nAs a result, you can trigger the evaluation of a generator expression using the [`next`](https://docs.python.org/3/library/functions.html#next) builtin function.\nFor a deferred expression you should use [`evaluate`](index.html#reframe.utility.sanity.evaluate) instead.\n* A generator object is iterable, whereas a deferrable object will be iterable if and only if the result of its evaluation is iterable.\n\nNote\n\nTechnically, a deferrable object is iterable, too, since it provides the `__iter__` method.\nThat’s why you can include it in iteration expressions. However, it delegates this call to the result of its evaluation.\n\nHere is an example demonstrating this difference:\n```\n>>> for i in gsize(l): print(i)\n...\n2 2\n>>> for i in dsize(l): print(i)\n...\n2 Traceback (most recent call last):\n File \"<stdin>\", line 1, in <module>\n File \"/users/karakasv/Devel/reframe/reframe/core/deferrable.py\", line 73, in __iter__\n return iter(self.evaluate())\nTypeError: 'int' object is not iterable\n```\nNotice how the iteration works fine with the generator object, whereas with the deferrable function, the iteration call is delegated to the result of the evaluation, which is not an iterable, therefore yielding [`TypeError`](https://docs.python.org/3/library/exceptions.html#TypeError).\nNotice also, the printout of `2` in the iteration over the deferrable expression, which shows that it has been evaluated.\n\n### ReFrame Manuals[¶](#reframe-manuals)\n\n#### Command Line Reference[¶](#command-line-reference)\n\n##### Synopsis[¶](#synopsis)\n\nreframe [OPTION]... ACTION[¶](#cmdoption-arg-reframe)\n\n##### Description[¶](#description)\n\nReFrame provides both a [programming interface](regression_test_api.html) for writing regression tests and a command-line interface for managing and running the tests, which is detailed here.\nThe `reframe` command is part of ReFrame’s frontend.\nThis frontend is responsible for loading and running regression tests written in ReFrame.\nReFrame executes tests by sending them down to a well defined pipeline.\nThe implementation of the different stages of this pipeline is part of ReFrame’s core architecture, but the frontend is responsible for driving this pipeline and executing tests through it.\nThere are three basic phases that the frontend goes through, which are described briefly in the following.\n\n###### Test discovery and test loading[¶](#test-discovery-and-test-loading)\n\nThis is the very first phase of the frontend.\nReFrame will search for tests in its *check search path* and will load them.\nWhen ReFrame loads a test, it actually *instantiates* it, meaning that it will call its `__init__()` method unconditionally whether this test is meant to run on the selected system or not.\nThis is something that writers of regression tests should bear in mind.\n\n-c, --checkpath=PATH[¶](#cmdoption-c)\nA filesystem path where ReFrame should search for tests.\n\n`PATH` can be a directory or a single test file.\nIf it is a directory, ReFrame will search for test files inside this directory load all tests found in them.\nThis option can be specified multiple times, in which case each `PATH` will be searched in order.\n\nThe check search path can also be set using the [`RFM_CHECK_SEARCH_PATH`](#envvar-RFM_CHECK_SEARCH_PATH) environment variable or the [`check_search_path`](index.html#config.general.check_search_path) general configuration parameter.\n\n-R, --recursive[¶](#cmdoption-R)\nSearch for test files recursively in directories found in the check search path.\n\nThis option can also be set using the [`RFM_CHECK_SEARCH_RECURSIVE`](#envvar-RFM_CHECK_SEARCH_RECURSIVE) environment variable or the [`check_search_recursive`](index.html#config.general.check_search_recursive) general configuration parameter.\n\n###### Test filtering[¶](#test-filtering)\n\nAfter all tests in the search path have been loaded, they are first filtered by the selected system.\nAny test that is not valid for the current system, it will be filtered out.\nThe current system is either auto-selected or explicitly specified with the [`--system`](#cmdoption-system) option.\nTests can be filtered by different attributes and there are specific command line options for achieving this.\nA common characteristic of all test filtering options is that if a test is selected, then all its dependencies will be selected, too, regardless if they match the filtering criteria or not.\nThis happens recursively so that if test `T1` depends on `T2` and `T2` depends on `T3`, then selecting `T1` would also select `T2` and `T3`.\n\n--cpu-only[¶](#cmdoption-cpu-only)\nSelect tests that do not target GPUs.\n\nThese are all tests with `num_gpus_per_node` equals to zero This option and [`--gpu-only`](#cmdoption-gpu-only) are mutually exclusive.\n\nThe [`--gpu-only`](#cmdoption-gpu-only) and [`--cpu-only`](#cmdoption-cpu-only) check only the value of the `num_gpus_per_node` attribute of tests.\nThe value of this attribute is not required to be non-zero for GPU tests.\nTests may or may not make use of it.\n\n--failed[¶](#cmdoption-failed)\nSelect only the failed test cases for a previous run.\n\nThis option can only be used in combination with the [`--restore-session`](#cmdoption-restore-session).\nTo rerun the failed cases from the last run, you can use `reframe --restore-session --failed -r`.\n\nNew in version 3.4.\n\n--gpu-only[¶](#cmdoption-gpu-only)\nSelect tests that can run on GPUs.\n\nThese are all tests with `num_gpus_per_node` greater than zero.\nThis option and [`--cpu-only`](#cmdoption-cpu-only) are mutually exclusive.\n\n--maintainer=MAINTAINER[¶](#cmdoption-maintainer)\nFilter tests by maintainer.\n\n`MAINTAINER` is interpreted as a [Python Regular Expression](https://docs.python.org/3/library/re.html); all tests that have at least a matching maintainer will be selected.\n`MAINTAINER` being a regular expression has the implication that `--maintainer 'foo'` will select also tests that define `'foobar'` as a maintainer.\nTo restrict the selection to tests defining only `'foo'`, you should use `--maintainer 'foo$'`.\n\nThis option may be specified multiple times, in which case only tests defining or matching *all* maintainers will be selected.\n\nNew in version 3.9.1.\n\nChanged in version 4.1.0: The `MAINTAINER` pattern is matched anywhere in the maintainer’s name and not at its beginning.\nIf you want to match at the beginning of the name, you should prepend `^`.\n\n-n, --name=NAME[¶](#cmdoption-n)\nFilter tests by name.\n\n`NAME` is interpreted as a [Python Regular Expression](https://docs.python.org/3/library/re.html);\nany test whose *display name* matches `NAME` will be selected.\nThe display name of a test encodes also any parameterization information.\nSee [Test Naming Scheme](#test-naming-scheme) for more details on how the tests are automatically named by the framework.\n\nBefore matching, any whitespace will be removed from the display name of the test.\n\nThis option may be specified multiple times, in which case tests with *any* of the specified names will be selected:\n`-n NAME1 -n NAME2` is therefore equivalent to `-n 'NAME1|NAME2'`.\n\nIf the special notation `<test_name>@<variant_num>` is passed as the `NAME` argument, then an exact match will be performed selecting the variant `variant_num` of the test `test_name`.\n\nYou may also select a test by its hash code using the notation `/<test-hash>` for the `NAME` argument.\n\nNote\n\nFixtures cannot be selected.\n\nChanged in version 3.10.0: The option’s behaviour was adapted and extended in order to work with the updated test naming scheme.\n\nChanged in version 4.0.0: Support selecting tests by their hash code.\n\nChanged in version 4.1.0: The `NAME` pattern is matched anywhere in the test name and not at its beginning.\nIf you want to match at the beginning of a test name, you should prepend `^`.\n\n-p, --prgenv=NAME[¶](#cmdoption-p)\nFilter tests by programming environment.\n\n`NAME` is interpreted as a [Python Regular Expression](https://docs.python.org/3/library/re.html);\nany test for which at least one valid programming environment is matching `NAME` will be selected.\n\nThis option may be specified multiple times, in which case only tests matching all of the specified programming environments will be selected.\n\n--skip-prgenv-check[¶](#cmdoption-skip-prgenv-check)\nDo not filter tests against programming environments.\n\nEven if the [`-p`](#cmdoption-p) option is not specified, ReFrame will filter tests based on the programming environments defined for the currently selected system.\nThis option disables that filter completely.\n\n--skip-system-check[¶](#cmdoption-skip-system-check)\nDo not filter tests against the selected system.\n\n-T, --exclude-tag=TAG[¶](#cmdoption-T)\nExclude tests by tags.\n\n`TAG` is interpreted as a [Python Regular Expression](https://docs.python.org/3/library/re.html);\nany test with tags matching `TAG` will be excluded.\n\nThis option may be specified multiple times, in which case tests with *any* of the specified tags will be excluded:\n`-T TAG1 -T TAG2` is therefore equivalent to `-T 'TAG1|TAG2'`.\n\nChanged in version 4.1.0: The `TAG` pattern is matched anywhere in the tag name and not at its beginning.\nIf you want to match at the beginning of a tag, you should prepend `^`.\n\n-t, --tag=TAG[¶](#cmdoption-0)\nFilter tests by tag.\n\n`TAG` is interpreted as a [Python Regular Expression](https://docs.python.org/3/library/re.html); all tests that have at least a matching tag will be selected.\n`TAG` being a regular expression has the implication that `-t 'foo'` will select also tests that define `'foobar'` as a tag.\nTo restrict the selection to tests defining only `'foo'`, you should use `-t 'foo$'`.\n\nThis option may be specified multiple times, in which case only tests defining or matching *all* tags will be selected.\n\nChanged in version 4.1.0: The `TAG` pattern is matched anywhere in the tag name and not at its beginning.\nIf you want to match at the beginning of a tag, you should prepend `^`.\n\n-x, --exclude=NAME[¶](#cmdoption-x)\nExclude tests by name.\n\n`NAME` is interpreted as a [Python Regular Expression](https://docs.python.org/3/library/re.html);\nany test whose name matches `NAME` will be excluded.\n\nThis option may be specified multiple times, in which case tests with *any* of the specified names will be excluded:\n`-x NAME1 -x NAME2` is therefore equivalent to `-x 'NAME1|NAME2'`.\n\nChanged in version 4.1.0: The `NAME` pattern is matched anywhere in the test name and not at its beginning.\nIf you want to match at the beginning of a test name, you should prepend `^`.\n\n###### Test actions[¶](#test-actions)\n\nReFrame will finally act upon the selected tests.\nThere are currently two actions that can be performed on tests: (a) list the tests and (b) execute the tests.\nAn action must always be specified.\n\n--ci-generate=FILE[¶](#cmdoption-ci-generate)\nDo not run the tests, but generate a Gitlab [child pipeline](https://docs.gitlab.com/ee/ci/parent_child_pipelines.html) specification in `FILE`.\n\nYou can set up your Gitlab CI to use the generated file to run every test as a separate Gitlab job respecting test dependencies.\nFor more information, have a look in [Integrating into a CI pipeline](index.html#generate-ci-pipeline).\n\nNew in version 3.4.1.\n\n--describe[¶](#cmdoption-describe)\nPrint a detailed description of the [selected tests](#test-filtering) in JSON format and exit.\n\nNote\n\nThe generated test description corresponds to its state after it has been initialized.\nIf any of its attributes are changed or set during its execution, their updated values will not be shown by this listing.\n\nNew in version 3.10.0.\n\n--dry-run[¶](#cmdoption-dry-run)\nDry run the selected tests.\n\nThe dry-run mode will try to execute as much of the test pipeline as possible.\nMore specifically, the tests will not be submitted and will not be run for real,\nbut their stage directory will be prepared and the corresponding job script will be emitted.\nSimilarly, the sanity and performance functions will not be evaluated but all the preparation will happen.\nTests run in dry-run mode will not fail unless there is a programming error in the test or if the test tries to use a resource that is not produced in dry run mode (e.g., access the standard output or a resource produced by a dependency outside any sanity or performance function).\nIn this case, users can call the [`is_dry_run()`](index.html#reframe.core.pipeline.RegressionTest.is_dry_run) method in their test and take a specific action if the test is run in dry-run mode.\n\nNew in version 4.1.\n\n-L, --list-detailed[=T|C][¶](#cmdoption-L)\nList selected tests providing more details for each test.\n\nThe unique id of each test (see also [`unique_name`](index.html#reframe.core.pipeline.RegressionTest.unique_name)) as well as the file where each test is defined are printed.\n\nThis option accepts optionally a single argument denoting what type of listing is requested.\nPlease refer to [`-l`](#cmdoption-1) for an explanation of this argument.\n\nNew in version 3.10.0: Support for different types of listing is added.\n\nChanged in version 4.0.5: The variable names to which fixtures are bound are also listed.\nSee [Test Naming Scheme](#test-naming-scheme) for more information.\n\n-l, --list[=T|C][¶](#cmdoption-1)\nList selected tests and their dependencies.\n\nThis option accepts optionally a single argument denoting what type of listing is requested.\nThere are two types of possible listings:\n\n* *Regular test listing* (`T`, the default): This type of listing lists the tests and their dependencies or fixtures using their [`display_name`](index.html#reframe.core.pipeline.RegressionTest.display_name). A test that is listed as a dependency of another test will not be listed separately.\n* *Concretized test case listing* (`C`): This type of listing lists the exact test cases and their dependencies as they have been concretized for the current system and environment combinations.\nThis listing shows practically the exact test DAG that will be executed.\n\nNew in version 3.10.0: Support for different types of listing is added.\n\nChanged in version 4.0.5: The variable names to which fixtures are bound are also listed.\nSee [Test Naming Scheme](#test-naming-scheme) for more information.\n\n--list-tags[¶](#cmdoption-list-tags)\nList the unique tags of the selected tests.\n\nThe tags are printed in alphabetical order.\n\nNew in version 3.6.0.\n\n-r, --run[¶](#cmdoption-2)\nExecute the selected tests.\n\nIf more than one action options are specified, the precedence order is the following:\n\n> ```\n> --describe > --list-detailed > --list > --list-tags > --ci-generate\n> ```\n###### Options controlling ReFrame output[¶](#options-controlling-reframe-output)\n\n--compress-report[¶](#cmdoption-compress-report)\nCompress the generated run report (see [`--report-file`](#cmdoption-report-file)).\nThe generated report is a JSON file formatted in a human readable form.\nIf this option is enabled, the generated JSON file will be a single stream of text without additional spaces or new lines.\n\nThis option can also be set using the [`RFM_COMPRESS_REPORT`](#envvar-RFM_COMPRESS_REPORT) environment variable or the [`compress_report`](index.html#config.general.compress_report) general configuration parameter.\n\nNew in version 3.12.0.\n\n--dont-restage[¶](#cmdoption-dont-restage)\nDo not restage a test if its stage directory exists.\nNormally, if the stage directory of a test exists, ReFrame will remove it and recreate it.\nThis option disables this behavior.\n\nThis option can also be set using the [`RFM_CLEAN_STAGEDIR`](#envvar-RFM_CLEAN_STAGEDIR) environment variable or the [`clean_stagedir`](index.html#config.general.clean_stagedir) general configuration parameter.\n\nNew in version 3.1.\n\n--keep-stage-files[¶](#cmdoption-keep-stage-files)\nKeep test stage directories even for tests that finish successfully.\n\nThis option can also be set using the [`RFM_KEEP_STAGE_FILES`](#envvar-RFM_KEEP_STAGE_FILES) environment variable or the [`keep_stage_files`](index.html#config.general.keep_stage_files) general configuration parameter.\n\n-o, --output=DIR[¶](#cmdoption-o)\nDirectory prefix for test output files.\n\nWhen a test finishes successfully, ReFrame copies important output files to a test-specific directory for future reference.\nThis test-specific directory is of the form `{output_prefix}/{system}/{partition}/{environment}/{test_name}`,\nwhere `output_prefix` is set by this option.\nThe test files saved in this directory are the following:\n\n* The ReFrame-generated build script, if not a run-only test.\n* The standard output and standard error of the build phase, if not a run-only test.\n* The ReFrame-generated job script, if not a compile-only test.\n* The standard output and standard error of the run phase, if not a compile-only test.\n* Any additional files specified by the `keep_files` regression test attribute.\n\nThis option can also be set using the [`RFM_OUTPUT_DIR`](#envvar-RFM_OUTPUT_DIR) environment variable or the `outputdir` system configuration parameter.\n\n--perflogdir=DIR[¶](#cmdoption-perflogdir)\nDirectory prefix for logging performance data.\n\nThis option is relevant only to the `filelog` [logging handler](config_reference.html#the-filelog-log-handler).\n\nThis option can also be set using the [`RFM_PERFLOG_DIR`](#envvar-RFM_PERFLOG_DIR) environment variable or the [`basedir`](index.html#config.logging.handlers_perflog..filelog..basedir) logging handler configuration parameter.\n\n--prefix=DIR[¶](#cmdoption-prefix)\nGeneral directory prefix for ReFrame-generated directories.\n\nThe base stage and output directories (see below) will be specified relative to this prefix if not specified explicitly.\n\nThis option can also be set using the [`RFM_PREFIX`](#envvar-RFM_PREFIX) environment variable or the [`prefix`](index.html#config.systems.prefix) system configuration parameter.\n\n--report-file=FILE[¶](#cmdoption-report-file)\nThe file where ReFrame will store its report.\n\nThe `FILE` argument may contain the special placeholder `{sessionid}`, in which case ReFrame will generate a new report each time it is run by appending a counter to the report file.\n\nThis option can also be set using the [`RFM_REPORT_FILE`](#envvar-RFM_REPORT_FILE) environment variable or the [`report_file`](index.html#config.general.report_file) general configuration parameter.\n\nNew in version 3.1.\n\n--report-junit=FILE[¶](#cmdoption-report-junit)\nInstruct ReFrame to generate a JUnit XML report in `FILE`.\n\nThe generated report adheres to the XSD schema [here](https://github.com/windyroad/JUnit-Schema/blob/master/JUnit.xsd) where each retry is treated as an individual testsuite.\n\nThis option can also be set using the [`RFM_REPORT_JUNIT`](#envvar-RFM_REPORT_JUNIT) environment variable or the [`report_junit`](index.html#config.general.report_junit) general configuration parameter.\n\nNew in version 3.6.0.\n\nChanged in version 3.6.1: Added support for retries in the JUnit XML report.\n\n-s, --stage=DIR[¶](#cmdoption-s)\nDirectory prefix for staging test resources.\n\nReFrame does not execute tests from their original source directory.\nInstead it creates a test-specific stage directory and copies all test resources there.\nIt then changes to that directory and executes the test.\nThis test-specific directory is of the form `{stage_prefix}/{system}/{partition}/{environment}/{test_name}`,\nwhere `stage_prefix` is set by this option.\nIf a test finishes successfully, its stage directory will be removed.\n\nThis option can also be set using the [`RFM_STAGE_DIR`](#envvar-RFM_STAGE_DIR) environment variable or the [`stagedir`](index.html#config.systems.stagedir) system configuration parameter.\n\n--save-log-files[¶](#cmdoption-save-log-files)\nSave ReFrame log files in the output directory before exiting.\n\nOnly log files generated by `file` [log handlers](config_reference.html#the-file-log-handler) will be copied.\n\nThis option can also be set using the [`RFM_SAVE_LOG_FILES`](#envvar-RFM_SAVE_LOG_FILES) environment variable or the [`save_log_files`](index.html#config.general.save_log_files) general configuration parameter.\n\n--timestamp [TIMEFMT][¶](#cmdoption-timestamp)\nAppend a timestamp to the output and stage directory prefixes.\n\n`TIMEFMT` can be any valid *[strftime(3)](http://man7.org/linux/man-pages/man3/strftime.3.html)* time format.\nIf not specified, `TIMEFMT` is set to `%FT%T`.\n\nThis option can also be set using the [`RFM_TIMESTAMP_DIRS`](#envvar-RFM_TIMESTAMP_DIRS) environment variable or the [`timestamp_dirs`](index.html#config.general.timestamp_dirs) general configuration parameter.\n\n###### Options controlling ReFrame execution[¶](#options-controlling-reframe-execution)\n\n--disable-hook=HOOK[¶](#cmdoption-disable-hook)\nDisable the pipeline hook named `HOOK` from all the tests that will run.\n\nThis feature is useful when you have implemented test workarounds as pipeline hooks, in which case you can quickly disable them from the command line.\nThis option may be specified multiple times in order to disable multiple hooks at the same time.\n\nNew in version 3.2.\n\n--distribute[=NODESTATE][¶](#cmdoption-distribute)\nDistribute the selected tests on all the nodes in state `NODESTATE` in their respective valid partitions.\n\nReFrame will parameterize and run the tests on the selected nodes.\nEffectively, it will dynamically create new tests that inherit from the original tests and add a new parameter named `$nid` which contains the list of nodes that the test must run on.\nThe new tests are named with the following pattern `{orig_test_basename}_{partition_fullname}`.\n\nWhen determining the list of nodes to distribute the selected tests, ReFrame will take into account any job options passed through the [`-J`](#cmdoption-J) option.\n\nYou can optionally specify the state of the nodes to consider when distributing the test through the `NODESTATE` argument:\n\n* `all`: Tests will run on all the nodes of their respective valid partitions regardless of the nodes’ state.\n* `idle`: Tests will run on all *idle* nodes of their respective valid partitions.\n* `NODESTATE`: Tests will run on all the nodes in state `NODESTATE` of their respective valid partitions.\nIf `NODESTATE` is not specified, `idle` will be assumed.\n\nThe state of the nodes will be determined once, before beginning the execution of the tests, so it might be different at the time the tests are actually submitted.\n\nNote\n\nCurrently, only single-node jobs can be distributed and only local or the Slurm-based backends support this feature.\n\nNote\n\nDistributing tests with dependencies is not supported, but you can distribute tests that use fixtures.\n\nNew in version 3.11.0.\n\n--exec-order=ORDER[¶](#cmdoption-exec-order)\nImpose an execution order for the independent tests.\nThe `ORDER` argument can take one of the following values:\n\n* `name`: Order tests by their display name.\n* `rname`: Order tests by their display name in reverse order.\n* `uid`: Order tests by their unique name.\n* `ruid`: Order tests by their unique name in reverse order.\n* `random`: Randomize the order of execution.\n\nIf this option is not specified the order of execution of independent tests is implementation defined.\nThis option can be combined with any of the listing options ([`-l`](#cmdoption-1) or [`-L`](#cmdoption-L)) to list the tests in the order.\n\nNew in version 4.0.0.\n\n--exec-policy=POLICY[¶](#cmdoption-exec-policy)\nThe execution policy to be used for running tests.\n\nThere are two policies defined:\n\n* `serial`: Tests will be executed sequentially.\n* `async`: Tests will be executed asynchronously.\nThis is the default policy.\n\nThe `async` execution policy executes the build and run phases of tests asynchronously by submitting their associated jobs in a non-blocking way.\nReFrame’s runtime monitors the progress of each test and will resume the pipeline execution of an asynchronously spawned test as soon as its build or run phase have finished.\nNote that the rest of the pipeline stages are still executed sequentially in this policy.\n\nConcurrency can be controlled by setting the [`max_jobs`](index.html#config.systems.partitions.max_jobs) system partition configuration parameter.\nAs soon as the concurrency limit is reached, ReFrame will first poll the status of all its pending tests to check if any execution slots have been freed up.\nIf there are tests that have finished their build or run phase, ReFrame will keep pushing tests for execution until the concurrency limit is reached again.\nIf no execution slots are available, ReFrame will throttle job submission.\n\n--max-retries=NUM[¶](#cmdoption-max-retries)\nThe maximum number of times a failing test can be retried.\n\nThe test stage and output directories will receive a `_retry<N>` suffix every time the test is retried.\n\n--maxfail=NUM[¶](#cmdoption-maxfail)\nThe maximum number of failing test cases before the execution is aborted.\n\nAfter `NUM` failed test cases the rest of the test cases will be aborted.\nThe counter of the failed test cases is reset to 0 in every retry.\n\n--mode=MODE[¶](#cmdoption-mode)\nReFrame execution mode to use.\n\nAn execution mode is simply a predefined set of options that is set in the [`modes`](index.html#modes) [configuration parameter](index.html#exec-mode-config).\nAdditional options can be passed to the command line, in which case they will be combined with the options defined in the selected execution mode.\nMore specifically, any additional ReFrame options will be *appended* to the command line options of the selected mode.\nAs a result, if a normal option is specified both inside the execution mode and the in the command line, the command line option will take precedence.\nOn the other hand, if an option that is allowed to be specified multiple times, e.g., the [`-S`](#cmdoption-S) option, is passed both inside the execution mode and in the command line, their values will be combined.\nFor example, if the execution mode `foo` defines `-S modules=foo`, the invocation `--mode=foo -S num_tasks=10` is the equivalent of `-S modules=foo -S num_tasks=10`.\n\nChanged in version 4.1: Options that can be specified multiple times are now combined between execution modes and the command line.\n\n--repeat=N[¶](#cmdoption-repeat)\nRepeat the selected tests `N` times.\nThis option can be used in conjunction with the [`--distribute`](#cmdoption-distribute) option in which case the selected tests will be repeated multiple times and distributed on individual nodes of the system’s partitions.\n\nNote\n\nRepeating tests with dependencies is not supported, but you can repeat tests that use fixtures.\n\nNew in version 3.12.0.\n\n--restore-session [REPORT1[,REPORT2,...]][¶](#cmdoption-restore-session)\nRestore a testing session that has run previously.\n\n`REPORT1` etc. are a run report files generated by ReFrame.\nIf a report is not given, ReFrame will pick the last report file found in the default location of report files (see the [`--report-file`](#cmdoption-report-file) option).\nIf passed alone, this option will simply rerun all the test cases that have run previously based on the report file data.\nIt is more useful to combine this option with any of the [test filtering](#test-filtering) options, in which case only the selected test cases will be executed.\nThe difference in test selection process when using this option is that the dependencies of the selected tests will not be selected for execution, as they would normally, but they will be restored.\nFor example, if test `T1` depends on `T2` and `T2` depends on `T3`, then running `reframe -n T1 -r` would cause both `T2` and `T3` to run.\nHowever, by doing `reframe -n T1 --restore-session -r`, only `T1` would run and its immediate dependence `T2` will be restored.\nThis is useful when you have deep test dependencies or some of the tests in the dependency chain are very time consuming.\n\nMultiple reports may be passed as a comma-separated list.\nReFrame will try to restore any required test case by looking it up in each report sequentially.\nIf it cannot find it, it will issue an error and exit.\n\nNote\n\nIn order for a test case to be restored, its stage directory must be present.\nThis is not a problem when rerunning a failed case, since the stage directories of its dependencies are automatically kept, but if you want to rerun a successful test case, you should make sure to have run with the [`--keep-stage-files`](#cmdoption-keep-stage-files) option.\n\nNew in version 3.4.\n\nChanged in version 3.6.1: Multiple report files are now accepted.\n\n-S, --setvar=[TEST.]VAR=VAL[¶](#cmdoption-S)\nSet variable `VAR` in all tests or optionally only in test `TEST` to `VAL`.\n\n`TEST` can have the form `[TEST.][FIXT.]*`, in which case `VAR` will be set in fixture `FIXT` of `TEST`.\nNote that this syntax is recursive on fixtures, so that a variable can be set in a fixture arbitrarily deep.\n`TEST` prefix refers to the test class name, *not* the test name and `FIXT` refers to the fixture *variable name* inside the referenced test, i.e., the test variable to which the fixture is bound.\nThe fixture variable name is referred to as `'<varname>` when listing tests with the [`-l`](#cmdoption-1) and [`-L`](#cmdoption-L) options.\n\nMultiple variables can be set at the same time by passing this option multiple times.\nThis option *cannot* change arbitrary test attributes, but only test variables declared with the `variable` built-in.\nIf an attempt is made to change an inexistent variable or a test parameter, a warning will be issued.\n\nReFrame will try to convert `VAL` to the type of the variable.\nIf it does not succeed, a warning will be issued and the variable will not be set.\n`VAL` can take the special value `@none` to denote that the variable must be set to [`None`](https://docs.python.org/3/library/constants.html#None).\nBoolean variables can be set in one of the following ways:\n\n* By passing `true`, `yes` or `1` to set them to `True`.\n* By passing `false`, `no` or `0` to set them to `False`.\n\nPassing any other value will issue an error.\n\nNote\n\nBoolean variables in a test must be declared of type [`Bool`](index.html#reframe.utility.typecheck.Bool) and *not* of the built-in [`bool`](https://docs.python.org/3/library/functions.html#bool) type, in order to adhere to the aforementioned behaviour.\nIf a variable is defined as [`bool`](https://docs.python.org/3/library/functions.html#bool) there is no way you can set it to [`False`](https://docs.python.org/3/library/constants.html#False), since all strings in Python evaluate to [`True`](https://docs.python.org/3/library/constants.html#True).\n\nSequence and mapping types can also be set from the command line by using the following syntax:\n\n* Sequence types: `-S seqvar=1,2,3,4`\n* Mapping types: `-S mapvar=a:1,b:2,c:3`\n\nConversions to arbitrary objects are also supported.\nSee [`ConvertibleType`](index.html#reframe.utility.typecheck.ConvertibleType) for more details.\n\nVariable assignments passed from the command line happen *before* the test is instantiated and is the exact equivalent of assigning a new value to the variable *at the end* of the test class body.\nThis has a number of implications that users of this feature should be aware of:\n\n* In the following test, `num_tasks` will have always the value `1` regardless of any command-line assignment of the variable `foo`:\n```\n@rfm.simple_test class my_test(rfm.RegressionTest):\n foo = variable(int, value=1)\n num_tasks = foo\n```\nTip\n\nIn cases where the class body expresses logic as a function of a variable and this variable, as well as its dependent logic, need to be controlled externally, the variable’s default value (i.e. the value set through the value argument) may be modified as follows through an environment variable and not through the -S option:\n```\nimport os\n\n@rfm.simple_test class my_test(rfm.RegressionTest):\n max_nodes = variable(int, value=int(os.getenv('MAX_NODES', 1)))\n # Parameterise number of nodes\n num_nodes = parameter((1 << i for i in range(0, int(max_nodes))))\n```\n* If the variable is set in any pipeline hook, the command line assignment will have an effect until the variable assignment in the pipeline hook is reached.\nThe variable will be then overwritten.\n* The [test filtering](#test-filtering) happens *after* a test is instantiated, so the only way to scope a variable assignment is to prefix it with the test class name.\nHowever, this has some positive side effects:\n\n\t+ Passing `-S valid_systems='*'` and `-S valid_prog_environs='*'` is the equivalent of passing the [`--skip-system-check`](#cmdoption-skip-system-check) and [`--skip-prgenv-check`](#cmdoption-skip-prgenv-check) options.\n\t+ Users could alter the behavior of tests based on tag values that they pass from the command line, by changing the behavior of a test in a post-init hook based on the value of the [`tags`](index.html#reframe.core.pipeline.RegressionTest.tags) attribute.\n\t+ Users could force a test with required variables to run if they set these variables from the command line.\n\tFor example, the following test could only be run if invoked with `-S num_tasks=<NUM>`:\n```\n@rfm.simple_test class my_test(rfm.RegressionTest):\n num_tasks = required\n```\nNew in version 3.8.0.\n\nChanged in version 3.9.3: Proper handling of boolean variables.\n\nChanged in version 3.11.1: Allow setting variables in fixtures.\n\n--skip-performance-check[¶](#cmdoption-skip-performance-check)\nSkip performance checking phase.\n\nThe phase is completely skipped, meaning that performance data will *not* be logged.\n\n--skip-sanity-check[¶](#cmdoption-skip-sanity-check)\nSkip sanity checking phase.\n\n###### Options controlling job submission[¶](#options-controlling-job-submission)\n\n-J, --job-option=OPTION[¶](#cmdoption-J)\nPass `OPTION` directly to the job scheduler backend.\n\nThe syntax of `OPTION` is `-J key=value`.\nIf `OPTION` starts with `-` it will be passed verbatim to the backend job scheduler.\nIf `OPTION` starts with `#` it will be emitted verbatim in the job script.\nOtherwise, ReFrame will pass `--key value` or `-k value` (if `key` is a single character) to the backend scheduler.\nAny job options specified with this command-line option will be emitted after any job options specified in the [`access`](index.html#config.systems.partitions.access) system partition configuration parameter.\n\nEspecially for the Slurm backends, constraint options, such as `-J constraint=value`, `-J C=value`, `-J --constraint=value` or `-J -C=value`, are going to be combined with any constraint options specified in the [`access`](index.html#config.systems.partitions.access) system partition configuration parameter.\nFor example, if `-C x` is specified in the [`access`](index.html#config.systems.partitions.access) and `-J C=y` is passed to the command-line, ReFrame will pass `-C x&y` as a constraint to the scheduler.\nNotice, however, that if constraint options are specified through multiple [`-J`](#cmdoption-J) options, only the last one will be considered.\nIf you wish to completely overwrite any constraint options passed in [`access`](index.html#config.systems.partitions.access), you should consider passing explicitly the Slurm directive with `-J '#SBATCH --constraint=new'`.\n\nChanged in version 3.0: This option has become more flexible.\n\nChanged in version 3.1: Use `&` to combine constraints.\n\n###### Flexible node allocation[¶](#flexible-node-allocation)\n\nReFrame can automatically set the number of tasks of a test, if its [`num_tasks`](index.html#reframe.core.pipeline.RegressionTest.num_tasks) attribute is set to a value less than or equal to zero.\nThis scheme is conveniently called *flexible node allocation* and is valid only for the Slurm backend.\nWhen allocating nodes automatically, ReFrame will take into account all node limiting factors, such as partition [`access`](index.html#config.systems.partitions.access) options, and any job submission control options described above.\nNodes from this pool are allocated according to different policies.\nIf no node can be selected, the test will be marked as a failure with an appropriate message.\n\n--flex-alloc-nodes=POLICY[¶](#cmdoption-flex-alloc-nodes)\nSet the flexible node allocation policy.\n\nAvailable values are the following:\n\n* `all`: Flexible tests will be assigned as many tasks as needed in order to span over *all* the nodes of the node pool.\n* `STATE`: Flexible tests will be assigned as many tasks as needed in order to span over the nodes that are currently in state `STATE`.\nQuerying of the node state and submission of the test job are two separate steps not executed atomically.\nIt is therefore possible that the number of tasks assigned does not correspond to the actual nodes in the given state.\n\nIf this option is not specified, the default allocation policy for flexible tests is ‘idle’.\n* Any positive integer: Flexible tests will be assigned as many tasks as needed in order to span over the specified number of nodes from the node pool.\n\nChanged in version 3.1: It is now possible to pass an arbitrary node state as a flexible node allocation parameter.\n\n###### Options controlling ReFrame environment[¶](#options-controlling-reframe-environment)\n\nReFrame offers the ability to dynamically change its environment as well as the environment of tests.\nIt does so by leveraging the selected system’s environment modules system.\n\n-M, --map-module=MAPPING[¶](#cmdoption-M)\nApply a module mapping.\n\nReFrame allows manipulating test modules on-the-fly using module mappings.\nA module mapping has the form `old_module: module1 [module2]...` and will cause ReFrame to replace a module with another list of modules upon load time.\nFor example, the mapping `foo: foo/1.2` will load module `foo/1.2` whenever module `foo` needs to be loaded.\nA mapping may also be self-referring, e.g., `gnu: gnu gcc/10.1`, however cyclic dependencies in module mappings are not allowed and ReFrame will issue an error if it detects one.\nThis option is especially useful for running tests using a newer version of a software or library.\n\nThis option may be specified multiple times, in which case multiple mappings will be applied.\n\nThis option can also be set using the [`RFM_MODULE_MAPPINGS`](#envvar-RFM_MODULE_MAPPINGS) environment variable or the [`module_mappings`](index.html#config.general.module_mappings) general configuration parameter.\n\nChanged in version 3.3: If the mapping replaces a module collection, all new names must refer to module collections, too.\n\nSee also\n\nModule collections with [Environment Modules](https://modules.readthedocs.io/en/latest/MIGRATING.html#module-collection) and [Lmod](https://lmod.readthedocs.io/en/latest/010_user.html#user-collections).\n\n-m, --module=NAME[¶](#cmdoption-3)\nLoad environment module `NAME` before acting on any tests.\n\nThis option may be specified multiple times, in which case all specified modules will be loaded in order.\nReFrame will *not* perform any automatic conflict resolution.\n\nThis option can also be set using the [`RFM_USER_MODULES`](#envvar-RFM_USER_MODULES) environment variable or the [`user_modules`](index.html#config.general.user_modules) general configuration parameter.\n\n--module-mappings=FILE[¶](#cmdoption-module-mappings)\nA file containing module mappings.\n\nEach line of the file contains a module mapping in the form described in the [`-M`](#cmdoption-M) option.\nThis option may be combined with the [`-M`](#cmdoption-M) option, in which case module mappings specified will be applied additionally.\n\nThis option can also be set using the [`RFM_MODULE_MAP_FILE`](#envvar-RFM_MODULE_MAP_FILE) environment variable or the [`module_map_file`](index.html#config.general.module_map_file) general configuration parameter.\n\n--module-path=PATH[¶](#cmdoption-module-path)\nManipulate the `MODULEPATH` environment variable before acting on any tests.\n\nIf `PATH` starts with the `-` character, it will be removed from the `MODULEPATH`, whereas if it starts with the `+` character, it will be added to it.\nIn all other cases, `PATH` will completely override MODULEPATH.\nThis option may be specified multiple times, in which case all the paths specified will be added or removed in order.\n\nNew in version 3.3.\n\n--non-default-craype[¶](#cmdoption-non-default-craype)\nTest a non-default Cray Programming Environment.\n\nSince CDT 19.11, this option can be used in conjunction with [`-m`](#cmdoption-3), which will load the target CDT.\nFor example:\n```\nreframe -m cdt/20.03 --non-default-craype -r\n```\nThis option causes ReFrame to properly set the `LD_LIBRARY_PATH` for such cases.\nIt will emit the following code after all the environment modules of a test have been loaded:\n```\nexport LD_LIBRARY_PATH=$CRAY_LD_LIBRARY_PATH:$LD_LIBRARY_PATH\n```\nThis option can also be set using the [`RFM_NON_DEFAULT_CRAYPE`](#envvar-RFM_NON_DEFAULT_CRAYPE) environment variable or the [`non_default_craype`](index.html#config.general.non_default_craype) general configuration parameter.\n\n--purge-env[¶](#cmdoption-purge-env)\nUnload all environment modules before acting on any tests.\n\nThis will unload also sticky Lmod modules.\n\nThis option can also be set using the [`RFM_PURGE_ENVIRONMENT`](#envvar-RFM_PURGE_ENVIRONMENT) environment variable or the [`purge_environment`](index.html#config.general.purge_environment) general configuration parameter.\n\n-u, --unload-module=NAME[¶](#cmdoption-u)\nUnload environment module `NAME` before acting on any tests.\n\nThis option may be specified multiple times, in which case all specified modules will be unloaded in order.\n\nThis option can also be set using the [`RFM_UNLOAD_MODULES`](#envvar-RFM_UNLOAD_MODULES) environment variable or the [`unload_modules`](index.html#config.general.unload_modules) general configuration parameter.\n\n###### Miscellaneous options[¶](#miscellaneous-options)\n\n-C, --config-file=FILE[¶](#cmdoption-C)\nUse `FILE` as configuration file for ReFrame.\n\nThis option can be passed multiple times, in which case multiple configuration files will be read and loaded successively.\nThe base of the configuration chain is always the builtin configuration file, namely the `${RFM_INSTALL_PREFIX}/reframe/core/settings.py`.\nAt any point, the user can “break” the chain of configuration files by prefixing the configuration file name with a colon as in the following example: `-C :/path/to/new_config.py`.\nThis will ignore any previously loaded configuration file and will only load the one specified.\nNote, however, that the builtin configuration file cannot be overriden;\nIt will always be loaded first in the chain.\n\nThis option can also be set using the [`RFM_CONFIG_FILES`](#envvar-RFM_CONFIG_FILES) environment variable.\n\nIn order to determine its final configuration, ReFrame first loads the builtin configuration file unconditionally and then starts looking for possible configuration file locations defined in the [`RFM_CONFIG_PATH`](#envvar-RFM_CONFIG_PATH) environment variable.\nFor each directory defined in the [`RFM_CONFIG_PATH`](#envvar-RFM_CONFIG_PATH), ReFrame looks for a file named `settings.py` or `settings.json` inside it and loads it.\nIf both a `settings.py` and a `settings.json` files are found, the Python configuration will be preferred.\nReFrame, finally, processes any configuration files specified in the command line or in the [`RFM_CONFIG_FILES`](#envvar-RFM_CONFIG_FILES) environment variable.\n\nChanged in version 4.0.0.\n\n--detect-host-topology[=FILE][¶](#cmdoption-detect-host-topology)\nDetect the local host processor topology, store it to `FILE` and exit.\n\nIf no `FILE` is specified, the standard output will be used.\n\nNew in version 3.7.0.\n\n--failure-stats[¶](#cmdoption-failure-stats)\nPrint failure statistics at the end of the run.\n\n-h, --help[¶](#cmdoption-h)\nPrint a short help message and exit.\n\n--nocolor[¶](#cmdoption-nocolor)\nDisable output coloring.\n\nThis option can also be set using the [`RFM_COLORIZE`](#envvar-RFM_COLORIZE) environment variable or the [`colorize`](index.html#config.general.colorize) general configuration parameter.\n\n--performance-report[¶](#cmdoption-performance-report)\nPrint a performance report for all the performance tests that have been run.\n\nThe report shows the performance values retrieved for the different performance variables defined in the tests.\n\n-q, --quiet[¶](#cmdoption-q)\nDecrease the verbosity level.\n\nThis option can be specified multiple times.\nEvery time this option is specified, the verbosity level will be decreased by one.\nThis option can be combined arbitrarily with the [`-v`](#cmdoption-4) option, in which case the final verbosity level will be determined by the final combination.\nFor example, specifying `-qv` will not change the verbosity level, since the two options cancel each other, but `-qqv` is equivalent to `-q`.\nFor a list of ReFrame’s verbosity levels, see the description of the [`-v`](#cmdoption-4) option.\n\nNew in version 3.9.3.\n\n--show-config [PARAM][¶](#cmdoption-show-config)\nShow the value of configuration parameter `PARAM` as this is defined for the currently selected system and exit.\n\nThe parameter value is printed in JSON format.\nIf `PARAM` is not specified or if it set to `all`, the whole configuration for the currently selected system will be shown.\nConfiguration parameters are formatted as a path navigating from the top-level configuration object to the actual parameter.\nThe `/` character acts as a selector of configuration object properties or an index in array objects.\nThe `@` character acts as a selector by name for configuration objects that have a `name` property.\nHere are some example queries:\n\n* Retrieve all the partitions of the current system:\n```\nreframe --show-config=systems/0/partitions\n```\n* Retrieve the job scheduler of the partition named `default`:\n```\nreframe --show-config=systems/0/partitions/@default/scheduler\n```\n* Retrieve the check search path for system `foo`:\n```\nreframe --system=foo --show-config=general/0/check_search_path\n```\n--system=NAME[¶](#cmdoption-system)\nLoad the configuration for system `NAME`.\n\nThe `NAME` must be a valid system name in the configuration file.\nIt may also have the form `SYSNAME:PARTNAME`, in which case the configuration of system `SYSNAME` will be loaded, but as if it had `PARTNAME` as its sole partition.\nOf course, `PARTNAME` must be a valid partition of system `SYSNAME`.\nIf this option is not specified, ReFrame will try to pick the correct configuration entry automatically.\nIt does so by trying to match the hostname of the current machine again the hostname patterns defined in the [`hostnames`](index.html#config.systems.hostnames) system configuration parameter.\nThe system with the first match becomes the current system.\n\nThis option can also be set using the [`RFM_SYSTEM`](#envvar-RFM_SYSTEM) environment variable.\n\n--upgrade-config-file=OLD[:NEW][¶](#cmdoption-upgrade-config-file)\nConvert the old-style configuration file `OLD`, place it into the new file `NEW` and exit.\n\nIf a new file is not given, a file in the system temporary directory will be created.\n\n-V, --version[¶](#cmdoption-V)\nPrint version and exit.\n\n-v, --verbose[¶](#cmdoption-4)\nIncrease verbosity level of output.\n\nThis option can be specified multiple times.\nEvery time this option is specified, the verbosity level will be increased by one.\nThere are the following message levels in ReFrame listed in increasing verbosity order:\n`critical`, `error`, `warning`, `info`, `verbose` and `debug`.\nThe base verbosity level of the output is defined by the [`level`](index.html#config.logging.handlers.level) stream logging handler configuration parameter.\n\nThis option can also be set using the [`RFM_VERBOSE`](#envvar-RFM_VERBOSE) environment variable or the [`verbose`](index.html#config.general.verbose) general configuration parameter.\n\n##### Test Naming Scheme[¶](#test-naming-scheme)\n\nNew in version 3.10.0.\n\nThis section describes the test naming scheme.\nThis scheme has superseded the old one in ReFrame 4.0.\n\nEach ReFrame test is assigned a unique name, which will be used internally by the framework to reference the test.\nAny test-specific path component will use that name, too.\nIt is formed as follows for the various types of tests:\n\n* *Regular tests*: The unique name is simply the test class name.\nThis implies that you cannot load two tests with the same class name within the same run session even if these tests reside in separate directories.\n* *Parameterized tests*: The unique name is formed by the test class name followed by an `_` and the variant number of the test.\nEach point in the parameter space of the test is assigned a unique variant number.\n* *Fixtures*: The unique name is formed by the test class name followed by an `_` and a hash.\nThe hash is constructed by combining the information of the fixture variant (if the fixture is parameterized), the fixture’s scope and any fixture variables that were explicitly set.\n\nSince unique names can be cryptic, they are not listed by the [`-l`](#cmdoption-1) option, but are listed when a detailed listing is requested by using the [`-L`](#cmdoption-L) option.\n\nA human readable version of the test name, which is called the *display name*, is also constructed for each test.\nThis name encodes all the parameterization information as well as the fixture-specific information (scopes, variables).\nThe format of the display name is the following in BNF notation:\n```\n<display_name> ::= <test_class_name> (<params>)* (<scope>)?\n<params> ::= \"%\" <parametrization> \"=\" <pvalue>\n<parametrization> ::= (<fname> \".\")* <pname>\n<scope> ::= \"~\" <scope_descr>\n<scope_descr> ::= <first> (\"+\" <second>)*\n\n<test_class_name> ::= (* as in Python *)\n<fname> ::= (* string *)\n<pname> ::= (* string *)\n<pvalue> ::= (* string *)\n<first> ::= (* string *)\n<second> ::= (* string *)\n```\nThe following is an example of a fictitious complex test that is itself parameterized and depends on parameterized fixtures as well.\n```\nimport reframe as rfm\n\nclass MyFixture(rfm.RunOnlyRegressionTest):\n p = parameter([1, 2])\n\nclass X(rfm.RunOnlyRegressionTest):\n foo = variable(int, value=1)\n\n@rfm.simple_test class TestA(rfm.RunOnlyRegressionTest):\n f = fixture(MyFixture, scope='test', action='join')\n x = parameter([3, 4])\n t = fixture(MyFixture, scope='test')\n l = fixture(X, scope='environment', variables={'foo': 10})\n valid_systems = ['*']\n valid_prog_environs = ['*']\n```\nHere is how this test is listed where the various components of the display name can be seen:\n```\n- TestA %x=4 %l.foo=10 %t.p=2 /8804be5d\n ^MyFixture %p=1 ~TestA_3 't 'f /f027ee75\n ^MyFixture %p=2 ~TestA_3 't 'f /830323a4\n ^X %foo=10 ~generic:default+builtin 'l /7dae3cc5\n- TestA %x=3 %l.foo=10 %t.p=2 /89f6f5d1\n ^MyFixture %p=1 ~TestA_2 't 'f /02368516\n ^MyFixture %p=2 ~TestA_2 't 'f /854b99b5\n ^X %foo=10 ~generic:default+builtin 'l /7dae3cc5\n- TestA %x=4 %l.foo=10 %t.p=1 /af9b2941\n ^MyFixture %p=2 ~TestA_1 't 'f /f0383f7f\n ^MyFixture %p=1 ~TestA_1 't 'f /d07f4281\n ^X %foo=10 ~generic:default+builtin 'l /7dae3cc5\n- TestA %x=3 %l.foo=10 %t.p=1 /a9e50aa3\n ^MyFixture %p=2 ~TestA_0 't 'f /b894ab05\n ^MyFixture %p=1 ~TestA_0 't 'f /ca376ca8\n ^X %foo=10 ~generic:default+builtin 'l /7dae3cc5 Found 4 check(s)\n```\nNotice that the variable name to which every fixture is bound in its parent test is also listed as `'<varname>`.\nThis is useful for setting variables down the fixture hierarchy using the [`-S`](#cmdoption-S) option.\n\nDisplay names may not always be unique.\nAssume the following test:\n```\nclass MyTest(RegressionTest):\n p = parameter([1, 1, 1])\n```\nThis generates three different tests with different unique names, but their display name is the same for all: `MyTest %p=1`.\nNotice that this example leads to a name conflict with the old naming scheme, since all tests would be named `MyTest_1`.\n\nEach test is also associated with a hash code that is derived from the test name, its parameters and their values.\nAs in the example listing above, the hash code of each test is printed with the [`-l`](#cmdoption-1) option and individual tests can be selected by their hash using the [`-n`](#cmdoption-n) option, e.g., `-n /1c51609b`.\nThe stage and output directories, as well as the performance log file of the `filelog` [performance log handler](config_reference.html#the-filelog-log-handler) will use the hash code for the test-specific directories and files.\nThis might lead to conflicts for tests as the one above when executing them with the asynchronous execution policy, but ensures consistency of performance record files when parameter values are added to or deleted from a test parameter.\nMore specifically, the test’s hash will not change if a new parameter value is added or deleted or even if the parameter values are shuffled.\nTest variants on the other side are more volatile and can change with such changes.\nAlso users should not rely on how the variant numbers are assigned to a test, as this is an implementation detail.\n\nChanged in version 4.0.0: A hash code is associated with each test.\n\n###### Differences from the old naming scheme[¶](#differences-from-the-old-naming-scheme)\n\nPrior to version 3.10, ReFrame used to encode the parameter values of an instance of parameterized test in its name.\nIt did so by taking the string representation of the value and replacing any non-alphanumeric character with an underscore.\nThis could lead to very large and hard to read names when a test defined multiple parameters or the parameter type was more complex.\nVery large test names meant also very large path names which could also lead to problems and random failures.\nFixtures followed a similar naming pattern making them hard to debug.\n\n##### Environment[¶](#environment)\n\nSeveral aspects of ReFrame can be controlled through environment variables.\nUsually environment variables have counterparts in command line options or configuration parameters.\nIn such cases, command-line options take precedence over environment variables, which in turn precede configuration parameters.\nBoolean environment variables can have any value of `true`, `yes`, `y` (case insensitive) or `1` to denote true and any value of `false`, `no`, `n` (case insensitive) or `0` to denote false.\n\nChanged in version 3.9.2: Values `1` and `0` are now valid for boolean environment variables.\n\nHere is an alphabetical list of the environment variables recognized by ReFrame.\nWhenever an environment variable is associated with a configuration option, its default value is omitted as it is the same.\n\nRFM_AUTODETECT_FQDN[¶](#envvar-RFM_AUTODETECT_FQDN)\nUse the fully qualified domain name as the hostname.\nThis is a boolean variable and defaults to `0`.\n\n| Associated command line option | N/A |\n| Associated configuration parameter | N/A |\n\nNew in version 3.11.0.\n\nChanged in version 4.0.0: This variable now defaults to `0`.\n\nRFM_AUTODETECT_METHOD[¶](#envvar-RFM_AUTODETECT_METHOD)\nMethod to use for detecting the current system and pick the right configuration.\nThe following values can be used:\n\n* `hostname`: The `hostname` command will be used to detect the current system.\nThis is the default value, if not specified.\n\n| Associated command line option | N/A |\n| Associated configuration parameter | N/A |\n\nNew in version 3.11.0.\n\nRFM_AUTODETECT_XTHOSTNAME[¶](#envvar-RFM_AUTODETECT_XTHOSTNAME)\nUse `/etc/xthostname` file, if present, to retrieve the current system’s name.\nIf the file cannot be found, the hostname will be retrieved using the `hostname` command.\nThis is a boolean variable and defaults to `0`.\n\nThis option meaningful for Cray systems.\n\n| Associated command line option | N/A |\n| Associated configuration parameter | N/A |\n\nNew in version 3.11.0.\n\nChanged in version 4.0.0: This variable now defaults to `0`.\n\nRFM_CHECK_SEARCH_PATH[¶](#envvar-RFM_CHECK_SEARCH_PATH)\nA colon-separated list of filesystem paths where ReFrame should search for tests.\n\n| Associated command line option | [`-c`](#cmdoption-c) |\n| Associated configuration parameter | [`check_search_path`](index.html#config.general.check_search_path) |\n\nRFM_CHECK_SEARCH_RECURSIVE[¶](#envvar-RFM_CHECK_SEARCH_RECURSIVE)\nSearch for test files recursively in directories found in the check search path.\n\n| Associated command line option | [`-R`](#cmdoption-R) |\n| Associated configuration parameter | [`check_search_recursive`](index.html#config.general.check_search_recursive) |\n\nRFM_CLEAN_STAGEDIR[¶](#envvar-RFM_CLEAN_STAGEDIR)\nClean stage directory of tests before populating it.\n\nNew in version 3.1.\n\n| Associated command line option | [`--dont-restage`](#cmdoption-dont-restage) |\n| Associated configuration parameter | [`clean_stagedir`](index.html#config.general.clean_stagedir) |\n\nRFM_COLORIZE[¶](#envvar-RFM_COLORIZE)\nEnable output coloring.\n\n| Associated command line option | [`--nocolor`](#cmdoption-nocolor) |\n| Associated configuration parameter | [`colorize`](index.html#config.general.colorize) |\n\nRFM_COMPRESS_REPORT[¶](#envvar-RFM_COMPRESS_REPORT)\nCompress the generated run report file.\n\n| Associated command line option | [`--compress-report`](#cmdoption-compress-report) |\n| Associated configuration parameter | [`compress_report`](index.html#config.general.compress_report) |\n\nNew in version 3.12.0.\n\nRFM_CONFIG_FILE[¶](#envvar-RFM_CONFIG_FILE)\nSet the configuration file for ReFrame.\n\n| Associated command line option | [`-C`](#cmdoption-C) |\n| Associated configuration parameter | N/A |\n\nDeprecated since version 4.0.0: Please use the [`RFM_CONFIG_FILES`](#envvar-RFM_CONFIG_FILES) instead.\n\nRFM_CONFIG_FILES[¶](#envvar-RFM_CONFIG_FILES)\nA colon-separated list of configuration files to load.\nRefer to the documentation of the [`--config-file`](#cmdoption-C) option for a detailed description on how ReFrame loads its configuration.\n\n| Associated command line option | [`-C`](#cmdoption-C) |\n| Associated configuration parameter | N/A |\n\nNew in version 4.0.0.\n\nRFM_CONFIG_PATH[¶](#envvar-RFM_CONFIG_PATH)\nA colon-separated list of directories that contain ReFrame configuration files.\nRefer to the documentation of the [`--config-file`](#cmdoption-C) option for a detailed description on how ReFrame loads its configuration.\n\n| Associated command line option | N/A |\n| Associated configuration parameter | N/A |\n\nNew in version 4.0.0.\n\nRFM_GIT_TIMEOUT[¶](#envvar-RFM_GIT_TIMEOUT)\nTimeout value in seconds used when checking if a git repository exists.\n\n| Associated command line option | N/A |\n| Associated configuration parameter | [`git_timeout`](index.html#config.general.git_timeout) |\n\nNew in version 3.9.0.\n\nRFM_GRAYLOG_ADDRESS[¶](#envvar-RFM_GRAYLOG_ADDRESS)\nThe address of the Graylog server to send performance logs.\nThe address is specified in `host:port` format.\n\n| Associated command line option | N/A |\n| Associated configuration parameter | [`address`](index.html#config.logging.handlers..graylog..address) |\n\nNew in version 3.1.\n\nRFM_HTTPJSON_URL[¶](#envvar-RFM_HTTPJSON_URL)\nThe URL of the server to send performance logs in JSON format.\nThe URL is specified in `scheme://host:port/path` format.\n\n| Associated command line option | N/A |\n| Associated configuration parameter | [`url`](index.html#config.logging.handlers..httpjson..url) |\n\nNew in version 3.6.1.\n\nRFM_IGNORE_REQNODENOTAVAIL[¶](#envvar-RFM_IGNORE_REQNODENOTAVAIL)\nDo not treat specially jobs in pending state with the reason `ReqNodeNotAvail` (Slurm only).\n\n| Associated command line option | N/A |\n| Associated configuration parameter | [`ignore_reqnodenotavail`](index.html#config.systems.partitions.sched_options.ignore_reqnodenotavail) |\n\nRFM_KEEP_STAGE_FILES[¶](#envvar-RFM_KEEP_STAGE_FILES)\nKeep test stage directories even for tests that finish successfully.\n\n| Associated command line option | [`--keep-stage-files`](#cmdoption-keep-stage-files) |\n| Associated configuration parameter | [`keep_stage_files`](index.html#config.general.keep_stage_files) |\n\nRFM_MODULE_MAP_FILE[¶](#envvar-RFM_MODULE_MAP_FILE)\nA file containing module mappings.\n\n| Associated command line option | [`--module-mappings`](#cmdoption-module-mappings) |\n| Associated configuration parameter | [`module_map_file`](index.html#config.general.module_map_file) |\n\nRFM_MODULE_MAPPINGS[¶](#envvar-RFM_MODULE_MAPPINGS)\nA comma-separated list of module mappings.\n\n| Associated command line option | [`-M`](#cmdoption-M) |\n| Associated configuration parameter | [`module_mappings`](index.html#config.general.module_mappings) |\n\nRFM_NON_DEFAULT_CRAYPE[¶](#envvar-RFM_NON_DEFAULT_CRAYPE)\nTest a non-default Cray Programming Environment.\n\n| Associated command line option | [`--non-default-craype`](#cmdoption-non-default-craype) |\n| Associated configuration parameter | [`non_default_craype`](index.html#config.general.non_default_craype) |\n\nRFM_OUTPUT_DIR[¶](#envvar-RFM_OUTPUT_DIR)\nDirectory prefix for test output files.\n\n| Associated command line option | [`-o`](#cmdoption-o) |\n| Associated configuration parameter | [`outputdir`](index.html#config.systems.outputdir) |\n\nRFM_PERF_INFO_LEVEL[¶](#envvar-RFM_PERF_INFO_LEVEL)\nLogging level at which the immediate performance information is logged.\n\n| Associated command line option | n/a |\n| Associated configuration parameter | [`perf_info_level`](index.html#config.general.perf_info_level) |\n\nRFM_PERFLOG_DIR[¶](#envvar-RFM_PERFLOG_DIR)\nDirectory prefix for logging performance data.\n\n| Associated command line option | [`--perflogdir`](#cmdoption-perflogdir) |\n| Associated configuration parameter | [`basedir`](index.html#config.logging.handlers..filelog..basedir) |\n\nRFM_PIPELINE_TIMEOUT[¶](#envvar-RFM_PIPELINE_TIMEOUT)\nTimeout in seconds for advancing the pipeline in the asynchronous execution policy.\nSee [Tweaking the throughput and interactivity of test jobs in the asynchronous execution policy](index.html#pipeline-timeout) for more guidance on how to set this.\n\n| Associated command line option | N/A |\n| Associated configuration parameter | [`pipeline_timeout`](index.html#config.general.pipeline_timeout) |\n\nNew in version 3.10.0.\n\nRFM_PREFIX[¶](#envvar-RFM_PREFIX)\nGeneral directory prefix for ReFrame-generated directories.\n\n| Associated command line option | [`--prefix`](#cmdoption-prefix) |\n| Associated configuration parameter | [`prefix`](index.html#config.systems.prefix) |\n\nRFM_PURGE_ENVIRONMENT[¶](#envvar-RFM_PURGE_ENVIRONMENT)\nUnload all environment modules before acting on any tests.\n\n| Associated command line option | [`--purge-env`](#cmdoption-purge-env) |\n| Associated configuration parameter | [`purge_environment`](index.html#config.general.purge_environment) |\n\nRFM_REMOTE_DETECT[¶](#envvar-RFM_REMOTE_DETECT)\nAuto-detect processor information of remote partitions as well.\n\n| Associated command line option | N/A |\n| Associated configuration parameter | [`remote_detect`](index.html#config.general.remote_detect) |\n\nNew in version 3.7.0.\n\nRFM_REMOTE_WORKDIR[¶](#envvar-RFM_REMOTE_WORKDIR)\nThe temporary directory prefix that will be used to create a fresh ReFrame clone, in order to auto-detect the processor information of a remote partition.\n\n| Associated command line option | N/A |\n| Associated configuration parameter | [`remote_workdir`](index.html#config.general.remote_workdir) |\n\nNew in version 3.7.0.\n\nRFM_REPORT_FILE[¶](#envvar-RFM_REPORT_FILE)\nThe file where ReFrame will store its report.\n\nNew in version 3.1.\n\n| Associated command line option | [`--report-file`](#cmdoption-report-file) |\n| Associated configuration parameter | [`report_file`](index.html#config.general.report_file) |\n\nRFM_REPORT_JUNIT[¶](#envvar-RFM_REPORT_JUNIT)\nThe file where ReFrame will generate a JUnit XML report.\n\nNew in version 3.6.0.\n\n| Associated command line option | [`--report-junit`](#cmdoption-report-junit) |\n| Associated configuration parameter | [`report_junit`](index.html#config.general.report_junit) |\n\nRFM_RESOLVE_MODULE_CONFLICTS[¶](#envvar-RFM_RESOLVE_MODULE_CONFLICTS)\nResolve module conflicts automatically.\n\nNew in version 3.6.0.\n\n| Associated command line option | N/A |\n| Associated configuration parameter | [`resolve_module_conflicts`](index.html#config.general.resolve_module_conflicts) |\n\nRFM_SAVE_LOG_FILES[¶](#envvar-RFM_SAVE_LOG_FILES)\nSave ReFrame log files in the output directory before exiting.\n\n| Associated command line option | [`--save-log-files`](#cmdoption-save-log-files) |\n| Associated configuration parameter | [`save_log_files`](index.html#config.general.save_log_files) |\n\nRFM_STAGE_DIR[¶](#envvar-RFM_STAGE_DIR)\nDirectory prefix for staging test resources.\n\n| Associated command line option | [`-s`](#cmdoption-s) |\n| Associated configuration parameter | [`stagedir`](index.html#config.systems.stagedir) |\n\nRFM_SYSLOG_ADDRESS[¶](#envvar-RFM_SYSLOG_ADDRESS)\nThe address of the Syslog server to send performance logs.\nThe address is specified in `host:port` format.\nIf no port is specified, the address refers to a UNIX socket.\n\n| Associated command line option | N/A |\n| Associated configuration parameter | [`address`](index.html#config.logging.handlers..syslog..address) |\n\nNew in version 3.1.\n\nRFM_SYSTEM[¶](#envvar-RFM_SYSTEM)\nSet the current system.\n\n| Associated command line option | [`--system`](#cmdoption-system) |\n| Associated configuration parameter | N/A |\n\nRFM_TIMESTAMP_DIRS[¶](#envvar-RFM_TIMESTAMP_DIRS)\nAppend a timestamp to the output and stage directory prefixes.\n\n| Associated command line option | [`--timestamp`](#cmdoption-timestamp) |\n| Associated configuration parameter | [`timestamp_dirs`](index.html#config.general.timestamp_dirs) |\n\nRFM_TRAP_JOB_ERRORS[¶](#envvar-RFM_TRAP_JOB_ERRORS)\nTrap job errors in submitted scripts and fail tests automatically.\n\n| Associated configuration parameter | [`trap_job_errors`](index.html#config.general.trap_job_errors) |\n\nNew in version 3.9.0.\n\nRFM_UNLOAD_MODULES[¶](#envvar-RFM_UNLOAD_MODULES)\nA comma-separated list of environment modules to be unloaded before acting on any tests.\n\n| Associated command line option | [`-u`](#cmdoption-u) |\n| Associated configuration parameter | [`unload_modules`](index.html#config.general.unload_modules) |\n\nRFM_USE_LOGIN_SHELL[¶](#envvar-RFM_USE_LOGIN_SHELL)\nUse a login shell for the generated job scripts.\n\n| Associated command line option | N/A |\n| Associated configuration parameter | [`use_login_shell`](index.html#config.general.use_login_shell) |\n\nRFM_USER_MODULES[¶](#envvar-RFM_USER_MODULES)\nA comma-separated list of environment modules to be loaded before acting on any tests.\n\n| Associated command line option | [`-m`](#cmdoption-3) |\n| Associated configuration parameter | [`user_modules`](index.html#config.general.user_modules) |\n\nRFM_VERBOSE[¶](#envvar-RFM_VERBOSE)\nSet the verbosity level of output.\n\n| Associated command line option | [`-v`](#cmdoption-4) |\n| Associated configuration parameter | [`verbose`](index.html#config.general.verbose) |\n\n##### Configuration File[¶](#configuration-file)\n\nThe configuration file of ReFrame defines the systems and environments to test as well as parameters controlling its behavior.\nUpon start up ReFrame checks for configuration files in the following locations in that order:\n\n1. `$HOME/.reframe/settings.{py,json}`\n2. `$RFM_INSTALL_PREFIX/settings.{py,json}`\n3. `/etc/reframe.d/settings.{py,json}`\n\nReFrame accepts configuration files either in Python or JSON syntax.\nIf both are found in the same location, the Python file will be preferred.\n\nThe `RFM_INSTALL_PREFIX` environment variable refers to the installation directory of ReFrame.\nUsers have no control over this variable.\nIt is always set by the framework upon startup.\n\nIf no configuration file can be found in any of the predefined locations, ReFrame will fall back to a generic configuration that allows it to run on any system.\nThis configuration file is located in [`reframe/core/settings.py`](https://github.com/reframe-hpc/reframe/blob/master/reframe/core/settings.py).\nUsers may *not* modify this file.\n\nFor a complete reference of the configuration, please refer to [`reframe.settings(8)`](config_reference.html) man page.\n\n##### Reporting Bugs[¶](#reporting-bugs)\n\nFor bugs, feature request, help, please open an issue on Github: <<https://github.com/reframe-hpc/reframe>##### See Also[¶](#see-also)\n\nSee full documentation online: <<https://reframe-hpc.readthedocs.io/>#### Configuration Reference[¶](#configuration-reference)\n\nReFrame’s behavior can be configured through its configuration file (see [Configuring ReFrame for Your Site](index.html#document-configure)), environment variables and command-line options.\nAn option can be specified via multiple paths (e.g., a configuration file parameter and an environment variable), in which case command-line options precede environment variables, which in turn precede configuration file options.\nThis section provides a complete reference guide of the configuration options of ReFrame that can be set in its configuration file or specified using environment variables.\n\nReFrame’s configuration is in JSON syntax.\nThe full schema describing it can be found in [`reframe/schemas/config.json`](https://github.com/reframe-hpc/reframe/blob/master/reframe/schemas/config.json) file.\nThe final configuration for ReFrame is validated against this schema.\n\nThe syntax we use to describe the different configuration objects follows the convention: `OBJECT[.OBJECT]*.PROPERTY`.\nEven if a configuration object contains a list of other objects, this is not reflected in the above syntax, as all objects in a certain list are homogeneous.\nFor example, by `systems.partitions.name` we designate the `name` property of any partition object inside the `partitions` property of any system object inside the top level `systems` object.\nIf we were to use indices, that would be rewritten as `systems[i].partitions[j].name` where `i` indexes the systems and `j` indexes the partitions of the i-th system.\nFor cases, where the objects in a list are not homogeneous, e.g., the logging handlers, we surround the object type with `..`.\nFor example, the `logging.handlers..filelog..name` syntax designates the `name` attribute of the `filelog` logging handler.\n\n##### Top-level Configuration[¶](#top-level-configuration)\n\nThe top-level configuration object is essentially the full configuration of ReFrame.\nIt consists of the following properties, which we also call conventionally *configuration sections*:\n\nsystems[¶](#systems)\n\nRequired Yes\n\nA list of [system configuration objects](#system-configuration).\n\nenvironments[¶](#environments)\n\nRequired Yes\n\nA list of [environment configuration objects](#environment-configuration).\n\nlogging[¶](#logging)\n\nRequired Yes\n\nA list of [logging configuration objects](#logging-configuration).\n\nmodes[¶](#modes)\n\nRequired No\n\nA list of [execution mode configuration objects](#execution-mode-configuration).\n\ngeneral[¶](#general)\n\nRequired No\n\nA list of [general configuration objects](#general-configuration).\n\nWarning\n\nChanged in version 4.0.0: The `schedulers` section is removed.\nScheduler options should be set per partition using the [`sched_options`](#config.systems.partitions.sched_options) attribute.\n\n##### System Configuration[¶](#system-configuration)\n\nsystems.name[¶](#config.systems.name)\n\nRequired Yes\n\nThe name of this system.\nOnly alphanumeric characters, dashes (`-`) and underscores (`_`) are allowed.\n\nsystems.descr[¶](#config.systems.descr)\n\nRequired No\n\nDefault\n`\"\"`\n\nThe description of this system.\n\nsystems.hostnames[¶](#config.systems.hostnames)\n\nRequired Yes\n\nA list of hostname regular expression patterns in Python [syntax](https://docs.python.org/3.8/library/re.html), which will be used by the framework in order to automatically select a system configuration.\nFor the auto-selection process, see [here](configure.html#picking-a-system-configuration).\n\nsystems.max_local_jobs[¶](#config.systems.max_local_jobs)\nThe maximum number of forced local build or run jobs allowed.\n\nForced local jobs run within the execution context of ReFrame.\n\nRequired No\n\nDefault\n`8`\n\nNew in version 3.10.0.\n\nsystems.modules_system[¶](#config.systems.modules_system)\n\nRequired No\n\nDefault\n`\"nomod\"`\n\nThe modules system that should be used for loading environment modules on this system.\nAvailable values are the following:\n\n* `tmod`: The classic Tcl implementation of the [environment modules](https://sourceforge.net/projects/modules/files/Modules/modules-3.2.10/) (version 3.2).\n* `tmod31`: The classic Tcl implementation of the [environment modules](https://sourceforge.net/projects/modules/files/Modules/modules-3.2.10/) (version 3.1).\nA separate backend is required for Tmod 3.1, because Python bindings are different from Tmod 3.2.\n* `tmod32`: A synonym of `tmod`.\n* `tmod4`: The [new environment modules](http://modules.sourceforge.net/) implementation (versions older than 4.1 are not supported).\n* `lmod`: The [Lua implementation](https://lmod.readthedocs.io/en/latest/) of the environment modules.\n* `spack`: [Spack](https://spack.readthedocs.io/en/latest/)’s built-in mechanism for managing modules.\n* `nomod`: This is to denote that no modules system is used by this system.\n\nNew in version 3.4: The `spack` backend is added.\n\nsystems.modules[¶](#config.systems.modules)\n\nRequired No\n\nDefault\n`[]`\n\nA list of [environment module objects](#module-objects) to be loaded always when running on this system.\nThese modules modify the ReFrame environment.\nThis is useful in cases where a particular module is needed, for example, to submit jobs on a specific system.\n\nsystems.env_vars[¶](#config.systems.env_vars)\n\nRequired No\n\nDefault\n`[]`\n\nA list of environment variables to be set always when running on this system.\nThese variables modify the ReFrame environment.\nEach environment variable is specified as a two-element list containing the variable name and its value.\nYou may reference other environment variables when defining an environment variable here.\nReFrame will expand its value.\nVariables are set after the environment modules are loaded.\n\nNew in version 4.0.0.\n\nsystems.variables[¶](#config.systems.variables)\n\nDeprecated since version 4.0.0: Please use [`env_vars`](#config.systems.env_vars) instead.\nIf specified in conjunction with [`env_vars`](#config.systems.env_vars), it will be ignored.\n\nsystems.prefix[¶](#config.systems.prefix)\n\nRequired No\n\nDefault\n`\".\"`\n\nDirectory prefix for a ReFrame run on this system.\nAny directories or files produced by ReFrame will use this prefix, if not specified otherwise.\n\nsystems.stagedir[¶](#config.systems.stagedir)\n\nRequired No\n\nDefault\n`\"${RFM_PREFIX}/stage\"`\n\nStage directory prefix for this system.\nThis is the directory prefix, where ReFrame will create the stage directories for each individual test case.\n\nsystems.outputdir[¶](#config.systems.outputdir)\n\nRequired No\n\nDefault\n`\"${RFM_PREFIX}/output\"`\n\nOutput directory prefix for this system.\nThis is the directory prefix, where ReFrame will save information about the successful tests.\n\nsystems.resourcesdir[¶](#config.systems.resourcesdir)\n\nRequired No\n\nDefault\n`\".\"`\n\nDirectory prefix where external test resources (e.g., large input files) are stored.\nYou may reference this prefix from within a regression test by accessing the [`resourcesdir`](index.html#reframe.core.systems.System.resourcesdir) attribute of the current system.\n\nsystems.partitions[¶](#config.systems.partitions)\n\nRequired Yes\n\nA list of [system partition configuration objects](#system-partition-configuration).\nThis list must have at least one element.\n\nsystems.sched_options[¶](#config.systems.sched_options)\n\nRequired No\n\nDefault\n`{}`\n\nScheduler options for the local scheduler that is associated with the ReFrame’s execution context.\nTo understand the difference between the different execution contexts, please refer to “[Where each pipeline stage is executed?](index.html#execution-contexts)”\nFor the available scheduler options, see the [`sched_options`](#config.systems.partitions.sched_options) in the partition configuration below.\n\nNew in version 4.1.\n\nWarning\n\nThis option is broken in 4.0.\n\n###### System Partition Configuration[¶](#system-partition-configuration)\n\nsystems.partitions.name[¶](#config.systems.partitions.name)\n\nRequired Yes\n\nThe name of this partition.\nOnly alphanumeric characters, dashes (`-`) and underscores (`_`) are allowed.\n\nsystems.partitions.descr[¶](#config.systems.partitions.descr)\n\nRequired No\n\nDefault\n`\"\"`\n\nThe description of this partition.\n\nsystems.partitions.scheduler[¶](#config.systems.partitions.scheduler)\n\nRequired Yes\n\nThe job scheduler that will be used to launch jobs on this partition.\nSupported schedulers are the following:\n\n* `local`: Jobs will be launched locally without using any job scheduler.\n* `flux`: Jobs will be launched using the [Flux Framework](https://flux-framework.org/) scheduler.\n* `oar`: Jobs will be launched using the [OAR](https://oar.imag.fr/) scheduler.\n* `pbs`: Jobs will be launched using the [PBS Pro](https://en.wikipedia.org/wiki/Portable_Batch_System) scheduler.\n* `sge`: Jobs will be launched using the [Sun Grid Engine](https://arc.liv.ac.uk/SGE/htmlman/manuals.html) scheduler.\n* `slurm`: Jobs will be launched using the [Slurm](https://www.schedmd.com/) scheduler.\nThis backend requires job accounting to be enabled in the target system.\nIf not, you should consider using the `squeue` backend below.\n* `squeue`: Jobs will be launched using the [Slurm](https://www.schedmd.com/) scheduler.\nThis backend does not rely on job accounting to retrieve job statuses, but ReFrame does its best to query the job state as reliably as possible.\n* `torque`: Jobs will be launched using the [Torque](https://en.wikipedia.org/wiki/TORQUE) scheduler.\n* `lsf`: Jobs will be launched using the [LSF](https://www.ibm.com/docs/en/spectrum-lsf/10.1.0?topic=lsf-session-scheduler) scheduler.\n\nNew in version 3.7.2: Support for the SGE scheduler is added.\n\nNew in version 3.8.2: Support for the OAR scheduler is added.\n\nNew in version 3.11.0: Support for the LSF scheduler is added.\n\nNote\n\nThe way that multiple node jobs are submitted using the SGE scheduler can be very site-specific.\nFor this reason, the `sge` scheduler backend does not try to interpret any related arguments, e.g., `num_tasks`, `num_tasks_per_node` etc.\nUsers must specify how these resources are to be requested by setting the [`resources`](#config.systems.partitions.resources) partition configuration parameter and then request them from inside a test using the [`extra_resources`](index.html#reframe.core.pipeline.RegressionTest.extra_resources) test attribute.\nHere is an example configuration for a system partition named `foo` that defines different ways for submitting MPI-only, OpenMP-only and MPI+OpenMP jobs:\n```\n{\n 'name': 'foo',\n 'scheduler': 'sge',\n 'resources': [\n {\n 'name': 'smp',\n 'options': ['-pe smp {num_slots}']\n },\n {\n 'name': 'mpi',\n 'options': ['-pe mpi {num_slots}']\n },\n {\n 'name': 'mpismp',\n 'options': ['-pe mpismp {num_slots}']\n }\n ]\n}\n```\nEach test then can request the different type of slots as follows:\n```\nself.extra_resouces = {\n 'smp': {'num_slots': self.num_cpus_per_task},\n 'mpi': {'num_slots': self.num_tasks},\n 'mpismp': {'num_slots': self.num_tasks*self.num_cpus_per_task}\n}\n```\nNotice that defining [`extra_resources`](index.html#reframe.core.pipeline.RegressionTest.extra_resources) allows the test to be portable to other systems that have different schedulers;\nthe [`extra_resources`](index.html#reframe.core.pipeline.RegressionTest.extra_resources) will be simply ignored in this case and the scheduler backend will interpret the different test fields in the appropriate way.\n\nsystems.partitions.sched_options[¶](#config.systems.partitions.sched_options)\n\nRequired No\n\nDefault\n`{}`\n\nScheduler-specific options for this partition.\nSee below for the available options.\n\nNew in version 4.1.\n\nWarning\n\nThis option is broken in 4.0.\n\nsystems.partitions.sched_options.ignore_reqnodenotavail[¶](#config.systems.partitions.sched_options.ignore_reqnodenotavail)\n\nRequired No\n\nDefault\n`false`\n\nIgnore the `ReqNodeNotAvail` Slurm state.\n\nIf a job associated to a test is in pending state with the Slurm reason `ReqNodeNotAvail` and a list of unavailable nodes is also specified, ReFrame will check the status of the nodes and, if all of them are indeed down, it will cancel the job.\nSometimes, however, when Slurm’s backfill algorithm takes too long to compute, Slurm will set the pending reason to `ReqNodeNotAvail` and mark all system nodes as unavailable, causing ReFrame to kill the job.\nIn such cases, you may set this parameter to `true` to avoid this.\n\nThis option is relevant for the Slurm backends only.\n\nsystems.partitions.sched_options.job_submit_timeout[¶](#config.systems.partitions.sched_options.job_submit_timeout)\n\nRequired No\n\nDefault\n`60`\n\nTimeout in seconds for the job submission command.\n\nIf timeout is reached, the test issuing that command will be marked as a failure.\n\nsystems.partitions.sched_options.resubmit_on_errors[¶](#config.systems.partitions.sched_options.resubmit_on_errors)\n\nRequired No\n\nDefault\n`[]`\n\nIf any of the listed errors occur, try to resubmit the job after some seconds.\n\nAs an example, you could have ReFrame trying to resubmit a job in case that the maximum submission limit per user is reached by setting this field to `[\"QOSMaxSubmitJobPerUserLimit\"]`.\nYou can ignore multiple errors at the same time if you add more error strings in the list.\n\nThis option is relevant for the Slurm backends only.\n\nNew in version 3.4.1.\n\nWarning\n\nJob submission is a synchronous operation in ReFrame.\nIf this option is set, ReFrame’s execution will block until the error conditions specified in this list are resolved.\nNo other test would be able to proceed.\n\nsystems.partitions.sched_options.use_nodes_option[¶](#config.systems.partitions.sched_options.use_nodes_option)\n\nRequired No\n\nDefault\n`false`\n\nAlways emit the `--nodes` Slurm option in the preamble of the job script.\n\nThis option is relevant for the Slurm backends only.\n\nsystems.partitions.launcher[¶](#config.systems.partitions.launcher)\n\nRequired Yes\n\nThe parallel job launcher that will be used in this partition to launch parallel programs.\nAvailable values are the following:\n\n* `alps`: Parallel programs will be launched using the [Cray ALPS](https://pubs.cray.com/content/S-2393/CLE%205.2.UP03/cle-xc-system-administration-guide-s-2393-5203-xc/the-aprun-client) `aprun` command.\n* `ibrun`: Parallel programs will be launched using the `ibrun` command.\nThis is a custom parallel program launcher used at [TACC](https://portal.tacc.utexas.edu/user-guides/stampede2).\n* `local`: No parallel program launcher will be used.\nThe program will be launched locally.\n* `lrun`: Parallel programs will be launched using [LC Launcher](https://hpc.llnl.gov/training/tutorials/using-lcs-sierra-system#lrun)’s `lrun` command.\n* `lrun-gpu`: Parallel programs will be launched using [LC Launcher](https://hpc.llnl.gov/training/tutorials/using-lcs-sierra-system#lrun)’s `lrun -M \"-gpu\"` command that enables the CUDA-aware Spectrum MPI.\n* `mpirun`: Parallel programs will be launched using the `mpirun` command.\n* `mpiexec`: Parallel programs will be launched using the `mpiexec` command.\n* `srun`: Parallel programs will be launched using [Slurm](https://slurm.schedmd.com/srun.html)’s `srun` command.\n* `srunalloc`: Parallel programs will be launched using [Slurm](https://slurm.schedmd.com/srun.html)’s `srun` command, but job allocation options will also be emitted.\nThis can be useful when combined with the `local` job scheduler.\n* `ssh`: Parallel programs will be launched using SSH.\nThis launcher uses the partition’s [`access`](#config.systems.partitions.access) property in order to determine the remote host and any additional options to be passed to the SSH client.\nThe ssh command will be launched in “batch mode,” meaning that password-less access to the remote host must be configured.\nHere is an example configuration for the ssh launcher:\n```\n{\n 'name': 'foo'\n 'scheduler': 'local',\n 'launcher': 'ssh'\n 'access': ['-l admin', 'remote.host'],\n 'environs': ['builtin'],\n}\n```\n* `upcrun`: Parallel programs will be launched using the [UPC](https://upc.lbl.gov/) `upcrun` command.\n* `upcxx-run`: Parallel programs will be launched using the [UPC++](https://bitbucket.org/berkeleylab/upcxx/wiki/Home) `upcxx-run` command.\n\nTip\n\nNew in version 4.0.0: ReFrame also allows you to register your own custom launchers simply by defining them in the configuration.\nYou can follow a small tutorial [here](tutorial_advanced.html#adding-a-custom-launcher-to-a-partition).\n\nsystems.partitions.access[¶](#config.systems.partitions.access)\n\nRequired No\n\nDefault\n`[]`\n\nA list of job scheduler options that will be passed to the generated job script for gaining access to that logical partition.\n\nsystems.partitions.environs[¶](#config.systems.partitions.environs)\n\n> required\n> No\n> default\n> `[]`\nA list of environment names that ReFrame will use to run regression tests on this partition.\nEach environment must be defined in the [`environments`](#environments) section of the configuration and the definition of the environment must be valid for this partition.\n\nsystems.partitions.container_platforms[¶](#config.systems.partitions.container_platforms)\n\nRequired No\n\nDefault\n`[]`\n\nA list for [container platform configuration objects](#container-platform-configuration).\nThis will allow launching regression tests that use containers on this partition.\n\nsystems.partitions.modules[¶](#config.systems.partitions.modules)\n\n> required\n> No\n> default\n> `[]`\nA list of [environment module objects](#module-objects) to be loaded before running a regression test on this partition.\n\nsystems.partitions.time_limit[¶](#config.systems.partitions.time_limit)\n\nRequired No\n\nDefault\n`null`\n\nThe time limit for the jobs submitted on this partition.\nWhen the value is `null`, no time limit is applied.\n\nsystems.partitions.env_vars[¶](#config.systems.partitions.env_vars)\n\nRequired No\n\nDefault\n`[]`\n\nA list of environment variables to be set before running a regression test on this partition.\nEach environment variable is specified as a two-element list containing the variable name and its value.\nYou may reference other environment variables when defining an environment variable here.\nReFrame will expand its value.\nVariables are set after the environment modules are loaded.\n\nNew in version 4.0.0.\n\nsystems.partitions.variables[¶](#config.systems.partitions.variables)\n\nDeprecated since version 4.0.0: Please use [`env_vars`](#config.systems.partitions.env_vars) instead.\nIf specified in conjunction with [`env_vars`](#config.systems.partitions.env_vars), it will be ignored.\n\nsystems.partitions.max_jobs[¶](#config.systems.partitions.max_jobs)\n\nRequired No\n\nDefault\n`8`\n\nThe maximum number of concurrent regression tests that may be active (i.e., not completed) on this partition.\nThis option is relevant only when ReFrame executes with the [asynchronous execution policy](pipeline.html#execution-policies).\n\nsystems.partitions.prepare_cmds[¶](#config.systems.partitions.prepare_cmds)\n\nRequired No\n\nDefault\n`[]`\n\nList of shell commands to be emitted before any environment loading commands are emitted.\n\nNew in version 3.5.0.\n\nsystems.partitions.resources[¶](#config.systems.partitions.resources)\n\nRequired No\n\nDefault\n`[]`\n\nA list of job scheduler [resource specification](config_reference.html#custom-job-scheduler-resources) objects.\n\nsystems.partitions.processor[¶](#config.systems.partitions.processor)\n\nRequired No\n\nDefault\n`{}`\n\nProcessor information for this partition stored in a [processor info object](#processor-info).\nIf not set, ReFrame will try to auto-detect this information (see [Auto-detecting processor information](index.html#proc-autodetection) for more information).\n\nNew in version 3.5.0.\n\nChanged in version 3.7.0: ReFrame is now able to detect the processor information automatically.\n\nsystems.partitions.devices[¶](#config.systems.partitions.devices)\n\nRequired No\n\nDefault\n`[]`\n\nA list with [device info objects](#device-info) for this partition.\n\nNew in version 3.5.0.\n\nsystems.partitions.features[¶](#config.systems.partitions.features)\n\nRequired No\n\nDefault\n`[]`\n\nUser defined features of the partition.\nThese are accessible through the [`features`](index.html#reframe.core.systems.SystemPartition.features) attribute of the [`current_partition`](index.html#reframe.core.pipeline.RegressionTest.current_partition) and can also be selected through the extended syntax of [`valid_systems`](index.html#reframe.core.pipeline.RegressionTest.valid_systems).\nThe values of this list must be alphanumeric strings starting with a non-digit character and may also contain a `-`.\n\nNew in version 3.11.0.\n\nsystems.partitions.extras[¶](#config.systems.partitions.extras)\n\nRequired No\n\nDefault\n`{}`\n\nUser defined attributes of the partition.\nThese are accessible through the [`extras`](index.html#reframe.core.systems.SystemPartition.extras) attribute of the [`current_partition`](index.html#reframe.core.pipeline.RegressionTest.current_partition) and can also be selected through the extended syntax of [`valid_systems`](index.html#reframe.core.pipeline.RegressionTest.valid_systems).\nThe attributes of this object must be alphanumeric strings starting with a non-digit character and their values can be of any type.\n\nNew in version 3.5.0.\n\n####### Container Platform Configuration[¶](#container-platform-configuration)\n\nReFrame can launch containerized applications, but you need to configure properly a system partition in order to do that by defining a container platform configuration.\n\nsystems.partitions.container_platforms.type[¶](#config.systems.partitions.container_platforms.type)\n\nRequired Yes\n\nThe type of the container platform.\nAvailable values are the following:\n\n* `Apptainer`: The [Apptainer](https://apptainer.org/) container runtime.\n* `Docker`: The [Docker](https://www.docker.com/) container runtime.\n* `Sarus`: The [Sarus](https://sarus.readthedocs.io/) container runtime.\n* `Shifter`: The [Shifter](https://github.com/NERSC/shifter) container runtime.\n* `Singularity`: The [Singularity](https://sylabs.io/) container runtime.\n\nsystems.partitions.container_platforms.default[¶](#config.systems.partitions.container_platforms.default)\n\nRequired No\n\nIf set to `true`, this is the default container platform of this partition.\nIf not specified, the default container platform is assumed to be the first in the list of [`container_platforms`](#config.systems.partitions.container_platforms).\n\nNew in version 3.12.0.\n\nsystems.partitions.container_platforms.modules[¶](#config.systems.partitions.container_platforms.modules)\n\nRequired No\n\nDefault\n`[]`\n\nA list of [environment module objects](#module-objects) to be loaded when running containerized tests using this container platform.\n\nsystems.partitions.container_platforms.env_vars[¶](#config.systems.partitions.container_platforms.env_vars)\n\nRequired No\n\nDefault\n`[]`\n\nList of environment variables to be set when running containerized tests using this container platform.\nEach environment variable is specified as a two-element list containing the variable name and its value.\nYou may reference other environment variables when defining an environment variable here.\nReFrame will expand its value.\nVariables are set after the environment modules are loaded.\n\nNew in version 4.0.0.\n\nsystems.partitions.container_platforms.variables[¶](#config.systems.partitions.container_platforms.variables)\n\nDeprecated since version 4.0.0: Please use [`env_vars`](#config.systems.partitions.container_platforms.env_vars) instead.\nIf specified in conjunction with [`env_vars`](#config.systems.partitions.container_platforms.env_vars), it will be ignored.\n\n####### Custom Job Scheduler Resources[¶](#custom-job-scheduler-resources)\n\nReFrame allows you to define custom scheduler resources for each partition that you can then transparently access through the [`extra_resources`](index.html#reframe.core.pipeline.RegressionTest.extra_resources) attribute of a regression test.\n\nsystems.partitions.resources.name[¶](#config.systems.partitions.resources.name)\n\n> required\n> Yes\nThe name of this resources.\nThis name will be used to request this resource in a regression test’s [`extra_resources`](index.html#reframe.core.pipeline.RegressionTest.extra_resources).\n\nsystems.partitions.resources.options[¶](#config.systems.partitions.resources.options)\n\n> required\n> No\n> default\n> `[]`\n> A list of options to be passed to this partition’s job scheduler.\n> The option strings can contain placeholders of the form `{placeholder_name}`.\n> These placeholders may be replaced with concrete values by a regression test through the [`extra_resources`](index.html#reframe.core.pipeline.RegressionTest.extra_resources) attribute.\n> For example, one could define a `gpu` resource for a multi-GPU system that uses Slurm as follows:\n> ```\n> 'resources': [\n> {\n> 'name': 'gpu',\n> 'options': ['--gres=gpu:{num_gpus_per_node}']\n> }\n> ]\n> ```\n> A regression test then may request this resource as follows:\n> ```\n> self.extra_resources = {'gpu': {'num_gpus_per_node': '8'}}\n> ```\n> And the generated job script will have the following line in its preamble:\n> ```\n> #SBATCH --gres=gpu:8\n> ```\n> A resource specification may also start with `#PREFIX`, in which case `#PREFIX` will replace the standard job script prefix of the backend scheduler of this partition.\n> This is useful in cases of job schedulers like Slurm, that allow alternative prefixes for certain features.\n> An example is the [DataWarp](https://www.nersc.gov/assets/Uploads/dw-overview-overby.pdf) functionality of Slurm which is supported by the `#DW` prefix.\n> One could then define DataWarp related resources as follows:\n> ```\n> 'resources': [\n> {\n> 'name': 'datawarp',\n> 'options': [\n> '#DW jobdw capacity={capacity} access_mode={mode} type=scratch',\n> '#DW stage_out source={out_src} destination={out_dst} type={stage_filetype}'\n> ]\n> }\n> ]\n> ```\n> A regression test that needs to make use of that resource, it can set its [`extra_resources`](index.html#reframe.core.pipeline.RegressionTest.extra_resources) as follows:\n> ```\n> self.extra_resources = {\n> 'datawarp': {\n> 'capacity': '100GB',\n> 'mode': 'striped',\n> 'out_src': '$DW_JOB_STRIPED/name',\n> 'out_dst': '/my/file',\n> 'stage_filetype': 'file'\n> }\n> }\n> ```\nNote\n\nFor the `pbs` and `torque` backends, options accepted in the [`access`](#config.systems.partitions.access) and [`resources`](#config.systems.partitions.resources) attributes may either refer to actual `qsub` options or may be just resources specifications to be passed to the `-l` option.\nThe backend assumes a `qsub` option, if the options passed in these attributes start with a `-`.\n\n##### Environment Configuration[¶](#environment-configuration)\n\nEnvironments defined in this section will be used for running regression tests.\nThey are associated with [system partitions](#system-partition-configuration).\n\nenvironments.name[¶](#config.environments.name)\n\nRequired Yes\n\nThe name of this environment.\n\nenvironments.modules[¶](#config.environments.modules)\n\nRequired No\n\nDefault\n`[]`\n\nA list of [environment module objects](#module-objects) to be loaded when this environment is loaded.\n\nenvironments.env_vars[¶](#config.environments.env_vars)\n\nRequired No\n\nDefault\n`[]`\n\nA list of environment variables to be set when loading this environment.\nEach environment variable is specified as a two-element list containing the variable name and its value.\nYou may reference other environment variables when defining an environment variable here.\nReFrame will expand its value.\nVariables are set after the environment modules are loaded.\n\nNew in version 4.0.0.\n\nenvironments.variables[¶](#config.environments.variables)\n\nDeprecated since version 4.0.0: Please use [`env_vars`](#config.environments.env_vars) instead.\nIf specified in conjunction with [`env_vars`](#config.environments.env_vars), it will be ignored.\n\nenvironments.features[¶](#config.environments.features)\n\nRequired No\n\nDefault\n`[]`\n\nUser defined features of the environment.\nThese are accessible through the [`features`](index.html#reframe.core.environments.Environment.features) attribute of the [`current_environ`](index.html#reframe.core.pipeline.RegressionTest.current_environ) and can also be selected through the extended syntax of [`valid_prog_environs`](index.html#reframe.core.pipeline.RegressionTest.valid_prog_environs).\nThe values of this list must be alphanumeric strings starting with a non-digit character and may also contain a `-`.\n\nNew in version 3.11.0.\n\nenvironments.extras[¶](#config.environments.extras)\n\nRequired No\n\nDefault\n`{}`\n\nUser defined attributes of the environment.\nThese are accessible through the `extras` attribute of the [`current_environ`](index.html#reframe.core.pipeline.RegressionTest.current_environ) and can also be selected through the extended syntax of [`valid_prog_environs`](index.html#reframe.core.pipeline.RegressionTest.valid_prog_environs).\nThe attributes of this object must be alphanumeric strings starting with a non-digit character and their values can be of any type.\n\nNew in version 3.9.1.\n\nenvironments.cc[¶](#config.environments.cc)\n\nRequired No\n\nDefault\n`\"cc\"`\n\nThe C compiler to be used with this environment.\n\nenvironments.cxx[¶](#config.environments.cxx)\n\nRequired No\n\nDefault\n`\"CC\"`\n\nThe C++ compiler to be used with this environment.\n\nenvironments.ftn[¶](#config.environments.ftn)\n\nRequired No\n\nDefault\n`\"ftn\"`\n\nThe Fortran compiler to be used with this environment.\n\nenvironments.cppflags[¶](#config.environments.cppflags)\n\nRequired No\n\nDefault\n`[]`\n\nA list of C preprocessor flags to be used with this environment by default.\n\nenvironments.cflags[¶](#config.environments.cflags)\n\nRequired No\n\nDefault\n`[]`\n\nA list of C flags to be used with this environment by default.\n\nenvironments.cxxflags[¶](#config.environments.cxxflags)\n\nRequired No\n\nDefault\n`[]`\n\nA list of C++ flags to be used with this environment by default.\n\nenvironments.fflags[¶](#config.environments.fflags)\n\nRequired No\n\nDefault\n`[]`\n\nA list of Fortran flags to be used with this environment by default.\n\nenvironments.ldflags[¶](#config.environments.ldflags)\n\nRequired No\n\nDefault\n`[]`\n\nA list of linker flags to be used with this environment by default.\n\nenvironments.target_systems[¶](#config.environments.target_systems)\n\nRequired No\n\nDefault\n`[\"*\"]`\n\nA list of systems or system/partitions combinations that this environment definition is valid for.\nA `*` entry denotes any system.\nIn case of multiple definitions of an environment, the most specific to the current system partition will be used.\nFor example, if the current system/partition combination is `daint:mc`, the second definition of the `PrgEnv-gnu` environment will be used:\n```\n'environments': [\n {\n 'name': 'PrgEnv-gnu',\n 'modules': ['PrgEnv-gnu']\n },\n {\n 'name': 'PrgEnv-gnu',\n 'modules': ['PrgEnv-gnu', 'openmpi'],\n 'cc': 'mpicc',\n 'cxx': 'mpicxx',\n 'ftn': 'mpif90',\n 'target_systems': ['daint:mc']\n }\n]\n```\nHowever, if the current system was `daint:gpu`, the first definition would be selected, despite the fact that the second definition is relevant for another partition of the same system.\nTo better understand this, ReFrame resolves definitions in a hierarchical way.\nIt first looks for definitions for the current partition, then for the containing system and, finally, for global definitions (the `*` pseudo-system).\n\n##### Logging Configuration[¶](#logging-configuration)\n\nLogging in ReFrame is handled by logger objects which further delegate message to *logging handlers* which are eventually responsible for emitting or sending the log records to their destinations.\nYou may define different logger objects per system but *not* per partition.\n\nlogging.level[¶](#config.logging.level)\n\nRequired No\n\nDefault\n`\"undefined\"`\n\nThe level associated with this logger object.\nThere are the following levels in decreasing severity order:\n\n* `critical`: Catastrophic errors; the framework cannot proceed with its execution.\n* `error`: Normal errors; the framework may or may not proceed with its execution.\n* `warning`: Warning messages.\n* `info`: Informational messages.\n* `verbose`: More informational messages.\n* `debug`: Debug messages.\n* `debug2`: Further debug messages.\n* `undefined`: This is the lowest level; does not filter any message.\n\nIf a message is logged by the framework, its severity level will be checked by the logger and if it is higher from the logger’s level, it will be passed down to its handlers.\n\nNew in version 3.3: The `debug2` and `undefined` levels are added.\n\nChanged in version 3.3: The default level is now `undefined`.\n\nlogging.handlers[¶](#config.logging.handlers)\n\nRequired Yes\n\nA list of logging handlers responsible for handling normal framework output.\n\nlogging.handlers_perflog[¶](#config.logging.handlers_perflog)\n\nRequired Yes\n\nA list of logging handlers responsible for handling performance data from tests.\n\nlogging.perflog_compat[¶](#config.logging.perflog_compat)\n\nRequired No\n\nDefault\n`false`\n\nEmit a separate log record for each performance variable.\nSet this option to `true` if you want to keep compatibility with the performance logging prior to ReFrame 4.0.\n\nlogging.target_systems[¶](#config.logging.target_systems)\n\nRequired No\n\nDefault\n`[\"*\"]`\n\nA list of systems or system/partitions combinations that this logging configuration is valid for.\nFor a detailed description of this property, have a look at the [`target_systems`](#config.environments.target_systems) definition for environments.\n\n###### Common logging handler properties[¶](#common-logging-handler-properties)\n\nAll logging handlers share the following set of common attributes:\n\nlogging.handlers.type[¶](#config.logging.handlers.type)\n\nlogging.handlers_perflog.type[¶](#config.logging.handlers_perflog.type)\n\nRequired Yes\n\nThe type of handler.\nThere are the following types available:\n\n* `file`: This handler sends log records to file.\nSee [here](#the-file-log-handler) for more details.\n* `filelog`: This handler sends performance log records to files.\nSee [here](#the-filelog-log-handler) for more details.\n* `graylog`: This handler sends performance log records to Graylog.\nSee [here](#the-graylog-log-handler) for more details.\n* `stream`: This handler sends log records to a file stream.\nSee [here](#the-stream-log-handler) for more details.\n* `syslog`: This handler sends log records to a Syslog facility.\nSee [here](#the-syslog-log-handler) for more details.\n* `httpjson`: This handler sends log records in JSON format using HTTP post requests.\nSee [here](#the-httpjson-log-handler) for more details.\n\nlogging.handlers.level[¶](#config.logging.handlers.level)\n\nlogging.handlers_perflog.level[¶](#config.logging.handlers_perflog.level)\n\nRequired No\n\nDefault\n`\"info\"`\n\nThe [log level](#config.logging.level) associated with this handler.\n\nlogging.handlers.format[¶](#config.logging.handlers.format)\n\nlogging.handlers_perflog.format[¶](#config.logging.handlers_perflog.format)\n\nRequired No\n\nDefault\n`\"%(message)s\"`\n\nLog record format string.\n\nReFrame accepts all log record attributes from Python’s [logging](https://docs.python.org/3.8/library/logging.html#logrecord-attributes) mechanism and adds the following attributes:\n\n:header: “Log record attribute”, “Description”[¶](#id2)\n\n| `%(check_build_locally)s` | The value of the [`build_locally`](index.html#reframe.core.pipeline.RegressionTest.build_locally) attribute. |\n| `%(check_build_time_limit)s` | The value of the [`build_time_limit`](index.html#reframe.core.pipeline.RegressionTest.build_time_limit) attribute. |\n| `%(check_descr)s` | The value of the [`descr`](index.html#reframe.core.pipeline.RegressionTest.descr) attribute. |\n| `%(check_display_name)s` | The value of the [`display_name`](index.html#reframe.core.pipeline.RegressionTest.display_name) attribute. |\n| `%(check_environ)s` | The name of the test’s [`current_environ`](index.html#reframe.core.pipeline.RegressionTest.current_environ). |\n| `%(check_env_vars)s` | The value of the [`env_vars`](index.html#reframe.core.pipeline.RegressionTest.env_vars) attribute. |\n| `%(check_exclusive_access)s` | The value of the [`exclusive_access`](index.html#reframe.core.pipeline.RegressionTest.exclusive_access) attribute. |\n| `%(check_executable)s` | The value of the [`executable`](index.html#reframe.core.pipeline.RegressionTest.executable) attribute. |\n| `%(check_executable_opts)s` | The value of the [`executable_opts`](index.html#reframe.core.pipeline.RegressionTest.executable_opts) attribute. |\n| `%(check_extra_resources)s` | The value of the [`extra_resources`](index.html#reframe.core.pipeline.RegressionTest.extra_resources) attribute. |\n| `%(check_hashcode)s` | The unique hash associated with this test. |\n| `%(check_info)s` | Various information about this test; essentially the return value of the test’s [`info()`](index.html#reframe.core.pipeline.RegressionTest.info) function. |\n| `%(check_job_completion_time)s` | Same as the `(check_job_completion_time_unix)s` but formatted according to `datefmt`. |\n| `%(check_job_completion_time_unix)s` | The completion time of the associated run job (see [`completion_time`](index.html#reframe.core.schedulers.Job.completion_time)). |\n| `%(check_job_exitcode)s` | The exit code of the associated run job. |\n| `%(check_job_nodelist)s` | The list of nodes that the associated run job has run on. |\n| `%(check_jobid)s` | The ID of the associated run job. |\n| `%(check_keep_files)s` | The value of the [`keep_files`](index.html#reframe.core.pipeline.RegressionTest.keep_files) attribute. |\n| `%(check_local)s` | The value of the [`local`](index.html#reframe.core.pipeline.RegressionTest.local) attribute. |\n| `%(check_maintainers)s` | The value of the [`maintainers`](index.html#reframe.core.pipeline.RegressionTest.maintainers) attribute. |\n| `%(check_max_pending_time)s` | The value of the [`max_pending_time`](index.html#reframe.core.pipeline.RegressionTest.max_pending_time) attribute. |\n| `%(check_modules)s` | The value of the [`modules`](index.html#reframe.core.pipeline.RegressionTest.modules) attribute. |\n| `%(check_name)s` | The value of the [`name`](index.html#reframe.core.pipeline.RegressionTest.name) attribute. |\n| `%(check_num_cpus_per_task)s` | The value of the [`num_cpus_per_task`](index.html#reframe.core.pipeline.RegressionTest.num_cpus_per_task) attribute. |\n| `%(check_num_gpus_per_node)s` | The value of the [`num_gpus_per_node`](index.html#reframe.core.pipeline.RegressionTest.num_gpus_per_node) attribute. |\n| `%(check_num_tasks)s` | The value of the [`num_tasks`](index.html#reframe.core.pipeline.RegressionTest.num_tasks) attribute. |\n| `%(check_num_tasks_per_core)s` | The value of the [`num_tasks_per_core`](index.html#reframe.core.pipeline.RegressionTest.num_tasks_per_core) attribute. |\n| `%(check_num_tasks_per_node)s` | The value of the [`num_tasks_per_node`](index.html#reframe.core.pipeline.RegressionTest.num_tasks_per_node) attribute. |\n| `%(check_num_tasks_per_socket)s` | The value of the [`num_tasks_per_socket`](index.html#reframe.core.pipeline.RegressionTest.num_tasks_per_socket) attribute. |\n| `%(check_outputdir)s` | The value of the [`outputdir`](index.html#reframe.core.pipeline.RegressionTest.outputdir) attribute. |\n| `%(check_partition)s` | The name of the test’s [`current_partition`](index.html#reframe.core.pipeline.RegressionTest.current_partition). |\n| `%(check_perfvalues)s` | All the performance variables of the test combined. These will be formatted according to [`format_perfvars`](#config.logging.handlers_perflog.format_perfvars). |\n| `%(check_postbuild_cmds)s` | The value of the [`postbuild_cmds`](index.html#reframe.core.pipeline.RegressionTest.postbuild_cmds) attribute. |\n| `%(check_postrun_cmds)s` | The value of the [`postrun_cmds`](index.html#reframe.core.pipeline.RegressionTest.postrun_cmds) attribute. |\n| `%(check_prebuild_cmds)s` | The value of the [`prebuild_cmds`](index.html#reframe.core.pipeline.RegressionTest.prebuild_cmds) attribute. |\n| `%(check_prefix)s` | The value of the [`prefix`](index.html#reframe.core.pipeline.RegressionTest.prefix) attribute. |\n| `%(check_prerun_cmds)s` | The value of the [`prerun_cmds`](index.html#reframe.core.pipeline.RegressionTest.prerun_cmds) attribute. |\n| `%(check_readonly_files)s` | The value of the [`readonly_files`](index.html#reframe.core.pipeline.RegressionTest.readonly_files) attribute. |\n| `%(check_short_name)s` | The value of the [`short_name`](index.html#reframe.core.pipeline.RegressionTest.short_name) attribute. |\n| `%(check_sourcepath)s` | The value of the [`sourcepath`](index.html#reframe.core.pipeline.RegressionTest.sourcepath) attribute. |\n| `%(check_sourcesdir)s` | The value of the [`sourcesdir`](index.html#reframe.core.pipeline.RegressionTest.sourcesdir) attribute. |\n| `%(check_stagedir)s` | The value of the [`stagedir`](index.html#reframe.core.pipeline.RegressionTest.stagedir) attribute. |\n| `%(check_strict_check)s` | The value of the [`strict_check`](index.html#reframe.core.pipeline.RegressionTest.strict_check) attribute. |\n| `%(check_system)s` | The name of the test’s [`current_system`](index.html#reframe.core.pipeline.RegressionTest.current_system). |\n| `%(check_tags)s` | The value of the [`tags`](index.html#reframe.core.pipeline.RegressionTest.tags) attribute. |\n| `%(check_time_limit)s` | The value of the [`time_limit`](index.html#reframe.core.pipeline.RegressionTest.time_limit) attribute. |\n| `%(check_unique_name)s` | The value of the [`unique_name`](index.html#reframe.core.pipeline.RegressionTest.unique_name) attribute. |\n| `%(check_use_multithreading)s` | The value of the [`use_multithreading`](index.html#reframe.core.pipeline.RegressionTest.use_multithreading) attribute. |\n| `%(check_valid_prog_environs)s` | The value of the [`valid_prog_environs`](index.html#reframe.core.pipeline.RegressionTest.valid_prog_environs) attribute. |\n| `%(check_valid_systems)s` | The value of the [`valid_systems`](index.html#reframe.core.pipeline.RegressionTest.valid_systems) attribute. |\n| `%(check_variables)s` | DEPRECATED: Please use `%(check_env_vars)s` instead. |\n| `%(osuser)s` | The name of the OS user running ReFrame. |\n| `%(osgroup)s` | The name of the OS group running ReFrame. |\n| `%(version)s` | The ReFrame version. |\n\nReFrame allows you to log any test variable, parameter or property if they are marked as “loggable”.\nThe log record attribute will have the form `%(check_NAME)s` where `NAME` is the variable name, the parameter name or the property name that is marked as loggable.\n\nNew in version 3.3: Allow arbitrary test attributes to be logged.\n\nNew in version 3.4.2: Allow arbitrary job attributes to be logged.\n\nChanged in version 3.11.0: Limit the number of attributes that can be logged. User attributes or properties must be explicitly marked as “loggable” in order to be selectable for logging.\n\nlogging.handlers.format_perfvars[¶](#config.logging.handlers.format_perfvars)\n\nlogging.handlers_perflog.format_perfvars[¶](#config.logging.handlers_perflog.format_perfvars)\nFormat specifier for logging the performance variables.\n\nThis defines how the `%(check_perfvalues)s` will be formatted.\nSince a test may define multiple performance variables, the formatting specified in this field will be repeated for each performance variable sequentially in the *same* line.\n\nImportant\n\nThe last character of this format will be interpreted as the final delimiter of the formatted performance variables to the rest of the record.\n\nThe following log record attributes are defined additionally by this format specifier:\n\n| Log record attribute | Description |\n| --- | --- |\n| `%(check_perf_lower_thres)s` | The lower threshold of the logged performance variable. |\n| `%(check_perf_ref)s` | The reference value of the logged performance variable. |\n| `%(check_perf_unit)s` | The measurement unit of the logged performance variable. |\n| `%(check_perf_upper)s` | The upper thresholds of the logged performance variable. |\n| `%(check_perf_value)s` | The actual value of the logged performance variable. |\n| `%(check_perf_var)s` | The name of the logged performance variable. |\n\nImportant\n\nReFrame versions prior to 4.0 logged a separate line for each performance variable and the `%(check_perf_*)s` attributes could be used directly in the `format`.\nYou can re-enable this behavior by setting the [`config.logging.perflog_compat`](#config.logging.perflog_compat) logging configuration parameter.\n\nNew in version 4.0.0.\n\nlogging.handlers.datefmt[¶](#config.logging.handlers.datefmt)\n\nlogging.handlers_perflog.datefmt\n\nRequired No\n\nDefault\n`\"%FT%T\"`\n\nTime format to be used for printing timestamps fields.\nThere are two timestamp fields available: `%(asctime)s` and `%(check_job_completion_time)s`.\nIn addition to the format directives supported by the standard library’s [time.strftime()](https://docs.python.org/3.8/library/time.html#time.strftime) function, ReFrame allows you to use the `%:z` directive – a GNU `date` extension – that will print the time zone difference in a RFC3339 compliant way, i.e., `+/-HH:MM` instead of `+/-HHMM`.\n\n###### The `file` log handler[¶](#the-file-log-handler)\n\nThis log handler handles output to normal files.\nThe additional properties for the `file` handler are the following:\n\nlogging.handlers..file..name[¶](#config.logging.handlers..file..name)\n\nlogging.handlers_perflog..file..name[¶](#config.logging.handlers_perflog..file..name)\n\nRequired No\n\nThe name of the file where this handler will write log records.\nIf not specified, ReFrame will create a log file prefixed with `rfm-` in the system’s temporary directory.\n\nChanged in version 3.3: The `name` parameter is no more required and the default log file resides in the system’s temporary directory.\n\nlogging.handlers..file..append[¶](#config.logging.handlers..file..append)\n\nlogging.handlers_perflog..file..ppend[¶](#config.logging.handlers_perflog..file..ppend)\n\nRequired No\n\nDefault\n`false`\n\nControls whether this handler should append to its file or not.\n\nlogging.handlers..file..timestamp[¶](#config.logging.handlers..file..timestamp)\n\nlogging.handlers_perflog..file..timestamp[¶](#config.logging.handlers_perflog..file..timestamp)\n\nRequired No\n\nDefault\n`false`\n\nAppend a timestamp to this handler’s log file.\nThis property may also accept a date format as described in the [`datefmt`](#config.logging.handlers.datefmt) property.\nIf the handler’s [`name`](#config.logging.handlers..file..name) property is set to `filename.log` and this property is set to `true` or to a specific timestamp format, the resulting log file will be `filename_<timestamp>.log`.\n\n###### The `filelog` log handler[¶](#the-filelog-log-handler)\n\nThis handler is meant primarily for performance logging and logs the performance of a test in one or more files.\nThe additional properties for the `filelog` handler are the following:\n\nlogging.handlers..filelog..basedir[¶](#config.logging.handlers..filelog..basedir)\n\nlogging.handlers_perflog..filelog..basedir[¶](#config.logging.handlers_perflog..filelog..basedir)\n\nRequired No\n\nDefault\n`\"./perflogs\"`\n\nThe base directory of performance data log files.\n\nlogging.handlers..filelog..prefix[¶](#config.logging.handlers..filelog..prefix)\n\nlogging.handlers_perflog..filelog..prefix[¶](#config.logging.handlers_perflog..filelog..prefix)\n\nRequired Yes\n\nThis is a directory prefix (usually dynamic), appended to the [`basedir`](#config.logging.handlers..filelog..basedir), where the performance logs of a test will be stored.\nThis attribute accepts any of the check-specific [formatting placeholders](#config.logging.handlers_perflog.format).\nThis allows to create dynamic paths based on the current system, partition and/or programming environment a test executes with.\nFor example, a value of `%(check_system)s/%(check_partition)s` would generate the following structure of performance log files:\n```\n{basedir}/\n system1/\n partition1/\n test_short_name.log\n partition2/\n test_short_name.log\n ...\n system2/\n ...\n```\nlogging.handlers..filelog..append[¶](#config.logging.handlers..filelog..append)\n\nlogging.handlers_perflog..filelog..append[¶](#config.logging.handlers_perflog..filelog..append)\n\nRequired No\n\nDefault\n`true`\n\nOpen each log file in append mode.\n\nChanged in version 4.0.0: The `filelog` handler is very cautious when generating a test log file: if a change is detected in the information that is being logged, the hanlder will not append to the same file, but it will instead create a new one, saving the old file using the `.h<N>` suffix, where `N` is an integer that is increased every time a new file is being created due to such changes.\nExamples of changes in the logged information are when the log record format changes or a new performance metric is added, deleted or has its name changed.\nThis behavior guarantees that each log file is consistent and it will not break existing parsers.\n\n###### The `graylog` log handler[¶](#the-graylog-log-handler)\n\nThis handler sends log records to a [Graylog](https://www.graylog.org/) server.\nThe additional properties for the `graylog` handler are the following:\n\nlogging.handlers..graylog..address[¶](#config.logging.handlers..graylog..address)\n\nlogging.handlers_perflog..graylog..address[¶](#config.logging.handlers_perflog..graylog..address)\n\nRequired Yes\n\nThe address of the Graylog server defined as `host:port`.\n\nlogging.handlers..graylog..extras[¶](#config.logging.handlers..graylog..extras)\n\nlogging.handlers_perflog..graylog..extras[¶](#config.logging.handlers_perflog..graylog..extras)\n\nRequired No\n\nDefault\n`{}`\n\nA set of optional key/value pairs to be passed with each log record to the server.\nThese may depend on the server configuration.\n\nThis log handler uses internally [pygelf](https://pypi.org/project/pygelf/).\nIf `pygelf` is not available, this log handler will be ignored.\n[GELF](http://docs.graylog.org/en/latest/pages/gelf.html) is a format specification for log messages that are sent over the network.\nThe `graylog` handler sends log messages in JSON format using an HTTP POST request to the specified address.\nMore details on this log format may be found [here](http://docs.graylog.org/en/latest/pages/gelf.html#gelf-payload-specification).\nAn example configuration of this handler for performance logging is shown here:\n```\n{\n 'type': 'graylog',\n 'address': 'graylog-server:12345',\n 'level': 'info',\n 'format': '%(message)s',\n 'extras': {\n 'facility': 'reframe',\n 'data-version': '1.0'\n }\n}\n```\nAlthough the [`format`](#config.logging.handlers.format) attribute is defined for this handler, it is not only the log message that will be transmitted the Graylog server.\nThis handler transmits the whole log record, meaning that all the information will be available and indexable at the remote end.\n\n###### The `stream` log handler[¶](#the-stream-log-handler)\n\nThis handler sends log records to a file stream.\nThe additional properties for the `stream` handler are the following:\n\nlogging.handlers..stream..name[¶](#config.logging.handlers..stream..name)\n\nlogging.handlers_perflog..stream..name[¶](#config.logging.handlers_perflog..stream..name)\n\nRequired No\n\nDefault\n`\"stdout\"`\n\nThe name of the file stream to send records to.\nThere are only two available streams:\n\n* `stdout`: the standard output.\n* `stderr`: the standard error.\n\n###### The `syslog` log handler[¶](#the-syslog-log-handler)\n\nThis handler sends log records to UNIX syslog.\nThe additional properties for the `syslog` handler are the following:\n\nlogging.handlers..syslog..socktype[¶](#config.logging.handlers..syslog..socktype)\n\nlogging.handlers_perflog..syslog..socktype[¶](#config.logging.handlers_perflog..syslog..socktype)\n\nRequired No\n\nDefault\n`\"udp\"`\n\nThe socket type where this handler will send log records to.\nThere are two socket types:\n\n* `udp`: A UDP datagram socket.\n* `tcp`: A TCP stream socket.\n\nlogging.handlers..syslog..facility[¶](#config.logging.handlers..syslog..facility)\n\nlogging.handlers_perflog..syslog..facility[¶](#config.logging.handlers_perflog..syslog..facility)\n\nRequired No\n\nDefault\n`\"user\"`\n\nThe Syslog facility where this handler will send log records to.\nThe list of supported facilities can be found [here](https://docs.python.org/3.8/library/logging.handlers.html#logging.handlers.SysLogHandler.encodePriority).\n\nlogging.handlers..syslog..address[¶](#config.logging.handlers..syslog..address)\n\nlogging.handlers_perflog..syslog..address[¶](#config.logging.handlers_perflog..syslog..address)\n\nRequired Yes\n\nThe socket address where this handler will connect to.\nThis can either be of the form `<host>:<port>` or simply a path that refers to a Unix domain socket.\n\n###### The `httpjson` log handler[¶](#the-httpjson-log-handler)\n\nThis handler sends log records in JSON format to a server using HTTP POST requests.\nThe additional properties for the `httpjson` handler are the following:\n\nlogging.handlers..httpjson..url[¶](#config.logging.handlers..httpjson..url)\n\nlogging.handlers_perflog..httpjson..url[¶](#config.logging.handlers_perflog..httpjson..url)\n\nRequired Yes\n\nThe URL to be used in the HTTP(S) request server.\n\nlogging.handlers..httpjson..extras[¶](#config.logging.handlers..httpjson..extras)\n\nlogging.handlers_perflog..httpjson..extras[¶](#config.logging.handlers_perflog..httpjson..extras)\n\nRequired No\n\nDefault\n`{}`\n\nA set of optional key/value pairs to be passed with each log record to the server.\nThese may depend on the server configuration.\n\nlogging.handlers..httpjson..ignore_keys[¶](#config.logging.handlers..httpjson..ignore_keys)\n\nlogging.handlers_perflog..httpjson..ignore_keys[¶](#config.logging.handlers_perflog..httpjson..ignore_keys)\n\nRequired No\n\nDefault\n`[]`\n\nThese keys will be excluded from the log record that will be sent to the server.\n\nThe `httpjson` handler sends log messages in JSON format using an HTTP POST request to the specified URL.\n\nAn example configuration of this handler for performance logging is shown here:\n```\n{\n 'type': 'httpjson',\n 'url': 'http://httpjson-server:12345/rfm',\n 'level': 'info',\n 'extras': {\n 'facility': 'reframe',\n 'data-version': '1.0'\n },\n 'ignore_keys': ['check_perfvalues']\n}\n```\nThis handler transmits the whole log record, meaning that all the information will be available and indexable at the remote end.\n\nlogging.handlers..httpjson..debug[¶](#config.logging.handlers..httpjson..debug)\n\nlogging.handlers_perflog..httpjson..debug[¶](#config.logging.handlers_perflog..httpjson..debug)\n\nRequired No\n\nDefault\n`false`\n\nIf set, the `httpjson` handler will not attempt to send the data to the server, but it will instead dump the JSON record in the current directory.\nThe filename has the following form: `httpjson_record_<timestamp>.json`.\n\nNew in version 4.1.\n\nlogging.handlers..httpjson..json_formatter[¶](#config.logging.handlers..httpjson..json_formatter)\n\nlogging.handlers_perflog..httpjson..json_formatter[¶](#config.logging.handlers_perflog..httpjson..json_formatter)\nA callable for converting the log record into JSON.\n\nThe formatter’s signature is the following:\n\njson_formatter(*record: [object](https://docs.python.org/3/library/functions.html#object)*, *extras: Dict[[str](https://docs.python.org/3/library/stdtypes.html#str), [str](https://docs.python.org/3/library/stdtypes.html#str)]*, *ignore_keys: Set[[str](https://docs.python.org/3/library/stdtypes.html#str)]*) → [str](https://docs.python.org/3/library/stdtypes.html#str)[¶](#config.logging.handlers_perflog..httpjson.json_formatter)\n\nParameters\n* **record** – The prepared log record.\nThe log record is a simple Python object with all the attributes listed in [`format`](#config.logging.handlers.format), as well as all the default Python [log record](https://docs.python.org/3.8/library/logging.html#logrecord-attributes) attributes.\nIn addition to those, there is also the special `__rfm_check__` attribute that contains a reference to the actual test for which the performance is being logged.\n* **extras** – Any extra attributes specified in [`extras`](#config.logging.handlers..httpjson..extras).\n* **ignore_keys** – The set of keys specified in [`ignore_keys`](#config.logging.handlers..httpjson..ignore_keys).\nReFrame always adds the default Python log record attributes in this set.\n\nReturns A string representation of the JSON record to be sent to the server or [`None`](https://docs.python.org/3/library/constants.html#None) if the record should not be sent to the server.\n\nNote\n\nThis configuration parameter can only be used in a Python configuration file.\n\nNew in version 4.1.\n\n##### Execution Mode Configuration[¶](#execution-mode-configuration)\n\nReFrame allows you to define groups of command line options that are collectively called *execution modes*.\nAn execution mode can then be selected from the command line with the [`--mode`](index.html#cmdoption-mode) option.\nThe options of an execution mode will be passed to ReFrame as if they were specified in the command line.\n\nmodes.name[¶](#config.modes.name)\n\nRequired Yes\n\nThe name of this execution mode.\nThis can be used with the [`--mode`](index.html#cmdoption-mode) command line option to invoke this mode.\n\nmodes.options[¶](#config.modes.options)\n\nRequired No\n\nDefault\n`[]`\n\nThe command-line options associated with this execution mode.\n\nmodes.target_systems[¶](#config.modes.target_systems)\n\nRequired No\n\nDefault\n`[\"*\"]`\n\nA list of systems *only* that this execution mode is valid for.\nFor a detailed description of this property, have a look at the [`target_systems`](#config.environments.target_systems) definition for environments.\n\n##### General Configuration[¶](#general-configuration)\n\ngeneral.check_search_path[¶](#config.general.check_search_path)\n\nRequired No\n\nDefault\n`[\"${RFM_INSTALL_PREFIX}/checks/\"]`\n\nA list of paths (files or directories) where ReFrame will look for regression test files.\nIf the search path is set through the environment variable, it should be a colon separated list.\nIf specified from command line, the search path is constructed by specifying multiple times the command line option.\n\ngeneral.check_search_recursive[¶](#config.general.check_search_recursive)\n\nRequired No\n\nDefault\n`false`\n\nSearch directories in the [search path](#general.check_search_path) recursively.\n\ngeneral.clean_stagedir[¶](#config.general.clean_stagedir)\n\nRequired No\n\nDefault\n`true`\n\nClean stage directory of tests before populating it.\n\nNew in version 3.1.\n\ngeneral.colorize[¶](#config.general.colorize)\n\nRequired No\n\nDefault\n`true`\n\nUse colors in output.\nThe command-line option sets the configuration option to `false`.\n\ngeneral.compress_report[¶](#config.general.compress_report)\n\nRequired No\n\nDefault\n`false`\n\nCompress the generated run report file.\nSee the documentation of the [`--compress-report`](index.html#cmdoption-compress-report) option for more information.\n\nNew in version 3.12.0.\n\ngeneral.git_timeout[¶](#config.general.git_timeout)\n\nRequired No\n\nDefault 5\n\nTimeout value in seconds used when checking if a git repository exists.\n\ngeneral.dump_pipeline_progress[¶](#config.general.dump_pipeline_progress)\nDump pipeline progress for the asynchronous execution policy in `pipeline-progress.json`.\nThis option is meant for debug purposes only.\n\nRequired No\n\nDefault\n`False`\n\nNew in version 3.10.0.\n\ngeneral.pipeline_timeout[¶](#config.general.pipeline_timeout)\nTimeout in seconds for advancing the pipeline in the asynchronous execution policy.\n\nReFrame’s asynchronous execution policy will try to advance as many tests as possible in their pipeline, but some tests may take too long to proceed (e.g., due to copying of large files) blocking the advancement of previously started tests.\nIf this timeout value is exceeded and at least one test has progressed, ReFrame will stop processing new tests and it will try to further advance tests that have already started.\nSee [Tweaking the throughput and interactivity of test jobs in the asynchronous execution policy](index.html#pipeline-timeout) for more guidance on how to set this.\n\nRequired No\n\nDefault\n`10`\n\nNew in version 3.10.0.\n\ngeneral.perf_info_level[¶](#config.general.perf_info_level)\n\nRequired No\n\nDefault\n`\"info\"`\n\nThe log level at which the immediate performance info will be printed.\n\nAs soon as a performance test is finished, ReFrame will log its performance on the standard output immediately.\nThis option controls at which verbosity level this info will appear.\n\nFor a list of available log levels, refer to the [`level`](#config.logging.level) logger configuration parameter.\n\nNew in version 4.0.0.\n\ngeneral.remote_detect[¶](#config.general.remote_detect)\n\nRequired No\n\nDefault\n`false`\n\nTry to auto-detect processor information of remote partitions as well.\nThis may slow down the initialization of the framework, since it involves submitting auto-detection jobs to the remote partitions.\nFor more information on how ReFrame auto-detects processor information, you may refer to [Auto-detecting processor information](index.html#proc-autodetection).\n\nNew in version 3.7.0.\n\ngeneral.remote_workdir[¶](#config.general.remote_workdir)\n\nRequired No\n\nDefault\n`\".\"`\n\nThe temporary directory prefix that will be used to create a fresh ReFrame clone, in order to auto-detect the processor information of a remote partition.\n\nNew in version 3.7.0.\n\ngeneral.ignore_check_conflicts[¶](#config.general.ignore_check_conflicts)\n\nRequired No\n\nDefault\n`false`\n\nIgnore test name conflicts when loading tests.\n\nDeprecated since version 3.8.0: This option will be removed in a future version.\n\ngeneral.trap_job_errors[¶](#config.general.trap_job_errors)\n\nRequired No\n\nDefault\n`false`\n\nTrap command errors in the generated job scripts and let them exit immediately.\n\nNew in version 3.2.\n\ngeneral.keep_stage_files[¶](#config.general.keep_stage_files)\n\nRequired No\n\nDefault\n`false`\n\nKeep stage files of tests even if they succeed.\n\ngeneral.module_map_file[¶](#config.general.module_map_file)\n\nRequired No\n\nDefault\n`\"\"`\n\nFile containing module mappings.\n\ngeneral.module_mappings[¶](#config.general.module_mappings)\n\nRequired No\n\nDefault\n`[]`\n\nA list of module mappings.\nIf specified through the environment variable, the mappings must be separated by commas.\nIf specified from command line, multiple module mappings are defined by passing the command line option multiple times.\n\ngeneral.non_default_craype[¶](#config.general.non_default_craype)\n\nRequired No\n\nDefault\n`false`\n\nTest a non-default Cray Programming Environment.\nThis will emit some special instructions in the generated build and job scripts.\nSee also [`--non-default-craype`](index.html#cmdoption-non-default-craype) for more details.\n\ngeneral.purge_environment[¶](#config.general.purge_environment)\n\nRequired No\n\nDefault\n`false`\n\nPurge any loaded environment modules before running any tests.\n\ngeneral.report_file[¶](#config.general.report_file)\n\nRequired No\n\nDefault\n`\"${HOME}/.reframe/reports/run-report-{sessionid}.json\"`\n\nThe file where ReFrame will store its report.\n\nNew in version 3.1.\n\nChanged in version 3.2: Default value has changed to avoid generating a report file per session.\n\nChanged in version 4.0.0: Default value was reverted back to generate a new file per run.\n\ngeneral.report_junit[¶](#config.general.report_junit)\n\nRequired No\n\nDefault\n`null`\n\nThe file where ReFrame will store its report in JUnit format.\nThe report adheres to the XSD schema [here](https://github.com/windyroad/JUnit-Schema/blob/master/JUnit.xsd).\n\nNew in version 3.6.0.\n\ngeneral.resolve_module_conflicts[¶](#config.general.resolve_module_conflicts)\n\nRequired No\n\nDefault\n`true`\n\nReFrame by default resolves any module conflicts and emits the right sequence of `module unload` and `module load` commands, in order to load the requested modules.\nThis option disables this behavior if set to `false`.\n\nYou should avoid using this option for modules system that cannot handle module conflicts automatically, such as early Tmod verions.\n\nDisabling the automatic module conflict resolution, however, can be useful when modules in a remote system partition are not present on the host where ReFrame runs.\nIn order to resolve any module conflicts and generate the right load sequence of modules, ReFrame loads temporarily the requested modules and tracks any conflicts along the way.\nBy disabling this option, ReFrame will simply emit the requested `module load` commands without attempting to load any module.\n\nNew in version 3.6.0.\n\ngeneral.save_log_files[¶](#config.general.save_log_files)\n\nRequired No\n\nDefault\n`false`\n\nSave any log files generated by ReFrame to its output directory\n\ngeneral.target_systems[¶](#config.general.target_systems)\n\nRequired No\n\nDefault\n`[\"*\"]`\n\nA list of systems or system/partitions combinations that these general options are valid for.\nFor a detailed description of this property, have a look at the [`target_systems`](#config.environments.target_systems) definition for environments.\n\ngeneral.timestamp_dirs[¶](#config.general.timestamp_dirs)\n\nRequired No\n\nDefault\n`\"\"`\n\nAppend a timestamp to ReFrame directory prefixes.\nValid formats are those accepted by the [time.strftime()](https://docs.python.org/3.8/library/time.html#time.strftime) function.\nIf specified from the command line without any argument, `\"%FT%T\"` will be used as a time format.\n\ngeneral.unload_modules[¶](#config.general.unload_modules)\n\nRequired No\n\nDefault\n`[]`\n\nA list of [environment module objects](#module-objects) to unload before executing any test.\nIf specified using an the environment variable, a space separated list of modules is expected.\nIf specified from the command line, multiple modules can be passed by passing the command line option multiple times.\n\ngeneral.use_login_shell[¶](#config.general.use_login_shell)\n\nRequired No\n\nDefault\n`false`\n\nUse a login shell for the generated job scripts.\nThis option will cause ReFrame to emit `-l` in the shebang of shell scripts.\nThis option, if set to `true`, may cause ReFrame to fail, if the shell changes permanently to a different directory during its start up.\n\ngeneral.user_modules[¶](#config.general.user_modules)\n\nRequired No\n\nDefault\n`[]`\n\nA list of [environment module objects](#module-objects) to be loaded before executing any test.\nIf specified using an the environment variable, a space separated list of modules is expected.\nIf specified from the command line, multiple modules can be passed by passing the command line option multiple times.\n\ngeneral.verbose[¶](#config.general.verbose)\n\nRequired No\n\nDefault 0\n\nSet the verbosity level of the output.\nThe higher the number, the more verbose the output will be.\nIf set to a negative number, this will decrease the verbosity level.\n\n##### Module Objects[¶](#module-objects)\n\nNew in version 3.3.\n\nA *module object* in ReFrame’s configuration represents an environment module.\nIt can either be a simple string or a JSON object with the following attributes:\n\nenvironments.modules.name[¶](#config.environments.modules.name)\n\nsystems.modules.name[¶](#config.systems.modules.name)\n\nsystems.partitions.modules.name[¶](#config.systems.partitions.modules.name)\n\nsystems.partitions.container_platforms.modules.name[¶](#config.systems.partitions.container_platforms.modules.name)\n\nRequired Yes\n\nThe name of the module.\n\nenvironments.modules.collection[¶](#config.environments.modules.collection)\n\nsystems.modules.collection[¶](#config.systems.modules.collection)\n\nsystems.partitions.modules.collection[¶](#config.systems.partitions.modules.collection)\n\nsystems.partitions.container_platforms.modules.collection[¶](#config.systems.partitions.container_platforms.modules.collection)\n\nRequired No\n\nDefault\n`false`\n\nA boolean value indicating whether this module refers to a module collection.\nModule collections are treated differently from simple modules when loading.\n\nenvironments.modules.path[¶](#config.environments.modules.path)\n\nsystems.modules.path[¶](#config.systems.modules.path)\n\nsystems.partitions.modules.path[¶](#config.systems.partitions.modules.path)\n\nsystems.partitions.container_platforms.modules.path[¶](#config.systems.partitions.container_platforms.modules.path)\n\nRequired No\n\nDefault\n`null`\n\nIf the module is not present in the default `MODULEPATH`, the module’s location can be specified here.\nReFrame will make sure to set and restore the `MODULEPATH` accordingly for loading the module.\n\nNew in version 3.5.0.\n\nSee also\n\nModule collections with [Environment Modules](https://modules.readthedocs.io/en/latest/MIGRATING.html#module-collection) and [Lmod](https://lmod.readthedocs.io/en/latest/010_user.html#user-collections).\n\n##### Processor Info[¶](#processor-info)\n\nNew in version 3.5.0.\n\nA *processor info object* in ReFrame’s configuration is used to hold information about the processor of a system partition and is made available to the tests through the [`processor`](index.html#reframe.core.systems.SystemPartition.processor) attribute of the [`current_partition`](index.html#reframe.core.pipeline.RegressionTest.current_partition).\n\nNote\n\nIn the following the term *logical CPUs* refers to the smallest processing unit recognized by the OS.\nDepending on the microarchitecture, this can either be a core or a hardware thread in processors that support simultaneous multithreading and this feature is enabled.\nTherefore, properties such as `num_cpus_per_core` may have a value greater than one.\n\nsystems.partitions.processor.arch[¶](#config.systems.partitions.processor.arch)\n\nRequired No\n\nDefault\n`None`\n\nThe microarchitecture of the processor.\n\nsystems.partitions.processor.num_cpus[¶](#config.systems.partitions.processor.num_cpus)\n\nRequired No\n\nDefault\n`None`\n\nNumber of logical CPUs.\n\nsystems.partitions.processor.num_cpus_per_core[¶](#config.systems.partitions.processor.num_cpus_per_core)\n\nRequired No\n\nDefault\n`None`\n\nNumber of logical CPUs per core.\n\nsystems.partitions.processor.num_cpus_per_socket[¶](#config.systems.partitions.processor.num_cpus_per_socket)\n\nRequired No\n\nDefault\n`None`\n\nNumber of logical CPUs per socket.\n\nsystems.partitions.processor.num_sockets[¶](#config.systems.partitions.processor.num_sockets)\n\nRequired No\n\nDefault\n`None`\n\nNumber of sockets.\n\nsystems.partitions.processor.topology[¶](#config.systems.partitions.processor.topology)\n\nRequired No\n\nDefault\n`None`\n\nProcessor topology.\nAn example follows:\n```\n'topology': {\n 'numa_nodes': ['0x000000ff'],\n 'sockets': ['0x000000ff'],\n 'cores': ['0x00000003', '0x0000000c',\n '0x00000030', '0x000000c0'],\n 'caches': [\n {\n 'type': 'L3',\n 'size': 6291456,\n 'linesize': 64,\n 'associativity': 0,\n 'num_cpus': 8,\n 'cpusets': ['0x000000ff']\n },\n {\n 'type': 'L2',\n 'size': 262144,\n 'linesize': 64,\n 'associativity': 4,\n 'num_cpus': 2,\n 'cpusets': ['0x00000003', '0x0000000c',\n '0x00000030', '0x000000c0']\n },\n {\n 'type': 'L1',\n 'size': 32768,\n 'linesize': 64,\n 'associativity': 0,\n 'num_cpus': 2,\n 'cpusets': ['0x00000003', '0x0000000c',\n '0x00000030', '0x000000c0']\n }\n ]\n}\n```\n##### Device Info[¶](#device-info)\n\nNew in version 3.5.0.\n\nA *device info object* in ReFrame’s configuration is used to hold information about a specific type of devices in a system partition and is made available to the tests through the [`devices`](index.html#reframe.core.systems.SystemPartition.processor) attribute of the [`current_partition`](index.html#reframe.core.pipeline.RegressionTest.current_partition).\n\nsystems.partitions.devices.type[¶](#config.systems.partitions.devices.type)\n\nRequired No\n\nDefault\n`None`\n\nThe type of the device, for example `\"gpu\"`.\n\nsystems.partitions.devices.arch\n\nRequired No\n\nDefault\n`None`\n\nThe microarchitecture of the device.\n\nsystems.partitions.devices.num_devices[¶](#config.systems.partitions.devices.num_devices)\n\nRequired No\n\nDefault\n`None`\n\nNumber of devices of this type inside the system partition.\n\n#### Programming APIs[¶](#programming-apis)\n\n##### Test API Reference[¶](#test-api-reference)\n\nThis page provides a reference guide of the ReFrame API for writing regression tests covering all the relevant details.\nInternal data structures and APIs are covered only to the extent that this might be helpful to the final user of the framework.\n\n###### Test Base Classes[¶](#module-reframe.core.pipeline)\n\n*class* reframe.core.pipeline.CompileOnlyRegressionTest(**args*, ***kwargs*)[[source]](_modules/reframe/core/pipeline.html#CompileOnlyRegressionTest)[¶](#reframe.core.pipeline.CompileOnlyRegressionTest)\nBases: [`RegressionTest`](#reframe.core.pipeline.RegressionTest)\n\nBase class for compile-only regression tests.\n\nThese tests are by default local and will skip the run phase of the regression test pipeline.\n\nThe standard output and standard error of the test will be set to those of the compilation stage.\n\nThis class is also directly available under the top-level `reframe`\nmodule.\n\nrun()[[source]](_modules/reframe/core/pipeline.html#CompileOnlyRegressionTest.run)[¶](#reframe.core.pipeline.CompileOnlyRegressionTest.run)\nThe run stage of the regression test pipeline.\n\nImplemented as no-op.\n\nrun_wait()[[source]](_modules/reframe/core/pipeline.html#CompileOnlyRegressionTest.run_wait)[¶](#reframe.core.pipeline.CompileOnlyRegressionTest.run_wait)\nWait for this test to finish.\n\nImplemented as no-op\n\nsetup(*partition*, *environ*, ***job_opts*)[[source]](_modules/reframe/core/pipeline.html#CompileOnlyRegressionTest.setup)[¶](#reframe.core.pipeline.CompileOnlyRegressionTest.setup)\nThe setup stage of the regression test pipeline.\n\nSimilar to the [`RegressionTest.setup()`](#reframe.core.pipeline.RegressionTest.setup), except that no run job is created for this test.\n\n*property* stderr[¶](#reframe.core.pipeline.CompileOnlyRegressionTest.stderr)\nThe name of the file containing the standard error of the test.\n\nThis is set during the [`setup()`](#reframe.core.pipeline.CompileOnlyRegressionTest.setup) phase.\n\nThis attribute is evaluated lazily, so it can by used inside sanity expressions.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or `None` if a run job has not yet been created.\n\n*property* stdout[¶](#reframe.core.pipeline.CompileOnlyRegressionTest.stdout)\nThe name of the file containing the standard output of the test.\n\nThis is set during the [`setup()`](#reframe.core.pipeline.CompileOnlyRegressionTest.setup) phase.\n\nThis attribute is evaluated lazily, so it can by used inside sanity expressions.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or `None` if a run job has not yet been created.\n\n*class* reframe.core.pipeline.RegressionMixin(**args*, ***kwargs*)[[source]](_modules/reframe/core/pipeline.html#RegressionMixin)[¶](#reframe.core.pipeline.RegressionMixin)\nBases: [`object`](https://docs.python.org/3/library/functions.html#object)\n\nBase mixin class for regression tests.\n\nMultiple inheritance from more than one\n[`RegressionTest`](#reframe.core.pipeline.RegressionTest) class is not allowed in ReFrame. Hence, mixin classes provide the flexibility to bundle reusable test add-ons, leveraging the metaclass magic implemented in\n`RegressionTestMeta`. Using this metaclass allows mixin classes to use powerful ReFrame features, such as hooks, parameters or variables.\n\nNew in version 3.4.2.\n\n*class* reframe.core.pipeline.RegressionTest(**args*, ***kwargs*)[[source]](_modules/reframe/core/pipeline.html#RegressionTest)[¶](#reframe.core.pipeline.RegressionTest)\nBases: [`RegressionMixin`](#reframe.core.pipeline.RegressionMixin), `JSONSerializable`\n\nBase class for regression tests.\n\nAll regression tests must eventually inherit from this class.\nThis class provides the implementation of the pipeline phases that the regression test goes through during its lifetime.\n\nThis class accepts parameters at the *class definition*, i.e., the test class can be defined as follows:\n```\nclass MyTest(RegressionTest, param='foo', ...):\n```\nwhere `param` is one of the following:\n\nParameters\n* **pin_prefix** – lock the test prefix to the directory where the current class lives.\n* **require_version** – a list of ReFrame version specifications that this test is allowed to run. A version specification string can have one of the following formats:\n\n\t+ `VERSION`: Specifies a single version.\n\t+ `{OP}VERSION`, where `{OP}` can be any of `>`, `>=`, `<`,\n\t`<=`, `==` and `!=`. For example, the version specification\n\tstring `'>=3.5.0'` will allow the following test to be loaded\n\tonly by ReFrame 3.5.0 and higher. The `==VERSION` specification\n\tis the equivalent of `VERSION`.\n\t+ `V1..V2`: Specifies a range of versions.\nThe test will be selected if *any* of the versions is satisfied, even if the versions specifications are conflicting.\n* **special** – allow pipeline stage methods to be overriden in this class.\n\nNote\n\nChanged in version 2.19: Base constructor takes no arguments.\n\nNew in version 3.3: The `pin_prefix` class definition parameter is added.\n\nNew in version 3.7.0: The `require_verion` class definition parameter is added.\n\nWarning\n\nChanged in version 3.4.2: Multiple inheritance with a shared common ancestor is not allowed.\n\nbuild_locally *= True*[¶](#reframe.core.pipeline.RegressionTest.build_locally)\n\nNew in version 3.3.\n\nAlways build the source code for this test locally. If set to\n`False`, ReFrame will spawn a build job on the partition where the test will run. Setting this to `False` is useful when cross-compilation is not supported on the system where ReFrame is run.\nNormally, ReFrame will mark the test as a failure if the spawned job exits with a non-zero exit code. However, certain scheduler backends,\nsuch as the `squeue` do not set it. In such cases, it is the user’s responsibility to check whether the build phase failed by adding an appropriate sanity check.\n\nType boolean\n\nDefault\n`True`\n\nbuild_system *= None*[¶](#reframe.core.pipeline.RegressionTest.build_system)\n\nNew in version 2.14.\n\nThe build system to be used for this test.\nIf not specified, the framework will try to figure it out automatically based on the value of [`sourcepath`](#reframe.core.pipeline.RegressionTest.sourcepath).\n\nThis field may be set using either a string referring to a concrete build system class name\n(see [build systems](#build-systems)) or an instance of\n[`reframe.core.buildsystems.BuildSystem`](#reframe.core.buildsystems.BuildSystem). The former is the recommended way.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or [`reframe.core.buildsystems.BuildSystem`](#reframe.core.buildsystems.BuildSystem).\n\nDefault\n`None`.\n\nbuild_time_limit *= None*[¶](#reframe.core.pipeline.RegressionTest.build_time_limit)\n\nNew in version 3.5.1.\n\nThe time limit for the build job of the regression test.\n\nIt is specified similarly to the [`time_limit`](#reframe.core.pipeline.RegressionTest.time_limit) attribute.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or [`float`](https://docs.python.org/3/library/functions.html#float) or [`int`](https://docs.python.org/3/library/functions.html#int)\n\nDefault\n`None`\n\ncheck_performance()[[source]](_modules/reframe/core/pipeline.html#RegressionTest.check_performance)[¶](#reframe.core.pipeline.RegressionTest.check_performance)\nThe performance checking phase of the regression test pipeline.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – If the performance check fails.\n\nWarning\n\nChanged in version 3.0: You may not override this method directly unless you are in special test. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\nChanged in version 3.4: Overriding this method directly in no longer allowed. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\ncheck_sanity()[[source]](_modules/reframe/core/pipeline.html#RegressionTest.check_sanity)[¶](#reframe.core.pipeline.RegressionTest.check_sanity)\nThe sanity checking phase of the regression test pipeline.\n\nRaises\n* [**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – If the sanity check fails.\n* [**reframe.core.exceptions.ReframeSyntaxError**](index.html#reframe.core.exceptions.ReframeSyntaxError) – If the sanity function cannot be resolved due to ambiguous syntax.\n\nWarning\n\nChanged in version 3.0: You may not override this method directly unless you are in special test. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\nChanged in version 3.4: Overriding this method directly in no longer allowed. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\ncleanup(*remove_files=False*)[[source]](_modules/reframe/core/pipeline.html#RegressionTest.cleanup)[¶](#reframe.core.pipeline.RegressionTest.cleanup)\nThe cleanup phase of the regression test pipeline.\n\nParameters\n**remove_files** – If `True`, the stage directory associated with this test will be removed.\n\nWarning\n\nChanged in version 3.0: You may not override this method directly unless you are in special test. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\nChanged in version 3.4: Overriding this method directly in no longer allowed. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\ncompile()[[source]](_modules/reframe/core/pipeline.html#RegressionTest.compile)[¶](#reframe.core.pipeline.RegressionTest.compile)\nThe compilation phase of the regression test pipeline.\n\nRaises\n[**reframe.core.exceptions.ReframeError**](index.html#reframe.core.exceptions.ReframeError) – In case of errors.\n\nWarning\n\nChanged in version 3.0: You may not override this method directly unless you are in special test. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\nChanged in version 3.4: Overriding this method directly in no longer allowed. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\ncompile_complete()[[source]](_modules/reframe/core/pipeline.html#RegressionTest.compile_complete)[¶](#reframe.core.pipeline.RegressionTest.compile_complete)\nCheck if the build phase has completed.\n\nReturns\n`True` if the associated build job has finished,\n`False` otherwise.\n\nIf no job descriptor is yet associated with this test,\n`True` is returned.\n\nRaises\n[**reframe.core.exceptions.ReframeError**](index.html#reframe.core.exceptions.ReframeError) – In case of errors.\n\ncompile_wait()[[source]](_modules/reframe/core/pipeline.html#RegressionTest.compile_wait)[¶](#reframe.core.pipeline.RegressionTest.compile_wait)\nWait for compilation phase to finish.\n\nNew in version 2.13.\n\nWarning\n\nChanged in version 3.0: You may not override this method directly unless you are in special test. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\nChanged in version 3.4: Overriding this method directly in no longer allowed. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\ncontainer_platform *= _NoRuntime*[¶](#reframe.core.pipeline.RegressionTest.container_platform)\n\nNew in version 2.20.\n\nThe container platform to be used for launching this test.\n\nThis field is set automatically by the default container runtime associated with the current system partition. Users may also set this,\nexplicitly overriding any partition setting. If the\n[`image`](#reframe.core.containers.ContainerPlatform.image) attribute of\n[`container_platform`](#reframe.core.pipeline.RegressionTest.container_platform) is set, then the test will run inside a container using the specified container runtime.\n```\nself.container_platform = 'Singularity'\nself.container_platform.image = 'docker://ubuntu:18.04'\nself.container_platform.command = 'cat /etc/os-release'\n```\nIf the test will run inside a container, the [`executable`](#reframe.core.pipeline.RegressionTest.executable) and\n[`executable_opts`](#reframe.core.pipeline.RegressionTest.executable_opts) attributes are ignored. The container platform’s\n[`command`](#reframe.core.containers.ContainerPlatform.command) will be used instead.\n\nNote\n\nOnly the run phase of the test will run inside the container.\nIf you enable the containerized run in a non run-only test, the compilation phase will still run natively.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or\n[`ContainerPlatform`](#reframe.core.containers.ContainerPlatform).\n\nDefault the container runtime specified in the current system partition’s configuration (see also\n[Container Platform Configuration](index.html#container-platform-configuration)).\n\nChanged in version 3.12.0: This field is now set automatically from the current partition’s configuration.\n\n*property* current_environ[¶](#reframe.core.pipeline.RegressionTest.current_environ)\nThe programming environment that the regression test is currently executing with.\n\nThis is set by the framework during the [`setup()`](#reframe.core.pipeline.RegressionTest.setup) phase.\n\nType\n[`reframe.core.environments.ProgEnvironment`](#reframe.core.environments.ProgEnvironment).\n\n*property* current_partition[¶](#reframe.core.pipeline.RegressionTest.current_partition)\nThe system partition the regression test is currently executing on.\n\nThis is set by the framework during the [`setup()`](#reframe.core.pipeline.RegressionTest.setup) phase.\n\nType\n[`reframe.core.systems.SystemPartition`](#reframe.core.systems.SystemPartition).\n\n*property* current_system[¶](#reframe.core.pipeline.RegressionTest.current_system)\nThe system the regression test is currently executing on.\n\nThis is set by the framework during the initialization phase.\n\nType\n[`reframe.core.systems.System`](#reframe.core.systems.System).\n\ndepends_on(*target*, *how=None*, **args*, ***kwargs*)[[source]](_modules/reframe/core/pipeline.html#RegressionTest.depends_on)[¶](#reframe.core.pipeline.RegressionTest.depends_on)\nAdd a dependency to another test.\n\nParameters\n* **target** – The name of the test that this one will depend on.\n* **how** – A callable that defines how the test cases of this test depend on the the test cases of the target test.\nThis callable should accept two arguments:\n\n\t+ The source test case (i.e., a test case of this test)\n\trepresented as a two-element tuple containing the names of the\n\tpartition and the environment of the current test case.\n\t+ Test destination test case (i.e., a test case of the target\n\ttest) represented as a two-element tuple containing the names of\n\tthe partition and the environment of the current target test\n\tcase.\nIt should return `True` if a dependency between the source and destination test cases exists, `False` otherwise.\n\nThis function will be called multiple times by the framework when the test DAG is constructed, in order to determine the connectivity of the two tests.\n\nIn the following example, this test depends on `T1` when their partitions match, otherwise their test cases are independent.\n```\ndef by_part(src, dst):\n p0, _ = src\n p1, _ = dst\n return p0 == p1\n\nself.depends_on('T0', how=by_part)\n```\nThe framework offers already a set of predefined relations between the test cases of inter-dependent tests. See the\n[`reframe.utility.udeps`](index.html#module-reframe.utility.udeps) for more details.\n\nThe default `how` function is\n[`reframe.utility.udeps.by_case()`](index.html#reframe.utility.udeps.by_case), where test cases on different partitions and environments are independent.\n\nSee also\n\n* [How Test Dependencies Work In ReFrame](index.html#document-dependencies)\n* [Test Case Dependencies Management](index.html#test-case-deps-management)\n\nNew in version 2.21.\n\nChanged in version 3.3: Dependencies between test cases from different partitions are now\n allowed. The `how` argument now accepts a callable.\n\nDeprecated since version 3.3: Passing an integer to the `how` argument as well as using the\n`subdeps` argument is deprecated.\n\nChanged in version 4.0.0: Passing an integer to the `how` argument is no longer supported.\n\ndescr[¶](#reframe.core.pipeline.RegressionTest.descr)\nA detailed description of the test.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`''`\n\nChanged in version 4.0: The default value is now the empty string.\n\n*property* display_name[¶](#reframe.core.pipeline.RegressionTest.display_name)\nA human-readable version of the name this test.\n\nThis name contains a string representation of the various parameters of this specific test variant.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nNote\n\nThe display name may not be unique.\n\nNew in version 3.10.0.\n\nenv_vars *= {}*[¶](#reframe.core.pipeline.RegressionTest.env_vars)\nEnvironment variables to be set before running this test.\n\nThe value of the environment variables can be of any type. ReFrame will invoke `str()` on it whenever it needs to emit it in a script.\n\nType\n`Dict[str, object]`\n\nDefault\n`{}`\n\nNew in version 4.0.0.\n\nexclusive_access *= False*[¶](#reframe.core.pipeline.RegressionTest.exclusive_access)\nSpecify whether this test needs exclusive access to nodes.\n\nType boolean\n\nDefault\n`False`\n\nexecutable[¶](#reframe.core.pipeline.RegressionTest.executable)\nThe name of the executable to be launched during the run phase.\n\nIf this variable is undefined when entering the compile pipeline stage, it will be set to `os.path.join('.', self.unique_name)`.\nClasses that override the compile stage may leave this variable undefined.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`required`\n\nChanged in version 3.7.3: Default value changed from `os.path.join('.', self.unique_name)` to\n`required`.\n\nexecutable_opts *= []*[¶](#reframe.core.pipeline.RegressionTest.executable_opts)\nList of options to be passed to the [`executable`](#reframe.core.pipeline.RegressionTest.executable).\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nextra_resources *= {}*[¶](#reframe.core.pipeline.RegressionTest.extra_resources)\n\nNew in version 2.8.\n\nExtra resources for this test.\n\nThis field is for specifying custom resources needed by this test. These resources are defined in the [configuration](config_reference.html#.systems[].partitions[].resources) of a system partition. For example, assume that two additional resources, named\n`gpu` and `datawarp`, are defined in the configuration file as follows:\n```\n'resources': [\n {\n 'name': 'gpu',\n 'options': ['--gres=gpu:{num_gpus_per_node}']\n },\n {\n 'name': 'datawarp',\n 'options': [\n '#DW jobdw capacity={capacity}',\n '#DW stage_in source={stagein_src}'\n ]\n }\n]\n```\nA regression test may then instantiate the above resources by setting the [`extra_resources`](#reframe.core.pipeline.RegressionTest.extra_resources) attribute as follows:\n```\nself.extra_resources = {\n 'gpu': {'num_gpus_per_node': 2}\n 'datawarp': {\n 'capacity': '100GB',\n 'stagein_src': '/foo'\n }\n}\n```\nThe generated batch script (for Slurm) will then contain the following lines:\n```\n#SBATCH --gres=gpu:2\n#DW jobdw capacity=100GB\n#DW stage_in source=/foo\n```\nNotice that if the resource specified in the configuration uses an alternative directive prefix (in this case `#DW`), this will replace the standard prefix of the backend scheduler (in this case `#SBATCH`)\n\nIf the resource name specified in this variable does not match a resource name in the partition configuration, it will be simply ignored.\nThe [`num_gpus_per_node`](#reframe.core.pipeline.RegressionTest.num_gpus_per_node) attribute translates internally to the\n`_rfm_gpu` resource, so that setting\n`self.num_gpus_per_node = 2` is equivalent to the following:\n```\nself.extra_resources = {'_rfm_gpu': {'num_gpus_per_node': 2}}\n```\nType\n`Dict[str, Dict[str, object]]`\n\nDefault\n`{}`\n\nNote\n\nChanged in version 2.9: A new more powerful syntax was introduced that allows also custom job script directive prefixes.\n\n*property* fixture_variant[¶](#reframe.core.pipeline.RegressionTest.fixture_variant)\nThe point in the fixture space for the test.\n\nThis can be seen as an index to the fixture space representing a unique combination of the fixture variants. This number is directly mapped from `variant_num`.\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\ngetdep(*target*, *environ=None*, *part=None*)[[source]](_modules/reframe/core/pipeline.html#RegressionTest.getdep)[¶](#reframe.core.pipeline.RegressionTest.getdep)\nRetrieve the test case of a target dependency.\n\nThis is a low-level method. The `@require_deps` decorators should be preferred.\n\nParameters\n* **target** – The name of the target dependency to be retrieved.\n* **environ** – The name of the programming environment that will be used to retrieve the test case of the target test. If `None`,\n[`RegressionTest.current_environ`](#reframe.core.pipeline.RegressionTest.current_environ) will be used.\n\nNew in version 2.21.\n\nChanged in version 3.8.0: Setting `environ` or `part` to `'*'` will skip the match check on the environment and partition, respectively.\n\ninfo()[[source]](_modules/reframe/core/pipeline.html#RegressionTest.info)[¶](#reframe.core.pipeline.RegressionTest.info)\nProvide live information for this test.\n\nThis method is used by the front-end to print the status message during the test’s execution. This function is also called to provide the message for the [check_info](config_reference.html#.logging[].handlers[].format) logging attribute.\nBy default, it returns a message reporting the test name, the current partition and the current programming environment that the test is currently executing on.\n\nNew in version 2.10.\n\nReturns a string with an informational message about this test\n\nNote\n\nWhen overriding this method, you should pay extra attention on how you use the [`RegressionTest`](#reframe.core.pipeline.RegressionTest)’s attributes, because this method may be called at any point of the test’s lifetime.\n\nis_dry_run()[[source]](_modules/reframe/core/pipeline.html#RegressionTest.is_dry_run)[¶](#reframe.core.pipeline.RegressionTest.is_dry_run)\nCheck if the test runs in dry-run mode.\n\nNew in version 4.1.\n\nis_fixture()[[source]](_modules/reframe/core/pipeline.html#RegressionTest.is_fixture)[¶](#reframe.core.pipeline.RegressionTest.is_fixture)\nCheck if the test is a fixture.\n\nis_local()[[source]](_modules/reframe/core/pipeline.html#RegressionTest.is_local)[¶](#reframe.core.pipeline.RegressionTest.is_local)\nCheck if the test will execute locally.\n\nA test executes locally if the [`local`](#reframe.core.pipeline.RegressionTest.local) attribute is set or if the current partition’s scheduler does not support job submission.\n\nis_performance_check()[[source]](_modules/reframe/core/pipeline.html#RegressionTest.is_performance_check)[¶](#reframe.core.pipeline.RegressionTest.is_performance_check)\nReturn [`True`](https://docs.python.org/3/library/constants.html#True) if the test is a performance test.\n\n*property* job[¶](#reframe.core.pipeline.RegressionTest.job)\nThe job descriptor associated with this test.\n\nThis is set by the framework during the [`setup()`](#reframe.core.pipeline.RegressionTest.setup) phase.\n\nType\n[`reframe.core.schedulers.Job`](#reframe.core.schedulers.Job).\n\nkeep_files *= []*[¶](#reframe.core.pipeline.RegressionTest.keep_files)\nList of files to be kept after the test finishes.\n\nBy default, the framework saves the standard output, the standard error and the generated shell script that was used to run this test.\n\nThese files will be copied over to the test’s output directory during the [`cleanup()`](#reframe.core.pipeline.RegressionTest.cleanup) phase.\n\nDirectories are also accepted in this field.\n\nRelative path names are resolved against the stage directory.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nChanged in version 3.3: This field accepts now also file glob patterns.\n\nlocal *= False*[¶](#reframe.core.pipeline.RegressionTest.local)\nAlways execute this test locally.\n\nType boolean\n\nDefault\n`False`\n\n*property* logger[¶](#reframe.core.pipeline.RegressionTest.logger)\nA logger associated with this test.\n\nYou can use this logger to log information for your test.\n\nmaintainers *= []*[¶](#reframe.core.pipeline.RegressionTest.maintainers)\nList of people responsible for this test.\n\nWhen the test fails, this contact list will be printed out.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nmax_pending_time *= None*[¶](#reframe.core.pipeline.RegressionTest.max_pending_time)\n\nNew in version 3.0.\n\nThe maximum time a job can be pending before starting running.\n\nTime duration is specified as of the [`time_limit`](#reframe.core.pipeline.RegressionTest.time_limit) attribute.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or [`datetime.timedelta`](https://docs.python.org/3/library/datetime.html#datetime.timedelta)\n\nDefault\n`None`\n\nmodules *= []*[¶](#reframe.core.pipeline.RegressionTest.modules)\nList of modules to be loaded before running this test.\n\nThese modules will be loaded during the [`setup()`](#reframe.core.pipeline.RegressionTest.setup) phase.\n\nType\n`List[str]` or `Dict[str, object]`\n\nDefault\n`[]`\n\n*property* name[¶](#reframe.core.pipeline.RegressionTest.name)\nThe name of the test.\n\nThis is an alias of [`display_name`](#reframe.core.pipeline.RegressionTest.display_name).\n\nnum_cpus_per_task *= None*[¶](#reframe.core.pipeline.RegressionTest.num_cpus_per_task)\nNumber of CPUs per task required by this test.\n\nIgnored if `None`.\n\nType integral or `None`\n\nDefault\n`None`\n\nnum_gpus_per_node *= None*[¶](#reframe.core.pipeline.RegressionTest.num_gpus_per_node)\nNumber of GPUs per node required by this test.\nThis attribute is translated internally to the `_rfm_gpu` resource.\nFor more information on test resources, have a look at the\n[`extra_resources`](#reframe.core.pipeline.RegressionTest.extra_resources) attribute.\n\nType integral or `None`\n\nDefault\n`None`\n\nChanged in version 4.0.0: The default value changed to `None`.\n\nnum_tasks *= 1*[¶](#reframe.core.pipeline.RegressionTest.num_tasks)\nNumber of tasks required by this test.\n\nIf the number of tasks is set to zero or a negative value, ReFrame will try to flexibly allocate the number of tasks based on the command line option [`--flex-alloc-nodes`](index.html#cmdoption-flex-alloc-nodes). A negative number is used to indicate the minimum number of tasks required for the test. In this case the minimum number of tasks is the absolute value of the number, while Setting [`num_tasks`](#reframe.core.pipeline.RegressionTest.num_tasks) to zero is equivalent to setting it to\n[`-num_tasks_per_node`](#reframe.core.pipeline.RegressionTest.num_tasks_per_node).\n\nSetting [`num_tasks`](#reframe.core.pipeline.RegressionTest.num_tasks) to [`None`](https://docs.python.org/3/library/constants.html#None) has a scheduler-specific interpretation, but in principle, passes the responsibility of producing a correct job script to the user by setting the appropriate scheduler options. More specifically, the different backends interpret the\n[`None`](https://docs.python.org/3/library/constants.html#None) [`num_tasks`](#reframe.core.pipeline.RegressionTest.num_tasks) as follows:\n\n* `flux`: not applicable.\n* `local`: not applicable.\n* `lsf`: Neither the `-nnodes` nor the `-n` will be emitted.\n* `oar`: Resets it to `1`.\n* `pbs`: Resets it to `1`.\n* `sge`: not applicable.\n* `slurm`: Neither the `--ntasks` nor the `--nodes` option (if the\n[`use_nodes_option`](index.html#config.systems.partitions.sched_options.use_nodes_option) is specified) will be emitted.\n* `squeue`: See `slurm` backend.\n* `torque`: See `pbs` backend.\n\nType integral or [`None`](https://docs.python.org/3/library/constants.html#None)\n\nDefault\n`1`\n\nNote\n\nChanged in version 2.15: Added support for flexible allocation of the number of tasks if the number of tasks is set to `0`.\n\nChanged in version 2.16: Negative [`num_tasks`](#reframe.core.pipeline.RegressionTest.num_tasks) is allowed for specifying the minimum number of required tasks by the test.\n\nChanged in version 4.1: Allow [`num_tasks`](#reframe.core.pipeline.RegressionTest.num_tasks) to be [`None`](https://docs.python.org/3/library/constants.html#None).\n\nnum_tasks_per_core *= None*[¶](#reframe.core.pipeline.RegressionTest.num_tasks_per_core)\nNumber of tasks per core required by this test.\n\nIgnored if `None`.\n\nType integral or `None`\n\nDefault\n`None`\n\nnum_tasks_per_node *= None*[¶](#reframe.core.pipeline.RegressionTest.num_tasks_per_node)\nNumber of tasks per node required by this test.\n\nIgnored if `None`.\n\nType integral or `None`\n\nDefault\n`None`\n\nnum_tasks_per_socket *= None*[¶](#reframe.core.pipeline.RegressionTest.num_tasks_per_socket)\nNumber of tasks per socket required by this test.\n\nIgnored if `None`.\n\nType integral or `None`\n\nDefault\n`None`\n\n*property* outputdir[¶](#reframe.core.pipeline.RegressionTest.outputdir)\nThe output directory of the test.\n\nThis is set during the [`setup()`](#reframe.core.pipeline.RegressionTest.setup) phase.\n\nNew in version 2.13.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str).\n\n*property* param_variant[¶](#reframe.core.pipeline.RegressionTest.param_variant)\nThe point in the parameter space for the test.\n\nThis can be seen as an index to the paraemter space representing a unique combination of the parameter values. This number is directly mapped from `variant_num`.\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\nperf_patterns[¶](#reframe.core.pipeline.RegressionTest.perf_patterns)\nPatterns for verifying the performance of this test.\n\nIf set to `None`, no performance checking will be performed.\n\nType A dictionary with keys of type [`str`](https://docs.python.org/3/library/stdtypes.html#str) and deferrable expressions (i.e., the result of a [sanity function](index.html#document-deferrable_functions_reference)) as values.\n`None` is also allowed.\n\nDefault\n`None`\n\nWarning\n\nYou are advised to follow the new syntax for defining performance variables in your tests using either the [`@performance_function`](#reframe.core.builtins.performance_function) builtin or the\n[`perf_variables`](#reframe.core.pipeline.RegressionTest.perf_variables), as [`perf_patterns`](#reframe.core.pipeline.RegressionTest.perf_patterns) will likely be deprecated in the future.\n\nperf_variables *= {}*[¶](#reframe.core.pipeline.RegressionTest.perf_variables)\nThe performance variables associated with the test.\n\nIn this context, a performance variable is a key-value pair, where the key is the desired variable name and the value is the deferred performance expression (i.e. the result of a [deferrable performance function](index.html#deferrable-performance-functions)) that computes or extracts the performance variable’s value.\n\nBy default, ReFrame will populate this field during the test’s instantiation with all the member functions decorated with the\n[`@performance_function`](#reframe.core.builtins.performance_function) decorator.\nIf no performance functions are present in the class, no performance checking or reporting will be carried out.\n\nThis mapping may be extended or replaced by other performance variables that may be defined in any pipeline hook executing before the performance stage. To this end, deferred performance functions can be created inline using the utility\n[`make_performance_function()`](index.html#reframe.utility.sanity.make_performance_function).\n\nRefer to the [ReFrame Tutorials](index.html#document-tutorials) for concrete usage examples.\n\nType A dictionary with keys of type [`str`](https://docs.python.org/3/library/stdtypes.html#str) and deferred performance expressions as values (see\n[Deferrable performance functions](index.html#deferrable-performance-functions)).\n\nDefault Collection of performance variables associated to each of the member functions decorated with the [`@performance_function`](#reframe.core.builtins.performance_function)\ndecorator.\n\nNew in version 3.8.0.\n\npostbuild_cmds *= []*[¶](#reframe.core.pipeline.RegressionTest.postbuild_cmds)\n\nNew in version 3.0.\n\nList of shell commands to be executed after a successful compilation.\n\nThese commands are emitted in the script after the actual build commands generated by the selected [build system](#reframe.core.pipeline.RegressionTest.build_system).\n\nType\n`List[str]`\n\nDefault\n`[]`\n\npostrun_cmds *= []*[¶](#reframe.core.pipeline.RegressionTest.postrun_cmds)\n\nNew in version 3.0.\n\nList of shell commands to execute after the parallel launch command.\n\nSee [`prerun_cmds`](#reframe.core.pipeline.RegressionTest.prerun_cmds) for a more detailed description of the semantics.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nprebuild_cmds *= []*[¶](#reframe.core.pipeline.RegressionTest.prebuild_cmds)\n\nNew in version 3.0.\n\nList of shell commands to be executed before compiling.\n\nThese commands are emitted in the build script before the actual build commands generated by the selected [build system](#reframe.core.pipeline.RegressionTest.build_system).\n\nType\n`List[str]`\n\nDefault\n`[]`\n\n*property* prefix[¶](#reframe.core.pipeline.RegressionTest.prefix)\nThe prefix directory of the test.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str).\n\nprerun_cmds *= []*[¶](#reframe.core.pipeline.RegressionTest.prerun_cmds)\n\nNew in version 3.0.\n\nList of shell commands to execute before the parallel launch command.\n\nThese commands do not execute in the context of ReFrame.\nInstead, they are emitted in the generated job script just before the actual job launch command.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nreadonly_files *= []*[¶](#reframe.core.pipeline.RegressionTest.readonly_files)\nList of files or directories (relative to the [`sourcesdir`](#reframe.core.pipeline.RegressionTest.sourcesdir)) that will be symlinked in the stage directory and not copied.\n\nYou can use this variable to avoid copying very large files to the stage directory.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nreference *= {}*[¶](#reframe.core.pipeline.RegressionTest.reference)\nThe set of reference values for this test.\n\nThe reference values are specified as a scoped dictionary keyed on the performance variables defined in [`perf_patterns`](#reframe.core.pipeline.RegressionTest.perf_patterns) and scoped under the system/partition combinations.\nThe reference itself is a four-tuple that contains the reference value,\nthe lower and upper thresholds and the measurement unit.\n\nAn example follows:\n```\nself.reference = {\n 'sys0:part0': {\n 'perfvar0': (50, -0.1, 0.1, 'Gflop/s'),\n 'perfvar1': (20, -0.1, 0.1, 'GB/s')\n },\n 'sys0:part1': {\n 'perfvar0': (100, -0.1, 0.1, 'Gflop/s'),\n 'perfvar1': (40, -0.1, 0.1, 'GB/s')\n }\n}\n```\nTo better understand how to set the performance reference tuple, here are some examples with both positive and negative reference values:\n\n> | **Performance Tuple** | **Expected** | **Lowest** | **Highest** |\n> | `(100, -0.01, 0.02, 'MB/s')` | 100 MB/s | 99 MB/s | 102 MB/s |\n> | `(100, -0.01, None, 'MB/s')` | 100 MB/s | 99 MB/s | inf MB/s |\n> | `(100, None, 0.02, 'MB/s')` | 100 MB/s | -inf MB/s | 102 MB/s |\n> | `(-100, -0.01, 0.02, 'C')` | -100 C | -101 C | -98 C |\n> | `(-100, -0.01, None, 'C')` | -100 C | -101 C | inf C |\n> | `(-100, None, 0.02, 'C')` | -100 C | -inf C | -98 C |\nDuring the performance stage of the pipeline, the reference tuple elements, except the unit, are passed to the\n[`assert_reference()`](index.html#reframe.utility.sanity.assert_reference) function along with the obtained performance value in order to actually assess whether the test passes the performance check or not.\n\nType A scoped dictionary with system names as scopes, performance variables as keys and reference tuples as values.\nThe elements of reference tuples cannot be deferrable expressions.\n\nNote\n\nChanged in version 3.0: The measurement unit is required. The user should explicitly specify `None` if no unit is available.\n\nChanged in version 3.8.0: The unit in the reference tuple is again optional, but it is recommended to use it for clarity.\n\nChanged in version 4.0.0: Deferrable expressions are not allowed in reference tuples.\n\nrequire_reference *= False*[¶](#reframe.core.pipeline.RegressionTest.require_reference)\nRequire that a reference is defined for each system that this test is run on.\n\nIf this is set and a reference is not found for the current system, the test will fail.\n\nType boolean\n\nDefault\n`False`\n\nNew in version 4.0.0.\n\nrun()[[source]](_modules/reframe/core/pipeline.html#RegressionTest.run)[¶](#reframe.core.pipeline.RegressionTest.run)\nThe run phase of the regression test pipeline.\n\nThis call is non-blocking.\nIt simply submits the job associated with this test and returns.\n\nWarning\n\nChanged in version 3.0: You may not override this method directly unless you are in special test. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\nChanged in version 3.4: Overriding this method directly in no longer allowed. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\nrun_complete()[[source]](_modules/reframe/core/pipeline.html#RegressionTest.run_complete)[¶](#reframe.core.pipeline.RegressionTest.run_complete)\nCheck if the run phase has completed.\n\nReturns\n`True` if the associated job has finished,\n`False` otherwise.\n\nIf no job descriptor is yet associated with this test,\n`True` is returned.\n\nRaises\n[**reframe.core.exceptions.ReframeError**](index.html#reframe.core.exceptions.ReframeError) – In case of errors.\n\nWarning\n\nYou may not override this method directly unless you are in special test. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\nChanged in version 3.4: Overriding this method directly in no longer allowed. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\nrun_wait()[[source]](_modules/reframe/core/pipeline.html#RegressionTest.run_wait)[¶](#reframe.core.pipeline.RegressionTest.run_wait)\nWait for the run phase of this test to finish.\n\nRaises\n[**reframe.core.exceptions.ReframeError**](index.html#reframe.core.exceptions.ReframeError) – In case of errors.\n\nWarning\n\nYou may not override this method directly unless you are in special test. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\nChanged in version 3.4: Overriding this method directly in no longer allowed. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\nsanity_patterns[¶](#reframe.core.pipeline.RegressionTest.sanity_patterns)\nRefer to the [ReFrame Tutorials](index.html#document-tutorials) for concrete usage examples.\n\nIf not set, a sanity error may be raised during sanity checking if no other sanity checking functions already exist.\n\nType A deferrable expression (i.e., the result of a [sanity function](index.html#document-deferrable_functions_reference))\n\nDefault\n`required`\n\nNote\n\nChanged in version 2.9: The default behaviour has changed and it is now considered a sanity failure if this attribute is set to `required`.\n\nIf a test doesn’t care about its output, this must be stated explicitly as follows:\n```\nself.sanity_patterns = sn.assert_true(1)\n```\nChanged in version 3.6: The default value has changed from `None` to `required`.\n\nset_var_default(*name*, *value*)[[source]](_modules/reframe/core/pipeline.html#RegressionTest.set_var_default)[¶](#reframe.core.pipeline.RegressionTest.set_var_default)\nSet the default value of a variable if variable is undefined.\n\nA variable is undefined if it is declared and required and no value is yet assigned to it.\n\nParameters\n* **name** – The name of the variable.\n* **value** – The value to set the variable to.\n\nRaises\n[**ValueError**](https://docs.python.org/3/library/exceptions.html#ValueError) – If the variable does not exist\n\nNew in version 3.10.1.\n\nsetup(*partition*, *environ*, ***job_opts*)[[source]](_modules/reframe/core/pipeline.html#RegressionTest.setup)[¶](#reframe.core.pipeline.RegressionTest.setup)\nThe setup phase of the regression test pipeline.\n\nParameters\n* **partition** – The system partition to set up this test for.\n* **environ** – The environment to set up this test for.\n* **job_opts** – Options to be passed through to the backend scheduler.\nWhen overriding this method users should always pass through\n`job_opts` to the base class method.\n\nRaises\n[**reframe.core.exceptions.ReframeError**](index.html#reframe.core.exceptions.ReframeError) – In case of errors.\n\nWarning\n\nChanged in version 3.0: You may not override this method directly unless you are in special test. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\nChanged in version 3.4: Overriding this method directly in no longer allowed. See [here](migration_2_to_3.html#force-override-a-pipeline-method) for more details.\n\n*property* short_name[¶](#reframe.core.pipeline.RegressionTest.short_name)\nA short version of the test’s display name.\n\nThe shortened version coincides with the [`unique_name`](#reframe.core.pipeline.RegressionTest.unique_name) for simple tests and combines the test’s class name and a hash code for parameterised tests.\n\nNew in version 4.0.0.\n\nskip(*msg=None*)[[source]](_modules/reframe/core/pipeline.html#RegressionTest.skip)[¶](#reframe.core.pipeline.RegressionTest.skip)\nSkip test.\n\nParameters\n**msg** – A message explaining why the test was skipped.\n\nNew in version 3.5.1.\n\nskip_if(*cond*, *msg=None*)[[source]](_modules/reframe/core/pipeline.html#RegressionTest.skip_if)[¶](#reframe.core.pipeline.RegressionTest.skip_if)\nSkip test if condition is true.\n\nParameters\n* **cond** – The condition to check for skipping the test.\n* **msg** – A message explaining why the test was skipped.\n\nNew in version 3.5.1.\n\nskip_if_no_procinfo(*msg=None*)[[source]](_modules/reframe/core/pipeline.html#RegressionTest.skip_if_no_procinfo)[¶](#reframe.core.pipeline.RegressionTest.skip_if_no_procinfo)\nSkip test if no processor topology information is available.\n\nThis method has effect only if called after the `setup` stage.\n\nParameters\n**msg** – A message explaining why the test was skipped.\nIf not specified, a default message will be used.\n\nNew in version 3.9.1.\n\nsourcepath[¶](#reframe.core.pipeline.RegressionTest.sourcepath)\nThe path to the source file or source directory of the test.\n\nIt must be a path relative to the [`sourcesdir`](#reframe.core.pipeline.RegressionTest.sourcesdir), pointing to a subfolder or a file contained in [`sourcesdir`](#reframe.core.pipeline.RegressionTest.sourcesdir). This applies also in the case where [`sourcesdir`](#reframe.core.pipeline.RegressionTest.sourcesdir) is a Git repository.\n\nIf it refers to a regular file, this file will be compiled using the\n[`SingleSource`](#reframe.core.buildsystems.SingleSource) build system.\nIf it refers to a directory, ReFrame will try to infer the build system to use for the project and will fall back in using the [`Make`](#reframe.core.buildsystems.Make) build system, if it cannot find a more specific one.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`''`\n\nsourcesdir *= src*[¶](#reframe.core.pipeline.RegressionTest.sourcesdir)\nThe directory containing the test’s resources.\n\nThis directory may be specified with an absolute path or with a path relative to the location of the test. Its contents will always be copied to the stage directory of the test.\n\nThis attribute may also accept a URL, in which case ReFrame will treat it as a Git repository and will try to clone its contents in the stage directory of the test.\n\nIf set to `None`, the test has no resources an no action is taken.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or `None`\n\nDefault\n`'src'` if such a directory exists at the test level,\notherwise `None`\n\nNote\n\nChanged in version 2.9: Allow `None` values to be set also in regression tests with a compilation phase\n\nChanged in version 2.10: Support for Git repositories was added.\n\nChanged in version 3.0: Default value is now conditionally set to either `'src'` or\n`None`.\n\n*property* stagedir[¶](#reframe.core.pipeline.RegressionTest.stagedir)\nThe stage directory of the test.\n\nThis is set during the [`setup()`](#reframe.core.pipeline.RegressionTest.setup) phase.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str).\n\n*property* stderr[¶](#reframe.core.pipeline.RegressionTest.stderr)\nThe name of the file containing the standard error of the test.\n\nThis is set during the [`setup()`](#reframe.core.pipeline.RegressionTest.setup) phase.\n\nThis attribute is evaluated lazily, so it can by used inside sanity expressions.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or `None` if a run job has not yet been created.\n\n*property* stdout[¶](#reframe.core.pipeline.RegressionTest.stdout)\nThe name of the file containing the standard output of the test.\n\nThis is set during the [`setup()`](#reframe.core.pipeline.RegressionTest.setup) phase.\n\nThis attribute is evaluated lazily, so it can by used inside sanity expressions.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or `None` if a run job has not yet been created.\n\nstrict_check *= True*[¶](#reframe.core.pipeline.RegressionTest.strict_check)\nMark this test as a strict performance test.\n\nIf a test is marked as non-strict, the performance checking phase will always succeed, unless the `--strict` command-line option is passed when invoking ReFrame.\n\nType boolean\n\nDefault\n`True`\n\ntags *= set()*[¶](#reframe.core.pipeline.RegressionTest.tags)\nSet of tags associated with this test.\n\nThis test can be selected from the frontend using any of these tags.\n\nType\n`Set[str]`\n\nDefault an empty set\n\ntime_limit *= None*[¶](#reframe.core.pipeline.RegressionTest.time_limit)\nTime limit for this test.\n\nTime limit is specified as a string in the form\n`<days>d<hours>h<minutes>m<seconds>s` or as number of seconds. If set to `None`, the\n[`time_limit`](#reframe.core.systems.SystemPartition.time_limit) of the current system partition will be used.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or [`float`](https://docs.python.org/3/library/functions.html#float) or [`int`](https://docs.python.org/3/library/functions.html#int)\n\nDefault\n`None`\n\nNote\n\nChanged in version 2.15: This attribute may be set to `None`.\n\nWarning\n\nChanged in version 3.0: The old syntax using a `(h, m, s)` tuple is deprecated.\n\nChanged in version 3.2: - The old syntax using a `(h, m, s)` tuple is dropped.\n- Support of timedelta objects is dropped.\n- Number values are now accepted.\n\nChanged in version 3.5.1: The default value is now `None` and it can be set globally per partition via the configuration.\n\n*property* unique_name[¶](#reframe.core.pipeline.RegressionTest.unique_name)\nThe unique name of this test.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nNew in version 3.10.0.\n\nuse_multithreading *= None*[¶](#reframe.core.pipeline.RegressionTest.use_multithreading)\nSpecify whether this tests needs simultaneous multithreading enabled.\n\nIgnored if `None`.\n\nType boolean or `None`\n\nDefault\n`None`\n\nvalid_prog_environs[¶](#reframe.core.pipeline.RegressionTest.valid_prog_environs)\nList of programming environments supported by this test.\n\nThe syntax of this attribute is exactly the same as of the\n[`valid_systems`](#reframe.core.pipeline.RegressionTest.valid_systems) except that the `a:b` entries are invalid.\n\nType\n`List[str]`\n\nDefault\n`required`\n\nSee also\n\n* [Environment features](config_reference.html#environments-.features)\n* [Environment extras](config_reference.html#environments-.extras)\n\nChanged in version 2.12: Programming environments can now be specified using wildcards.\n\nChanged in version 2.17: Support for wildcards is dropped.\n\nChanged in version 3.3: Default value changed from `[]` to `None`.\n\nChanged in version 3.6: Default value changed from `None` to `required`.\n\nChanged in version 3.11.0: Extend syntax to support features and key/value pairs.\n\nvalid_systems[¶](#reframe.core.pipeline.RegressionTest.valid_systems)\nList of systems or system features or system properties required by this test.\n\nEach entry in this list is a requirement and can have one of the following forms:\n\n* `sysname`: The test is valid for system named `sysname`.\n* `sysname:partname`: The test is valid for the partition `partname`\nof system `sysname`.\n* `*`: The test is valid for any system.\n* `*:partname`: The test is valid for any partition named `partname`\nin any system.\n* `+feat`: The test is valid for all partitions that define feature\n`feat` as a feature.\n* `-feat`: The test is valid for all partitions that do not define feature `feat` as a feature.\n* `%key=val`: The test is valid for all partitions that define the extra property `key` with the value `val`.\n\nMultiple features and key/value pairs can be included in a single entry of the [`valid_systems`](#reframe.core.pipeline.RegressionTest.valid_systems) list, in which case an AND operation on these constraints is implied. For example, the test defining the following will be valid for all systems that have define both `feat1`\nand `feat2` and set `foo=1`\n```\nvalid_systems = ['+feat1 +feat2 %foo=1']\n```\nFor key/value pairs comparisons, ReFrame will automatically convert the value in the key/value spec to the type of the value of the corresponding entry in the partitions `extras` property. In the above example, if the type of `foo` property is integer, `1` will be converted to an integer value. If a conversion to the target type is not possible, then the requested key/value pair is not matched.\n\nMultiple entries in the [`valid_systems`](#reframe.core.pipeline.RegressionTest.valid_systems) list are implicitly ORed,\nsuch that the following example implies that the test is valid for either `sys1` or for any other system that does not define `feat`.\n```\nvalid_systems = ['sys1', '-feat']\n```\nType\n`List[str]`\n\nDefault\n`None`\n\nSee also\n\n* [System partition features](config_reference.html#systems-.partitions-.features)\n* [System partition extras](config_reference.html#systems-.partitions-.extras)\n\nChanged in version 3.3: Default value changed from `[]` to `None`.\n\nChanged in version 3.6: Default value changed from `None` to `required`.\n\nChanged in version 3.11.0: Extend syntax to support features and key/value pairs.\n\nvariables *= {}*[¶](#reframe.core.pipeline.RegressionTest.variables)\nEnvironment variables to be set before running this test.\n\nThis is an alias of [`env_vars`](#reframe.core.pipeline.RegressionTest.env_vars).\n\nDeprecated since version 4.0.0: Please use [`env_vars`](#reframe.core.pipeline.RegressionTest.env_vars) instead.\n\n*property* variant_num[¶](#reframe.core.pipeline.RegressionTest.variant_num)\nThe variant number of the test.\n\nThis number should be treated as a unique ID representing a unique combination of the available parameter and fixture variants.\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\n*class* reframe.core.pipeline.RunOnlyRegressionTest(**args*, ***kwargs*)[[source]](_modules/reframe/core/pipeline.html#RunOnlyRegressionTest)[¶](#reframe.core.pipeline.RunOnlyRegressionTest)\nBases: [`RegressionTest`](#reframe.core.pipeline.RegressionTest)\n\nBase class for run-only regression tests.\n\nThis class is also directly available under the top-level `reframe`\nmodule.\n\ncompile()[[source]](_modules/reframe/core/pipeline.html#RunOnlyRegressionTest.compile)[¶](#reframe.core.pipeline.RunOnlyRegressionTest.compile)\nThe compilation phase of the regression test pipeline.\n\nThis is a no-op for this type of test.\n\ncompile_wait()[[source]](_modules/reframe/core/pipeline.html#RunOnlyRegressionTest.compile_wait)[¶](#reframe.core.pipeline.RunOnlyRegressionTest.compile_wait)\nWait for compilation phase to finish.\n\nThis is a no-op for this type of test.\n\nrun()[[source]](_modules/reframe/core/pipeline.html#RunOnlyRegressionTest.run)[¶](#reframe.core.pipeline.RunOnlyRegressionTest.run)\nThe run phase of the regression test pipeline.\n\nThe resources of the test are copied to the stage directory and the rest of execution is delegated to the [`RegressionTest.run()`](#reframe.core.pipeline.RegressionTest.run).\n\nsetup(*partition*, *environ*, ***job_opts*)[[source]](_modules/reframe/core/pipeline.html#RunOnlyRegressionTest.setup)[¶](#reframe.core.pipeline.RunOnlyRegressionTest.setup)\nThe setup stage of the regression test pipeline.\n\nSimilar to the [`RegressionTest.setup()`](#reframe.core.pipeline.RegressionTest.setup), except that no build job is created for this test.\n\n###### Test Decorators[¶](#test-decorators)\n\n@reframe.core.decorators.simple_test[[source]](_modules/reframe/core/decorators.html#simple_test)[¶](#reframe.core.decorators.simple_test)\nClass decorator for registering tests with ReFrame.\n\nThe decorated class must derive from\n[`reframe.core.pipeline.RegressionTest`](#reframe.core.pipeline.RegressionTest). This decorator is also available directly under the `reframe` module.\n\nNew in version 2.13.\n\n###### Builtins[¶](#builtins)\n\nNew in version 3.4.2.\n\nReFrame test base classes and, in particular, the [`reframe.core.pipeline.RegressionMixin`](#reframe.core.pipeline.RegressionMixin) class, define a set of functions and decorators that can be used to define essential test elements, such as variables, parameters, fixtures, pipeline hooks etc.\nThese are called *builtins* because they are directly available for use inside the test class body that is being defined without the need to import any module.\nHowever, almost all of these builtins are also available from the `reframe.core.builtins` module.\nThe use of this module is required only when creating new tests programmatically using the [`make_test()`](#reframe.core.meta.make_test) function.\n\nreframe.core.pipeline.RegressionMixin.bind(*func*, *name=None*)[¶](#reframe.core.pipeline.reframe.core.pipeline.RegressionMixin.bind)\nBind a free function to a regression test.\n\nBy default, the function is bound with the same name as the free function.\nHowever, the function can be bound using a different name with the `name` argument.\n\nParameters\n* **func** – external function to be bound to a class.\n* **name** – bind the function under a different name.\n\nNote\n\nThis is the only builtin that is not available through the `reframe.core.builtins` module.\nThe reason is that the [`bind()`](#reframe.core.pipeline.reframe.core.pipeline.RegressionMixin.bind) method needs to access the class namespace directly in order to bind the free function to the class.\n\nNew in version 3.6.2.\n\n@reframe.core.builtins.deferrable[[source]](_modules/reframe/core/deferrable.html#deferrable)[¶](#reframe.core.builtins.deferrable)\nConvert the decorated function to a deferred expression.\n\nSee [Deferrable Functions Reference](index.html#deferrable-functions) for further information on deferrable functions.\n\nreframe.core.builtins.fixture(*cls*, ***, *scope='test'*, *action='fork'*, *variants='all'*, *variables=None*)[¶](#reframe.core.builtins.fixture)\nInsert a new fixture in the current test.\n\nA fixture is a regression test that creates, prepares and/or manages a resource for another regression test. Fixtures may contain other fixtures and so on, forming a directed acyclic graph. A parent fixture (or a regular regression test) requires the resources managed by its child fixtures in order to run, and it may only access these fixture resources after its `setup` pipeline stage. The execution of parent fixtures is postponed until all their respective children have completed execution.\nHowever, the destruction of the resources managed by a fixture occurs in reverse order, only after all the parent fixtures have been destroyed.\nThis destruction of resources takes place during the `cleanup` pipeline stage of the regression test. Fixtures must not define the members\n[`valid_systems`](#reframe.core.pipeline.RegressionTest.valid_systems) and\n[`valid_prog_environs`](#reframe.core.pipeline.RegressionTest.valid_prog_environs). These variables are defined based on the values specified in the parent test,\nensuring that the fixture runs with a suitable system partition and programming environment combination. A fixture’s\n[`name`](#reframe.core.pipeline.RegressionTest.name) attribute may be internally mangled depending on the arguments passed during the fixture declaration. Hence, manually setting or modifying the\n[`name`](#reframe.core.pipeline.RegressionTest.name) attribute in the fixture class is disallowed, and breaking this restriction will result in undefined behavior.\n\nWarning\n\nThe fixture name mangling is considered an internal framework mechanism and it may change in future versions without any notice. Users must not express any logic in their tests that relies on a given fixture name mangling scheme.\n\nBy default, the resources managed by a fixture are private to the parent test. However, it is possible to share these resources across different tests by passing the appropriate fixture `scope` argument. The different scope levels are independent from each other and a fixture only executes once per scope, where all the tests that belong to that same scope may use the same resources managed by a given fixture instance. The available scopes are:\n\n> * **session**: This scope encloses all the tests and fixtures that run\n> in the full ReFrame session. This may include tests that use different\n> system partition and programming environment combinations. The fixture\n> class must derive from\n> [`RunOnlyRegressionTest`](#reframe.core.pipeline.RunOnlyRegressionTest) to avoid any\n> implicit dependencies on the partition or the programming environment\n> used.\n> * **partition**: This scope spans across a single system partition. This\n> may include different tests that run on the same partition but use\n> different programming environments. Fixtures with this scope must be\n> independent of the programming environment, which restricts the\n> fixture class to derive from\n> [`RunOnlyRegressionTest`](#reframe.core.pipeline.RunOnlyRegressionTest).\n> * **environment**: The extent of this scope covers a single combination\n> of system partition and programming environment. Since the fixture is\n> guaranteed to have the same partition and programming environment as\n> the parent test, the fixture class can be any derived class from\n> [`RegressionTest`](#reframe.core.pipeline.RegressionTest). * **test**: This scope\n> covers a single instance of the parent test, where the resources\n> provided by the fixture are exclusive to each parent test instance.\n> The fixture class can be any derived class from\n> [`RegressionTest`](#reframe.core.pipeline.RegressionTest).\nRather than specifying the scope at the fixture class definition, ReFrame fixtures set the scope level from the consumer side (i.e. when used by another test or fixture). A test may declare multiple fixtures using the same class, where fixtures with different scopes are guaranteed to point to different instances of the fixture class. On the other hand, when two or more fixtures use the same fixture class and have the same scope, these different fixtures will point to the same underlying resource if the fixtures refer to the same [variant](#test-variants) of the fixture class. The example below illustrates the different fixture scope usages:\n```\nclass MyFixture(rfm.RunOnlyRegressionTest):\n my_var = variable(int, value=1)\n ...\n\n@rfm.simple_test class TestA(rfm.RegressionTest):\n valid_systems = ['p1', 'p2']\n valid_prog_environs = ['e1', 'e2']\n\n # Fixture shared throughout the full session\n f1 = fixture(MyFixture, scope='session')\n\n # Fixture shared for each supported partition\n f2 = fixture(MyFixture, scope='partition')\n\n # Fixture shared for each supported part+environ\n f3 = fixture(MyFixture, scope='environment')\n\n # Fixture private evaluation of MyFixture\n f4 = fixture(MyFixture, scope='test')\n ...\n\n@rfm.simple_test class TestB(rfm.RegressionTest):\n valid_systems = ['p1']\n valid_prog_environs = ['e1']\n\n # Another private instance of MyFixture\n f1 = fixture(MyFixture, scope='test')\n\n # Same as f3 in TestA for p1 + e1\n f2 = fixture(MyFixture, scope='environment')\n\n # Same as f1 in TestA\n f3 = fixture(MyFixture, scope='session')\n ...\n\n @run_after('setup')\n def access_fixture_resources(self):\n # Dummy pipeline hook to illustrate fixture resource access\n assert self.f1.my_var is not self.f2.my_var\n assert self.f1.my_var is not self.f3.my_var\n```\n`TestA` supports two different valid systems and another two valid programming environments. Assuming that both environments are supported by each of the system partitions `'p1'` and `'p2'`, this test will execute a total of four times. This test uses the very simple\n`MyFixture` fixture multiple times using different scopes, where fixture `f1` (session scope) will be shared across the four test instances, and fixture `f4` (test scope) will be executed once per test instance. On the other hand, `f2` (partition scope) will run once per partition supported by test `TestA`, and the multiple per-partition executions (i.e. for each programming environment) will share the same underlying resource for `f2`. Lastly, `f3` will run a total of four times, which is once per partition and environment combination. This simple `TestA` shows how multiple instances from the same test can share resources, but the real power behind fixtures is illustrated with\n`TestB`, where this resource sharing is extended across different tests. For simplicity, `TestB` only supports a single partition\n`'p1'` and programming environment `'e1'`, and similarly to\n`TestA`, `f1` (test scope) causes a private evaluation of the fixture `MyFixture`. However, the resources managed by fixtures\n`f2` (environment scope) and `f3` (session scope) are shared with\n`Test1`.\n\nFixtures are treated by ReFrame as first-class ReFrame tests, which means that these classes can use the same built-in functionalities as in regular tests decorated with\n[`@rfm.simple_test`](#reframe.core.decorators.simple_test). This includes the `parameter()`\nbuilt-in, where fixtures may have more than one\n[variant](#test-variants). When this occurs, a parent test may select to either treat a parameterized fixture as a test parameter, or instead,\nto gather all the fixture variants from a single instance of the parent test. In essence, fixtures implement fork-join model whose behavior may be controlled through the `action` argument. This argument may be set to one of the following options:\n\n> * **fork**: This option parameterizes the parent test as a function of\n> the fixture variants. The fixture handle will resolve to a single\n> instance of the fixture.\n> * **join**: This option gathers all the variants from a fixture into a\n> single instance of the parent test. The fixture handle will point to a\n> list containing all the fixture variants.\nA test may declare multiple fixtures with different `action` options,\nwhere the default `action` option is `'fork'`. The example below illustrates the behavior of these two different options.\n```\nclass ParamFix(rfm.RegressionTest):\n p = parameter(range(5)) # A simple test parameter\n ...\n\n@rfm.simple_test class TestC(rfm.RegressionTest):\n # Parameterize TestC for each ParamFix variant\n f = fixture(ParamFix, action='fork')\n ...\n\n @run_after('setup')\n def access_fixture_resources(self):\n print(self.f.p) # Prints the fixture's variant parameter value\n\n@rfm.simple_test class TestD(rfm.RegressionTest):\n # Gather all fixture variants into a single test\n f = fixture(ParamFix, action='join')\n ...\n\n @run_after('setup')\n def reduce_range(self):\n # Sum all the values of p for each fixture variant\n res = functools.reduce(lambda x, y: x+y,\n (fix.p for fix in self.f))\n n = len(self.f)-1\n assert res == (n*n + n)/2\n```\nHere `ParamFix` is a simple fixture class with a single parameter.\nWhen the test `TestC` uses this fixture with a `'fork'` action,\nthe test is implicitly parameterized over each variant of\n`ParamFix`. Hence, when the `access_fixture_resources()`\npost-setup hook accesses the fixture `f`, it only access a single instance of the `ParamFix` fixture. On the other hand, when this same fixture is used with a `'join'` action by `TestD`, the test is not parameterized and all the `ParamFix` instances are gathered into `f` as a list. Thus, the post-setup pipeline hook\n`reduce_range()` can access all the fixture variants and compute a reduction of the different `p` values.\n\nWhen declaring a fixture, a parent test may select a subset of the fixture variants through the `variants` argument. This variant selection can be done by either passing an iterable containing valid variant indices (see\n[Test variants](#test-variants) for further information on how the test variants are indexed), or instead, passing a mapping with the parameter name (of the fixture class) as keys and filtering functions as values. These filtering functions are unary functions that return the value of a boolean expression on the values of the specified parameter, and they all must evaluate to `True` for at least one of the fixture class variants.\nSee the example below for an illustration on how to filter-out fixture variants.\n```\nclass ComplexFixture(rfm.RegressionTest):\n # A fixture with 400 different variants.\n p0 = parameter(range(100))\n p1 = parameter(['a', 'b', 'c', 'd'])\n ...\n\n@rfm.simple_test class TestE(rfm.RegressionTest):\n # Select the fixture variants with boolean conditions\n foo = fixture(ComplexFixture,\n variants={'p0': lambda x: x<10,\n 'p1': lambda x: x=='d'})\n\n # Select the fixture variants by index\n bar = fixture(ComplexFixture, variants=range(300,310))\n ...\n```\nA parent test may also specify the value of different variables in the fixture class to be set before its instantiation. Each variable must have been declared in the fixture class with the\n`variable()` built-in,\notherwise it is silently ignored. This variable specification is equivalent to deriving a new class from the fixture class, and setting these variable values in the class body of a newly derived class.\nTherefore, when fixture declarations use the same fixture class and pass different values to the `variables` argument, the fixture class is interpreted as a different class for each of these fixture declarations.\nSee the example below.\n```\nclass Fixture(rfm.RegressionTest):\n v = variable(int, value=1)\n ...\n\n@rfm.simple_test class TestF(rfm.RegressionTest):\n foo = fixture(Fixture)\n bar = fixture(Fixture, variables={'v':5})\n baz = fixture(Fixture, variables={'v':10})\n ...\n\n @run_after('setup')\n def print_fixture_variables(self):\n print(self.foo.v) # Prints 1\n print(self.bar.v) # Prints 5\n print(self.baz.v) # Prints 10\n```\nThe test `TestF` declares the fixtures `foo`, `bar` and `baz`\nusing the same `Fixture` class. If no variables were set in `bar`\nand `baz`, this would result into the same fixture being declared multiple times in the same scope (implicitly set to `'test'`), which would lead to a single instance of `Fixture` being referred to by\n`foo`, `bar` and `baz`. However, in this case ReFrame identifies that the declared fixtures pass different values to the `variables`\nargument in the fixture declaration, and executes these three fixtures separately.\n\nNote\n\nMappings passed to the `variables` argument that define the same class variables in different order are interpreted as the same value.\nThe two fixture declarations below are equivalent, and both `foo` and\n`bar` will point to the same instance of the fixture class\n`MyResource`.\n```\nfoo = fixture(MyResource, variables={'a':1, 'b':2})\nbar = fixture(MyResource, variables={'b':2, 'a':1})\n```\n**Early access to fixture objects**\n\nThe test instance represented by a fixture can be accessed fully from within a test only after the setup stage. The reason for that is that fixtures eventually translate into test dependencies and access to the parent dependencies cannot happen before the this stage.\n\nHowever, it is often useful, especially in the case of parameterized fixtures, to be able to access the fixture parameters earlier, e.g., in a post-init hook in order to properly set the\n[`valid_systems`](#reframe.core.pipeline.RegressionTest.valid_systems) and\n[`valid_prog_environs`](#reframe.core.pipeline.RegressionTest.valid_prog_environs) of the test. These attributes cannot be set later than the test’s initialization in order to have an effect.\n\nFor this reason, early access to fixture objects is allowed *only* for retrieving their parameters.\n```\nclass Fixture(rfm.RegressionTest):\n x = parameter([1, 2, 3])\n\nclass Test(rfm.RunOnlyRegressionTest):\n foo = fixture(Fixture)\n executable = './myexec'\n valid_prog_environs = ['*']\n\n @run_after('init')\n def early_access(self):\n # Only fixture parameters can be accessed here!\n if self.foo.x == 1:\n self.valid_systems = ['sys1]\n else:\n self.valid_systems = ['sys2']\n\n @run_after('setup')\n def normal_access(self):\n # Any test attribute of the associated fixture test can be\n # accessed here\n self.executable_opts = [\n '-i', os.path.join(self.foo.stagedir, 'input.txt')\n ]\n```\nDuring test initialization, ReFrame binds the `foo` name to a proxy object that holds the parameterization of the target fixture. This proxy object is recursive, so that if fixture `foo` contained another fixture named `bar`, it would allow you to access any parameters of that fixture with `self.foo.bar.param`.\n\nDuring the test setup stage, the `foo`’s binding changes and it is now bound to the exact test instance that was executed for the target test instance.\n\nParameters\n* **cls** – A class derived from\n[`RegressionTest`](#reframe.core.pipeline.RegressionTest) that manages a given resource. The base from this class may be further restricted to other derived classes of [`RegressionTest`](#reframe.core.pipeline.RegressionTest)\ndepending on the `scope` parameter.\n* **scope** – Sets the extent to which other regression tests may share the resources managed by a fixture. The available scopes are, from more to less restrictive, `'test'`, `'environment'`,\n`'partition'` and `'session'`. By default a fixture’s scope is set to `'test'`, which makes the resource private to the test that uses the fixture. This means that when multiple regression tests use the same fixture class with a `'test'` scope, the fixture will run once per regression test. When the scope is set to `'environment'`, the resources managed by the fixture are shared across all the tests that use the fixture and run on the same system partition and use the same programming environment. When the scope is set to `'partition'`, the resources managed by the fixture are shared instead across all the tests that use the fixture and run on the same system partition.\nLastly, when the scope is set to `'session'`, the resources managed by the fixture are shared across the full ReFrame session. Fixtures with either `'partition'` or `'session'` scopes may be shared across different regression tests under different programming environments, and for this reason, when using these two scopes, the fixture class `cls` is required to derive from\n[`RunOnlyRegressionTest`](#reframe.core.pipeline.RunOnlyRegressionTest).\n* **action** – Set the behavior of a parameterized fixture to either\n`'fork'` or `'join'`. With a `'fork'` action, a parameterized fixture effectively parameterizes the regression test. On the other hand, a `'join'` action gathers all the fixture variants into the same instance of the regression test. By default, the `action`\nparameter is set to `'fork'`.\n* **variants** – Filter or sub-select a subset of the variants from a parameterized fixture. This argument can be either an iterable with the indices from the desired variants, or a mapping containing unary functions that return the value of a boolean expression on the values of a given parameter.\n* **variables** – Mapping to set the values of fixture’s variables. The variables are set after the fixture class has been created (i.e. after the class body has executed) and before the fixture class is instantiated.\n\nNew in version 3.9.0.\n\nChanged in version 3.11.0: Allow early access of fixture objects.\n\n@reframe.core.builtins.loggable_as(*name*)[[source]](_modules/reframe/core/builtins.html#loggable_as)[¶](#reframe.core.builtins.loggable_as)\nMark a property as loggable.\n\nParameters\n**name** – An alternative name that will be used for logging this property. If [`None`](https://docs.python.org/3/library/constants.html#None), the name of the decorated property will be used.\n\nRaises\n[**ValueError**](https://docs.python.org/3/library/exceptions.html#ValueError) – if the decorated function is not a property.\n\nNew in version 3.10.2.\n\n@reframe.core.builtins.loggable[¶](#reframe.core.builtins.loggable)\nEquivalent to [`loggable_as(None)`](#reframe.core.builtins.loggable_as).\n\nreframe.core.builtins.parameter(*values=None*, *inherit_params=False*, *filter_params=None*, *fmt=None*, *loggable=False*)[¶](#reframe.core.builtins.parameter)\nInserts a new test parameter.\n\nAt the class level, these parameters are stored in a separate namespace referred to as the *parameter space*. If a parameter with a matching name is already present in the parameter space of a parent class, the existing parameter values will be combined with those provided by this method following the inheritance behavior set by the arguments `inherit_params`\nand `filter_params`. Instead, if no parameter with a matching name exists in any of the parent parameter spaces, a new regression test parameter is created. A regression test can be parameterized as follows:\n```\nclass Foo(rfm.RegressionTest):\n variant = parameter(['A', 'B'])\n\n # print(variant)\n # Error: a parameter may only be accessed from the class instance\n\n @run_after('init')\n def do_something(self):\n if self.variant == 'A':\n do_this()\n else:\n do_other()\n```\nOne of the most powerful features of these built-in functions is that they store their input information at the class level. However, a parameter may only be accessed from the class instance and accessing it directly from the class body is disallowed. With this approach, extending or specializing an existing parameterized regression test becomes straightforward, since the test attribute additions and modifications made through built-in functions in the parent class are automatically inherited by the child test. For instance, continuing with the example above, one could override the `do_something()` hook in the `Foo`\nregression test as follows:\n```\nclass Bar(Foo):\n @run_after('init')\n def do_something(self):\n if self.variant == 'A':\n override_this()\n else:\n override_other()\n```\nMoreover, a derived class may extend, partially extend and/or modify the parameter values provided in the base class as shown below.\n```\nclass ExtendVariant(Bar):\n # Extend the full set of inherited variant parameter values\n # to ['A', 'B', 'C']\n variant = parameter(['C'], inherit_params=True)\n\nclass PartiallyExtendVariant(Bar):\n # Extend a subset of the inherited variant parameter values\n # to ['A', 'D']\n variant = parameter(['D'], inherit_params=True,\n filter_params=lambda x: x[:1])\n\nclass ModifyVariant(Bar):\n # Modify the variant parameter values to ['AA', 'BA']\n variant = parameter(inherit_params=True,\n filter_params=lambda x: map(lambda y: y+'A', x))\n```\nA parameter with no values is referred to as an *abstract parameter* (i.e.\na parameter that is declared but not defined). Therefore, classes with at least one abstract parameter are considered abstract classes.\n```\nclass AbstractA(Bar):\n variant = parameter()\n\nclass AbstractB(Bar):\n variant = parameter(inherit_params=True, filter_params=lambda x: [])\n```\nParameters\n* **values** – An iterable containing the parameter values.\n* **inherit_params** – If [`True`](https://docs.python.org/3/library/constants.html#True), the parameter values defined in any base class will be inherited. In this case, the parameter values provided in the current class will extend the set of inherited parameter values. If the parameter does not exist in any of the parent parameter spaces, this option has no effect.\n* **filter_params** – Function to filter/modify the inherited parameter values that may have been provided in any of the parent parameter spaces. This function must accept a single iterable argument and return an iterable. It will be called with the inherited parameter values and it must return the filtered set of parameter values. This function will only have an effect if used with\n`inherit_params=True`.\n* **fmt** – A formatting function that will be used to format the values of this parameter in the test’s\n[`display_name`](#reframe.core.pipeline.RegressionTest.display_name). This function should take as argument the parameter value and return a string representation of the value. If the returned value is not a string, it will be converted using the `str()` function.\n* **loggable** – Mark this parameter as loggable. If [`True`](https://docs.python.org/3/library/constants.html#True), this parameter will become a log record attribute under the name\n`check_NAME`, where `NAME` is the name of the parameter.\n\nReturns A new test parameter.\n\nNew in version 3.10.0: The `fmt` argument is added.\n\nNew in version 3.11.0: The `loggable` argument is added.\n\n@reframe.core.builtins.performance_function(*unit*, ***, *perf_key=None*)[[source]](_modules/reframe/core/builtins.html#performance_function)[¶](#reframe.core.builtins.performance_function)\nDecorate a test member function to mark it as a performance metric function.\n\nThis decorator converts the decorated method into a performance deferrable function (see “[Deferrable performance functions](index.html#deferrable-performance-functions)” for more details)\nwhose evaluation is deferred to the performance stage of the regression test. The decorated function must take a single argument without a default value (i.e. `self`) and any number of arguments with default values. A test may decorate multiple member functions as performance functions,\nwhere each of the decorated functions must be provided with the unit of the performance quantity to be extracted from the test. Any performance function may be overridden in a derived class and multiple bases may define their own performance functions. In the event of a name conflict,\nthe derived class will follow Python’s [MRO](https://docs.python.org/3/library/stdtypes.html#class.__mro__) to choose the appropriate performance function. However, defining more than one performance function with the same name in the same class is disallowed.\n\nThe full set of performance functions of a regression test is stored under\n[`perf_variables`](#reframe.core.pipeline.RegressionTest.perf_variables) as key-value pairs, where, by default, the key is the name of the decorated member function, and the value is the deferred performance function itself.\nOptionally, the key under which a performance function is stored in\n[`perf_variables`](#reframe.core.pipeline.RegressionTest.perf_variables) can be customised by passing the desired key as the `perf_key` argument to this decorator.\n\nParameters\n**unit** – A string representing the measurement unit of this metric.\n\nNew in version 3.8.0.\n\n@reframe.core.builtins.require_deps[[source]](_modules/reframe/core/hooks.html#require_deps)[¶](#reframe.core.builtins.require_deps)\nDecorator to denote that a function will use the test dependencies.\n\nThe arguments of the decorated function must be named after the dependencies that the function intends to use. The decorator will bind the arguments to a partial realization of the\n[`getdep()`](#reframe.core.pipeline.RegressionTest.getdep) function, such that conceptually the new function arguments will be the following:\n```\nnew_arg = functools.partial(getdep, orig_arg_name)\n```\nThe converted arguments are essentially functions accepting a single argument, which is the target test’s programming environment.\nAdditionally, this decorator will attach the function to run *after* the test’s setup phase, but *before* any other “post-setup” pipeline hook.\n\nWarning\n\nChanged in version 3.7.0: Using this functionality from the `reframe` or\n`reframe.core.decorators` modules is now deprecated. You should use the built-in function described here.\n\nChanged in version 4.0.0: You may only use this function as framework built-in.\n\n@reframe.core.builtins.run_after(*stage*)[[source]](_modules/reframe/core/builtins.html#run_after)[¶](#reframe.core.builtins.run_after)\nAttach the decorated function after a certain pipeline stage.\n\nThis is analogous to `run_before()`, except that the hook will execute right after the stage it was attached to. This decorator also supports `'init'` as a valid `stage` argument, where in this case, the hook will execute right after the test is initialized (i.e.\nafter the `__init__()` method is called) and before entering the test’s pipeline. In essence, a post-init hook is equivalent to defining additional `__init__()` functions in the test. The following code\n```\nclass MyTest(rfm.RegressionTest):\n @run_after('init')\n def foo(self):\n self.x = 1\n```\nis equivalent to\n```\nclass MyTest(rfm.RegressionTest):\n def __init__(self):\n self.x = 1\n```\nChanged in version 3.5.2: Add support for post-init hooks.\n\n@reframe.core.builtins.run_before(*stage*)[[source]](_modules/reframe/core/builtins.html#run_before)[¶](#reframe.core.builtins.run_before)\nAttach the decorated function before a certain pipeline stage.\n\nThe function will run just before the specified pipeline stage and it cannot accept any arguments except `self`. This decorator can be stacked, in which case the function will be attached to multiple pipeline stages. See above for the valid `stage` argument values.\n\nParameters\n**stage** – The pipeline stage where this function will be attached to.\nSee [Pipeline Hooks](#pipeline-hooks) for the list of valid stage values.\n\n@reframe.core.builtins.sanity_function[[source]](_modules/reframe/core/builtins.html#sanity_function)[¶](#reframe.core.builtins.sanity_function)\nDecorate a test member function to mark it as a sanity check.\n\nThis decorator will convert the given function into a\n`deferrable()` and mark it to be executed during the test’s sanity stage. When this decorator is used, manually assigning a value to `sanity_patterns` in the test is not allowed.\n\nDecorated functions may be overridden by derived classes, and derived classes may also decorate a different method as the test’s sanity function. Decorating multiple member functions in the same class is not allowed. However, a `RegressionTest` may inherit from multiple\n`RegressionMixin` classes with their own sanity functions. In this case, the derived class will follow Python’s [MRO](https://docs.python.org/3/library/stdtypes.html#class.__mro__) to find a suitable sanity function.\n\nNew in version 3.7.0.\n\nreframe.core.builtins.variable(**args*, ***kwargs*)[¶](#reframe.core.builtins.variable)\nInsert a new test variable.\n\nDeclaring a test variable through the [`variable()`](#reframe.core.builtins.variable) built-in allows for a more robust test implementation than if the variables were just defined as regular test attributes (e.g. `self.a = 10`). Using variables declared through the [`variable()`](#reframe.core.builtins.variable) built-in guarantees that these regression test variables will not be redeclared by any child class, while also ensuring that any values that may be assigned to such variables comply with its original declaration. In essence, declaring test variables with the [`variable()`](#reframe.core.builtins.variable) built-in removes any potential test errors that might be caused by accidentally overriding a class attribute. See the example below.\n```\nclass Foo(rfm.RegressionTest):\n my_var = variable(int, value=8)\n not_a_var = my_var - 4\n\n @run_after('init')\n def access_vars(self):\n print(self.my_var) # prints 8.\n # self.my_var = 'override' # Error: my_var must be an int!\n self.not_a_var = 'override' # This will work, but is dangerous!\n self.my_var = 10 # tests may also assign values the standard way\n```\nHere, the argument `value` in the [`variable()`](#reframe.core.builtins.variable) built-in sets the default value for the variable. This value may be accessed directly from the class body, as long as it was assigned before either in the same class body or in the class body of a parent class. This behavior extends the standard Python data model, where a regular class attribute from a parent class is never available in the class body of a child class. Hence, using the [`variable()`](#reframe.core.builtins.variable) built-in enables us to directly use or modify any variables that may have been declared upstream the class inheritance chain, without altering their original value at the parent class level.\n```\nclass Bar(Foo):\n print(my_var) # prints 8\n # print(not_a_var) # This is standard Python and raises a NameError\n\n # Since my_var is available, we can also update its value:\n my_var = 4\n\n # Bar inherits the full declaration of my_var with the original\n # type-checking.\n # my_var = 'override' # Wrong type error again!\n\n @run_after('init')\n def access_vars(self):\n print(self.my_var) # prints 4\n print(self.not_a_var) # prints 4\n\nprint(Foo.my_var) # prints 8 print(Bar.my_var) # prints 4\n```\nHere, `Bar` inherits the variables from `Foo` and can see that `my_var` has already been declared in the parent class. Therefore,\nthe value of `my_var` is updated ensuring that the new value complies to the original variable declaration. However, the value of `my_var` at\n`Foo` remains unchanged.\n\nThese examples above assumed that a default value can be provided to the variables in the bases tests, but that might not always be the case. For example, when writing a test library, one might want to leave some variables undefined and force the user to set these when using the test.\nAs shown in the example below, imposing such requirement is as simple as not passing any `value` to the [`variable()`](#reframe.core.builtins.variable) built-in, which marks the given variable as *required*.\n```\n# Test as written in the library class EchoBaseTest(rfm.RunOnlyRegressionTest):\n what = variable(str)\n\n valid_systems = ['*']\n valid_prog_environs = ['*']\n\n @run_before('run')\n def set_executable(self):\n self.executable = f'echo {self.what}'\n\n @sanity_function\n def assert_what(self):\n return sn.assert_found(fr'{self.what}')\n\n# Test as written by the user\n@rfm.simple_test class HelloTest(EchoBaseTest):\n what = 'Hello'\n\n# A parameterized test with type-checking\n@rfm.simple_test class FoodTest(EchoBaseTest):\n param = parameter(['Bacon', 'Eggs'])\n\n @run_after('init')\n def set_vars_with_params(self):\n self.what = self.param\n```\nSimilarly to a variable with a value already assigned to it, the value of a required variable may be set either directly in the class body, on the\n`__init__()` method, or in any other hook before it is referenced.\nOtherwise an error will be raised indicating that a required variable has not been set. Conversely, a variable with a default value already assigned to it can be made required by assigning it the `required` keyword.\nHowever, this `required` keyword is only available in the class body.\n```\nclass MyRequiredTest(HelloTest):\n what = required\n```\nRunning the above test will cause the `set_exec_and_sanity()` hook from `EchoBaseTest` to throw an error indicating that the variable\n`what` has not been set.\n\nFinally, variables may alias each other. If a variable is an alias of another one it behaves in the exact same way as its target. If a change is made to the target variable, this is reflected to the alias and vice versa. However, alias variables are independently loggable: an alias may be logged but not its target and vice versa. Aliased variables are useful when you want to rename a variable and you want to keep the old one for compatibility reasons.\n\nParameters\n* **types** – the supported types for the variable.\n* **value** – the default value assigned to the variable. If no value is provided, the variable is set as `required`.\n* **field** – the field validator to be used for this variable. If no field argument is provided, it defaults to\n`reframe.core.fields.TypedField`. The provided field validator by this argument must derive from `reframe.core.fields.Field`.\n* **alias** – the target variable if this variable is an alias. This must refer to an already declared variable and neither default value nor a field can be specified for an alias variable.\n* **loggable** – Mark this variable as loggable. If [`True`](https://docs.python.org/3/library/constants.html#True), this variable will become a log record attribute under the name\n`check_NAME`, where `NAME` is the name of the variable.\n* **kwargs** – keyword arguments to be forwarded to the constructor of the field validator.\n\nReturns A new test variable.\n\nNew in version 3.10.2: The `loggable` argument is added.\n\nNew in version 4.0.0: Alias variable are introduced.\n\nChanged in version 3.7.0: Expose [`@deferrable`](#reframe.core.builtins.deferrable) as a builtin.\n\nChanged in version 3.11.0: Builtins are now available also through the `reframe.core.builtins` module.\n\n###### Pipeline Hooks[¶](#pipeline-hooks)\n\nReFrame provides a mechanism to allow attaching arbitrary functions to run before or after a given stage of the execution pipeline.\nThis is achieved through the [`@run_before`](#reframe.core.builtins.run_before) and [`@run_after`](#reframe.core.builtins.run_after) test builtins.\nOnce attached to a given stage, these functions are referred to as *pipeline hooks*.\nA hook may be attached to multiple pipeline stages and multiple hooks may also be attached to the same pipeline stage.\nPipeline hooks attached to multiple stages will be executed on each pipeline stage the hook was attached to.\nPipeline stages with multiple hooks attached will execute these hooks in the order in which they were attached to the given pipeline stage.\nA derived class will inherit all the pipeline hooks defined in its bases, except for those whose hook function is overridden by the derived class.\nA function that overrides a pipeline hook from any of the base classes will not be a pipeline hook unless the overriding function is explicitly reattached to any pipeline stage.\nIn the event of a name clash arising from multiple inheritance, the inherited pipeline hook will be chosen following Python’s [MRO](https://docs.python.org/3/library/stdtypes.html#class.__mro__).\n\nA function may be attached to any of the following stages (listed in order of execution): `init`, `setup`, `compile`, `run`, `sanity`, `performance` and `cleanup`.\nThe `init` stage refers to the test’s instantiation and it runs before entering the execution pipeline.\nTherefore, a test function cannot be attached to run before the `init` stage.\nHooks attached to any other stage will run exactly before or after this stage executes.\nSo although a “post-init” and a “pre-setup” hook will both run *after* a test has been initialized and *before* the test goes through the first pipeline stage, they will execute in different times:\nthe post-init hook will execute *right after* the test is initialized.\nThe framework will then continue with other activities and it will execute the pre-setup hook *just before* it schedules the test for executing its setup stage.\n\nPipeline hooks are executed in reverse MRO order, i.e., the hooks of the least specialized class will be executed first.\nIn the following example, `BaseTest.x()` will execute before `DerivedTest.y()`:\n```\nclass BaseTest(rfm.RegressionTest):\n @run_after('setup')\n def x(self):\n '''Hook x'''\n\nclass DerivedTest(BaseTeset):\n @run_after('setup')\n def y(self):\n '''Hook y'''\n```\nSee also\n\n* [`@run_before`](#reframe.core.builtins.run_before), [`@run_after`](#reframe.core.builtins.run_after) decorators\n\nNote\n\nPipeline hooks do not execute in the test’s stage directory, but in the directory that ReFrame executes in.\nHowever, the test’s [`stagedir`](#reframe.core.pipeline.RegressionTest.stagedir) can be accessed by explicitly changing the working directory from within the hook function itself (see the [`change_dir`](index.html#reframe.utility.osext.change_dir) utility for further details):\n```\nimport reframe.utility.osext as osext\n\nclass MyTest(rfm.RegressionTest):\n ...\n @run_after('run')\n def my_post_run_hook(self):\n # Access the stage directory\n with osext.change_dir(self.stagedir):\n ...\n```\nWarning\n\nChanged in version 3.7.0: Declaring pipeline hooks using the same name functions from the `reframe` or `reframe.core.decorators` modules is now deprecated.\nYou should use the builtin functions described in the [Builtins](#builtins) section..\n\nChanged in version 4.0.0: Pipeline hooks can only be defined through the built-in functions described in this section.\n\nWarning\n\nChanged in version 3.9.2: Execution of pipeline hooks until this version was implementation-defined.\nIn practice, hooks of a derived class were executed before those of its parents.\n\nThis version defines the execution order of hooks, which now follows a strict reverse MRO order, so that parent hooks will execute before those of derived classes.\nTests that relied on the execution order of hooks might break with this change.\n\n###### Test variants[¶](#test-variants)\n\nThrough the [`parameter()`](#reframe.core.builtins.parameter) and [`fixture()`](#reframe.core.builtins.fixture) builtins, a regression test may store multiple versions or variants of a regression test at the class level.\nDuring class creation, the test’s parameter and fixture spaces are constructed and combined, assigning a unique index to each of the available test variants.\nIn most cases, the user does not need to be aware of all the internals related to this variant indexing, since ReFrame will run by default all the available variants for each of the registered tests.\nOn the other hand, in more complex use cases such as setting dependencies across different test variants, or when performing some complex variant sub-selection on a fixture declaration, the user may need to access some of this low-level information related to the variant indexing.\nTherefore, classes that derive from the base [`RegressionMixin`](#reframe.core.pipeline.RegressionMixin) provide classmethods and properties to query these data.\n\nWarning\n\nWhen selecting test variants through their variant index, no index ordering should ever be assumed, being the user’s responsibility to ensure on each ReFrame run that the selected index corresponds to the desired parameter and/or fixture variants.\n\nRegressionMixin.num_variants[¶](#reframe.core.pipeline.RegressionMixin.num_variants)\nTotal number of variants of the test.\n\n*classmethod* RegressionMixin.get_variant_nums(***conditions*)[¶](#reframe.core.pipeline.RegressionMixin.get_variant_nums)\nGet the variant numbers that meet the specified conditions.\n\nThe given conditions enable filtering the parameter space of the test.\nFiltering the fixture space is not allowed.\n```\n# Filter out the test variants where my_param is greater than 3 cls.get_variant_nums(my_param=lambda x: x < 4)\n```\nThe returned list of variant numbers can be passed to\n[`variant_name()`](#reframe.core.pipeline.RegressionMixin.variant_name) in order to retrieve the actual test name.\n\nParameters\n**conditions** – keyword arguments where the key is the test parameter name and the value is either a single value or a unary function that evaluates to [`True`](https://docs.python.org/3/library/constants.html#True) if the parameter point must be kept, [`False`](https://docs.python.org/3/library/constants.html#False) otherwise. If a single value is passed this is implicitly converted to the equality function, such that\n```\nget_variant_nums(p=10)\n```\nis equivalent to\n```\nget_variant_nums(p=lambda x: x == 10)\n```\n*classmethod* RegressionMixin.variant_name(*variant_num=None*)[¶](#reframe.core.pipeline.RegressionMixin.variant_name)\nReturn the name of the test variant with a specific variant number.\n\nParameters\n**variant_num** – An integer in the range of `[0, cls.num_variants)`.\n\n###### Dynamic Creation of Tests[¶](#dynamic-creation-of-tests)\n\nNew in version 3.10.0.\n\nreframe.core.meta.make_test(*name*, *bases*, *body*, *methods=None*, ***kwargs*)[[source]](_modules/reframe/core/meta.html#make_test)[¶](#reframe.core.meta.make_test)\nDefine a new test class programmatically.\n\nUsing this method is completely equivalent to using the [`class`](https://docs.python.org/3/reference/compound_stmts.html#class)\nto define the test class. More specifically, the following:\n```\nfrom reframe.core.meta import make_test\n\nhello_cls = make_test(\n 'HelloTest', (rfm.RunOnlyRegressionTest,),\n {\n 'valid_systems': ['*'],\n 'valid_prog_environs': ['*'],\n 'executable': 'echo',\n 'sanity_patterns': sn.assert_true(1)\n }\n)\n```\nis completely equivalent to\n```\nclass HelloTest(rfm.RunOnlyRegressionTest):\n valid_systems = ['*']\n valid_prog_environs = ['*']\n executable = 'echo'\n sanity_patterns: sn.assert_true(1)\n\nhello_cls = HelloTest\n```\nTest [builtins](#builtins) can also be used when defining the body of the test by accessing them through the `reframe.core.builtins`.\nMethods can also be bound to the newly created tests using the `methods`\nargument. The following is an example:\n```\nimport reframe.core.builtins as builtins from reframe.core.meta import make_test\n\ndef set_message(obj):\n obj.executable_opts = [obj.message]\n\ndef validate(obj):\n return sn.assert_found(obj.message, obj.stdout)\n\nhello_cls = make_test(\n 'HelloTest', (rfm.RunOnlyRegressionTest,),\n {\n 'valid_systems': ['*'],\n 'valid_prog_environs': ['*'],\n 'executable': 'echo',\n 'message': builtins.variable(str)\n },\n methods=[\n builtins.run_before('run')(set_message),\n builtins.sanity_function(validate)\n ]\n)\n```\nParameters\n* **name** – The name of the new test class.\n* **bases** – A tuple of the base classes of the class that is being created.\n* **body** – A mapping of key/value pairs that will be inserted as class attributes in the newly created class.\n* **methods** – A list of functions to be bound as methods to the class that is being created. The functions will be bound with their original name.\n* **kwargs** – Any keyword arguments to be passed to the\n`RegressionTestMeta` metaclass.\n\nNew in version 3.10.0.\n\nChanged in version 3.11.0: Added the `methods` arguments.\n\n###### Environments and Systems[¶](#module-reframe.core.environments)\n\n*class* reframe.core.environments.Environment(*name*, *modules=None*, *env_vars=None*, *extras=None*, *features=None*)[[source]](_modules/reframe/core/environments.html#Environment)[¶](#reframe.core.environments.Environment)\nBases: `JSONSerializable`\n\nThis class abstracts away an environment to run regression tests.\n\nIt is simply a collection of modules to be loaded and environment variables to be set when this environment is loaded by the framework.\n\nWarning\n\nUsers may not create [`Environment`](#reframe.core.environments.Environment) objects directly.\n\n*property* env_vars[¶](#reframe.core.environments.Environment.env_vars)\nThe environment variables associated with this environment.\n\nType\n`OrderedDict[str, str]`\n\nNew in version 4.0.0.\n\n*property* extras[¶](#reframe.core.environments.Environment.extras)\nUser defined properties specified in the configuration.\n\nNew in version 3.9.1.\n\nType\n`Dict[str, object]`\n\n*property* features[¶](#reframe.core.environments.Environment.features)\nUsed defined features specified in the configuration.\n\nNew in version 3.11.0.\n\nType\n`List[str]`\n\n*property* modules[¶](#reframe.core.environments.Environment.modules)\nThe modules associated with this environment.\n\nType\n`List[str]`\n\n*property* modules_detailed[¶](#reframe.core.environments.Environment.modules_detailed)\nA view of the modules associated with this environment in a detailed format.\n\nEach module is represented as a dictionary with the following attributes:\n\n* `name`: the name of the module.\n* `collection`: `True` if the module name refers to a module collection.\n\nType\n`List[Dict[str, object]]`\n\nNew in version 3.3.\n\n*property* name[¶](#reframe.core.environments.Environment.name)\nThe name of this environment.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*property* variables[¶](#reframe.core.environments.Environment.variables)\nThe environment variables associated with this environment.\n\nDeprecated since version 4.0.0: Please [`env_vars`](#reframe.core.environments.Environment.env_vars) instead.\n\n*class* reframe.core.environments.ProgEnvironment(*name*, *modules=None*, *env_vars=None*, *extras=None*, *features=None*, *cc='cc'*, *cxx='CC'*, *ftn='ftn'*, *nvcc='nvcc'*, *cppflags=None*, *cflags=None*, *cxxflags=None*, *fflags=None*, *ldflags=None*, ***kwargs*)[[source]](_modules/reframe/core/environments.html#ProgEnvironment)[¶](#reframe.core.environments.ProgEnvironment)\nBases: [`Environment`](#reframe.core.environments.Environment)\n\nA class representing a programming environment.\n\nThis type of environment adds also properties for retrieving the compiler and compilation flags.\n\nWarning\n\nUsers may not create [`ProgEnvironment`](#reframe.core.environments.ProgEnvironment) objects directly.\n\n*property* cc[¶](#reframe.core.environments.ProgEnvironment.cc)\nThe C compiler of this programming environment.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*property* cflags[¶](#reframe.core.environments.ProgEnvironment.cflags)\nThe C compiler flags of this programming environment.\n\nType\n`List[str]`\n\n*property* cppflags[¶](#reframe.core.environments.ProgEnvironment.cppflags)\nThe preprocessor flags of this programming environment.\n\nType\n`List[str]`\n\n*property* cxx[¶](#reframe.core.environments.ProgEnvironment.cxx)\nThe C++ compiler of this programming environment.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*property* cxxflags[¶](#reframe.core.environments.ProgEnvironment.cxxflags)\nThe C++ compiler flags of this programming environment.\n\nType\n`List[str]`\n\n*property* fflags[¶](#reframe.core.environments.ProgEnvironment.fflags)\nThe Fortran compiler flags of this programming environment.\n\nType\n`List[str]`\n\n*property* ftn[¶](#reframe.core.environments.ProgEnvironment.ftn)\nThe Fortran compiler of this programming environment.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*property* ldflags[¶](#reframe.core.environments.ProgEnvironment.ldflags)\nThe linker flags of this programming environment.\n\nType\n`List[str]`\n\n*class* reframe.core.environments._EnvironmentSnapshot(*name='env_snapshot'*)[[source]](_modules/reframe/core/environments.html#_EnvironmentSnapshot)[¶](#reframe.core.environments._EnvironmentSnapshot)\nBases: [`Environment`](#reframe.core.environments.Environment)\n\nAn environment snapshot.\n\nrestore()[[source]](_modules/reframe/core/environments.html#_EnvironmentSnapshot.restore)[¶](#reframe.core.environments._EnvironmentSnapshot.restore)\nRestore this environment snapshot.\n\nreframe.core.environments.snapshot()[[source]](_modules/reframe/core/environments.html#snapshot)[¶](#reframe.core.environments.snapshot)\nCreate an environment snapshot\n\nReturns An instance of [`_EnvironmentSnapshot`](#reframe.core.environments._EnvironmentSnapshot).\n\n*class* reframe.core.systems.DeviceInfo(*info*)[[source]](_modules/reframe/core/systems.html#DeviceInfo)[¶](#reframe.core.systems.DeviceInfo)\nBases: `_ReadOnlyInfo`, `JSONSerializable`\n\nA representation of a device inside ReFrame.\n\nYou can access all the keys of the [device configuration object](config_reference.html#device-info).\n\nNew in version 3.5.0.\n\nWarning\n\nUsers may not create [`DeviceInfo`](#reframe.core.systems.DeviceInfo) objects directly.\n\n*property* device_type[¶](#reframe.core.systems.DeviceInfo.device_type)\nThe type of the device.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or `None`\n\n*property* info[¶](#reframe.core.systems.DeviceInfo.info)\nAll the available information from the configuration.\n\nType\n[`dict`](https://docs.python.org/3/library/stdtypes.html#dict)\n\n*property* num_devices[¶](#reframe.core.systems.DeviceInfo.num_devices)\nNumber of devices of this type.\n\nIt will return 1 if it wasn’t set in the configuration.\n\nType integral\n\n*class* reframe.core.systems.ProcessorInfo(*info*)[[source]](_modules/reframe/core/systems.html#ProcessorInfo)[¶](#reframe.core.systems.ProcessorInfo)\nBases: `_ReadOnlyInfo`, `JSONSerializable`\n\nA representation of a processor inside ReFrame.\n\nYou can access all the keys of the [processor configuration object](config_reference.html#processor-info).\n\nNew in version 3.5.0.\n\nWarning\n\nUsers may not create [`ProcessorInfo`](#reframe.core.systems.ProcessorInfo) objects directly.\n\n*property* info[¶](#reframe.core.systems.ProcessorInfo.info)\nAll the available information from the configuration.\n\nType\n[`dict`](https://docs.python.org/3/library/stdtypes.html#dict)\n\n*property* num_cores[¶](#reframe.core.systems.ProcessorInfo.num_cores)\nTotal number of cores.\n\nType integral or `None`\n\n*property* num_cores_per_numa_node[¶](#reframe.core.systems.ProcessorInfo.num_cores_per_numa_node)\nNumber of cores per NUMA node.\n\nType integral or `None`\n\n*property* num_cores_per_socket[¶](#reframe.core.systems.ProcessorInfo.num_cores_per_socket)\nNumber of cores per socket.\n\nType integral or `None`\n\n*property* num_numa_nodes[¶](#reframe.core.systems.ProcessorInfo.num_numa_nodes)\nNumber of NUMA nodes.\n\nType integral or `None`\n\n*class* reframe.core.systems.System(*name*, *descr*, *hostnames*, *modules_system*, *preload_env*, *prefix*, *outputdir*, *resourcesdir*, *stagedir*, *partitions*)[[source]](_modules/reframe/core/systems.html#System)[¶](#reframe.core.systems.System)\nBases: `JSONSerializable`\n\nA representation of a system inside ReFrame.\n\nWarning\n\nUsers may not create [`System`](#reframe.core.systems.System) objects directly.\n\n*property* descr[¶](#reframe.core.systems.System.descr)\nThe description of this system.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*property* hostnames[¶](#reframe.core.systems.System.hostnames)\nThe hostname patterns associated with this system.\n\nType\n`List[str]`\n\njson()[[source]](_modules/reframe/core/systems.html#System.json)[¶](#reframe.core.systems.System.json)\nReturn a JSON object representing this system.\n\n*property* modules_system[¶](#reframe.core.systems.System.modules_system)\nThe modules system name associated with this system.\n\nType\n[`reframe.core.modules.ModulesSystem`](#reframe.core.modules.ModulesSystem)\n\n*property* name[¶](#reframe.core.systems.System.name)\nThe name of this system.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*property* outputdir[¶](#reframe.core.systems.System.outputdir)\nThe ReFrame output directory prefix associated with this system.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*property* partitions[¶](#reframe.core.systems.System.partitions)\nThe system partitions associated with this system.\n\nType\n`List[SystemPartition]`\n\n*property* prefix[¶](#reframe.core.systems.System.prefix)\nThe ReFrame prefix associated with this system.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*property* preload_environ[¶](#reframe.core.systems.System.preload_environ)\nThe environment to load whenever ReFrame runs on this system.\n\nNew in version 2.19.\n\nType\n[`reframe.core.environments.Environment`](#reframe.core.environments.Environment)\n\n*property* resourcesdir[¶](#reframe.core.systems.System.resourcesdir)\nGlobal resources directory for this system.\n\nThis directory may be used for storing large files related to regression tests. The value of this directory is controlled by the\n[resourcesdir](config_reference.html#.systems[].resourcesdir)\nconfiguration parameter.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*property* stagedir[¶](#reframe.core.systems.System.stagedir)\nThe ReFrame stage directory prefix associated with this system.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*class* reframe.core.systems.SystemPartition(***, *parent*, *name*, *sched_type*, *launcher_type*, *descr*, *access*, *container_runtime*, *container_environs*, *resources*, *local_env*, *environs*, *max_jobs*, *prepare_cmds*, *processor*, *devices*, *extras*, *features*, *time_limit*)[[source]](_modules/reframe/core/systems.html#SystemPartition)[¶](#reframe.core.systems.SystemPartition)\nBases: `JSONSerializable`\n\nA representation of a system partition inside ReFrame.\n\nWarning\n\nUsers may not create [`SystemPartition`](#reframe.core.systems.SystemPartition) objects directly.\n\n*property* access[¶](#reframe.core.systems.SystemPartition.access)\nThe scheduler options for accessing this system partition.\n\nType\n`List[str]`\n\n*property* container_environs[¶](#reframe.core.systems.SystemPartition.container_environs)\nEnvironments associated with the different container platforms.\n\nType\n`Dict[str, Environment]`\n\n*property* container_runtime[¶](#reframe.core.systems.SystemPartition.container_runtime)\nThe default container runtime of this partition.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or `None`\n\n*property* descr[¶](#reframe.core.systems.SystemPartition.descr)\nThe description of this partition.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*property* devices[¶](#reframe.core.systems.SystemPartition.devices)\nA list of devices in the current partition.\n\nNew in version 3.5.0.\n\nType\n`List[reframe.core.systems.DeviceInfo]`\n\nenvironment(*name*)[[source]](_modules/reframe/core/systems.html#SystemPartition.environment)[¶](#reframe.core.systems.SystemPartition.environment)\nReturn the partition environment named `name`.\n\n*property* environs[¶](#reframe.core.systems.SystemPartition.environs)\nThe programming environments associated with this system partition.\n\nType\n`List[ProgEnvironment]`\n\n*property* extras[¶](#reframe.core.systems.SystemPartition.extras)\nUser defined properties associated with this partition.\n\nThese extras are defined in the configuration.\n\nNew in version 3.5.0.\n\nType\n`Dict[str, object]`\n\n*property* features[¶](#reframe.core.systems.SystemPartition.features)\nUser defined features associated with this partition.\n\nThese features are defined in the configuration.\n\nNew in version 3.11.0.\n\nType\n`List[str]`\n\n*property* fullname[¶](#reframe.core.systems.SystemPartition.fullname)\nReturn the fully-qualified name of this partition.\n\nThe fully-qualified name is of the form\n`<parent-system-name>:<partition-name>`.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\njson()[[source]](_modules/reframe/core/systems.html#SystemPartition.json)[¶](#reframe.core.systems.SystemPartition.json)\nReturn a JSON object representing this system partition.\n\n*property* launcher_type[¶](#reframe.core.systems.SystemPartition.launcher_type)\nThe type of the backend launcher of this partition.\n\nNew in version 3.2.\n\nType a subclass of [`reframe.core.launchers.JobLauncher`](#reframe.core.launchers.JobLauncher).\n\n*property* local_env[¶](#reframe.core.systems.SystemPartition.local_env)\nThe local environment associated with this partition.\n\nType\n`Environment`\n\n*property* max_jobs[¶](#reframe.core.systems.SystemPartition.max_jobs)\nThe maximum number of concurrent jobs allowed on this partition.\n\nType integral\n\n*property* name[¶](#reframe.core.systems.SystemPartition.name)\nThe name of this partition.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*property* prepare_cmds[¶](#reframe.core.systems.SystemPartition.prepare_cmds)\nCommands to be emitted before loading the modules.\n\nType\n`List[str]`\n\n*property* processor[¶](#reframe.core.systems.SystemPartition.processor)\nProcessor information for the current partition.\n\nNew in version 3.5.0.\n\nType\n[`reframe.core.systems.ProcessorInfo`](#reframe.core.systems.ProcessorInfo)\n\n*property* resources[¶](#reframe.core.systems.SystemPartition.resources)\nThe resources template strings associated with this partition.\n\nThis is a dictionary, where the key is the name of a resource and the value is the scheduler options or directives associated with this resource.\n\nType\n`Dict[str, List[str]]`\n\n*property* scheduler[¶](#reframe.core.systems.SystemPartition.scheduler)\nThe backend scheduler of this partition.\n\nType\n`reframe.core.schedulers.JobScheduler`.\n\nNote\n\nChanged in version 2.8: Prior versions returned a string representing the scheduler and job launcher combination.\n\nChanged in version 3.2: The property now stores a `JobScheduler` instance.\n\nselect_devices(*devtype*)[[source]](_modules/reframe/core/systems.html#SystemPartition.select_devices)[¶](#reframe.core.systems.SystemPartition.select_devices)\nReturn all devices of the requested type:\n\nParameters\n**devtype** – The type of the device info objects to return.\n\nReturns A list of [`DeviceInfo`](#reframe.core.systems.DeviceInfo) objects of the specified type.\n\n*property* time_limit[¶](#reframe.core.systems.SystemPartition.time_limit)\nThe time limit that will be used when submitting jobs to this partition.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or [`None`](https://docs.python.org/3/library/constants.html#None)\n\nNew in version 3.11.0.\n\n###### Jobs and Parallel Launchers[¶](#module-reframe.core.schedulers)\n\n*class* reframe.core.schedulers.Job(**args*, ***kwargs*)[[source]](_modules/reframe/core/schedulers.html#Job)[¶](#reframe.core.schedulers.Job)\nBases: `JSONSerializable`\n\nA job descriptor.\n\nA job descriptor is created by the framework after the “setup” phase and is associated with the test.\n\nWarning\n\nUsers may not create a job descriptor directly.\n\n*property* cli_options[¶](#reframe.core.schedulers.Job.cli_options)\nThe scheduler options passed through the [`-J`](index.html#cmdoption-J) command line options.\n\n*property* completion_time[¶](#reframe.core.schedulers.Job.completion_time)\nThe completion time of this job as a floating point number expressed in seconds since the epoch, in UTC.\n\nThis attribute is `None` if the job hasn’t been finished yet,\nor if ReFrame runtime hasn’t perceived it yet.\n\nThe accuracy of this timestamp depends on the backend scheduler.\nThe `slurm` scheduler backend relies on job accounting and returns the actual termination time of the job. The rest of the backends report as completion time the moment when the framework realizes that the spawned job has finished. In this case, the accuracy depends on the execution policy used. If tests are executed with the serial execution policy, this is close to the real completion time, but if the asynchronous execution policy is used, it can differ significantly.\n\nType\n[`float`](https://docs.python.org/3/library/functions.html#float) or `None`\n\n*property* exception[¶](#reframe.core.schedulers.Job.exception)\nThe last exception that this job encountered.\n\nThe scheduler will raise this exception the next time the status of this job is queried.\n\nexclusive_access *= False*[¶](#reframe.core.schedulers.Job.exclusive_access)\nRequest exclusive access on the nodes for this job.\n\nType\n[`bool`](https://docs.python.org/3/library/functions.html#bool)\n\nDefault\n`false`\n\nNote\n\nThis attribute is set by the framework just before submitting the job based on the test information.\n\nNew in version 3.11.0.\n\n*property* exitcode[¶](#reframe.core.schedulers.Job.exitcode)\nThe exit code of this job.\n\nThis may or may not be set depending on the scheduler backend.\n\nNew in version 2.21.\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int) or `None`\n\n*property* jobid[¶](#reframe.core.schedulers.Job.jobid)\nThe ID of this job.\n\nNew in version 2.21.\n\nChanged in version 3.2: Job ID type is now a string.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or `None`\n\nlauncher[¶](#reframe.core.schedulers.Job.launcher)\nThe (parallel) program launcher that will be used to launch the\n(parallel) executable of this job.\n\nUsers are allowed to explicitly set the current job launcher, but this is only relevant in rare situations, such as when you want to wrap the current launcher command. For this specific scenario, you may have a look at the [`reframe.core.launchers.LauncherWrapper`](#reframe.core.launchers.LauncherWrapper) class.\n\nThe following example shows how you can replace the current partition’s launcher for this test with the “local” launcher:\n```\nfrom reframe.core.backends import getlauncher\n\n@run_after('setup')\ndef set_launcher(self):\n self.job.launcher = getlauncher('local')()\n```\nType\n[`reframe.core.launchers.JobLauncher`](#reframe.core.launchers.JobLauncher)\n\nmax_pending_time *= None*[¶](#reframe.core.schedulers.Job.max_pending_time)\nMaximum pending time for this job.\n\nSee [`reframe.core.pipeline.RegressionTest.max_pending_time`](#reframe.core.pipeline.RegressionTest.max_pending_time) for more details.\n\nNote\n\nThis attribute is set by the framework just before submitting the job based on the test information.\n\nNew in version 3.11.0.\n\n*property* name[¶](#reframe.core.schedulers.Job.name)\nThe name of this job.\n\n*property* nodelist[¶](#reframe.core.schedulers.Job.nodelist)\nThe list of node names assigned to this job.\n\nThis attribute is `None` if no nodes are assigned to the job yet.\nThis attribute is set reliably only for the `slurm` backend, i.e.,\nSlurm *with* accounting enabled.\nThe `squeue` scheduler backend, i.e., Slurm *without* accounting,\nmight not set this attribute for jobs that finish very quickly.\nFor the `local` scheduler backend, this returns an one-element list containing the hostname of the current host.\n\nThis attribute might be useful in a flexible regression test for determining the actual nodes that were assigned to the test.\nFor more information on flexible node allocation, see the\n[`--flex-alloc-nodes`](index.html#cmdoption-flex-alloc-nodes) command-line option.\n\nThis attribute is *not* supported by the `pbs` scheduler backend.\n\nNew in version 2.17.\n\nType\n`List[str]` or `None`\n\nnum_cpus_per_task *= None*[¶](#reframe.core.schedulers.Job.num_cpus_per_task)\nNumber of processing elements associated with each task for this job.\n\nType integral or `NoneType`\n\nDefault\n`None`\n\nNote\n\nThis attribute is set by the framework just before submitting the job based on the test information.\n\nNew in version 3.11.0.\n\nnum_tasks *= 1*[¶](#reframe.core.schedulers.Job.num_tasks)\nNumber of tasks for this job.\n\nType integral\n\nDefault\n`1`\n\nNote\n\nThis attribute is set by the framework just before submitting the job based on the test information.\n\nNew in version 3.11.0.\n\nChanged in version 4.1: Allow [`None`](https://docs.python.org/3/library/constants.html#None) values\n\nnum_tasks_per_core *= None*[¶](#reframe.core.schedulers.Job.num_tasks_per_core)\nNumber of tasks per core for this job.\n\nType integral or `NoneType`\n\nDefault\n`None`\n\nNote\n\nThis attribute is set by the framework just before submitting the job based on the test information.\n\nNew in version 3.11.0.\n\nnum_tasks_per_node *= None*[¶](#reframe.core.schedulers.Job.num_tasks_per_node)\nNumber of tasks per node for this job.\n\nType integral or `NoneType`\n\nDefault\n`None`\n\nNote\n\nThis attribute is set by the framework just before submitting the job based on the test information.\n\nNew in version 3.11.0.\n\nnum_tasks_per_socket *= None*[¶](#reframe.core.schedulers.Job.num_tasks_per_socket)\nNumber of tasks per socket for this job.\n\nType integral or `NoneType`\n\nDefault\n`None`\n\nNote\n\nThis attribute is set by the framework just before submitting the job based on the test information.\n\nNew in version 3.11.0.\n\noptions *= []*[¶](#reframe.core.schedulers.Job.options)\nArbitrary options to be passed to the backend job scheduler.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\npin_nodes *= []*[¶](#reframe.core.schedulers.Job.pin_nodes)\nPin the jobs on the given nodes.\n\nThe list of nodes will be transformed to a suitable string and be passed to the scheduler’s options. Currently it will have an effect only for the Slurm scheduler.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nNew in version 3.11.0.\n\n*property* sched_access[¶](#reframe.core.schedulers.Job.sched_access)\nThe partition’s [`access`](index.html#config.systems.partitions.access)\noptions.\n\n*property* sched_flex_alloc_nodes[¶](#reframe.core.schedulers.Job.sched_flex_alloc_nodes)\nThe argument of the [`--flex-alloc-nodes`](index.html#cmdoption-flex-alloc-nodes) command line option.\n\n*property* scheduler[¶](#reframe.core.schedulers.Job.scheduler)\nThe scheduler where this job is assigned to.\n\n*property* script_filename[¶](#reframe.core.schedulers.Job.script_filename)\nThe filename of the generated job script.\n\n*property* state[¶](#reframe.core.schedulers.Job.state)\nThe state of this job.\n\nThe value of this field is scheduler-specific.\n\nNew in version 2.21.\n\nType\n:class`str` or `None`\n\n*property* stderr[¶](#reframe.core.schedulers.Job.stderr)\nThe file where the standard error of the job is saved.\n\n*property* stdout[¶](#reframe.core.schedulers.Job.stdout)\nThe file where the standard output of the job is saved.\n\n*property* submit_time[¶](#reframe.core.schedulers.Job.submit_time)\nThe submission time of this job as a floating point number expressed in seconds since the epoch, in UTC.\n\nThis attribute is `None` if the job hasn’t been submitted yet.\n\nThis attribute is set right after the job is submitted and can vary significantly from the time the jobs starts running, depending on the scheduler.\n\nType\n[`float`](https://docs.python.org/3/library/functions.html#float) or `None`\n\ntime_limit *= None*[¶](#reframe.core.schedulers.Job.time_limit)\nTime limit for this job.\n\nSee [`reframe.core.pipeline.RegressionTest.time_limit`](#reframe.core.pipeline.RegressionTest.time_limit) for more details.\n\nNote\n\nThis attribute is set by the framework just before submitting the job based on the test information.\n\nNew in version 3.11.0.\n\nuse_smt *= None*[¶](#reframe.core.schedulers.Job.use_smt)\nEnable SMT for this job.\n\nType\n[`bool`](https://docs.python.org/3/library/functions.html#bool) or `NoneType`\n\nDefault\n`None`\n\nNote\n\nThis attribute is set by the framework just before submitting the job based on the test information.\n\nNew in version 3.11.0.\n\n*property* workdir[¶](#reframe.core.schedulers.Job.workdir)\nThe working directory for this job.\n\n*class* reframe.core.launchers.JobLauncher[[source]](_modules/reframe/core/launchers.html#JobLauncher)[¶](#reframe.core.launchers.JobLauncher)\nBases: [`ABC`](https://docs.python.org/3/library/abc.html#abc.ABC)\n\nAbstract base class for job launchers.\n\nA job launcher is the executable that actually launches a distributed program to multiple nodes, e.g., `mpirun`, `srun` etc.\n\nNote\n\nChanged in version 4.0.0: Users may create job launchers directly.\n\nChanged in version 2.8: Job launchers do not get a reference to a job during their initialization.\n\n*abstract* command(*job*)[[source]](_modules/reframe/core/launchers.html#JobLauncher.command)[¶](#reframe.core.launchers.JobLauncher.command)\nThe launcher command to be emitted for a specific job.\n\nLauncher backends provide concrete implementations of this method.\n\nParameters\n**job** – A job descriptor.\n\nReturns the basic launcher command as a list of tokens.\n\noptions[¶](#reframe.core.launchers.JobLauncher.options)\nList of options to be passed to the job launcher invocation.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nrun_command(*job*)[[source]](_modules/reframe/core/launchers.html#JobLauncher.run_command)[¶](#reframe.core.launchers.JobLauncher.run_command)\nThe full launcher command to be emitted for a specific job.\n\nThis includes any user options.\n\nParameters\n**job** – a job descriptor.\n\nReturns the launcher command as a string.\n\n*class* reframe.core.launchers.LauncherWrapper(*target_launcher*, *wrapper_command*, *wrapper_options=[]*)[[source]](_modules/reframe/core/launchers.html#LauncherWrapper)[¶](#reframe.core.launchers.LauncherWrapper)\nBases: [`JobLauncher`](#reframe.core.launchers.JobLauncher)\n\nWrap a launcher object so as to modify its invocation.\n\nThis is useful for parallel debuggers. For example, to launch a regression test using the [ARM DDT](https://www.arm.com/products/development-tools/server-and-hpc/forge)\ndebugger, you can do the following:\n```\n@run_after('setup')\ndef set_launcher(self):\n self.job.launcher = LauncherWrapper(self.job.launcher, 'ddt',\n ['--offline'])\n```\nIf the current system partition uses native Slurm for job submission, this setup will generate the following command in the submission script:\n```\nddt --offline srun <test_executable>\n```\nIf the current partition uses `mpirun` instead, it will generate\n```\nddt --offline mpirun -np <num_tasks> ... <test_executable>\n```\nParameters\n* **target_launcher** – The launcher to wrap.\n* **wrapper_command** – The wrapper command.\n* **wrapper_options** – List of options to pass to the wrapper command.\n\ncommand(*job*)[[source]](_modules/reframe/core/launchers.html#LauncherWrapper.command)[¶](#reframe.core.launchers.LauncherWrapper.command)\nThe launcher command to be emitted for a specific job.\n\nLauncher backends provide concrete implementations of this method.\n\nParameters\n**job** – A job descriptor.\n\nReturns the basic launcher command as a list of tokens.\n\nreframe.core.backends.getlauncher(*name*)[¶](#reframe.core.backends.getlauncher)\nRetrieve the [`reframe.core.launchers.JobLauncher`](#reframe.core.launchers.JobLauncher) concrete implementation for a parallel launcher backend.\n\nParameters\n**name** – The registered name of the launcher backend.\n\nreframe.core.backends.getscheduler(*name*)[¶](#reframe.core.backends.getscheduler)\nRetrieve the `reframe.core.schedulers.JobScheduler` concrete implementation for a scheduler backend.\n\nParameters\n**name** – The registered name of the scheduler backend.\n\n###### Runtime Services[¶](#module-reframe.core.runtime)\n\n*class* reframe.core.runtime.RuntimeContext(*site_config*)[[source]](_modules/reframe/core/runtime.html#RuntimeContext)[¶](#reframe.core.runtime.RuntimeContext)\nBases: [`object`](https://docs.python.org/3/library/functions.html#object)\n\nThe runtime context of the framework.\n\nThere is a single instance of this class globally in the framework.\n\nNew in version 2.13.\n\nget_option(*option*, *default=None*)[[source]](_modules/reframe/core/runtime.html#RuntimeContext.get_option)[¶](#reframe.core.runtime.RuntimeContext.get_option)\nGet a configuration option.\n\nParameters\n* **option** – The option to be retrieved.\n* **default** – The value to return if `option` cannot be retrieved.\n\nReturns The value of the option.\n\nChanged in version 3.11.0: Add `default` named argument.\n\n*property* modules_system[¶](#reframe.core.runtime.RuntimeContext.modules_system)\nThe environment modules system used in the current host.\n\nType\n[`reframe.core.modules.ModulesSystem`](#reframe.core.modules.ModulesSystem).\n\n*property* output_prefix[¶](#reframe.core.runtime.RuntimeContext.output_prefix)\nThe output directory prefix.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*property* stage_prefix[¶](#reframe.core.runtime.RuntimeContext.stage_prefix)\nThe stage directory prefix.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*property* system[¶](#reframe.core.runtime.RuntimeContext.system)\nThe current host system.\n\nType\n[`reframe.core.systems.System`](#reframe.core.systems.System)\n\nreframe.core.runtime.is_env_loaded(*environ*)[[source]](_modules/reframe/core/runtime.html#is_env_loaded)[¶](#reframe.core.runtime.is_env_loaded)\nCheck if environment is loaded.\n\nParameters\n**environ** ([*Environment*](index.html#reframe.core.environments.Environment)) – Environment to check for.\n\nReturns\n`True` if this environment is loaded, `False`\notherwise.\n\nreframe.core.runtime.loadenv(**environs*)[[source]](_modules/reframe/core/runtime.html#loadenv)[¶](#reframe.core.runtime.loadenv)\nLoad environments in the current Python context.\n\nParameters\n**environs** (*List**[*[*Environment*](index.html#reframe.core.environments.Environment)*]*) – A list of environments to load.\n\nReturns A tuple containing snapshot of the current environment upon entry to this function and a list of shell commands required to load the environments.\n\nReturn type Tuple[[_EnvironmentSnapshot](index.html#reframe.core.environments._EnvironmentSnapshot), List[[str](https://docs.python.org/3/library/stdtypes.html#str)]]\n\n*class* reframe.core.runtime.module_use(**paths*)[[source]](_modules/reframe/core/runtime.html#module_use)[¶](#reframe.core.runtime.module_use)\nBases: [`object`](https://docs.python.org/3/library/functions.html#object)\n\nContext manager for temporarily modifying the module path.\n\nreframe.core.runtime.runtime()[[source]](_modules/reframe/core/runtime.html#runtime)[¶](#reframe.core.runtime.runtime)\nGet the runtime context of the framework.\n\nNew in version 2.13.\n\nReturns A [`reframe.core.runtime.RuntimeContext`](#reframe.core.runtime.RuntimeContext) object.\n\n*class* reframe.core.runtime.temp_config(*system*)[[source]](_modules/reframe/core/runtime.html#temp_config)[¶](#reframe.core.runtime.temp_config)\nBases: [`object`](https://docs.python.org/3/library/functions.html#object)\n\nContext manager to temporarily switch to specific configuration.\n\n*class* reframe.core.runtime.temp_environment(*modules=None*, *env_vars=None*)[[source]](_modules/reframe/core/runtime.html#temp_environment)[¶](#reframe.core.runtime.temp_environment)\nBases: [`object`](https://docs.python.org/3/library/functions.html#object)\n\nContext manager to temporarily change the environment.\n\n###### Modules Systems[¶](#modules-systems)\n\n*class* reframe.core.modules.ModulesSystem(*backend*)[[source]](_modules/reframe/core/modules.html#ModulesSystem)[¶](#reframe.core.modules.ModulesSystem)\nBases: [`object`](https://docs.python.org/3/library/functions.html#object)\n\nA modules system.\n\navailable_modules(*substr=None*)[[source]](_modules/reframe/core/modules.html#ModulesSystem.available_modules)[¶](#reframe.core.modules.ModulesSystem.available_modules)\nReturn a list of available modules that contain `substr` in their name.\n\nReturn type List[[str](https://docs.python.org/3/library/stdtypes.html#str)]\n\nconflicted_modules(*name*, *collection=False*, *path=None*)[[source]](_modules/reframe/core/modules.html#ModulesSystem.conflicted_modules)[¶](#reframe.core.modules.ModulesSystem.conflicted_modules)\nReturn the list of the modules conflicting with module `name`.\n\nIf module `name` resolves to multiple real modules, then the returned list will be the concatenation of the conflict lists of all the real modules.\n\nParameters\n* **name** – The name of the module.\n* **collection** – The module is a “module collection” (TMod4/LMod only).\n* **path** – The path where the module resides if not in the default\n`MODULEPATH`.\n\nReturns A list of conflicting module names.\n\nChanged in version 3.3: The `collection` argument is added.\n\nChanged in version 3.5.0: The `path` argument is added.\n\nemit_load_commands(*name*, *collection=False*, *path=None*)[[source]](_modules/reframe/core/modules.html#ModulesSystem.emit_load_commands)[¶](#reframe.core.modules.ModulesSystem.emit_load_commands)\nReturn the appropriate shell commands for loading a module.\n\nModule mappings are not taken into account by this function.\n\nParameters\n* **name** – The name of the module to load.\n* **collection** – The module is a “module collection” (TMod4/LMod only)\n* **path** – The path where the module resides if not in the default\n`MODULEPATH`.\n\nReturns A list of shell commands.\n\nChanged in version 3.3: The `collection` argument was added and module mappings are no more taken into account by this function.\n\nChanged in version 3.5.0: The `path` argument is added.\n\nemit_unload_commands(*name*, *collection=False*, *path=None*)[[source]](_modules/reframe/core/modules.html#ModulesSystem.emit_unload_commands)[¶](#reframe.core.modules.ModulesSystem.emit_unload_commands)\nReturn the appropriate shell commands for unloading a module.\n\nModule mappings are not taken into account by this function.\n\nParameters\n* **name** – The name of the module to unload.\n* **collection** – The module is a “module collection” (TMod4/LMod only)\n* **path** – The path where the module resides if not in the default\n`MODULEPATH`.\n\nReturns A list of shell commands.\n\nChanged in version 3.3: The `collection` argument was added and module mappings are no more taken into account by this function.\n\nChanged in version 3.5.0: The `path` argument is added.\n\nexecute(*cmd*, **args*)[[source]](_modules/reframe/core/modules.html#ModulesSystem.execute)[¶](#reframe.core.modules.ModulesSystem.execute)\nExecute an arbitrary module command.\n\nParameters\n* **cmd** – The command to execute, e.g., `load`, `restore` etc.\n* **args** – The arguments to pass to the command.\n\nReturns The command output.\n\nis_module_loaded(*name*)[[source]](_modules/reframe/core/modules.html#ModulesSystem.is_module_loaded)[¶](#reframe.core.modules.ModulesSystem.is_module_loaded)\nCheck if module `name` is loaded.\n\nIf module `name` refers to multiple real modules, this method will return `True` only if all the referees are loaded.\n\nload_module(*name*, *collection=False*, *path=None*, *force=False*)[[source]](_modules/reframe/core/modules.html#ModulesSystem.load_module)[¶](#reframe.core.modules.ModulesSystem.load_module)\nLoad the module `name`.\n\nParameters\n* **collection** – The module is a “module collection” (TMod4/Lmod only)\n* **path** – The path where the module resides if not in the default\n`MODULEPATH`.\n* **force** – If set, forces the loading, unloading first any conflicting modules currently loaded. If module `name` refers to multiple real modules, all of the target modules will be loaded.\n\nReturns A list of two-element tuples, where each tuple contains the module that was loaded and the list of modules that had to be unloaded first due to conflicts. This list will be normally of size one, but it can be longer if there is mapping that maps module `name` to multiple other modules.\n\nChanged in version 3.3: - The `collection` argument is added.\n- This function now returns a list of tuples.\n\nChanged in version 3.5.0: - The `path` argument is added.\n- The `force` argument is now the last argument.\n\nloaded_modules()[[source]](_modules/reframe/core/modules.html#ModulesSystem.loaded_modules)[¶](#reframe.core.modules.ModulesSystem.loaded_modules)\nReturn a list of loaded modules.\n\nReturn type List[[str](https://docs.python.org/3/library/stdtypes.html#str)]\n\n*property* name[¶](#reframe.core.modules.ModulesSystem.name)\nThe name of this module system.\n\n*property* searchpath[¶](#reframe.core.modules.ModulesSystem.searchpath)\nThe module system search path as a list of directories.\n\nsearchpath_add(**dirs*)[[source]](_modules/reframe/core/modules.html#ModulesSystem.searchpath_add)[¶](#reframe.core.modules.ModulesSystem.searchpath_add)\nAdd `dirs` to the module system search path.\n\nsearchpath_remove(**dirs*)[[source]](_modules/reframe/core/modules.html#ModulesSystem.searchpath_remove)[¶](#reframe.core.modules.ModulesSystem.searchpath_remove)\nRemove `dirs` from the module system search path.\n\nunload_all()[[source]](_modules/reframe/core/modules.html#ModulesSystem.unload_all)[¶](#reframe.core.modules.ModulesSystem.unload_all)\nUnload all loaded modules.\n\nunload_module(*name*, *collection=False*, *path=None*)[[source]](_modules/reframe/core/modules.html#ModulesSystem.unload_module)[¶](#reframe.core.modules.ModulesSystem.unload_module)\nUnload module `name`.\n\nParameters\n* **name** – The name of the module to unload. If module `name` is resolved to multiple real modules, all the referred to modules will be unloaded in reverse order.\n* **collection** – The module is a “module collection” (TMod4 only)\n* **path** – The path where the module resides if not in the default\n`MODULEPATH`.\n\nChanged in version 3.3: The `collection` argument was added.\n\nChanged in version 3.5.0: The `path` argument is added.\n\n*property* version[¶](#reframe.core.modules.ModulesSystem.version)\nThe version of this module system.\n\n###### Build Systems[¶](#build-systems)\n\nNew in version 2.14.\n\nReFrame delegates the compilation of the regression test to a build system.\nBuild systems in ReFrame are entities that are responsible for generating the necessary shell commands for compiling a code.\nEach build system defines a set of attributes that users may set in order to customize their compilation.\nAn example usage is the following:\n```\nself.build_system = 'SingleSource'\nself.build_system.cflags = ['-fopenmp']\n```\nUsers simply set the build system to use in their regression tests and then they configure it.\nIf no special configuration is needed for the compilation, users may completely ignore the build systems.\nReFrame will automatically pick one based on the regression test attributes and will try to compile the code.\n\nAll build systems in ReFrame derive from the abstract base class [`reframe.core.buildsystems.BuildSystem`](#reframe.core.buildsystems.BuildSystem).\nThis class defines a set of common attributes, such us compilers, compilation flags etc. that all subclasses inherit.\nIt is up to the concrete build system implementations on how to use or not these attributes.\n\n*class* reframe.core.buildsystems.Autotools(**args*, ***kwargs*)[[source]](_modules/reframe/core/buildsystems.html#Autotools)[¶](#reframe.core.buildsystems.Autotools)\nBases: [`ConfigureBasedBuildSystem`](#reframe.core.buildsystems.ConfigureBasedBuildSystem)\n\nA build system for compiling Autotools-based projects.\n\nThis build system will emit the following commands:\n\n1. Create a build directory if `builddir` is not `None` and change to it.\n2. Invoke `configure` to configure the project by setting the corresponding flags for compilers and compiler flags.\n3. Issue `make` to compile the code.\n\nconfiguredir *= .*[¶](#reframe.core.buildsystems.Autotools.configuredir)\nThe directory of the configure script.\n\nThis can be changed to do an out of source build without copying the entire source tree.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`'.'`\n\n*class* reframe.core.buildsystems.BuildSystem(**args*, ***kwargs*)[[source]](_modules/reframe/core/buildsystems.html#BuildSystem)[¶](#reframe.core.buildsystems.BuildSystem)\nBases: [`object`](https://docs.python.org/3/library/functions.html#object)\n\nThe abstract base class of any build system.\n\nConcrete build systems inherit from this class and must override the\n`emit_build_commands()` abstract function.\n\ncc[¶](#reframe.core.buildsystems.BuildSystem.cc)\nThe C compiler to be used.\nIf empty and [`flags_from_environ`](#reframe.core.buildsystems.BuildSystem.flags_from_environ) is `True`,\nthe compiler defined in the current programming environment will be used.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`''`\n\ncflags *= []*[¶](#reframe.core.buildsystems.BuildSystem.cflags)\nThe C compiler flags to be used.\nIf empty and [`flags_from_environ`](#reframe.core.buildsystems.BuildSystem.flags_from_environ) is `True`,\nthe corresponding flags defined in the current programming environment will be used.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\ncppflags *= []*[¶](#reframe.core.buildsystems.BuildSystem.cppflags)\nThe preprocessor flags to be used.\nIf empty and [`flags_from_environ`](#reframe.core.buildsystems.BuildSystem.flags_from_environ) is `True`,\nthe corresponding flags defined in the current programming environment will be used.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\ncxx[¶](#reframe.core.buildsystems.BuildSystem.cxx)\nThe C++ compiler to be used.\nIf empty and [`flags_from_environ`](#reframe.core.buildsystems.BuildSystem.flags_from_environ) is `True`,\nthe compiler defined in the current programming environment will be used.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`''`\n\ncxxflags *= []*[¶](#reframe.core.buildsystems.BuildSystem.cxxflags)\nThe C++ compiler flags to be used.\nIf empty and [`flags_from_environ`](#reframe.core.buildsystems.BuildSystem.flags_from_environ) is `True`,\nthe corresponding flags defined in the current programming environment will be used.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nfflags *= []*[¶](#reframe.core.buildsystems.BuildSystem.fflags)\nThe Fortran compiler flags to be used.\nIf empty and [`flags_from_environ`](#reframe.core.buildsystems.BuildSystem.flags_from_environ) is `True`,\nthe corresponding flags defined in the current programming environment will be used.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nflags_from_environ *= True*[¶](#reframe.core.buildsystems.BuildSystem.flags_from_environ)\nSet compiler and compiler flags from the current programming environment if not specified otherwise.\n\nType\n[`bool`](https://docs.python.org/3/library/functions.html#bool)\n\nDefault\n`True`\n\nftn[¶](#reframe.core.buildsystems.BuildSystem.ftn)\nThe Fortran compiler to be used.\nIf empty and [`flags_from_environ`](#reframe.core.buildsystems.BuildSystem.flags_from_environ) is `True`,\nthe compiler defined in the current programming environment will be used.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`''`\n\nldflags *= []*[¶](#reframe.core.buildsystems.BuildSystem.ldflags)\nThe linker flags to be used.\nIf empty and [`flags_from_environ`](#reframe.core.buildsystems.BuildSystem.flags_from_environ) is `True`,\nthe corresponding flags defined in the current programming environment will be used.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nnvcc[¶](#reframe.core.buildsystems.BuildSystem.nvcc)\nThe CUDA compiler to be used.\nIf empty and [`flags_from_environ`](#reframe.core.buildsystems.BuildSystem.flags_from_environ) is `True`,\nthe compiler defined in the current programming environment will be used.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`''`\n\n*class* reframe.core.buildsystems.BuildSystemMeta(*name*, *bases*, *namespace*, ***kwargs*)[[source]](_modules/reframe/core/buildsystems.html#BuildSystemMeta)[¶](#reframe.core.buildsystems.BuildSystemMeta)\nBases: `RegressionTestMeta`, [`ABCMeta`](https://docs.python.org/3/library/abc.html#abc.ABCMeta)\n\nBuild systems metaclass.\n\n*class* reframe.core.buildsystems.CMake(**args*, ***kwargs*)[[source]](_modules/reframe/core/buildsystems.html#CMake)[¶](#reframe.core.buildsystems.CMake)\nBases: [`ConfigureBasedBuildSystem`](#reframe.core.buildsystems.ConfigureBasedBuildSystem)\n\nA build system for compiling CMake-based projects.\n\nThis build system will emit the following commands:\n\n1. Create a build directory if `builddir` is not `None` and change to it.\n2. Invoke `cmake` to configure the project by setting the corresponding CMake flags for compilers and compiler flags.\n3. Issue `make` to compile the code.\n\n*class* reframe.core.buildsystems.ConfigureBasedBuildSystem(**args*, ***kwargs*)[[source]](_modules/reframe/core/buildsystems.html#ConfigureBasedBuildSystem)[¶](#reframe.core.buildsystems.ConfigureBasedBuildSystem)\nBases: [`BuildSystem`](#reframe.core.buildsystems.BuildSystem)\n\nAbstract base class for configured-based build systems.\n\nbuilddir *= None*[¶](#reframe.core.buildsystems.ConfigureBasedBuildSystem.builddir)\nThe CMake build directory, where all the generated files will be placed.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`None`\n\nconfig_opts *= []*[¶](#reframe.core.buildsystems.ConfigureBasedBuildSystem.config_opts)\nAdditional configuration options to be passed to the CMake invocation.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nmake_opts *= []*[¶](#reframe.core.buildsystems.ConfigureBasedBuildSystem.make_opts)\nOptions to be passed to the subsequent `make` invocation.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nmax_concurrency *= 1*[¶](#reframe.core.buildsystems.ConfigureBasedBuildSystem.max_concurrency)\nSame as for the [`Make`](#reframe.core.buildsystems.Make) build system.\n\nType integer\n\nDefault\n`1`\n\nsrcdir *= None*[¶](#reframe.core.buildsystems.ConfigureBasedBuildSystem.srcdir)\nThe top-level directory of the code.\n\nThis is set automatically by the framework based on the\n[`reframe.core.pipeline.RegressionTest.sourcepath`](#reframe.core.pipeline.RegressionTest.sourcepath) attribute.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`None`\n\n*class* reframe.core.buildsystems.CustomBuild(**args*, ***kwargs*)[[source]](_modules/reframe/core/buildsystems.html#CustomBuild)[¶](#reframe.core.buildsystems.CustomBuild)\nBases: [`BuildSystem`](#reframe.core.buildsystems.BuildSystem)\n\nCustom build system.\n\nThis build system backend allows users to use custom build scripts to build the test code. It does not do any interpretation of the current test environment and it simply runs the supplied [`commands`](#reframe.core.buildsystems.CustomBuild.commands).\n\nUsers should use this build system with caution, because environment management, reproducibility and any potential side effects are all controlled by the user’s custom build system.\n\nNew in version 3.11.0.\n\ncommands[¶](#reframe.core.buildsystems.CustomBuild.commands)\nThe commands to run for building the test code.\n\nType\n`List[str]`\n\n*class* reframe.core.buildsystems.EasyBuild(**args*, ***kwargs*)[[source]](_modules/reframe/core/buildsystems.html#EasyBuild)[¶](#reframe.core.buildsystems.EasyBuild)\nBases: [`BuildSystem`](#reframe.core.buildsystems.BuildSystem)\n\nA build system for building test code using [EasyBuild](https://easybuild.io/).\n\nReFrame will use EasyBuild to build and install the code in the test’s stage directory by default. ReFrame uses environment variables to configure EasyBuild for running, so users can pass additional options to the `eb` command and modify the default behaviour.\n\nNew in version 3.5.0.\n\neasyconfigs *= []*[¶](#reframe.core.buildsystems.EasyBuild.easyconfigs)\nThe list of easyconfig files to build and install.\nThis field is required.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nemit_package *= False*[¶](#reframe.core.buildsystems.EasyBuild.emit_package)\nInstruct EasyBuild to emit a package for the built software.\nThis will essentially pass the `--package` option to `eb`.\n\nType\n[`bool`](https://docs.python.org/3/library/functions.html#bool)\n\nDefault\n`False`\n\noptions *= []*[¶](#reframe.core.buildsystems.EasyBuild.options)\nOptions to pass to the `eb` command.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\npackage_opts *= {}*[¶](#reframe.core.buildsystems.EasyBuild.package_opts)\nOptions controlling the package creation from EasyBuild.\nFor each key/value pair of this dictionary, ReFrame will pass\n`--package-{key}={val}` to the EasyBuild invocation.\n\nType\n`Dict[str, str]`\n\nDefault\n`{}`\n\nprefix *= easybuild*[¶](#reframe.core.buildsystems.EasyBuild.prefix)\nDefault prefix for the EasyBuild installation.\n\nRelative paths will be appended to the stage directory of the test.\nReFrame will set the following environment variables before running EasyBuild.\n```\nexport EASYBUILD_BUILDPATH={prefix}/build export EASYBUILD_INSTALLPATH={prefix}\nexport EASYBUILD_PREFIX={prefix}\nexport EASYBUILD_SOURCEPATH={prefix}\n```\nUsers can change these defaults by passing specific options to the\n`eb` command.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`easybuild`\n\n*class* reframe.core.buildsystems.Make(**args*, ***kwargs*)[[source]](_modules/reframe/core/buildsystems.html#Make)[¶](#reframe.core.buildsystems.Make)\nBases: [`BuildSystem`](#reframe.core.buildsystems.BuildSystem)\n\nA build system for compiling codes using `make`.\n\nThe generated build command has the following form:\n```\nmake -j [N] [-f MAKEFILE] [-C SRCDIR] CC=\"X\" CXX=\"X\" FC=\"X\" NVCC=\"X\" CPPFLAGS=\"X\" CFLAGS=\"X\" CXXFLAGS=\"X\" FCFLAGS=\"X\" LDFLAGS=\"X\" OPTIONS\n```\nThe compiler and compiler flags variables will only be passed if they are not `None`.\nTheir value is determined by the corresponding attributes of\n[`BuildSystem`](#reframe.core.buildsystems.BuildSystem).\nIf you want to completely disable passing these variables to the `make`\ninvocation, you should make sure not to set any of the correspoding attributes and set also the [`BuildSystem.flags_from_environ`](#reframe.core.buildsystems.BuildSystem.flags_from_environ) flag to\n`False`.\n\nmakefile *= None*[¶](#reframe.core.buildsystems.Make.makefile)\nInstruct build system to use this Makefile.\nThis option is useful when having non-standard Makefile names.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`None`\n\nmax_concurrency *= 1*[¶](#reframe.core.buildsystems.Make.max_concurrency)\nLimit concurrency for `make` jobs.\nThis attribute controls the `-j` option passed to `make`.\nIf not `None`, `make` will be invoked as `make -j max_concurrency`.\nOtherwise, it will invoked as `make -j`.\n\nType integer\n\nDefault\n`1`\n\nNote\n\nChanged in version 2.19: The default value is now `1`\n\noptions *= []*[¶](#reframe.core.buildsystems.Make.options)\nAppend these options to the `make` invocation.\nThis variable is also useful for passing variables or targets to\n`make`.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nsrcdir *= None*[¶](#reframe.core.buildsystems.Make.srcdir)\nThe top-level directory of the code.\n\nThis is set automatically by the framework based on the\n[`reframe.core.pipeline.RegressionTest.sourcepath`](#reframe.core.pipeline.RegressionTest.sourcepath) attribute.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`None`\n\n*class* reframe.core.buildsystems.SingleSource(**args*, ***kwargs*)[[source]](_modules/reframe/core/buildsystems.html#SingleSource)[¶](#reframe.core.buildsystems.SingleSource)\nBases: [`BuildSystem`](#reframe.core.buildsystems.BuildSystem)\n\nA build system for compiling a single source file.\n\nThe generated build command will have the following form:\n```\nCOMP CPPFLAGS XFLAGS SRCFILE -o EXEC LDFLAGS\n```\n* `COMP` is the required compiler for compiling `SRCFILE`.\nThis build system will automatically detect the programming language of the source file and pick the correct compiler.\nSee also the [`SingleSource.lang`](#reframe.core.buildsystems.SingleSource.lang) attribute.\n* `CPPFLAGS` are the preprocessor flags and are passed to any compiler.\n* `XFLAGS` is any of `CFLAGS`, `CXXFLAGS` or `FCFLAGS` depending on the programming language of the source file.\n* `SRCFILE` is the source file to be compiled.\nThis is set up automatically by the framework.\nSee also the [`SingleSource.srcfile`](#reframe.core.buildsystems.SingleSource.srcfile) attribute.\n* `EXEC` is the executable to be generated.\nThis is also set automatically by the framework.\nSee also the [`SingleSource.executable`](#reframe.core.buildsystems.SingleSource.executable) attribute.\n* `LDFLAGS` are the linker flags.\n\nFor CUDA codes, the language assumed is C++ (for the compilation flags) and the compiler used is [`BuildSystem.nvcc`](#reframe.core.buildsystems.BuildSystem.nvcc).\n\nexecutable *= None*[¶](#reframe.core.buildsystems.SingleSource.executable)\nThe executable file to be generated.\n\nThis is set automatically by the framework based on the\n[`reframe.core.pipeline.RegressionTest.executable`](#reframe.core.pipeline.RegressionTest.executable) attribute.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or `None`\n\ninclude_path *= []*[¶](#reframe.core.buildsystems.SingleSource.include_path)\nThe include path to be used for this compilation.\n\nAll the elements of this list will be appended to the\n[`BuildSystem.cppflags`](#reframe.core.buildsystems.BuildSystem.cppflags), by prepending to each of them the `-I`\noption.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\nlang *= None*[¶](#reframe.core.buildsystems.SingleSource.lang)\nThe programming language of the file that needs to be compiled.\nIf not specified, the build system will try to figure it out automatically based on the extension of the source file.\nThe automatically detected extensions are the following:\n\n> * C: .c and .upc.\n> * C++: .cc, .cp, .cxx, .cpp, .CPP, .c++ and .C.\n> * Fortran: .f, .for, .ftn, .F, .FOR, .fpp, .FPP, .FTN,\n> .f90, .f95, .f03, .f08, .F90, .F95, .F03 and .F08.\n> * CUDA: .cu.\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or `None`\n\nsrcfile *= None*[¶](#reframe.core.buildsystems.SingleSource.srcfile)\nThe source file to compile.\nThis is automatically set by the framework based on the\n[`reframe.core.pipeline.RegressionTest.sourcepath`](#reframe.core.pipeline.RegressionTest.sourcepath) attribute.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or `None`\n\n*class* reframe.core.buildsystems.Spack(**args*, ***kwargs*)[[source]](_modules/reframe/core/buildsystems.html#Spack)[¶](#reframe.core.buildsystems.Spack)\nBases: [`BuildSystem`](#reframe.core.buildsystems.BuildSystem)\n\nA build system for building test code using [Spack](https://spack.io/).\n\nReFrame will use a user-provided Spack environment in order to build and test a set of specs.\n\nNew in version 3.6.1.\n\nemit_load_cmds *= True*[¶](#reframe.core.buildsystems.Spack.emit_load_cmds)\nEmit the necessary `spack load` commands before running the test.\n\nType\n[`bool`](https://docs.python.org/3/library/functions.html#bool)\n\nDefault\n[`True`](https://docs.python.org/3/library/constants.html#True)\n\nenvironment *= None*[¶](#reframe.core.buildsystems.Spack.environment)\nThe Spack environment to use for building this test.\n\nReFrame will activate and install this environment.\nThis environment will also be used to run the test.\n```\nspack env activate -V -d <environment directory>\n```\nReFrame looks for environments in the test’s\n[`sourcesdir`](#reframe.core.pipeline.RegressionTest.sourcesdir).\n\nIf this field is None, the default, the environment name will be automatically set to rfm_spack_env.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or `None`\n\nDefault\n`None`\n\nNote\n\nChanged in version 3.7.3: The field is no longer required and the Spack environment will be automatically created if not provided.\n\ninstall_opts *= []*[¶](#reframe.core.buildsystems.Spack.install_opts)\nOptions to pass to `spack install`\n\nType\n`List[str]`\n\nDefault\n`[]`\n\ninstall_tree *= None*[¶](#reframe.core.buildsystems.Spack.install_tree)\nThe directory where Spack will install the packages requested by this test.\n\nAfter activating the Spack environment, ReFrame will set the install_tree Spack configuration in the given environment with the following command:\n```\nspack config add \"config:install_tree:root:<install tree>\"\n```\nRelative paths are resolved against the test’s stage directory. If this field and the Spack environment are both None, the default, the install directory will be automatically set to opt/spack. If this field None but the Spack environment is not, then install_tree will not be set automatically and the install tree of the given environment will not be overridden.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or `None`\n\nDefault\n`None`\n\nNew in version 3.7.3.\n\nspecs *= []*[¶](#reframe.core.buildsystems.Spack.specs)\nA list of additional specs to build and install within the given environment.\n\nReFrame will add the specs to the active environment by emititing the following command:\n```\nspack add spec1 spec2 ... specN\n```\nIf no spec is passed, ReFrame will simply install what is prescribed by the environment.\n\nType\n`List[str]`\n\nDefault\n`[]`\n\n###### Container Platforms[¶](#container-platforms)\n\nNew in version 2.20.\n\n*class* reframe.core.containers.Apptainer[[source]](_modules/reframe/core/containers.html#Apptainer)[¶](#reframe.core.containers.Apptainer)\nBases: [`Singularity`](#reframe.core.containers.Singularity)\n\nContainer platform backend for running containers with [Apptainer](https://apptainer.org/).\n\nNew in version 4.0.0.\n\n*class* reframe.core.containers.ContainerPlatform[[source]](_modules/reframe/core/containers.html#ContainerPlatform)[¶](#reframe.core.containers.ContainerPlatform)\nBases: [`ABC`](https://docs.python.org/3/library/abc.html#abc.ABC)\n\nThe abstract base class of any container platform.\n\ncommand[¶](#reframe.core.containers.ContainerPlatform.command)\nThe command to be executed within the container.\n\nIf no command is given, then the default command of the corresponding container image is going to be executed.\n\nNew in version 3.5.0: Changed the attribute name from commands to command and its type to a string.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or `None`\n\nDefault\n`None`\n\n*classmethod* create(*name*)[[source]](_modules/reframe/core/containers.html#ContainerPlatform.create)[¶](#reframe.core.containers.ContainerPlatform.create)\nFactory method to create a new container by name.\n\nimage[¶](#reframe.core.containers.ContainerPlatform.image)\nThe container image to be used for running the test.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or `None`\n\nDefault\n`None`\n\nmount_points[¶](#reframe.core.containers.ContainerPlatform.mount_points)\nList of mount point pairs for directories to mount inside the container.\n\nEach mount point is specified as a tuple of\n`(/path/in/host, /path/in/container)`. The stage directory of the ReFrame test is always mounted under `/rfm_workdir` inside the container, independelty of this field.\n\nType\n`list[tuple[str, str]]`\n\nDefault\n`[]`\n\noptions[¶](#reframe.core.containers.ContainerPlatform.options)\nAdditional options to be passed to the container runtime when executed.\n\nType\n`list[str]`\n\nDefault\n`[]`\n\npull_image[¶](#reframe.core.containers.ContainerPlatform.pull_image)\nPull the container image before running.\n\nThis does not have any effect for the Singularity container platform.\n\nNew in version 3.5.\n\nType\n[`bool`](https://docs.python.org/3/library/functions.html#bool)\n\nDefault\n`True`\n\nworkdir[¶](#reframe.core.containers.ContainerPlatform.workdir)\nThe working directory of ReFrame inside the container.\n\nThis is the directory where the test’s stage directory is mounted inside the container. This directory is always mounted regardless if\n[`mount_points`](#reframe.core.containers.ContainerPlatform.mount_points) is set or not.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`/rfm_workdir`\n\nChanged in version 3.12.0: This attribute is no more deprecated.\n\n*class* reframe.core.containers.Docker[[source]](_modules/reframe/core/containers.html#Docker)[¶](#reframe.core.containers.Docker)\nBases: [`ContainerPlatform`](#reframe.core.containers.ContainerPlatform)\n\nContainer platform backend for running containers with [Docker](https://www.docker.com/).\n\n*class* reframe.core.containers.Sarus[[source]](_modules/reframe/core/containers.html#Sarus)[¶](#reframe.core.containers.Sarus)\nBases: [`ContainerPlatform`](#reframe.core.containers.ContainerPlatform)\n\nContainer platform backend for running containers with [Sarus](https://sarus.readthedocs.io).\n\nwith_mpi[¶](#reframe.core.containers.Sarus.with_mpi)\nEnable MPI support when launching the container.\n\nType boolean\n\nDefault\n`False`\n\n*class* reframe.core.containers.Shifter[[source]](_modules/reframe/core/containers.html#Shifter)[¶](#reframe.core.containers.Shifter)\nBases: [`Sarus`](#reframe.core.containers.Sarus)\n\nContainer platform backend for running containers with [Shifter](https://www.nersc.gov/research-and-development/user-defined-images/).\n\n*class* reframe.core.containers.Singularity[[source]](_modules/reframe/core/containers.html#Singularity)[¶](#reframe.core.containers.Singularity)\nBases: [`ContainerPlatform`](#reframe.core.containers.ContainerPlatform)\n\nContainer platform backend for running containers with [Singularity](https://sylabs.io/).\n\nwith_cuda[¶](#reframe.core.containers.Singularity.with_cuda)\nEnable CUDA support when launching the container.\n\nType boolean\n\nDefault\n`False`\n\n###### The `reframe` module[¶](#the-reframe-module)\n\nThe `reframe` module offers direct access to the basic test classes, constants and decorators.\n\n*class* reframe.CompileOnlyRegressionTest[¶](#reframe.core.containers.reframe.CompileOnlyRegressionTest)\nSee [`reframe.core.pipeline.CompileOnlyRegressionTest`](#reframe.core.pipeline.CompileOnlyRegressionTest).\n\n*class* reframe.RegressionTest[¶](#reframe.core.containers.reframe.RegressionTest)\nSee [`reframe.core.pipeline.RegressionTest`](#reframe.core.pipeline.RegressionTest).\n\n*class* reframe.RunOnlyRegressionTest[¶](#reframe.core.containers.reframe.RunOnlyRegressionTest)\nSee [`reframe.core.pipeline.RunOnlyRegressionTest`](#reframe.core.pipeline.RunOnlyRegressionTest).\n\n@reframe.simple_test[¶](#reframe.core.containers.reframe.simple_test)\nSee [`@reframe.core.decorators.simple_test`](#reframe.core.decorators.simple_test).\n\n###### Mapping of Test Attributes to Job Scheduler Backends[¶](#mapping-of-test-attributes-to-job-scheduler-backends)\n\n| Test attribute | Slurm option | Torque option | PBS option |\n| --- | --- | --- | --- |\n| `num_tasks` | `--ntasks`1 | `-l nodes={num_tasks//num_tasks_per_node}:ppn={num_tasks_per_node*num_cpus_per_task}` | `-l select={num_tasks//num_tasks_per_node}:mpiprocs={num_tasks_per_node}:ncpus={num_tasks_per_node*num_cpus_per_task}` |\n| `num_tasks_per_node` | `--ntasks-per-node` | see `num_tasks` | see `num_tasks` |\n| `num_tasks_per_core` | `--ntasks-per-core` | n/a | n/a |\n| `num_tasks_per_socket` | `--ntasks-per-socket` | n/a | n/a |\n| `num_cpus_per_task` | `--cpus-per-task` | see `num_tasks` | see `num_tasks` |\n| `time_limit` | `--time=hh:mm:ss` | `-l walltime=hh:mm:ss` | `-l walltime=hh:mm::ss` |\n| `exclusive_access` | `--exclusive` | n/a | n/a |\n| `use_smt` | `--hint=[no]multithread` | n/a | n/a |\n\nIf any of the attributes is set to `None` it will not be emitted at all in the job script.\nIn cases that the attribute is required, it will be set to `1`.\n\n1 The `--nodes` option may also be emitted if the [`use_nodes_option`](index.html#config.systems.partitions.sched_options.use_nodes_option) scheduler configuration parameter is set.\n\n##### Deferrable Functions Reference[¶](#deferrable-functions-reference)\n\n*Deferrable functions* are the functions whose execution may be postponed to a later time after they are called.\nThe key characteristic of these functions is that they store their arguments when they are called, and the execution itself does not occur until the function is evaluated either explicitly or implicitly.\n\nReFrame provides an ample set of deferrable utilities and it also allows users to write their own deferrable functions when needed.\nPlease refer to “[Understanding the Mechanism of Deferrable Functions](index.html#document-deferrables)” for a hands-on explanation on how deferrable functions work and how to create custom deferrable functions.\n\n###### Explicit evaluation of deferrable functions[¶](#explicit-evaluation-of-deferrable-functions)\n\nDeferrable functions may be evaluated at any time by calling `evaluate()` on their return value or by passing the deferred function itself to the [`evaluate()`](#reframe.utility.sanity.evaluate) free function.\nThese `evaluate()` functions take an optional [`bool`](https://docs.python.org/3/library/functions.html#bool) argument `cache`, which can be used to cache the evaluation of the deferrable function.\nHence, if caching is enabled on a given deferrable function, any subsequent calls to `evaluate()` will simply return the previously cached results.\n\nChanged in version 3.8.0: Support of cached evaluation is added.\n\n###### Implicit evaluation of deferrable functions[¶](#implicit-evaluation-of-deferrable-functions)\n\nDeferrable functions may also be evaluated implicitly in the following situations:\n\n* When you try to get their truthy value by either explicitly or implicitly calling `bool` on their return value.\nThis implies that when you include the result of a deferrable function in an [`if`](https://docs.python.org/3/reference/compound_stmts.html#if) statement or when you apply the [`and`](https://docs.python.org/3/reference/expressions.html#and), [`or`](https://docs.python.org/3/reference/expressions.html#or) or [`not`](https://docs.python.org/3/reference/expressions.html#not) operators, this will trigger their immediate evaluation.\n* When you try to iterate over their result.\nThis implies that including the result of a deferrable function in a [`for`](https://docs.python.org/3/reference/compound_stmts.html#for) statement will trigger its evaluation immediately.\n* When you try to explicitly or implicitly get its string representation by calling `str` on its result.\nThis implies that printing the return value of a deferrable function will automatically trigger its evaluation.\n\n###### Categories of deferrable functions[¶](#categories-of-deferrable-functions)\n\nCurrently ReFrame provides three broad categories of deferrable functions:\n\n1. Deferrable replacements of certain Python built-in functions.\nThese functions simply delegate their execution to the actual built-ins.\n2. Assertion functions.\nThese functions are used to assert certain conditions and they either return `True` or raise [`SanityError`](index.html#reframe.core.exceptions.SanityError) with a message describing the error.\nUsers may provide their own formatted messages through the `msg` argument.\nFor example, in the following call to `assert_eq()` the `{0}` and `{1}` placeholders will obtain the actual arguments passed to the assertion function.\n```\nassert_eq(a, 1, msg=\"{0} is not equal to {1}\")\n```\nIf in the user provided message more placeholders are used than the arguments of the assert function (except the `msg` argument), no argument substitution will be performed in the user message.\n3. Utility functions.\nThey include, but are not limited to, functions to iterate over regex matches in a file, extracting and converting values from regex matches, computing statistical information on series of data etc.\n\n####### Deferrable performance functions[¶](#deferrable-performance-functions)\n\nNew in version 3.8.0.\n\nDeferrable performance functions are a special type of deferrable functions which are intended for measuring a given quantity.\nTherefore, this kind of deferrable functions have an associated unit that can be used to interpret the return values from these functions.\nThe unit of a deferrable performance function can be accessed through the public member `unit`.\nRegular deferrable functions can be promoted to deferrable performance functions using the [`make_performance_function()`](#reframe.utility.sanity.make_performance_function) utility.\nAlso, this utility allows to create performance functions directly from any callable.\n\n###### List of deferrable functions and utilities[¶](#list-of-deferrable-functions-and-utilities)\n\n@reframe.utility.sanity.deferrable(*func*)[¶](#reframe.utility.sanity.deferrable)\nDeferrable decorator.\n\nConverts the decorated free function into a deferrable function.\n```\nimport reframe.utility.sanity as sn\n\n@sn.deferrable def myfunc(*args):\n do_sth()\n```\nreframe.utility.sanity.abs(*x*)[[source]](_modules/reframe/utility/sanity.html#abs)[¶](#reframe.utility.sanity.abs)\nReplacement for the built-in [`abs()`](https://docs.python.org/3/library/functions.html#abs) function.\n\nreframe.utility.sanity.all(*iterable*)[[source]](_modules/reframe/utility/sanity.html#all)[¶](#reframe.utility.sanity.all)\nReplacement for the built-in [`all()`](https://docs.python.org/3/library/functions.html#all) function.\n\nreframe.utility.sanity.allx(*iterable*)[[source]](_modules/reframe/utility/sanity.html#allx)[¶](#reframe.utility.sanity.allx)\nSame as the built-in [`all()`](https://docs.python.org/3/library/functions.html#all) function, except that it returns `False` if `iterable` is empty.\n\nNew in version 2.13.\n\nreframe.utility.sanity.and_(*a*, *b*)[[source]](_modules/reframe/utility/sanity.html#and_)[¶](#reframe.utility.sanity.and_)\nDeferrable version of the [`and`](https://docs.python.org/3/reference/expressions.html#and) operator.\n\nReturns\n`a and b`.\n\nreframe.utility.sanity.any(*iterable*)[[source]](_modules/reframe/utility/sanity.html#any)[¶](#reframe.utility.sanity.any)\nReplacement for the built-in [`any()`](https://docs.python.org/3/library/functions.html#any) function.\n\nreframe.utility.sanity.assert_bounded(*val*, *lower=None*, *upper=None*, *msg=None*)[[source]](_modules/reframe/utility/sanity.html#assert_bounded)[¶](#reframe.utility.sanity.assert_bounded)\nAssert that `lower <= val <= upper`.\n\nParameters\n* **val** – The value to check.\n* **lower** – The lower bound. If `None`, it defaults to `-inf`.\n* **upper** – The upper bound. If `None`, it defaults to `inf`.\n* **msg** – The error message to use if the assertion fails. You may use\n`{0}` … `{N}` as placeholders for the function arguments.\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails.\n\nreframe.utility.sanity.assert_eq(*a*, *b*, *msg=None*)[[source]](_modules/reframe/utility/sanity.html#assert_eq)[¶](#reframe.utility.sanity.assert_eq)\nAssert that `a == b`.\n\nParameters\n**msg** – The error message to use if the assertion fails. You may use\n`{0}` … `{N}` as placeholders for the function arguments.\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails.\n\nreframe.utility.sanity.assert_false(*x*, *msg=None*)[[source]](_modules/reframe/utility/sanity.html#assert_false)[¶](#reframe.utility.sanity.assert_false)\nAssert that `x` is evaluated to `False`.\n\nParameters\n**msg** – The error message to use if the assertion fails. You may use\n`{0}` … `{N}` as placeholders for the function arguments.\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails.\n\nreframe.utility.sanity.assert_found(*patt*, *filename*, *msg=None*, *encoding='utf-8'*)[[source]](_modules/reframe/utility/sanity.html#assert_found)[¶](#reframe.utility.sanity.assert_found)\nAssert that regex pattern `patt` is found in the file `filename`.\n\nParameters\n* **patt** – The regex pattern to search.\nAny standard Python [regular expression](https://docs.python.org/3/library/re.html#regular-expression-syntax)\nis accepted.\nThe [re.MULTILINE](https://docs.python.org/3/library/re.html#re.MULTILINE) flag is set for the pattern search.\n* **filename** – The name of the file to examine or a file descriptor as in\n[`open()`](https://docs.python.org/3/library/functions.html#open). Any [`OSError`](https://docs.python.org/3/library/exceptions.html#OSError) raised while processing the file will be propagated as a [`reframe.core.exceptions.SanityError`](index.html#reframe.core.exceptions.SanityError).\n* **msg** – The error message to use if the assertion fails. You may use\n`{0}` … `{N}` as placeholders for the function arguments.\n* **encoding** – The name of the encoding used to decode the file.\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails.\n\nreframe.utility.sanity.assert_found_s(*patt*, *string*, *msg=None*)[[source]](_modules/reframe/utility/sanity.html#assert_found_s)[¶](#reframe.utility.sanity.assert_found_s)\nAssert that regex pattern `patt` is found in the string `string`.\n\nParameters\n* **patt** – as in [`assert_found()`](#reframe.utility.sanity.assert_found).\n* **string** – The string to examine.\n* **msg** – as in [`assert_found()`](#reframe.utility.sanity.assert_found). You may use\n`{0}` … `{N}` as placeholders for the function arguments.\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails.\n\nNew in version 3.4.1.\n\nreframe.utility.sanity.assert_ge(*a*, *b*, *msg=None*)[[source]](_modules/reframe/utility/sanity.html#assert_ge)[¶](#reframe.utility.sanity.assert_ge)\nAssert that `a >= b`.\n\nParameters\n**msg** – The error message to use if the assertion fails. You may use\n`{0}` … `{N}` as placeholders for the function arguments.\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails.\n\nreframe.utility.sanity.assert_gt(*a*, *b*, *msg=None*)[[source]](_modules/reframe/utility/sanity.html#assert_gt)[¶](#reframe.utility.sanity.assert_gt)\nAssert that `a > b`.\n\nParameters\n**msg** – The error message to use if the assertion fails. You may use\n`{0}` … `{N}` as placeholders for the function arguments.\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails.\n\nreframe.utility.sanity.assert_in(*item*, *container*, *msg=None*)[[source]](_modules/reframe/utility/sanity.html#assert_in)[¶](#reframe.utility.sanity.assert_in)\nAssert that `item` is in `container`.\n\nParameters\n**msg** – The error message to use if the assertion fails. You may use\n`{0}` … `{N}` as placeholders for the function arguments.\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails.\n\nreframe.utility.sanity.assert_le(*a*, *b*, *msg=None*)[[source]](_modules/reframe/utility/sanity.html#assert_le)[¶](#reframe.utility.sanity.assert_le)\nAssert that `a <= b`.\n\nParameters\n**msg** – The error message to use if the assertion fails. You may use\n`{0}` … `{N}` as placeholders for the function arguments.\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails.\n\nreframe.utility.sanity.assert_lt(*a*, *b*, *msg=None*)[[source]](_modules/reframe/utility/sanity.html#assert_lt)[¶](#reframe.utility.sanity.assert_lt)\nAssert that `a < b`.\n\nParameters\n**msg** – The error message to use if the assertion fails. You may use\n`{0}` … `{N}` as placeholders for the function arguments.\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails.\n\nreframe.utility.sanity.assert_ne(*a*, *b*, *msg=None*)[[source]](_modules/reframe/utility/sanity.html#assert_ne)[¶](#reframe.utility.sanity.assert_ne)\nAssert that `a != b`.\n\nParameters\n**msg** – The error message to use if the assertion fails. You may use\n`{0}` … `{N}` as placeholders for the function arguments.\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails.\n\nreframe.utility.sanity.assert_not_found(*patt*, *filename*, *msg=None*, *encoding='utf-8'*)[[source]](_modules/reframe/utility/sanity.html#assert_not_found)[¶](#reframe.utility.sanity.assert_not_found)\nAssert that regex pattern `patt` is not found in the file\n`filename`.\n\nThis is the inverse of [`assert_found()`](#reframe.utility.sanity.assert_found).\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails.\n\nreframe.utility.sanity.assert_not_found_s(*patt*, *string*, *msg=None*)[[source]](_modules/reframe/utility/sanity.html#assert_not_found_s)[¶](#reframe.utility.sanity.assert_not_found_s)\nAssert that regex pattern `patt` is not found in `string`.\n\nThis is the inverse of [`assert_found_s()`](#reframe.utility.sanity.assert_found_s).\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails.\n\nNew in version 3.4.1.\n\nreframe.utility.sanity.assert_not_in(*item*, *container*, *msg=None*)[[source]](_modules/reframe/utility/sanity.html#assert_not_in)[¶](#reframe.utility.sanity.assert_not_in)\nAssert that `item` is not in `container`.\n\nParameters\n**msg** – The error message to use if the assertion fails. You may use\n`{0}` … `{N}` as placeholders for the function arguments.\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails.\n\nreframe.utility.sanity.assert_reference(*val*, *ref*, *lower_thres=None*, *upper_thres=None*, *msg=None*)[[source]](_modules/reframe/utility/sanity.html#assert_reference)[¶](#reframe.utility.sanity.assert_reference)\nAssert that value `val` respects the reference value `ref`.\n\nParameters\n* **val** – The value to check.\n* **ref** – The reference value.\n* **lower_thres** – The lower threshold value expressed as a negative decimal fraction of the reference value. Must be in [-1, 0] for ref >= 0.0 and in [-inf, 0] for ref < 0.0.\nIf `None`, no lower thresholds is applied.\n* **upper_thres** – The upper threshold value expressed as a decimal fraction of the reference value. Must be in [0, inf] for ref >= 0.0 and in [0, 1] for ref < 0.0.\nIf `None`, no upper thresholds is applied.\n* **msg** – The error message to use if the assertion fails. You may use\n`{0}` … `{N}` as placeholders for the function arguments.\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails or if the lower and upper thresholds do not have appropriate values.\n\nreframe.utility.sanity.assert_true(*x*, *msg=None*)[[source]](_modules/reframe/utility/sanity.html#assert_true)[¶](#reframe.utility.sanity.assert_true)\nAssert that `x` is evaluated to `True`.\n\nParameters\n**msg** – The error message to use if the assertion fails. You may use\n`{0}` … `{N}` as placeholders for the function arguments.\n\nReturns\n`True` on success.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – if assertion fails.\n\nreframe.utility.sanity.avg(*iterable*)[[source]](_modules/reframe/utility/sanity.html#avg)[¶](#reframe.utility.sanity.avg)\nReturn the average of all the elements of `iterable`.\n\nreframe.utility.sanity.chain(**iterables*)[[source]](_modules/reframe/utility/sanity.html#chain)[¶](#reframe.utility.sanity.chain)\nReplacement for the [`itertools.chain()`](https://docs.python.org/3/library/itertools.html#itertools.chain)\nfunction.\n\nreframe.utility.sanity.contains(*seq*, *key*)[[source]](_modules/reframe/utility/sanity.html#contains)[¶](#reframe.utility.sanity.contains)\nDeferrable version of the [`in`](https://docs.python.org/3/reference/expressions.html#in) operator.\n\nReturns\n`key in seq`.\n\nreframe.utility.sanity.count(*iterable*)[[source]](_modules/reframe/utility/sanity.html#count)[¶](#reframe.utility.sanity.count)\nReturn the element count of `iterable`.\n\nThis is similar to the built-in [`len()`](https://docs.python.org/3/library/functions.html#len), except that it can also handle any argument that supports iteration, including generators.\n\nreframe.utility.sanity.count_uniq(*iterable*)[[source]](_modules/reframe/utility/sanity.html#count_uniq)[¶](#reframe.utility.sanity.count_uniq)\nReturn the unique element count of `iterable`.\n\nreframe.utility.sanity.defer(*x*)[[source]](_modules/reframe/utility/sanity.html#defer)[¶](#reframe.utility.sanity.defer)\nDefer the evaluation of variable `x`.\n\nNew in version 2.21.\n\nreframe.utility.sanity.enumerate(*iterable*, *start=0*)[[source]](_modules/reframe/utility/sanity.html#enumerate)[¶](#reframe.utility.sanity.enumerate)\nReplacement for the built-in\n[`enumerate()`](https://docs.python.org/3/library/functions.html#enumerate) function.\n\nreframe.utility.sanity.evaluate(*expr*, *cache=False*)[[source]](_modules/reframe/utility/sanity.html#evaluate)[¶](#reframe.utility.sanity.evaluate)\nEvaluate a deferred expression.\n\nIf `expr` is not a deferred expression, it will be returned as is.\nIf `expr` is a deferred expression and `cache` is `True`, the results of the deferred expression will be cached and subsequent calls to [`evaluate()`](#reframe.utility.sanity.evaluate) on this deferred expression (when `cache=False`)\nwill simply return the previously cached result.\n\nParameters\n* **expr** – The expression to be evaluated.\n* **cache** – Cache the result of this evaluation.\n\nNote\n\nWhen the `cache` argument is passed as `True`, a deferred expression will always be evaluated and its results will be re-cached.\nThis may replace any other results that may have been cached in previous evaluations.\n\nNew in version 2.21.\n\nChanged in version 3.8.0: The `cache` argument is added.\n\nreframe.utility.sanity.extractall(*patt*, *filename*, *tag=0*, *conv=None*, *encoding='utf-8'*)[[source]](_modules/reframe/utility/sanity.html#extractall)[¶](#reframe.utility.sanity.extractall)\nExtract all values from the capturing group `tag` of a matching regex\n`patt` in the file `filename`.\n\nParameters\n* **patt** – The regex pattern to search.\nAny standard Python [regular expression](https://docs.python.org/3/library/re.html#regular-expression-syntax)\nis accepted.\nThe [re.MULTILINE](https://docs.python.org/3/library/re.html#re.MULTILINE) flag is set for the pattern search.\n* **filename** – The name of the file to examine or a file descriptor as in\n[`open()`](https://docs.python.org/3/library/functions.html#open).\n* **encoding** – The name of the encoding used to decode the file.\n* **tag** – The regex capturing group to be extracted.\nGroup `0` refers always to the whole match.\nSince the file is processed line by line, this means that group `0`\nreturns the whole line that was matched.\n* **conv** – A callable or iterable of callables taking a single argument and returning a new value.\nIf not an iterable, it will be used to convert the extracted values for all the capturing groups specified in `tag`.\nOtherwise, each conversion function will be used to convert the value extracted from the corresponding capturing group in `tag`.\nIf more conversion functions are supplied than the corresponding capturing groups in `tag`, the last conversion function will be used for the additional capturing groups.\n\nReturns A list of tuples of converted values extracted from the capturing groups specified in `tag`, if `tag` is an iterable.\nOtherwise, a list of the converted values extracted from the single capturing group specified in `tag`.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – In case of errors.\n\nChanged in version 3.1: Multiple regex capturing groups are now supporetd via `tag` and multiple conversion functions can be used in `conv`.\n\nreframe.utility.sanity.extractall_s(*patt*, *string*, *tag=0*, *conv=None*)[[source]](_modules/reframe/utility/sanity.html#extractall_s)[¶](#reframe.utility.sanity.extractall_s)\n\nExtract all values from the capturing group `tag` of a matching regex`patt` in `string`.\n\narg patt as in [`extractall()`](#reframe.utility.sanity.extractall).\n\n‘ :arg string: The string to examine.\narg tag as in [`extractall()`](#reframe.utility.sanity.extractall).\n\narg conv as in [`extractall()`](#reframe.utility.sanity.extractall).\n\nreturns same as [`extractall()`](#reframe.utility.sanity.extractall).\n\nNew in version 3.4.1.\n\nreframe.utility.sanity.extractiter(*patt*, *filename*, *tag=0*, *conv=None*, *encoding='utf-8'*)[[source]](_modules/reframe/utility/sanity.html#extractiter)[¶](#reframe.utility.sanity.extractiter)\nGet an iterator over the values extracted from the capturing group\n`tag` of a matching regex `patt` in the file `filename`.\n\nThis function is equivalent to [`extractall()`](#reframe.utility.sanity.extractall) except that it returns a generator object, instead of a list, which you can use to iterate over the extracted values.\n\nreframe.utility.sanity.extractiter_s(*patt*, *string*, *tag=0*, *conv=None*)[[source]](_modules/reframe/utility/sanity.html#extractiter_s)[¶](#reframe.utility.sanity.extractiter_s)\nGet an iterator over the values extracted from the capturing group\n`tag` of a matching regex `patt` in `string`.\n\nThis function is equivalent to [`extractall_s()`](#reframe.utility.sanity.extractall_s) except that it returns a generator object, instead of a list, which you can use to iterate over the extracted values.\n\nNew in version 3.4.1.\n\nreframe.utility.sanity.extractsingle(*patt*, *filename*, *tag=0*, *conv=None*, *item=0*, *encoding='utf-8'*)[[source]](_modules/reframe/utility/sanity.html#extractsingle)[¶](#reframe.utility.sanity.extractsingle)\nExtract a single value from the capturing group `tag` of a matching regex `patt` in the file `filename`.\n\nThis function is equivalent to `extractall(patt, filename, tag,\nconv)[item]`, except that it raises a `SanityError` if `item` is out of bounds.\n\nParameters\n* **patt** – as in [`extractall()`](#reframe.utility.sanity.extractall).\n* **filename** – as in [`extractall()`](#reframe.utility.sanity.extractall).\n* **encoding** – as in [`extractall()`](#reframe.utility.sanity.extractall).\n* **tag** – as in [`extractall()`](#reframe.utility.sanity.extractall).\n* **conv** – as in [`extractall()`](#reframe.utility.sanity.extractall).\n* **item** – the specific element to extract.\n\nReturns The extracted value.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – In case of errors.\n\nreframe.utility.sanity.extractsingle_s(*patt*, *string*, *tag=0*, *conv=None*, *item=0*)[[source]](_modules/reframe/utility/sanity.html#extractsingle_s)[¶](#reframe.utility.sanity.extractsingle_s)\nExtract a single value from the capturing group `tag` of a matching regex `patt` in `string`.\n\nThis function is equivalent to `extractall_s(patt, string, tag,\nconv)[item]`, except that it raises a `SanityError` if `item` is out of bounds.\n\nParameters\n* **patt** – as in [`extractall_s()`](#reframe.utility.sanity.extractall_s).\n* **string** – as in [`extractall_s()`](#reframe.utility.sanity.extractall_s).\n* **tag** – as in [`extractall_s()`](#reframe.utility.sanity.extractall_s).\n* **conv** – as in [`extractall_s()`](#reframe.utility.sanity.extractall_s).\n* **item** – the specific element to extract.\n\nReturns The extracted value.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – In case of errors.\n\nNew in version 3.4.1.\n\nreframe.utility.sanity.filter(*function*, *iterable*)[[source]](_modules/reframe/utility/sanity.html#filter)[¶](#reframe.utility.sanity.filter)\nReplacement for the built-in\n[`filter()`](https://docs.python.org/3/library/functions.html#filter) function.\n\nreframe.utility.sanity.findall(*patt*, *filename*, *encoding='utf-8'*)[[source]](_modules/reframe/utility/sanity.html#findall)[¶](#reframe.utility.sanity.findall)\nGet all matches of regex `patt` in `filename`.\n\nParameters\n* **patt** – The regex pattern to search.\nAny standard Python [regular expression](https://docs.python.org/3/library/re.html#regular-expression-syntax)\nis accepted.\nThe [re.MULTILINE](https://docs.python.org/3/library/re.html#re.MULTILINE) flag is set for the pattern search.\n* **filename** – The name of the file to examine.\n* **encoding** – The name of the encoding used to decode the file.\n\nReturns A list of raw [regex match objects](https://docs.python.org/3/library/re.html#match-objects).\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – In case an [`OSError`](https://docs.python.org/3/library/exceptions.html#OSError) is raised while processing `filename`.\n\nreframe.utility.sanity.findall_s(*patt*, *string*)[[source]](_modules/reframe/utility/sanity.html#findall_s)[¶](#reframe.utility.sanity.findall_s)\nGet all matches of regex `patt` in `string`.\n\nParameters\n* **patt** – as in [`findall()`](#reframe.utility.sanity.findall)\n* **string** – The string to examine.\n\nReturns same as `finall()`.\n\nNew in version 3.4.1.\n\nreframe.utility.sanity.finditer(*patt*, *filename*, *encoding='utf-8'*)[[source]](_modules/reframe/utility/sanity.html#finditer)[¶](#reframe.utility.sanity.finditer)\nGet an iterator over the matches of the regex `patt` in `filename`.\n\nThis function is equivalent to [`findall()`](#reframe.utility.sanity.findall) except that it returns a generator object instead of a list, which you can use to iterate over the raw matches.\n\nreframe.utility.sanity.finditer_s(*patt*, *string*)[[source]](_modules/reframe/utility/sanity.html#finditer_s)[¶](#reframe.utility.sanity.finditer_s)\nGet an iterator over the matches of the regex `patt` in `string`.\n\nThis function is equivalent to [`findall_s()`](#reframe.utility.sanity.findall_s) except that it returns a generator object instead of a list, which you can use to iterate over the raw matches.\n\nNew in version 3.4.1.\n\nreframe.utility.sanity.getattr(*obj*, *attr*, **args*)[[source]](_modules/reframe/utility/sanity.html#getattr)[¶](#reframe.utility.sanity.getattr)\nReplacement for the built-in\n[`getattr()`](https://docs.python.org/3/library/functions.html#getattr) function.\n\nreframe.utility.sanity.getitem(*container*, *item*)[[source]](_modules/reframe/utility/sanity.html#getitem)[¶](#reframe.utility.sanity.getitem)\nGet `item` from `container`.\n\n`container` may refer to any container that can be indexed.\n\nRaises\n[**reframe.core.exceptions.SanityError**](index.html#reframe.core.exceptions.SanityError) – In case `item` cannot be retrieved from `container`.\n\nreframe.utility.sanity.glob(*pathname*, ***, *recursive=False*)[[source]](_modules/reframe/utility/sanity.html#glob)[¶](#reframe.utility.sanity.glob)\nReplacement for the [`glob.glob()`](https://docs.python.org/3/library/glob.html#glob.glob) function.\n\nreframe.utility.sanity.hasattr(*obj*, *name*)[[source]](_modules/reframe/utility/sanity.html#hasattr)[¶](#reframe.utility.sanity.hasattr)\nReplacement for the built-in\n[`hasattr()`](https://docs.python.org/3/library/functions.html#hasattr) function.\n\nreframe.utility.sanity.iglob(*pathname*, *recursive=False*)[[source]](_modules/reframe/utility/sanity.html#iglob)[¶](#reframe.utility.sanity.iglob)\nReplacement for the [`glob.iglob()`](https://docs.python.org/3/library/glob.html#glob.iglob)\nfunction.\n\nreframe.utility.sanity.len(*s*)[[source]](_modules/reframe/utility/sanity.html#len)[¶](#reframe.utility.sanity.len)\nReplacement for the built-in [`len()`](https://docs.python.org/3/library/functions.html#len) function.\n\nreframe.utility.sanity.make_performance_function(*func*, *unit*, **args*, ***kwargs*)[[source]](_modules/reframe/utility/sanity.html#make_performance_function)[¶](#reframe.utility.sanity.make_performance_function)\nConvert a callable or deferred expression into a performance function.\n\nIf `func` is a deferred expression, the performance function will be built by extending this deferred expression into a deferred performance expression. Otherwise, a new deferred performance expression will be created from the function `func()`. The argument `unit` is the unit associated with the deferrable performance expression, and `*args` and\n`**kwargs` are the arguments to be captured by this deferred expression.\nSee\n[deferrable functions reference](index.html#document-deferrable_functions_reference)\nfor further information on deferrable functions.\n\nNew in version 3.8.0.\n\nreframe.utility.sanity.map(*function*, **iterables*)[[source]](_modules/reframe/utility/sanity.html#map)[¶](#reframe.utility.sanity.map)\nReplacement for the built-in [`map()`](https://docs.python.org/3/library/functions.html#map) function.\n\nreframe.utility.sanity.max(**args*)[[source]](_modules/reframe/utility/sanity.html#max)[¶](#reframe.utility.sanity.max)\nReplacement for the built-in [`max()`](https://docs.python.org/3/library/functions.html#max) function.\n\nreframe.utility.sanity.min(**args*)[[source]](_modules/reframe/utility/sanity.html#min)[¶](#reframe.utility.sanity.min)\nReplacement for the built-in [`min()`](https://docs.python.org/3/library/functions.html#min) function.\n\nreframe.utility.sanity.not_(*a*)[[source]](_modules/reframe/utility/sanity.html#not_)[¶](#reframe.utility.sanity.not_)\nDeferrable version of the [`not`](https://docs.python.org/3/reference/expressions.html#not) operator.\n\nReturns\n`not a`.\n\nreframe.utility.sanity.or_(*a*, *b*)[[source]](_modules/reframe/utility/sanity.html#or_)[¶](#reframe.utility.sanity.or_)\nDeferrable version of the [`or`](https://docs.python.org/3/reference/expressions.html#or) operator.\n\nReturns\n`a or b`.\n\nreframe.utility.sanity.path_exists(*path*)[[source]](_modules/reframe/utility/sanity.html#path_exists)[¶](#reframe.utility.sanity.path_exists)\nReplacement for the [`os.path.exists()`](https://docs.python.org/3/library/os.path.html#os.path.exists) function.\n\nNew in version 3.4.\n\nreframe.utility.sanity.path_isdir(*path*)[[source]](_modules/reframe/utility/sanity.html#path_isdir)[¶](#reframe.utility.sanity.path_isdir)\nReplacement for the [`os.path.isdir()`](https://docs.python.org/3/library/os.path.html#os.path.isdir) function.\n\nNew in version 3.4.\n\nreframe.utility.sanity.path_isfile(*path*)[[source]](_modules/reframe/utility/sanity.html#path_isfile)[¶](#reframe.utility.sanity.path_isfile)\nReplacement for the [`os.path.isfile()`](https://docs.python.org/3/library/os.path.html#os.path.isfile) function.\n\nNew in version 3.4.\n\nreframe.utility.sanity.path_islink(*path*)[[source]](_modules/reframe/utility/sanity.html#path_islink)[¶](#reframe.utility.sanity.path_islink)\nReplacement for the [`os.path.islink()`](https://docs.python.org/3/library/os.path.html#os.path.islink) function.\n\nNew in version 3.4.\n\nreframe.utility.sanity.print(*obj*, ***, *sep=' '*, *end='\\n'*, *file=None*, *flush=False*)[[source]](_modules/reframe/utility/sanity.html#print)[¶](#reframe.utility.sanity.print)\nReplacement for the built-in [`print()`](https://docs.python.org/3/library/functions.html#print) function.\n\nThe only difference is that this function takes a *single* object argument and it returns that, so that you can use it transparently inside a complex sanity expression. For example, you could write the following to print the matches returned from the [`extractall()`](#reframe.utility.sanity.extractall) function:\n```\n@sanity_function def my_sanity_fn(self):\n return sn.assert_eq(\n sn.count(sn.print(sn.extractall(...))), 10\n )\n```\nIf `file` is None, [`print()`](#reframe.utility.sanity.print) will print its arguments to the standard output. Unlike the builtin [`print()`](https://docs.python.org/3/library/functions.html#print)\nfunction, we don’t bind the `file` argument to `sys.stdout` by default. This would capture `sys.stdout` at the time this function is defined and would prevent it from seeing changes to `sys.stdout`,\nsuch as redirects, in the future.\n\nChanged in version 3.4: This function accepts now a single object argument in contrast to the built-in [`print()`](https://docs.python.org/3/library/functions.html#print) function, which accepts multiple.\n\nreframe.utility.sanity.reversed(*seq*)[[source]](_modules/reframe/utility/sanity.html#reversed)[¶](#reframe.utility.sanity.reversed)\nReplacement for the built-in\n[`reversed()`](https://docs.python.org/3/library/functions.html#reversed) function.\n\nreframe.utility.sanity.round(*number*, **args*)[[source]](_modules/reframe/utility/sanity.html#round)[¶](#reframe.utility.sanity.round)\nReplacement for the built-in\n[`round()`](https://docs.python.org/3/library/functions.html#round) function.\n\nreframe.utility.sanity.setattr(*obj*, *name*, *value*)[[source]](_modules/reframe/utility/sanity.html#setattr)[¶](#reframe.utility.sanity.setattr)\nReplacement for the built-in\n[`setattr()`](https://docs.python.org/3/library/functions.html#setattr) function.\n\nreframe.utility.sanity.sorted(*iterable*, **args*)[[source]](_modules/reframe/utility/sanity.html#sorted)[¶](#reframe.utility.sanity.sorted)\nReplacement for the built-in\n[`sorted()`](https://docs.python.org/3/library/functions.html#sorted) function.\n\nreframe.utility.sanity.sum(*iterable*, **args*)[[source]](_modules/reframe/utility/sanity.html#sum)[¶](#reframe.utility.sanity.sum)\nReplacement for the built-in [`sum()`](https://docs.python.org/3/library/functions.html#sum) function.\n\nreframe.utility.sanity.zip(**iterables*)[[source]](_modules/reframe/utility/sanity.html#zip)[¶](#reframe.utility.sanity.zip)\nReplacement for the built-in [`zip()`](https://docs.python.org/3/library/functions.html#zip) function.\n\n##### Utility Functions[¶](#utility-functions)\n\nNew in version 3.3.\n\nThis is a collection of utility functions and classes that are used by the framework but can also be useful when writing regression tests.\nFunctions or classes marked as draft should be used with caution, since they might change or be replaced without a deprecation warning.\n\n###### General Utilities[¶](#module-reframe.utility)\n\n*class* reframe.utility.MappingView(*mapping*)[[source]](_modules/reframe/utility.html#MappingView)[¶](#reframe.utility.MappingView)\nBases: [`Mapping`](https://docs.python.org/3/library/collections.abc.html#collections.abc.Mapping)\n\nA read-only view of a mapping.\n\nSee [`collections.abc.Mapping`](https://docs.python.org/3/library/collections.abc.html#collections.abc.Mapping) for a list of supported of operations.\n\nget(*key*, *default=None*)[[source]](_modules/reframe/utility.html#MappingView.get)[¶](#reframe.utility.MappingView.get)\nReturn the value mapped to `key` or `default`, if `key` does not exist.\n\nParameters\n* **key** – The key to look up.\n* **default** – The default value to return if the key is not present.\n\nReturns The value associated to the requested key.\n\nitems()[[source]](_modules/reframe/utility.html#MappingView.items)[¶](#reframe.utility.MappingView.items)\nReturn a set-like object providing a view on the underlying mapping’s items.\n\nkeys()[[source]](_modules/reframe/utility.html#MappingView.keys)[¶](#reframe.utility.MappingView.keys)\nReturn a set-like object providing a view on the underlying mapping’s keys.\n\nvalues()[[source]](_modules/reframe/utility.html#MappingView.values)[¶](#reframe.utility.MappingView.values)\nReturn a set-like object providing a view on the underlying mapping’s values.\n\n*class* reframe.utility.OrderedSet(**args*)[[source]](_modules/reframe/utility.html#OrderedSet)[¶](#reframe.utility.OrderedSet)\nBases: [`MutableSet`](https://docs.python.org/3/library/collections.abc.html#collections.abc.MutableSet)\n\nAn ordered set.\n\nThis container behaves like a normal set but remembers the insertion order of its elements. It can also inter-operate with standard Python sets.\n\nOperations between ordered sets respect the order of the elements of the operands. For example, if `x` and `y` are both ordered sets, then `x\n| y` will be a new ordered set with the (unique) elements of `x` and\n`y` in the order they appear in `x` and `y`. The same holds for all the other set operations.\n\nadd(*elem*)[[source]](_modules/reframe/utility.html#OrderedSet.add)[¶](#reframe.utility.OrderedSet.add)\nSee same method in [`set`](https://docs.python.org/3/library/stdtypes.html#set).\n\nclear()[[source]](_modules/reframe/utility.html#OrderedSet.clear)[¶](#reframe.utility.OrderedSet.clear)\nSee same method in [`set`](https://docs.python.org/3/library/stdtypes.html#set).\n\ndifference(**others*)[[source]](_modules/reframe/utility.html#OrderedSet.difference)[¶](#reframe.utility.OrderedSet.difference)\nSee same method in [`set`](https://docs.python.org/3/library/stdtypes.html#set).\n\ndiscard(*elem*)[[source]](_modules/reframe/utility.html#OrderedSet.discard)[¶](#reframe.utility.OrderedSet.discard)\nSee same method in [`set`](https://docs.python.org/3/library/stdtypes.html#set).\n\nintersection(**others*)[[source]](_modules/reframe/utility.html#OrderedSet.intersection)[¶](#reframe.utility.OrderedSet.intersection)\nSee same method in [`set`](https://docs.python.org/3/library/stdtypes.html#set).\n\nisdisjoint(*other*)[[source]](_modules/reframe/utility.html#OrderedSet.isdisjoint)[¶](#reframe.utility.OrderedSet.isdisjoint)\nSee same method in [`set`](https://docs.python.org/3/library/stdtypes.html#set).\n\nissubset(*other*)[[source]](_modules/reframe/utility.html#OrderedSet.issubset)[¶](#reframe.utility.OrderedSet.issubset)\nSee same method in [`set`](https://docs.python.org/3/library/stdtypes.html#set).\n\nissuperset(*other*)[[source]](_modules/reframe/utility.html#OrderedSet.issuperset)[¶](#reframe.utility.OrderedSet.issuperset)\nSee same method in [`set`](https://docs.python.org/3/library/stdtypes.html#set).\n\npop()[[source]](_modules/reframe/utility.html#OrderedSet.pop)[¶](#reframe.utility.OrderedSet.pop)\nSee same method in [`set`](https://docs.python.org/3/library/stdtypes.html#set).\n\nremove(*elem*)[[source]](_modules/reframe/utility.html#OrderedSet.remove)[¶](#reframe.utility.OrderedSet.remove)\nSee same method in [`set`](https://docs.python.org/3/library/stdtypes.html#set).\n\nsymmetric_difference(*other*)[[source]](_modules/reframe/utility.html#OrderedSet.symmetric_difference)[¶](#reframe.utility.OrderedSet.symmetric_difference)\nSee same method in [`set`](https://docs.python.org/3/library/stdtypes.html#set).\n\nunion(**others*)[[source]](_modules/reframe/utility.html#OrderedSet.union)[¶](#reframe.utility.OrderedSet.union)\nSee same method in [`set`](https://docs.python.org/3/library/stdtypes.html#set).\n\n*class* reframe.utility.ScopedDict(*mapping={}*, *scope_sep=':'*, *global_scope='*'*)[[source]](_modules/reframe/utility.html#ScopedDict)[¶](#reframe.utility.ScopedDict)\nBases: [`UserDict`](https://docs.python.org/3/library/collections.html#collections.UserDict)\n\nThis is a special dictionary that imposes scopes on its keys.\n\nWhen a key is not found, it will be searched up in the scope hierarchy.\nIf not found even at the global scope, a [`KeyError`](https://docs.python.org/3/library/exceptions.html#KeyError) will be raised.\n\nA scoped dictionary is initialized using a two-level normal dictionary that defines the different scopes and the keys inside them. Scopes can be nested by concatenating them using the `:` separator by default:\n`scope:subscope`. Below is an example of a scoped dictionary that also demonstrates key lookup:\n```\nd = ScopedDict({\n 'a': {'k1': 1, 'k2': 2},\n 'a:b': {'k1': 3, 'k3': 4},\n '*': {'k1': 7, 'k3': 9, 'k4': 10}\n})\n\nassert d['a:k1'] == 1 # resolved in the scope 'a'\nassert d['a:k3'] == 9 # resolved in the global scope assert d['a:b:k1'] == 3 # resolved in the scope 'a:b'\nassert d['a:b:k2'] == 2 # resolved in the scope 'a'\nassert d['a:b:k4'] == 10 # resolved in the global scope d['a:k5'] # KeyError d['*:k2'] # KeyError\n```\nIf no scope is specified in the key lookup, the global scope is assumed.\nFor example, `d['k1']` will return `7`. The syntaxes `d[':k1']` and\n`d['*:k1']` are all equivalent.\nIf you try to retrieve a whole scope, e.g., `d['a:b']`,\n[`KeyError`](https://docs.python.org/3/library/exceptions.html#KeyError) will be raised.\nFor retrieving scopes, you should use the [`scope()`](#reframe.utility.ScopedDict.scope) function.\n\nKey deletion follows the same resolution mechanism as key retrieval,\nexcept that you are allowed to delete whole scopes. For example, `del d['*']` will delete the global scope, such that subsequent access of\n`d['a:k3']` will raise a [`KeyError`](https://docs.python.org/3/library/exceptions.html#KeyError). If a key specification matches both a key and scope, the key will be deleted and not the scope.\n\nParameters\n* **mapping** – A two-level mapping of the form\n```\n{\n scope1: {k1: v1, k2: v2},\n scope2: {k1: v1, k3: v3}\n}\n```\nBoth the scope keys and the actual dictionary keys must be strings, otherwise a [`TypeError`](https://docs.python.org/3/library/exceptions.html#TypeError) will be raised.\n* **scope_sep** – A character that separates the scopes.\n* **global_scope** – A key that represents the global scope.\n\n*property* global_scope_mark[¶](#reframe.utility.ScopedDict.global_scope_mark)\nThe key representing the global scope of this dictionary.\n\nscope(*name*)[[source]](_modules/reframe/utility.html#ScopedDict.scope)[¶](#reframe.utility.ScopedDict.scope)\nRetrieve a whole scope.\n\nParameters\n**scope** – The name of the scope to retrieve.\n\nReturns A dictionary with the keys that are within the requested scope.\n\n*property* scope_separator[¶](#reframe.utility.ScopedDict.scope_separator)\nThe scope separator of this dictionary.\n\nupdate(*other*)[[source]](_modules/reframe/utility.html#ScopedDict.update)[¶](#reframe.utility.ScopedDict.update)\nUpdate this dictionary from the values of a two-level mapping as described above.\n\nParameters\n**other** – A two-level mapping defining scopes and keys.\n\n*class* reframe.utility.SequenceView(*container*)[[source]](_modules/reframe/utility.html#SequenceView)[¶](#reframe.utility.SequenceView)\nBases: [`Sequence`](https://docs.python.org/3/library/collections.abc.html#collections.abc.Sequence)\n\nA read-only view of a sequence.\n\nSee [`collections.abc.Sequence`](https://docs.python.org/3/library/collections.abc.html#collections.abc.Sequence) for a list of supported of operations.\n\nParameters\n**container** – The container to create a view on.\n\nRaises\n[**TypeError**](https://docs.python.org/3/library/exceptions.html#TypeError) – If the container does not fulfill the\n[`collections.abc.Sequence`](https://docs.python.org/3/library/collections.abc.html#collections.abc.Sequence) interface.\n\nNote\n\nYou can concatenate a [`SequenceView`](#reframe.utility.SequenceView) with a container of the same type as the underlying container of the view, in which case a new container with the concatenated elements will be returned.\n\ncount(*value*)[[source]](_modules/reframe/utility.html#SequenceView.count)[¶](#reframe.utility.SequenceView.count)\nCount occurrences of `value` in the container.\n\nParameters\n**value** – The value to search for.\n\nReturns The number of occurrences.\n\nindex(*value*, *start=0*, *stop=None*)[[source]](_modules/reframe/utility.html#SequenceView.index)[¶](#reframe.utility.SequenceView.index)\nReturn the first index of `value`.\n\nParameters\n* **value** – The value to search for.\n* **start** – The position where the search starts.\n* **stop** – The position where the search stops. The element at this position is not looked at. If `None`, this equals to the sequence’s length.\n\nReturns The index of the first element found that equals `value`.\n\nRaises\n[**ValueError**](https://docs.python.org/3/library/exceptions.html#ValueError) – if the value is not present.\n\nreframe.utility.allx(*iterable*)[[source]](_modules/reframe/utility.html#allx)[¶](#reframe.utility.allx)\nSame as the built-in [`all()`](https://docs.python.org/3/library/functions.html#all), except that it returns\n`False` if `iterable` is empty.\n\nreframe.utility.attr_validator(*validate_fn*)[[source]](_modules/reframe/utility.html#attr_validator)[¶](#reframe.utility.attr_validator)\nValidate object attributes recursively.\n\nThis returns a function which you can call with the object to check. It will return `True` if the `validate_fn()` returns `True`\nfor all object attributes recursively. If the object to be validated is an iterable, its elements will be validated individually.\n\nParameters\n**validate_fn** – A callable that validates an object. It takes a single argument, which is the object to validate.\n\nReturns A validation function that will perform the actual validation.\nIt accepts a single argument, which is the object to validate. It returns a two-element tuple, containing the result of the validation as a boolean and a formatted string indicating the faulty attribute.\n\nNote\n\nObjects defining `__slots__` are passed directly to the\n`validate_fn` function.\n\nNew in version 3.3.\n\nreframe.utility.cache_return_value(*fn*)[[source]](_modules/reframe/utility.html#cache_return_value)[¶](#reframe.utility.cache_return_value)\nDecorator that caches the return value of the decorated function.\n\nThe function will only be called once and then the cached value will be returned each time.\n\nreframe.utility.decamelize(*s*, *delim='_'*)[[source]](_modules/reframe/utility.html#decamelize)[¶](#reframe.utility.decamelize)\nDecamelize a string.\n\nFor example, `MyBaseClass` will be converted to `my_base_class`.\nThe delimiter may be changed by setting the `delim` argument.\n\nParameters\n* **s** – A string in camel notation.\n* **delim** – The delimiter that will be used to separate words.\n\nReturns The converted string.\n\nreframe.utility.find_modules(*substr*, *environ_mapping=None*)[[source]](_modules/reframe/utility.html#find_modules)[¶](#reframe.utility.find_modules)\nReturn all modules in the current system that contain `substr` in their name.\n\nThis function is a generator and will yield tuples of partition,\nenvironment and module combinations for each partition of the current system and for each environment of a partition.\n\nThe `environ_mapping` argument allows you to map module name patterns to ReFrame environments. This is useful for flat module name schemes, in order to avoid incompatible combinations of modules and environments.\n\nYou can use this function to parametrize regression tests over the available environment modules. The following example will generate tests for all the available `netcdf` packages in the system:\n```\n@rfm.simple_test class MyTest(rfm.RegressionTest):\n module_info = parameter(find_modules('netcdf'))\n\n @run_after('init')\n def apply_module_info(self):\n s, e, m = self.module_info\n self.valid_systems = [s]\n self.valid_prog_environs = [e]\n self.modules = [m]\n ...\n```\nThe following example shows the use of `environ_mapping` with flat module name schemes. In this example, the toolchain for which the package was built is encoded in the module’s name. Using the `environ_mapping`\nargument we can map module name patterns to ReFrame environments, so that invalid combinations are pruned:\n```\nmy_find_modules = functools.partial(find_modules, environ_mapping={\n r'.*CrayGNU.*': 'PrgEnv-gnu',\n r'.*CrayIntel.*': 'PrgEnv-intel',\n r'.*CrayCCE.*': 'PrgEnv-cray'\n})\n\n@rfm.simple_test class MyTest(rfm.RegressionTest):\n module_info = parameter(my_find_modules('GROMACS'))\n\n @run_after('init')\n def apply_module_info(self):\n s, e, m = self.module_info\n self.valid_systems = [s]\n self.valid_prog_environs = [e]\n self.modules = [m]\n ...\n```\nParameters\n* **substr** – A substring that the returned module names must contain.\n* **environ_mapping** – A dictionary mapping regular expressions to environment names.\n\nReturns An iterator that iterates over tuples of the module, partition and environment name combinations that were found.\n\nreframe.utility.import_module_from_file(*filename*, *force=False*)[[source]](_modules/reframe/utility.html#import_module_from_file)[¶](#reframe.utility.import_module_from_file)\nImport module from file.\n\nParameters\n* **filename** – The path to the filename of a Python module.\n* **force** – Force reload of module in case it is already loaded.\n\nReturns The loaded Python module.\n\nreframe.utility.is_copyable(*obj*)[[source]](_modules/reframe/utility.html#is_copyable)[¶](#reframe.utility.is_copyable)\nCheck if an object can be copied with [`copy.deepcopy()`](https://docs.python.org/3/library/copy.html#copy.deepcopy), without performing the copy.\n\nThis is a superset of [`is_picklable()`](#reframe.utility.is_picklable). It returns `True` also in the following cases:\n\n* The object defines a `__copy__()` method.\n* The object defines a `__deepcopy__()` method.\n* The object is a function.\n* The object is a builtin type.\n\nNew in version 3.3.\n\nreframe.utility.is_picklable(*obj*)[[source]](_modules/reframe/utility.html#is_picklable)[¶](#reframe.utility.is_picklable)\nCheck if an object can be pickled.\n\nNew in version 3.3.\n\nreframe.utility.is_trivially_callable(*fn*, ***, *non_def_args=0*)[[source]](_modules/reframe/utility.html#is_trivially_callable)[¶](#reframe.utility.is_trivially_callable)\nCheck that a callable object is trivially callable.\n\nAn object is trivially callable when it can be invoked by providing just an expected number of non-default arguments to its call method. For example, (non-static) member functions expect a single argument without a default value, which will passed as `cls` or `self` during invocation depending on whether the function is a classmethod or not, respectively.\nOn the other hand, member functions that are static methods are not passed any values by default when invoked. Therefore, these functions can only be trivially callable when their call method expects no arguments by default.\n\nParameters\n* **fn** – A callable to be tested if its trivially callable.\n* **non_def_args** – The number of non-default arguments the callable\n`fn` expects when invoked.\n\nReturns This function returns [`True`](https://docs.python.org/3/library/constants.html#True) if the expected number of arguments matches the value of `non_def_args`. Otherwise, it returns\n[`False`](https://docs.python.org/3/library/constants.html#False).\n\nreframe.utility.longest(**iterables*)[[source]](_modules/reframe/utility.html#longest)[¶](#reframe.utility.longest)\nReturn the longest sequence.\n\nThis function raises a [`TypeError`](https://docs.python.org/3/library/exceptions.html#TypeError) if any of the iterables is not [`Sized`](https://bit.ly/3QqJlmw).\n\nParameters\n**iterables** – The iterables to check.\n\nReturns The longest iterable.\n\nreframe.utility.nodelist_abbrev(*nodes*)[[source]](_modules/reframe/utility.html#nodelist_abbrev)[¶](#reframe.utility.nodelist_abbrev)\nCreate an abbreviated string representation of the node list.\n\nFor example, the node list\n```\n['nid001', 'nid002', 'nid010', 'nid011', 'nid012', 'nid510', 'nid511']\n```\nwill be abbreviated as follows:\n```\nnid00[1-2],nid0[10-12],nid51[0-1]\n```\nNew in version 3.5.3.\n\nParameters\n**nodes** – The node list to abbreviate.\n\nReturns The abbreviated list representation.\n\nreframe.utility.nodelist_expand(*nodespec*)[[source]](_modules/reframe/utility.html#nodelist_expand)[¶](#reframe.utility.nodelist_expand)\nExpand the nodes in `nodespec` to a list of nodes.\n\nParameters\n**nodespec** – A node specification as the one returned by\n[`nodelist_abbrev()`](#reframe.utility.nodelist_abbrev)\n\nReturns The list of nodes corresponding to the given node specification.\n\nNew in version 4.0.0.\n\nreframe.utility.ppretty(*value*, *htchar=' '*, *lfchar='\\n'*, *indent=4*, *basic_offset=0*, *repr=<built-in function repr>*)[[source]](_modules/reframe/utility.html#ppretty)[¶](#reframe.utility.ppretty)\nFormat value in a pretty way.\n\nIf value is a container, this function will recursively format the container’s elements.\n\nParameters\n* **value** – The value to be formatted.\n* **htchar** – Horizontal-tab character.\n* **lfchar** – Linefeed character.\n* **indent** – Number of `htchar` characters for every indentation level.\n* **basic_offset** – Basic offset for the representation, any additional indentation space is added to the `basic_offset`.\n* **repr** – A [`repr()`](#reframe.utility.repr)-like function that will be used for printing values. This function is allowed to accept all the arguments of\n[`ppretty()`](#reframe.utility.ppretty) except the `repr` argument.\n\nReturns A formatted string of `value`.\n\nreframe.utility.repr(*obj*, *htchar=' '*, *lfchar='\\n'*, *indent=4*, *basic_offset=0*)[[source]](_modules/reframe/utility.html#repr)[¶](#reframe.utility.repr)\nA [`repr()`](https://docs.python.org/3/library/functions.html#repr) replacement function for debugging purposes printing all object attributes recursively.\n\nThis function does not follow the standard [`repr()`](https://docs.python.org/3/library/functions.html#repr) convention, but it prints each object as a set of key/value pairs along with its memory location. It also keeps track of the already visited objects, and abbreviates their representation.\n\nParameters\n**obj** – The object to be dumped.\nFor the rest of the arguments, see [`ppretty()`](#reframe.utility.ppretty).\n\nReturns The formatted object dump.\n\nreframe.utility.shortest(**iterables*)[[source]](_modules/reframe/utility.html#shortest)[¶](#reframe.utility.shortest)\nReturn the shortest sequence.\n\nThis function raises a [`TypeError`](https://docs.python.org/3/library/exceptions.html#TypeError) if any of the iterables is not [`Sized`](https://bit.ly/3QqJlmw).\n\nParameters\n**iterables** – The iterables to check.\n\nReturns The shortest iterable.\n\n*class* reframe.utility.temp_setattr(*obj*, *attr*, *val*)[[source]](_modules/reframe/utility.html#temp_setattr)[¶](#reframe.utility.temp_setattr)\nBases: [`object`](https://docs.python.org/3/library/functions.html#object)\n\nContext manager to temporarily change the attribute value of an object.\n\n###### System Utilities[¶](#module-reframe.utility.osext)\n\n*class* reframe.utility.osext.change_dir(*dir_name*)[[source]](_modules/reframe/utility/osext.html#change_dir)[¶](#reframe.utility.osext.change_dir)\nBases: [`object`](https://docs.python.org/3/library/functions.html#object)\n\nContext manager to temporarily change the current working directory.\n\nParameters\n**dir_name** – The directory to temporarily change to.\n\nreframe.utility.osext.concat_files(*dst*, **files*, *sep='\\n'*, *overwrite=False*)[[source]](_modules/reframe/utility/osext.html#concat_files)[¶](#reframe.utility.osext.concat_files)\nConcatenate `files` into `dst`.\n\nParameters\n* **dst** – The name of the output file.\n* **files** – The files to concatenate.\n* **sep** – The separator to use during concatenation.\n* **overwrite** – Overwrite the `output` file if it already exists.\n\nRaises\n* [**TypeError**](https://docs.python.org/3/library/exceptions.html#TypeError) – In case `files` it not an iterable object.\n* [**ValueError**](https://docs.python.org/3/library/exceptions.html#ValueError) – In case `output` already exists and ovewrite is\n`False`.\n\nreframe.utility.osext.copytree(*src*, *dst*, *symlinks=False*, *ignore=None*, *copy_function=<function copy2>*, *ignore_dangling_symlinks=False*, *dirs_exist_ok=False*)[[source]](_modules/reframe/utility/osext.html#copytree)[¶](#reframe.utility.osext.copytree)\nCompatibility version of [`shutil.copytree()`](https://docs.python.org/3/library/shutil.html#shutil.copytree) for Python < 3.8.\n\nThis function will automatically delegate to [`shutil.copytree()`](https://docs.python.org/3/library/shutil.html#shutil.copytree)\nfor Python versions >= 3.8.\n\nreframe.utility.osext.copytree_virtual(*src*, *dst*, *file_links=None*, *symlinks=False*, *copy_function=<function copy2>*, *ignore_dangling_symlinks=False*, *dirs_exist_ok=False*)[[source]](_modules/reframe/utility/osext.html#copytree_virtual)[¶](#reframe.utility.osext.copytree_virtual)\nCopy `src` to `dst`, but create symlinks for the files listed in\n`file_links`.\n\nIf `file_links` is empty or `None`, this is equivalent to\n[`copytree()`](#reframe.utility.osext.copytree). The rest of the arguments are passed as-is to\n[`copytree()`](#reframe.utility.osext.copytree). Paths in `file_links` must be relative to `src`.\nIf you try to pass `'.'` in `file_links`, an [`OSError`](https://docs.python.org/3/library/exceptions.html#OSError) will be raised.\n\nreframe.utility.osext.cray_cdt_version()[[source]](_modules/reframe/utility/osext.html#cray_cdt_version)[¶](#reframe.utility.osext.cray_cdt_version)\nReturn either the Cray Development Toolkit (CDT) version, the Cray Programming Environment (CPE) version or `None` if the version cannot be retrieved.\n\nreframe.utility.osext.cray_cle_info(*filename='/etc/opt/cray/release/cle-release'*)[[source]](_modules/reframe/utility/osext.html#cray_cle_info)[¶](#reframe.utility.osext.cray_cle_info)\nReturn the Cray Linux Environment (CLE) release information.\n\nParameters\n**filename** – The file that contains the CLE release information\n\nReturns A named tuple with the following attributes that correspond to the release information: `release`, `build`, `date`,\n`arch`, `network`, `patchset`.\n\nreframe.utility.osext.expandvars(*s*)[[source]](_modules/reframe/utility/osext.html#expandvars)[¶](#reframe.utility.osext.expandvars)\nExpand environment variables in `s` and perform any command substitution.\n\nThis function is the same as [`os.path.expandvars()`](https://docs.python.org/3/library/os.path.html#os.path.expandvars), except that it also recognizes the syntax of shell command substitution: `$(cmd)` or\n``cmd``.\n\nreframe.utility.osext.follow_link(*path*)[[source]](_modules/reframe/utility/osext.html#follow_link)[¶](#reframe.utility.osext.follow_link)\nReturn the final target of a symlink chain.\n\nIf `path` is not a symlink, it will be returned as is.\n\nreframe.utility.osext.force_remove_file(*filename*)[[source]](_modules/reframe/utility/osext.html#force_remove_file)[¶](#reframe.utility.osext.force_remove_file)\nRemove filename ignoring [`FileNotFoundError`](https://docs.python.org/3/library/exceptions.html#FileNotFoundError).\n\nreframe.utility.osext.git_clone(*url*, *targetdir=None*, *opts=None*, *timeout=5*)[[source]](_modules/reframe/utility/osext.html#git_clone)[¶](#reframe.utility.osext.git_clone)\nClone a git repository from a URL.\n\nParameters\n* **url** – The URL to clone from.\n* **opts** – List of options to be passed to the git clone command\n* **timeout** – Timeout in seconds when checking if the url is a valid repository.\n* **targetdir** – The directory where the repository will be cloned to. If\n`None`, a new directory will be created with the repository name as if `git clone {url}` was issued.\n\nreframe.utility.osext.git_repo_exists(*url*, *timeout=5*)[[source]](_modules/reframe/utility/osext.html#git_repo_exists)[¶](#reframe.utility.osext.git_repo_exists)\nCheck if URL refers to a valid Git repository.\n\nParameters\n* **url** – The URL to check.\n* **timeout** – Timeout in seconds.\n\nReturns\n`True` if URL is a Git repository, `False`\notherwise or if timeout is reached.\n\nreframe.utility.osext.git_repo_hash(*commit='HEAD'*, *short=True*, *wd=None*)[[source]](_modules/reframe/utility/osext.html#git_repo_hash)[¶](#reframe.utility.osext.git_repo_hash)\nReturn the SHA1 hash of a Git commit.\n\nParameters\n* **commit** – The commit to look at.\n* **short** – Return a short hash. This always corresponds to the first 8 characters of the long hash. We don’t rely on Git for the short hash,\nsince depending on the version it might return either 7 or 8 characters.\n* **wd** – Change to this directory before retrieving the hash. If `None`,\nReFrame’s install prefix will be used.\n\nReturns The Git commit hash or `None` if the hash could not be retrieved.\n\nreframe.utility.osext.inpath(*entry*, *pathvar*)[[source]](_modules/reframe/utility/osext.html#inpath)[¶](#reframe.utility.osext.inpath)\nCheck if entry is in path.\n\nParameters\n* **entry** – The entry to look for.\n* **pathvar** – A path variable in the form ‘entry1:entry2:entry3’.\n\nReturns\n`True` if the entry exists in the path variable,\n`False` otherwise.\n\nreframe.utility.osext.is_interactive()[[source]](_modules/reframe/utility/osext.html#is_interactive)[¶](#reframe.utility.osext.is_interactive)\nCheck if the current Python session is interactive.\n\nreframe.utility.osext.is_url(*s*)[[source]](_modules/reframe/utility/osext.html#is_url)[¶](#reframe.utility.osext.is_url)\nCheck if string is a URL.\n\nreframe.utility.osext.mkstemp_path(**args*, ***kwargs*)[[source]](_modules/reframe/utility/osext.html#mkstemp_path)[¶](#reframe.utility.osext.mkstemp_path)\nCreate a temporary file and return its path.\n\nThis is a wrapper to [`tempfile.mkstemp()`](https://docs.python.org/3/library/tempfile.html#tempfile.mkstemp) except that it closes the temporary file as soon as it creates it and returns the path.\n\n`args` and `kwargs` passed through to [`tempfile.mkstemp()`](https://docs.python.org/3/library/tempfile.html#tempfile.mkstemp).\n\nreframe.utility.osext.osgroup()[[source]](_modules/reframe/utility/osext.html#osgroup)[¶](#reframe.utility.osext.osgroup)\nReturn the group name of the current OS user.\n\nIf the group name cannot be retrieved, `None` will be returned.\n\nreframe.utility.osext.osuser()[[source]](_modules/reframe/utility/osext.html#osuser)[¶](#reframe.utility.osext.osuser)\nReturn the name of the current OS user.\n\nIf the user name cannot be retrieved, `None` will be returned.\n\nreframe.utility.osext.rmtree(**args*, *max_retries=3*, ***kwargs*)[[source]](_modules/reframe/utility/osext.html#rmtree)[¶](#reframe.utility.osext.rmtree)\nPersistent version of [`shutil.rmtree()`](https://docs.python.org/3/library/shutil.html#shutil.rmtree).\n\nIf [`shutil.rmtree()`](https://docs.python.org/3/library/shutil.html#shutil.rmtree) fails with `ENOTEMPTY` or `EBUSY`, ignore the error and retry up to `max_retries` times to delete the directory.\n\nThis version of [`rmtree()`](#reframe.utility.osext.rmtree) is mostly provided to work around a race condition between when `sacct` reports a job as completed and when the Slurm epilog runs. See [gh #291](https://github.com/reframe-hpc/reframe/issues/291) for more information.\nFurthermore, it offers a work around for NFS file systems where stale file handles may be present during the [`rmtree()`](#reframe.utility.osext.rmtree) call, causing it to throw a busy device/resource error. See [gh #712](https://github.com/reframe-hpc/reframe/issues/712) for more information.\n\n`args` and `kwargs` are passed through to [`shutil.rmtree()`](https://docs.python.org/3/library/shutil.html#shutil.rmtree).\n\nIf `onerror` is specified in `kwargs` and it is not `None`, this function is completely equivalent to [`shutil.rmtree()`](https://docs.python.org/3/library/shutil.html#shutil.rmtree).\n\nParameters\n* **args** – Arguments to be passed through to [`shutil.rmtree()`](https://docs.python.org/3/library/shutil.html#shutil.rmtree).\n* **max_reties** – Maximum number of retries if the target directory cannot be deleted.\n* **kwargs** – Keyword arguments to be passed through to\n[`shutil.rmtree()`](https://docs.python.org/3/library/shutil.html#shutil.rmtree).\n\nreframe.utility.osext.run_command(*cmd*, *check=False*, *timeout=None*, ***kwargs*)[[source]](_modules/reframe/utility/osext.html#run_command)[¶](#reframe.utility.osext.run_command)\nRun command synchronously.\n\nThis function will block until the command executes or the timeout is reached. It essentially calls [`run_command_async()`](#reframe.utility.osext.run_command_async) and waits for the command’s completion.\n\nParameters\n* **cmd** – The command to execute as a string or a sequence. See\n[`run_command_async()`](#reframe.utility.osext.run_command_async) for more details.\n* **check** – Raise an error if the command exits with a non-zero exit code.\n* **timeout** – Timeout in seconds.\n* **kwargs** – Keyword arguments to be passed [`run_command_async()`](#reframe.utility.osext.run_command_async).\n\nReturns A [`subprocess.CompletedProcess`](https://docs.python.org/3/library/subprocess.html#subprocess.CompletedProcess) object with information about the command’s outcome.\n\nRaises\n* [**reframe.core.exceptions.SpawnedProcessError**](index.html#reframe.core.exceptions.SpawnedProcessError) – If `check`\nis `True` and the command fails.\n* [**reframe.core.exceptions.SpawnedProcessTimeout**](index.html#reframe.core.exceptions.SpawnedProcessTimeout) – If the command times out.\n\nreframe.utility.osext.run_command_async(*cmd*, *stdout=-1*, *stderr=-1*, *shell=False*, *log=True*, ***popen_args*)[[source]](_modules/reframe/utility/osext.html#run_command_async)[¶](#reframe.utility.osext.run_command_async)\nRun command asynchronously.\n\nA wrapper to [`subprocess.Popen`](https://docs.python.org/3/library/subprocess.html#subprocess.Popen) with the following tweaks:\n\n* It always passes `universal_newlines=True` to `Popen`.\n* If `shell=False` and `cmd` is a string, it will lexically split\n`cmd` using `shlex.split(cmd)`.\n\nParameters\n* **cmd** – The command to run either as a string or a sequence of arguments.\n* **stdout** – Same as the corresponding argument of `Popen`.\nDefault is [`subprocess.PIPE`](https://docs.python.org/3/library/subprocess.html#subprocess.PIPE).\n* **stderr** – Same as the corresponding argument of `Popen`.\nDefault is [`subprocess.PIPE`](https://docs.python.org/3/library/subprocess.html#subprocess.PIPE).\n* **shell** – Same as the corresponding argument of `Popen`.\n* **log** – Log the execution of the command through ReFrame’s logging facility.\n* **popen_args** – Any additional arguments to be passed to\n`Popen`.\n\nReturns A new `Popen` object.\n\nreframe.utility.osext.samefile(*path1*, *path2*)[[source]](_modules/reframe/utility/osext.html#samefile)[¶](#reframe.utility.osext.samefile)\nCheck if paths refer to the same file.\n\nIf paths exist, this is equivalent to [`os.path.samefile()`](https://docs.python.org/3/library/os.path.html#os.path.samefile). If only one of the paths exists and is a symbolic link, it will be followed and its final target will be compared to the other path. If both paths do not exist, a simple string comparison will be performed (after the paths have been normalized).\n\nreframe.utility.osext.subdirs(*dirname*, *recurse=False*)[[source]](_modules/reframe/utility/osext.html#subdirs)[¶](#reframe.utility.osext.subdirs)\nGet the list of subdirectories of `dirname` including `dirname`.\n\nIf `recurse` is `True`, this function will retrieve all subdirectories in pre-order.\n\nParameters\n* **dirname** – The directory to start searching.\n* **recurse** – If `True`, then recursively search for subdirectories.\n\nReturns The list of subdirectories found.\n\nreframe.utility.osext.unique_abs_paths(*paths*, *prune_children=True*)[[source]](_modules/reframe/utility/osext.html#unique_abs_paths)[¶](#reframe.utility.osext.unique_abs_paths)\nGet the unique absolute paths from a given list of `paths`.\n\nParameters\n* **paths** – An iterable of paths.\n* **prune_children** – Discard paths that are children of other paths in the list.\n\nRaises\n[**TypeError**](https://docs.python.org/3/library/exceptions.html#TypeError) – In case `paths` it not an iterable object.\n\n###### Type Checking Utilities[¶](#module-reframe.utility.typecheck)\n\nDynamic recursive type checking of collections.\n\nThis module defines types for collections, such as lists, dictionaries etc.,\nthat you can use with the [`isinstance()`](https://docs.python.org/3/library/functions.html#isinstance) builtin function to recursively type check all the elements of the collection. Suppose you have a list of integers, suchs as `[1, 2, 3]`, the following checks should be true:\n```\nl = [1, 2, 3]\nassert isinstance(l, List[int]) == True assert isinstance(l, List[float]) == False\n```\nAggregate types can be combined in an arbitrary depth, so that you can type check any complex data strcture:\n```\nd = {'a': [1, 2], 'b': [3, 4]}\nassert isisntance(d, Dict) == True assert isisntance(d, Dict[str, List[int]]) == True\n```\nThis module offers the following aggregate types:\n\nList[T]\nA list with elements of type `T`.\n\nSet[T]\nA set with elements of type `T`.\n\nDict[K,V]\nA dictionary with keys of type `K` and values of type `V`.\n\nTuple[T]\nA tuple with elements of type `T`.\n\nTuple[T1,T2,...,Tn]\nA tuple with `n` elements, whose types are exactly `T1`,\n`T2`, …, `Tn` in that order.\n\nStr[patt]\nA string type whose members are all the strings matching the regular expression `patt`.\n\n####### Implementation details[¶](#implementation-details)\n\nInternally, this module leverages metaclasses and the\n`__isinstancecheck__()` method to customize the behaviour of the\n[`isinstance()`](https://docs.python.org/3/library/functions.html#isinstance) builtin.\n\nBy implementing also the `__getitem__()` accessor method, this module follows the look-and-feel of the type hints proposed in [PEP484](https://www.python.org/dev/peps/pep-0484/). This method returns a new type that is a subtype of the base container type. Using the facilities of\n[`abc.ABCMeta`](https://docs.python.org/3/library/abc.html#abc.ABCMeta), builtin types, such as [`list`](https://docs.python.org/3/library/stdtypes.html#list),\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) etc. are registered as subtypes of the base container types offered by this module. The type hierarchy of the types defined in this module is the following (example shown for `List`, but it is analogous for the rest of the types):\n```\n type\n |\n |\n |\n List\n / |\n / |\n / |\nlist List[T]\n```\nIn the above example `T` may refer to any type, so that\n`List[List[int]]` is an instance of `List`, but not an instance of `List[int]`.\n\n*class* reframe.utility.typecheck.Bool(**args*, ***kwargs*)[[source]](_modules/reframe/utility/typecheck.html#Bool)[¶](#reframe.utility.typecheck.Bool)\nBases: [`object`](https://docs.python.org/3/library/functions.html#object)\n\nA boolean type accepting implicit conversions from strings.\n\nThis type represents a boolean value but allows implicit conversions from\n[`str`](https://docs.python.org/3/library/stdtypes.html#str). More specifically, the following conversions are supported:\n\n* The strings `'yes'`, `'true'` and `'1'` are converted to `True`.\n* The strings `'no'`, `'false'` and `'0'` are converted to\n`False`.\n\nThe built-in [`bool`](https://docs.python.org/3/library/functions.html#bool) type is registered as a subclass of this type.\n\nBoolean test variables that are meant to be set properly from the command line must be declared of this type and not [`bool`](https://docs.python.org/3/library/functions.html#bool).\n\n*class* reframe.utility.typecheck.ConvertibleType(*name*, *bases*, *namespace*, ***kwargs*)[[source]](_modules/reframe/utility/typecheck.html#ConvertibleType)[¶](#reframe.utility.typecheck.ConvertibleType)\nBases: [`ABCMeta`](https://docs.python.org/3/library/abc.html#abc.ABCMeta)\n\nA type that support conversions from other types.\n\nThis is a metaclass that allows classes that use it to support arbitrary conversions from other types using a cast-like syntax without having to change their constructor:\n```\nnew_obj = convertible_type(another_type)\n```\nFor example, a class whose constructor accepts and [`int`](https://docs.python.org/3/library/functions.html#int) may need to support a cast-from-string conversion. This is particular useful if you want a custom-typed test\n`variable` to be able to be set from the command line using the [`-S`](index.html#cmdoption-S) option.\n\nIn order to support such conversions, a class must use this metaclass and define a class method, named as `__rfm_cast_<type>__`, for each of the type conversion that needs to support .\n\nThe following is an example of a class `X` that its normal constructor accepts two arguments but it also allows conversions from string:\n```\nclass X(metaclass=ConvertibleType):\n def __init__(self, x, y):\n self.data = (x, y)\n\n @classmethod\n def __rfm_cast_str__(cls, s):\n return X(*(int(x) for x in s.split(',', maxsplit=1)))\n\n assert X(2, 3).data == X('2,3').data\n```\nNew in version 3.8.0.\n\n*class* reframe.utility.typecheck.Duration(**args*, ***kwargs*)[[source]](_modules/reframe/utility/typecheck.html#Duration)[¶](#reframe.utility.typecheck.Duration)\nBases: [`object`](https://docs.python.org/3/library/functions.html#object)\n\nA float type that represents duration in seconds.\n\nThis type supports the following implicit conversions:\n\n* From integer values\n* From string values in the form of `<days>d<hours>h<minutes>m<seconds>s`\n\nNew in version 4.2.\n\n###### Test Case Dependencies Management[¶](#module-reframe.utility.udeps)\n\nManaging the test case “micro-dependencies” between two tests.\n\nThis module defines a set of basic functions that can be used with the `how`\nargument of the [`reframe.core.pipeline.RegressionTest.depends_on()`](index.html#reframe.core.pipeline.RegressionTest.depends_on)\nfunction to control how the individual dependencies between the test cases of two tests are formed.\n\nAll functions take two arguments, the source and destination vertices of an edge in the test case dependency subgraph that connects two tests. In the relation *“T0 depends on T1”*, the source are the test cases of “T0” and the destination are the test cases of “T1.” The source and destination arguments are two-element tuples containing the names of the partition and the environment of the corresponding test cases. These functions return\n`True` if there is an edge connecting the two test cases or\n`False` otherwise.\n\nA `how` function will be called by the framework multiple times when the test DAG is built. More specifically, for each test dependency relation, it will be called once for each test case combination of the two tests.\n\nThe `how` functions essentially split the test case subgraph of two dependent tests into fully connected components based on the values of their supported partitions and environments.\n\nThe [How Test Dependencies Work In ReFrame](index.html#document-dependencies) page contains more information about test dependencies and shows visually the test case subgraph connectivity that the different\n`how` functions described here achieve.\n\nNew in version 3.3.\n\nreframe.utility.udeps.by_case(*src*, *dst*)[[source]](_modules/reframe/utility/udeps.html#by_case)[¶](#reframe.utility.udeps.by_case)\nThe test cases of two dependent tests will be split by partition and by environment.\n\nTest cases from different partitions and different environments are independent.\n\nreframe.utility.udeps.by_env(*src*, *dst*)[[source]](_modules/reframe/utility/udeps.html#by_env)[¶](#reframe.utility.udeps.by_env)\nThe test cases of two dependent tests will be split by environment.\n\nTest cases from different environments are independent.\n\nreframe.utility.udeps.by_part(*src*, *dst*)[[source]](_modules/reframe/utility/udeps.html#by_part)[¶](#reframe.utility.udeps.by_part)\nThe test cases of two dependent tests will be split by partition.\n\nTest cases from different partitions are independent.\n\nreframe.utility.udeps.by_xcase(*src*, *dst*)[[source]](_modules/reframe/utility/udeps.html#by_xcase)[¶](#reframe.utility.udeps.by_xcase)\nThe test cases of two dependent tests will be split by the exclusive disjunction (XOR) of their partitions and environments.\n\nTest cases from the same environment and the same partition are independent.\n\nreframe.utility.udeps.by_xenv(*src*, *dst*)[[source]](_modules/reframe/utility/udeps.html#by_xenv)[¶](#reframe.utility.udeps.by_xenv)\nThe test cases of two dependent tests will be split by the exclusive disjunction (XOR) of their environments.\n\nTest cases from the same environment are independent.\n\nreframe.utility.udeps.by_xpart(*src*, *dst*)[[source]](_modules/reframe/utility/udeps.html#by_xpart)[¶](#reframe.utility.udeps.by_xpart)\nThe test cases of two dependent tests will be split by the exclusive disjunction (XOR) of their partitions.\n\nTest cases from the same partition are independent.\n\nreframe.utility.udeps.fully(*src*, *dst*)[[source]](_modules/reframe/utility/udeps.html#fully)[¶](#reframe.utility.udeps.fully)\nThe test cases of two dependent tests will be fully connected.\n\n##### ReFrame Errors[¶](#reframe-errors)\n\nWhen writing ReFrame tests, you don’t need to check for any exceptions raised.\nThe runtime will take care of finalizing your test and continuing execution.\n\nDealing with ReFrame errors is only useful if you are extending ReFrame’s functionality, either by modifying its core or by creating new regression test base classes for fulfilling your specific needs.\n\nWarning\n\nThis API is considered a developer’s API, so it can change from version to version without a deprecation warning.\n\n*exception* reframe.core.exceptions.AbortTaskError(**args*)[[source]](_modules/reframe/core/exceptions.html#AbortTaskError)[¶](#reframe.core.exceptions.AbortTaskError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised by the runtime inside a regression task to denote that it has been aborted due to an external reason (e.g., keyboard interrupt, fatal error in other places etc.)\n\n*exception* reframe.core.exceptions.BuildError(*stdout*, *stderr*, *prefix=None*)[[source]](_modules/reframe/core/exceptions.html#BuildError)[¶](#reframe.core.exceptions.BuildError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when a build fails.\n\n*exception* reframe.core.exceptions.BuildSystemError(**args*)[[source]](_modules/reframe/core/exceptions.html#BuildSystemError)[¶](#reframe.core.exceptions.BuildSystemError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when a build system is not configured properly.\n\n*exception* reframe.core.exceptions.CommandLineError(**args*)[[source]](_modules/reframe/core/exceptions.html#CommandLineError)[¶](#reframe.core.exceptions.CommandLineError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when an error in command-line arguments occurs.\n\n*exception* reframe.core.exceptions.ConfigError(**args*)[[source]](_modules/reframe/core/exceptions.html#ConfigError)[¶](#reframe.core.exceptions.ConfigError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when a configuration error occurs.\n\n*exception* reframe.core.exceptions.ContainerError(**args*)[[source]](_modules/reframe/core/exceptions.html#ContainerError)[¶](#reframe.core.exceptions.ContainerError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when a container platform is not configured properly.\n\n*exception* reframe.core.exceptions.DependencyError(**args*)[[source]](_modules/reframe/core/exceptions.html#DependencyError)[¶](#reframe.core.exceptions.DependencyError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when a dependency problem is encountered.\n\n*exception* reframe.core.exceptions.EnvironError(**args*)[[source]](_modules/reframe/core/exceptions.html#EnvironError)[¶](#reframe.core.exceptions.EnvironError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when an error related to an environment occurs.\n\n*exception* reframe.core.exceptions.FailureLimitError(**args*)[[source]](_modules/reframe/core/exceptions.html#FailureLimitError)[¶](#reframe.core.exceptions.FailureLimitError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when the limit of test failures has been reached.\n\n*exception* reframe.core.exceptions.ForceExitError(**args*)[[source]](_modules/reframe/core/exceptions.html#ForceExitError)[¶](#reframe.core.exceptions.ForceExitError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when ReFrame execution must be forcefully ended,\ne.g., after a SIGTERM was received.\n\n*exception* reframe.core.exceptions.JobBlockedError(*msg=None*, *jobid=None*)[[source]](_modules/reframe/core/exceptions.html#JobBlockedError)[¶](#reframe.core.exceptions.JobBlockedError)\nBases: [`JobError`](#reframe.core.exceptions.JobError)\n\nRaised by job schedulers when a job is blocked indefinitely.\n\n*exception* reframe.core.exceptions.JobError(*msg=None*, *jobid=None*)[[source]](_modules/reframe/core/exceptions.html#JobError)[¶](#reframe.core.exceptions.JobError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised for job related errors.\n\n*property* jobid[¶](#reframe.core.exceptions.JobError.jobid)\nThe job ID of the job that encountered the error.\n\n*exception* reframe.core.exceptions.JobNotStartedError(*msg=None*, *jobid=None*)[[source]](_modules/reframe/core/exceptions.html#JobNotStartedError)[¶](#reframe.core.exceptions.JobNotStartedError)\nBases: [`JobError`](#reframe.core.exceptions.JobError)\n\nRaised when trying an operation on a unstarted job.\n\n*exception* reframe.core.exceptions.JobSchedulerError(**args*)[[source]](_modules/reframe/core/exceptions.html#JobSchedulerError)[¶](#reframe.core.exceptions.JobSchedulerError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when a job scheduler encounters an error condition.\n\n*exception* reframe.core.exceptions.LoggingError(**args*)[[source]](_modules/reframe/core/exceptions.html#LoggingError)[¶](#reframe.core.exceptions.LoggingError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when an error related to logging has occurred.\n\n*exception* reframe.core.exceptions.NameConflictError(**args*)[[source]](_modules/reframe/core/exceptions.html#NameConflictError)[¶](#reframe.core.exceptions.NameConflictError)\nBases: [`RegressionTestLoadError`](#reframe.core.exceptions.RegressionTestLoadError)\n\nRaised when there is a name clash in the test suite.\n\n*exception* reframe.core.exceptions.PerformanceError(**args*)[[source]](_modules/reframe/core/exceptions.html#PerformanceError)[¶](#reframe.core.exceptions.PerformanceError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised to denote an error in performance checking, e.g., when a performance reference is not met.\n\n*exception* reframe.core.exceptions.PipelineError(**args*)[[source]](_modules/reframe/core/exceptions.html#PipelineError)[¶](#reframe.core.exceptions.PipelineError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when a condition prevents the regression test pipeline to continue and the error may not be described by another more specific exception.\n\n*exception* reframe.core.exceptions.ReframeBaseError(**args*)[[source]](_modules/reframe/core/exceptions.html#ReframeBaseError)[¶](#reframe.core.exceptions.ReframeBaseError)\nBases: [`BaseException`](https://docs.python.org/3/library/exceptions.html#BaseException)\n\nBase exception for any ReFrame error.\n\nThis exception base class offers a specialized `__str__()` method that concatenates the messages of a chain of exceptions by inspecting their\n`__cause__` field. For example, the following piece of code will print `error message 2: error message 1`:\n```\n from reframe.core.exceptions import *\n\n def foo():\n raise ReframeError('error message 1)\n\n def bar():\n try:\n foo()\n except ReframeError as e:\n raise ReframeError('error message 2') from e\n\nif __name__ == '__main__':\n try:\n bar()\n except Exception as e:\n print(e)\n```\n*exception* reframe.core.exceptions.ReframeError(**args*)[[source]](_modules/reframe/core/exceptions.html#ReframeError)[¶](#reframe.core.exceptions.ReframeError)\nBases: [`ReframeBaseError`](#reframe.core.exceptions.ReframeBaseError), [`Exception`](https://docs.python.org/3/library/exceptions.html#Exception)\n\nBase exception for soft errors.\n\nSoft errors may be treated by simply printing the exception’s message and trying to continue execution if possible.\n\n*exception* reframe.core.exceptions.ReframeFatalError(**args*)[[source]](_modules/reframe/core/exceptions.html#ReframeFatalError)[¶](#reframe.core.exceptions.ReframeFatalError)\nBases: [`ReframeBaseError`](#reframe.core.exceptions.ReframeBaseError)\n\nA fatal framework error.\n\nExecution must be aborted.\n\n*exception* reframe.core.exceptions.ReframeSyntaxError(**args*)[[source]](_modules/reframe/core/exceptions.html#ReframeSyntaxError)[¶](#reframe.core.exceptions.ReframeSyntaxError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when the syntax of regression tests is incorrect.\n\n*exception* reframe.core.exceptions.RegressionTestLoadError(**args*)[[source]](_modules/reframe/core/exceptions.html#RegressionTestLoadError)[¶](#reframe.core.exceptions.RegressionTestLoadError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when the regression test cannot be loaded.\n\n*exception* reframe.core.exceptions.SanityError(**args*)[[source]](_modules/reframe/core/exceptions.html#SanityError)[¶](#reframe.core.exceptions.SanityError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised to denote an error in sanity checking.\n\n*exception* reframe.core.exceptions.SkipTestError(**args*)[[source]](_modules/reframe/core/exceptions.html#SkipTestError)[¶](#reframe.core.exceptions.SkipTestError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when a test needs to be skipped.\n\n*exception* reframe.core.exceptions.SpawnedProcessError(*args*, *stdout*, *stderr*, *exitcode*)[[source]](_modules/reframe/core/exceptions.html#SpawnedProcessError)[¶](#reframe.core.exceptions.SpawnedProcessError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when a spawned OS command has failed.\n\n*property* command[¶](#reframe.core.exceptions.SpawnedProcessError.command)\nThe command that the spawned process tried to execute.\n\n*property* exitcode[¶](#reframe.core.exceptions.SpawnedProcessError.exitcode)\nThe exit code of the process.\n\n*property* stderr[¶](#reframe.core.exceptions.SpawnedProcessError.stderr)\nThe standard error of the process as a string.\n\n*property* stdout[¶](#reframe.core.exceptions.SpawnedProcessError.stdout)\nThe standard output of the process as a string.\n\n*exception* reframe.core.exceptions.SpawnedProcessTimeout(*args*, *stdout*, *stderr*, *timeout*)[[source]](_modules/reframe/core/exceptions.html#SpawnedProcessTimeout)[¶](#reframe.core.exceptions.SpawnedProcessTimeout)\nBases: [`SpawnedProcessError`](#reframe.core.exceptions.SpawnedProcessError)\n\nRaised when a spawned OS command has timed out.\n\n*property* timeout[¶](#reframe.core.exceptions.SpawnedProcessTimeout.timeout)\nThe timeout of the process.\n\n*exception* reframe.core.exceptions.StatisticsError(**args*)[[source]](_modules/reframe/core/exceptions.html#StatisticsError)[¶](#reframe.core.exceptions.StatisticsError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised to denote an error in dealing with statistics.\n\n*exception* reframe.core.exceptions.TaskDependencyError(**args*)[[source]](_modules/reframe/core/exceptions.html#TaskDependencyError)[¶](#reframe.core.exceptions.TaskDependencyError)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised inside a regression task by the runtime when one of its dependencies has failed.\n\n*exception* reframe.core.exceptions.TaskExit(**args*)[[source]](_modules/reframe/core/exceptions.html#TaskExit)[¶](#reframe.core.exceptions.TaskExit)\nBases: [`ReframeError`](#reframe.core.exceptions.ReframeError)\n\nRaised when a regression task must exit the pipeline prematurely.\n\nreframe.core.exceptions.is_exit_request(*exc_type*, *exc_value*, *tb*)[[source]](_modules/reframe/core/exceptions.html#is_exit_request)[¶](#reframe.core.exceptions.is_exit_request)\nCheck if the error is a request to exit.\n\nreframe.core.exceptions.is_severe(*exc_type*, *exc_value*, *tb*)[[source]](_modules/reframe/core/exceptions.html#is_severe)[¶](#reframe.core.exceptions.is_severe)\nCheck if exception is a severe one.\n\nreframe.core.exceptions.is_user_error(*exc_type*, *exc_value*, *tb*)[[source]](_modules/reframe/core/exceptions.html#is_user_error)[¶](#reframe.core.exceptions.is_user_error)\nCheck if error is a user programming error.\n\nA user error is any of [`AttributeError`](https://docs.python.org/3/library/exceptions.html#AttributeError), [`NameError`](https://docs.python.org/3/library/exceptions.html#NameError),\n[`ModuleNotFoundError`](https://docs.python.org/3/library/exceptions.html#ModuleNotFoundError), [`TypeError`](https://docs.python.org/3/library/exceptions.html#TypeError) or\n[`ValueError`](https://docs.python.org/3/library/exceptions.html#ValueError) and the exception isthrown from user context.\n\nreframe.core.exceptions.user_frame(*exc_type*, *exc_value*, *tb*)[[source]](_modules/reframe/core/exceptions.html#user_frame)[¶](#reframe.core.exceptions.user_frame)\nReturn a user frame from the exception’s traceback.\n\nAs user frame is considered the first frame that is outside from\n`reframe` module.\n\nReturns A frame object or `None` if no user frame was found.\n\nreframe.core.exceptions.what(*exc_type*, *exc_value*, *tb*)[[source]](_modules/reframe/core/exceptions.html#what)[¶](#reframe.core.exceptions.what)\nA short description of the error.\n\n### ReFrame Test Library (experimental)[¶](#reframe-test-library-experimental)\n\nThis is a collection of generic tests that you can either run out-of-the-box by specializing them for your system using the [`-S`](index.html#cmdoption-S) option or create your site-specific tests by building upon them.\n\n#### Microbenchmarks[¶](#microbenchmarks)\n\n##### OSU microbenchmarks[¶](#osu-microbenchmarks)\n\nThere are two final parameterized tests that represent the various OSU benchmarks:\n\n> * The `osu_run` test that runs the benchmarks only.\n> This assumes that the OSU microbenchmarks are installed and available.\n> * The `osu_build_run` test that builds and runs the benchmarks.\n> This test uses two fixtures in total: one to build the tests and one to fetch them.\nDepending on your setup you can select the most appropriate final test.\nThe benchmarks define various variables with a reasonable default value that affect the execution of the benchmark.\nFor collective communication benchmarks, setting the `num_tasks` is required.\nAll tests set `num_tasks_per_node` to `1` by default.\n\n###### Examples[¶](#examples)\n\nRun the run-only version of the point to point bandwidth benchmark:\n```\nreframe -n 'osu_run.*benchmark_info=mpi.pt2pt.osu_bw' -S modules=my-osu-benchmarks -S valid_systems=mysystem -S valid_prog_environs=myenv -l\n```\nBuild and run the CUDA-aware version of the allreduce benchmark.\n```\nreframe -n 'osu_build_run.*benchmark_info=mpi.collective.osu_allreduce.*build_type=cuda' -S device_buffers=cuda -S num_tasks=16 -S valid_systems=sys:part -S valid_prog_environs=myenv -l\n```\n*class* hpctestlib.microbenchmarks.mpi.osu.build_osu_benchmarks(**args*, ***kwargs*)[[source]](_modules/hpctestlib/microbenchmarks/mpi/osu.html#build_osu_benchmarks)[¶](#hpctestlib.microbenchmarks.mpi.osu.build_osu_benchmarks)\nBases: [`CompileOnlyRegressionTest`](index.html#reframe.core.pipeline.CompileOnlyRegressionTest)\n\nFixture for building the OSU benchmarks\n\nbuild_type *= <reframe.core.parameters.TestParam object>*[¶](#hpctestlib.microbenchmarks.mpi.osu.build_osu_benchmarks.build_type)\nBuild variant parameter.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nValues\n`'cpu', 'cuda', 'rocm', 'openacc'`\n\nosu_benchmarks *= <reframe.core.fixtures.TestFixture object>*[¶](#hpctestlib.microbenchmarks.mpi.osu.build_osu_benchmarks.osu_benchmarks)\nThe fixture object that retrieves the benchmarks\n\nType\n[`fetch_osu_benchmarks`](#hpctestlib.microbenchmarks.mpi.osu.fetch_osu_benchmarks)\n\nScope\n*session*\n\n*class* hpctestlib.microbenchmarks.mpi.osu.fetch_osu_benchmarks(**args*, ***kwargs*)[[source]](_modules/hpctestlib/microbenchmarks/mpi/osu.html#fetch_osu_benchmarks)[¶](#hpctestlib.microbenchmarks.mpi.osu.fetch_osu_benchmarks)\nBases: [`RunOnlyRegressionTest`](index.html#reframe.core.pipeline.RunOnlyRegressionTest)\n\nFixture for fetching the OSU benchmarks.\n\nversion *= 5.9*[¶](#hpctestlib.microbenchmarks.mpi.osu.fetch_osu_benchmarks.version)\nThe version of the benchmarks to fetch.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`'5.9'`\n\n*class* hpctestlib.microbenchmarks.mpi.osu.osu_benchmark(**args*, ***kwargs*)[[source]](_modules/hpctestlib/microbenchmarks/mpi/osu.html#osu_benchmark)[¶](#hpctestlib.microbenchmarks.mpi.osu.osu_benchmark)\nBases: [`RunOnlyRegressionTest`](index.html#reframe.core.pipeline.RunOnlyRegressionTest)\n\nOSU benchmark test base class.\n\nbenchmark_info *= <reframe.core.parameters.TestParam object>*[¶](#hpctestlib.microbenchmarks.mpi.osu.osu_benchmark.benchmark_info)\nParameter indicating the available benchmark to execute.\n\nType 2-element tuple containing the benchmark name and whether latency or bandwidth is to be measured.\n\nValues\n`mpi.collective.osu_alltoall`,\n`mpi.collective.osu_allreduce`,\n`mpi.pt2pt.osu_bw`,\n`mpi.pt2pt.osu_latency`\n\ndevice_buffers *= cpu*[¶](#hpctestlib.microbenchmarks.mpi.osu.osu_benchmark.device_buffers)\nDevice buffers.\n\nUse accelerator device buffers.\nValid values are `cpu`, `cuda`, `openacc` or `rocm`.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`'cpu'`\n\nmessage_size[¶](#hpctestlib.microbenchmarks.mpi.osu.osu_benchmark.message_size)\nMaximum message size.\n\nBoth the performance and the sanity checks will be done for this message size.\n\nThis value is set to `8` for latency benchmarks and to `4194304` for bandwidth benchmarks.\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\nnum_iters *= 1000*[¶](#hpctestlib.microbenchmarks.mpi.osu.osu_benchmark.num_iters)\nNumber of iterations.\n\nThis value is passed to the excutable through the -i option.\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\nDefault\n`1000`\n\nnum_tasks[¶](#hpctestlib.microbenchmarks.mpi.osu.osu_benchmark.num_tasks)\nNumber of tasks to use.\n\nThis variable is required.\nIt is set to `2` for point to point benchmarks, but it is undefined for collective benchmarks\n\nRequired Yes\n\nnum_warmup_iters *= 10*[¶](#hpctestlib.microbenchmarks.mpi.osu.osu_benchmark.num_warmup_iters)\nNumber of warmup iterations.\n\nThis value is passed to the excutable through the -x option.\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\nDefault\n`10`\n\n*class* hpctestlib.microbenchmarks.mpi.osu.osu_build_run(**args*, ***kwargs*)[[source]](_modules/hpctestlib/microbenchmarks/mpi/osu.html#osu_build_run)[¶](#hpctestlib.microbenchmarks.mpi.osu.osu_build_run)\nBases: [`osu_benchmark`](#hpctestlib.microbenchmarks.mpi.osu.osu_benchmark)\n\nOSU benchmark test (build and run)\n\nosu_binaries *= <reframe.core.fixtures.TestFixture object>*[¶](#hpctestlib.microbenchmarks.mpi.osu.osu_build_run.osu_binaries)\nThe fixture object that builds the OSU binaries\n\nType\n[`build_osu_benchmarks`](#hpctestlib.microbenchmarks.mpi.osu.build_osu_benchmarks)\n\nScope\n*environment*\n\n*class* hpctestlib.microbenchmarks.mpi.osu.osu_run(**args*, ***kwargs*)[[source]](_modules/hpctestlib/microbenchmarks/mpi/osu.html#osu_run)[¶](#hpctestlib.microbenchmarks.mpi.osu.osu_run)\nBases: [`osu_benchmark`](#hpctestlib.microbenchmarks.mpi.osu.osu_benchmark)\n\nRun-only OSU benchmark test\n\n##### GPU benchmarks[¶](#module-hpctestlib.microbenchmarks.gpu.gpu_burn)\n\n*class* hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_build(**args*, ***kwargs*)[[source]](_modules/hpctestlib/microbenchmarks/gpu/gpu_burn.html#gpu_burn_build)[¶](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_build)\nBases: [`CompileOnlyRegressionTest`](index.html#reframe.core.pipeline.CompileOnlyRegressionTest)\n\nFixture for building the GPU burn benchmark.\n\nSummary[¶](#id1)\n\n| Variables | Parameters | Fixtures |\n| --- | --- | --- |\n| * [`gpu_arch`](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_build.gpu_arch)\n* [`gpu_build`](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_build.gpu_build)\n | *None* | *None* |\n\ngpu_arch *= None*[¶](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_build.gpu_arch)\nSet the GPU architecture.\n\nThis variable will be passed to the compiler to generate the arch-specific code.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str) or [`None`](https://docs.python.org/3/library/constants.html#None)\n\nDefault\n`None`\n\ngpu_build *= None*[¶](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_build.gpu_build)\nSet the build option to either `'cuda'` or `'hip'`.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`'cuda'`\n\n*class* hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_check(**args*, ***kwargs*)[[source]](_modules/hpctestlib/microbenchmarks/gpu/gpu_burn.html#gpu_burn_check)[¶](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_check)\nBases: [`RunOnlyRegressionTest`](index.html#reframe.core.pipeline.RunOnlyRegressionTest)\n\nGPU burn benchmark.\n\nThis benchmark runs continuously GEMM, either single or double precision,\non a selected set of GPUs on the node where the benchmark runs.\n\nThe floating point precision of the computations, the duration of the benchmark as well as the list of GPU devices that the benchmark will run on can be controlled through test variables.\n\nThis benchmark tries to build the benchmark code through the\n[`gpu_burn_build`](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_build) fixture.\n\nThis benchmark sets the\n[`num_gpus_per_node`](index.html#reframe.core.pipeline.RegressionTest.num_gpus_per_node) test attribute, if not already set, based on the number of devices with `type\n== 'gpu'` defined in the corresponding partition configuration.\nSimilarly, this benchmark will use the `arch` device configuration attribute to set the `gpu_arch` variable, if this is not already set by the user.\n\nSummary[¶](#id2)\n\n| Variables | Parameters | Metrics | Fixtures | System features | Environment features |\n| --- | --- | --- | --- | --- | --- |\n| * [`use_dp`](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_check.use_dp)\n* [`duration`](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_check.duration)\n* [`devices`](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_check.devices)\n | *None* | * [`gpu_perf_min`](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_check.gpu_perf_min)\n* [`gpu_temp_max`](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_check.gpu_temp_max)\n | * [`gpu_burn_build`](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_build) `[E]`\n | `+gpu` | `+cuda` OR `+hip` |\n\ndevices *= []*[¶](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_check.devices)\nList of device IDs to run the benchmark on.\n\nIf empty, the benchmark will run on all the available devices.\n\nType\n`List[int]`\n\nDefault\n`[]`\n\nduration *= 10*[¶](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_check.duration)\nDuration of the benchmark in seconds.\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\nDefault\n`10`\n\ngpu_perf_min()[[source]](_modules/hpctestlib/microbenchmarks/gpu/gpu_burn.html#gpu_burn_check.gpu_perf_min)[¶](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_check.gpu_perf_min)\nLowest performance recorded among all the selected devices.\n\ngpu_temp_max()[[source]](_modules/hpctestlib/microbenchmarks/gpu/gpu_burn.html#gpu_burn_check.gpu_temp_max)[¶](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_check.gpu_temp_max)\nMaximum temperature recorded among all the selected devices.\n\nuse_dp *= True*[¶](#hpctestlib.microbenchmarks.gpu.gpu_burn.gpu_burn_check.use_dp)\nUse double-precision arithmetic when running the benchmark.\n\nType\n[`bool`](https://docs.python.org/3/library/functions.html#bool)\n\nDefault\n`True`\n\n#### Scientific Applications[¶](#module-hpctestlib.sciapps.amber.nve)\n\n*class* hpctestlib.sciapps.amber.nve.amber_nve_check(**args*, ***kwargs*)[[source]](_modules/hpctestlib/sciapps/amber/nve.html#amber_nve_check)[¶](#hpctestlib.sciapps.amber.nve.amber_nve_check)\nBases: [`RunOnlyRegressionTest`](index.html#reframe.core.pipeline.RunOnlyRegressionTest)\n\nAmber NVE test.\n\n[Amber](https://ambermd.org/) is a suite of biomolecular simulation programs. It began in the late 1970’s, and is maintained by an active development community.\n\nThis test is parametrized over the benchmark type (see\n[`benchmark_info`](#hpctestlib.sciapps.amber.nve.amber_nve_check.benchmark_info)) and the variant of the code (see `variant`).\nEach test instance executes the benchmark, validates numerically its output and extracts and reports a performance metric.\n\nassert_energy_readout()[[source]](_modules/hpctestlib/sciapps/amber/nve.html#amber_nve_check.assert_energy_readout)[¶](#hpctestlib.sciapps.amber.nve.amber_nve_check.assert_energy_readout)\nAssert that the obtained energy meets the required tolerance.\n\n*property* bench_name[¶](#hpctestlib.sciapps.amber.nve.amber_nve_check.bench_name)\nThe benchmark name.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nbenchmark_info *= <reframe.core.parameters.TestParam object>*[¶](#hpctestlib.sciapps.amber.nve.amber_nve_check.benchmark_info)\nParameter pack encoding the benchmark information.\n\nThe first element of the tuple refers to the benchmark name,\nthe second is the energy reference and the third is the tolerance threshold.\n\nType Tuple[str, float, float]\n\nValues\n```\n[\n ('Cellulose_production_NVE', -443246.0, 5.0E-05),\n ('FactorIX_production_NVE', -234188.0, 1.0E-04),\n ('JAC_production_NVE_4fs', -44810.0, 1.0E-03),\n ('JAC_production_NVE', -58138.0, 5.0E-04)\n]\n```\n*property* energy_ref[¶](#hpctestlib.sciapps.amber.nve.amber_nve_check.energy_ref)\nThe energy reference value for this benchmark.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*property* energy_tol[¶](#hpctestlib.sciapps.amber.nve.amber_nve_check.energy_tol)\nThe energy tolerance value for this benchmark.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\ninput_file[¶](#hpctestlib.sciapps.amber.nve.amber_nve_check.input_file)\nThe input file to use.\n\nThis is set to `mdin.CPU` or `mdin.GPU` depending on the test variant during initialization.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nRequired Yes\n\nnum_tasks[¶](#hpctestlib.sciapps.amber.nve.amber_nve_check.num_tasks)\nSee [`num_tasks`](index.html#reframe.core.pipeline.RegressionTest.num_tasks).\n\nThe `mpi` variant of the test requires `num_tasks > 1`.\n\nRequired Yes\n\noutput_file *= amber.out*[¶](#hpctestlib.sciapps.amber.nve.amber_nve_check.output_file)\nThe output file to pass to the Amber executable.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nRequired No\n\nDefault\n`'amber.out'`\n\nperf()[[source]](_modules/hpctestlib/sciapps/amber/nve.html#amber_nve_check.perf)[¶](#hpctestlib.sciapps.amber.nve.amber_nve_check.perf)\nThe performance of the benchmark expressed in `ns/day`.\n\n*class* hpctestlib.sciapps.gromacs.benchmarks.gromacs_check(**args*, ***kwargs*)[[source]](_modules/hpctestlib/sciapps/gromacs/benchmarks.html#gromacs_check)[¶](#hpctestlib.sciapps.gromacs.benchmarks.gromacs_check)\nBases: [`RunOnlyRegressionTest`](index.html#reframe.core.pipeline.RunOnlyRegressionTest)\n\nGROMACS benchmark test.\n\n[GROMACS](https://www.gromacs.org/) is a versatile package to perform molecular dynamics, i.e. simulate the Newtonian equations of motion for systems with hundreds to millions of particles.\n\nThe benchmarks consist on a set of different inputs files that vary in the number of atoms and can be found in the following repository, which is also versioned: <https://github.com/victorusu/GROMACS_Benchmark_Suite/>.\n\nEach test instance validates numerically its output and extracts and reports a performance metric.\n\nassert_energy_readout()[[source]](_modules/hpctestlib/sciapps/gromacs/benchmarks.html#gromacs_check.assert_energy_readout)[¶](#hpctestlib.sciapps.gromacs.benchmarks.gromacs_check.assert_energy_readout)\nAssert that the obtained energy meets the benchmark tolerances.\n\n*property* bench_name[¶](#hpctestlib.sciapps.gromacs.benchmarks.gromacs_check.bench_name)\nThe benchmark name.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nbenchmark_info *= <reframe.core.parameters.TestParam object>*[¶](#hpctestlib.sciapps.gromacs.benchmarks.gromacs_check.benchmark_info)\nParameter pack encoding the benchmark information.\n\nThe first element of the tuple refers to the benchmark name,\nthe second is the energy reference and the third is the tolerance threshold.\n\nType Tuple[str, float, float]\n\nValues\n\nbenchmark_version *= 1.0.0*[¶](#hpctestlib.sciapps.gromacs.benchmarks.gromacs_check.benchmark_version)\nThe version of the benchmark suite to use.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`'1.0.0'`\n\n*property* energy_ref[¶](#hpctestlib.sciapps.gromacs.benchmarks.gromacs_check.energy_ref)\nThe energy reference value for this benchmark.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\n*property* energy_tol[¶](#hpctestlib.sciapps.gromacs.benchmarks.gromacs_check.energy_tol)\nThe energy tolerance value for this benchmark.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nnb_impl *= <reframe.core.parameters.TestParam object>*[¶](#hpctestlib.sciapps.gromacs.benchmarks.gromacs_check.nb_impl)\nParameter encoding the implementation of the non-bonded calculations\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nValues\n`['cpu', 'gpu']`\n\n#### Data Analytics[¶](#module-hpctestlib.data_analytics.spark.spark_checks)\n\n*class* hpctestlib.data_analytics.spark.spark_checks.compute_pi_check(**args*, ***kwargs*)[[source]](_modules/hpctestlib/data_analytics/spark/spark_checks.html#compute_pi_check)[¶](#hpctestlib.data_analytics.spark.spark_checks.compute_pi_check)\nBases: [`RunOnlyRegressionTest`](index.html#reframe.core.pipeline.RunOnlyRegressionTest)\n\nTest Apache Spark by computing PI.\n\nApache Spark is a unified analytics engine for large-scale data processing. It provides high-level APIs in Java, Scala, Python and R, and an optimized engine that supports general execution graphs. It also supports a rich set of higher-level tools including Spark SQL for SQL and structured data processing, MLlib for machine learning, GraphX for graph processing, and Structured Streaming for incremental computation and stream processing (see spark.apache.org).\n\nThis test checks that Spark is functioning correctly. To do this, it is necessary to define the tolerance of acceptable deviation. The tolerance is used to check that the computations are executed correctly, by comparing the value of pi calculated to the one obtained from the math library. The default assumption is that Spark is already installed on the system under test.\n\nassert_pi_readout()[[source]](_modules/hpctestlib/data_analytics/spark/spark_checks.html#compute_pi_check.assert_pi_readout)[¶](#hpctestlib.data_analytics.spark.spark_checks.compute_pi_check.assert_pi_readout)\nAssert that the obtained pi value meets the specified tolerances.\n\nexec_cores *= 1*[¶](#hpctestlib.data_analytics.spark.spark_checks.compute_pi_check.exec_cores)\nThe number of cores per each Spark executor\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\nRequired No\n\nDefault 1\n\nexecutor_memory[¶](#hpctestlib.data_analytics.spark.spark_checks.compute_pi_check.executor_memory)\nAmount of memory to use per executor process, following the JVM memory strings convention, i.e a number with a size unit suffix\n(“k”, “m”, “g” or “t”) (e.g. 512m, 2g)\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nRequired Yes\n\nnum_workers *= 1*[¶](#hpctestlib.data_analytics.spark.spark_checks.compute_pi_check.num_workers)\nThe number of Spark workers per node\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\nRequired No\n\nDefault 1\n\nspark_local_dirs *= /tmp*[¶](#hpctestlib.data_analytics.spark.spark_checks.compute_pi_check.spark_local_dirs)\nThe local directories used by Spark\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nRequired No\n\nDefault\n‘/tmp’\n\nspark_prefix[¶](#hpctestlib.data_analytics.spark.spark_checks.compute_pi_check.spark_prefix)\nThe Spark installation prefix path\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nRequired Yes\n\ntolerance *= 0.01*[¶](#hpctestlib.data_analytics.spark.spark_checks.compute_pi_check.tolerance)\nThe absolute tolerance of the computed value of PI\n\nType\n[`float`](https://docs.python.org/3/library/functions.html#float)\n\nRequired No\n\nDefault 0.01\n\nvariant *= <reframe.core.parameters.TestParam object>*[¶](#hpctestlib.data_analytics.spark.spark_checks.compute_pi_check.variant)\nParameter encoding the variant of the test.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nValues\n`['spark', 'pyspark']`\n\n#### Python[¶](#module-hpctestlib.python.numpy.numpy_ops)\n\n*class* hpctestlib.python.numpy.numpy_ops.numpy_ops_check(**args*, ***kwargs*)[[source]](_modules/hpctestlib/python/numpy/numpy_ops.html#numpy_ops_check)[¶](#hpctestlib.python.numpy.numpy_ops.numpy_ops_check)\nBases: [`RunOnlyRegressionTest`](index.html#reframe.core.pipeline.RunOnlyRegressionTest)\n\nNumPy basic operations test.\n\n[NumPy](https://numpy.org/) is the fundamental package for scientific computing in Python.\nIt provides a multidimensional array object, various derived objects\n(such as masked arrays and matrices), and an assortment of routines for fast operations on arrays, including mathematical, logical, shape manipulation, sorting, selecting, I/O, discrete Fourier transforms,\nbasic linear algebra, basic statistical operations, random simulation and much more.\n\nThis test test performs some fundamental NumPy linear algebra operations\n(matrix product, SVD, Cholesky decomposition, eigendecomposition, and inverse matrix calculation) and users the execution time as a performance metric. The default assumption is that NumPy is already installed on the currest system.\n\ntime_cholesky()[[source]](_modules/hpctestlib/python/numpy/numpy_ops.html#numpy_ops_check.time_cholesky)[¶](#hpctestlib.python.numpy.numpy_ops.numpy_ops_check.time_cholesky)\nTime of the `cholesky` kernel in seconds.\n\ntime_dot()[[source]](_modules/hpctestlib/python/numpy/numpy_ops.html#numpy_ops_check.time_dot)[¶](#hpctestlib.python.numpy.numpy_ops.numpy_ops_check.time_dot)\nTime of the `dot` kernel in seconds.\n\ntime_eigendec()[[source]](_modules/hpctestlib/python/numpy/numpy_ops.html#numpy_ops_check.time_eigendec)[¶](#hpctestlib.python.numpy.numpy_ops.numpy_ops_check.time_eigendec)\nTime of the `eigendec` kernel in seconds.\n\ntime_inv()[[source]](_modules/hpctestlib/python/numpy/numpy_ops.html#numpy_ops_check.time_inv)[¶](#hpctestlib.python.numpy.numpy_ops.numpy_ops_check.time_inv)\nTime of the `inv` kernel in seconds.\n\ntime_svd()[[source]](_modules/hpctestlib/python/numpy/numpy_ops.html#numpy_ops_check.time_svd)[¶](#hpctestlib.python.numpy.numpy_ops.numpy_ops_check.time_svd)\nTime of the `svd` kernel in seconds.\n\n#### Interactive Computing[¶](#module-hpctestlib.interactive.jupyter.ipcmagic)\n\n*class* hpctestlib.interactive.jupyter.ipcmagic.ipcmagic_check(**args*, ***kwargs*)[[source]](_modules/hpctestlib/interactive/jupyter/ipcmagic.html#ipcmagic_check)[¶](#hpctestlib.interactive.jupyter.ipcmagic.ipcmagic_check)\nBases: [`RunOnlyRegressionTest`](index.html#reframe.core.pipeline.RunOnlyRegressionTest)\n\nTest ipcmagic via a distributed TensorFlow training with ipyparallel.\n\n[ipcmagic](https://github.com/eth-cscs/ipcluster_magic) is a Python package and collection of CLI scripts for controlling clusters for Jupyter. For more information, please have a look\n[here](https://user.cscs.ch/tools/interactive/jupyterlab/).\n\nThis test checks the ipcmagic performance.\nTo do this, a single-layer neural network is trained against a noisy linear function. The parameters of the fitted linear function are returned in the end along with the resulting loss function. The default assumption is that ipcmagic is already installed on the system under test.\n\nassert_successful_execution()[[source]](_modules/hpctestlib/interactive/jupyter/ipcmagic.html#ipcmagic_check.assert_successful_execution)[¶](#hpctestlib.interactive.jupyter.ipcmagic.ipcmagic_check.assert_successful_execution)\nChecks that the program is running on 2 different nodes (hostnames are different), that IPCMagic is configured and returns the correct end-of-program message (returns the slope parameter in the end).\n\n#### Machine Learning[¶](#module-hpctestlib.ml.tensorflow.horovod)\n\n*class* hpctestlib.ml.tensorflow.horovod.tensorflow_cnn_check(**args*, ***kwargs*)[[source]](_modules/hpctestlib/ml/tensorflow/horovod.html#tensorflow_cnn_check)[¶](#hpctestlib.ml.tensorflow.horovod.tensorflow_cnn_check)\nBases: [`RunOnlyRegressionTest`](index.html#reframe.core.pipeline.RunOnlyRegressionTest)\n\nRun a synthetic CNN benchmark with TensorFlow2 and Horovod.\n\nTensorFlow is an end-to-end open source platform for machine learning. It has a comprehensive, flexible ecosystem of tools, libraries and community resources that lets researchers push the state-of-the-art in ML and developers easily build and deploy ML powered applications. For more information, refer to <https://www.tensorflow.org/>.\n\nHorovod is a distributed deep learning training framework for TensorFlow,\nKeras, PyTorch, and Apache MXNet. The goal of Horovod is to make distributed deep learning fast and easy to use. For more information refer to <https://github.com/horovod/horovod>.\n\nThis test runs the Horovod `tensorflow2_synthentic_benchmark.py`\nexample, checks its sanity and extracts the GPU performance.\n\nbatch_size *= 32*[¶](#hpctestlib.ml.tensorflow.horovod.tensorflow_cnn_check.batch_size)\nThe size of the batch used during the learning of models.\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\nDefault\n`32`\n\nbenchmark_version *= v0.21.0*[¶](#hpctestlib.ml.tensorflow.horovod.tensorflow_cnn_check.benchmark_version)\nThe version of Horovod to use.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`'v0.21.0'`\n\nmodel *= InceptionV3*[¶](#hpctestlib.ml.tensorflow.horovod.tensorflow_cnn_check.model)\nThe name of the model to use for this benchmark.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`'InceptionV3'`\n\nnum_batches_per_iter *= 5*[¶](#hpctestlib.ml.tensorflow.horovod.tensorflow_cnn_check.num_batches_per_iter)\nThe number of batches per iteration.\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\nDefault\n`5`\n\nnum_iters *= 5*[¶](#hpctestlib.ml.tensorflow.horovod.tensorflow_cnn_check.num_iters)\nThe number of iterations.\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\nDefault\n`5`\n\nnum_warmup_batches *= 5*[¶](#hpctestlib.ml.tensorflow.horovod.tensorflow_cnn_check.num_warmup_batches)\nThe number of warmup batches\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\nDefault\n`5`\n\nthroughput_iteration()[[source]](_modules/hpctestlib/ml/tensorflow/horovod.html#tensorflow_cnn_check.throughput_iteration)[¶](#hpctestlib.ml.tensorflow.horovod.tensorflow_cnn_check.throughput_iteration)\nThe average GPU throughput per iteration in `images/s`.\n\nthroughput_total()[[source]](_modules/hpctestlib/ml/tensorflow/horovod.html#tensorflow_cnn_check.throughput_total)[¶](#hpctestlib.ml.tensorflow.horovod.tensorflow_cnn_check.throughput_total)\nThe total GPU throughput of the benchmark in `images/s`.\n\n*class* hpctestlib.ml.pytorch.horovod.pytorch_cnn_check(**args*, ***kwargs*)[[source]](_modules/hpctestlib/ml/pytorch/horovod.html#pytorch_cnn_check)[¶](#hpctestlib.ml.pytorch.horovod.pytorch_cnn_check)\nBases: [`RunOnlyRegressionTest`](index.html#reframe.core.pipeline.RunOnlyRegressionTest)\n\nRun a synthetic CNN benchmark with PyTorch and Horovod.\n\nPyTorch is a Python package that provides tensor computation like NumPy with strong GPU acceleration and deep neural networks built on a tape-based autograd system. For more information, refer to\n<https://pytorch.org/>.\n\nHorovod is a distributed deep learning training framework for TensorFlow,\nKeras, PyTorch, and Apache MXNet. The goal of Horovod is to make distributed deep learning fast and easy to use. For more information refer to <https://github.com/horovod/horovod>.\n\nThis test runs the Horovod `pytorch_synthentic_benchmark.py`\nexample, checks its sanity and extracts the GPU performance.\n\nbatch_size *= 64*[¶](#hpctestlib.ml.pytorch.horovod.pytorch_cnn_check.batch_size)\nThe size of the batch used during the learning of models.\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\nDefault\n`64`\n\nbenchmark_version *= v0.21.0*[¶](#hpctestlib.ml.pytorch.horovod.pytorch_cnn_check.benchmark_version)\nThe version of Horovod to use.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`'v0.21.0'`\n\nmodel *= inception_v3*[¶](#hpctestlib.ml.pytorch.horovod.pytorch_cnn_check.model)\nThe name of the model to use for this benchmark.\n\nType\n[`str`](https://docs.python.org/3/library/stdtypes.html#str)\n\nDefault\n`'inception_v3'`\n\nnum_batches_per_iter *= 5*[¶](#hpctestlib.ml.pytorch.horovod.pytorch_cnn_check.num_batches_per_iter)\nThe number of batches per iteration.\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\nDefault\n`5`\n\nnum_iters *= 5*[¶](#hpctestlib.ml.pytorch.horovod.pytorch_cnn_check.num_iters)\nThe number of iterations.\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\nDefault\n`5`\n\nnum_warmup_batches *= 5*[¶](#hpctestlib.ml.pytorch.horovod.pytorch_cnn_check.num_warmup_batches)\nThe number of warmup batches\n\nType\n[`int`](https://docs.python.org/3/library/functions.html#int)\n\nDefault\n`5`\n\nthroughput_iteration()[[source]](_modules/hpctestlib/ml/pytorch/horovod.html#pytorch_cnn_check.throughput_iteration)[¶](#hpctestlib.ml.pytorch.horovod.pytorch_cnn_check.throughput_iteration)\nThe average GPU throughput per iteration in `images/s`.\n\nthroughput_total()[[source]](_modules/hpctestlib/ml/pytorch/horovod.html#pytorch_cnn_check.throughput_total)[¶](#hpctestlib.ml.pytorch.horovod.pytorch_cnn_check.throughput_total)\nThe total GPU throughput of the benchmark in `images/s`."}}},{"rowIdx":59177,"cells":{"project":{"kind":"string","value":"nerodia"},"source":{"kind":"string","value":"readthedoc"},"language":{"kind":"string","value":"Python"},"content":{"kind":"string","value":"Nerodia 0.5.0 documentation\n\n[Nerodia](index.html#document-index)\n\n---\n\n---\n\nNerodia[¶](#nerodia)\n===\n\nNerodia is a Python port of the Watir ruby gem. <https://github.com/watir/watirInstallation[¶](#installation)\n---\n\n### Supported Python Versions[¶](#supported-python-versions)\n\n* Python 2.7\n* Python 3.4+\n\n### Installing[¶](#installing)\n\nIf you have [pip](https://pip.pypa.io/) on your system, you can simply install or upgrade:\n```\npip install -U nerodia\n```\nAlternately, you can download the source distribution from [PyPI](http://pypi.python.org/pypi/nerodia) (e.g. nerodia-1.0.0.tar.gz), unarchive it, and run:\n```\npython setup.py install\n```\nExamples[¶](#examples)\n---\n\n* [Navigate to a Website](#navigate-to-a-website)\n* [Perform a Google Search](#perform-a-google-search)\n* [Select a Checkbox](#select-a-checkbox)\n* [Elements in Frames](#elements-in-frames)\n\n### Navigate to a Website[¶](#navigate-to-a-website)\n```\nfrom nerodia.browser import Browser\n\nbrowser = Browser(browser='firefox')\nbrowser.goto('watir.com')\n\nbrowser.close()\n```\n### Perform a Google Search[¶](#perform-a-google-search)\n```\nfrom nerodia.browser import Browser\n\nbrowser = Browser(browser='firefox')\nbrowser.goto('google.com')\n\nsearch_input = browser.text_field(title='Search')\nsearch_input.value = 'nerodia'\nbrowser.button(value='Google Search').click()\n\nbrowser.close()\n```\n### Select a Checkbox[¶](#select-a-checkbox)\n```\nfrom nerodia.browser import Browser\n\nbrowser = Browser(browser='firefox')\nbrowser.goto('the-internet.herokuapp.com/checkboxes')\n\ncheckbox1 = browser.checkbox()\ncheckbox1.set()\n\nbrowser.close()\n```\n### Elements in Frames[¶](#elements-in-frames)\n```\nfrom nerodia.browser import Browser\n\nbrowser = Browser(browser='firefox')\nbrowser.goto('the-internet.herokuapp.com/iframe')\n\nprint(browser.iframe().p().text)\nprint(browser.link(css='#page-footer a').text)\n\nbrowser.close()\n```\nResult:\n```\n> Your content goes here.\n> Elemental Selenium\n```\nDifferences from Watir[¶](#differences-from-watir)\n---\n\nThe goal of this project is to be as close to Watir as possible. In terms of functionality, it is equivalent; however, there are some syntax differences due to the nature of Python.\n\n### Containers[¶](#containers)\n\nThe following containers cannot be used because either the singular or plural version is reserved by Python.\n\n> | Watir | Nerodia |\n> | --- | --- |\n> | a | link |\n> | as | links |\n> | del | delete |\n> | dels | deletes |\n> | i | ital |\n> | is | itals |\n### Locators[¶](#locators)\n\nThe following locators cannot be used because they are reserved by Python.\n\n> | Watir | Nerodia |\n> | --- | --- |\n> | class | class_name |\n> | for | N/A* |\n> **This locator is only possible via the below options.*\nAlternatively, if you are only using one locator you can pass them as individual arguments:\n```\nbrowser.div('class', 'spam')\n```\nA third option is to use a dictionary and unpack into the container:\n```\nlocator = {'class': 'spam', 'index': 1}\nbrowser.div(**locator)\n```\n### Blocks[¶](#blocks)\n\nSince Python does not have blocks, alternate methods are required.\n\n#### Context[¶](#context)\n\nFor cases where we want to perform some actions inside of a different browser context without completely switching to that context, we use the context manager.\n\nConsider the following Window switching Watir code:\n```\nbrowser.window(title: 'Spam and Ham!').use do\n browser.button(id: 'close').click end\n```\nIn Nerodia, the equivalent would be:\n```\nwith browser.window(title='Spam and Ham!'):\n browser.button(id='close').click()\n```\nThe same would go for frames.\n\n#### Waits[¶](#waits)\n\nFor waits, we need to use `lambdas` or `closures`.\n\nConsider the following wait Watir code:\n```\nbtn = browser.button(id: 'btn')\nbtn.wait_until(timeout: 2, interval: 0.5) { btn.enabled }\nbtn.click\n```\nIn Nerodia, the equivalent would be:\n```\nbtn = browser.button(id='btn')\nbtn.wait_until(timeout=2, interval=0.5, method=lambda e: e.enabled)\nbtn.click()\n```\nAlso, `while` is reserved in Python. Therefore, the Nerodia equivalent of Watir’s `Wait.while` is `Wait.until_not`\n\nNerodia API Documentation\n\nnerodia[¶](#nerodia)\n---\n\n### nerodia package[¶](#nerodia-package)\n\n#### Subpackages[¶](#subpackages)\n\n##### nerodia.elements package[¶](#nerodia-elements-package)\n\n###### Submodules[¶](#submodules)\n\n###### nerodia.elements.area module[¶](#nerodia-elements-area-module)\n\n###### nerodia.elements.button module[¶](#nerodia-elements-button-module)\n\n###### nerodia.elements.cell module[¶](#nerodia-elements-cell-module)\n\n###### nerodia.elements.check_box module[¶](#nerodia-elements-check-box-module)\n\n###### nerodia.elements.d_list module[¶](#nerodia-elements-d-list-module)\n\n###### nerodia.elements.element module[¶](#nerodia-elements-element-module)\n\n###### nerodia.elements.file_field module[¶](#nerodia-elements-file-field-module)\n\n###### nerodia.elements.font module[¶](#nerodia-elements-font-module)\n\n###### nerodia.elements.form module[¶](#nerodia-elements-form-module)\n\n###### nerodia.elements.hidden module[¶](#nerodia-elements-hidden-module)\n\n###### nerodia.elements.html_elements module[¶](#nerodia-elements-html-elements-module)\n\n###### nerodia.elements.i_frame module[¶](#nerodia-elements-i-frame-module)\n\n###### nerodia.elements.image module[¶](#nerodia-elements-image-module)\n\n###### nerodia.elements.input module[¶](#nerodia-elements-input-module)\n\n###### nerodia.elements.link module[¶](#nerodia-elements-link-module)\n\n###### nerodia.elements.list module[¶](#nerodia-elements-list-module)\n\n###### nerodia.elements.option module[¶](#nerodia-elements-option-module)\n\n###### nerodia.elements.radio module[¶](#nerodia-elements-radio-module)\n\n###### nerodia.elements.radio_set module[¶](#nerodia-elements-radio-set-module)\n\n###### nerodia.elements.row module[¶](#nerodia-elements-row-module)\n\n###### nerodia.elements.select module[¶](#nerodia-elements-select-module)\n\n###### nerodia.elements.svg_elements module[¶](#nerodia-elements-svg-elements-module)\n\n###### nerodia.elements.table module[¶](#nerodia-elements-table-module)\n\n###### nerodia.elements.table_cell module[¶](#nerodia-elements-table-cell-module)\n\n###### nerodia.elements.table_data_cell module[¶](#nerodia-elements-table-data-cell-module)\n\n###### nerodia.elements.table_row module[¶](#nerodia-elements-table-row-module)\n\n###### nerodia.elements.table_section module[¶](#nerodia-elements-table-section-module)\n\n###### nerodia.elements.text_area module[¶](#nerodia-elements-text-area-module)\n\n###### nerodia.elements.text_field module[¶](#nerodia-elements-text-field-module)\n\n###### Module contents[¶](#module-contents)\n\n##### nerodia.wait package[¶](#nerodia-wait-package)\n\n###### Submodules[¶](#submodules)\n\n###### nerodia.wait.timer module[¶](#nerodia-wait-timer-module)\n\n###### nerodia.wait.wait module[¶](#nerodia-wait-wait-module)\n\n###### Module contents[¶](#module-contents)\n\n#### Submodules[¶](#submodules)\n\n#### nerodia.after_hooks module[¶](#nerodia-after-hooks-module)\n\n#### nerodia.alert module[¶](#nerodia-alert-module)\n\n#### nerodia.browser module[¶](#nerodia-browser-module)\n\nSee also\n\nSee `nerodia.container.Container` for all possible element containers.\n\nSee also\n\nSee `nerodia.has_window.HasWindow` for accessing windows.\n\nSee also\n\nSee `nerodia.wait.wait.Waitable` for wait methods.\n\n#### nerodia.container module[¶](#nerodia-container-module)\n\n#### nerodia.cookies module[¶](#nerodia-cookies-module)\n\n#### nerodia.element_collection module[¶](#nerodia-element-collection-module)\n\n#### nerodia.has_window module[¶](#nerodia-has-window-module)\n\n#### nerodia.logger module[¶](#nerodia-logger-module)\n\n#### nerodia.screenshot module[¶](#nerodia-screenshot-module)\n\n#### nerodia.window module[¶](#nerodia-window-module)"}}},{"rowIdx":59178,"cells":{"project":{"kind":"string","value":"github.com/gomodule/redigo/redis"},"source":{"kind":"string","value":"go"},"language":{"kind":"string","value":"Go"},"content":{"kind":"string","value":"None\n\nDocumentation\n [¶](#section-documentation)\n---\n\n \n### Overview [¶](#pkg-overview)\n\n* [Connections](#hdr-Connections)\n* [Executing Commands](#hdr-Executing_Commands)\n* [Pipelining](#hdr-Pipelining)\n* [Concurrency](#hdr-Concurrency)\n* [Publish and Subscribe](#hdr-Publish_and_Subscribe)\n* [Reply Helpers](#hdr-Reply_Helpers)\n* [Errors](#hdr-Errors)\n\nPackage redis is a client for the Redis database.\n\nThe Redigo FAQ (<https://github.com/gomodule/redigo/wiki/FAQ>) contains more documentation about this package.\n\n#### Connections [¶](#hdr-Connections)\n\nThe Conn interface is the primary interface for working with Redis.\nApplications create connections by calling the Dial, DialWithTimeout or NewConn functions. In the future, functions will be added for creating sharded and other types of connections.\n\nThe application must call the connection Close method when the application is done with the connection.\n\n#### Executing Commands [¶](#hdr-Executing_Commands)\n\nThe Conn interface has a generic method for executing Redis commands:\n```\nDo(commandName string, args ...interface{}) (reply interface{}, err error)\n```\nThe Redis command reference (<http://redis.io/commands>) lists the available commands. An example of using the Redis APPEND command is:\n```\nn, err := conn.Do(\"APPEND\", \"key\", \"value\")\n```\nThe Do method converts command arguments to bulk strings for transmission to the server as follows:\n```\nGo Type Conversion\n[]byte Sent as is string Sent as is int, int64 strconv.FormatInt(v)\nfloat64 strconv.FormatFloat(v, 'g', -1, 64)\nbool true -> \"1\", false -> \"0\"\nnil \"\"\nall other types fmt.Fprint(w, v)\n```\nRedis command reply types are represented using the following Go types:\n```\nRedis type Go type error redis.Error integer int64 simple string string bulk string []byte or nil if value not present.\narray []interface{} or nil if value not present.\n```\nUse type assertions or the reply helper functions to convert from interface{} to the specific Go type for the command result.\n\n#### Pipelining [¶](#hdr-Pipelining)\n\nConnections support pipelining using the Send, Flush and Receive methods.\n```\nSend(commandName string, args ...interface{}) error Flush() error Receive() (reply interface{}, err error)\n```\nSend writes the command to the connection's output buffer. Flush flushes the connection's output buffer to the server. Receive reads a single reply from the server. The following example shows a simple pipeline.\n```\nc.Send(\"SET\", \"foo\", \"bar\")\nc.Send(\"GET\", \"foo\")\nc.Flush()\nc.Receive() // reply from SET v, err = c.Receive() // reply from GET\n```\nThe Do method combines the functionality of the Send, Flush and Receive methods. The Do method starts by writing the command and flushing the output buffer. Next, the Do method receives all pending replies including the reply for the command just sent by Do. If any of the received replies is an error,\nthen Do returns the error. If there are no errors, then Do returns the last reply. If the command argument to the Do method is \"\", then the Do method will flush the output buffer and receive pending replies without sending a command.\n\nUse the Send and Do methods to implement pipelined transactions.\n```\nc.Send(\"MULTI\")\nc.Send(\"INCR\", \"foo\")\nc.Send(\"INCR\", \"bar\")\nr, err := c.Do(\"EXEC\")\nfmt.Println(r) // prints [1, 1]\n```\n#### Concurrency [¶](#hdr-Concurrency)\n\nConnections support one concurrent caller to the Receive method and one concurrent caller to the Send and Flush methods. No other concurrency is supported including concurrent calls to the Do and Close methods.\n\nFor full concurrent access to Redis, use the thread-safe Pool to get, use and release a connection from within a goroutine. Connections returned from a Pool have the concurrency restrictions described in the previous paragraph.\n\n#### Publish and Subscribe [¶](#hdr-Publish_and_Subscribe)\n\nUse the Send, Flush and Receive methods to implement Pub/Sub subscribers.\n```\nc.Send(\"SUBSCRIBE\", \"example\")\nc.Flush()\nfor {\n reply, err := c.Receive()\n if err != nil {\n return err\n }\n // process pushed message\n}\n```\nThe PubSubConn type wraps a Conn with convenience methods for implementing subscribers. The Subscribe, PSubscribe, Unsubscribe and PUnsubscribe methods send and flush a subscription management command. The receive method converts a pushed message to convenient types for use in a type switch.\n```\npsc := redis.PubSubConn{Conn: c}\npsc.Subscribe(\"example\")\nfor {\n switch v := psc.Receive().(type) {\n case redis.Message:\n fmt.Printf(\"%s: message: %s\\n\", v.Channel, v.Data)\n case redis.Subscription:\n fmt.Printf(\"%s: %s %d\\n\", v.Channel, v.Kind, v.Count)\n case error:\n return v\n }\n}\n```\n#### Reply Helpers [¶](#hdr-Reply_Helpers)\n\nThe Bool, Int, Bytes, String, Strings and Values functions convert a reply to a value of a specific type. To allow convenient wrapping of calls to the connection Do and Receive methods, the functions take a second argument of type error. If the error is non-nil, then the helper function returns the error. If the error is nil, the function converts the reply to the specified type:\n```\nexists, err := redis.Bool(c.Do(\"EXISTS\", \"foo\"))\nif err != nil {\n // handle error return from c.Do or type conversion error.\n}\n```\nThe Scan function converts elements of a array reply to Go types:\n```\nvar value1 int var value2 string reply, err := redis.Values(c.Do(\"MGET\", \"key1\", \"key2\"))\nif err != nil {\n // handle error\n}\n if _, err := redis.Scan(reply, &value1, &value2); err != nil {\n // handle error\n}\n```\n#### Errors [¶](#hdr-Errors)\n\nConnection methods return error replies from the server as type redis.Error.\n\nCall the connection Err() method to determine if the connection encountered non-recoverable error such as a network error or protocol parsing error. If Err() returns a non-nil value, then the connection is not usable and should be closed.\n\nExample (Zpop) [¶](#example-package-Zpop)\n\nThis example implements ZPOP as described at\n<http://redis.io/topics/transactions> using WATCH/MULTI/EXEC and scripting.\n```\npackage main\n\nimport (\n\t\"fmt\"\n\n\t\"github.com/gomodule/redigo/redis\"\n)\n\n// zpop pops a value from the ZSET key using WATCH/MULTI/EXEC commands.\nfunc zpop(c redis.Conn, key string) (result string, err error) {\n\n\tdefer func() {\n\t\t// Return connection to normal state on error.\n\t\tif err != nil {\n\t\t\tc.Do(\"DISCARD\") // nolint: errcheck\n\t\t}\n\t}()\n\n\t// Loop until transaction is successful.\n\tfor {\n\t\tif _, err := c.Do(\"WATCH\", key); err != nil {\n\t\t\treturn \"\", err\n\t\t}\n\n\t\tmembers, err := redis.Strings(c.Do(\"ZRANGE\", key, 0, 0))\n\t\tif err != nil {\n\t\t\treturn \"\", err\n\t\t}\n\t\tif len(members) != 1 {\n\t\t\treturn \"\", redis.ErrNil\n\t\t}\n\n\t\tif err = c.Send(\"MULTI\"); err != nil {\n\t\t\treturn \"\", err\n\t\t}\n\t\tif err = c.Send(\"ZREM\", key, members[0]); err != nil {\n\t\t\treturn \"\", err\n\t\t}\n\t\tqueued, err := c.Do(\"EXEC\")\n\t\tif err != nil {\n\t\t\treturn \"\", err\n\t\t}\n\n\t\tif queued != nil {\n\t\t\tresult = members[0]\n\t\t\tbreak\n\t\t}\n\t}\n\n\treturn result, nil\n}\n\n// zpopScript pops a value from a ZSET.\nvar zpopScript = redis.NewScript(1, `\n local r = redis.call('ZRANGE', KEYS[1], 0, 0)\n if r ~= nil then\n r = r[1]\n redis.call('ZREM', KEYS[1], r)\n end\n return r\n`)\n\n// This example implements ZPOP as described at\n// http://redis.io/topics/transactions using WATCH/MULTI/EXEC and scripting.\nfunc main() {\n\tc, err := dial()\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tdefer c.Close()\n\n\t// Add test data using a pipeline.\n\n\tfor i, member := range []string{\"red\", \"blue\", \"green\"} {\n\t\tif err = c.Send(\"ZADD\", \"zset\", i, member); err != nil {\n\t\t\tfmt.Println(err)\n\t\t\treturn\n\t\t}\n\t}\n\n\tif _, err := c.Do(\"\"); err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\n\t// Pop using WATCH/MULTI/EXEC\n\n\tv, err := zpop(c, \"zset\")\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tfmt.Println(v)\n\n\t// Pop using a script.\n\n\tv, err = redis.String(zpopScript.Do(c, \"zset\"))\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tfmt.Println(v)\n\n}\n```\n```\nOutput:\n\nred blue\n```\nShare Format\nRun\n\n### Index [¶](#pkg-index)\n\n* [Variables](#pkg-variables)\n* [func Bool(reply interface{}, err error) (bool, error)](#Bool)\n* [func ByteSlices(reply interface{}, err error) ([][]byte, error)](#ByteSlices)\n* [func Bytes(reply interface{}, err error) ([]byte, error)](#Bytes)\n* [func DoContext(c Conn, ctx context.Context, cmd string, args ...interface{}) (interface{}, error)](#DoContext)\n* [func DoWithTimeout(c Conn, timeout time.Duration, cmd string, args ...interface{}) (interface{}, error)](#DoWithTimeout)\n* [func Float64(reply interface{}, err error) (float64, error)](#Float64)\n* [func Float64Map(result interface{}, err error) (map[string]float64, error)](#Float64Map)\n* [func Float64s(reply interface{}, err error) ([]float64, error)](#Float64s)\n* [func Int(reply interface{}, err error) (int, error)](#Int)\n* [func Int64(reply interface{}, err error) (int64, error)](#Int64)\n* [func Int64Map(result interface{}, err error) (map[string]int64, error)](#Int64Map)\n* [func Int64s(reply interface{}, err error) ([]int64, error)](#Int64s)\n* [func IntMap(result interface{}, err error) (map[string]int, error)](#IntMap)\n* [func Ints(reply interface{}, err error) ([]int, error)](#Ints)\n* [func MultiBulk(reply interface{}, err error) ([]interface{}, error)](#MultiBulk)deprecated\n* [func Positions(result interface{}, err error) ([]*[2]float64, error)](#Positions)\n* [func ReceiveContext(c Conn, ctx context.Context) (interface{}, error)](#ReceiveContext)\n* [func ReceiveWithTimeout(c Conn, timeout time.Duration) (interface{}, error)](#ReceiveWithTimeout)\n* [func Scan(src []interface{}, dest ...interface{}) ([]interface{}, error)](#Scan)\n* [func ScanSlice(src []interface{}, dest interface{}, fieldNames ...string) error](#ScanSlice)\n* [func ScanStruct(src []interface{}, dest interface{}) error](#ScanStruct)\n* [func String(reply interface{}, err error) (string, error)](#String)\n* [func StringMap(reply interface{}, err error) (map[string]string, error)](#StringMap)\n* [func Strings(reply interface{}, err error) ([]string, error)](#Strings)\n* [func Uint64(reply interface{}, err error) (uint64, error)](#Uint64)\n* [func Uint64Map(result interface{}, err error) (map[string]uint64, error)](#Uint64Map)\n* [func Uint64s(reply interface{}, err error) ([]uint64, error)](#Uint64s)\n* [func Values(reply interface{}, err error) ([]interface{}, error)](#Values)\n* [type Args](#Args)\n* + [func (args Args) Add(value ...interface{}) Args](#Args.Add)\n\t+ [func (args Args) AddFlat(v interface{}) Args](#Args.AddFlat)\n* [type Argument](#Argument)\n* [type Conn](#Conn)\n* + [func Dial(network, address string, options ...DialOption) (Conn, error)](#Dial)\n\t+ [func DialContext(ctx context.Context, network, address string, options ...DialOption) (Conn, error)](#DialContext)\n\t+ [func DialTimeout(network, address string, ...) (Conn, error)](#DialTimeout)deprecated\n\t+ [func DialURL(rawurl string, options ...DialOption) (Conn, error)](#DialURL)\n\t+ [func DialURLContext(ctx context.Context, rawurl string, options ...DialOption) (Conn, error)](#DialURLContext)\n\t+ [func NewConn(netConn net.Conn, readTimeout, writeTimeout time.Duration) Conn](#NewConn)\n\t+ [func NewLoggingConn(conn Conn, logger *log.Logger, prefix string) Conn](#NewLoggingConn)\n\t+ [func NewLoggingConnFilter(conn Conn, logger *log.Logger, prefix string, skip func(cmdName string) bool) Conn](#NewLoggingConnFilter)\n* [type ConnWithContext](#ConnWithContext)\n* [type ConnWithTimeout](#ConnWithTimeout)\n* [type DialOption](#DialOption)\n* + [func DialClientName(name string) DialOption](#DialClientName)\n\t+ [func DialConnectTimeout(d time.Duration) DialOption](#DialConnectTimeout)\n\t+ [func DialContextFunc(f func(ctx context.Context, network, addr string) (net.Conn, error)) DialOption](#DialContextFunc)\n\t+ [func DialDatabase(db int) DialOption](#DialDatabase)\n\t+ [func DialKeepAlive(d time.Duration) DialOption](#DialKeepAlive)\n\t+ [func DialNetDial(dial func(network, addr string) (net.Conn, error)) DialOption](#DialNetDial)\n\t+ [func DialPassword(password string) DialOption](#DialPassword)\n\t+ [func DialReadTimeout(d time.Duration) DialOption](#DialReadTimeout)\n\t+ [func DialTLSConfig(c *tls.Config) DialOption](#DialTLSConfig)\n\t+ [func DialTLSHandshakeTimeout(d time.Duration) DialOption](#DialTLSHandshakeTimeout)\n\t+ [func DialTLSSkipVerify(skip bool) DialOption](#DialTLSSkipVerify)\n\t+ [func DialUseTLS(useTLS bool) DialOption](#DialUseTLS)\n\t+ [func DialUsername(username string) DialOption](#DialUsername)\n\t+ [func DialWriteTimeout(d time.Duration) DialOption](#DialWriteTimeout)\n* [type Error](#Error)\n* + [func (err Error) Error() string](#Error.Error)\n* [type Latency](#Latency)\n* + [func Latencies(result interface{}, err error) ([]Latency, error)](#Latencies)\n* [type LatencyHistory](#LatencyHistory)\n* + [func LatencyHistories(result interface{}, err error) ([]LatencyHistory, error)](#LatencyHistories)\n* [type Message](#Message)\n* [type Pong](#Pong)\n* [type Pool](#Pool)\n* + [func NewPool(newFn func() (Conn, error), maxIdle int) *Pool](#NewPool)deprecated\n* + [func (p *Pool) ActiveCount() int](#Pool.ActiveCount)\n\t+ [func (p *Pool) Close() error](#Pool.Close)\n\t+ [func (p *Pool) Get() Conn](#Pool.Get)\n\t+ [func (p *Pool) GetContext(ctx context.Context) (Conn, error)](#Pool.GetContext)\n\t+ [func (p *Pool) IdleCount() int](#Pool.IdleCount)\n\t+ [func (p *Pool) Stats() PoolStats](#Pool.Stats)\n* [type PoolStats](#PoolStats)\n* [type PubSubConn](#PubSubConn)\n* + [func (c PubSubConn) Close() error](#PubSubConn.Close)\n\t+ [func (c PubSubConn) PSubscribe(channel ...interface{}) error](#PubSubConn.PSubscribe)\n\t+ [func (c PubSubConn) PUnsubscribe(channel ...interface{}) error](#PubSubConn.PUnsubscribe)\n\t+ [func (c PubSubConn) Ping(data string) error](#PubSubConn.Ping)\n\t+ [func (c PubSubConn) Receive() interface{}](#PubSubConn.Receive)\n\t+ [func (c PubSubConn) ReceiveContext(ctx context.Context) interface{}](#PubSubConn.ReceiveContext)\n\t+ [func (c PubSubConn) ReceiveWithTimeout(timeout time.Duration) interface{}](#PubSubConn.ReceiveWithTimeout)\n\t+ [func (c PubSubConn) Subscribe(channel ...interface{}) error](#PubSubConn.Subscribe)\n\t+ [func (c PubSubConn) Unsubscribe(channel ...interface{}) error](#PubSubConn.Unsubscribe)\n* [type Scanner](#Scanner)\n* [type Script](#Script)\n* + [func NewScript(keyCount int, src string) *Script](#NewScript)\n* + [func (s *Script) Do(c Conn, keysAndArgs ...interface{}) (interface{}, error)](#Script.Do)\n\t+ [func (s *Script) DoContext(ctx context.Context, c Conn, keysAndArgs ...interface{}) (interface{}, error)](#Script.DoContext)\n\t+ [func (s *Script) Hash() string](#Script.Hash)\n\t+ [func (s *Script) Load(c Conn) error](#Script.Load)\n\t+ [func (s *Script) Send(c Conn, keysAndArgs ...interface{}) error](#Script.Send)\n\t+ [func (s *Script) SendHash(c Conn, keysAndArgs ...interface{}) error](#Script.SendHash)\n* [type SlowLog](#SlowLog)\n* + [func SlowLogs(result interface{}, err error) ([]SlowLog, error)](#SlowLogs)\n* [type Subscription](#Subscription)\n\n#### Examples [¶](#pkg-examples)\n\n* [Package (Zpop)](#example-package-Zpop)\n* [Args](#example-Args)\n* [Bool](#example-Bool)\n* [Dial](#example-Dial)\n* [Dial (Acl)](#example-Dial-Acl)\n* [DialContext](#example-DialContext)\n* [DialURL](#example-DialURL)\n* [Int](#example-Int)\n* [Ints](#example-Ints)\n* [PubSubConn](#example-PubSubConn)\n* [Scan](#example-Scan)\n* [ScanSlice](#example-ScanSlice)\n* [Script](#example-Script)\n* [String](#example-String)\n\n### Constants [¶](#pkg-constants)\n\nThis section is empty.\n\n### Variables [¶](#pkg-variables)\n\n```\nvar ErrNil = [errors](/errors).[New](/errors#New)(\"redigo: nil returned\")\n```\nErrNil indicates that a reply value is nil.\n\n```\nvar ErrPoolExhausted = [errors](/errors).[New](/errors#New)(\"redigo: connection pool exhausted\")\n```\nErrPoolExhausted is returned from a pool connection method (Do, Send,\nReceive, Flush, Err) when the maximum number of database connections in the pool has been reached.\n\n### Functions [¶](#pkg-functions)\n\n#### \nfunc [Bool](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L204) [¶](#Bool)\n```\nfunc Bool(reply interface{}, err [error](/builtin#error)) ([bool](/builtin#bool), [error](/builtin#error))\n```\nBool is a helper that converts a command reply to a boolean. If err is not equal to nil, then Bool returns false, err. Otherwise Bool converts the reply to boolean as follows:\n```\nReply type Result integer value != 0, nil bulk string strconv.ParseBool(reply)\nnil false, ErrNil other false, error\n```\nExample [¶](#example-Bool)\n```\npackage main\n\nimport (\n\t\"fmt\"\n\n\t\"github.com/gomodule/redigo/redis\"\n)\n\n// dial wraps DialDefaultServer() with a more suitable function name for examples.\nfunc dial() (redis.Conn, error) {\n\treturn redis.DialDefaultServer()\n}\n\nfunc main() {\n\tc, err := dial()\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tdefer c.Close()\n\n\tif _, err = c.Do(\"SET\", \"foo\", 1); err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\texists, err := redis.Bool(c.Do(\"EXISTS\", \"foo\"))\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\n\tfmt.Printf(\"%#v\\n\", exists)\n}\n```\n```\nOutput:\n\ntrue\n```\nShare Format\nRun\n\n#### \nfunc [ByteSlices](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L321) [¶](#ByteSlices)\n```\nfunc ByteSlices(reply interface{}, err [error](/builtin#error)) ([][][byte](/builtin#byte), [error](/builtin#error))\n```\nByteSlices is a helper that converts an array command reply to a [][]byte.\nIf err is not equal to nil, then ByteSlices returns nil, err. Nil array items are stay nil. ByteSlices returns an error if an array item is not a bulk string or nil.\n\n#### \nfunc [Bytes](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L178) [¶](#Bytes)\n```\nfunc Bytes(reply interface{}, err [error](/builtin#error)) ([][byte](/builtin#byte), [error](/builtin#error))\n```\nBytes is a helper that converts a command reply to a slice of bytes. If err is not equal to nil, then Bytes returns nil, err. Otherwise Bytes converts the reply to a slice of bytes as follows:\n```\nReply type Result bulk string reply, nil simple string []byte(reply), nil nil nil, ErrNil other nil, error\n```\n#### \nfunc [DoContext](https://github.com/gomodule/redigo/blob/v1.8.9/redis/redis.go#L126) [¶](#DoContext)\n\nadded in v1.8.6\n```\nfunc DoContext(c [Conn](#Conn), ctx [context](/context).[Context](/context#Context), cmd [string](/builtin#string), args ...interface{}) (interface{}, [error](/builtin#error))\n```\nDoContext sends a command to server and returns the received reply.\nmin(ctx,DialReadTimeout()) will be used as the deadline.\nThe connection will be closed if DialReadTimeout() timeout or ctx timeout or ctx canceled when this function is running.\nDialReadTimeout() timeout return err can be checked by errors.Is(err, os.ErrDeadlineExceeded).\nctx timeout return err context.DeadlineExceeded.\nctx canceled return err context.Canceled.\n\n#### \nfunc [DoWithTimeout](https://github.com/gomodule/redigo/blob/v1.8.9/redis/redis.go#L137) [¶](#DoWithTimeout)\n```\nfunc DoWithTimeout(c [Conn](#Conn), timeout [time](/time).[Duration](/time#Duration), cmd [string](/builtin#string), args ...interface{}) (interface{}, [error](/builtin#error))\n```\nDoWithTimeout executes a Redis command with the specified read timeout. If the connection does not satisfy the ConnWithTimeout interface, then an error is returned.\n\n#### \nfunc [Float64](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L127) [¶](#Float64)\n```\nfunc Float64(reply interface{}, err [error](/builtin#error)) ([float64](/builtin#float64), [error](/builtin#error))\n```\nFloat64 is a helper that converts a command reply to 64 bit float. If err is not equal to nil, then Float64 returns 0, err. Otherwise, Float64 converts the reply to a float64 as follows:\n```\nReply type Result bulk string parsed reply, nil nil 0, ErrNil other 0, error\n```\n#### \nfunc [Float64Map](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L487) [¶](#Float64Map)\n\nadded in v1.8.9\n```\nfunc Float64Map(result interface{}, err [error](/builtin#error)) (map[[string](/builtin#string)][float64](/builtin#float64), [error](/builtin#error))\n```\nFloat64Map is a helper that converts an array of strings (alternating key, value)\ninto a map[string]float64. The HGETALL commands return replies in this format.\nRequires an even number of values in result.\n\n#### \nfunc [Float64s](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L277) [¶](#Float64s)\n```\nfunc Float64s(reply interface{}, err [error](/builtin#error)) ([][float64](/builtin#float64), [error](/builtin#error))\n```\nFloat64s is a helper that converts an array command reply to a []float64. If err is not equal to nil, then Float64s returns nil, err. Nil array items are converted to 0 in the output slice. Floats64 returns an error if an array item is not a bulk string or nil.\n\n#### \nfunc [Int](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L36) [¶](#Int)\n```\nfunc Int(reply interface{}, err [error](/builtin#error)) ([int](/builtin#int), [error](/builtin#error))\n```\nInt is a helper that converts a command reply to an integer. If err is not equal to nil, then Int returns 0, err. Otherwise, Int converts the reply to an int as follows:\n```\nReply type Result integer int(reply), nil bulk string parsed reply, nil nil 0, ErrNil other 0, error\n```\nExample [¶](#example-Int)\n```\npackage main\n\nimport (\n\t\"fmt\"\n\n\t\"github.com/gomodule/redigo/redis\"\n)\n\n// dial wraps DialDefaultServer() with a more suitable function name for examples.\nfunc dial() (redis.Conn, error) {\n\treturn redis.DialDefaultServer()\n}\n\nfunc main() {\n\tc, err := dial()\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tdefer c.Close()\n\n\t_, err = c.Do(\"SET\", \"k1\", 1)\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tn, err := redis.Int(c.Do(\"GET\", \"k1\"))\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tfmt.Printf(\"%#v\\n\", n)\n\tn, err = redis.Int(c.Do(\"INCR\", \"k1\"))\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tfmt.Printf(\"%#v\\n\", n)\n}\n```\n```\nOutput:\n\n1 2\n```\nShare Format\nRun\n\n#### \nfunc [Int64](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L67) [¶](#Int64)\n```\nfunc Int64(reply interface{}, err [error](/builtin#error)) ([int64](/builtin#int64), [error](/builtin#error))\n```\nInt64 is a helper that converts a command reply to 64 bit integer. If err is not equal to nil, then Int64 returns 0, err. Otherwise, Int64 converts the reply to an int64 as follows:\n```\nReply type Result integer reply, nil bulk string parsed reply, nil nil 0, ErrNil other 0, error\n```\n#### \nfunc [Int64Map](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L464) [¶](#Int64Map)\n```\nfunc Int64Map(result interface{}, err [error](/builtin#error)) (map[[string](/builtin#string)][int64](/builtin#int64), [error](/builtin#error))\n```\nInt64Map is a helper that converts an array of strings (alternating key, value)\ninto a map[string]int64. The HGETALL commands return replies in this format.\nRequires an even number of values in result.\n\n#### \nfunc [Int64s](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L341) [¶](#Int64s)\n```\nfunc Int64s(reply interface{}, err [error](/builtin#error)) ([][int64](/builtin#int64), [error](/builtin#error))\n```\nInt64s is a helper that converts an array command reply to a []int64.\nIf err is not equal to nil, then Int64s returns nil, err. Nil array items are stay nil. Int64s returns an error if an array item is not a bulk string or nil.\n\n#### \nfunc [IntMap](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L441) [¶](#IntMap)\n```\nfunc IntMap(result interface{}, err [error](/builtin#error)) (map[[string](/builtin#string)][int](/builtin#int), [error](/builtin#error))\n```\nIntMap is a helper that converts an array of strings (alternating key, value)\ninto a map[string]int. The HGETALL commands return replies in this format.\nRequires an even number of values in result.\n\n#### \nfunc [Ints](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L365) [¶](#Ints)\n```\nfunc Ints(reply interface{}, err [error](/builtin#error)) ([][int](/builtin#int), [error](/builtin#error))\n```\nInts is a helper that converts an array command reply to a []int.\nIf err is not equal to nil, then Ints returns nil, err. Nil array items are stay nil. Ints returns an error if an array item is not a bulk string or nil.\n\nExample [¶](#example-Ints)\n```\npackage main\n\nimport (\n\t\"fmt\"\n\n\t\"github.com/gomodule/redigo/redis\"\n)\n\n// dial wraps DialDefaultServer() with a more suitable function name for examples.\nfunc dial() (redis.Conn, error) {\n\treturn redis.DialDefaultServer()\n}\n\nfunc main() {\n\tc, err := dial()\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tdefer c.Close()\n\n\t_, err = c.Do(\"SADD\", \"set_with_integers\", 4, 5, 6)\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tints, err := redis.Ints(c.Do(\"SMEMBERS\", \"set_with_integers\"))\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tfmt.Printf(\"%#v\\n\", ints)\n}\n```\n```\nOutput:\n\n[]int{4, 5, 6}\n```\nShare Format\nRun\n\n#### \n\n func [MultiBulk](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L224)\ndeprecated\n```\nfunc MultiBulk(reply interface{}, err [error](/builtin#error)) ([]interface{}, [error](/builtin#error))\n```\nMultiBulk is a helper that converts an array command reply to a []interface{}.\n\nDeprecated: Use Values instead.\n\n#### \nfunc [Positions](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L509) [¶](#Positions)\n```\nfunc Positions(result interface{}, err [error](/builtin#error)) ([]*[2][float64](/builtin#float64), [error](/builtin#error))\n```\nPositions is a helper that converts an array of positions (lat, long)\ninto a [][2]float64. The GEOPOS command returns replies in this format.\n\n#### \nfunc [ReceiveContext](https://github.com/gomodule/redigo/blob/v1.8.9/redis/redis.go#L151) [¶](#ReceiveContext)\n\nadded in v1.8.6\n```\nfunc ReceiveContext(c [Conn](#Conn), ctx [context](/context).[Context](/context#Context)) (interface{}, [error](/builtin#error))\n```\nReceiveContext receives a single reply from the Redis server.\nmin(ctx,DialReadTimeout()) will be used as the deadline.\nThe connection will be closed if DialReadTimeout() timeout or ctx timeout or ctx canceled when this function is running.\nDialReadTimeout() timeout return err can be checked by strings.Contains(e.Error(), \"io/timeout\").\nctx timeout return err context.DeadlineExceeded.\nctx canceled return err context.Canceled.\n\n#### \nfunc [ReceiveWithTimeout](https://github.com/gomodule/redigo/blob/v1.8.9/redis/redis.go#L162) [¶](#ReceiveWithTimeout)\n```\nfunc ReceiveWithTimeout(c [Conn](#Conn), timeout [time](/time).[Duration](/time#Duration)) (interface{}, [error](/builtin#error))\n```\nReceiveWithTimeout receives a reply with the specified read timeout. If the connection does not satisfy the ConnWithTimeout interface, then an error is returned.\n\n#### \nfunc [Scan](https://github.com/gomodule/redigo/blob/v1.8.9/redis/scan.go#L328) [¶](#Scan)\n```\nfunc Scan(src []interface{}, dest ...interface{}) ([]interface{}, [error](/builtin#error))\n```\nScan copies from src to the values pointed at by dest.\n\nScan uses RedisScan if available otherwise:\n\nThe values pointed at by dest must be an integer, float, boolean, string,\n[]byte, interface{} or slices of these types. Scan uses the standard strconv package to convert bulk strings to numeric and boolean types.\n\nIf a dest value is nil, then the corresponding src value is skipped.\n\nIf a src element is nil, then the corresponding dest value is not modified.\n\nTo enable easy use of Scan in a loop, Scan returns the slice of src following the copied values.\n\nExample [¶](#example-Scan)\n```\nc, err := dial()\nif err != nil {\n\tfmt.Println(err)\n\treturn\n}\ndefer c.Close()\n\nif err = c.Send(\"HMSET\", \"album:1\", \"title\", \"Red\", \"rating\", 5); err != nil {\n\tfmt.Println(err)\n\treturn\n}\nif err = c.Send(\"HMSET\", \"album:2\", \"title\", \"Earthbound\", \"rating\", 1); err != nil {\n\tfmt.Println(err)\n\treturn\n}\nif err = c.Send(\"HMSET\", \"album:3\", \"title\", \"Beat\"); err != nil {\n\tfmt.Println(err)\n\treturn\n}\nif err = c.Send(\"LPUSH\", \"albums\", \"1\"); err != nil {\n\tfmt.Println(err)\n\treturn\n}\nif err = c.Send(\"LPUSH\", \"albums\", \"2\"); err != nil {\n\tfmt.Println(err)\n\treturn\n}\nif err = c.Send(\"LPUSH\", \"albums\", \"3\"); err != nil {\n\tfmt.Println(err)\n\treturn\n}\nvalues, err := redis.Values(c.Do(\"SORT\", \"albums\",\n\t\"BY\", \"album:*->rating\",\n\t\"GET\", \"album:*->title\",\n\t\"GET\", \"album:*->rating\"))\nif err != nil {\n\tfmt.Println(err)\n\treturn\n}\n\nfor len(values) > 0 {\n\tvar title string\n\trating := -1 // initialize to illegal value to detect nil.\n\tvalues, err = redis.Scan(values, &title, &rating)\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tif rating == -1 {\n\t\tfmt.Println(title, \"not-rated\")\n\t} else {\n\t\tfmt.Println(title, rating)\n\t}\n}\n```\n```\nOutput:\n\nBeat not-rated Earthbound 1 Red 5\n```\n#### \nfunc [ScanSlice](https://github.com/gomodule/redigo/blob/v1.8.9/redis/scan.go#L547) [¶](#ScanSlice)\n```\nfunc ScanSlice(src []interface{}, dest interface{}, fieldNames ...[string](/builtin#string)) [error](/builtin#error)\n```\nScanSlice scans src to the slice pointed to by dest.\n\nIf the target is a slice of types which implement Scanner then the custom RedisScan method is used otherwise the following rules apply:\n\nThe elements in the dest slice must be integer, float, boolean, string, struct or pointer to struct values.\n\nStruct fields must be integer, float, boolean or string values. All struct fields are used unless a subset is specified using fieldNames.\n\nExample [¶](#example-ScanSlice)\n```\nc, err := dial()\nif err != nil {\n\tfmt.Println(err)\n\treturn\n}\ndefer c.Close()\n\nif err = c.Send(\"HMSET\", \"album:1\", \"title\", \"Red\", \"rating\", 5); err != nil {\n\tfmt.Println(err)\n\treturn\n}\nif err = c.Send(\"HMSET\", \"album:2\", \"title\", \"Earthbound\", \"rating\", 1); err != nil {\n\tfmt.Println(err)\n\treturn\n}\nif err = c.Send(\"HMSET\", \"album:3\", \"title\", \"Beat\", \"rating\", 4); err != nil {\n\tfmt.Println(err)\n\treturn\n}\nif err = c.Send(\"LPUSH\", \"albums\", \"1\"); err != nil {\n\tfmt.Println(err)\n\treturn\n}\nif err = c.Send(\"LPUSH\", \"albums\", \"2\"); err != nil {\n\tfmt.Println(err)\n\treturn\n}\nif err = c.Send(\"LPUSH\", \"albums\", \"3\"); err != nil {\n\tfmt.Println(err)\n\treturn\n}\nvalues, err := redis.Values(c.Do(\"SORT\", \"albums\",\n\t\"BY\", \"album:*->rating\",\n\t\"GET\", \"album:*->title\",\n\t\"GET\", \"album:*->rating\"))\nif err != nil {\n\tfmt.Println(err)\n\treturn\n}\n\nvar albums []struct {\n\tTitle string\n\tRating int\n}\nif err := redis.ScanSlice(values, &albums); err != nil {\n\tfmt.Println(err)\n\treturn\n}\nfmt.Printf(\"%v\\n\", albums)\n```\n```\nOutput:\n\n[{Earthbound 1} {Beat 4} {Red 5}]\n```\n#### \nfunc [ScanStruct](https://github.com/gomodule/redigo/blob/v1.8.9/redis/scan.go#L490) [¶](#ScanStruct)\n```\nfunc ScanStruct(src []interface{}, dest interface{}) [error](/builtin#error)\n```\nScanStruct scans alternating names and values from src to a struct. The HGETALL and CONFIG GET commands return replies in this format.\n\nScanStruct uses exported field names to match values in the response. Use\n'redis' field tag to override the name:\n```\nField int `redis:\"myName\"`\n```\nFields with the tag redis:\"-\" are ignored.\n\nEach field uses RedisScan if available otherwise:\nInteger, float, boolean, string and []byte fields are supported. Scan uses the standard strconv package to convert bulk string values to numeric and boolean types.\n\nIf a src element is nil, then the corresponding field is not modified.\n\n#### \nfunc [String](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L152) [¶](#String)\n```\nfunc String(reply interface{}, err [error](/builtin#error)) ([string](/builtin#string), [error](/builtin#error))\n```\nString is a helper that converts a command reply to a string. If err is not equal to nil, then String returns \"\", err. Otherwise String converts the reply to a string as follows:\n```\nReply type Result bulk string string(reply), nil simple string reply, nil nil \"\", ErrNil other \"\", error\n```\nExample [¶](#example-String)\n```\npackage main\n\nimport (\n\t\"fmt\"\n\n\t\"github.com/gomodule/redigo/redis\"\n)\n\n// dial wraps DialDefaultServer() with a more suitable function name for examples.\nfunc dial() (redis.Conn, error) {\n\treturn redis.DialDefaultServer()\n}\n\nfunc main() {\n\tc, err := dial()\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tdefer c.Close()\n\n\t_, err = c.Do(\"SET\", \"hello\", \"world\")\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\ts, err := redis.String(c.Do(\"GET\", \"hello\"))\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tfmt.Printf(\"%#v\\n\", s)\n}\n```\n```\nOutput:\n\n\"world\"\n```\nShare Format\nRun\n\n#### \nfunc [StringMap](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L418) [¶](#StringMap)\n```\nfunc StringMap(reply interface{}, err [error](/builtin#error)) (map[[string](/builtin#string)][string](/builtin#string), [error](/builtin#error))\n```\nStringMap is a helper that converts an array of strings (alternating key, value)\ninto a map[string]string. The HGETALL and CONFIG GET commands return replies in this format.\nRequires an even number of values in result.\n\n#### \nfunc [Strings](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L298) [¶](#Strings)\n```\nfunc Strings(reply interface{}, err [error](/builtin#error)) ([][string](/builtin#string), [error](/builtin#error))\n```\nStrings is a helper that converts an array command reply to a []string. If err is not equal to nil, then Strings returns nil, err. Nil array items are converted to \"\" in the output slice. Strings returns an error if an array item is not a bulk string or nil.\n\n#### \nfunc [Uint64](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L98) [¶](#Uint64)\n```\nfunc Uint64(reply interface{}, err [error](/builtin#error)) ([uint64](/builtin#uint64), [error](/builtin#error))\n```\nUint64 is a helper that converts a command reply to 64 bit unsigned integer.\nIf err is not equal to nil, then Uint64 returns 0, err. Otherwise, Uint64 converts the reply to an uint64 as follows:\n```\nReply type Result\n+integer reply, nil bulk string parsed reply, nil nil 0, ErrNil other 0, error\n```\n#### \nfunc [Uint64Map](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L571) [¶](#Uint64Map)\n\nadded in v1.7.2\n```\nfunc Uint64Map(result interface{}, err [error](/builtin#error)) (map[[string](/builtin#string)][uint64](/builtin#uint64), [error](/builtin#error))\n```\nUint64Map is a helper that converts an array of strings (alternating key, value)\ninto a map[string]uint64. The HGETALL commands return replies in this format.\nRequires an even number of values in result.\n\n#### \nfunc [Uint64s](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L548) [¶](#Uint64s)\n\nadded in v1.7.2\n```\nfunc Uint64s(reply interface{}, err [error](/builtin#error)) ([][uint64](/builtin#uint64), [error](/builtin#error))\n```\nUint64s is a helper that converts an array command reply to a []uint64.\nIf err is not equal to nil, then Uint64s returns nil, err. Nil array items are stay nil. Uint64s returns an error if an array item is not a bulk string or nil.\n\n#### \nfunc [Values](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L234) [¶](#Values)\n```\nfunc Values(reply interface{}, err [error](/builtin#error)) ([]interface{}, [error](/builtin#error))\n```\nValues is a helper that converts an array command reply to a []interface{}.\nIf err is not equal to nil, then Values returns nil, err. Otherwise, Values converts the reply as follows:\n```\nReply type Result array reply, nil nil nil, ErrNil other nil, error\n```\n### Types [¶](#pkg-types)\n\n#### \ntype [Args](https://github.com/gomodule/redigo/blob/v1.8.9/redis/scan.go#L626) [¶](#Args)\n```\ntype Args []interface{}\n```\nArgs is a helper for constructing command arguments from structured values.\n\nExample [¶](#example-Args)\n```\nc, err := dial()\nif err != nil {\n\tfmt.Println(err)\n\treturn\n}\ndefer c.Close()\n\nvar p1, p2 struct {\n\tTitle string `redis:\"title\"`\n\tAuthor string `redis:\"author\"`\n\tBody string `redis:\"body\"`\n}\n\np1.Title = \"Example\"\np1.Author = \"Gary\"\np1.Body = \"Hello\"\n\nif _, err := c.Do(\"HMSET\", redis.Args{}.Add(\"id1\").AddFlat(&p1)...); err != nil {\n\tfmt.Println(err)\n\treturn\n}\n\nm := map[string]string{\n\t\"title\": \"Example2\",\n\t\"author\": \"Steve\",\n\t\"body\": \"Map\",\n}\n\nif _, err := c.Do(\"HMSET\", redis.Args{}.Add(\"id2\").AddFlat(m)...); err != nil {\n\tfmt.Println(err)\n\treturn\n}\n\nfor _, id := range []string{\"id1\", \"id2\"} {\n\n\tv, err := redis.Values(c.Do(\"HGETALL\", id))\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\n\tif err := redis.ScanStruct(v, &p2); err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\n\tfmt.Printf(\"%+v\\n\", p2)\n}\n```\n```\nOutput:\n\n{Title:Example Author:Gary Body:Hello}\n{Title:Example2 Author:Steve Body:Map}\n```\n#### \nfunc (Args) [Add](https://github.com/gomodule/redigo/blob/v1.8.9/redis/scan.go#L629) [¶](#Args.Add)\n```\nfunc (args [Args](#Args)) Add(value ...interface{}) [Args](#Args)\n```\nAdd returns the result of appending value to args.\n\n#### \nfunc (Args) [AddFlat](https://github.com/gomodule/redigo/blob/v1.8.9/redis/scan.go#L646) [¶](#Args.AddFlat)\n```\nfunc (args [Args](#Args)) AddFlat(v interface{}) [Args](#Args)\n```\nAddFlat returns the result of appending the flattened value of v to args.\n\nMaps are flattened by appending the alternating keys and map values to args.\n\nSlices are flattened by appending the slice elements to args.\n\nStructs are flattened by appending the alternating names and values of exported fields to args. If v is a nil struct pointer, then nothing is appended. The 'redis' field tag overrides struct field names. See ScanStruct for more information on the use of the 'redis' field tag.\n\nOther types are appended to args as is.\npanics if v includes a recursive anonymous struct.\n\n#### \ntype [Argument](https://github.com/gomodule/redigo/blob/v1.8.9/redis/redis.go#L52) [¶](#Argument)\n```\ntype Argument interface {\n // RedisArg returns a value to be encoded as a bulk string per the\n\t// conversions listed in the section 'Executing Commands'.\n\t// Implementations should typically return a []byte or string.\n\tRedisArg() interface{}\n}\n```\nArgument is the interface implemented by an object which wants to control how the object is converted to Redis bulk strings.\n\n#### \ntype [Conn](https://github.com/gomodule/redigo/blob/v1.8.9/redis/redis.go#L29) [¶](#Conn)\n```\ntype Conn interface {\n // Close closes the connection.\n\tClose() [error](/builtin#error)\n\n // Err returns a non-nil value when the connection is not usable.\n\tErr() [error](/builtin#error)\n\n // Do sends a command to the server and returns the received reply.\n\t// This function will use the timeout which was set when the connection is created\n\tDo(commandName [string](/builtin#string), args ...interface{}) (reply interface{}, err [error](/builtin#error))\n\n // Send writes the command to the client's output buffer.\n\tSend(commandName [string](/builtin#string), args ...interface{}) [error](/builtin#error)\n\n // Flush flushes the output buffer to the Redis server.\n\tFlush() [error](/builtin#error)\n\n // Receive receives a single reply from the Redis server\n\tReceive() (reply interface{}, err [error](/builtin#error))\n}\n```\nConn represents a connection to a Redis server.\n\n#### \nfunc [Dial](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L212) [¶](#Dial)\n```\nfunc Dial(network, address [string](/builtin#string), options ...[DialOption](#DialOption)) ([Conn](#Conn), [error](/builtin#error))\n```\nDial connects to the Redis server at the given network and address using the specified options.\n\nExample [¶](#example-Dial)\n\nConnect to local instance of Redis running on the default port.\n```\npackage main\n\nimport (\n\t\"github.com/gomodule/redigo/redis\"\n)\n\nfunc main() {\n\tc, err := redis.Dial(\"tcp\", \":6379\")\n\tif err != nil {\n\t\t// handle error\n\t}\n\tdefer c.Close()\n}\n```\n```\nOutput:\n```\nShare Format\nRun\n\nExample (Acl) [¶](#example-Dial-Acl)\n\nConnect to an Redis instance using the Redis ACL system\n```\npackage main\n\nimport (\n\t\"github.com/gomodule/redigo/redis\"\n)\n\nfunc main() {\n\tc, err := redis.Dial(\"tcp\", \"localhost:6379\",\n\t\tredis.DialUsername(\"username\"),\n\t\tredis.DialPassword(\"password\"),\n\t)\n\tif err != nil {\n\t\t// handle error\n\t}\n\tdefer c.Close()\n}\n```\n```\nOutput:\n```\nShare Format\nRun\n\n#### \nfunc [DialContext](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L224) [¶](#DialContext)\n\nadded in v1.8.2\n```\nfunc DialContext(ctx [context](/context).[Context](/context#Context), network, address [string](/builtin#string), options ...[DialOption](#DialOption)) ([Conn](#Conn), [error](/builtin#error))\n```\nDialContext connects to the Redis server at the given network and address using the specified options and context.\n\nExample [¶](#example-DialContext)\n\nConnect to local instance of Redis running on the default port using the provided context.\n```\npackage main\n\nimport (\n\t\"context\"\n\n\t\"github.com/gomodule/redigo/redis\"\n)\n\nfunc main() {\n\tctx := context.Background()\n\tc, err := redis.DialContext(ctx, \"tcp\", \":6379\")\n\tif err != nil {\n\t\t// handle error\n\t}\n\tdefer c.Close()\n}\n```\n```\nOutput:\n```\nShare Format\nRun\n\n#### \n\n func [DialTimeout](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L65)\ndeprecated\n```\nfunc DialTimeout(network, address [string](/builtin#string), connectTimeout, readTimeout, writeTimeout [time](/time).[Duration](/time#Duration)) ([Conn](#Conn), [error](/builtin#error))\n```\nDialTimeout acts like Dial but takes timeouts for establishing the connection to the server, writing a command and reading a reply.\n\nDeprecated: Use Dial with options instead.\n\n#### \nfunc [DialURL](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L320) [¶](#DialURL)\n```\nfunc DialURL(rawurl [string](/builtin#string), options ...[DialOption](#DialOption)) ([Conn](#Conn), [error](/builtin#error))\n```\nDialURL wraps DialURLContext using context.Background.\n\nExample [¶](#example-DialURL)\n\nConnect to remote instance of Redis using a URL.\n```\npackage main\n\nimport (\n\t\"os\"\n\n\t\"github.com/gomodule/redigo/redis\"\n)\n\nfunc main() {\n\tc, err := redis.DialURL(os.Getenv(\"REDIS_URL\"))\n\tif err != nil {\n\t\t// handle connection error\n\t}\n\tdefer c.Close()\n}\n```\n```\nOutput:\n```\nShare Format\nRun\n\n#### \nfunc [DialURLContext](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L329) [¶](#DialURLContext)\n\nadded in v1.8.6\n```\nfunc DialURLContext(ctx [context](/context).[Context](/context#Context), rawurl [string](/builtin#string), options ...[DialOption](#DialOption)) ([Conn](#Conn), [error](/builtin#error))\n```\nDialURLContext connects to a Redis server at the given URL using the Redis URI scheme. URLs should follow the draft IANA specification for the scheme (<https://www.iana.org/assignments/uri-schemes/prov/redis>).\n\n#### \nfunc [NewConn](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L395) [¶](#NewConn)\n```\nfunc NewConn(netConn [net](/net).[Conn](/net#Conn), readTimeout, writeTimeout [time](/time).[Duration](/time#Duration)) [Conn](#Conn)\n```\nNewConn returns a new Redigo connection for the given net connection.\n\n#### \nfunc [NewLoggingConn](https://github.com/gomodule/redigo/blob/v1.8.9/redis/log.go#L30) [¶](#NewLoggingConn)\n```\nfunc NewLoggingConn(conn [Conn](#Conn), logger *[log](/log).[Logger](/log#Logger), prefix [string](/builtin#string)) [Conn](#Conn)\n```\nNewLoggingConn returns a logging wrapper around a connection.\n\n#### \nfunc [NewLoggingConnFilter](https://github.com/gomodule/redigo/blob/v1.8.9/redis/log.go#L38) [¶](#NewLoggingConnFilter)\n```\nfunc NewLoggingConnFilter(conn [Conn](#Conn), logger *[log](/log).[Logger](/log#Logger), prefix [string](/builtin#string), skip func(cmdName [string](/builtin#string)) [bool](/builtin#bool)) [Conn](#Conn)\n```\nNewLoggingConnFilter returns a logging wrapper around a connection and a filter function.\n\n#### \ntype [ConnWithContext](https://github.com/gomodule/redigo/blob/v1.8.9/redis/redis.go#L97) [¶](#ConnWithContext)\n\nadded in v1.8.6\n```\ntype ConnWithContext interface {\n\t[Conn](#Conn)\n\n // DoContext sends a command to server and returns the received reply.\n\t// min(ctx,DialReadTimeout()) will be used as the deadline.\n\t// The connection will be closed if DialReadTimeout() timeout or ctx timeout or ctx canceled when this function is running.\n\t// DialReadTimeout() timeout return err can be checked by errors.Is(err, os.ErrDeadlineExceeded).\n\t// ctx timeout return err context.DeadlineExceeded.\n\t// ctx canceled return err context.Canceled.\n\tDoContext(ctx [context](/context).[Context](/context#Context), commandName [string](/builtin#string), args ...interface{}) (reply interface{}, err [error](/builtin#error))\n\n // ReceiveContext receives a single reply from the Redis server.\n\t// min(ctx,DialReadTimeout()) will be used as the deadline.\n\t// The connection will be closed if DialReadTimeout() timeout or ctx timeout or ctx canceled when this function is running.\n\t// DialReadTimeout() timeout return err can be checked by errors.Is(err, os.ErrDeadlineExceeded).\n\t// ctx timeout return err context.DeadlineExceeded.\n\t// ctx canceled return err context.Canceled.\n\tReceiveContext(ctx [context](/context).[Context](/context#Context)) (reply interface{}, err [error](/builtin#error))\n}\n```\nConnWithContext is an optional interface that allows the caller to control the command's life with context.\n\n#### \ntype [ConnWithTimeout](https://github.com/gomodule/redigo/blob/v1.8.9/redis/redis.go#L84) [¶](#ConnWithTimeout)\n```\ntype ConnWithTimeout interface {\n\t[Conn](#Conn)\n\n // DoWithTimeout sends a command to the server and returns the received reply.\n\t// The timeout overrides the readtimeout set when dialing the connection.\n\tDoWithTimeout(timeout [time](/time).[Duration](/time#Duration), commandName [string](/builtin#string), args ...interface{}) (reply interface{}, err [error](/builtin#error))\n\n // ReceiveWithTimeout receives a single reply from the Redis server.\n\t// The timeout overrides the readtimeout set when dialing the connection.\n\tReceiveWithTimeout(timeout [time](/time).[Duration](/time#Duration)) (reply interface{}, err [error](/builtin#error))\n}\n```\nConnWithTimeout is an optional interface that allows the caller to override a connection's default read timeout. This interface is useful for executing the BLPOP, BRPOP, BRPOPLPUSH, XREAD and other commands that block at the server.\n\nA connection's default read timeout is set with the DialReadTimeout dial option. Applications should rely on the default timeout for commands that do not block at the server.\n\nAll of the Conn implementations in this package satisfy the ConnWithTimeout interface.\n\nUse the DoWithTimeout and ReceiveWithTimeout helper functions to simplify use of this interface.\n\n#### \ntype [DialOption](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L73) [¶](#DialOption)\n```\ntype DialOption struct {\n\t// contains filtered or unexported fields\n}\n```\nDialOption specifies an option for dialing a Redis server.\n\n#### \nfunc [DialClientName](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L180) [¶](#DialClientName)\n\nadded in v1.7.2\n```\nfunc DialClientName(name [string](/builtin#string)) [DialOption](#DialOption)\n```\nDialClientName specifies a client name to be used by the Redis server connection.\n\n#### \nfunc [DialConnectTimeout](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L118) [¶](#DialConnectTimeout)\n```\nfunc DialConnectTimeout(d [time](/time).[Duration](/time#Duration)) [DialOption](#DialOption)\n```\nDialConnectTimeout specifies the timeout for connecting to the Redis server when no DialNetDial option is specified.\nIf no DialConnectTimeout option is specified then the default is 30 seconds.\n\n#### \nfunc [DialContextFunc](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L148) [¶](#DialContextFunc)\n\nadded in v1.8.2\n```\nfunc DialContextFunc(f func(ctx [context](/context).[Context](/context#Context), network, addr [string](/builtin#string)) ([net](/net).[Conn](/net#Conn), [error](/builtin#error))) [DialOption](#DialOption)\n```\nDialContextFunc specifies a custom dial function with context for creating TCP connections, otherwise a net.Dialer customized via the other options is used.\nDialContextFunc overrides DialConnectTimeout and DialKeepAlive.\n\n#### \nfunc [DialDatabase](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L155) [¶](#DialDatabase)\n```\nfunc DialDatabase(db [int](/builtin#int)) [DialOption](#DialOption)\n```\nDialDatabase specifies the database to select when dialing a connection.\n\n#### \nfunc [DialKeepAlive](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L128) [¶](#DialKeepAlive)\n```\nfunc DialKeepAlive(d [time](/time).[Duration](/time#Duration)) [DialOption](#DialOption)\n```\nDialKeepAlive specifies the keep-alive period for TCP connections to the Redis server when no DialNetDial option is specified.\nIf zero, keep-alives are not enabled. If no DialKeepAlive option is specified then the default of 5 minutes is used to ensure that half-closed TCP sessions are detected.\n\n#### \nfunc [DialNetDial](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L137) [¶](#DialNetDial)\n```\nfunc DialNetDial(dial func(network, addr [string](/builtin#string)) ([net](/net).[Conn](/net#Conn), [error](/builtin#error))) [DialOption](#DialOption)\n```\nDialNetDial specifies a custom dial function for creating TCP connections, otherwise a net.Dialer customized via the other options is used.\nDialNetDial overrides DialConnectTimeout and DialKeepAlive.\n\n#### \nfunc [DialPassword](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L163) [¶](#DialPassword)\n```\nfunc DialPassword(password [string](/builtin#string)) [DialOption](#DialOption)\n```\nDialPassword specifies the password to use when connecting to the Redis server.\n\n#### \nfunc [DialReadTimeout](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L102) [¶](#DialReadTimeout)\n```\nfunc DialReadTimeout(d [time](/time).[Duration](/time#Duration)) [DialOption](#DialOption)\n```\nDialReadTimeout specifies the timeout for reading a single command reply.\n\n#### \nfunc [DialTLSConfig](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L188) [¶](#DialTLSConfig)\n```\nfunc DialTLSConfig(c *[tls](/crypto/tls).[Config](/crypto/tls#Config)) [DialOption](#DialOption)\n```\nDialTLSConfig specifies the config to use when a TLS connection is dialed.\nHas no effect when not dialing a TLS connection.\n\n#### \nfunc [DialTLSHandshakeTimeout](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L95) [¶](#DialTLSHandshakeTimeout)\n\nadded in v1.8.3\n```\nfunc DialTLSHandshakeTimeout(d [time](/time).[Duration](/time#Duration)) [DialOption](#DialOption)\n```\nDialTLSHandshakeTimeout specifies the maximum amount of time waiting to wait for a TLS handshake. Zero means no timeout.\nIf no DialTLSHandshakeTimeout option is specified then the default is 30 seconds.\n\n#### \nfunc [DialTLSSkipVerify](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L196) [¶](#DialTLSSkipVerify)\n```\nfunc DialTLSSkipVerify(skip [bool](/builtin#bool)) [DialOption](#DialOption)\n```\nDialTLSSkipVerify disables server name verification when connecting over TLS. Has no effect when not dialing a TLS connection.\n\n#### \nfunc [DialUseTLS](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L204) [¶](#DialUseTLS)\n```\nfunc DialUseTLS(useTLS [bool](/builtin#bool)) [DialOption](#DialOption)\n```\nDialUseTLS specifies whether TLS should be used when connecting to the server. This option is ignore by DialURL.\n\n#### \nfunc [DialUsername](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L172) [¶](#DialUsername)\n\nadded in v1.8.2\n```\nfunc DialUsername(username [string](/builtin#string)) [DialOption](#DialOption)\n```\nDialUsername specifies the username to use when connecting to the Redis server when Redis ACLs are used.\nA DialPassword must also be passed otherwise this option will have no effect.\n\n#### \nfunc [DialWriteTimeout](https://github.com/gomodule/redigo/blob/v1.8.9/redis/conn.go#L109) [¶](#DialWriteTimeout)\n```\nfunc DialWriteTimeout(d [time](/time).[Duration](/time#Duration)) [DialOption](#DialOption)\n```\nDialWriteTimeout specifies the timeout for writing a single command.\n\n#### \ntype [Error](https://github.com/gomodule/redigo/blob/v1.8.9/redis/redis.go#L24) [¶](#Error)\n```\ntype Error [string](/builtin#string)\n```\nError represents an error returned in a command reply.\n\n#### \nfunc (Error) [Error](https://github.com/gomodule/redigo/blob/v1.8.9/redis/redis.go#L26) [¶](#Error.Error)\n```\nfunc (err [Error](#Error)) Error() [string](/builtin#string)\n```\n#### \ntype [Latency](https://github.com/gomodule/redigo/blob/v1.8.9/redis/redis.go#L192) [¶](#Latency)\n\nadded in v1.8.9\n```\ntype Latency struct {\n // Name of the latest latency spike event.\n\tName [string](/builtin#string)\n\n // Time of the latest latency spike for the event.\n\tTime [time](/time).[Time](/time#Time)\n\n // Latest is the latest recorded latency for the named event.\n\tLatest [time](/time).[Duration](/time#Duration)\n\n // Max is the maximum latency for the named event.\n\tMax [time](/time).[Duration](/time#Duration)\n}\n```\nLatency represents a redis LATENCY LATEST.\n\n#### \nfunc [Latencies](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L651) [¶](#Latencies)\n\nadded in v1.8.9\n```\nfunc Latencies(result interface{}, err [error](/builtin#error)) ([][Latency](#Latency), [error](/builtin#error))\n```\nLatencies is a helper that parses the LATENCY LATEST command output and return the slice of Latency values.\n\n#### \ntype [LatencyHistory](https://github.com/gomodule/redigo/blob/v1.8.9/redis/redis.go#L207) [¶](#LatencyHistory)\n\nadded in v1.8.9\n```\ntype LatencyHistory struct {\n // Time is the unix timestamp at which the event was processed.\n\tTime [time](/time).[Time](/time#Time)\n\n // ExecutationTime is the amount of time needed for the command execution.\n\tExecutionTime [time](/time).[Duration](/time#Duration)\n}\n```\nLatencyHistory represents a redis LATENCY HISTORY.\n\n#### \nfunc [LatencyHistories](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L703) [¶](#LatencyHistories)\n\nadded in v1.8.9\n```\nfunc LatencyHistories(result interface{}, err [error](/builtin#error)) ([][LatencyHistory](#LatencyHistory), [error](/builtin#error))\n```\nLatencyHistories is a helper that parse the LATENCY HISTORY command output and returns a LatencyHistory slice.\n\n#### \ntype [Message](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pubsub.go#L36) [¶](#Message)\n```\ntype Message struct {\n // The originating channel.\n\tChannel [string](/builtin#string)\n\n // The matched pattern, if any\n\tPattern [string](/builtin#string)\n\n // The message data.\n\tData [][byte](/builtin#byte)\n}\n```\nMessage represents a message notification.\n\n#### \ntype [Pong](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pubsub.go#L48) [¶](#Pong)\n```\ntype Pong struct {\n Data [string](/builtin#string)\n}\n```\nPong represents a pubsub pong notification.\n\n#### \ntype [Pool](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pool.go#L120) [¶](#Pool)\n```\ntype Pool struct {\n // Dial is an application supplied function for creating and configuring a\n\t// connection.\n\t//\n\t// The connection returned from Dial must not be in a special state\n\t// (subscribed to pubsub channel, transaction started, ...).\n\tDial func() ([Conn](#Conn), [error](/builtin#error))\n\n // DialContext is an application supplied function for creating and configuring a\n\t// connection with the given context.\n\t//\n\t// The connection returned from Dial must not be in a special state\n\t// (subscribed to pubsub channel, transaction started, ...).\n\tDialContext func(ctx [context](/context).[Context](/context#Context)) ([Conn](#Conn), [error](/builtin#error))\n\n // TestOnBorrow is an optional application supplied function for checking\n\t// the health of an idle connection before the connection is used again by\n\t// the application. Argument t is the time that the connection was returned\n\t// to the pool. If the function returns an error, then the connection is\n\t// closed.\n\tTestOnBorrow func(c [Conn](#Conn), t [time](/time).[Time](/time#Time)) [error](/builtin#error)\n\n // Maximum number of idle connections in the pool.\n\tMaxIdle [int](/builtin#int)\n\n // Maximum number of connections allocated by the pool at a given time.\n\t// When zero, there is no limit on the number of connections in the pool.\n\tMaxActive [int](/builtin#int)\n\n // Close connections after remaining idle for this duration. If the value\n\t// is zero, then idle connections are not closed. Applications should set\n\t// the timeout to a value less than the server's timeout.\n\tIdleTimeout [time](/time).[Duration](/time#Duration)\n\n // If Wait is true and the pool is at the MaxActive limit, then Get() waits\n\t// for a connection to be returned to the pool before returning.\n\tWait [bool](/builtin#bool)\n\n // Close connections older than this duration. If the value is zero, then\n\t// the pool does not close connections based on age.\n\tMaxConnLifetime [time](/time).[Duration](/time#Duration)\n\t// contains filtered or unexported fields\n}\n```\nPool maintains a pool of connections. The application calls the Get method to get a connection from the pool and the connection's Close method to return the connection's resources to the pool.\n\nThe following example shows how to use a pool in a web application. The application creates a pool at application startup and makes it available to request handlers using a package level variable. The pool configuration used here is an example, not a recommendation.\n```\nfunc newPool(addr string) *redis.Pool {\n return &redis.Pool{\n MaxIdle: 3,\n IdleTimeout: 240 * time.Second,\n // Dial or DialContext must be set. When both are set, DialContext takes precedence over Dial.\n Dial: func () (redis.Conn, error) { return redis.Dial(\"tcp\", addr) },\n }\n}\n\nvar (\n pool *redis.Pool\n redisServer = flag.String(\"redisServer\", \":6379\", \"\")\n)\n\nfunc main() {\n flag.Parse()\n pool = newPool(*redisServer)\n ...\n}\n```\nA request handler gets a connection from the pool and closes the connection when the handler is done:\n```\nfunc serveHome(w http.ResponseWriter, r *http.Request) {\n conn := pool.Get()\n defer conn.Close()\n ...\n}\n```\nUse the Dial function to authenticate connections with the AUTH command or select a database with the SELECT command:\n```\npool := &redis.Pool{\n // Other pool configuration not shown in this example.\n Dial: func () (redis.Conn, error) {\n c, err := redis.Dial(\"tcp\", server)\n if err != nil {\n return nil, err\n }\n if _, err := c.Do(\"AUTH\", password); err != nil {\n c.Close()\n return nil, err\n }\n if _, err := c.Do(\"SELECT\", db); err != nil {\n c.Close()\n return nil, err\n }\n return c, nil\n },\n}\n```\nUse the TestOnBorrow function to check the health of an idle connection before the connection is returned to the application. This example PINGs connections that have been idle more than a minute:\n```\npool := &redis.Pool{\n // Other pool configuration not shown in this example.\n TestOnBorrow: func(c redis.Conn, t time.Time) error {\n if time.Since(t) < time.Minute {\n return nil\n }\n _, err := c.Do(\"PING\")\n return err\n },\n}\n```\n#### \n\n func [NewPool](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pool.go#L175)\ndeprecated\n```\nfunc NewPool(newFn func() ([Conn](#Conn), [error](/builtin#error)), maxIdle [int](/builtin#int)) *[Pool](#Pool)\n```\nNewPool creates a new pool.\n\nDeprecated: Initialize the Pool directly as shown in the example.\n\n#### \nfunc (*Pool) [ActiveCount](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pool.go#L300) [¶](#Pool.ActiveCount)\n```\nfunc (p *[Pool](#Pool)) ActiveCount() [int](/builtin#int)\n```\nActiveCount returns the number of connections in the pool. The count includes idle connections and connections in use.\n\n#### \nfunc (*Pool) [Close](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pool.go#L316) [¶](#Pool.Close)\n```\nfunc (p *[Pool](#Pool)) Close() [error](/builtin#error)\n```\nClose releases the resources used by the pool.\n\n#### \nfunc (*Pool) [Get](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pool.go#L184) [¶](#Pool.Get)\n```\nfunc (p *[Pool](#Pool)) Get() [Conn](#Conn)\n```\nGet gets a connection. The application must close the returned connection.\nThis method always returns a valid connection so that applications can defer error handling to the first use of the connection. If there is an error getting an underlying connection, then the connection Err, Do, Send, Flush and Receive methods return that error.\n\n#### \nfunc (*Pool) [GetContext](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pool.go#L198) [¶](#Pool.GetContext)\n```\nfunc (p *[Pool](#Pool)) GetContext(ctx [context](/context).[Context](/context#Context)) ([Conn](#Conn), [error](/builtin#error))\n```\nGetContext gets a connection using the provided context.\n\nThe provided Context must be non-nil. If the context expires before the connection is complete, an error is returned. Any expiration on the context will not affect the returned connection.\n\nIf the function completes without error, then the application must close the returned connection.\n\n#### \nfunc (*Pool) [IdleCount](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pool.go#L308) [¶](#Pool.IdleCount)\n```\nfunc (p *[Pool](#Pool)) IdleCount() [int](/builtin#int)\n```\nIdleCount returns the number of idle connections in the pool.\n\n#### \nfunc (*Pool) [Stats](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pool.go#L285) [¶](#Pool.Stats)\n```\nfunc (p *[Pool](#Pool)) Stats() [PoolStats](#PoolStats)\n```\nStats returns pool's statistics.\n\n#### \ntype [PoolStats](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pool.go#L268) [¶](#PoolStats)\n```\ntype PoolStats struct {\n // ActiveCount is the number of connections in the pool. The count includes\n\t// idle connections and connections in use.\n\tActiveCount [int](/builtin#int)\n // IdleCount is the number of idle connections in the pool.\n\tIdleCount [int](/builtin#int)\n\n // WaitCount is the total number of connections waited for.\n\t// This value is currently not guaranteed to be 100% accurate.\n\tWaitCount [int64](/builtin#int64)\n\n // WaitDuration is the total time blocked waiting for a new connection.\n\t// This value is currently not guaranteed to be 100% accurate.\n\tWaitDuration [time](/time).[Duration](/time#Duration)\n}\n```\nPoolStats contains pool statistics.\n\n#### \ntype [PubSubConn](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pubsub.go#L53) [¶](#PubSubConn)\n```\ntype PubSubConn struct {\n Conn [Conn](#Conn)\n}\n```\nPubSubConn wraps a Conn with convenience methods for subscribers.\n\nExample [¶](#example-PubSubConn)\n\nThis example shows how receive pubsub notifications with cancelation and health checks.\n```\n//go:build go1.7\n// +build go1.7\n\npackage main\n\nimport (\n\t\"context\"\n\t\"fmt\"\n\t\"time\"\n\n\t\"github.com/gomodule/redigo/redis\"\n)\n\n// listenPubSubChannels listens for messages on Redis pubsub channels. The\n// onStart function is called after the channels are subscribed. The onMessage\n// function is called for each message.\nfunc listenPubSubChannels(ctx context.Context, redisServerAddr string,\n\tonStart func() error,\n\tonMessage func(channel string, data []byte) error,\n\tchannels ...string) error {\n\t// A ping is set to the server with this period to test for the health of\n\t// the connection and server.\n\tconst healthCheckPeriod = time.Minute\n\n\tc, err := redis.Dial(\"tcp\", redisServerAddr,\n\t\t// Read timeout on server should be greater than ping period.\n\t\tredis.DialReadTimeout(healthCheckPeriod+10*time.Second),\n\t\tredis.DialWriteTimeout(10*time.Second))\n\tif err != nil {\n\t\treturn err\n\t}\n\tdefer c.Close()\n\n\tpsc := redis.PubSubConn{Conn: c}\n\n\tif err := psc.Subscribe(redis.Args{}.AddFlat(channels)...); err != nil {\n\t\treturn err\n\t}\n\n\tdone := make(chan error, 1)\n\n\t// Start a goroutine to receive notifications from the server.\n\tgo func() {\n\t\tfor {\n\t\t\tswitch n := psc.Receive().(type) {\n\t\t\tcase error:\n\t\t\t\tdone <- n\n\t\t\t\treturn\n\t\t\tcase redis.Message:\n\t\t\t\tif err := onMessage(n.Channel, n.Data); err != nil {\n\t\t\t\t\tdone <- err\n\t\t\t\t\treturn\n\t\t\t\t}\n\t\t\tcase redis.Subscription:\n\t\t\t\tswitch n.Count {\n\t\t\t\tcase len(channels):\n\t\t\t\t\t// Notify application when all channels are subscribed.\n\t\t\t\t\tif err := onStart(); err != nil {\n\t\t\t\t\t\tdone <- err\n\t\t\t\t\t\treturn\n\t\t\t\t\t}\n\t\t\t\tcase 0:\n\t\t\t\t\t// Return from the goroutine when all channels are unsubscribed.\n\t\t\t\t\tdone <- nil\n\t\t\t\t\treturn\n\t\t\t\t}\n\t\t\t}\n\t\t}\n\t}()\n\n\tticker := time.NewTicker(healthCheckPeriod)\n\tdefer ticker.Stop()\nloop:\n\tfor {\n\t\tselect {\n\t\tcase <-ticker.C:\n\t\t\t// Send ping to test health of connection and server. If\n\t\t\t// corresponding pong is not received, then receive on the\n\t\t\t// connection will timeout and the receive goroutine will exit.\n\t\t\tif err = psc.Ping(\"\"); err != nil {\n\t\t\t\tbreak loop\n\t\t\t}\n\t\tcase <-ctx.Done():\n\t\t\tbreak loop\n\t\tcase err := <-done:\n\t\t\t// Return error from the receive goroutine.\n\t\t\treturn err\n\t\t}\n\t}\n\n\t// Signal the receiving goroutine to exit by unsubscribing from all channels.\n\tif err := psc.Unsubscribe(); err != nil {\n\t\treturn err\n\t}\n\n\t// Wait for goroutine to complete.\n\treturn <-done\n}\n\nfunc publish() {\n\tc, err := dial()\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tdefer c.Close()\n\n\tif _, err = c.Do(\"PUBLISH\", \"c1\", \"hello\"); err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tif _, err = c.Do(\"PUBLISH\", \"c2\", \"world\"); err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\tif _, err = c.Do(\"PUBLISH\", \"c1\", \"goodbye\"); err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n}\n\n// This example shows how receive pubsub notifications with cancelation and\n// health checks.\nfunc main() {\n\tredisServerAddr, err := serverAddr()\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\n\tctx, cancel := context.WithCancel(context.Background())\n\n\terr = listenPubSubChannels(ctx,\n\t\tredisServerAddr,\n\t\tfunc() error {\n\t\t\t// The start callback is a good place to backfill missed\n\t\t\t// notifications. For the purpose of this example, a goroutine is\n\t\t\t// started to send notifications.\n\t\t\tgo publish()\n\t\t\treturn nil\n\t\t},\n\t\tfunc(channel string, message []byte) error {\n\t\t\tfmt.Printf(\"channel: %s, message: %s\\n\", channel, message)\n\n\t\t\t// For the purpose of this example, cancel the listener's context\n\t\t\t// after receiving last message sent by publish().\n\t\t\tif string(message) == \"goodbye\" {\n\t\t\t\tcancel()\n\t\t\t}\n\t\t\treturn nil\n\t\t},\n\t\t\"c1\", \"c2\")\n\n\tif err != nil {\n\t\tfmt.Println(err)\n\t\treturn\n\t}\n\n}\n```\n```\nOutput:\n\nchannel: c1, message: hello channel: c2, message: world channel: c1, message: goodbye\n```\nShare Format\nRun\n\n#### \nfunc (PubSubConn) [Close](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pubsub.go#L58) [¶](#PubSubConn.Close)\n```\nfunc (c [PubSubConn](#PubSubConn)) Close() [error](/builtin#error)\n```\nClose closes the connection.\n\n#### \nfunc (PubSubConn) [PSubscribe](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pubsub.go#L71) [¶](#PubSubConn.PSubscribe)\n```\nfunc (c [PubSubConn](#PubSubConn)) PSubscribe(channel ...interface{}) [error](/builtin#error)\n```\nPSubscribe subscribes the connection to the given patterns.\n\n#### \nfunc (PubSubConn) [PUnsubscribe](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pubsub.go#L89) [¶](#PubSubConn.PUnsubscribe)\n```\nfunc (c [PubSubConn](#PubSubConn)) PUnsubscribe(channel ...interface{}) [error](/builtin#error)\n```\nPUnsubscribe unsubscribes the connection from the given patterns, or from all of them if none is given.\n\n#### \nfunc (PubSubConn) [Ping](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pubsub.go#L100) [¶](#PubSubConn.Ping)\n```\nfunc (c [PubSubConn](#PubSubConn)) Ping(data [string](/builtin#string)) [error](/builtin#error)\n```\nPing sends a PING to the server with the specified data.\n\nThe connection must be subscribed to at least one channel or pattern when calling this method.\n\n#### \nfunc (PubSubConn) [Receive](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pubsub.go#L110) [¶](#PubSubConn.Receive)\n```\nfunc (c [PubSubConn](#PubSubConn)) Receive() interface{}\n```\nReceive returns a pushed message as a Subscription, Message, Pong or error.\nThe return value is intended to be used directly in a type switch as illustrated in the PubSubConn example.\n\n#### \nfunc (PubSubConn) [ReceiveContext](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pubsub.go#L123) [¶](#PubSubConn.ReceiveContext)\n\nadded in v1.8.9\n```\nfunc (c [PubSubConn](#PubSubConn)) ReceiveContext(ctx [context](/context).[Context](/context#Context)) interface{}\n```\nReceiveContext is like Receive, but it allows termination of the receive via a Context. If the call returns due to closure of the context's Done channel the underlying Conn will have been closed.\n\n#### \nfunc (PubSubConn) [ReceiveWithTimeout](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pubsub.go#L116) [¶](#PubSubConn.ReceiveWithTimeout)\n```\nfunc (c [PubSubConn](#PubSubConn)) ReceiveWithTimeout(timeout [time](/time).[Duration](/time#Duration)) interface{}\n```\nReceiveWithTimeout is like Receive, but it allows the application to override the connection's default timeout.\n\n#### \nfunc (PubSubConn) [Subscribe](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pubsub.go#L63) [¶](#PubSubConn.Subscribe)\n```\nfunc (c [PubSubConn](#PubSubConn)) Subscribe(channel ...interface{}) [error](/builtin#error)\n```\nSubscribe subscribes the connection to the specified channels.\n\n#### \nfunc (PubSubConn) [Unsubscribe](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pubsub.go#L80) [¶](#PubSubConn.Unsubscribe)\n```\nfunc (c [PubSubConn](#PubSubConn)) Unsubscribe(channel ...interface{}) [error](/builtin#error)\n```\nUnsubscribe unsubscribes the connection from the given channels, or from all of them if none is given.\n\n#### \ntype [Scanner](https://github.com/gomodule/redigo/blob/v1.8.9/redis/redis.go#L61) [¶](#Scanner)\n```\ntype Scanner interface {\n // RedisScan assigns a value from a Redis value. The argument src is one of\n\t// the reply types listed in the section `Executing Commands`.\n\t//\n\t// An error should be returned if the value cannot be stored without\n\t// loss of information.\n\tRedisScan(src interface{}) [error](/builtin#error)\n}\n```\nScanner is implemented by an object which wants to control its value is interpreted when read from Redis.\n\n#### \ntype [Script](https://github.com/gomodule/redigo/blob/v1.8.9/redis/script.go#L27) [¶](#Script)\n```\ntype Script struct {\n\t// contains filtered or unexported fields\n}\n```\nScript encapsulates the source, hash and key count for a Lua script. See\n<http://redis.io/commands/eval> for information on scripts in Redis.\n\nExample [¶](#example-Script)\n```\npackage main\n\nimport (\n\t\"github.com/gomodule/redigo/redis\"\n)\n\nvar (\n\tc redis.Conn\n\treply interface{}\n\terr error\n)\n\nfunc main() {\n\t// Initialize a package-level variable with a script.\n\tvar getScript = redis.NewScript(1, `return redis.call('get', KEYS[1])`)\n\n\t// In a function, use the script Do method to evaluate the script. The Do\n\t// method optimistically uses the EVALSHA command. If the script is not\n\t// loaded, then the Do method falls back to the EVAL command.\n\treply, err = getScript.Do(c, \"foo\")\n}\n```\n```\nOutput:\n```\nShare Format\nRun\n\n#### \nfunc [NewScript](https://github.com/gomodule/redigo/blob/v1.8.9/redis/script.go#L38) [¶](#NewScript)\n```\nfunc NewScript(keyCount [int](/builtin#int), src [string](/builtin#string)) *[Script](#Script)\n```\nNewScript returns a new script object. If keyCount is greater than or equal to zero, then the count is automatically inserted in the EVAL command argument list. If keyCount is less than zero, then the application supplies the count as the first value in the keysAndArgs argument to the Do, Send and SendHash methods.\n\n#### \nfunc (*Script) [Do](https://github.com/gomodule/redigo/blob/v1.8.9/redis/script.go#L80) [¶](#Script.Do)\n```\nfunc (s *[Script](#Script)) Do(c [Conn](#Conn), keysAndArgs ...interface{}) (interface{}, [error](/builtin#error))\n```\nDo evaluates the script. Under the covers, Do optimistically evaluates the script using the EVALSHA command. If the command fails because the script is not loaded, then Do evaluates the script using the EVAL command (thus causing the script to load).\n\n#### \nfunc (*Script) [DoContext](https://github.com/gomodule/redigo/blob/v1.8.9/redis/script.go#L64) [¶](#Script.DoContext)\n\nadded in v1.8.6\n```\nfunc (s *[Script](#Script)) DoContext(ctx [context](/context).[Context](/context#Context), c [Conn](#Conn), keysAndArgs ...interface{}) (interface{}, [error](/builtin#error))\n```\n#### \nfunc (*Script) [Hash](https://github.com/gomodule/redigo/blob/v1.8.9/redis/script.go#L60) [¶](#Script.Hash)\n```\nfunc (s *[Script](#Script)) Hash() [string](/builtin#string)\n```\nHash returns the script hash.\n\n#### \nfunc (*Script) [Load](https://github.com/gomodule/redigo/blob/v1.8.9/redis/script.go#L101) [¶](#Script.Load)\n```\nfunc (s *[Script](#Script)) Load(c [Conn](#Conn)) [error](/builtin#error)\n```\nLoad loads the script without evaluating it.\n\n#### \nfunc (*Script) [Send](https://github.com/gomodule/redigo/blob/v1.8.9/redis/script.go#L96) [¶](#Script.Send)\n```\nfunc (s *[Script](#Script)) Send(c [Conn](#Conn), keysAndArgs ...interface{}) [error](/builtin#error)\n```\nSend evaluates the script without waiting for the reply.\n\n#### \nfunc (*Script) [SendHash](https://github.com/gomodule/redigo/blob/v1.8.9/redis/script.go#L91) [¶](#Script.SendHash)\n```\nfunc (s *[Script](#Script)) SendHash(c [Conn](#Conn), keysAndArgs ...interface{}) [error](/builtin#error)\n```\nSendHash evaluates the script without waiting for the reply. The script is evaluated with the EVALSHA command. The application must ensure that the script is loaded by a previous call to Send, Do or Load methods.\n\n#### \ntype [SlowLog](https://github.com/gomodule/redigo/blob/v1.8.9/redis/redis.go#L171) [¶](#SlowLog)\n\nadded in v1.7.2\n```\ntype SlowLog struct {\n // ID is a unique progressive identifier for every slow log entry.\n\tID [int64](/builtin#int64)\n\n // Time is the unix timestamp at which the logged command was processed.\n\tTime [time](/time).[Time](/time#Time)\n\n // ExecutationTime is the amount of time needed for the command execution.\n\tExecutionTime [time](/time).[Duration](/time#Duration)\n\n // Args is the command name and arguments\n\tArgs [][string](/builtin#string)\n\n // ClientAddr is the client IP address (4.0 only).\n\tClientAddr [string](/builtin#string)\n\n // ClientName is the name set via the CLIENT SETNAME command (4.0 only).\n\tClientName [string](/builtin#string)\n}\n```\nSlowLog represents a redis SlowLog\n\n#### \nfunc [SlowLogs](https://github.com/gomodule/redigo/blob/v1.8.9/redis/reply.go#L593) [¶](#SlowLogs)\n\nadded in v1.7.2\n```\nfunc SlowLogs(result interface{}, err [error](/builtin#error)) ([][SlowLog](#SlowLog), [error](/builtin#error))\n```\nSlowLogs is a helper that parse the SLOWLOG GET command output and return the array of SlowLog\n\n#### \ntype [Subscription](https://github.com/gomodule/redigo/blob/v1.8.9/redis/pubsub.go#L24) [¶](#Subscription)\n```\ntype Subscription struct {\n // Kind is \"subscribe\", \"unsubscribe\", \"psubscribe\" or \"punsubscribe\"\n\tKind [string](/builtin#string)\n\n // The channel that was changed.\n\tChannel [string](/builtin#string)\n\n // The current number of subscriptions for connection.\n\tCount [int](/builtin#int)\n}\n```\nSubscription represents a subscribe or unsubscribe notification."}}},{"rowIdx":59179,"cells":{"project":{"kind":"string","value":"rainbow-code"},"source":{"kind":"string","value":"npm"},"language":{"kind":"string","value":"JavaScript"},"content":{"kind":"string","value":"Rainbow\n===\n\nRainbow is a code syntax highlighting library written in Javascript.\n\nIt was designed to be lightweight (~2.5kb), easy to use, and extendable.\n\nIt is completely themable via CSS.\n\nDemo\n---\n\nYou can see rainbow in action at <http://rainbowco.de>.\n\nYou can also build/download custom packages from there.\n\nContents\n---\n\n* [Quick Start](#quick-start)\n\t+ [Browser](#browser)\n\t+ [Node.js](#nodejs)\n\t\t- [Install rainbow](#install-rainbow)\n\t\t- [Highlight some code](#highlight-some-code)\n\t+ [Supported Browsers](#supported-browsers)\n\t+ [Supported Languages](#supported-languages)\n\t+ [Specifying a language](#specifying-a-language)\n\t+ [Themes](#themes)\n\t\t- [Rendering code blocks](#rendering-code-blocks)\n\t\t- [Adding custom rules for specific languages](#adding-custom-rules-for-specific-languages)\n\t+ [JavaScript API Documentation](#javascript-api-documentation)\n\t\t- [Rainbow.color](#rainbowcolor)\n\t\t\t* [Preventing automatic highlighting on page load](#preventing-automatic-highlighting-on-page-load)\n\t\t\t* [Extra options for color](#extra-options-for-color)\n\t\t\t\t+ [globalClass](#globalclass)\n\t\t- [Rainbow.extend](#rainbowextend)\n\t\t\t* [Extending existing languages](#extending-existing-languages)\n\t\t\t* [How code is highlighted](#how-code-is-highlighted)\n\t\t\t\t+ [Match by name](#match-by-name)\n\t\t\t\t+ [Match by group](#match-by-group)\n\t\t\t\t+ [Match by array of sub-patterns](#match-by-array-of-sub-patterns)\n\t\t\t\t+ [Match using another language](#match-using-another-language)\n\t\t\t* [Extending an existing language](#extending-an-existing-language)\n\t\t\t* [How Rainbow chooses a match](#how-rainbow-chooses-a-match)\n\t\t\t* [Known limitations](#known-limitations)\n\t\t\t\t+ [Regular expressions lookbehind assertions](#regular-expressions-lookbehind-assertions)\n\t\t\t\t+ [Regular expression subgroup matches](#regular-expression-subgroup-matches)\n\t\t- [Rainbow.addAlias](#rainbowaddalias)\n\t\t- [Rainbow.onHighlight](#rainbowonhighlight)\n\t\t- [Rainbow.remove](#rainbowremove)\n\t+ [Building](#building)\n\t\t- [Getting a local environment set up](#getting-a-local-environment-set-up)\n\t\t- [Build commands](#build-commands)\n\t\t\t* [gulp build](#gulp-build)\n\t\t\t* [gulp lint](#gulp-lint)\n\t\t\t* [gulp pack](#gulp-pack)\n\t\t\t* [gulp sass](#gulp-sass)\n\t\t\t* [gulp test](#gulp-test)\n\t\t\t* [gulp watch](#gulp-watch)\n\t+ [More Info](#more-info)\n\nQuick Start\n---\n\n### Browser\n\n1. Include some markup for code you want to be highlighted:\n```\n<pre><code data-language=\"python\">def openFile(path): file = open(path, \"r\") content = file.read() file.close() return content</code></pre>\n```\n2. Include a CSS theme file in the `<head>`:\n```\n<link href=\"/assets/css/theme.css\" rel=\"stylesheet\" type=\"text/css\">\n```\n3. Include rainbow.js and whatever languages you want before the closing `</body>`:\n```\n<script src=\"/assets/js/rainbow.js\"></script><script src=\"/assets/js/language/generic.js\"></script><script src=\"/assets/js/language/python.js\"></script>\n```\nBy default `dist/rainbow.min.js` comes with some popular languages bundled together with it.\n\n### Node.js\n\nRainbow 2.0 introduced support for node.js. All of the existing API methods should work, but there is also a new `Rainbow.colorSync` method for synchronous highlighting.\n\n#### Install rainbow\n```\nnpm install --save rainbow-code\n```\n#### Highlight some code\n```\nvar rainbow = require('rainbow-code');var highlighted = rainbow.colorSync('// So meta\\nrainbow.colorSync(\\'var helloWorld = true;\\');', 'javascript');console.log(highlighted);\n```\nSupported Browsers\n---\n\nRainbow 2.0 should work in the following browsers:\n\n| Chrome | Firefox | IE | Safari |\n| --- | --- | --- | --- |\n| 20+ | 13+ | 10+ | 6+ |\n\nFor older browsers you can download the legacy [1.2.0](https://github.com/ccampbell/rainbow/archive/1.2.0.zip) release.\n\nSupported Languages\n---\n\nCurrently supported languages are:\n\n* C\n* C#\n* Coffeescript\n* CSS\n* D\n* Go\n* Haskell\n* HTML\n* Java\n* JavaScript\n* JSON\n* Lua\n* PHP\n* Python\n* R\n* Ruby\n* Scheme\n* Shell\n* Smalltalk\n\nSpecifying a language\n---\n\nIn your markup the `data-language` attribute is used to specify what language to use for highlighting. For example:\n```\n<pre><code data-language=\"javascript\">var testing = true;</code></pre>\n```\nRainbow also supports the HTML5 style for specifying languages:\n```\n<pre><code class=\"language-javascript\">var testing = true;</code></pre>\n```\nAnd the Google prettify style:\n```\n<pre class=\"lang-javascript\"><code>var testing = true;</code></pre>\n```\nThemes\n---\n\nThemes are located in the `themes/sass` directory. They are written using sass so that common logic can be shared among all themes without having to duplicate it in each theme file. You should not edit the css files directly.\n\n`_base.sass` includes some default styles shared by almost every theme. `_init.sass` contains mixins and initialization logic that is shared by every theme.\n\n### Rendering code blocks\n\nAs of version 2.0 the themes use a clever trick to display the highlighted code. All code blocks default to `opacity: 0`, but an animation is triggered on page load to happen after a 2 second delay.\n\nThis means for users who do not have JavaScript enabled the code will fade in after 2 seconds. If JavaScript is enabled, the animation is stopped on load and the delay is reset to `0s`. That ensures that as soon as the code is done being highlighted it will be able to show up instantly. This is used to prevent a flash of unstyled text on page load and ensure that the code blocks only show up after they have been highlighted.\n\nThere is also a preload animation that will show up for any code block that takes longer than 300ms to load.\n\n### Adding custom rules for specific languages\n\nA SASS mixin was added to simplify defining styles that should only apply for a specific language. Using it looks like this:\n```\n@include language(\"html\") .support.operator color: #fff @include language((\"javascript\", \"js\")) .variable.super color: #66D9EF\n```\nYou can pass a single language or a list of languages.\n\nJavaScript API Documentation\n---\n\nRainbow has four public methods:\n\n* [Rainbow.color](#rainbowcolor)\n* [Rainbow.extend](#rainbowextend)\n* [Rainbow.addAlias](#rainbowaddalias)\n* [Rainbow.onHighlight](#rainbowonhighlight)\n* [Rainbow.remove](#rainbowremove)\n\n### Rainbow.color\n\nRainbow.color is used to highlight blocks of code.\n\nFor convenience, this method is called automatically to highlight all code blocks on the page when `DOMContentLoaded` fires. If you would like to highlight stuff that is not in the DOM you can use it on its own. There are three ways to use it.\n\n1. The first option is calling the color method on its own:\n```\nRainbow.color();\n```\nEach time this is called, Rainbow will look for matching `pre` blocks on the page that have not yet been highlighted and highlight them.\n\nYou can optionally pass a callback function that will fire when all the blocks have been highlighted.\n```\nRainbow.color(function() { console.log('The new blocks are now highlighted!');});\n```\n2. The second option is passing a specific element to the color method.\n\nIn this example we are creating a code block, highlighting it, then inserting it into the DOM:\n```\nvar div = document.createElement('div');div.innerHTML = '<pre><code data-language=\"javascript\">var foo = true;</code></pre>';Rainbow.color(div, function() { document.getElementById('something-else').appendChild(div;)});\n```\n3. The final option is passing in your code as a string to `Rainbow.color`.\n```\nRainbow.color('var foo = true;', 'javascript', function(highlightedCode) { console.log(highlightedCode);});\n```\n#### Preventing automatic highlighting on page load\n\nIf you want to prevent code on the page from being highlighted when the page loads you can set the `defer` property to `true`.\n```\nRainbow.defer = true;\n```\nNote that you have to set this before `DOMContentLoaded` fires or else it will not do anything.\n\n#### Extra options for color\n\nAs of right now there is one extra option for color.\n\n##### globalClass\n\nThis option allows you to have an extra class added to every span that Rainbow renders. This can be useful if you want to remove the classes in order to trigger a special effect of some sort.\n\nTo apply a global class you can add it in your markup:\n```\n<pre><code data-language=\"javascript\" data-global-class=\"animate\">var hello = true;</code></pre>\n```\nOr you can pass it into a `Rainbow.color` call like this:\n```\nRainbow.color('var hello = true;', { language: 'javascript', globalClass: 'animate'});\n```\n### Rainbow.extend\n\nRainbow.extend is used to define language grammars which are used to highlight the code you pass in. It can be used to define new languages or to extend existing languages.\n\nA very simple language grammer looks something like this:\n```\nRainbow.extend('example', [ { name: 'keyword', pattern: /function|return|continue|break/g }]);\n```\nAny pattern used with extend will take precedence over an existing pattern that matches the same block. It will also take precedence over any pattern that is included as part of the generic patterns.\n\nFor example if you were to call\n```\nRainbow.extend('example', [ { name: 'keyword.magic', pattern: /function/g }]);\n```\nThis would mean that function will be highlighted as `<span class=\"keyword magic\">function</span>`, but `return`, `continue`, and `break` will still be highlighted as just `<span class=\"keyword\">return</span>`, etc.\n\n#### Extending existing languages\n\nBy default languages are considered to be standalone, but if you specify an optional third parameter you can have your language inherit from another one.\n\nFor example the python language grammars inherit from the generic ones:\n```\nRainbow.extend('python', [ { name: 'constant.language', pattern: /True|False|None/g }], 'generic');\n```\nIf you wanted to remove the default boolean values you should be able to do something like this:\n```\nRainbow.extend('python', [ { name: '', pattern: /true|false/g }]);\n```\n#### How code is highlighted\n\nThe `name` value determines what classes will be added to a span that matches the pattern you specify. For example, if you name a pattern `constant.hex-color` the code that matches that pattern will be wrapped in `<span class=\"constant hex-color></span>`. This allows you to share common base styles but add more customization in your themes for more specific matches.\n\n#### How grammars are defined\n\n##### Match by name\n\nThe simplest way to define a grammar is to define a regular expression and a name to go along with that.\n```\nRainbow.extend([ { name: 'constant.boolean', pattern: /true|false/g }]);\n```\n##### Match by group\n\nThis allows you to take a more complicated regular expression and map certain parts of it to specific scopes.\n```\nRainbow.extend([ { matches: { 1: 'constant.boolean.true', 2: 'constant.boolean.false' }, pattern: /(true)|(false)/g }]);\n```\n##### Match by array of sub-patterns\n\nFor more complicated matches you may want to process code within another match group.\n```\nRainbow.extend([ { matches: [ { name: 'constant.boolean.true', pattern: /true/ }, { name: 'constant.boolean.false', pattern: /false/ } ], pattern: /true|false/g }]);\n```\n##### Match using another language\n\nSometimes a language supports other languages being embedded inside of it such as JavaScript inside HTML. Rainbow supports that out of the box. Here is an example of how you would highlight PHP tags inside of HTML code.\n```\nRainbow.extend('html', [ { name: 'source.php.embedded', matches: { 2: { language: 'php' } }, pattern: /<\\?(php)?([\\s\\S]*?)(\\?>)/gm }]);\n```\nYou should be able to nest sub-patterns as many levels deep as you would like.\n\n#### Extending an existing language\n\nIf you have a language specific pattern that you want highlighted, but it does not exist in the language syntax rules, you can add a rule in your own javascript.\n\nLet's say for example you want to highlight PHP's apc functions.\nYou can include the php language then in the markup on your page add:\n```\n<script> Rainbow.extend('php', [ { 'matches': { 1: 'support.function' }, 'pattern': /\\b(apc_(store|fetch|add|inc))(?=\\()/g } ]);</script```\n#### How Rainbow chooses a match\n\nIn general the best practice is to make your patterns as specific as possible (for example targetting specific keywords).\n\nWhen you create a new language it gets pushed to the front of whatever language it is inheriting from. This means whatever rules are added last will be checked against first.\n\nRainbow chooses the first match it finds for a block. If another pattern overlaps with a pattern that has already been chosen then it is ignored.\n\nThere is one exception to this. If a match that comes later is more specific (the start AND end points stretch beyond another pattern already matched) then the new match will take precedence and the old one will be discarded.\n\nThat means if you have a pattern that matches `function test()` and you add a pattern that matches `public function test()` the new one will be used instead.\n\n#### Known limitations\n\n##### Regular expressions lookbehind assertions\n\nJavascript does not allow positive or negative lookbehind assertions so this means it is possible to have conflicts with the starting and end positions of matches. If you want to match a pattern that ends with a specific character it is recommended that you use a positive lookahead for that character instead of including it in the regex. That allows the same character to be used at the start of another match group without overlapping.\n\n##### Regular expression subgroup matches\n\nYou cannot match part of a subgroup directly to a scope. For example if you have:\n```\n{ matches: { 1: 'name.attribute', 2: 'value' }, pattern: /(name=\\\"(.*?)\\\")/g}\n```\nThis will result in code `name=\"value\"` being highlighted as:\n```\n<span class=\"name attribute\">name=\"value\"</span>\n```\nYou see the value class never gets applied.\n\nTo achieve what you really want you would have to use a subpattern like this:\n```\n{ name: 'name.attribute', matches: [{ matches: { 1: 'value' }, pattern: /\\\"(.*?)\\\"/g }], pattern: /(name=\\\"(.*?)\\\")/g}\n```\nThis means the entire block is wrapped with `name.attribute` scope and then within that any part in double quotes will be wrapped as `value`.\n\nThat means the entire block will behighlighted as\n```\n<span class=\"name attribute\">name=\"<span class=\"value\">value</span>\"</span>\n```\nIn this example you could avoid subpatterns completely by using a regex like this to begin with:\n```\n/(name=)\\\"(.*?)\\\"/g\n```\n### Rainbow.addAlias\n\nThe addAlias function allows you to map a different name to a language. For example:\n```\nRainbow.addAlias('js', 'javascript');\n```\nThis allows you to highlight javascript code by using the language `js` instead of `javascript`.\n\n### Rainbow.onHighlight\n\nThis method notifies you as soon as a block of code has been highlighted.\n```\nRainbow.onHighlight(function(block, language) { console.log(block, 'for language', language, 'was highlighted');});\n```\nThe first parameter returns a reference to that code block in the DOM. The second parameter returns a string of the language being highlighted.\n\n### Rainbow.remove\n\nThis method allows you to remove syntax rules for a specific language. It is only really useful for development if you want to be able to reload language grammars on the fly without having to refresh the page.\n```\n// Remove all the javascript patternsRainbow.remove('javascript');\n```\nBuilding\n---\n\nRainbow is compiled using [rollup](https://github.com/rollup/rollup) and [buble](https://gitlab.com/Rich-Harris/buble).\n\n[Gulp](https://github.com/gulpjs/gulp) is used for all build related tasks.\n\n### Getting a local environment set up\n```\ngit clone git@github.com:ccampbell/rainbow.gitcd rainbownpm install\n```\n### Build commands\n\n#### gulp build\n\nThe build command is used to build a custom version of rainbow.js. If you run\n```\ngulp build\n```\nA file will be created at `dist/rainbow-custom.min.js` containing rainbow.js as well as popular languages. If you want to specify specific languages you can use:\n```\ngulp build --languages=html,css,php,javascript\n```\nIf you want a minimized version of rainbow without any languages you can pass\n```\ngulp build --languages=none\n```\nIf you want a minimized version with all languages you can pass\n```\ngulp build --languages=all\n```\n#### gulp lint\n\nThe lint command will check all the javascript files for things that do not match the styleguide from the `.eslintrc` file.\n\n#### gulp pack\n\nThe pack command will run a buble + rollup build and save the resulting file to `dist/rainbow.js`\n\n#### gulp sass\n\nThe sass command will compile all the rainbow themes\n\n#### gulp test\n\nThe test command will run the unit tests. You can pass the `--watch` flag to keep the tests running and have them rerun when you make changes. That is a little buggy though which has something to do with karma + rollup.\n\nYou can also use the `--browsers` flag to specify a browser to run the tests in. Currently only `PhantomJS` and `Chrome` are supported.\n\n#### gulp watch\n\nThe watch command will look at sass files and src js files and build the css or js (using `gulp sass` and `gulp pack`) every time you make a change.\n\nMore Info\n---\n\nIf you are looking for line number support you can try one of the following:\n\n* <https://github.com/Blender3D/rainbow.linenumbers.js>\n* <https://github.com/Sjeiti/rainbow.linenumbersYou can check out demos and build custom packages at [rainbowco.de](http://rainbowco.de).\nReadme\n---\n\n### Keywords\n\n* code\n* highlight\n* highlighter"}}},{"rowIdx":59180,"cells":{"project":{"kind":"string","value":"rebird"},"source":{"kind":"string","value":"cran"},"language":{"kind":"string","value":"R"},"content":{"kind":"string","value":"Package ‘rebird’\n October 14, 2022\nVersion 1.3.0\nTitle R Client for the eBird Database of Bird Observations\nDescription A programmatic client for the eBird database\n (<https://ebird.org/home>), including functions for searching for bird\n observations by geographic location (latitude, longitude), eBird\n hotspots, location identifiers, by notable sightings, by region, and by\n taxonomic name.\nDepends R (>= 2.10)\nLicense MIT + file LICENSE\nURL https://docs.ropensci.org/rebird/,\n https://github.com/ropensci/rebird\nBugReports https://github.com/ropensci/rebird/issues\nVignetteBuilder knitr\nImports methods, utils, stats, httr (>= 1.0.0), jsonlite, dplyr,\n assertthat\nSuggests knitr, rmarkdown, testthat, covr\nRoxygenNote 7.1.1\nX-schema.org-applicationCategory Data Access\nX-schema.org-keywords birds, birding, ebird, database, data, biology,\n observations, sightings, ornithology\nX-schema.org-isPartOf https://ropensci.org\nEncoding UTF-8\nNeedsCompilation no\nAuthor <NAME> [aut],\n <NAME> [aut] (<https://orcid.org/0000-0003-1444-9135>),\n <NAME> [aut],\n <NAME> [ctb],\n <NAME> [ctb],\n <NAME> [ctb],\n <NAME> [aut, cre] (<https://orcid.org/0000-0002-4147-5796>)\nMaintainer <NAME> <<EMAIL>>\nRepository CRAN\nDate/Publication 2021-09-20 07:00:02 UTC\nR topics documented:\nebirdchecklistfee... 2\nebirdfre... 4\nebirdge... 5\nebirdhistorica... 8\nebirdhotspo... 10\nebirdhotspotlis... 12\nebirdlo... 13\nebirdnotabl... 16\nebirdregio... 18\nebirdregionchec... 20\nebirdregioninf... 21\nebirdregionspecie... 22\nebirdsubregionlis... 23\nebirdtaxonom... 25\ngetlatln... 26\nnearestob... 27\nrebird-deprecate... 29\nspecies_cod... 29\n ebirdchecklistfeed Checklist feed on a date at a region or hotspot\nDescription\n Returns checklist-level information reported in a given region or hotspot. Note only bird informa-\n tion is species count.\nUsage\n ebirdchecklistfeed(loc, date, max = 10, sleep = 0, key = NULL, ...)\nArguments\n loc (required) Region code or locID (if a hotspot). Region code can be country\n code (e.g. \"US\"), subnational1 code (states/provinces, e.g. \"US-NV\"), or sub-\n national2 code (counties, e.g. \"US-VA-003\").\n date (required) Date of historic observation date formatted according to ISO 8601\n (e.g. ’YYYY-MM-DD’, or ’YYYY-MM-DD hh:mm’). Hours and minutes are\n excluded.\n max Maximum number of result rows to return in this request (between 1 and 200,\n default 10)\n sleep Time (in seconds) before function sends API call. The defaults is zero. Set this\n to a higher number if you are using this function in a loop with many API calls.\n key eBird API key. You can obtain one from https://ebird.org/api/keygen. We strongly\n recommend storing it in your .Renviron file as an environment variable called\n EBIRD_KEY.\n ... Curl options passed on to GET.\nValue\n A data.frame containing the collected information:\n \"locId\": unique identifier for the locations\n \"subId\": submission (checklist) identifier\n \"userDisplayName\": first and last name of the observer\n \"numSpecies\": number of species reported\n \"obsDt\": observation date formatted according to ISO 8601 (e.g. ’YYYY-MM-DD’, or ’YYYY-\n MM-DD hh:mm’). Hours and minutes are excluded if the observer did not report an observation\n time\n \"obsTime\": observation time (24hr)\n \"subID\": deprecated submission identifier\n \"loc\": delimited string of location descriptors\nAuthor(s)\n <NAME> <<EMAIL>>\nReferences\n http://ebird.org/\nExamples\n ## Not run:\n ebirdchecklistfeed(loc = \"L207391\", date = \"2020-03-24\", max = 10)\n ## End(Not run)\n ebirdfreq Download historical frequencies of bird observations from eBird\nDescription\n NOTE: Currently disabled.\nUsage\n ebirdfreq(\n loctype,\n loc,\n startyear = 1900,\n endyear = format(Sys.Date(), \"%Y\"),\n startmonth = 1,\n endmonth = 12,\n long = TRUE,\n ...\n )\nArguments\n loctype String with location type. Either \"states\", \"counties\", or \"hotspots\".\n loc String with location identifier. If querying states or provinces, the two letter\n country code followed by the two letter state code and separated by \"-\" (e.g.\n \"US-NY\"). If querying counties, is as in states/provinces, but appending county\n identifier after a dash. For counties in the US, the county codes is a 3-digit\n number specific to each state (e.g. Bronx County: \"US-NY-005\"). For coun-\n ties in Canada, county codes are two-letter identifiers (e.g. Metro Vancouver:\n \"CA-BC-GV\"). If querying hotspots then the unique identifier is a 6-digit num-\n ber prepended with an \"L\" (e.g. \"L196159\"). All these codes can be found\n by looking at the URL in each respective location/hotspot webpage (which are\n accessible through the \"Explore Data\" tab).\n startyear Starting year for query. Defaults to 1900.\n endyear Ending year for query. Defaults to current year specified by Sys.Date().\n startmonth Starting month for query as an integer (1-12). Defaults to January.\n endmonth Ending month for query as an integer (1-12). Defaults to December.\n long Logical, Should output be in long format? Defaults to TRUE. If FALSE then\n output will be in wide format.\n ... Curl options passed on to GET\nDetails\n This function was the only rebird function to not use the API and formulated a url-based query\n instead. Now you need to be logged into eBird to download the frequency data, but we can’t\n authenticate through R, so this function does not work. This functionality is likely to be added to\n the API in the future, so we are keeping the function in the meantime, but it throws an informative\n error, and provides the constructed url to obtain the frequency data manually through your browser.\nValue\n This function currently returns an error, but also provides the constructed url to manually obtain\n the data for the location and dates requested through your browser.\n A data frame containing the collected information. If in long format:\n \"monthQt\": month and week (eBird data divides each month by four weeks)\n \"comName\": species common name\n \"frequency\": proportion of times the species was seen in a specified week\n \"sampleSize\" number of complete eBird checklists submitted for specified given week @return If\n in wide format, then first column is the species list and all other columns are of individual weeks\n (four in each month). First row contains the number of complete checklists for each week.\nAuthor(s)\n <NAME> <<EMAIL>>, <NAME> <<EMAIL>>\nReferences\n http://ebird.org/\nExamples\n ## Not run:\n ebirdfreq(\"states\", \"US-NY\", 2014, 2014, 1, 12)\n ebirdfreq(\"counties\", \"CA-BC-GV\", 1900, 2015, 1, 3)\n ebirdfreq(\"hotspots\", \"L196159\", long=FALSE)\n ## End(Not run)\n ebirdgeo Sightings at location determined by latitude/longitude\nDescription\n Returns the most recent sighting date and specific location for the requested species of bird reported\n within the number of days specified and reported in the specified area.\nUsage\n ebirdgeo(\n species = NULL,\n lat = NULL,\n lng = NULL,\n dist = NULL,\n back = NULL,\n max = NULL,\n locale = NULL,\n provisional = FALSE,\n hotspot = FALSE,\n sleep = 0,\n key = NULL,\n ...\n )\nArguments\n species Species code of the species of interest. Scientific names can be specified if\n wrapped around the species_code function. Defaults to NULL, so sightings\n for all species are returned. See eBird taxonomy for more information: https:\n //ebird.org/science/use-ebird-data/the-ebird-taxonomy.\n lat Decimal latitude. value between -90.00 and 90.00, up to two decimal places of\n precision. Defaults to latitude based on IP.\n lng Decimal longitude. value between -180.00 and 180.00, up to two decimal places\n of precision. Defaults to longitude based on IP.\n dist Distance defining radius of interest from given lat/lng in kilometers (between 0\n and 50, defaults to 25).\n back Number of days back to look for observations (between 1 and 30, defaults to\n 14).\n max Maximum number of result rows to return in this request (between 1 and 10000,\n defaults to all).\n locale Language/locale of response (when translations are available). See https://docs.oracle.com/javase/6/docs/\n (defaults to en_US).\n provisional Should flagged records that have not been reviewed be included? (defaults to\n FALSE).\n hotspot Should results be limited to sightings at birding hotspots? (defaults to FALSE).\n sleep Time (in seconds) before function sends API call (defaults to zero. Set to higher\n number if you are using this function in a loop with many API calls).\n key eBird API key. You can obtain one from https://ebird.org/api/keygen. We strongly\n recommend storing it in your .Renviron file as an environment variable called\n EBIRD_KEY.\n ... Curl options passed on to GET\nValue\n A data.frame containing the collected information:\n \"comName\": species common name\n \"howMany\": number of individuals observed, NA if only presence was noted\n \"lat\": latitude of the location\n \"lng\": longitude of the location\n \"locID\": unique identifier for the location\n \"locName\": location name\n \"locationPrivate\": TRUE if location is not a birding hotspot\n \"obsDt\": observation date formatted according to ISO 8601 (e.g. ’YYYY-MM-DD’, or ’YYYY-\n MM-DD hh:mm’). Hours and minutes are excluded if the observer did not report an observation\n time.\n \"obsReviewed\": TRUE if observation has been reviewed, FALSE otherwise\n \"obsValid\": TRUE if observation has been deemed valid by either the automatic filters or a regional\n viewer, FALSE otherwise\n \"sciName\" species’ scientific name\nAuthor(s)\n <NAME> <<EMAIL>>, <NAME> <<EMAIL>>\nReferences\n http://ebird.org/\nExamples\n ## Not run:\n ebirdgeo('amegfi', 42, -76) # American Goldfinch\n ebirdgeo(species_code('spinus tristis'), 42, -76) # same as above\n ebirdgeo(lat=42, lng=-76, max=10, provisional=TRUE, hotspot=TRUE)\n ebirdgeo(species_code('Anas platyrhynchos'), 39, -121, max=5)\n library('httr')\n ebirdgeo(species_code('Anas platyrhynchos'), 39, -121, max=5, config=verbose())\n ebirdgeo(species_code('Anas platyrhynchos'), 39, -121, max=5, config=progress())\n # ebirdgeo(species_code('Anas platyrhynchos'), 39, -121, max=5, config=timeout(0.1))\n ## End(Not run)\n ebirdhistorical Historic observations on a date at a region or hotspot\nDescription\n Returns a list of taxa reported in a given region or hotspot on a specific date\nUsage\n ebirdhistorical(\n loc,\n date,\n sortKey = \"mrec\",\n categories = \"all\",\n max = 10000,\n fieldSet = \"simple\",\n provisional = FALSE,\n limitToHotspots = FALSE,\n sleep = 0,\n key = NULL,\n ...\n )\nArguments\n loc (required) Region code or locID (if a hotspot). Region code can be country\n code (e.g. \"US\"), subnational1 code (states/provinces, e.g. \"US-NV\"), or sub-\n national2 code (counties, e.g. \"US-VA-003\").\n date (required) Date of historic observation date formatted according to ISO 8601\n (e.g. ’YYYY-MM-DD’, or ’YYYY-MM-DD hh:mm’). Hours and minutes are\n excluded.\n sortKey [mrec|create] Whether to order results by latest observation date or by latest\n creation date. The default is by observation date.\n categories [domestic|form|hybrid|intergrade|issf|slash|species|spuh] This is useful for lim-\n iting results to certain taxonomic categories. The default is all. Multiple cate-\n gories may be comma-separated.\n max Maximum number of result rows to return in this request. (A number between 1\n and 10000. The default is 10000)\n fieldSet [simple|full] This is set to restrict results to either all or a subset of sighting\n fields. The default is simple.\n provisional Should flagged records that have not been reviewed be included?\n limitToHotspots\n Should results be limited to sightings at birding hotspots? The default is FALSE.\n sleep Time (in seconds) before function sends API call. The defaults is zero. Set this\n to a higher number if you are using this function in a loop with many API calls.\n key eBird API key. You can obtain one from https://ebird.org/api/keygen. We strongly\n recommend storing it in your .Renviron file as an environment variable called\n EBIRD_KEY.\n ... Curl options passed on to GET.\nValue\n A data.frame containing the collected information:\n \"speciesCode\": species codes\n \"comName\": species common names\n \"sciName\" species’ scientific names\n \"locId\": unique identifier for the locations\n \"locName\": location name\n \"obsDt\": observation date formatted according to ISO 8601 (e.g. ’YYYY-MM-DD’, or ’YYYY-\n MM-DD hh:mm’). Hours and minutes are excluded if the observer did not report an observation\n time\n \"howMany\": \"howMany\": number of individuals observed, NA if only presence was noted\n \"obsValid\": TRUE if observation has been deemed valid by either the automatic filters or a regional\n viewer, FALSE otherwise\n \"obsReviewed\": TRUE if observation has been reviewed, FALSE otherwise\n \"locationPrivate\": TRUE if location is not a birding hotspot\n \"subID\": submission ID\n \"subnational2Code\": county code (returned if simple=FALSE)\n \"subnational2Name\": county name (returned if simple=FALSE)\n \"subnational1Code\": state/province ISO code (returned if simple=FALSE)\n \"subnational1Name\": state/province name (returned if simple=FALSE)\n \"countryCode\": country ISO code (returned if simple=FALSE)\n \"countryName\": country name (returned if simple=FALSE)\n \"userDisplayName\": first and last name of the observer (returned if simple=FALSE)\n \"obsID\": observation ID (returned if simple=FALSE)\n \"checklistID\": checklist ID (returned if simple=FALSE)\n \"presenceNoted\": ’true’ if user marked presence but did not count the number of birds. ’false’\n otherwise (returned if simple=FALSE)\n \"hasComments\": ’true’ if comments are included (returned if simple=FALSE)\n \"hasRichMedia\": ’true’ if rich media (e.g. photos/sounds) are included (returned if simple=FALSE)\n \"firstName\": observer’s first name (returned if simple=FALSE)\n \"lastName\": observer’s last name (returned if simple=FALSE)\nAuthor(s)\n <NAME> <<EMAIL>>\nReferences\n http://ebird.org/\nExamples\n ## Not run:\n ebirdhistorical(loc = 'US-VA-003', date='2019-02-14',max=10)\n ebirdhistorical(loc = 'L196159', date='2019-02-14', fieldSet='full')\n ## End(Not run)\n ebirdhotspot Recent observations at hotspots\nDescription\n Returns the most recent sighting information reported in a given vector of hotspots.\nUsage\n ebirdhotspot(\n locID,\n species = NULL,\n back = NULL,\n max = NULL,\n locale = NULL,\n provisional = FALSE,\n sleep = 0,\n key = NULL,\n ...\n )\nArguments\n locID (required) Vector containing code(s) for up to 10 regions of interest; here, re-\n gions are the locIDs of hotspots. Values that are not valid or are not hotspots are\n ignored.\n species Scientific name of the species of interest (not case sensitive). Defaults to NULL,\n in which case sightings for all species are returned. See eBird taxonomy for\n more information: http://ebird.org/content/ebird/about/ebird-taxonomy\n back Number of days back to look for observations (between 1 and 30, defaults to\n 14).\n max Maximum number of result rows to return in this request (between 1 and 10000,\n defaults to all)\n locale Language/locale of response (when translations are available). See http://java.sun.com/javase/6/docs/api/j\n (defaults to en_US)\n provisional Should flagged records that have not been reviewed be included? (defaults to\n FALSE)\n sleep Time (in seconds) before function sends API call (defaults to zero. Set to higher\n number if you are using this function in a loop with many API calls).\n key eBird API key. You can obtain one from https://ebird.org/api/keygen. We strongly\n recommend storing it in your .Renviron file as an environment variable called\n EBIRD_KEY.\n ... Curl options passed on to GET\nValue\n A data.frame containing the collected information:\n \"comName\": species common name\n \"howMany\": number of individuals observed, NA if only presence was noted\n \"lat\": latitude of the location\n \"lng\": longitude of the location\n \"locID\": unique identifier for the location\n \"locName\": location name\n \"locationPrivate\": TRUE if location is not a birding hotspot\n \"obsDt\": observation date formatted according to ISO 8601 (e.g. ’YYYY-MM-DD’, or ’YYYY-\n MM-DD hh:mm’). Hours and minutes are excluded if the observer did not report an observation\n time.\n \"obsReviewed\": TRUE if observation has been reviewed, FALSE otherwise\n \"obsValid\": TRUE if observation has been deemed valid by either the automatic filters or a regional\n viewer, FALSE otherwise\n \"sciName\" species’ scientific name\nAuthor(s)\n <NAME> <<EMAIL>>\nReferences\n http://ebird.org/\nExamples\n ## Not run:\n ebirdhotspot(locID=c('L99381','L99382'), species='larus delawarensis')\n ebirdhotspot('L99381', max=10, provisional=TRUE)\n ## End(Not run)\n ebirdhotspotlist Hotspots in a region or nearby coordinates\nDescription\n Get the list of hotspots within a region, or within a radius of up to 50 kilometers, from a given set\n of coordinates.\nUsage\n ebirdhotspotlist(\n regionCode = NULL,\n lat = NULL,\n lng = NULL,\n dist = NULL,\n back = NULL,\n sleep = 0,\n key = NULL,\n ...\n )\nArguments\n regionCode The country, subnational1 or subnational2 code. If ‘regionCode‘ is provided\n then latitude and longitude are ignored.\n lat Decimal latitude. value between -90.00 and 90.00, up to two decimal places of\n precision. Defaults to latitude based on IP if neither ‘regionCode‘ nor ‘lat‘ and\n ‘lng‘ are provided.\n lng Decimal longitude. value between -180.00 and 180.00, up to two decimal places\n of precision. Defaults to longitude based on IP if neither ‘regionCode‘ nor ‘lat‘\n and ‘lng‘ are provided.\n dist The search radius from the given set of coordinates, in kilometers (between 0\n and 500, defaults to 25).\n back Only fetch hotspots which have been visited up to ’back’ days ago (defaults to\n ‘NULL‘).\n sleep Time (in seconds) before function sends API call (defaults to zero. Set to higher\n number if you are using this function in a loop with many API calls).\n key eBird API key. You can obtain one from https://ebird.org/api/keygen. We strongly\n recommend storing it in your .Renviron file as an environment variable called\n EBIRD_KEY.\n ... Curl options passed on to GET\nValue\n A data.frame with ten columns containing:\n \"locId\": unique identifier for the hotspot\n \"locName\": hotspot name\n \"countryCode\": country code\n \"subnational1Code\": subnational1 code (state/province level)\n \"subnational2Code\": subnational2 code (county/municipality level)\n \"lat\": latitude of the hotspot\n \"lng\": longitude of the hotspot\n \"latestObsDt\": Date of latest observation\n \"numSpeciesAllTime\": Total number of species recorded in the hotspot\nAuthor(s)\n <NAME> <<EMAIL>>, <NAME> <<EMAIL>>\nReferences\n http://ebird.org/\nExamples\n ## Not run:\n ebirdhotspotlist(\"CA-NS-HL\")\n ebirdhotspotlist(\"VA\")\n ebirdhotspotlist(lat = 30, lng = -90, dist = 10)\n library(httr)\n ebirdhotspotlist(\"CA-NS-HL\", config = verbose())\n ## End(Not run)\n ebirdloc Recent observations at a locality\nDescription\n Returns the most recent sighting information reported in a given vector of locations (including non-\n hotspots).\nUsage\n ebirdloc(\n locID,\n species = NULL,\n back = NULL,\n max = NULL,\n locale = NULL,\n provisional = FALSE,\n simple = TRUE,\n sleep = 0,\n key = NULL,\n ...\n )\nArguments\n locID (required) Vector containing code(s) for up to 10 regions of interest; here, values\n that are not hotspots are returned. Values that are not valid are ignored.\n species Scientific name of the species of interest (not case sensitive). Defaults to NULL,\n in which case sightings for all species are returned. See eBird taxonomy for\n more information: http://ebird.org/content/ebird/about/ebird-taxonomy\n back Number of days back to look for observations (between 1 and 30, defaults to\n 14).\n max Maximum number of result rows to return in this request (between 1 and 10000,\n defaults to all)\n locale Language/locale of response (when translations are available). See http://java.sun.com/javase/6/docs/api/j\n (defaults to en_US)\n provisional Should flagged records that have not been reviewed be included? (defaults to\n FALSE)\n simple Logical. Whether to return a simple (TRUE, default) or detailed (FALSE) set of\n results fields.\n sleep Time (in seconds) before function sends API call (defaults to zero. Set to higher\n number if you are using this function in a loop with many API calls).\n key ebird API key. You can obtain one from https://ebird.org/api/keygen. We strongly\n recommend storing it in your .Renviron file as an environment variable called\n EBIRD_KEY.\n ... Curl options passed on to GET\nValue\n A data.frame containing the collected information:\n \"comName\": species common name\n \"howMany\": number of individuals observed, NA if only presence was noted\n \"lat\": latitude of the location\n \"lng\": longitude of the location\n \"locID\": unique identifier for the location\n \"locName\": location name\n \"locationPrivate\": TRUE if location is not a birding hotspot\n \"obsDt\": observation date formatted according to ISO 8601 (e.g. ’YYYY-MM-DD’, or ’YYYY-\n MM-DD hh:mm’). Hours and minutes are excluded if the observer did not report an observation\n time.\n \"obsReviewed\": TRUE if observation has been reviewed, FALSE otherwise\n \"obsValid\": TRUE if observation has been deemed valid by either the automatic filters or a regional\n viewer, FALSE otherwise\n \"sciName\" species’ scientific name\n \"subnational2Code\": county code (returned if simple=FALSE)\n \"subnational2Name\": county name (returned if simple=FALSE)\n \"subnational1Code\": state/province ISO code (returned if simple=FALSE)\n \"subnational1Name\": state/province name (returned if simple=FALSE)\n \"countryCode\": country ISO code (returned if simple=FALSE)\n \"countryName\": country name (returned if simple=FALSE)\n \"userDisplayName\": first and last name of the observer (returned if simple=FALSE)\n \"firstName\": observer’s first name (returned if simple=FALSE)\n \"lastName\": observer’s last name (returned if simple=FALSE)\n \"subID\": submission ID (returned if simple=FALSE)\n \"obsID\": observation ID (returned if simple=FALSE)\n \"checklistID\": checklist ID (returned if simple=FALSE)\n \"presenceNoted\": ’true’ if user marked presence but did not count the number of birds. ’false’\n otherwise (returned if simple=FALSE)\nAuthor(s)\n <NAME> <<EMAIL>>\nReferences\n http://ebird.org/\nExamples\n ## Not run:\n ebirdloc(locID = c('L99381','L99382'))\n ebirdloc('L99381', 'Branta canadensis', provisional=TRUE)\n ## End(Not run)\n ebirdnotable Recent nearby notable observations\nDescription\n Returns the most recent notable observations by either latitude/longitude, hotspot or location ID, or\n particular region.\nUsage\n ebirdnotable(\n lat = NULL,\n lng = NULL,\n dist = NULL,\n locID = NULL,\n region = NULL,\n back = NULL,\n max = NULL,\n provisional = FALSE,\n hotspot = FALSE,\n simple = TRUE,\n sleep = 0,\n key = NULL,\n ...\n )\nArguments\n lat Decimal latitude. value between -90.00 and 90.00, up to two decimal places of\n precision.\n lng Decimal longitude. value between -180.00 and 180.00, up to two decimal places\n of precision.\n dist Distance defining radius of interest from given lat/lng in kilometers (between 0\n and 50, defaults to 25)\n locID Vector containing code(s) for up to 10 locations of interest.\n region Region code corresponding to selected region type. For supported region and\n coding, see https://confluence.cornell.edu/display/CLOISAPI/eBird-1.1-RegionCodeReference\n back Number of days back to look for observations (between 1 and 30, defaults to\n 14).\n max Maximum number of result rows to return in this request (between 1 and 10000,\n defaults to all).\n provisional Should flagged records that have not been reviewed be included? (defaults to\n FALSE)\n hotspot Should results be limited to sightings at birding hotspots? (defaults to FALSE).\n simple Logical. Whether to return a simple (TRUE, default) or detailed (FALSE) set of\n results fields.\n sleep Time (in seconds) before function sends API call (defaults to zero. Set to higher\n number if you are using this function in a loop with many API calls).\n key eBird API key. You can obtain one from https://ebird.org/api/keygen. We strongly\n recommend storing it in your .Renviron file as an environment variable called\n EBIRD_KEY.\n ... Curl options passed on to GET\nValue\n A data.frame containing the collected information:\n \"speciesCode\": species code\n \"comName\": species common name\n \"sciName\" species’ scientific name\n \"locId\": unique identifier for the location\n \"locName\": location name\n \"obsDt\": observation date formatted according to ISO 8601 (e.g. ’YYYY-MM-DD’, or ’YYYY-\n MM-DD hh:mm’). Hours and minutes are excluded if the observer did not report an observation\n time.\n \"howMany\": number of individuals observed, NA if only presence was noted\n \"lat\": latitude of the location\n \"lng\": longitude of the location\n \"obsValid\": TRUE if observation has been deemed valid by either the\n \"obsReviewed\": TRUE if observation has been reviewed, FALSE otherwise\n \"locationPrivate\": TRUE if location is not a birding hotspot automatic filters or a regional viewer,\n FALSE otherwise\n \"subId\": submission ID\n \"subnational2Code\": county code (returned if simple=FALSE)\n \"subnational2Name\": county name (returned if simple=FALSE)\n \"subnational1Code\": state/province ISO code (returned if simple=FALSE)\n \"subnational1Name\": state/province name (returned if simple=FALSE)\n \"countryCode\": country ISO code (returned if simple=FALSE)\n \"countryName\": country name (returned if simple=FALSE)\n \"userDisplayName\": observer’s eBird username (returned if simple=FALSE)\n \"obsID\": observation ID (returned if simple=FALSE)\n \"checklistID\": checklist ID (returned if simple=FALSE)\n \"presenceNoted\": ’true’ if user marked presence but did not count the number of birds. ’false’\n otherwise (returned if simple=FALSE)\n \"firstName\": observer’s first name (returned if simple=FALSE)\n \"lastName\": observer’s last name (returned if simple=FALSE)\nNote\n ebirdnotable requires that either latitude/longitude, location ID, or region be passed to the func-\n tion. Multiple entries will result in the most specific being used. If none is supplied, defaults to\n lat/lng based on your IP.\nAuthor(s)\n <NAME> <<EMAIL>>, <NAME> <<EMAIL>>\nReferences\n http://ebird.org/\nExamples\n ## Not run:\n ebirdnotable(lat=42, lng=-70)\n ebirdnotable(region='US', max=10)\n ebirdnotable(region='US-OH')\n ebirdnotable(region='CA-NS-HL')\n ebirdnotable(locID = c('L275836','L124345'))\n ## End(Not run)\n ebirdregion Recent observations at a region or hotspot\nDescription\n Returns the most recent sighting information reported in a given region or hotspot.\nUsage\n ebirdregion(\n loc,\n species = NULL,\n back = NULL,\n max = NULL,\n locale = NULL,\n provisional = FALSE,\n hotspot = FALSE,\n simple = TRUE,\n sleep = 0,\n key = NULL,\n ...\n )\nArguments\n loc (required) Region code or locID (for hotspots). Region code can be country code\n (e.g. \"US\"), subnational1 (states/provinces, e.g. \"US-NV\"), or subnational2\n code (counties, e.g. \"CA-BC-GV\").\n species eBird species code. See ebirdtaxonomy for a full list of scientific names, com-\n mon names, and species codes. Alternatively, you can wrap the scientific name\n in the species_code function which will return the eBird species code. Defaults\n to NULL, in which case sightings for all species are returned. See eBird tax-\n onomy for more information: https://ebird.org/science/use-ebird-data/the-ebird-\n taxonomy\n back Number of days back to look for observations (between 1 and 30, defaults to\n 14).\n max Maximum number of result rows to return in this request (between 1 and 10000,\n defaults to all)\n locale Language/locale of response (when translations are available). See https://\n docs.oracle.com/javase/6/docs/api/java/util/Locale.html and https:\n //support.ebird.org/en/support/solutions/articles/48000804865-bird-names-in-ebird\n (defaults to en_US).\n provisional Should flagged records that have not been reviewed be included? (defaults to\n FALSE)\n hotspot Should results be limited to sightings at birding hotspots? (defaults to FALSE).\n simple Logical. Whether to return a simple (TRUE, default) or detailed (FALSE) set of\n results fields. Detailed results are only available if loc is a locID.\n sleep Time (in seconds) before function sends API call (defaults to zero. Set to higher\n number if you are using this function in a loop with many API calls).\n key eBird API key. You can obtain one from https://ebird.org/api/keygen. We strongly\n recommend storing it in your .Renviron file as an environment variable called\n EBIRD_KEY to avoid having to constantly supply the key, and to avoid acciden-\n tally sharing it publicly.\n ... Curl options passed on to GET\nValue\n A data.frame containing the collected information:\n \"speciesCode\": species code\n \"comName\": species common name\n \"sciName\" species’ scientific name\n \"locID\": unique identifier for the location\n \"locName\": location name\n \"obsDt\": observation date formatted according to ISO 8601 (e.g. ’YYYY-MM-DD’, or ’YYYY-\n MM-DD hh:mm’). Hours and minutes are excluded if the observer did not report an observation\n time.\n \"howMany\": number of individuals observed, NA if only presence was noted\n \"lat\": latitude of the location\n \"lng\": longitude of the location\n \"obsValid\": TRUE if observation has been deemed valid by either the automatic filters or a regional\n viewer, FALSE otherwise\n \"obsReviewed\": TRUE if observation has been reviewed, FALSE otherwise\n \"locationPrivate\": TRUE if location is not a birding hotspot\n \"subId\": submission ID\nAuthor(s)\n <NAME> <<EMAIL>>\nReferences\n http://ebird.org/\nExamples\n ## Not run:\n ebirdregion(loc = 'US', species = 'btbwar')\n ebirdregion(loc = 'US', species = species_code('Setophaga caerulescens')) # same as above\n ebirdregion(loc = 'L196159', species = 'bkcchi', back = 30)\n ebirdregion('US-OH', max = 10, provisional = TRUE, hotspot = TRUE)\n ## End(Not run)\n ebirdregioncheck Check if a region type is valid\nDescription\n Check if a region type is valid\nUsage\n ebirdregioncheck(loc, key = NULL, ...)\nArguments\n loc The location code to be checked.\n key ebird API key. You can obtain one from https://ebird.org/api/keygen. We strongly\n recommend storing it in your .Renviron file as an environment variable called\n EBIRD_KEY.\n ... Curl options passed on to GET\nValue\n Logical.\nAuthor(s)\n <NAME> <<EMAIL>>, <NAME> <<EMAIL>>\nReferences\n http://ebird.org/\nExamples\n ## Not run:\n ebirdregioncheck(\"US\")\n ebirdregioncheck(\"CA-BC\")\n ebirdregioncheck(\"CA-BC-GV\")\n ## End(Not run)\n ebirdregioninfo Region and hotspot info\nDescription\n Region and hotspot info\nUsage\n ebirdregioninfo(loc, format = \"full\", key = NULL, ...)\nArguments\n loc The location or hotspot code to be checked. A single location only.\n format Different options for displaying hierarchy of the region’s name: [nameonly|namequal|detailed|detailednoq\n defaults to full. Not used for hotspots.\n key eBird API key. You can obtain one from https://ebird.org/api/keygen. We strongly\n recommend storing it in your .Renviron file as an environment variable called\n EBIRD_KEY.\n ... Curl options passed on to GET\nValue\n When region is a hotspot, a data frame (with some redundant information) containing:\n \"locId\", \"locID\": hotspot ID\n \"name\", \"locName\": hotspot name\n \"latitude\", \"longitude\", \"lat\", \"long\": hotspot latitude and longitude (point location)\n \"countryCode\", \"countryName\": code and name of the country where hotspot is located\n \"subnational1Code\", \"subnational1Name\": code and name of the subnational1 area (e.g. state or\n province) where hotspot is located\n \"subnational2Code\", \"subnational2Name\": code and name of the subnational2 area (e.g. county)\n where hotspot is located\n \"isHotspot\": logical, whether region is a hotspot (should always be TRUE)\n \"hierarchicalName\": full hotspot name including subnational1, subnational2, and country info\n When region is a subnational1, subnational2, or country code, a data frame containing:\n \"region\": name of the region, varies depending on value of \"format\" provided\n \"minX\", \"maxX\", \"minY\", \"maxY\": lat/long bounds of the region\nAuthor(s)\n <NAME> <<EMAIL>>, <NAME> <<EMAIL>>, <NAME>\n <<EMAIL>>\nReferences\n http://ebird.org/\nExamples\n ## Not run:\n ebirdregioninfo(\"US\")\n ebirdregioninfo(\"CA-BC-GV\")\n ebirdregioninfo(\"CA-BC-GV\", format = \"revdetailed\") # reverse order of region name\n ebirdregioninfo(\"L196159\")\n ## End(Not run)\n ebirdregionspecies Get a list of species codes ever seen in a location.\nDescription\n Returns the eBird codes for all species-level taxa recorded in a particular region or location. Codes\n are returned in taxonomic order.\nUsage\n ebirdregionspecies(location, key = NULL, ...)\nArguments\n location Any valid location, USFWS region, subnational2, subnational1, country, or cus-\n tom region code. (Location can be a hotspot or personal location).\n key eBird API key. You can obtain one from https://ebird.org/api/keygen. We strongly\n recommend storing it in your .Renviron file as an environment variable called\n EBIRD_KEY.\n ... Curl options passed on to GET\nValue\n A single column data.frame containing the collected information:\n \"speciesCode\": eBird species code, suitable for joining to the ebirdtaxonomy\nAuthor(s)\n <NAME> <<EMAIL>>\nReferences\n http://ebird.org/\nExamples\n ## Not run:\n ebirdregionspecies(\"GB\") # all in Great Britain\n ebirdregionspecies(\"GB-ENG\") # all in England\n ebirdregionspecies(\"GB-ENG-LND\") # all in London\n library(dplyr)\n taxonomy <- ebirdtaxonomy()\n localSpecies <- ebirdregionspecies(\"L5803024\") # specific hotspot\n inner_join(localSpecies, taxonomy)\n ## End(Not run)\n ebirdsubregionlist List sub-regions within a specified region.\nDescription\n List sub-regions within a specified region.\nUsage\n ebirdsubregionlist(\n regionType = c(\"country\", \"subnational1\", \"subnational2\"),\n parentRegionCode,\n key = NULL,\n ...\n )\nArguments\n regionType The type of region to search for. Must be one of ’country’, ’subnational1’ or\n ’subnational2’.\n parentRegionCode\n The region to search within. Must be a valid country or subnational1 code. If\n ‘regionType‘ is ’country’ then this parameter is ignored (since the search will\n automatically be world-wide).\n key eBird API key. You can obtain one from https://ebird.org/api/keygen. We strongly\n recommend storing it in your .Renviron file as an environment variable called\n EBIRD_KEY.\n ... Curl options passed on to GET\nValue\n A data.frame containing:\n \"code\": eBird code for the subregion\n \"name\": full name for the subregion\nAuthor(s)\n <NAME> <<EMAIL>>\nReferences\n http://ebird.org/\nExamples\n ## Not run:\n ebirdsubregionlist(\"country\")\n ebirdsubregionlist(\"subnational1\", \"US\")\n ebirdsubregionlist(\"subnational2\", \"US-NY\")\n ## End(Not run)\n ebirdtaxonomy eBird Taxonomy\nDescription\n Returns a data.frame of all taxa in the eBird taxonomy for the given combination of categories.\n Defaults to all categories. Any taxon with the category of ’species’ may be used as a parameter\n in service calls that take a species code. Any taxon not in this category will be rejected by these\n services at this time.\nUsage\n ebirdtaxonomy(cat = NULL, locale = NULL, key = NULL, ...)\nArguments\n cat Species category. String or character vector with one of more of: \"domestic\",\n \"form\", \"hybrid\", \"intergrade\", \"issf\", \"slash\", \"species\", \"spuh\". If not speci-\n fied, defaults to all. For more info about the meaning of species categories, see\n https://ebird.org/science/use-ebird-data/the-ebird-taxonomy.\n locale Language/locale of response (when translations are available). See https://\n docs.oracle.com/javase/6/docs/api/java/util/Locale.html and https:\n //support.ebird.org/en/support/solutions/articles/48000804865-bird-names-in-ebird\n (defaults to en_US).\n key eBird API key. You can obtain one from https://ebird.org/api/keygen. We strongly\n recommend storing it in your .Renviron file as an environment variable called\n EBIRD_KEY to avoid having to constantly supply the key, and to avoid acciden-\n tally sharing it publicly.\n ... Curl options passed on to GET\nValue\n A data.frame containing the collected information:\n \"sciName\": Taxon’s scientific name.\n \"comName\": Taxon’s common name.\n \"speciesCode\": Unique species code.\n \"category\": Taxon’s species category.\n \"taxonOrder\": Numeric value determining the order in which taxonomic lists are presented.\n \"bandingCodes\": Taxon’s ABA banding code(s).\n \"comNameCodes\": Taxon’s common name code(s).\n \"sciNameCodes\": Taxon’s scientific name code(s).\n \"order\": Taxon’s order.\n \"familyComName\": Family’s common name.\n \"familySciName\": Family’s scientific name.\n \"reportAs\": Species code to report taxon as.\n \"extinct\": Logical, whether the taxon is considered extinct.\n \"extinctYear\": Year taxon became extinct. Currently unavailable.\nAuthor(s)\n <NAME> <<EMAIL>>, <NAME> <<EMAIL>>\nReferences\n http://ebird.org/\nExamples\n ## Not run:\n ebirdtaxonomy()\n ebirdtaxonomy(cat = c(\"spuh\", \"slash\"))\n ## End(Not run)\n getlatlng get latitude and longitude from ip address\nDescription\n Returns the most recent and nearest reported sighting information with observations of a species.\nUsage\n getlatlng()\nValue\n a vector of length 2 with lat, lng in that order\nAuthor(s)\n <NAME> <<EMAIL>>\nReferences\n http://ipinfo.io\nExamples\n ## Not run:\n getlatlng()\n ## End(Not run)\n nearestobs Recent nearby observations of a species\nDescription\n Returns the most recent and nearest reported sighting information with observations of a species.\nUsage\n nearestobs(\n speciesCode,\n lat = NULL,\n lng = NULL,\n dist = NULL,\n back = NULL,\n max = NULL,\n locale = NULL,\n provisional = FALSE,\n hotspot = FALSE,\n sleep = 0,\n key = NULL,\n ...\n )\nArguments\n speciesCode (required) Species code of the species of interest. Scientific names can be speci-\n fied if wrapped around the species_code function. Defaults to NULL, so sight-\n ings for all species are returned. See eBird taxonomy for more information:\n https://ebird.org/science/use-ebird-data/the-ebird-taxonomy.\n lat Decimal latitude. value between -90.00 and 90.00, up to two decimal places of\n precision. Defaults to latitude based on IP.\n lng Decimal longitude. value between -180.00 and 180.00, up to two decimal places\n of precision. Defaults to longitude based on IP.\n dist Distance defining radius of interest from given lat/lng in kilometers (between 0\n and 50, defaults to 25)\n back Number of days back to look for observations (between 1 and 30, defaults to\n 14).\n max Maximum number of result rows to return in this request (between 1 and 10000,\n defaults to all).\n locale Language/locale of response (when translations are available). See https://\n docs.oracle.com/javase/6/docs/api/java/util/Locale.html (defaults to\n en_US).\n provisional Should flagged records that have not been reviewed be included? (defaults to\n FALSE).\n hotspot Should results be limited to sightings at birding hotspots? (defaults to FALSE).\n sleep Time (in seconds) before function sends API call (defaults to zero. Set to higher\n number if you are using this function in a loop with many API calls).\n key eBird API key. You can obtain one from https://ebird.org/api/keygen. We strongly\n recommend storing it in your .Renviron file as an environment variable called\n EBIRD_KEY.\n ... Curl options passed on to GET\nValue\n A data.frame containing the collected information:\n \"speciesCode\": species code\n \"comName\": species common name\n \"sciName\" species’ scientific name\n \"locId\": unique identifier for the location\n \"locName\": location name\n \"obsDt\": observation date formatted according to ISO 8601 (e.g. ’YYYY-MM-DD’, or ’YYYY-\n MM-DD hh:mm’). Hours and minutes are excluded if the observer did not report an observation\n time.\n \"howMany\": number of individuals observed, NA if only presence was noted\n \"lat\": latitude of the location.\n \"lng\": longitude of the location.\n \"obsValid\": TRUE if observation has been deemed valid by either the automatic filters or a regional\n viewer, FALSE otherwise\n \"obsReviewed\": TRUE if observation has been reviewed, FALSE otherwise\n \"locationPrivate\": TRUE if location is not a birding hotspot\n \"subId\": submission ID\nAuthor(s)\n <NAME> <<EMAIL>>, <NAME> <<EMAIL>>\nReferences\n http://ebird.org/\nExamples\n ## Not run:\n nearestobs('cangoo', 42, -76) # Canada Goose\n nearestobs(species_code('branta canadensis'), 42, -76) # Same as above\n nearestobs(species_code('branta canadensis'), 42, -76, max=10, provisional=TRUE, hotspot=TRUE)\n ## End(Not run)\n rebird-deprecated Deprecated functions in rebird\nDescription\n These functions still work but will be removed (defunct) in the next version.\nDetails\n • ebirdregioncheck: Deprecated: ’ebirdregioncheck’ will be removed in the next version of\n rebird. Use ’ebirdregioninfo’ instead.\n • ebirdloc: Deprecated: ’ebirdloc’ will be removed in the next version of rebird as it might\n not be supported in the new eBird API. Use ’ebirdregion’ instead.\n • ebirdhotspot: Deprecated: ’ebirdhotspot’ will be removed in the next version of rebird as it\n might not be supported in the new eBird API. Use ’ebirdregion’ instead.\n species_code Return species code\nDescription\n Returns the species code for a given scientific name. Uses an internally-stored version of the tax-\n onomy. Also provides a message with the common name, scientific name, and species code of the\n species.\nUsage\n species_code(sciname = NULL)\nArguments\n sciname (required) Character string of length 1 with the scientific name to look for. Case\n insensitive.\nValue\n A character string with the eBird species code.\nAuthor(s)\n <NAME> <<EMAIL>>\nReferences\n http://ebird.org/\n30 species_code\nExamples\n species_code(\"Anhinga anhinga\")"}}},{"rowIdx":59181,"cells":{"project":{"kind":"string","value":"rebus"},"source":{"kind":"string","value":"cran"},"language":{"kind":"string","value":"R"},"content":{"kind":"string","value":"Package ‘rebus.datetimes’\n November 3, 2022\nTitle Date and Time Extensions for the 'rebus' Package\nVersion 0.0-2\nDate 2022-10-16\nAuthor <NAME> [aut, cre]\nMaintainer <NAME> <<EMAIL>>\nDescription Build regular expressions piece by piece using human readable code.\n This package contains date and time functionality, and is primarily intended\n to be used by package developers.\nDepends R (>= 3.1.0)\nImports rebus.base\nSuggests stats, testthat\nLicense Unlimited\nCollate 'imports.R' 'datetime-constants.R' 'datetime.R'\n 'iso-classes.R'\nRoxygenNote 7.2.1\nNeedsCompilation no\nRepository CRAN\nDate/Publication 2022-11-03 20:40:06 UTC\nR topics documented:\nget_weekday... 2\nIsoClasse... 3\nOPT_LEADING_... 4\n get_weekdays Get the days of the week or months of the year\nDescription\n Get the names of the days of the week in a given locale.\nUsage\n get_weekdays(abbreviate = FALSE, locale = NULL, from = Sys.Date())\n get_months(abbreviate = FALSE, locale = NULL, from = Sys.Date())\nArguments\n abbreviate A logical value indicating whether or not to return abbreviated names (if they\n are available).\n locale A string denoting the name of the locale to retrieve names in, or NULL to use the\n current locale.\n from Date to use for the first day/month.\nValue\n A string containing a regular expression of the names of the days of the week, separated by pipes.\n The first day of the week will be the current day.\nNote\n See Sys.setlocale and https://stackoverflow.com/q/20960821/134830 and https://stackoverflow.\n com/q/26603564/134830 for how to specify the locale.\nExamples\n get_weekdays()\n get_weekdays(TRUE)\n get_months()\n get_months(TRUE)\n ## Not run:\n if(.Platform$OS.type == \"windows\")\n {\n get_weekdays(locale = \"French_France\")\n get_weekdays(TRUE, locale = \"French_France\")\n get_weekdays(locale = \"Arabic_Qatar\")\n get_weekdays(TRUE, locale = \"Arabic_Qatar\")\n } else\n {\n get_weekdays(locale = \"fr_FR.utf8\")\n get_weekdays(TRUE, locale = \"fr_FR.utf8\")\n get_weekdays(locale = \"ar_QA.utf8\")\n get_weekdays(TRUE, locale = \"ar_QA.utf8\")\n }\n ## End(Not run)\n IsoClasses ISO 8601 date-time classes\nDescription\n Match ISO 8601 date and time classes.\nUsage\n iso_date(lo, hi, char_class = TRUE)\n iso_time(lo, hi, char_class = TRUE)\n iso_datetime(lo, hi, char_class = TRUE)\nArguments\n lo A non-negative integer. Minimum number of repeats, when grouped.\n hi positive integer. Maximum number of repeats, when grouped.\n char_class TRUE or FALSE. Should the values be wrapped into a character class?\nValue\n A character vector representing part or all of a regular expression.\nReferences\n http://www.iso.org/iso/iso8601\nExamples\n iso_date()\n iso_time()\n iso_datetime()\n # With repetition\n iso_date(3, 6)\n iso_time(1, Inf)\n iso_datetime(0, Inf)\n # Without a class wrapper\n iso_date(char_class = FALSE)\n OPT_LEADING_0 Date-time regexes\nDescription\n Compound regex constants for matching ISO 8601 dates and times.\nUsage\n OPT_LEADING_0\n DTSEP\n CENTURY\n CENTURY_IN\n YEAR\n YEAR2\n YEAR4\n MONTH\n MONTH_IN\n WEEK_OF_YEAR\n WEEK_OF_YEAR_IN\n DAY\n DAY_IN\n DAY_SINGLE\n DAY_OF_YEAR\n DAY_OF_YEAR_IN\n WEEKDAY1\n WEEKDAY0\n HOUR24\nOPT_LEADING_0 5\n HOUR24_SINGLE\n HOUR24_IN\n HOUR12\n HOUR12_SINGLE\n HOUR12_IN\n MINUTE\n MINUTE_IN\n SECOND\n SECOND_IN\n FRACTIONAL_SECOND\n FRACTIONAL_SECOND_IN\n AM_PM\n TIMEZONE_OFFSET\n TIMEZONE\n ISO_DATE\n ISO_DATE_IN\n ISO_TIME\n ISO_TIME_IN\n ISO_DATETIME\n ISO_DATETIME_IN\n YMD\n YMD_IN\n YDM\n YDM_IN\n MYD\n MYD_IN\n MDY\n MDY_IN\n DYM\n DYM_IN\n DMY\n DMY_IN\n HMS\n HMS_IN\n HM\n HM_IN\n MS\n MS_IN\n datetime(x, locale = NULL, io = c(\"output\", \"input\"))\nArguments\n x A strptime-style date-time format string.\n locale A string specifying a locale.\n io Are you trying to match output or input? The latter is less strict about leading\n zeroes and spaces.\nFormat\n An object of class character of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\n An object of class regex (inherits from character) of length 1.\nValue\n A character vector representing part or all of a regular expression.\nNote\n \"%O[dHImMUVwWy]\", \"%E[cCyYxX]\", \"%x\", \"%X\" and \"%+\" are supposed to be locale-dependent\n upon output, but implementing this in an OS-portable way seems to be much more effort than it’s\n worth.\nSee Also\n strptime that describes formatting codes, ClassGroups, Sys.setlocale\nExamples\n datetime(\"%m/%d/%Y\") # match US style dates\n twelve_or_twentyfour <- rebus.base::or(\"%H\", \"%I%p\")\n datetime(twelve_or_twentyfour) # match hours in 24h or 12h format\n ## Not run:\n # week days and months can be matched in any locale\n if(.Platform$OS.type == \"windows\")\n {\n fr_FR <- \"French_France\"\n ar_QA <- \"Arabic_Qatar\"\n } else if(Sys.info()[\"sysname\"] != \"Darwin\") # mac\n {\n fr_FR <- \"fr_FR\"\n ar_QA <- \"ar_QA\"\n } else if(Sys.info()[\"sysname\"] != \"Linux\")\n {\n fr_FR <- \"fr_FR.utf8\"\n ar_QA <- \"ar_QA.utf8\"\n }\n datetime(\"%a %A %b %B\", fr_FR)\n datetime(\"%a %A %b %B\", ar_QA)\n # All letter tokens. Lots of output.\n x <- paste0(\"%\", c(letters, LETTERS))\n stats::setNames(datetime(x), x)\n ## End(Not run)\n # Individual date-time components\n DTSEP # optional selected punctuation or space\n CENTURY # exactly two digits\n YEAR # one to four digits\n YEAR2 # exactly two digits\n YEAR4 # exactly four digits\n MONTH # number from 1 to 12, leading zero\n WEEK_OF_YEAR # number from 0 to 53, leading zero\n DAY # number from 1 to 31, leading zero\n DAY_SINGLE # leading space\n HOUR24 # 24 hour clock, leading zero\n HOUR12 # 12 hour clock, leading zero\n HOUR24_SINGLE # 24 hour clock, leading space\n HOUR12_SINGLE # 12 hour clock, leading space\n MINUTE # number from 0 to 59, leading zero\n SECOND # number from 0 to 61 (leap seconds), leading zero\n FRACTIONAL_SECOND # a second optional decimal point and up to 6 digits\n AM_PM # AM or PM, any case\n TIMEZONE_OFFSET # optional plus or minus, then four digits\n TIMEZONE # Any value returned by OlsonNames()\n # ISO 8601 formats\n ISO_DATE # %Y-%m-%d\n ISO_TIME # %H:%M:%S\n ISO_DATETIME # ISO_DATE followed by ISO_TIME, separated by space or \"T\".\n # Compound forms, separated by DTSEP\n YMD\n YDM\n MYD\n MDY\n DYM\n DMY\n HMS\n HM\n MS\n # Some forms have less strict alternatives for input (with an '_IN' suffix).\n CENTURY_IN\n MONTH_IN\n WEEK_OF_YEAR_IN\n DAY_IN\n HOUR24_IN\n HOUR12_IN\n MINUTE_IN\n SECOND_IN\n FRACTIONAL_SECOND_IN\n ISO_DATE_IN\n ISO_TIME_IN\n ISO_DATETIME_IN\n YMD_IN\n YDM_IN\n MYD_IN\n MDY_IN\n DYM_IN\n DMY_IN\n HMS_IN\n HM_IN\n MS_IN\n dates <- seq(as.Date(\"2000-01-01\"), as.Date(\"2001-01-01\"), \"1 day\")\n datetimes <- seq(as.POSIXct(Sys.Date()), as.POSIXct(Sys.Date() + 1), \"1 sec\")\n times <- substring(format(datetimes), 12, 19)\n stopifnot(\n all(grepl(ISO_DATE, dates)),\n all(grepl(ISO_TIME, times)),\n all(grepl(ISO_DATETIME, datetimes))\n )\n non_dates <- c(\n \"2000-13-01\", \"2000-01-32\", \"2000-00-01\", \"2000-01-00\"\n )\n non_times <- c(\n \"24:00:00\", \"23:60:59\", \"23:59:62\", \"23 59 59\"\n )\n stopifnot(\n all(!grepl(ISO_DATE, non_dates)),\n all(!grepl(ISO_TIME, non_times))\n )"}}},{"rowIdx":59182,"cells":{"project":{"kind":"string","value":"GITHUB_raaminz_training.zip_unzipped_6_-_Http_Status_Codes_And_Exception_Handling.pdf"},"source":{"kind":"string","value":"free_programming_book"},"language":{"kind":"string","value":"Unknown"},"content":{"kind":"string","value":"Http Status Codes\n& Exception Handling\nPresented by <NAME> Twitter: @zare88r\nHTTP Status Codes\n Http Status header\nHTTP Status Codes\n1xx\n100 continue\n101 Switching Protocols\n2xx\n200 OK\nRedirection 3xx\n303 SEE OTHER\nHTTP Status Codes\n 4\n400 Bad Request\n401 Unauthorized\n403 Forbidden\n 5\nhttps://httpstatuses.com/\n500 Internal Server Error\n502 Bad Gateway\n503 Service Unavailable\nservlet status code\nresponse.setStatus(HttpServletResponse.SC_CONTINUE);\nresponse.sendError(HttpServletResponse.SC_BAD_REQUEST);\nresponse.sendError(HttpServletResponse.SC_UNAUTHORIZED,\n\"First Login /login.html\");\nCustomized error pages web.xml\n<error-page>\n<error-code>401</error-code>\n<location>/unauthorized.html</location>\n</error-page>\n<error-page>\n<error-code>404</error-code>\n<location>/notfound.html</location>\n</error-page>\nExceptions\n ServletException\n servlet\n . status code 500\n UnavailableException\n Servlet\n status code 503\n status code 404 .\nException Handler\n<error-page>\n<exception-type>java.lang.Throwable</exception-type>\n<location>/errorHandler</location>\n</error-page>\nscope servlet engine\n request exception servlet : javax.servlet.error.servlet_name\n 500 :javax.servlet.error.status_code\n exception : javax.servlet.error.exception\n location :javax.servlet.error.request_uri"}}},{"rowIdx":59183,"cells":{"project":{"kind":"string","value":"EMpeaksR"},"source":{"kind":"string","value":"cran"},"language":{"kind":"string","value":"R"},"content":{"kind":"string","value":"Package ‘EMpeaksR’\n March 29, 2023\nType Package\nTitle Conducting the Peak Fitting Based on the EM Algorithm\nVersion 0.3.1\nDescription The peak fitting of spectral data is performed by using the frame work of EM algo-\n rithm. We adapted the EM algorithm for the peak fitting of spectral data set by consider-\n ing the weight of the intensity corresponding to the measurement energy steps (Mat-\n <NAME>., <NAME>., <NAME>., Na-\n <NAME>., & <NAME>. (2019, 2021 and 2023) <doi:10.1080/14686996.2019.1620123>, <doi:10.1080/27660400.2021.189944\n age efficiently estimates the parameters of Gaussian mixture model during iterative calcula-\n tion between E-step and M-step, and the parameters are converged to a local optimal solu-\n tion. This package can support the investigation of peak shift with two advan-\n tages: (1) a large amount of data can be processed at high speed; and (2) stable and auto-\n matic calculation can be easily performed.\nLicense MIT + file LICENSE\nLanguage en-US\nEncoding UTF-8\nRoxygenNote 7.1.1\nSuggests testthat (>= 3.0.0)\nConfig/testthat/edition 3\nNeedsCompilation no\nAuthor <NAME> [aut, cre]\nMaintainer <NAME> <<EMAIL>>\nRepository CRAN\nDate/Publication 2023-03-29 13:40:02 UTC\nR topics documented:\nshow_dsgmm_curv... 2\nshow_gmm_curv... 4\nshow_lmm_curv... 6\nshow_pvmm_curv... 8\nshow_pvmm_lback_curv... 10\nspect_em_dsgm... 12\nspect_em_gm... 15\nspect_em_lm... 17\nspect_em_pvm... 19\nspect_em_pvmm_lbac... 21\n show_dsgmm_curve Visualization of the result of spect_em_dsgmm\nDescription\n Visualization of the result of spect_em_dsgmm().\nUsage\n show_dsgmm_curve(spect_em_dsgmm_res,\n x,\n y,\n mix_ratio_init,\n mu_init,\n sigma_init,\n alpha_init,\n eta_init)\nArguments\n spect_em_dsgmm_res\n data set obtained by spect_em_dsgmm()\n x measurement steps\n y intensity\n mix_ratio_init initial values of the mixture ratio of the components\n mu_init initial values of the mean of the components\n sigma_init initial values of the standard deviation of the components\n alpha_init initial values of the asymmetric parameter of the components\n eta_init initial values of the mixing ratio of Gauss and Lorentz distribution\nDetails\n Perform a visualization of fitting curve estimated by Doniach-Sunjic-Gauss mixture model.\nValue\n Show the fitting curve and variation of the parameters.\nReferences\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2019). Spectrum adapted\n expectation-maximization algorithm for high-throughput peak shift analysis. Science and technol-\n ogy of advanced materials, 20(1), 733-745.\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2021). Spectrum adapted\n expectation-conditional maximization algorithm for extending high–throughput peak separation\n method in XPS analysis. Science and Technology of Advanced Materials: Methods, 1(1), 45-55.\nExamples\n #generating the synthetic spectral data based on three component Doniach-Sunjic-Gauss mixture model.\n x <- seq(0, 100, by = 0.5)\n true_mu <- c(20, 50, 80)\n true_sigma <- c(3, 3, 3)\n true_alpha <- c(0.1, 0.3, 0.1)\n true_eta <- c(0.4, 0.6, 0.1)\n true_mix_ratio <- rep(1/3, 3)\n degree <- 4\n #trancated Doniach-Sunjic-Gauss\n truncated_dsg <- function(x, mu, sigma, alpha, eta) {\n ((eta*(((gamma(1-alpha)) /\n ((x-mu)^2+(sqrt(2*log(2))*sigma)^2)^((1-alpha)/2)) *\n cos((pi*alpha/2)+(1-alpha)*atan((x-mu) /\n (sqrt(2*log(2))*sigma))))) + (1-eta)*dnorm(x, mu, sigma)) /\n sum( ((eta*(((gamma(1-alpha)) /\n ((x-mu)^2+(sqrt(2*log(2))*sigma)^2)^((1-alpha)/2)) *\n cos((pi*alpha/2)+(1-alpha)*atan((x-mu) /\n (sqrt(2*log(2))*sigma))))) + (1-eta)*dnorm(x, mu, sigma)))\n }\n y <- c(true_mix_ratio[1]*truncated_dsg(x = x,\n mu = true_mu[1],\n sigma = true_sigma[1],\n alpha = true_alpha[1],\n eta = true_eta[1])*10^degree +\n true_mix_ratio[2]*truncated_dsg(x = x,\n mu = true_mu[2],\n sigma = true_sigma[2],\n alpha = true_alpha[2],\n eta = true_eta[2])*10^degree +\n true_mix_ratio[3]*truncated_dsg(x = x,\n mu = true_mu[3],\n sigma = true_sigma[3],\n alpha = true_alpha[3],\n eta = true_eta[3])*10^degree)\n plot(y~x, main = \"genrated synthetic spectral data\")\n #Peak fitting by EMpeaksR\n #Initial values\n K <- 3\n mix_ratio_init <- c(0.2, 0.4, 0.4)\n mu_init <- c(20, 40, 70)\n sigma_init <- c(4, 3, 2)\n alpha_init <- c(0.3, 0.2, 0.4)\n eta_init <- c(0.5, 0.4, 0.3)\n #Coducting calculation\n SP_ECM_DSG_res <- spect_em_dsgmm(x = x,\n y = y,\n mu = mu_init,\n sigma = sigma_init,\n alpha = alpha_init,\n eta = eta_init,\n mix_ratio = mix_ratio_init,\n conv.cri = 1e-2,\n maxit = 2000)\n #Plot fitting curve and trace plot of parameters\n show_dsgmm_curve(SP_ECM_DSG_res,\n x,\n y,\n mix_ratio_init,\n mu_init,\n sigma_init,\n alpha_init,\n eta_init)\n #Showing the result of spect_em_dsgmm()\n print(cbind(c(mu_init),\n c(sigma_init),\n c(alpha_init),\n c(eta_init),\n c(mix_ratio_init)))\n print(cbind(SP_ECM_DSG_res$mu,\n SP_ECM_DSG_res$sigma,\n SP_ECM_DSG_res$alpha,\n SP_ECM_DSG_res$eta,\n SP_ECM_DSG_res$mix_ratio))\n print(cbind(true_mu,\n true_sigma,\n true_alpha,\n true_eta,\n true_mix_ratio))\n show_gmm_curve Visualization of the result of spect_em_gmm\nDescription\n Visualization of the result of spect_em_gmm().\nUsage\n show_gmm_curve(spect_em_gmm_res, x, y, mix_ratio_init, mu_init, sigma_init)\nArguments\n spect_em_gmm_res\n data set obtained by spect_em_gmm()\n x measurement steps\n y intensity\n mix_ratio_init initial values of the mixture ratio of the components\n mu_init initial values of the mean of the components\n sigma_init initial values of the standard deviation of the components\nDetails\n Perform a visualization of fitting curve estimated by Gaussian mixture model.\nValue\n Show the fitting curve and variation of the parameters.\nReferences\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2019). Spectrum adapted\n expectation-maximization algorithm for high-throughput peak shift analysis. Science and technol-\n ogy of advanced materials, 20(1), 733-745.\nExamples\n #generating the synthetic spectral data based on three component Gausian mixture model.\n x <- seq(0, 100, by = 0.5)\n true_mu <- c(35, 50, 65)\n true_sigma <- c(3, 3, 3)\n true_mix_ratio <- rep(1/3, 3)\n degree <- 4\n y <- c(true_mix_ratio[1] * dnorm(x = x, mean = true_mu[1], sd = true_sigma[1])*10^degree +\n true_mix_ratio[2] * dnorm(x = x, mean = true_mu[2], sd = true_sigma[2])*10^degree +\n true_mix_ratio[3] * dnorm(x = x, mean = true_mu[3], sd = true_sigma[3])*10^degree)\n plot(y~x, main = \"genrated synthetic spectral data\")\n #Peak fitting by EMpeaksR\n #Initial values\n K <- 3\n mix_ratio_init <- c(0.2, 0.4, 0.4)\n mu_init <- c(20, 40, 70)\n sigma_init <- c(2, 5, 4)\n #Coducting calculation\n SP_EM_G_res <- spect_em_gmm(x, y, mu = mu_init, sigma = sigma_init, mix_ratio = mix_ratio_init,\n conv.cri = 1e-2, maxit = 2000)\n #Plot fitting curve and trace plot of parameters\n show_gmm_curve(SP_EM_G_res, x, y, mix_ratio_init, mu_init, sigma_init)\n #Showing the result of spect_em_gmm()\n print(cbind(c(mu_init), c(sigma_init), c(mix_ratio_init)))\n print(cbind(SP_EM_G_res$mu, SP_EM_G_res$sigma, SP_EM_G_res$mix_ratio))\n print(cbind(true_mu, true_sigma, true_mix_ratio))\n show_lmm_curve Visualization of the result of spect_em_lmm\nDescription\n Visualization of the result of spect_em_lmm().\nUsage\n show_lmm_curve(spect_em_lmm_res, x, y, mix_ratio_init, mu_init, gam_init)\nArguments\n spect_em_lmm_res\n data set obtained by spect_em_lmm()\n x measurement steps\n y intensity\n mix_ratio_init initial values of the mixture ratio of the components\n mu_init initial values of the mean of the components\n gam_init initial values of the scale parameter of the components\nDetails\n Perform a visualization of fitting curve estimated by Lorentz mixture model.\nValue\n Show the fitting curve and variation of the parameters.\nReferences\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2019). Spectrum adapted\n expectation-maximization algorithm for high-throughput peak shift analysis. Science and technol-\n ogy of advanced materials, 20(1), 733-745.\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2021). Spectrum adapted\n expectation-conditional maximization algorithm for extending high–throughput peak separation\n method in XPS analysis. Science and Technology of Advanced Materials: Methods, 1(1), 45-55.\nExamples\n #generating the synthetic spectral data based on three component Lorentz mixture model.\n x <- seq(0, 100, by = 0.5)\n true_mu <- c(35, 50, 65)\n true_gam <- c(3, 3, 3)\n true_mix_ratio <- rep(1/3, 3)\n degree <- 4\n #Normalized Lorentz distribution\n dCauchy <- function(x, mu, gam) {\n (dcauchy(x, mu, gam)) / sum(dcauchy(x, mu, gam))\n }\n y <- c(true_mix_ratio[1] * dCauchy(x = x, mu = true_mu[1], gam = true_gam[1])*10^degree +\n true_mix_ratio[2] * dCauchy(x = x, mu = true_mu[2], gam = true_gam[2])*10^degree +\n true_mix_ratio[3] * dCauchy(x = x, mu = true_mu[3], gam = true_gam[3])*10^degree)\n plot(y~x, main = \"genrated synthetic spectral data\")\n #Peak fitting by EMpeaksR\n #Initial values\n K <- 3\n mix_ratio_init <- c(0.2, 0.4, 0.4)\n mu_init <- c(20, 40, 70)\n gam_init <- c(2, 5, 4)\n #Coducting calculation\n SP_ECM_L_res <- spect_em_lmm(x, y, mu = mu_init, gam = gam_init, mix_ratio = mix_ratio_init,\n conv.cri = 1e-2, maxit = 2000)\n #Plot fitting curve and trace plot of parameters\n show_lmm_curve(SP_ECM_L_res, x, y, mix_ratio_init, mu_init, gam_init)\n #Showing the result of spect_em_lmm()\n print(cbind(c(mu_init), c(gam_init), c(mix_ratio_init)))\n print(cbind(SP_ECM_L_res$mu, SP_ECM_L_res$gam, SP_ECM_L_res$mix_ratio))\n print(cbind(true_mu, true_gam, true_mix_ratio))\n show_pvmm_curve Visualization of the result of spect_em_pvmm\nDescription\n Visualization of the result of spect_em_pvmm().\nUsage\n show_pvmm_curve(spect_em_pvmm_res, x, y, mix_ratio_init, mu_init, sigma_init, eta_init)\nArguments\n spect_em_pvmm_res\n data set obtained by spect_em_pvmm()\n x measurement steps\n y intensity\n mix_ratio_init initial values of the mixture ratio of the components\n mu_init initial values of the mean of the components\n sigma_init initial values of the standard deviation of the components\n eta_init initial values of the mixing ratio of Gauss and Lorentz distribution\nDetails\n Perform a visualization of fitting curve estimated by Pseudo-Voigt mixture model.\nValue\n Show the fitting curve and variation of the parameters.\nReferences\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2019). Spectrum adapted\n expectation-maximization algorithm for high-throughput peak shift analysis. Science and technol-\n ogy of advanced materials, 20(1), 733-745.\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2021). Spectrum adapted\n expectation-conditional maximization algorithm for extending high–throughput peak separation\n method in XPS analysis. Science and Technology of Advanced Materials: Methods, 1(1), 45-55.\nExamples\n #generating the synthetic spectral data based on three component Pseudo-Voigt mixture model.\n x <- seq(0, 100, by = 0.5)\n true_mu <- c(35, 50, 65)\n true_sigma <- c(3, 3, 3)\n true_eta <- c(0.3, 0.8, 0.5)\n true_mix_ratio <- rep(1/3, 3)\n degree <- 4\n #Normalized Pseudo-Voigt distribution\n truncated_pv <- function(x, mu, sigma, eta) {\n (eta*dcauchy(x, mu, sqrt(2*log(2))*sigma) + (1-eta)*dnorm(x, mu, sigma)) /\n sum(eta*dcauchy(x, mu, sqrt(2*log(2))*sigma) + (1-eta)*dnorm(x, mu, sigma))\n }\n y <- c(true_mix_ratio[1]*truncated_pv(x = x,\n mu = true_mu[1],\n sigma = true_sigma[1],\n eta = true_eta[1])*10^degree +\n true_mix_ratio[2]*truncated_pv(x = x,\n mu = true_mu[2],\n sigma = true_sigma[2],\n eta = true_eta[2])*10^degree +\n true_mix_ratio[3]*truncated_pv(x = x,\n mu = true_mu[3],\n sigma = true_sigma[3],\n eta = true_eta[3])*10^degree)\n plot(y~x, main = \"genrated synthetic spectral data\")\n #Peak fitting by EMpeaksR\n #Initial values\n K <- 3\n mix_ratio_init <- c(0.2, 0.4, 0.4)\n mu_init <- c(20, 40, 70)\n sigma_init <- c(2, 5, 4)\n eta_init <- c(0.5, 0.4, 0.3)\n #Coducting calculation\n SP_ECM_PV_res <- spect_em_pvmm(x = x,\n y = y,\n mu = mu_init,\n sigma = sigma_init,\n eta = eta_init,\n mix_ratio = mix_ratio_init,\n conv.cri = 1e-2,\n maxit = 2000)\n #Plot fitting curve and trace plot of parameters\n show_pvmm_curve(SP_ECM_PV_res, x, y, mix_ratio_init, mu_init, sigma_init, eta_init)\n #Showing the result of spect_em_pvmm()\n print(cbind(c(mu_init), c(sigma_init), c(eta_init), c(mix_ratio_init)))\n print(cbind(SP_ECM_PV_res$mu, SP_ECM_PV_res$sigma, SP_ECM_PV_res$eta, SP_ECM_PV_res$mix_ratio))\n print(cbind(true_mu, true_sigma, true_eta, true_mix_ratio))\n show_pvmm_lback_curve Visualization of the result of spect_em_pvmm_lback\nDescription\n Visualization of the result of spect_em_pvmm_lback().\nUsage\n show_pvmm_lback_curve(spect_em_pvmm_lback_res,\n x, y,\n mix_ratio_init,\n mu_init,\n sigma_init,\n eta_init,\n x_lower,\n x_upper)\nArguments\n spect_em_pvmm_lback_res\n data set obtained by spect_em_pvmm_lback()\n x measurement steps\n y intensity\n mu_init initial values of the mean of the components\n sigma_init initial values of the standard deviation of the components\n eta_init initial values of the mixing ratio of Gauss and Lorentz distribution\n mix_ratio_init initial values of the mixture ratio of the components\n x_lower lower limit of the measurement steps. Default is a minimum of x\n x_upper upper limit of the measurement steps. Default is a maximum of x\nDetails\n Perform a visualization of fitting curve estimated by pseudo-Voigt mixture model with a linear\n background.\nValue\n Show the fitting curve and variation of the parameters.\nReferences\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2019). Spectrum adapted\n expectation-maximization algorithm for high-throughput peak shift analysis. Science and technol-\n ogy of advanced materials, 20(1), 733-745.\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2021). Spectrum adapted\n expectation-conditional maximization algorithm for extending high–throughput peak separation\n method in XPS analysis. Science and Technology of Advanced Materials: Methods, 1(1), 45-55.\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2023). High-throughput XPS\n spectrum modeling with autonomous background subtraction for 3 d 5/2 peak mapping of SnS.\n Science and Technology of Advanced Materials: Methods, 3(1), 2159753.\nExamples\n #generating the synthetic spectral data based on three component Pseudo-Voigt mixture model.\n x <- seq(0, 100, by = 0.5)\n K <- 3\n true_mu <- c(35, 50, 65)\n true_sigma <- c(3, 3, 3)\n true_mix_ratio <- c(0.5/3, 0.5/3, 0.5/3, 0.5)\n true_eta <- c(0.4, 0.6, 0.1)\n degree <- 4\n #Normalized Pseudo-Voigt distribution\n truncated_pv <- function(x, mu, sigma, eta) {\n (eta*dcauchy(x, mu, sqrt(2*log(2))*sigma) + (1-eta)*dnorm(x, mu, sigma)) /\n sum(eta*dcauchy(x, mu, sqrt(2*log(2))*sigma) + (1-eta)*dnorm(x, mu, sigma))\n }\n y <- c(true_mix_ratio[1]*truncated_pv(x = x,\n mu = true_mu[1],\n sigma = true_sigma[1],\n eta = true_eta[1])*10^degree +\n true_mix_ratio[2]*truncated_pv(x = x,\n mu = true_mu[2],\n sigma = true_sigma[2],\n eta = true_eta[2])*10^degree +\n true_mix_ratio[3]*truncated_pv(x = x,\n mu = true_mu[3],\n sigma = true_sigma[3],\n eta = true_eta[3])*10^degree +\n true_mix_ratio[4]*(c(500*x + 15000) / sum(500*x + 15000))*10^degree)\n plot(y~x, main = \"genrated synthetic spectral data\")\n #Peak fitting by EMpeaksR\n #Initial values\n mu_init <- c(30, 40, 60)\n sigma_init <- c(4, 4, 4)\n mix_ratio_init <- rep(1/(length(mu_init)+3), length(mu_init)+3)\n eta_init <- c(1, 1, 1)\n #Coducting calculation\n SP_ECM_PV_LBACK_res <- spect_em_pvmm_lback(x = x,\n y = y,\n mu = mu_init,\n sigma = sigma_init,\n eta = eta_init,\n mix_ratio = mix_ratio_init,\n x_lower = min(x),\n x_upper = max(x),\n #Plot fitting curve and trace plot of parameters\n show_pvmm_lback_curve(spect_em_pvmm_lback_res = SP_ECM_PV_LBACK_res,\n x = x,\n y = y,\n mix_ratio_init = mix_ratio_init,\n mu_init = mu_init,\n sigma_init = sigma_init,\n eta_init = eta_init,\n x_lower = min(x),\n x_upper = max(x))\n #Showing the result of spect_em_pvmm_lback()\n print(cbind(SP_ECM_PV_LBACK_res$mu, SP_ECM_PV_LBACK_res$sigma, SP_ECM_PV_LBACK_res$eta,\n SP_ECM_PV_LBACK_res$mix_ratio[1:K]))\n print(cbind(true_mu, true_sigma, true_eta, true_mix_ratio[1:K]))\n spect_em_dsgmm Spectrum adapted ECM algorithm by DSGMM\nDescription\n Perform a peak fitting based on the spectrum adapted ECM algorithm by Doniach-Sunjic-Gauss\n mixture model.\nUsage\n spect_em_dsgmm(x, y, mu, sigma, alpha, eta, mix_ratio, conv.cri, maxit)\nArguments\n x measurement steps\n y intensity\n mu mean of the components\n sigma standard deviation of the components\n alpha asymmetric parameter of the component\n eta mixing ratio of Gauss and Lorentz distribution\n mix_ratio mixture ratio of the components\n conv.cri criterion of the convergence\n maxit maximum number of the iteration\nDetails\n Peak fitting is conducted by spectrum adapted ECM algorithm.\nValue\n mu estimated mean of the components\n sigma estimated standard deviation of the components\n alpha estimated asymmetric parameter of the components\n eta estimated mixing ratio of Gauss and Lorentz distribution\n mix_ratio estimated mixture ratio of the components\n it number of the iteration to reach the convergence\n LL variation of the weighted log likelihood values\n MU variation of mu\n SIGMA variation of sigma\n ALPHA variation of alpha\n ETA variation of beta\n MIX_RATIO variation of mix_ratio\n W_K decomposed component of the spectral data\n convergence message for the convergence in the calculation\n cal_time calculation time to complete the peak fitting. Unit is seconds\nReferences\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2019). Spectrum adapted\n expectation-maximization algorithm for high-throughput peak shift analysis. Science and technol-\n ogy of advanced materials, 20(1), 733-745.\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2021). Spectrum adapted\n expectation-conditional maximization algorithm for extending high–throughput peak separation\n method in XPS analysis. Science and Technology of Advanced Materials: Methods, 1(1), 45-55.\nExamples\n #generating the synthetic spectral data based on three component Doniach-Sunjic-Gauss mixture model.\n x <- seq(0, 100, by = 0.5)\n true_mu <- c(20, 50, 80)\n true_sigma <- c(3, 3, 3)\n true_alpha <- c(0.1, 0.3, 0.1)\n true_eta <- c(0.4, 0.6, 0.1)\n true_mix_ratio <- rep(1/3, 3)\n degree <- 4\n #trancated Doniach-Sunjic-Gauss\n truncated_dsg <- function(x, mu, sigma, alpha, eta) {\n ((eta*(((gamma(1-alpha)) /\n ((x-mu)^2+(sqrt(2*log(2))*sigma)^2)^((1-alpha)/2)) *\n cos((pi*alpha/2)+(1-alpha)*atan((x-mu) /\n (sqrt(2*log(2))*sigma))))) + (1-eta)*dnorm(x, mu, sigma)) /\n sum( ((eta*(((gamma(1-alpha)) /\n ((x-mu)^2+(sqrt(2*log(2))*sigma)^2)^((1-alpha)/2)) *\n cos((pi*alpha/2)+(1-alpha)*atan((x-mu) /\n (sqrt(2*log(2))*sigma))))) + (1-eta)*dnorm(x, mu, sigma)))\n }\n y <- c(true_mix_ratio[1]*truncated_dsg(x = x,\n mu = true_mu[1],\n sigma = true_sigma[1],\n alpha = true_alpha[1],\n eta = true_eta[1])*10^degree +\n true_mix_ratio[2]*truncated_dsg(x = x,\n mu = true_mu[2],\n sigma = true_sigma[2],\n alpha = true_alpha[2],\n eta = true_eta[2])*10^degree +\n true_mix_ratio[3]*truncated_dsg(x = x,\n mu = true_mu[3],\n sigma = true_sigma[3],\n alpha = true_alpha[3],\n eta = true_eta[3])*10^degree)\n plot(y~x, main = \"genrated synthetic spectral data\")\n #Peak fitting by EMpeaksR\n #Initial values\n K <- 3\n mix_ratio_init <- c(0.2, 0.4, 0.4)\n mu_init <- c(20, 40, 70)\n sigma_init <- c(4, 3, 2)\n alpha_init <- c(0.3, 0.2, 0.4)\n eta_init <- c(0.5, 0.4, 0.3)\n #Coducting calculation\n SP_ECM_DSG_res <- spect_em_dsgmm(x = x,\n y = y,\n mu = mu_init,\n sigma = sigma_init,\n alpha = alpha_init,\n eta = eta_init,\n mix_ratio = mix_ratio_init,\n conv.cri = 1e-2,\n maxit = 2000)\n #Plot fitting curve and trace plot of parameters\n show_dsgmm_curve(SP_ECM_DSG_res,\n x,\n y,\n mix_ratio_init,\n mu_init,\n sigma_init,\n alpha_init,\n eta_init)\n #Showing the result of spect_em_dsgmm()\n print(cbind(c(mu_init),\n c(sigma_init),\n c(alpha_init),\n c(eta_init),\n c(mix_ratio_init)))\n print(cbind(SP_ECM_DSG_res$mu,\n SP_ECM_DSG_res$sigma,\n SP_ECM_DSG_res$alpha,\n SP_ECM_DSG_res$eta,\n SP_ECM_DSG_res$mix_ratio))\n print(cbind(true_mu,\n true_sigma,\n true_alpha,\n true_eta,\n true_mix_ratio))\n spect_em_gmm Spectrum adapted EM algorithm by GMM\nDescription\n Perform a peak fitting based on the spectrum adapted EM algorithm by Gaussian mixture model.\nUsage\n spect_em_gmm(x, y, mu, sigma, mix_ratio, conv.cri, maxit)\nArguments\n x measurement steps\n y intensity\n mu mean of the components\n sigma standard deviation of the components\n mix_ratio mixture ratio of the components\n conv.cri criterion of the convergence\n maxit maximum number of the iteration\nDetails\n Peak fitting is conducted by spectrum adapted EM algorithm.\nValue\n mu estimated mean of the components\n sigma estimated standard deviation of the components\n mix_ratio estimated mixture ratio of the components\n it number of the iteration to reach the convergence\n LL variation of the weighted log likelihood values\n MU variation of mu\n SIGMA variation of sigma\n MIX_RATIO variation of mix_ratio\n W_K decomposed component of the spectral data\n convergence message for the convergence in the calculation\n cal_time calculation time to complete the peak fitting. Unit is seconds\nReferences\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2019). Spectrum adapted\n expectation-maximization algorithm for high-throughput peak shift analysis. Science and technol-\n ogy of advanced materials, 20(1), 733-745.\nExamples\n #generating the synthetic spectral data based on three component Gausian mixture model.\n x <- seq(0, 100, by = 0.5)\n true_mu <- c(35, 50, 65)\n true_sigma <- c(3, 3, 3)\n true_mix_ratio <- rep(1/3, 3)\n degree <- 4\n y <- c(true_mix_ratio[1] * dnorm(x = x, mean = true_mu[1], sd = true_sigma[1])*10^degree +\n true_mix_ratio[2] * dnorm(x = x, mean = true_mu[2], sd = true_sigma[2])*10^degree +\n true_mix_ratio[3] * dnorm(x = x, mean = true_mu[3], sd = true_sigma[3])*10^degree)\n plot(y~x, main = \"genrated synthetic spectral data\")\n #Peak fitting by EMpeaksR\n #Initial values\n K <- 3\n mix_ratio_init <- c(0.2, 0.4, 0.4)\n mu_init <- c(20, 40, 70)\n sigma_init <- c(2, 5, 4)\n #Coducting calculation\n SP_EM_G_res <- spect_em_gmm(x, y, mu = mu_init, sigma = sigma_init, mix_ratio = mix_ratio_init,\n conv.cri = 1e-2, maxit = 2000)\n #Plot fitting curve and trace plot of parameters\n show_gmm_curve(SP_EM_G_res, x, y, mix_ratio_init, mu_init, sigma_init)\n #Showing the result of spect_em_gmm()\n print(cbind(c(mu_init), c(sigma_init), c(mix_ratio_init)))\n print(cbind(SP_EM_G_res$mu, SP_EM_G_res$sigma, SP_EM_G_res$mix_ratio))\n print(cbind(true_mu, true_sigma, true_mix_ratio))\n spect_em_lmm Spectrum adapted ECM algorithm by LMM\nDescription\n Perform a peak fitting based on the spectrum adapted ECM algorithm by Lorentz mixture model.\nUsage\n spect_em_lmm(x, y, mu, gam, mix_ratio, conv.cri, maxit)\nArguments\n x measurement steps\n y intensity\n mu mean of the components\n gam scale parameter of the components\n mix_ratio mixture ratio of the components\n conv.cri criterion of the convergence\n maxit maximum number of the iteration\nDetails\n Peak fitting is conducted by spectrum adapted ECM algorithm.\nValue\n mu estimated mean of the components\n gam estimated scale parameter of the components\n mix_ratio estimated mixture ratio of the components\n it number of the iteration to reach the convergence\n LL variation of the weighted log likelihood values\n MU variation of mu\n GAM variation of gam\n MIX_RATIO variation of mix_ratio\n W_K decomposed component of the spectral data\n convergence message for the convergence in the calculation\n cal_time calculation time to complete the peak fitting. Unit is seconds\nReferences\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2019). Spectrum adapted\n expectation-maximization algorithm for high-throughput peak shift analysis. Science and technol-\n ogy of advanced materials, 20(1), 733-745.\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2021). Spectrum adapted\n expectation-conditional maximization algorithm for extending high–throughput peak separation\n method in XPS analysis. Science and Technology of Advanced Materials: Methods, 1(1), 45-55.\nExamples\n #generating the synthetic spectral data based on three component Lorentz mixture model.\n x <- seq(0, 100, by = 0.5)\n true_mu <- c(35, 50, 65)\n true_gam <- c(3, 3, 3)\n true_mix_ratio <- rep(1/3, 3)\n degree <- 4\n #Normalized Lorentz distribution\n dCauchy <- function(x, mu, gam) {\n (dcauchy(x, mu, gam)) / sum(dcauchy(x, mu, gam))\n }\n y <- c(true_mix_ratio[1] * dCauchy(x = x, mu = true_mu[1], gam = true_gam[1])*10^degree +\n true_mix_ratio[2] * dCauchy(x = x, mu = true_mu[2], gam = true_gam[2])*10^degree +\n true_mix_ratio[3] * dCauchy(x = x, mu = true_mu[3], gam = true_gam[3])*10^degree)\n plot(y~x, main = \"genrated synthetic spectral data\")\n #Peak fitting by EMpeaksR\n #Initial values\n K <- 3\n mix_ratio_init <- c(0.2, 0.4, 0.4)\n mu_init <- c(20, 40, 70)\n gam_init <- c(2, 5, 4)\n #Coducting calculation\n SP_ECM_L_res <- spect_em_lmm(x, y, mu = mu_init, gam = gam_init, mix_ratio = mix_ratio_init,\n conv.cri = 1e-2, maxit = 2000)\n #Plot fitting curve and trace plot of parameters\n show_lmm_curve(SP_ECM_L_res, x, y, mix_ratio_init, mu_init, gam_init)\n #Showing the result of spect_em_lmm()\n print(cbind(c(mu_init), c(gam_init), c(mix_ratio_init)))\n print(cbind(SP_ECM_L_res$mu, SP_ECM_L_res$gam, SP_ECM_L_res$mix_ratio))\n print(cbind(true_mu, true_gam, true_mix_ratio))\n spect_em_pvmm Spectrum adapted ECM algorithm by PVMM\nDescription\n Perform a peak fitting based on the spectrum adapted ECM algorithm by Pseudo-Voigt mixture\n model.\nUsage\n spect_em_pvmm(x, y, mu, sigma, eta, mix_ratio, conv.cri, maxit)\nArguments\n x measurement steps\n y intensity\n mu mean of the components\n sigma standard deviation of the components\n eta mixing ratio of Gauss and Lorentz distribution\n mix_ratio mixture ratio of the components\n conv.cri criterion of the convergence\n maxit maximum number of the iteration\nDetails\n Peak fitting is conducted by spectrum adapted ECM algorithm.\nValue\n mu estimated mean of the components\n sigma estimated standard deviation of the components\n eta estimated mixing ratio of Gauss and Lorentz distribution\n mix_ratio estimated mixture ratio of the components\n it number of the iteration to reach the convergence\n LL variation of the weighted log likelihood values\n MU variation of mu\n SIGMA variation of sigma\n ETA variation of beta\n MIX_RATIO variation of mix_ratio\n W_K decomposed component of the spectral data\n convergence message for the convergence in the calculation\n cal_time calculation time to complete the peak fitting. Unit is seconds\nReferences\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2019). Spectrum adapted\n expectation-maximization algorithm for high-throughput peak shift analysis. Science and technol-\n ogy of advanced materials, 20(1), 733-745.\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2021). Spectrum adapted\n expectation-conditional maximization algorithm for extending high–throughput peak separation\n method in XPS analysis. Science and Technology of Advanced Materials: Methods, 1(1), 45-55.\nExamples\n #generating the synthetic spectral data based on three component Pseudo-Voigt mixture model.\n x <- seq(0, 100, by = 0.5)\n true_mu <- c(35, 50, 65)\n true_sigma <- c(3, 3, 3)\n true_eta <- c(0.3, 0.8, 0.5)\n true_mix_ratio <- rep(1/3, 3)\n degree <- 4\n #Normalized Pseudo-Voigt distribution\n truncated_pv <- function(x, mu, sigma, eta) {\n (eta*dcauchy(x, mu, sqrt(2*log(2))*sigma) + (1-eta)*dnorm(x, mu, sigma)) /\n sum(eta*dcauchy(x, mu, sqrt(2*log(2))*sigma) + (1-eta)*dnorm(x, mu, sigma))\n }\n y <- c(true_mix_ratio[1]*truncated_pv(x = x,\n mu = true_mu[1],\n sigma = true_sigma[1],\n eta = true_eta[1])*10^degree +\n true_mix_ratio[2]*truncated_pv(x = x,\n mu = true_mu[2],\n sigma = true_sigma[2],\n eta = true_eta[2])*10^degree +\n true_mix_ratio[3]*truncated_pv(x = x,\n mu = true_mu[3],\n sigma = true_sigma[3],\n eta = true_eta[3])*10^degree)\n plot(y~x, main = \"genrated synthetic spectral data\")\n #Peak fitting by EMpeaksR\n #Initial values\n K <- 3\n mix_ratio_init <- c(0.2, 0.4, 0.4)\n mu_init <- c(20, 40, 70)\n sigma_init <- c(2, 5, 4)\n eta_init <- c(0.5, 0.4, 0.3)\n #Coducting calculation\n SP_ECM_PV_res <- spect_em_pvmm(x = x,\n y = y,\n mu = mu_init,\n sigma = sigma_init,\n eta = eta_init,\n mix_ratio = mix_ratio_init,\n conv.cri = 1e-2,\n maxit = 2000)\n #Plot fitting curve and trace plot of parameters\n show_pvmm_curve(SP_ECM_PV_res, x, y, mix_ratio_init, mu_init, sigma_init, eta_init)\n #Showing the result of spect_em_pvmm()\n print(cbind(c(mu_init), c(sigma_init), c(eta_init), c(mix_ratio_init)))\n print(cbind(SP_ECM_PV_res$mu, SP_ECM_PV_res$sigma, SP_ECM_PV_res$eta, SP_ECM_PV_res$mix_ratio))\n print(cbind(true_mu, true_sigma, true_eta, true_mix_ratio))\n spect_em_pvmm_lback Spectrum adapted ECM algorithm by PVMM with a linear back-\n ground\nDescription\n Perform a peak fitting based on the spectrum adapted ECM algorithm by pseudo-Voigt mixture\n model with a linear background.\nUsage\n spect_em_pvmm_lback(x, y, mu, sigma, eta, mix_ratio, x_lower, x_upper, conv.cri, maxit)\nArguments\n x measurement steps\n y intensity\n mu mean of the components\n sigma standard deviation of the components\n eta mixing ratio of Gauss and Lorentz distribution\n mix_ratio mixture ratio of the components\n x_lower lower limit of the measurement steps. Default is a minimum of x\n x_upper upper limit of the measurement steps. Default is a maximum of x\n conv.cri criterion of the convergence\n maxit maximum number of the iteration\nDetails\n Peak fitting is conducted by spectrum adapted ECM algorithm.\nValue\n mu estimated mean of the components\n sigma estimated standard deviation of the components\n eta estimated mixing ratio of Gauss and Lorentz distribution\n mix_ratio estimated mixture ratio of the components\n it number of the iteration to reach the convergence\n LL variation of the weighted log likelihood values\n MU variation of mu\n SIGMA variation of sigma\n ETA variation of beta\n MIX_RATIO variation of mix_ratio\n W_K decomposed component of the spectral data\n convergence message for the convergence in the calculation\n cal_time calculation time to complete the peak fitting. Unit is seconds\nReferences\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2019). Spectrum adapted\n expectation-maximization algorithm for high-throughput peak shift analysis. Science and technol-\n ogy of advanced materials, 20(1), 733-745.\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2021). Spectrum adapted\n expectation-conditional maximization algorithm for extending high–throughput peak separation\n method in XPS analysis. Science and Technology of Advanced Materials: Methods, 1(1), 45-55.\n <NAME>., <NAME>., <NAME>., <NAME>., & <NAME>. (2023). High-throughput XPS\n spectrum modeling with autonomous background subtraction for 3 d 5/2 peak mapping of SnS.\n Science and Technology of Advanced Materials: Methods, 3(1), 2159753.\nExamples\n #generating the synthetic spectral data based on three component Pseudo-Voigt mixture model.\n x <- seq(0, 100, by = 0.5)\n K <- 3\n true_mu <- c(35, 50, 65)\n true_sigma <- c(3, 3, 3)\n true_mix_ratio <- c(0.5/3, 0.5/3, 0.5/3, 0.5)\n true_eta <- c(0.4, 0.6, 0.1)\n degree <- 4\n #Normalized Pseudo-Voigt distribution\n truncated_pv <- function(x, mu, sigma, eta) {\n (eta*dcauchy(x, mu, sqrt(2*log(2))*sigma) + (1-eta)*dnorm(x, mu, sigma)) /\n sum(eta*dcauchy(x, mu, sqrt(2*log(2))*sigma) + (1-eta)*dnorm(x, mu, sigma))\n }\n y <- c(true_mix_ratio[1]*truncated_pv(x = x,\n mu = true_mu[1],\n sigma = true_sigma[1],\n eta = true_eta[1])*10^degree +\n true_mix_ratio[2]*truncated_pv(x = x,\n mu = true_mu[2],\n sigma = true_sigma[2],\n eta = true_eta[2])*10^degree +\n true_mix_ratio[3]*truncated_pv(x = x,\n mu = true_mu[3],\n sigma = true_sigma[3],\n eta = true_eta[3])*10^degree +\n true_mix_ratio[4]*(c(500*x + 15000) / sum(500*x + 15000))*10^degree)\n plot(y~x, main = \"genrated synthetic spectral data\")\n #Peak fitting by EMpeaksR\n #Initial values\n mu_init <- c(30, 40, 60)\n sigma_init <- c(4, 4, 4)\n mix_ratio_init <- rep(1/(length(mu_init)+3), length(mu_init)+3)\n eta_init <- c(1, 1, 1)\n #Coducting calculation\n SP_ECM_PV_LBACK_res <- spect_em_pvmm_lback(x = x,\n y = y,\n mu = mu_init,\n sigma = sigma_init,\n eta = eta_init,\n mix_ratio = mix_ratio_init,\n x_lower = min(x),\n x_upper = max(x),\n conv.cri = 1e-2,\n maxit = 2000)\n #Plot fitting curve and trace plot of parameters\n show_pvmm_lback_curve(spect_em_pvmm_lback_res = SP_ECM_PV_LBACK_res,\n x = x,\n y = y,\n mix_ratio_init = mix_ratio_init,\n mu_init = mu_init,\n sigma_init = sigma_init,\n eta_init = eta_init,\n x_lower = min(x),\n x_upper = max(x))\n #Showing the result of spect_em_pvmm_lback()\n print(cbind(SP_ECM_PV_LBACK_res$mu, SP_ECM_PV_LBACK_res$sigma, SP_ECM_PV_LBACK_res$eta,\n SP_ECM_PV_LBACK_res$mix_ratio[1:K]))\n print(cbind(true_mu, true_sigma, true_eta, true_mix_ratio[1:K]))"}}},{"rowIdx":59184,"cells":{"project":{"kind":"string","value":"@mattleff/react-native-track-player"},"source":{"kind":"string","value":"npm"},"language":{"kind":"string","value":"JavaScript"},"content":{"kind":"string","value":"### 📢 The Android side of RN track player is currently being rewritten with Kotlin and [KotlinAudio](https://github.com/DoubleSymmetry/KotlinAudio). Help us test! [More information here.](https://github.com/DoubleSymmetry/react-native-track-player/discussions/1264)\n\n---\n\nA fully-fledged audio module created for music apps. Provides audio playback, external media controls, background mode and more!\n\n* [Documentation](https://react-native-track-player.js.org)\n\t+ [Installation](https://react-native-track-player.js.org/docs/basics/installation/)\n\t+ [Getting Started](https://react-native-track-player.js.org/docs/basics/getting-started/)\n\t+ [API Docs](https://react-native-track-player.js.org/docs/api/events)\n\t+ [Platform Support](https://react-native-track-player.js.org/docs/basics/platform-support)\n\t+ [Background Mode](https://react-native-track-player.js.org/docs/basics/background-mode)\n\t+ [Build Preferences](https://react-native-track-player.js.org/docs/basics/build-preferences)\n\t+ [v2 Migration Guide](https://react-native-track-player.js.org/docs/v2-migration)\n* [Sponsors](#sponsors)\n* [Features](#features)\n* [Why another music module?](#why-another-music-module)\n* [Example Setup](#example-setup)\n* [Core Team ✨](#core-team-)\n* [Special Thanks ✨](#special-thanks-)\n* [I Have A Bug/Feature Request](#contributing)\n* [Community](#Community)\n\nNot sure where to start?\n\n1. Try [Getting Started](https://react-native-track-player.js.org/docs/basics/getting-started).\n2. Peruse the [API Docs](https://react-native-track-player.js.org/docs/api/events).\n3. Run the [Example Project](https://github.com/react-native-kit/react-native-track-player/blob/HEAD/example).\n\nSponsors\n---\n\nreact-native-track-player is made possible by the generosity of the sponsors below, and many other [individual backers](https://react-native-track-player.js.org/docs/sponsors#backers). Sponsoring directly impacts the longevity of this project.\n\n#### 🥇 Gold sponsors ($2000+ total contributions)\n\n| |\n| --- |\n| [**Radio Garden**](http://radio.garden/) |\n\n#### 🥈 Silver sponsors ($200+ per month)\n\n| | |\n| --- | --- |\n| [**Voxist**](http://www.voxist.com/) | [**Evergrace**](https://evergrace.co) |\n\n#### 🥉 Bronze sponsors ($500+ total contributions)\n\n[Become the first bronze sponsor!](https://github.com/sponsors/DoubleSymmetry)\n\n#### ✨ Contributing sponsors ($25+ per month)\n\n| |\n| --- |\n| \n**Podverse** |\n\n---\n\nFeatures\n---\n\n* **Lightweight** - Optimized to use the least amount of resources according to your needs\n* **Feels native** - As everything is built together, it follows the same design principles as real music apps do\n* **Multi-platform** - Supports Android, iOS and Windows\n* **Media Controls support** - Provides events for controlling the app from a Bluetooth device, the lock screen, a notification, a smartwatch or even a car\n* **Local or network, files or streams** - It doesn't matter where the media belongs, we've got you covered\n* **Adaptive bitrate streaming support** - Support for DASH, HLS or SmoothStreaming\n* **Caching support** - Cache media files to play them again without an internet connection\n* **Background support** - Keep playing audio even after the app is in background\n* **Fully Customizable** - Even the notification icons are customizable!\n* **Supports React Hooks 🎣** - Includes React Hooks for common use-cases so you don't have to write them\n\nWhy another music module?\n---\n\nAfter trying to team up modules like `react-native-sound`, `react-native-music-controls` and `react-native-google-cast`, I've noticed, that their structure and the way should be tied together can cause a lot of problems (mainly on Android). Those can heavily affect the app stability and user experience.\n\nAll audio modules (like `react-native-sound`) don't play in a separated service on Android, which should **only** be used for simple audio tracks in the foreground (such as sound effects, voice messages, etc.)\n\n`react-native-music-controls` is meant for apps using those audio modules, but it has a few problems: the audio isn't tied directly to the controls. It can be pretty useful for casting (such as Chromecast).\n\n`react-native-google-cast` works pretty well and also supports custom receivers, but it has fewer player controls, it's harder to integrate and still uses the Cast SDK v2.\n\nExample Setup\n---\n\nFirst please take a look at the [Getting Started](https://react-native-track-player.js.org/docs/basics/getting-started/) guide, but a basic example of how to play a track:\n```\nimport TrackPlayer from 'react-native-track-player';\n\nconst start = async () => {\n // Set up the player\n await TrackPlayer.setupPlayer();\n\n // Add a track to the queue\n await TrackPlayer.add({\n id: 'trackId',\n url: require('track.mp3'),\n title: 'Track Title',\n artist: 'Track Artist',\n artwork: require('track.png')\n });\n\n // Start playing it\n await TrackPlayer.play();\n};\nstart();\n```\nCore Team ✨\n---\n\n| | | |\n| --- | --- | --- |\n| [**<NAME>**](https://github.com/dcvz) | [**<NAME>**](https://github.com/mpivchev) | [**<NAME>**](https://github.com/jspizziri) |\n\nSpecial Thanks ✨\n---\n\n| | |\n| --- | --- |\n| [**<NAME>**](https://github.com/Guichaguri) | [**<NAME>**](https://github.com/curiousdustin) |\n\nContributing\n---\n\nYou want this package to be awesome and we want to deliver on that. As you know already you can just [File A Ticket](#file-a-ticket), but thats not actually the best way for you to get what you need (read on to see why). The best way is for you to [Be A Champion](#be-a-champion) and [dive into the code](#where-do-you-start).\n\n#### File A Ticket\n\nThe reality is that filing a ticket isn't always enough. **This is probably only going to work if your issue aligns with both the interests *and* the resources available** to the core team. Here are the things that align with our *interests*\nin order of priority.\n\n1. Fixing ***widespread, common, and critical Bugs***.\n2. Fixing ***uncommon but necessary Bugs***.\n3. Introducing new ***Features that have broad value***.\n\nNow keep in mind available resources. Long story short, the thing you care about needs to be cared about by either a lot of other people, or by us.\n\n**BUT!** There's another and, arguably even ***better way*** that helps you get what you need faster: [Be A Champion](#be-a-champion).\n\n#### Be A Champion\n\nBeing a *champion* makes it easy for us to help you. Which is what we all want!\nSo how can you be a champion? [Sponsor the Project](https://github.com/sponsors/DoubleSymmetry) or *be willing to write some code*.\n\n**If *you're willing*** to write some code ***we're willing*** to:\n\n* Open a design discussion, give feedback, and approve something that works.\n* Provide guidance in the implementation journey.\n\nSo, in a nutshell, let us know you're willing to do the work and ask for a little guidance, and watch the things you care about get done faster than anyone else.\nThe best help will be given to those who are willing to help themselves.\n\n###### You don't have experience you say? It's OK!\n\nYou may be thinking that you can't help because you know nothing about native iOS or Android or maybe even React code. But we're willing to help guide you.\n\nIf you're up for that task then we can help you understand native code and how React Native works.\n\nThe only way you go from *not-knowing* to *knowing* is by learning. Learning isn't something you should be ashamed of nor is it something you should be scared of.\n\n#### Where Do You Start?\n\nOur goal is to make it as easy as possible for you to make changes to the library.\nAll the documentation on how to work on the library and it's dependencies is\n[located in this Guide](https://github.com/react-native-kit/react-native-track-player/blob/HEAD/example/README.md)\n\nCommunity\n---\n\nYou can find us as part of the [React Native Track Player](https://discordapp.com/invite/ya2XDCR):\n\n* `# introductions` - Come greet the newest members of this group!\n* `# support` - Ask members of the community to trouble shoot issues with your app and make recommendations.\n* `# app-anouncements` - Tell the community about the app you made with this project!\n* `# releases` - Stay updated about the latest releases and dev efforts on the project.\nReadme\n---\n\n### Keywords\n\n* react\n* react-native\n* track-player\n* audio-player\n* audio\n* player\n* music\n* controls\n* chromecast\n* android\n* ios\n* windows\n* hooks"}}},{"rowIdx":59185,"cells":{"project":{"kind":"string","value":"kraftkit.sh/unikraft"},"source":{"kind":"string","value":"go"},"language":{"kind":"string","value":"Go"},"content":{"kind":"string","value":"None\n\nDocumentation\n [¶](#section-documentation)\n---\n\n \n### Overview [¶](#pkg-overview)\n\nSPDX-License-Identifier: BSD-3-Clause Copyright (c) 2022, Unikraft GmbH and The KraftKit Authors.\nLicensed under the BSD-3-Clause License (the \"License\").\nYou may not use this file except in compliance with the License.\n\nSPDX-License-Identifier: BSD-3-Clause Copyright (c) 2022, Unikraft GmbH and The KraftKit Authors.\nLicensed under the BSD-3-Clause License (the \"License\").\nYou may not use this file except in compliance with the License.\n\nSPDX-License-Identifier: BSD-3-Clause Copyright (c) 2022, Unikraft GmbH and The KraftKit Authors.\nLicensed under the BSD-3-Clause License (the \"License\").\nYou may not use this file except in compliance with the License.\n\n### Index [¶](#pkg-index)\n\n* [Constants](#pkg-constants)\n* [func ComponentTypes() map[string]ComponentType](#ComponentTypes)\n* [func IsFileUnikraftUnikernel(path string) (bool, error)](#IsFileUnikraftUnikernel)\n* [func PlaceComponent(workdir string, t ComponentType, name string) (string, error)](#PlaceComponent)\n* [func TypeNameVersion(entity Nameable) string](#TypeNameVersion)\n* [func WithContext(ctx context.Context, val *Context) context.Context](#WithContext)\n* [type ComponentType](#ComponentType)\n* + [func GuessTypeNameVersion(input string) (ComponentType, string, string, error)](#GuessTypeNameVersion)\n* + [func (ct ComponentType) Plural() string](#ComponentType.Plural)\n* [type Context](#Context)\n* + [func FromContext(ctx context.Context) *Context](#FromContext)\n* [type Nameable](#Nameable)\n* [type ProvidedSyscall](#ProvidedSyscall)\n* + [func NewProvidedSyscall(export string) *ProvidedSyscall](#NewProvidedSyscall)\n\n### Constants [¶](#pkg-constants)\n\n```\nconst (\n\t// Environmental variables recognized by Unikraft's build system.\n UK_NAME = \"CONFIG_UK_NAME\"\n UK_DEFNAME = \"CONFIG_UK_DEFNAME\"\n UK_CONFIG = \"CONFIG_UK_CONFIG\"\n UK_FULLVERSION = \"CONFIG_UK_FULLVERSION\"\n UK_CODENAME = \"CONFIG_UK_CODENAME\"\n UK_ARCH = \"CONFIG_UK_ARCH\"\n UK_BASE = \"CONFIG_UK_BASE\"\n UK_APP = \"CONFIG_UK_APP\"\n UK_DEFCONFIG = \"UK_DEFCONFIG\"\n KCONFIG_APP_DIR = \"KCONFIG_APP_DIR\"\n KCONFIG_LIB_DIR = \"KCONFIG_LIB_DIR\"\n KCONFIG_LIB_IN = \"KCONFIG_LIB_IN\"\n KCONFIG_PLAT_DIR = \"KCONFIG_PLAT_DIR\"\n KCONFIG_PLAT_IN = \"KCONFIG_PLAT_IN\"\n\n\t// Filenames which represent ecosystem files\n Config_uk = \"Config.uk\"\n Exportsyms_uk = \"exportsyms.uk\"\n Linker_uk = \"Linker.uk\"\n Localsyms_uk = \"localsyms.uk\"\n Makefile_uk = \"Makefile.uk\"\n\n // Standard Makefile.uk variables\n\tUK_PROVIDED_SYSCALLS = \"UK_PROVIDED_SYSCALLS\"\n\n\t// Built-in paths\n VendorDir = \".unikraft\"\n BuildDir = \".unikraft/build\"\n)\n```\n### Variables [¶](#pkg-variables)\n\nThis section is empty.\n\n### Functions [¶](#pkg-functions)\n\n#### \nfunc [ComponentTypes](https://github.com/unikraft/kraftkit/blob/v0.6.7/unikraft/type.go#L26) [¶](#ComponentTypes)\n```\nfunc ComponentTypes() map[[string](/builtin#string)][ComponentType](#ComponentType)\n```\n#### \nfunc [IsFileUnikraftUnikernel](https://github.com/unikraft/kraftkit/blob/v0.6.7/unikraft/utils.go#L14) [¶](#IsFileUnikraftUnikernel)\n\nadded in v0.6.0\n```\nfunc IsFileUnikraftUnikernel(path [string](/builtin#string)) ([bool](/builtin#bool), [error](/builtin#error))\n```\nIsFileUnikraftUnikernel is a utility method that determines whether the provided input file is a Unikraft unikernel. The file is checked with a number of known facts about the kernel image built with Unikraft.\n\n#### \nfunc [PlaceComponent](https://github.com/unikraft/kraftkit/blob/v0.6.7/unikraft/type.go#L94) [¶](#PlaceComponent)\n```\nfunc PlaceComponent(workdir [string](/builtin#string), t [ComponentType](#ComponentType), name [string](/builtin#string)) ([string](/builtin#string), [error](/builtin#error))\n```\nPlaceComponent is a universal source of truth for identifying the path to place a component\n\n#### \nfunc [TypeNameVersion](https://github.com/unikraft/kraftkit/blob/v0.6.7/unikraft/type.go#L111) [¶](#TypeNameVersion)\n\nadded in v0.4.0\n```\nfunc TypeNameVersion(entity [Nameable](#Nameable)) [string](/builtin#string)\n```\nTypeNameVersion returns the canonical name of the component using the format\n<TYPE>/<NAME>:<VERSION#### \nfunc [WithContext](https://github.com/unikraft/kraftkit/blob/v0.6.7/unikraft/context.go#L22) [¶](#WithContext)\n\nadded in v0.4.0\n```\nfunc WithContext(ctx [context](/context).[Context](/context#Context), val *[Context](#Context)) [context](/context).[Context](/context#Context)\n```\nWithContext\n\n### Types [¶](#pkg-types)\n\n#### \ntype [ComponentType](https://github.com/unikraft/kraftkit/blob/v0.6.7/unikraft/type.go#L15) [¶](#ComponentType)\n```\ntype ComponentType [string](/builtin#string)\n```\n```\nconst (\n ComponentTypeUnknown [ComponentType](#ComponentType) = \"unknown\"\n ComponentTypeCore [ComponentType](#ComponentType) = \"core\"\n ComponentTypeArch [ComponentType](#ComponentType) = \"arch\"\n ComponentTypePlat [ComponentType](#ComponentType) = \"plat\"\n ComponentTypeLib [ComponentType](#ComponentType) = \"lib\"\n ComponentTypeApp [ComponentType](#ComponentType) = \"app\"\n)\n```\n#### \nfunc [GuessTypeNameVersion](https://github.com/unikraft/kraftkit/blob/v0.6.7/unikraft/type.go#L64) [¶](#GuessTypeNameVersion)\n```\nfunc GuessTypeNameVersion(input [string](/builtin#string)) ([ComponentType](#ComponentType), [string](/builtin#string), [string](/builtin#string), [error](/builtin#error))\n```\nGuessNameAndType attempts to parse the input string, which could be formatted such that its type, name and version are present\n\n#### \nfunc (ComponentType) [Plural](https://github.com/unikraft/kraftkit/blob/v0.6.7/unikraft/type.go#L40) [¶](#ComponentType.Plural)\n```\nfunc (ct [ComponentType](#ComponentType)) Plural() [string](/builtin#string)\n```\n#### \ntype [Context](https://github.com/unikraft/kraftkit/blob/v0.6.7/unikraft/context.go#L15) [¶](#Context)\n\nadded in v0.4.0\n```\ntype Context struct {\n UK_NAME [string](/builtin#string)\n UK_BASE [string](/builtin#string)\n BUILD_DIR [string](/builtin#string)\n}\n```\nContext is a general-purpose context container for use within the Unikraft package. It mimics the environmental variables within Unikraft's main build system.\n\n#### \nfunc [FromContext](https://github.com/unikraft/kraftkit/blob/v0.6.7/unikraft/context.go#L26) [¶](#FromContext)\n\nadded in v0.4.0\n```\nfunc FromContext(ctx [context](/context).[Context](/context#Context)) *[Context](#Context)\n```\n#### \ntype [Nameable](https://github.com/unikraft/kraftkit/blob/v0.6.7/unikraft/type.go#L51) [¶](#Nameable)\n\nadded in v0.4.0\n```\ntype Nameable interface {\n // Type returns the entity's static component type.\n\tType() [ComponentType](#ComponentType)\n\n // Name returns the entity name.\n\tName() [string](/builtin#string)\n\n // Version returns the entity version.\n\tVersion() [string](/builtin#string)\n}\n```\nNameable represents an abstract interface which can be cast to structures which contain canonical information about a component. This allows us to generate a string representation of the entity.\n\n#### \ntype [ProvidedSyscall](https://github.com/unikraft/kraftkit/blob/v0.6.7/unikraft/syscall.go#L14) [¶](#ProvidedSyscall)\n\nadded in v0.4.0\n```\ntype ProvidedSyscall struct {\n Name [string](/builtin#string)\n Nargs [uint](/builtin#uint)\n}\n```\nProvidedSyscall is a simple structure which contains the syscall name and the number of arguments which are accepted.\n\n#### \nfunc [NewProvidedSyscall](https://github.com/unikraft/kraftkit/blob/v0.6.7/unikraft/syscall.go#L21) [¶](#NewProvidedSyscall)\n\nadded in v0.4.0\n```\nfunc NewProvidedSyscall(export [string](/builtin#string)) *[ProvidedSyscall](#ProvidedSyscall)\n```\nNewProvidedSyscall converts an exported `UK_PROVIDED_SYSCALL` entry into a standard `ProvidedSyscall` structure."}}},{"rowIdx":59186,"cells":{"project":{"kind":"string","value":"github.com/nicksherron/bashhub-server"},"source":{"kind":"string","value":"go"},"language":{"kind":"string","value":"Go"},"content":{"kind":"string","value":"README\n [¶](#section-readme)\n---\n\n### bashhub-server\n\n[](https://goreportcard.com/report/github.com/nicksherron/bashhub-server) [](https://travis-ci.org/nicksherron/bashhub-server) [](https://gitter.im/bashhub_server/community?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge)\n\nbashhub-server is a private cloud alternative for [bashhub-client](https://github.com/rcaloras/bashhub-client) with some added features like regex search.\n\n#### Features\n\n* Very simple drop in replacement for bashhub.com server and easy to [install](https://github.com/nicksherron/bashhub-server#installation) and get running with existing bashhub-client and bh command.\n* All the benefits of bashhub without having to send your shell history to a third-party\n* [Regex](https://github.com/nicksherron/bashhub-server#using-regex) search\n* [Import](https://github.com/nicksherron/bashhub-server#transferring-history-from-bashhubcom) old history from bashhub.com\n* Quickly connect any client with access to your server bashhub-sever address/port.\n* Written in Go so it's fast and is actively maintained\n* Salt hashed password encryption and jwt authentication protected endpoints\n\n#### Why?\n\nI love the idea behind bashhub. Having my shell history centralized and queryable from various systems whether it be my home computer or from an ssh session on a server is great. However, even with encryption,\nI was a little leary of sending my shell commands to a third-party server, so bashhub-server was created.\n\n#### Installation\n\n###### Homebrew or Linuxbrew\n```\n$ brew install bashhub-server/latest/bashhub-server\n```\n###### Docker\n```\n$ docker pull nicksherron/bashhub-server\n```\n###### Go\n\ngo modules are required\n```\n$ GO111MODULE=on go get -u github.com/nicksherron/bashhub-server\n```\n###### Releases\n\nBinaries for various os and architectures can be found in [releases](https://github.com/nicksherron/bashhub-server/releases).\nIf your system is not listed just submit an issue requesting your os and architecture.\n\n#### Usage\n```\n$ bashhub-server --help\n\nUsage:\n [flags]\n [command]\n\nAvailable Commands:\n help Help about any command\n transfer Transfer bashhub history from one server to another\n version Print the version number and build info\n\nFlags:\n -a, --addr string Ip and port to listen and serve on. (default \"http://0.0.0.0:8080\")\n --db string db location (sqlite or postgres)\n -h, --help help for this command\n --log string Set filepath for HTTP log. \"\" logs to stderr.\n\nUse \" [command] --help\" for more information about a command.\n```\n##### Running\n\nJust run the server\n```\n$ bashhub-server\n\n _ _ _ _\n| | | | | | | |\t\tversion: v0.2.1\n| |__ __ _ ___| |__ | |__ _ _| |\t\taddress: http://0.0.0.0:8080\n| '_ \\ / _' / __| '_ \\| '_ \\| | | | '_ \\\n| |_) | (_| __ \\ | | | | | | |_| | |_) |\n|_.__/ __,_|___/_| |_|_| |_|__,_|_.__/\n ___ ___ _ ____ _____ _ __\n/ __|/ _ \\ '__\\ \\ / / _ \\ '__|\n__ \\ __/ | \\ V / __/ |\n|___/___|_| _/ ___|_|\n\n2020/02/10 03:04:11 Listening and serving HTTP on http://0.0.0.0:8080\n```\nor on docker (backend database files are persisted in container volume `/data`)\n```\n$ docker run -d -p 8080:8080 --name bashhub-server nicksherron/bashhub-server \n```\nThen add `export BH_URL=http://localhost:8080` (or whatever you set your bashhub-server address to) to your .zshrc or .bashrc\n```\n$ echo \"export BH_URL=http://localhost:8080\" >> ~/.bashrc\n```\nor\n```\n$ echo \"export BH_URL=http://localhost:8080\" >> ~/.zshrc\n```\nThats it! Restart your shell and re-run bashhub setup.\n```\n$ $SHELL && bashhub setup\n```\n##### Changing default db\n\nBy default the backend db uses sqlite, with the location for each os shown below.\n\n| os | default |\n| --- | --- |\n| Unix | $XDG_CONFIG_HOME/bashhub-server/data.db OR $HOME/.config/bashhub-server/data.db |\n| Darwin | $HOME/Library/Application Support/bashhub-server/data.db |\n| Windows | %AppData%\\bashhub-server\\data.db |\n| Plan 9 | $home/lib/bashhub-server/data.db |\n\nTo set a different sqlite db file to use, run\n```\n$ bashhub-server --db path/to/file.db\n```\nPostgresql is also supported by bashhub-server. To use postgres specify the postgres uri in the --db flag with the following format\n```\n$ bashhub-server --db \"postgres://user:password@localhost:5432?sslmode=disable\"\n```\n##### Using Regex\n\nbashhub-server supports regex queries sent by the bh command (bashhub-client)\n\nWithout regex\n```\n$ bh bash\n\nbashhub setup docker pull nicksherron/bashhub-server bin/bashhub-server version untar bashhub-server_v0.1.0_darwin_amd64.tar.gz cd bashhub-server_v0.1.0_darwin_amd64\n./bashhub-server version make build && bin/bashhub-server cd bashhub-server brew install bashhub-server/latest/bashhub-server bashhub-server version bashhub-server --help\n```\nWith regex\n```\n$ bh \"^bash\"\n\nbashhub setup bashhub-server version bashhub-server --help\n```\nall commands with only 6 letters\n```\n$ bh \"^[a-zA-Z]{6}$\"\n\nwhoami ggpush\ngoland ggpull\n```\n##### Transferring history from bashhub.com\n\nYou can transfer your command history from one server to another with then `bashhub-server transfer`\ncommand.\n```\n$ bashhub-server transfer \\\n --src-user 'user' \\\n --src-pass 'password' \\\n --dst-user 'user' \\\n --dst-pass 'password' \n\ntransferring 872 / 8909 [-->____________________] 9.79% 45 inserts/sec\n```\nIf you're transferring from Bashhub.com they have a rate limit of 10 requests a seconds and you are limited to your last 10,000 commands.\n\nDocumentation\n [¶](#section-documentation)\n---\n\n\nThere is no documentation for this package."}}},{"rowIdx":59187,"cells":{"project":{"kind":"string","value":"SNSchart"},"source":{"kind":"string","value":"cran"},"language":{"kind":"string","value":"R"},"content":{"kind":"string","value":"Package ‘SNSchart’\n October 12, 2022\nType Package\nTitle Sequential Normal Scores in Statistical Process Management\nVersion 1.4.0\nAuthor <NAME> [aut],\n <NAME> [aut, cre],\n <NAME> [ctb]\nMaintainer <NAME> <<EMAIL>>\nDescription The methods discussed in this package are new non-parametric methods\n based on sequential normal scores 'SNS' (Conover et al (2017)\n <doi:10.1080/07474946.2017.1360091>), designed for sequences of observations,\n usually time series data, which may occur singly or in batches,\n and may be univariate or multivariate. These methods are designed\n to detect changes in the process, which may occur as changes in location\n (mean or median), changes in scale (standard deviation, or variance), or\n other changes of interest in the distribution of the observations,\n over the time observed. They usually apply to large data sets,\n so computations need to be simple enough to be done in a reasonable\n time on a computer, and easily updated as each new observation\n (or batch of observations) becomes available. Some examples and more detail\n in 'SNS' is presented in the work by Conover et al (2019) <arXiv:1901.04443>.\nDepends R (>= 2.10)\nLicense MIT + file LICENSE\nEncoding UTF-8\nLazyData true\nRoxygenNote 7.1.1\nImports parallel, stats, MASS\nSuggests testthat, knitr, rmarkdown\nVignetteBuilder knitr\nNeedsCompilation no\nRepository CRAN\nDate/Publication 2021-04-07 08:40:15 UTC\nR topics documented:\ncalibrateControlLimi... 2\ndataAlignmen... 6\nexample4... 7\nexample6... 8\nexample7... 8\nexample74... 9\nexample74... 9\nexample74... 10\nexample8... 10\nexample8... 11\nexample8... 11\nexample8... 12\nexample8... 12\nexample8... 13\nexample9... 13\nexample9... 14\ngetAR... 14\ngetDis... 17\ngetQuantil... 19\ngetR... 21\nmcalibrateControlLimi... 24\nmgetAR... 26\nmgetDis... 28\nmgetR... 29\nMN... 31\nMSN... 33\nN... 35\nSN... 36\nsran... 39\n calibrateControlLimit Calibration of the control limit for the selected chart\nDescription\n The methodology used to calibrate the control limit for the SNS chart depending on the selected\n chart\nUsage\n calibrateControlLimit(\n targetARL = NULL,\n targetMRL = NULL,\n n,\n m,\n theta = NULL,\n Ftheta = NULL,\n scoring = \"Z\",\n Chi2corrector = \"None\",\n dist,\n mu,\n sigma,\n dist.par = c(0, 1, 1),\n chart,\n chart.par,\n replicates = 50000,\n isParallel = TRUE,\n maxIter = 20,\n progress = TRUE,\n alignment = \"unadjusted\",\n constant = NULL,\n absolute = FALSE,\n isFixed = FALSE,\n rounding.factor = NULL\n )\nArguments\n targetARL scalar. is the target ARL to calibrate. By default is set to NULL\n targetMRL scalar. is the target ARL to calibrate. By default is set to NULL\n n scalar. Subroup size\n m scalar. Reference sample size\n theta scalar. Value corresponig with the Ftheta quantile.\n Ftheta scalar. Quantile of the data distribution. The values that take are between (0,1).\n scoring character string. If \"Z\" (normal scores) (default). If \"Z-SQ\" (normal scores\n squared).\n Chi2corrector character string. Only when scoring is Z-SQ. Select from\n • \"approx: Z^2*(m + 1 + 1.3)/(m+1).\n • \"exact\": Z^2/mean(Z).\n • \"none\": Z^2.\n If \"approx\" () (default). If \"exact\" (normal scores squared).\n dist character string. Select from:\n • \"Uniform: Continuous Uniform distribution .\n • \"Normal\": Normal distribution (default).\n • \"Normal2\": Squared Normal distribution (also known as Chi-squared).\n • \"DoubleExp\": Double exponential distribution (also known as Laplace dis-\n tribution).\n • \"DoubleExp2\": Double exponential squared distribution from a DoubleExp(0,1).\n • \"LogNormal\": Lognormal distribution.\n • \"Gamma\": Gamma distribution.\n • \"Weibull\": Weibull distribution.\n • \"t\": Student-t distribution.\n mu vector. Two elements, the first one is the mean of the reference sample and the\n second one is the mean of the monitoring sample.\n sigma vector. Two elements, the first one is the sd of the reference sample and the\n second one is the sd of the monitoring sample.\n dist.par vector. Distribution parameters. c(par.a, par.b). Default c(0,1).\n chart character string. Selected type of chart. Three options are available: Shewhart,\n CUSUM, EWMA\n chart.par vector. The size depends on the selected chart:\n Shewhart scheme: is c(k), where k comes from U CL = mu + kσ, LCL =\n mu − kσ.\n CUSUM scheme: is c(k, h, t) where k is the reference value and h is the\n control limit, and t is the type of the chart (1:positive, 2:negative, 3:two\n sides)\n EWMA scheme: is c(lambda, L), where lambda is the smoothing constant\n and L multiplies standard deviation to get the control limit\n replicates scalar. Number of replicates to get the ARL\n isParallel logical. If TRUE the code runs in parallel according to the number of cores in the\n computer,otherwise the code runs sequentially. Default TRUE.\n maxIter scalar. is a numeric. The maximum number of iteration to take the calibration\n before stops\n progress logical. If TRUE it shows the progress in the console.\n alignment character string. Aligment of the data X and Y. Select from\n • \"unadjusted\": nothing is sustracte from X and Y (default).\n • \"overallmean\": overall mean is sustracted from X and Y.\n • \"overallmedian\": overall median is sustracted from X and Y.\n • \"samplemean\": mean from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"samplemedian\": median from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"referencemean\": mean from Y is subtracted from X and Y.\n • \"referencemedian\": median from Y is subtracted from X and Y.\n • \"constantvalue\": a constant value is subtracted from X and Y.\n constant scalar. Only used when the alignment is selected \"constantvalue\". Default\n NULL.\n absolute logical. If TRUE, the absolute aligned values are obtained. (Default FALSE)\n isFixed logical. If TRUE the reference sample does not update, otherwise the reference\n sample is updated whenever the batch is in control.\n rounding.factor\n scalar. positive value that determine the range between two consecutive rounded\n values.\nValue\n Multiple output. Select by output$\n • objective.function: scalar. The best solution obtained, in terms of the target ARL or MRL\n • par.value: scalar. Which parameter of the chart reach this best solution\n • iter: scalar. In which iteration is found the objective function.\n • found: boolean. Is TRUE if in the maxIter is reached the desired +-5\nNote\n The argument chart.par in this function correspond to the initial parameters to start the calibration.\nExamples\n n <- 2 # subgroup size\n m <- 30 # reference-sample size\n dist <- \"Normal\" # distribution\n mu <- c(0, 0) # c(reference sample mean, monitoring sample mean)\n sigma <- c(1, 1) # c(reference sample sd, monitoring sample sd)\n #### Distribution parameters\n dist.par <- c(0, 1) # c(location, scale)\n #### Other Parameters\n replicates <- 2\n targetARL <- 370\n isParallel = FALSE\n #### Control chart parameters\n chart <- \"Shewhart\"\n chart.par <- c(3)\n shewhart <- calibrateControlLimit(\n targetARL = targetARL, targetMRL = NULL, n = n, m = m, theta = NULL,\n Ftheta = NULL, dist = dist, mu = mu, sigma = sigma, dist.par = dist.par, chart.par = chart.par,\n replicates = replicates, chart = chart, isParallel = isParallel\n )\n chart <- \"CUSUM\"\n chart.par <- c(0.5, 2.5, 3)\n cusum <- calibrateControlLimit(\n targetARL = targetARL, targetMRL = NULL, n = n, m = m, theta = NULL,\n Ftheta = NULL, dist = dist, mu = mu, sigma = sigma, dist.par = dist.par, chart.par = chart.par,\n replicates = replicates, chart = chart, isParallel = isParallel\n )\n chart <- \"EWMA\"\n chart.par <- c(0.2, 2.962)\n ewma <- calibrateControlLimit(\n targetARL = targetARL, targetMRL = NULL, n = n, m = m, theta = NULL,\n Ftheta = NULL, dist = dist, mu = mu, sigma = sigma, dist.par = dist.par, chart.par = chart.par,\n replicates = replicates, chart = chart, isParallel = isParallel\n )\n dataAlignment Alignment of the data\nDescription\n Align the monitoring sample X and the reference sample Y.\nUsage\n dataAlignment(\n X,\n Y,\n alignment = \"unadjusted\",\n constant = NULL,\n absolute = FALSE\n )\nArguments\n X vector. Monitoring sample.\n Y vector. Reference sample.\n alignment character string. Aligment of the data X and Y. Select from\n • \"unadjusted\": nothing is sustracte from X and Y (default).\n • \"overallmean\": overall mean is sustracted from X and Y.\n • \"overallmedian\": overall median is sustracted from X and Y.\n • \"samplemean\": mean from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"samplemedian\": median from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"referencemean\": mean from Y is subtracted from X and Y.\n • \"referencemedian\": median from Y is subtracted from X and Y.\n • \"constantvalue\": a constant value is subtracted from X and Y.\n constant scalar. Only used when the alignment is selected \"constantvalue\". Default\n NULL.\n absolute logical. If TRUE, the absolute aligned values are obtained. (Default FALSE)\nValue\n Multiple output. Select by output$\n • X: vector. Monitor sample with the alignment selected.\n • Y: vector. Reference sample with the alignment selected.\nExamples\n X = c(30, 45, 50)\n Y = c(20, 22, 25, 30, 70)\n dataAlignment(X,Y)\n example49 Data from Example 4.9 Qiu (2014).\nDescription\n A dataset containing the data set used in Example 4.9 of Qiu (2014).\nUsage\n example49\nFormat\n A data frame with 50 rows and 6 columns:\n Y1 Reference sample of the first data set. 10 batches are N(0,1)\n X1 Monitoring sample of the first data set. 10 batches are N(1,1).\n Y2 Reference sample of the second data set. 10 batches are N(0,1)\n X2 Monitoring sample of the second data set. 10 batches are N(0,2^2).\n X.id id of each observation of the batch for the second data set.\nSource\n https://users.phhp.ufl.edu/pqiu/research/book/spc/data/example49.dat\n example65 Data from Example 6.5 on page 246 Qiu (2014).\nDescription\n A dataset containing the data set used in Example 6.5 on page 246 of Qiu (2014).\nUsage\n example65\nFormat\n A data frame with 30 rows and 5 columns:\n x first 9 observations are the reference sample. Batch size equals to 1.\n Wn Wn\n Sn2 Sn2\n Bmax Bmax\n hn hn\nSource\n https://users.phhp.ufl.edu/pqiu/research/book/spc/data/example65.dat\n example71 Data from Example 7.1 Qiu (2014).\nDescription\n A dataset containing the data set used in Example 7.1 of Qiu (2014).\nUsage\n example71\nFormat\n The data (X1,X2,X3) consist of 30 observations each variable.\n X 1st batch.\n X.1 2nd batch.\n X.2 3rd batch.\nSource\n https://users.phhp.ufl.edu/pqiu/research/book/spc/data/example71.dat\n example74a Data from Example 7.4(a) Qiu (2014).\nDescription\n A dataset containing the data set used in Example 7.4(a) of Qiu (2014).\nUsage\n example74a\nFormat\n The data (X1,X2,X3) consist of 30 observations each variable.\n X 1st batch.\n X.1 2nd batch.\n X.2 3rd batch.\nSource\n https://users.phhp.ufl.edu/pqiu/research/book/spc/data/example74a.dat\n example74b Data from Example 7.4(b) Qiu (2014).\nDescription\n A dataset containing the data set used in Example 7.4(b) of Qiu (2014).\nUsage\n example74b\nFormat\n The data (X1,X2,X3) consist of 30 observations each variable.\n X 1st batch.\n X.1 2nd batch.\n X.2 3rd batch.\nSource\n https://users.phhp.ufl.edu/pqiu/research/book/spc/data/example74b.dat\n example74c Data from Example 7.4(c) Qiu (2014).\nDescription\n A dataset containing the data set used in Example 7.4(c) of Qiu (2014).\nUsage\n example74c\nFormat\n The data (X1,X2,X3) consist of 30 observations each variable.\n X 1st batch.\n X.1 2nd batch.\n X.2 3rd batch.\nSource\n https://users.phhp.ufl.edu/pqiu/research/book/spc/data/example74c.dat\n example81 Data from Example 8.1 on page 319 Qiu (2014).\nDescription\n A dataset containing the data set used in Example 8.1 on page 319 of Qiu (2014).\nUsage\n example81\nFormat\n A data frame with 300 rows (30 batches of size equals to 10)\n X observations of all batches\n X.id id of each observation of the batch\nSource\n https://users.phhp.ufl.edu/pqiu/research/book/spc/data/example81.dat\n example82 Data from Example 8.2 Qiu (2014).\nDescription\n A dataset containing the data set used in Example 8.2 on page 323 of Qiu (2014).\nUsage\n example82\nFormat\n A data frame with 150 rows (30 batches of size equals to 5)\n X observations of all batches\n X.id id of each observation of the batch\nSource\n https://users.phhp.ufl.edu/pqiu/research/book/spc/data/example82.dat\n example83 Data from Example 8.3 on page 326 Qiu (2014).\nDescription\n A dataset containing the data set used in Example 8.3 on page 326 of Qiu (2014).\nUsage\n example83\nFormat\n A data frame with 180 rows (30 batches of size equals to 6)\n X observations of all batches\n X.id id of each observation of the batch\nSource\n https://users.phhp.ufl.edu/pqiu/research/book/spc/data/example81.dat\n example84 Data from Example 8.4 Qiu (2014).\nDescription\n A dataset containing the data set used in Example 8.4 of Qiu (2014).\nUsage\n example84\nFormat\n A data frame with 150 rows (30 batches of size equals to 5)\n X observations of all batches\n X.id id of each observation of the batch\nSource\n https://users.phhp.ufl.edu/pqiu/research/book/spc/data/example84.dat\n example85 Data from Example 8.5 Qiu (2014).\nDescription\n A dataset containing the data set used in Example 8.5 of Qiu (2014).\nUsage\n example85\nFormat\n A data frame with 300 rows (30 batches of size equals to 10)\n X observations of all batches\n X.id id of each observation of the batch\nSource\n https://users.phhp.ufl.edu/pqiu/research/book/spc/data/example85.dat\n example87 Data from Example 8.7 on page 339 Qiu (2014).\nDescription\n A dataset containing the data set used in Example 8.7 on page 339 of Qiu (2014).\nUsage\n example87\nFormat\n A data frame with 86 rows (86 batches of size equals to 1)\n X observations of all batches\n X.id id of each observation of the batch\n Y reference sample of size equals to 14\nSource\n https://users.phhp.ufl.edu/pqiu/research/book/spc/data/example87.dat\n example91 Data from Example 9.1 on page 369 Qiu (2014).\nDescription\n A dataset containing the data set used in Example 9.1 on page 369 of Qiu (2014).\nUsage\n example91\nFormat\n The data (X,Y) consist of 20 batches with 50 observations in each batch.\n V1 1st batch.\n V2 2nd batch.\nSource\n https://users.phhp.ufl.edu/pqiu/research/book/spc/data/example91.dat\n example93 Data from Example 9.3 Qiu (2014).\nDescription\n A dataset containing the data set used in Example 9.3 of Qiu (2014).\nUsage\n example93\nFormat\n The data (X,Y) consist of 20 batches with 10 observations in each batch.\n X 1st batch.\n X.1 2nd batch.\nSource\n https://users.phhp.ufl.edu/pqiu/research/book/spc/data/example93.dat\n getARL Average Run Length (ARL)\nDescription\n Get the ARL getRL\nUsage\n getARL(\n n,\n m,\n theta = NULL,\n Ftheta = NULL,\n dist,\n mu,\n sigma,\n dist.par = c(0, 1, 1),\n chart,\n chart.par,\n scoring = \"Z\",\n Chi2corrector = \"None\",\n replicates = 10000,\n isParallel = TRUE,\n print.RL = FALSE,\n progress = FALSE,\n calibrate = FALSE,\n arl0 = 370,\n alignment = \"unadjusted\",\n constant = NULL,\n absolute = FALSE,\n isFixed = FALSE,\n rounding.factor = NULL\n )\nArguments\n n scalar. Subroup size\n m scalar. Reference sample size\n theta scalar. Value corresponig with the Ftheta quantile.\n Ftheta scalar. Quantile of the data distribution. The values that take are between (0,1).\n dist character string. Select from:\n • \"Uniform: Continuous Uniform distribution .\n • \"Normal\": Normal distribution (default).\n • \"Normal2\": Squared Normal distribution (also known as Chi-squared).\n • \"DoubleExp\": Double exponential distribution (also known as Laplace dis-\n tribution).\n • \"DoubleExp2\": Double exponential squared distribution from a DoubleExp(0,1).\n • \"LogNormal\": Lognormal distribution.\n • \"Gamma\": Gamma distribution.\n • \"Weibull\": Weibull distribution.\n • \"t\": Student-t distribution.\n mu vector. Two elements, the first one is the mean of the reference sample and the\n second one is the mean of the monitoring sample.\n sigma vector. Two elements, the first one is the sd of the reference sample and the\n second one is the sd of the monitoring sample.\n dist.par vector. Distribution parameters. c(par.a, par.b). Default c(0,1).\n chart character string. Selected type of chart. Three options are available: Shewhart,\n CUSUM, EWMA\n chart.par vector. The size depends on the selected chart:\n Shewhart scheme: is c(k), where k comes from U CL = mu + kσ, LCL =\n mu − kσ.\n CUSUM scheme: is c(k, h, t) where k is the reference value and h is the\n control limit, and t is the type of the chart (1:positive, 2:negative, 3:two\n sides)\n EWMA scheme: is c(lambda, L), where lambda is the smoothing constant\n and L multiplies standard deviation to get the control limit\n scoring character string. If \"Z\" (normal scores) (default). If \"Z-SQ\" (normal scores\n squared).\n Chi2corrector character string. Only when scoring is Z-SQ. Select from\n • \"approx: Z^2*(m + 1 + 1.3)/(m+1).\n • \"exact\": Z^2/mean(Z).\n • \"none\": Z^2.\n If \"approx\" () (default). If \"exact\" (normal scores squared).\n replicates scalar. Number of replicates to get the ARL\n isParallel logical. If TRUE the code runs in parallel according to the number of cores in the\n computer,otherwise the code runs sequentially. Default TRUE.\n print.RL logical. If TRUE return the vectors of RL for each iteration.\n progress logical. If TRUE it shows the progress in the console.\n calibrate logical. If TRUE the RL is limit to 10 times the target ARL.\n arl0 scalar. Expected value of the RL. Default 370.\n alignment character string. Aligment of the data X and Y. Select from\n • \"unadjusted\": nothing is sustracte from X and Y (default).\n • \"overallmean\": overall mean is sustracted from X and Y.\n • \"overallmedian\": overall median is sustracted from X and Y.\n • \"samplemean\": mean from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"samplemedian\": median from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"referencemean\": mean from Y is subtracted from X and Y.\n • \"referencemedian\": median from Y is subtracted from X and Y.\n • \"constantvalue\": a constant value is subtracted from X and Y.\n constant scalar. Only used when the alignment is selected \"constantvalue\". Default\n NULL.\n absolute logical. If TRUE, the absolute aligned values are obtained. (Default FALSE)\n isFixed logical. If TRUE the reference sample does not update, otherwise the reference\n sample is updated whenever the batch is in control.\n rounding.factor\n scalar. positive value that determine the range between two consecutive rounded\n values.\nValue\n Multiple output. Select by output$\n • ARL: scalar. Average Run Length for the RLs of all the replicates.\n • SDRL: scalar. Standard Deviation Run Length for the RL in all the replicates.\n • MRL: bolean. Median Run Length for the RLs of all the replicates.\n • QRL: vector. It retrieve the quantiles (0.05, 0.1, 0.2, 0.25, 0.5, 0.75, 0.8, 0.9, 0.95) for all the\n RLs.\nExamples\n n <- 5 # subgroup size\n m <- 100 # reference-sample size\n dist <- \"Normal\"\n mu <- c(0, 0) # c(reference sample mean, monitoring sample mean)\n sigma <- c(1, 1) # c(reference sample sd, monitoring sample sd)\n #### Normal distribution parameters\n dist.par <- c(0, 1) # c(location, scale)\n #### Other Parameters\n replicates <- 2\n print.RL <- TRUE\n isParallel <- FALSE\n calibrate <- FALSE\n progress <- TRUE\n arl0 <- 370\n #### Control chart parameters\n chart <- \"Shewhart\"\n chart.par <- c(3)\n shewhart <- getARL(n, m,\n theta = NULL, Ftheta = NULL, dist, mu, sigma, dist.par = dist.par,\n chart = chart, chart.par = chart.par, print.RL = print.RL,\n replicates = replicates, isParallel = isParallel,\n calibrate = calibrate, arl0 = arl0\n )\n chart <- \"CUSUM\"\n chart.par <- c(0.25, 4.4181, 3)\n cusum <- getARL(n, m,\n theta = NULL, Ftheta = NULL, dist, mu, sigma, dist.par = dist.par,\n chart = chart, chart.par = chart.par, print.RL = print.RL,\n replicates = replicates, isParallel = isParallel,\n calibrate = calibrate, arl0 = arl0\n )\n chart <- \"EWMA\"\n chart.par <- c(0.2, 2.962)\n shewhart <- getARL(n, m,\n theta = NULL, Ftheta = NULL, dist, mu, sigma, dist.par = dist.par,\n chart = chart, chart.par = chart.par, print.RL = print.RL,\n replicates = replicates, isParallel = isParallel,\n calibrate = calibrate, arl0 = arl0\n )\n getDist Random Observations Generator\nDescription\n Random observations generator selected from several distributions with user defined mean and\n variance.\nUsage\n getDist(\n n,\n dist,\n mu,\n sigma,\n par.location = 0,\n par.scale = 1,\n par.shape = 1,\n dist.par = NULL,\n rounding.factor = NULL\n )\nArguments\n n scalar. Number of observations to be generated.\n dist character string. Select from:\n • \"Uniform: Continuous Uniform distribution .\n • \"Normal\": Normal distribution (default).\n • \"Normal2\": Squared Normal distribution (also known as Chi-squared).\n • \"DoubleExp\": Double exponential distribution (also known as Laplace dis-\n tribution).\n • \"DoubleExp2\": Double exponential squared distribution from a DoubleExp(0,1).\n • \"LogNormal\": Lognormal distribution.\n • \"Gamma\": Gamma distribution.\n • \"Weibull\": Weibull distribution.\n • \"t\": Student-t distribution.\n mu scalar. Expected value of the desired distribution.\n sigma scalar. Standard deviation of the desired distribution.\n par.location scalar. Location parameter of the desired distribution. Default 0**.\n par.scale scalar. Scale parameter of the desired distribution. Default 1**.\n par.shape scalar. Shape parameter of the desired distribution, Default 1.\n dist.par vector. Overwrite par.location, par.scale, par.shape. Depends on the dis-\n tribution (default NULL):\n • \"Uniform: no matter how is defined always gives numbers between 0 and\n 1.\n • \"Normal\": c(location, scale).\n • \"Normal2\": c(location, scale).\n • \"DoubleExp\": c(location, scale).\n • \"DoubleExp2\": c(location, scale).\n • \"LogNormal\": c(location, scale).\n • \"Gamma\": c(scale, shape).\n • \"Weibull\": c(shape, scale).\n • \"t\": c(degrees of freedom).\n rounding.factor\n scalar. positive value that determine the range between two consecutive rounded\n values.\nValue\n A vector x with n observations generated following the selected distribution with its parameters.\n**Note\n • For \"Lognormal\", par.location and par.scale correspond to the location and scale param-\n eters of the normal distribution that generales the lognormal. Hence, in this case they are the\n logmean and the logsigma parameters\n • For \"Normal2\" and \"DoubleExp2\", par.location and par.scale correspond correspond\n to the location and scale parameters of the normal and double exponential that are used to\n generates their squared forms.\nExamples\n getDist(1, \"Normal\", 0, 1)\n getQuantile Obtain Quantile from Distribution Function\nDescription\n Get the quantile theta from several distributions with user defined mean and variance.\nUsage\n getQuantile(\n Ftheta,\n mu,\n sigma,\n dist,\n par.location = 0,\n par.scale = 1,\n par.shape = 1,\n dist.par = NULL\n )\nArguments\n Ftheta scalar. Quantile of the data distribution. The values that take are between (0,1).\n mu scalar. Expected value of the desired distribution.\n sigma scalar. Standard deviation of the desired distribution.\n dist character string. Select from:\n • \"Uniform: Continuous Uniform distribution .\n • \"Normal\": Normal distribution (default).\n • \"Normal2\": Squared Normal distribution (also known as Chi-squared).\n • \"DoubleExp\": Double exponential distribution (also known as Laplace dis-\n tribution).\n • \"DoubleExp2\": Double exponential squared distribution from a DoubleExp(0,1).\n • \"LogNormal\": Lognormal distribution.\n • \"Gamma\": Gamma distribution.\n • \"Weibull\": Weibull distribution.\n • \"t\": Student-t distribution.\n par.location scalar. Location parameter of the desired distribution. Default 0**.\n par.scale scalar. Scale parameter of the desired distribution. Default 1**.\n par.shape scalar. Shape parameter of the desired distribution, Default 1.\n dist.par vector. Overwrite par.location, par.scale, par.shape. Depends on the dis-\n tribution (default NULL):\n • \"Uniform: no matter how is defined always gives numbers between 0 and\n 1.\n • \"Normal\": c(location, scale).\n • \"Normal2\": c(location, scale).\n • \"DoubleExp\": c(location, scale).\n • \"DoubleExp2\": c(location, scale).\n • \"LogNormal\": c(location, scale).\n • \"Gamma\": c(scale, shape).\n • \"Weibull\": c(shape, scale).\n • \"t\": c(degrees of freedom).\nValue\n A quantile theta of the selected Ftheta distribution with its parameters.\nExamples\n getQuantile(0.5, 0, 1, \"Normal\")\n getRL Run Length\nDescription\n Get the run length\nUsage\n getRL(\n replica = 1,\n n,\n m,\n theta = NULL,\n Ftheta = NULL,\n dist,\n mu,\n sigma,\n dist.par = c(0, 1, 1),\n scoring = \"Z\",\n chart,\n chart.par,\n calibrate = FALSE,\n arl0 = 370,\n alignment = \"unadjusted\",\n constant = NULL,\n absolute = FALSE,\n isFixed = FALSE,\n Chi2corrector = \"None\",\n rounding.factor = NULL\n )\nArguments\n replica scalar. It is used for the parallel version of the function (parallel=TRUE). De-\n fault 1.\n n scalar. Subroup size\n m scalar. Reference sample size\n theta scalar. Value corresponig with the Ftheta quantile.\n Ftheta scalar. Quantile of the data distribution. The values that take are between (0,1).\n dist character string. Select from:\n • \"Uniform: Continuous Uniform distribution .\n • \"Normal\": Normal distribution (default).\n • \"Normal2\": Squared Normal distribution (also known as Chi-squared).\n • \"DoubleExp\": Double exponential distribution (also known as Laplace dis-\n tribution).\n • \"DoubleExp2\": Double exponential squared distribution from a DoubleExp(0,1).\n • \"LogNormal\": Lognormal distribution.\n • \"Gamma\": Gamma distribution.\n • \"Weibull\": Weibull distribution.\n • \"t\": Student-t distribution.\n mu vector. Two elements, the first one is the mean of the reference sample and the\n second one is the mean of the monitoring sample.\n sigma vector. Two elements, the first one is the sd of the reference sample and the\n second one is the sd of the monitoring sample.\n dist.par vector. Distribution parameters. c(par.a, par.b). Default c(0,1).\n scoring character string. If \"Z\" (normal scores) (default). If \"Z-SQ\" (normal scores\n squared).\n chart character string. Selected type of chart. Three options are available: Shewhart,\n CUSUM, EWMA\n chart.par vector. The size depends on the selected chart:\n Shewhart scheme: is c(k), where k comes from U CL = mu + kσ, LCL =\n mu − kσ.\n CUSUM scheme: is c(k, h, t) where k is the reference value and h is the\n control limit, and t is the type of the chart (1:positive, 2:negative, 3:two\n sides)\n EWMA scheme: is c(lambda, L), where lambda is the smoothing constant\n and L multiplies standard deviation to get the control limit\n calibrate logical. If TRUE the RL is limit to 10 times the target ARL.\n arl0 scalar. Expected value of the RL. Default 370.\n alignment character string. Aligment of the data X and Y. Select from\n • \"unadjusted\": nothing is sustracte from X and Y (default).\n • \"overallmean\": overall mean is sustracted from X and Y.\n • \"overallmedian\": overall median is sustracted from X and Y.\n • \"samplemean\": mean from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"samplemedian\": median from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"referencemean\": mean from Y is subtracted from X and Y.\n • \"referencemedian\": median from Y is subtracted from X and Y.\n • \"constantvalue\": a constant value is subtracted from X and Y.\n constant scalar. Only used when the alignment is selected \"constantvalue\". Default\n NULL.\n absolute logical. If TRUE, the absolute aligned values are obtained. (Default FALSE)\n isFixed logical. If TRUE the reference sample does not update, otherwise the reference\n sample is updated whenever the batch is in control.\n Chi2corrector character string. Only when scoring is Z-SQ. Select from\n • \"approx: Z^2*(m + 1 + 1.3)/(m+1).\n • \"exact\": Z^2/mean(Z).\n • \"none\": Z^2.\n If \"approx\" () (default). If \"exact\" (normal scores squared).\n rounding.factor\n scalar. positive value that determine the range between two consecutive rounded\n values.\nValue\n RL vector. The run length of the chart for the parameter setting.\nExamples\n n <- 5 # subgroup size\n m <- 100 # reference-sample size\n dist <- \"Normal\"\n mu <- c(0, 0) # c(reference sample mean, monitoring sample mean)\n sigma <- c(1, 1) # c(reference sample sd, monitoring sample sd)\n #### Distribution parameters\n dist.par <- c(0, 1, 1) # c(location, scale, shape)\n #### Other Parameters\n replicates <- 2\n print.RL <- TRUE\n calibrate <- FALSE\n progress <- TRUE\n arl0 <- 370\n #### Control chart parameters\n chart <- \"Shewhart\"\n chart.par <- c(3)\n shewhart <- getRL(1, n, m,\n theta = NULL, Ftheta = NULL,dist, mu, sigma, dist.par = dist.par,\n chart = chart, chart.par = chart.par, calibrate = calibrate, arl0 = arl0\n )\n chart <- \"CUSUM\"\n chart.par <- c(0.25, 4.4181, 3)\n cusum <- getRL(1, n, m,\n theta = NULL, Ftheta = NULL, dist, mu, sigma, dist.par = dist.par,\n chart = chart, chart.par = chart.par, calibrate = calibrate, arl0 = arl0\n )\n chart <- \"EWMA\"\n chart.par <- c(0.2, 2.962)\n shewhart <- getRL(1, n, m,\n theta = NULL, Ftheta = NULL,dist, mu, sigma, dist.par = dist.par,\n chart = chart, chart.par = chart.par, calibrate = calibrate, arl0 = arl0\n )\n mcalibrateControlLimit\n Calibration of the control limit for the selected chart\nDescription\n The methodology used to calibrate the control limit for the SNS chart depending on the selected\n chart\nUsage\n mcalibrateControlLimit(\n targetARL = NULL,\n targetMRL = NULL,\n n,\n m,\n nv,\n theta = NULL,\n Ftheta = NULL,\n dists = c(\"Normal\", \"Normal\"),\n mu = c(0, 0),\n sigma = NULL,\n dists.par = matrix(c(0, 1, 1, 0, 1, 1), ncol = 2),\n correlation = 0,\n chart = \"T2\",\n chart.par = c(10),\n replicates = 50000,\n isParallel = FALSE,\n maxIter = 20,\n progress = TRUE,\n alignment = \"unadjusted\",\n constant = NULL,\n absolute = FALSE\n )\nArguments\n targetARL scalar. is the target ARL to calibrate. By default is set to NULL\n targetMRL scalar. is the target ARL to calibrate. By default is set to NULL\n n scalar. Subroup size\n m scalar. Reference sample size\n nv scalar. Number of variables to be generated.\n theta vector. Value corresponding with the Ftheta quantile.\n Ftheta vector. Quantile of the data distribution. The values that take are between (0,1).\n dists list. Select the\n mu vector. Two elements of the vector the first one is the mean of the reference\n sample and the second one is the mean of the monitoring sample.\n sigma scalar. Standard deviation of the desired distribution.\n dists.par matrix For each variable (column), specify\n • par.location: Location parameter of the desired distribution. Default 0.\n • par.scale: Scale parameter of the desired distribution. Default 1.\n • par.shape: Shape parameter of the desired distribution, Default 1.\n The number of columns must be the same as the number of variables.\n correlation scalar. Corralation between variables.\n chart character string. Selected type of chart. One option available: \"T2\".\n T2 scheme: is c(k), where k comes from U CL = mu+kσ, LCL = mu−kσ.\n chart.par vector. Control limit and other parameters of the selected chart.\n replicates scalar. Number of replicates to get the ARL\n isParallel logical. If TRUE the code runs in parallel according to the number of cores in the\n computer,otherwise the code runs sequentially. Default TRUE.\n maxIter scalar. is a numeric. The maximum number of iteration to take the calibration\n before stops\n progress logical. If TRUE it shows the progress in the console.\n alignment character string. Aligment of the data X and Y. Select from\n • \"unadjusted\": nothing is sustracte from X and Y (default).\n • \"overallmean\": overall mean is sustracted from X and Y.\n • \"overallmedian\": overall median is sustracted from X and Y.\n • \"samplemean\": mean from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"samplemedian\": median from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"referencemean\": mean from Y is subtracted from X and Y.\n • \"referencemedian\": median from Y is subtracted from X and Y.\n • \"constantvalue\": a constant value is subtracted from X and Y.\n constant scalar. Only used when the alignment is selected \"constantvalue\". Default\n NULL.\n absolute logical. If TRUE, the absolute aligned values are obtained. (Default FALSE)\nValue\n Multiple output. Select by output$\n • objective.function: scalar. The best solution obtained, in terms of the target ARL or MRL\n • par.value: scalar. Which parameter of the chart reach this best solution\n • found: boolean. Is TRUE if in the maxIter is reached the desired +-5\nNote\n The argument chart.par in this function correspond to the initial parameters to start the calibration.\nExamples\n n <- 5 # subgroup size\n m <- 10 # reference-sample size\n dists <- c(\"Normal\", \"Normal\") # distribution\n mu <- c(0, 0) # c(reference sample mean, monitoring sample mean)\n nv <- 2 # number of variables\n #### Other Parameters\n replicates <- 2\n targetARL <- 200\n isParallel = FALSE\n maxIter <- 2\n #### Control chart parameters\n chart <- \"T2\"\n chart.par <- c(0.005)\n t2 <- mcalibrateControlLimit(targetARL = targetARL,n = n, m = m, nv = nv, theta = NULL,\n Ftheta = NULL, dists = dists, mu = mu, chart.par = chart.par,\n replicates = replicates, chart = chart, isParallel = isParallel,\n maxIter = maxIter\n )\n mgetARL Multivariate Average Run Length (ARL)\nDescription\n Get the ARL getRL\nUsage\n mgetARL(\n n,\n m,\n nv,\n theta = NULL,\n Ftheta = NULL,\n dists,\n dists.par = NULL,\n mu,\n sigma = NULL,\n chart = \"T2\",\n chart.par = c(0.005),\n correlation = 0,\n s = NULL,\n replicates = 10000,\n isParallel = TRUE,\n print.RL = FALSE,\n progress = FALSE,\n calibrate = FALSE,\n arl0 = 370,\n alignment = \"unadjusted\",\n constant = NULL,\n absolute = FALSE\n )\nArguments\n n scalar. Subroup size\n m scalar. Reference sample size\n nv scalar. Number of variables to be generated.\n theta vector. Value corresponding with the Ftheta quantile.\n Ftheta vector. Quantile of the data distribution. The values that take are between (0,1).\n dists list. Select the\n dists.par matrix For each variable (column), specify\n • par.location: Location parameter of the desired distribution. Default 0.\n • par.scale: Scale parameter of the desired distribution. Default 1.\n • par.shape: Shape parameter of the desired distribution, Default 1.\n The number of columns must be the same as the number of variables.\n mu vector. Two elements of the vector the first one is the mean of the reference\n sample and the second one is the mean of the monitoring sample.\n sigma scalar. Standard deviation of the desired distribution.\n chart character string. Selected type of chart. One option available: \"T2\".\n T2 scheme: is c(k), where k comes from U CL = mu+kσ, LCL = mu−kσ.\n chart.par vector. Control limit and other parameters of the selected chart.\n correlation scalar. Corralation between variables.\n s matrix. Correlation matrix of the variables\n replicates scalar. Number of replicates to get the ARL\n isParallel logical. If TRUE the code runs in parallel according to the number of cores in the\n computer,otherwise the code runs sequentially. Default TRUE.\n print.RL logical. If TRUE return the vectors of RL for each iteration.\n progress logical. If TRUE it shows the progress in the console.\n calibrate logical. If TRUE the RL is limit to 10 times the target ARL.\n arl0 scalar. Expected value of the RL. It is only used for stop the RL if exceeds 10\n times its value. Default 370.\n alignment character string. Aligment of the data X and Y. Select from\n • \"unadjusted\": nothing is sustracte from X and Y (default).\n • \"overallmean\": overall mean is sustracted from X and Y.\n • \"overallmedian\": overall median is sustracted from X and Y.\n • \"samplemean\": mean from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"samplemedian\": median from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"referencemean\": mean from Y is subtracted from X and Y.\n • \"referencemedian\": median from Y is subtracted from X and Y.\n • \"constantvalue\": a constant value is subtracted from X and Y.\n constant scalar. Only used when the alignment is selected \"constantvalue\". Default\n NULL.\n absolute logical. If TRUE, the absolute aligned values are obtained. (Default FALSE)\nValue\n Multiple output. Select by output$\n • ARL: scalar. Average Run Length for the RLs of all the replicates.\n • SDRL: scalar. Standard Deviation Run Length for the RL in all the replicates.\n • MRL: bolean. Median Run Length for the RLs of all the replicates.\n • QRL: vector. It retrieve the quantiles (0.05, 0.1, 0.2, 0.25, 0.5, 0.75, 0.8, 0.9, 0.95) for all the\n RLs.\nExamples\n mgetARL(replicates=5,n=5,m=100,nv=2,mu=c(0,0),\n dists = c(\"Normal\", \"Normal\"), dists.par = matrix(c(0,1,1,0,1,1), ncol=2),\n isParallel=FALSE)\n mgetDist Multivariate Random Observations Generetor\nDescription\n Multivariate Random observations generator selected from several distributions with user defined\n mean and variance.\nUsage\n mgetDist(\n n,\n nv,\n mu = 0,\n sigma = NULL,\n correlation = 0,\n s = NULL,\n dists = NULL,\n dists.par = NULL\n )\nArguments\n n scalar. Number of observations to be generated.\n nv scalar. Number of variables to be generated.\n mu scalar. Expected value of the desired distribution.\n sigma scalar. Standard deviation of the desired distribution.\n correlation scalar. Corralation between variables.\n s matrix. Correlation matrix of the variables\n dists list. Select the\n dists.par matrix For each variable (column), specify\n • par.location: Location parameter of the desired distribution. Default 0.\n • par.scale: Scale parameter of the desired distribution. Default 1.\n • par.shape: Shape parameter of the desired distribution, Default 1.\n The number of columns must be the same as the number of variables.\nValue\n A matrix x with n observations generated following the selected distribution with its parameters.\nExamples\n mgetDist(n=5, nv=2, dists=c(\"Normal\", \"Normal\"),dists.par= matrix(c(0,1,1,0,1,1), ncol=2))\n mgetRL Multivariate Run Length\nDescription\n Get the run length\nUsage\n mgetRL(\n replica = 1,\n n,\n m,\n nv,\n theta = NULL,\n Ftheta = NULL,\n dists,\n mu,\n sigma = NULL,\n dists.par = NULL,\n correlation = 0,\n s = NULL,\n chart = \"T2\",\n chart.par = c(0.005),\n null.dist = \"Chi\",\n alignment = \"unadjusted\",\n constant = NULL,\n absolute = FALSE,\n calibrate = FALSE,\n arl0 = 370\n )\nArguments\n replica scalar. It is used for the parallel version of the function (parallel=TRUE). De-\n fault 1.\n n scalar. Subroup size\n m scalar. Reference sample size\n nv scalar. Number of variables to be generated.\n theta vector. Value corresponding with the Ftheta quantile.\n Ftheta vector. Quantile of the data distribution. The values that take are between (0,1).\n dists list. Select the\n mu vector. Two elements of the vector the first one is the mean of the reference\n sample and the second one is the mean of the monitoring sample.\n sigma scalar. Standard deviation of the desired distribution.\n dists.par matrix For each variable (column), specify\n • par.location: Location parameter of the desired distribution. Default 0.\n • par.scale: Scale parameter of the desired distribution. Default 1.\n • par.shape: Shape parameter of the desired distribution, Default 1.\n The number of columns must be the same as the number of variables.\n correlation scalar. Corralation between variables.\n s matrix. Correlation matrix of the variables\n chart character string. Selected type of chart. One option available: \"T2\".\n T2 scheme: is c(k), where k comes from U CL = mu+kσ, LCL = mu−kσ.\n chart.par vector. Control limit and other parameters of the selected chart.\n null.dist character string. It is the null distribution choose from \"Chi\" or \"F\".\n alignment character string. Aligment of the data X and Y. Select from\n • \"unadjusted\": nothing is sustracte from X and Y (default).\n • \"overallmean\": overall mean is sustracted from X and Y.\n • \"overallmedian\": overall median is sustracted from X and Y.\n • \"samplemean\": mean from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"samplemedian\": median from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"referencemean\": mean from Y is subtracted from X and Y.\n • \"referencemedian\": median from Y is subtracted from X and Y.\n • \"constantvalue\": a constant value is subtracted from X and Y.\n constant scalar. Only used when the alignment is selected \"constantvalue\". Default\n NULL.\n absolute logical. If TRUE, the absolute aligned values are obtained. (Default FALSE)\n calibrate logical. If TRUE the RL is limit to 10 times the target ARL.\n arl0 scalar. Expected value of the RL. It is only used for stop the RL if exceeds 10\n times its value. Default 370.\nValue\n RL vector. The run length of the chart for the parameter setting.\nExamples\n mgetRL(n=5, m=10, nv=2, mu=c(0,0), dists = c(\"Normal\", \"Normal\"),\n dists.par = matrix(c(0,1,1,0,1,1), ncol=2))\n MNS Multivariate Normal Scores\nDescription\n Get conditional or unconditional multivariate normal score (NS) of observations (X) relative to pre-\n vious observations (Y).\nUsage\n MNS(\n X,\n Y = NULL,\n theta = NULL,\n Ftheta = NULL,\n scoring = \"Z\",\n alignment = \"unadjusted\",\n constant = NULL,\n absolute = FALSE\n )\nArguments\n X matrix or data.frame. New observations to obtain the normal scores.\n Y matrix or data.frame. If Y is not defined (no previous observation available,\n NULL), NS is relative to X. Default NULL.\n theta vector. Value corresponding with the Ftheta quantile.\n Ftheta vector. Quantile of the data distribution. The values that take are between (0,1).\n scoring character string. If \"Z\" (normal scores) (default). If \"Z-SQ\" (normal scores\n squared).\n alignment character string. Aligment of the data X and Y. Select from\n • \"unadjusted\": nothing is sustracte from X and Y (default).\n • \"overallmean\": overall mean is sustracted from X and Y.\n • \"overallmedian\": overall median is sustracted from X and Y.\n • \"samplemean\": mean from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"samplemedian\": median from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"referencemean\": mean from Y is subtracted from X and Y.\n • \"referencemedian\": median from Y is subtracted from X and Y.\n • \"constantvalue\": a constant value is subtracted from X and Y.\n constant scalar. Only used when the alignment is selected \"constantvalue\". Default\n NULL.\n absolute logical. If TRUE, the absolute aligned values are obtained. (Default FALSE)\nValue\n Multiple output. Select by output$\n • R: matrix. Multivariate Ranks for the X observations. If ties occurs, average ranks are used.\n • P: matrix. Multivariate Probability of the ranks for the X observations. Instead of Van Der\n Waerden normal scores where P = R/(n + 1), P = (R − 0.5)/n, where R stands for rank\n and P for the input evaluated in the inverse of a Standard Normal Distribution.\n • Z: matrix. Multivariate Normal scores for the X observations. Z if scoring is \"Z\" and Z 2 if\n scoring is \"Z-SQ\".\nExamples\n Y <- c(10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100)\n Y = matrix(Y, ncol=2)\n X <- c(30, 35, 45, 30, 35, 45)\n X = matrix(X, ncol=2)\n theta <- c(40, 40)\n Ftheta <- c(0.5, 0.5)\n # EXAMPLE CONDITIONAL\n MNS(X = X, Y = Y, theta = theta, Ftheta = Ftheta)\n MSNS Multivariate Sequential Normal Scores\nDescription\n Transform a matrix X into SNS using initial observations Y if available SNS follow the order of X.\nUsage\n MSNS(\n X,\n X.id,\n Y = NULL,\n theta = NULL,\n Ftheta = NULL,\n scoring = \"Z\",\n alignment = \"unadjusted\",\n constant = NULL,\n absolute = FALSE,\n chart = \"T2\",\n chart.par = c(0.005),\n null.dist = \"Chi\",\n isFixed = FALSE,\n omit.id = NULL,\n auto.omit.alarm = TRUE\n )\nArguments\n X matrix or data.frame. New observations to obtain the normal scores.\n X.id vector. The id of each column (variable) of the matrix X.\n Y matrix or data.frame. If Y is not defined (no previous observation available,\n NULL), NS is relative to X. Default NULL.\n theta vector. Value corresponding with the Ftheta quantile.\n Ftheta vector. Quantile of the data distribution. The values that take are between (0,1).\n scoring character string. If \"Z\" (normal scores) (default). If \"Z-SQ\" (normal scores\n squared).\n alignment character string. Aligment of the data X and Y. Select from\n • \"unadjusted\": nothing is sustracte from X and Y (default).\n • \"overallmean\": overall mean is sustracted from X and Y.\n • \"overallmedian\": overall median is sustracted from X and Y.\n • \"samplemean\": mean from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"samplemedian\": median from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"referencemean\": mean from Y is subtracted from X and Y.\n • \"referencemedian\": median from Y is subtracted from X and Y.\n • \"constantvalue\": a constant value is subtracted from X and Y.\n constant scalar. Only used when the alignment is selected \"constantvalue\". Default\n NULL.\n absolute logical. If TRUE, the absolute aligned values are obtained. (Default FALSE)\n chart character string. Selected type of chart. One option available: \"T2\".\n T2 scheme: is c(k), where k comes from U CL = mu+kσ, LCL = mu−kσ.\n chart.par vector. Control limit and other parameters of the selected chart.\n null.dist character string. It is the null distribution choose from \"Chi\" or \"F\".\n isFixed logical. If TRUE the reference sample does not update, otherwise the reference\n sample is updated when the batch is in control.\n omit.id vector. Elements of the vector are the id which are omitted in the analysis.\n auto.omit.alarm\n logical. Determine if OC signals are added (or not) to reference sample. By\n default is set to TRUE.\nValue\n Multiple output. Select by output$\n • coefficients: list. Two elements: n the number of observation per group in X and chart the\n selected chart to perform the analysis.\n • X: vector. New observations (Monitoring sample) to obtain the SNS.\n • Z: vector. SNS of the X monitoring sample.\n • T2: vector. T2 statistic for each of the groups in X.\n • X.id: vector. The id of each column (variable) of the matrix X.\n • UCL: vector. Upper control limit for each group in X.\nComments\n If ties, average ranks are used.\nSee Also\n MNS for multivariate normal scores\nExamples\n X = cbind(example91$X1, example91$X2)\n X.id = example91$X.id\n msns = MSNS(X, X.id)\n NS Normal Scores\nDescription\n Get conditional or unconditional normal score (NS) of observations (X) relative to previous obser-\n vations (Y).\nUsage\n NS(\n X,\n Y = NULL,\n theta = NULL,\n Ftheta = NULL,\n scoring = \"Z\",\n Chi2corrector = \"None\",\n alignment = \"unadjusted\",\n constant = NULL,\n absolute = FALSE\n )\nArguments\n X vector. New observations to obtain the N¡normal scores.\n Y vector. If Y is not defined (no previous observation available, NULL), NS is rela-\n tive to X. Default NULL.\n theta scalar. Value corresponig with the Ftheta quantile.\n Ftheta scalar. Quantile of the data distribution. The values that take are between (0,1).\n scoring character string. If \"Z\" (normal scores) (default). If \"Z-SQ\" (normal scores\n squared).\n Chi2corrector character string. Only when scoring is Z-SQ. Select from\n • \"approx: Z^2*(m + 1 + 1.3)/(m+1).\n • \"exact\": Z^2/mean(Z).\n • \"none\": Z^2.\n If \"approx\" () (default). If \"exact\" (normal scores squared).\n alignment character string. Aligment of the data X and Y. Select from\n • \"unadjusted\": nothing is sustracte from X and Y (default).\n • \"overallmean\": overall mean is sustracted from X and Y.\n • \"overallmedian\": overall median is sustracted from X and Y.\n • \"samplemean\": mean from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"samplemedian\": median from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"referencemean\": mean from Y is subtracted from X and Y.\n • \"referencemedian\": median from Y is subtracted from X and Y.\n • \"constantvalue\": a constant value is subtracted from X and Y.\n constant scalar. Only used when the alignment is selected \"constantvalue\". Default\n NULL.\n absolute logical. If TRUE, the absolute aligned values are obtained. (Default FALSE)\nValue\n Multiple output. Select by output$\n • R: vector. Ranks for the X observations. If ties occurs, average ranks are used.\n • P: vector. Probability of the ranks for the X observations. Instead of Van Der Waerden normal\n scores where P = R/(n + 1), P = (R − 0.5)/n, where R stands for rank and P for the input\n evaluated in the inverse of a Standard Normal Distribution.\n • Z: vector. Normal scores for the X observations. Z if scoring is \"Z\" and Z 2 if scoring is\n \"Z-SQ\".\nExamples\n Y <- c(10, 20, 30, 40, 50, 60, 70, 80, 90, 100)\n X <- c(30, 35, 45)\n theta <- 40\n Ftheta <- 0.5\n # EXAMPLE CONDITIONAL\n NS(X = X, Y = Y, theta = theta, Ftheta = Ftheta)\n # EXAMPLE UNCONDITIONAL\n theta <- NULL\n Ftheta <- NULL\n NS(X = X, Y = Y, theta = theta, Ftheta = Ftheta)\n SNS Sequential Normal Scores\nDescription\n Transform a vector X into SNS using initial observations Y if available SNS follow the order of X.\nUsage\n SNS(\n X,\n X.id,\n Y = NULL,\n theta = NULL,\n Ftheta = NULL,\n scoring = \"Z\",\n Chi2corrector = \"None\",\n alignment = \"unadjusted\",\n constant = NULL,\n absolute = FALSE,\n chart = \"Shewhart\",\n chart.par = c(3),\n snsRaw = FALSE,\n isFixed = FALSE,\n omit.id = NULL,\n auto.omit.alarm = TRUE\n )\nArguments\n X vector. New observations to obtain the N¡normal scores.\n X.id vector. The id of the vector X.\n Y vector. If Y is not defined (no previous observation available, NULL), NS is rela-\n tive to X. Default NULL.\n theta scalar. Value corresponig with the Ftheta quantile.\n Ftheta scalar. Quantile of the data distribution. The values that take are between (0,1).\n scoring character string. If \"Z\" (normal scores) (default). If \"Z-SQ\" (normal scores\n squared).\n Chi2corrector character string. Only when scoring is Z-SQ. Select from\n • \"approx: Z^2*(m + 1 + 1.3)/(m+1).\n • \"exact\": Z^2/mean(Z).\n • \"none\": Z^2.\n If \"approx\" () (default). If \"exact\" (normal scores squared).\n alignment character string. Aligment of the data X and Y. Select from\n • \"unadjusted\": nothing is sustracte from X and Y (default).\n • \"overallmean\": overall mean is sustracted from X and Y.\n • \"overallmedian\": overall median is sustracted from X and Y.\n • \"samplemean\": mean from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"samplemedian\": median from corresponding group (X and Y) is sustracted\n from its corresponing vector.\n • \"referencemean\": mean from Y is subtracted from X and Y.\n • \"referencemedian\": median from Y is subtracted from X and Y.\n • \"constantvalue\": a constant value is subtracted from X and Y.\n constant scalar. Only used when the alignment is selected \"constantvalue\". Default\n NULL.\n absolute logical. If TRUE, the absolute aligned values are obtained. (Default FALSE)\n chart character string. Selected type of chart. Three options are available: Shewhart,\n CUSUM, EWMA\n chart.par vector. The size depends on the selected chart:\n Shewhart scheme: is c(k), where k comes from U CL = mu + kσ, LCL =\n mu − kσ.\n CUSUM scheme: is c(k, h, t) where k is the reference value and h is the\n control limit, and t is the type of the chart (1:positive, 2:negative, 3:two\n sides)\n EWMA scheme: is c(lambda, L), where lambda is the smoothing constant\n and L multiplies standard deviation to get the control limit\n snsRaw logical. If TRUE return also the sns for each observation in vector X.\n isFixed logical. If TRUE the reference sample does not update, otherwise the reference\n sample is updated whenever the batch is in control.\n omit.id vector. Elements of the vector are the id which are omitted in the analysis.\n auto.omit.alarm\n logical. Determine if OC signals are added (or not) to reference sample. By\n default is set to TRUE.\nValue\n Multiple output. Select by output$\n • coefficients: list. Three elements: n the number of observation per group in X, chart the\n selected chart to perform the analysis, and chart.par the parameters of the selected chart.\n • R: vector. Ranks for the new observations (Monitoring sample).\n • X: vector. New observations (Monitoring sample) to obtain the SNS.\n • Z: vector. SNS of the X monitoring sample.\n • X.id: vector. The id of each column (variable) of the matrix X.\n • UCL: vector. Upper control limit for each group in X.\n • LCL: vector. Lower control limit for each group in X.\n • scoring: string. Selected score to evaluate SNS.\nComments\n If ties occur, average ranks are used.\nSee Also\n NS for normal scores\nExamples\n # EXAMPLE CONDITIONAL WITH REFERENCE SAMPLE\n Y <- c(10, 20, 30, 40, 50, 60, 70, 80, 90, 100)\n X <- c(30, 35, 45)\n theta <- 40\n Ftheta <- 0.5\n sample.id <- c(\"a\", \"b\", \"c\")\n SNS(X = X, X.id = sample.id, Y = Y, theta = theta, Ftheta = Ftheta)\n # EXAMPLE CONDITIONAL WITH REFERENCE SAMPLE\n Y <- c(10, 20, 30, 40, 50, 60, 70, 80, 90, 100)\n X <- c(30, 35, 45)\n theta <- 40\n Ftheta <- 0.5\n sample.id <- c(\"a\", \"b\", \"c\")\n SNS(X = X, X.id = sample.id, Y = Y, theta = theta, Ftheta = Ftheta)\n # EXAMPLE UNCONDITIONAL WITH REFERENCE SAMPLE\n Y <- c(10, 20, 30, 40, 50, 60, 70, 80, 90, 100)\n X <- c(30, 35, 45)\n theta <- NULL\n Ftheta <- NULL\n sample.id <- c(\"a\", \"b\", \"c\")\n SNS(X = X, X.id = sample.id, Y = Y, theta = theta, Ftheta = Ftheta)\n # EXAMPLE CONDITIONAL WITHOUT REFERENCE SAMPLE\n Y <- NULL # c(10,20,30,40,50,60,70,80,90,100)\n X <- c(30, 35, 45)\n theta <- 40\n Ftheta <- 0.5\n sample.id <- c(\"a\", \"b\", \"c\")\n SNS(X = X, X.id = sample.id, Y = Y, theta = theta, Ftheta = Ftheta)\n # EXAMPLE UNCONDITIONAL WITHOUT REFERENCE SAMPLE\n Y <- NULL\n X <- c(30, 35, 45)\n theta <- NULL\n Ftheta <- NULL\n sample.id <- c(\"a\", \"b\", \"c\")\n SNS(X = X, X.id = sample.id, Y = Y, theta = theta, Ftheta = Ftheta)\n srank Sequential Rank\nDescription\n Get the sequential rank of observations (X) relative to previous observations (Y).\nUsage\n srank(X, Y = NULL)\nArguments\n X vector. New observations to obtain the N¡normal scores.\n Y vector. If Y is not defined (no previous observation available, NULL), NS is rela-\n tive to X. Default NULL.\nValue\n vector. Sequentil Ranks for the X observations. If ties occurs, average of the ranks are used.\nExamples\n X <- c(30, 35, 45)\n srank(X)"}}},{"rowIdx":59188,"cells":{"project":{"kind":"string","value":"Rcapture"},"source":{"kind":"string","value":"cran"},"language":{"kind":"string","value":"R"},"content":{"kind":"string","value":"Package ‘Rcapture’\n October 12, 2022\nType Package\nTitle Loglinear Models for Capture-Recapture Experiments\nVersion 1.4-4\nDate 2022-05-03\nDescription Estimation of abundance and other demographic parameters for closed\n populations, open populations and the robust design in capture-recapture\n experiments using loglinear models.\nImports stats, graphics, utils\nLazyData true\nLicense GPL-2\nNeedsCompilation no\nAuthor <NAME> [aut, cre],\n <NAME> [aut]\nMaintainer <NAME> <<EMAIL>>\nRepository CRAN\nDate/Publication 2022-05-04 13:20:06 UTC\nR topics documented:\nRcapture-packag... 2\nBBS200... 7\nbuntin... 9\ncbir... 10\nclosed... 11\nclosedp.b... 16\nclosedp.custo... 19\nclosedp.Mt... 21\nclosedpC... 22\nclosedpM... 29\ndescriptiv... 32\nduc... 35\nhar... 36\nhistpo... 37\nHI... 39\nil... 40\nlesbia... 41\nmvol... 42\nopen... 44\nperiodhis... 47\nprofileC... 48\nrbvol... 50\nrobust... 51\nuifi... 54\n Rcapture-package Loglinear Models for Capture-Recapture Experiments\nDescription\n Estimation of abundance and of other demographic parameters for closed populations, open popu-\n lations and the robust design in capture-recapture experiments using loglinear models.\n This package focuses on closed populations. Since version 1.2-0, no new features have been added\n to open populations and robust design functions.\nDetails\n Package: Rcapture\n Type: Package\n Version: 1.4-4\n Date: 2022-05-03\n License: GPL-2\n SUMMARY OF Rcapture CONTENTS\n The Rcapture package contains nine capture-recapture data sets and the following functions:\n • Model fitting functions for\n > closed populations:\n – closedp functions: fit various loglinear models for abundance estimation,\n – closedpCI functions: fit one customized loglinear model and calculate a confidance in-\n terval for the abundance estimation,\n – closedpMS.t: fits various hierarchical loglinear models in a perspective of model selec-\n tion,\n – closedp.bc: performs bias corrections to the abundance estimations from customized\n loglinear models,\n – closedp.Mtb: fits model Mtb, which cannot be fitted by any other function, for abun-\n dance estimation;\n > open populations:\n – openp: computes various demographic parameters using a loglinear model;\n > robust design:\n – robustd functions: compute various demographic parameters and capture probabilities\n per period using a loglinear model.\n • Descriptive statistics functions:\n – descriptive: produces basic descriptive statistics for capture-recapture data;\n – uifit: produces fit statistics concerning the ui , i.e. the numbers of first captures on each\n capture occasion, for closed population models.\n • Data manipulation functions:\n – histpos functions: builds a matrix of observable capture histories;\n – periodhist: merges capture occasions.\n DESCRIPTION OF DATA SET FORMATS\n In capture-recapture experiments, the data collected consist of capture histories for captured units.\n A capture history is simply a serie of capture indicators for each capture event in the experiment.\n The capture history of one unit is expressed as a length t vector w = (w1 , . . . , wt ), where wj = 1\n if the unit is captured at the jth occasion and 0 if not. For closed populations, capture events are\n named capture occasions, whereas they are named capture periods for open populations.\n Capture-recapture data sets are given to Rcapture functions through the X argument. X must be a\n numeric matrix. Arguments dfreq and dtype indicate the format of the matrix. Each have two\n possible values, meaning that four data set formats are possible with Rcapture.\n FORMAT 1 - CAPTURE HISTORY PER UNIT\n If dfreq=FALSE and dtype=\"hist\" (the default), X has one row per unit captured in the experiment.\n Each row is an observed capture history. It must contain only zeros and ones; the number one indi-\n cates a capture. In this case, the number of columns in the table represents the number of capture\n occasions in the experiment (noted t). Here is an example of a data set of this type for t = 2:\n 11\n 11\n 10\n 10\n 10\n 10\n 01\n FORMAT 2 - AGGREGATED CAPTURE HISTORIES\n If dfreq=TRUE and dtype=\"hist\", X contains one row per observed capture history followed by its\n frequency. In that case, X has t+1 columns. As for format 1, the first t columns of X, identifying\n the capture histories, must contain only zeros and ones. The number one indicates a capture. In this\n format, the example data set is represented by the following matrix:\n 112\n 104\n 011\n If a possible capture history is not observed, it can appear in X with a frequency of zero, or it can\n simply be omitted.\n FORMAT 3 - NUMBER OF CAPTURES PER UNIT\n If dfreq=FALSE and dtype=\"nbcap\", X is a vector with the number of captures for every captured\n unit. Therefore, this format does not contain complete capture histories. Instead, capture histories\n are summarized through the number of captures. In this format, the example data set looks like:\n 2211111\n FORMAT 4 - AGGREGATED NUMBERS OF CAPTURES\n If dfreq=TRUE and dtype=\"nbcap\", X is a 2 columns matrix. The first column contains the observed\n numbers of captures, the second columns contains their frequencies. In this format, the example\n data is:\n 22\n 15\n DETAILS ABOUT FORMATS WITH NUMBERS OF CAPTURES\n Only few functions have the dtype argument. Functions without dtype argument accept only a data\n matrix X of the form dtype=\"hist\". So the first two formats listed above are the most common.\n Formats with dtype=\"nbcap\" are used for captures in continuous time (see below). They are also\n useful to reduce the size of the data set for experiments with a large number of capture occasions\n t (often with no units caught a large number of times). For theses formats, the number of capture\n occasions t cannot be deduced from X as it can be with dtype=\"hist\". One has no garanties that the\n larger number of captures observed is the total number of capture occasions. Therefore, if one gives\n a data matrix X with dtype=\"nbcap\", one must also provide t, the number of capture occasions, as\n an additional argument.\n For now, the data formats with dtype=\"nbcap\" are not generalized to the robust design. So dtype\n is not an argument of the robustd.0 function.\n CAPTURES IN CONTINUOUS TIME\n In some capture-recapture experiments, there is no well defined capture occasions. Captures occur\n in continuous time. The data set ill comes from such an experiment. Bohning and Schon (2005)\n call this type of capture-recapture data repeated counting data. These data sets always have the\n format dtype=\"nbcap\".\n We can estimate abundance for data of this type using the option t=Inf with the functions closedpCI.0\n and closedpCI.0. The function descriptive also accepts t=Inf. It modifies the y coordinate of\n the exploratory heterogeneity graph.\n DISTINCTION BETWEEN .t and .0 FUNCTIONS\n Capture recapture models for closed population aim at estimating the population size by modelling\n the probabilities of the different capture histories. The data available to fit these models consist\n of observed frequencies of capture histories. These frequencies are modeled in Rcapture using\n loglinear models for frequency tables. For functions with a name ending with .t (closedp.t,\n closedpCI.t, closedpMS.t and robustd.t), the observed values of the response variable is the\n vector of frequencies for every observable capture history. It has length 2t − 1.\n However, for a model without temporal effect (assuming that the probability of capturing a unit does\n not vary between capture occasions), all the information needed to fit the model can be found in ag-\n gregated data. Functions with a name ending with .0 (closedp.0, closedpCI.0 and robustd.0)\n fit models using as response variable the number of units captured i times, for i = 1, . . . , t. It is a\n vector of length t, which is much shorter than 2t − 1 for a large t. Because of an appropriate offset\n added to the model, the results from a .0 function match exaclty the results from a .t function for\n a corresponding capture-recapture model. Because .0 functions deal with smaller design matrix,\n they run faster than .t functions, but they cannot fit models with a temporal effect.\n FUNCTIONS USED FOR MODEL FITTING\n Most of the Rcapture functions use the function glm of the stats package to fit a loglinear model.\n However, the function optim, again of the stats, is also used by three functions:\n • closedpCI.t, closedpCI.0 and closedpMS.t: when a normal heterogeneous models is re-\n quested,\n • closedp.Mtb: because model Mtb does not have a loglinear form.\n ERRORS AND WARNINGS MANAGEMENT IN CLOSED POPULATION FUNCTIONS\n If an error occurs while executing a function fitting only one closed population model (closedpCI\n functions, closedp.bc, closedp.Mtb), the execution is stopped and the error message is printed\n (usual behavior in R). However, if an error occurs while fitting a model in a call to a closedp\n population function fitting more than one model (closedp functions, closedpMS.t), the execution\n of the call is not stopped. Instead, the row in the results table for the problematic model is filled with\n NA and the error message is stored in an output value. This value is called glm.err for closedp\n functions. It is however called fit.err for closedpMS.t because this function do not always use\n glm to fit the model (as mentionned above, optim is used for normal heterogeneous models).\n Warning messages while fitting a closed population model, if any, are stored in an output value\n called ’glm.warn’, ’fit.warn’ or ’optim.warn’ depending on the function. They are not printed in\n the console. To inform the user that a warning occured, the last column of the results table, named\n infoFit, contains a numerical code giving information about errors or warnings encountered. Here\n is a description of the meaning of the numerical code:\n 0 no error or warning occured while fitting the model;\n -1 an error occured while fitting the model;\n 1 a warning indicating that the model fit is questionnable occured (algorithm did not converge,\n non-positive sigma estimate for a normal heterogeneous model or large asymptotic bias);\n 2 the warning ’design matrix not of full rank’ occured, therefore some model’s coefficients are not\n estimable;\n 3 a warning not of type 1 or 2 occured (the glm warning ’fitted rates nummerically 0 occured’\n is often encountered with small frequencies, it does not always mean that the model fit is\n questionnable).\n The elements in the column infoFit can contain more than one number since more than one warn-\n ing can occur. For exemple, if infoFit takes the value 13 for a model, it means that at least one\n warning of type 1 and one warning of type 3 have occured. If more than one warning of the same\n type are encountered, the number representing the type of warning is not repeated in infoFit.\nAuthor(s)\n <NAME> <<EMAIL>> and <NAME>\nReferences\n <NAME>. and <NAME>. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\n <NAME>. and <NAME>. (2005) Nonparametric Maximum Likelihood Estimation of Population\n Size Based on the Counting Distribution. Journal of the Royal Statistical Society: Series C (Applied\n Statistics), 54(4), 721-737.\n <NAME>. (1987) Estimating the population size for capture-recapture data with unequal catchability.\n Biometrics, 43(4), 783–791.\n <NAME>. (1985) Example of the use of glim to analyze capture-recapture studies. In Lec-\n ture Notes in Statistics 29: Statistics in Ornithology, <NAME>. et <NAME>, New York,:\n Springer-Verlag, 242–274.\n <NAME>. (1989) Loglinear models for capture-recapture. Biometrics, 45, 395–413.\n <NAME>. (1992) Interval estimation for mark-recapture studies of closed populations. Bio-\n metrics, 48, 567–576.\n <NAME>. (1993) Variances of mark-recapture estimates. Biometrics, 49, 1188–1193.\n <NAME>. and <NAME>. (1991) Inference for Poisson and multinomial models for capture-\n recapture experiments. Biometrika, 78(4), 911–916.\n <NAME>. and <NAME>. (2001) Improved loglinear model estimators of abundance in capture-\n recapture experiments. Canadian Journal of Statistics, 29, 555–572.\n <NAME>. and <NAME>. (2004) Loglinear models for the robust design in mark-recapture exper-\n iments. Biometrics, 60, 100–107.\n <NAME>. and <NAME>. (2007) Applications and extensions of Chao’s moment estimator for\n the size of a closed population. Biometrics, 63(4), 999–1006.\n <NAME>. (2008) Why a time effect often has a limited impact on capture-recapture estimates in\n closed populations. Canadian Journal of Statistics, 36(1), 75–84.\nExamples\n # Here is an example on the lesbian data set.\n desc <- descriptive(lesbian, dfreq = TRUE)\n desc\n plot(desc)\n # 1612 out of 2185 individuals (74%) appear on one list only.\n # The exploratory heterogeneity graph are not quite linear.\n # Some heterogeneity in the units capture probabilities\n # seem present in the data set.\n closedp(lesbian, dfreq = TRUE)\n # According to the BIC, the best model is <NAME>.\n # Let's see if adding interactions between capture\n # histories to the model could improve the model's fit.\n closedpMS.t(lesbian, dfreq = TRUE, h = \"Darroch\")\n # According to the BIC, the best heterogeneous Darroch model\n # contains the double interactions 12, 13, 14.\n # Here is the profile likelihood confidence interval for the\n # abundance estimation from this model.\n closedpCI.t(lesbian, dfreq = TRUE, mX = \"[12,13,14]\", h = \"Darroch\")\n ####################################################\n # Example to illustrate warnings management in closed population functions.\n # Here is a capture-recapture data set one could encounter.\n crdata <- cbind(histpos.t(4), c(0,0,3,0,0,0,0,0,0,0,0,1,0,0,2))\n # This data set contains 4 capture occasions but only 6 captured units.\n # Fitting capture-recapture models on this data set is quite useless.\n # The population size should be very close to the sample size.\n # Such small frequencies in a capture-recapture data set should\n # lead to warnings when fitting a loglinear model on it.\n ex <- closedp.t(crdata, dfreq=TRUE)\n ex\n # Many models produce warnings of type 1 indicating that the model fit\n # is questionnable. The very large abundance estimation for some models\n # are another indicator of questionable model fits.\n # Details about the warnings are found in the glm.warn element of the output.\n ex$glm.warn\n BBS2001 Species Richness Data from the North American Breeding Bird Survey\n (BBS) in 2001\nDescription\n This data set contains species richness data from the North American Breeding Bird Survey (BBS)\n in 2001. The number of capture occasions t is 50.\nUsage\n BBS2001\nFormat\n 36 by 2 numeric matrix, with the following columns:\n nbcap Numbers of captures\n freq Observed frequencies for each number of captures\nDetails\n This data set is presented in Dorazio and Royle (2003). It comes from the North American Breeding\n Bird Survey (BBS). This survey includes more than 4000 39.4 km survey routes throughout North\n America. Along each survey route an observer stops at 50 equidistant locations and records the\n species identity and number of all birds heard or seen within a 3-minute period.\n The data in BBS2001 was collected along Route 123, in Maryland, in 2001. In this data set a capture\n occasion is in fact a location and the frequencies represent the number of species detected at i\n locations out of the 50 locations on that route. The frequencies for i = 37, . . . , 50 are null and do\n not appear in the data set.\nSource\n Dorazio, <NAME>. and <NAME>. (2003) Mixture models for estimating the size of a closed population\n when capture rates vary among individuals. Biometrics, 59, 351–364\nExamples\n desc <- descriptive(BBS2001, dfreq = TRUE, dtype = \"nbcap\", t = 50)\n plot(desc)\n # Note in this plot the convex shape typical of model Mh.\n cp <- closedp.0(BBS2001, dfreq = TRUE, dtype = \"nbcap\", t = 50, t0 = 20)\n cp\n plot(cp)\n # The Gamma estimator has the smallest deviance and all\n # its residuals are small.\n # Let's calculate a profile confidence interval for the gamma\n # estimator:\n closedpCI.0(BBS2001, dfreq = TRUE, dtype = \"nbcap\", t = 50, t0 = 20,\n m = \"Mh\", h = \"Gamma\", h.control = list(theta = 3.5))\n # One can also calculate a profile confidence interval for Chao's\n # lower bound:\n closedpCI.0(BBS2001, dfreq = TRUE, dtype = \"nbcap\", t = 50, t0 = 20,\n m = \"Mh\", h = \"Chao\")\n bunting Lazuli Bunting Data\nDescription\n This data set contains frequencies of capture histories for 8 years of observation of lazuli buntings.\nUsage\n bunting\nFormat\n 255 by 9 numeric matrix, with the following columns:\n p1, p2, p3, p4, p5, p6, p7, p8 Capture histories from eight periods\n freq Observed frequencies for each capture history\nDetails\n The data come from an eight-year (1973 to 1980) study by <NAME> of lazuli buntings\n wintering in Logan, Utah. This data set is analyzed in Burnham and al. (1987) and in Cormack\n (1993).\n Each row of this data set represents an observed capture history followed by its frequency.\nReferences\n <NAME>. and <NAME>. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\n <NAME>., <NAME>., <NAME>., <NAME>. and <NAME>. (1987) Design and\n analysis methods for fish survival experiments based on release-recapture. American Fisheries\n Society Monographs 5. Bethesda, Maryland.\n <NAME>. (1993) The flexibility of GLIM analyses of multiple recapture or resighting data.\n In <NAME> and <NAME>: Marked Individuals in the Study of Bird Population. Basel,\n Switzerland: <NAME>, 39–49.\nExamples\n descriptive(bunting, dfreq = TRUE)\n # 1430 birds among the 1681 birds seen (85%) were caught only once.\n # This suggests the presence of transient birds at each capture occasion.\n op.m1 <- openp(bunting, dfreq = TRUE)\n op.m1$model.fit\n plot(op.m1)\n # The residuals plot shows large residuals for the birds caught twice or\n # more while the residuals are small for birds caught once. The Jolly-Seber\n # model does not fit well and the likely presence of transients might\n # cause that. Let's remove the birds caught only once from the analysis.\n keep2 <- rowSums(histpos.t(8)) > 1\n op.m2<- openp(bunting, dfreq = TRUE, keep = keep2)\n op.m2$model.fit\n # The deviance drop of 94 for 6 degrees of freedom is highly significant.\n plot(op.m2)\n # The residual plot still shows Pearson residuals larger than 4. We redo\n # the analysis without the transient birds and without the large residuals.\n keep3p <- residuals(op.m2$glm, type = \"pearson\") < 4\n num3 <- ((1:255)[keep2])[keep3p]\n keep3 <- rep(FALSE, 255)\n keep3[num3] <- TRUE\n op.m3 <- openp(bunting, dfreq = TRUE, keep = keep3)\n cbind(op.m2$survivals, op.m3$survivals)\n # These changes have little impact on the survival estimates.\n # We now test the equality of the survival probabilities and estimate its\n # common value phi. Least squares estimates of phi and its standard error:\n siginv <- solve(op.m2$cov[8:12, 8:12])\n phi <- t(rep(1, 5)) %*% siginv %*% op.m2$survivals[2:6, 1] /\n (t(rep(1, 5)) %*% siginv %*% rep(1, 5))\n se <- 1 / sqrt(t(rep(1, 5)) %*% siginv %*% rep(1, 5))\n cbind(estimate = phi, stderr = se)\n # The chi-square goodness of fit statistic for a constant survival\n # and its pvalue are:\n chisq4 <- t(op.m2$survivals[2:6, 1] - as.vector(phi)) %*% siginv %*%\n (op.m2$survivals[2:6, 1] - as.vector(phi))\n cbind(chi2stat = chisq4, pvalue = 1 - pchisq(chisq4, df = 4))\n # The hypothesis of a constant survival is accepted.\n cbird Catbird Site Occupancy Data\nDescription\n This data set contains site occupancy data for catbirds. The number of capture occasions t is 11.\nUsage\n cbird\nFormat\n 6 by 2 numeric matrix, with the following columns:\n nbcap Numbers of captures\n freq Observed frequencies for each number of captures\nDetails\n This data set is presented in Royle (2006). It comes from an experiment in which 50 locations have\n been visited 11 times. The presence or absence of catbirds was noted at each visit. The experiment\n aims at estimating the number of locations (<=50) at which catbirds are present.\n In the data set cbird, the frequencies represent the number of locations at which cbirdirbs have\n been detected i times, out of t = 11 visits. The frequencies for i = 7, . . . , 11 are null and do not\n appear in the data set.\nSource\n <NAME>. (2006) Site occupancy models with heterogeneous detection probabilities. Biometrics,\n 62, 97–102\nExamples\n desc <- descriptive(cbird, dfreq = TRUE, dtype = \"nbcap\", t = 11)\n plot(desc)\n # The heterogeneity graph is mildly convex, mostly because of\n # 2 sites where catbirds have been seen six times.\n closedp.0(cbird, dfreq = TRUE, dtype = \"nbcap\", t = 11)\n # The residuals are OK. The occupancy rate (N/50) estimate\n # varies between 40.6% for M0 and 49.4% for Mh Gamma.\n # One could also try fitting M0 removing the 2 unusual sites:\n closedp.0(cbird, dfreq = TRUE, dtype = \"nbcap\", t = 11, t0 = 5)\n # M0 fits very well and the occupancy rate estimate is 42.6%.\n closedp Loglinear Models for Closed Population Capture-Recapture Experi-\n ments\nDescription\n The functions closedp.t and closedp.0 fit various loglinear models for closed populations in\n capture-recapture experiments. For back compatibility, closedp.t is also named closedp.\n closedp.t fits more models than closedp.0 but for data set with more than 20 capture occasions,\n the function might fail. However, closedp.0 works with fairly large data sets (see Details).\nUsage\n closedp(X, dfreq=FALSE, neg=TRUE, ...)\n closedp.t(X, dfreq=FALSE, neg=TRUE, ...)\n closedp.0(X, dfreq=FALSE, dtype=c(\"hist\",\"nbcap\"), t=NULL, t0=NULL,\n neg=TRUE, ...)\n ## S3 method for class 'closedp'\n print(x, ...)\n ## S3 method for class 'closedp'\n boxplot(x, main=\"Boxplots of Pearson Residuals\", ...)\n ## S3 method for class 'closedp'\n plot(x, main=\"Residual plots for some heterogeneity models\", ...)\nArguments\n X The matrix of the observed capture histories (see Rcapture-package for a de-\n scription of the accepted formats).\n dfreq A logical. By default FALSE, which means that X has one row per unit. If TRUE,\n it indicates that the matrix X contains frequencies in its last column.\n dtype A characters string, either \"hist\" or \"nbcap\", to specify the type of data.\n \"hist\", the default, means that X contains complete observed capture histories.\n \"nbcap\" means that X contains numbers of captures (see Rcapture-package\n for details on data formats).\n t Requested only if dtype=\"nbcap\". A numeric specifying the total number of\n capture occasions in the experiment. For closedp.0, the value t=Inf is ac-\n cepted. It indicates that captures occur in continuous time (see Rcapture-package).\n t0 A numeric. Models are fitted considering only the frequencies of units captured\n 1 to t0 times. By default, if t is not equal to Inf, t0=t. When t=Inf, the default\n value of t0 is the larger number of captures observed.\n neg If this option is set to TRUE, negative eta parameters in Chao’s lower bound\n models are set to zero (see Details).\n ... Further arguments to be passed to glm, print.default, boxplot.default or\n plot.default.\n x An object, produced by a closedp function, to print or to plot.\n main A main title for the plot.\nDetails\n closedp.t fits models M0, Mt, Mh Chao (LB), Mh Poisson2, Mh Darroch, Mh Gamma3.5, Mth\n Chao (LB), Mth Poisson2, Mth Darroch, Mth Gamma3.5, Mb and Mbh. closedp.0 fits only models\n M0, Mh Chao (LB), Mh Poisson2, Mh Darroch and Mh Gamma3.5. However, closedp.0 can\n be used with larger data sets than closedp.t. This is explained by the fact that closedp.t fits\n models using the frequencies of the observable capture histories (vector of size 2t − 1), whereas\n closedp.0 uses the numbers of units captured i times, for i = 1, . . . , t (vector of size t). See\n Rcapture-package for more details about the distinction between .t and .0 functions.\n Multinomial profile confidence intervals for the abundance are constructed by closedpCI.t and\n closedpCI.0.\n To calculate bias corrected abundance estimates, use the closedp.bc function.\n CHAO’S LOWER BOUND MODEL\n Chao’s (or LB) models estimate a lower bound for the abundance, both with a time effect (Mth\n Chao) and without one (Mh Chao). The estimate obtained under Mh Chao is Chao’s (1987) moment\n estimator. Rivest and Baillargeon (2007) exhibit a loglinear model underlying this estimator and\n provide a generalization to Mth. For these two models, a small deviance means that there is an\n heterogeneity in capture probabilities; it does not mean that the lower bound estimate is unbiased.\n To test whether a certain model for heterogeneity is adequate, one can conduct a likelihood ratio\n test by subtracting the deviance of Chao’s model to the deviance of the heterogeneous model under\n study. If this heterogeneous model includes a time effect, it must be compared to model Mth\n Chao. If it does not include a time effect, it must be compared to model Mh Chao. Under the\n null hypothesis of equivalence between the two models, the difference of deviances follows a chi-\n square distribution with degrees of freedom equal to the difference between the models’ degrees of\n freedom.\n Chao’s lower bound models contain t − 2 parameters, called eta parameters, for the heterogene-\n ity. These parameters should theoretically be greater or equal to zero (see Rivest and Baillargeon\n (2007)). When the argument neg is set to TRUE (the default), negative eta parameters are set to\n zero (to do so, columns are removed from the design matrix of the model). Degrees of freedom of\n Chao’s model increase when eta parameters are set to zero.\n OTHER MODELS FOR HETEROGENEITY\n Other models for heterogeneity are defined as follows :\n Model Column for heterogeneity in the design matrix\n Poisson2 2k − 1\n Darroch k 2 /2\n Gamma3.5 − log(3.5 + k) + log(3.5)\n where k is the number of captures. Poisson and Gamma models with alternative to the parameter\n defaults values 2 and 3.5 can be fitted with the closedpCI.t and closedpCI.0 functions.\n Darroch’s models for Mh and Mth are considered by Darroch et al. (1993) and Agresti (1994). Pois-\n son and Gamma models are discussed in Rivest and Baillargeon (2007). Poisson models typically\n yield smaller corrections for heterogeneity than Darroch’s model since the capture probabilities are\n bounded from below under these models. On the other hand, Gamma models can lead to very large\n estimators of abundance. We suggest considering this estimator only in experiments where very\n small capture probabilities are likely.\n PLOT METHODS AND FUNCTIONS\n The boxplot.closedp function produces boxplots of the Pearson residuals of the fitted loglinear\n models that converged.\n The plot.closedp function produces scatterplots of the Pearson residuals in terms of fi (number of\n units captured i times) for the heterogeneous models Mh Poisson2, Mh Darroch and Mh Gamma3.5\n if they converged.\nValue\n n The number of captured units.\n t The total number of capture occasions in the data matrix X. When captures oc-\n cur in continuous time (input argument t=Inf), this output value is the larger\n number of captures observed.\n t0 For closedp.0 only: the value of the argument t0 used in the computations.\n results A table containing, for every fitted model:\n abundance: the estimated population size,\n stderr: the standard error of the estimated population size,\n deviance: the model’s deviance,\n df: the number of degrees of freedom,\n AIC: the Akaike’s information criterion,\n BIC: the bayesian information criterion,\n infoFit: a numerical code giving information about error or warnings encoun-\n tered when fitting the model (see Rcapture-package for details).\n bias A vector, the asymptotic bias of the estimated population size for every fitted\n model.\n glm A list of the ’glm’ objects obtained from fitting models.\n glm.err A list of character string vectors. If the glm function generates an error when\n fitting a model, a copy of the error message is stored in glm.err$mname where\n mname is a short name to identify the model. A NULL list element means that no\n error occured for the considered model.\n glm.warn A list of character string vectors. If the glm function generates one or more warn-\n ings when fitting a model, a copy of these warnings are stored in glm.warn$mname\n where mname is a short name to identify the model. A NULL list element means\n that glm did not produce any warnings for the considered model.\n parameters Capture-recapture parameters estimates. It contains N, the estimated population\n size, and p or p1 to pt defined as follows for the different models :\n M0: the capture probability at any capture occasion,\n Mt: the capture probabilities for each capture occasion,\n Mh models: the average probability of capture,\n Mth models: the average probabilities of capture for each occasion,\n Mb and Mbh: the probability of first capture at any capture occasion.\n For models Mb and Mbh, it also contains c, the recapture probability at any\n capture occasion.\n neg.eta The position of the eta parameters set to zero in the loglinear parameter vector\n of models MhC and MthC. NULL if no parameters are set to zero.\n X A copy of the data given as input in the function call.\n dfreq A copy of the dfreq argument given in the function call.\nNote\n This function uses the glm function of the stats package.\nAuthor(s)\n <NAME> <<EMAIL>> and <NAME>\nReferences\n Agresti, A. (1994) Simple capture-recapture models permitting unequal catchability and variable\n sampling effort. Biometrics, 50, 494–500.\n Baillargeon, S. and Rivest, L.P. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\n Chao, A. (1987) Estimating the population size for capture-recapture data with unequal catcha-\n bililty. Biometrics, 45, 427–438.\n <NAME>., <NAME>., <NAME>. and <NAME>. (1993) A three sample multiple capture-\n recapture approach to the census population estimation with heterogeneous catchability. Journal of\n the American Statistical Association, 88, 1137–1148.\n <NAME>. and <NAME>. (2001) Improved loglinear model estimators of abundance in capture-\n recapture experiments. Canadian Journal of Statistics, 29, 555–572.\n <NAME>. and <NAME>. (2007) Applications and extensions of Chao’s moment estimator for\n the size of a closed population. Biometrics, 63(4), 999–1006.\n <NAME>. (1982) The Estimation of Animal Abundance and Related Parameters, 2nd edition,\n New York: Macmillan.\nSee Also\n closedpCI.t, closedpCI.0, closedp.bc, closedp.Mtb, closedpMS.t, uifit.\nExamples\n # hare data set\n hare.closedp <- closedp.t(hare)\n hare.closedp\n boxplot(hare.closedp)\n # Third primary period of mvole data set\n period3 <- mvole[, 11:15]\n closedp.t(period3)\n # BBS2001 data set\n BBS.closedp <- closedp.0(BBS2001, dfreq = TRUE, dtype = \"nbcap\", t = 50, t0 = 20)\n BBS.closedp\n plot(BBS.closedp)\n ### Seber (1982) p.107\n # When there is 2 capture occasions, the heterogeneity models cannot be fitted\n X <- matrix(c(1,1,167,1,0,781,0,1,254), byrow = TRUE, ncol = 3)\n closedp.t(X, dfreq = TRUE)\n ### Example of captures in continuous time\n # Illegal immigrants data set\n closedp.0(ill, dtype = \"nbcap\", dfreq = TRUE, t = Inf)\n closedp.bc Bias Correction for Closed Population Loglinear Models\nDescription\n This function applies a bias correction to the abundance estimations obtained by closed population\n models.\nUsage\n closedp.bc(X, dfreq=FALSE, dtype=c(\"hist\",\"nbcap\"), t=NULL, t0=t,\n m=c(\"M0\",\"Mt\",\"Mh\",\"Mth\",\"Mb\",\"Mbh\"), h=NULL, h.control=list(), ...)\n ## S3 method for class 'closedp.bc'\n print(x, ...)\nArguments\n X The matrix of the observed capture histories (see Rcapture-package for a de-\n scription of the accepted formats).\n dfreq A logical. By default FALSE, which means that X has one row per unit. If TRUE,\n it indicates that the matrix X contains frequencies in its last column.\n dtype A characters string, either \"hist\" or \"nbcap\", to specify the type of data. \"hist\",\n the default, means that X contains complete observed capture histories. \"nbcap\"\n means that X contains numbers of captures (see Rcapture-package for details\n on data formats). If m is \"Mt\", \"Mth\", \"Mb\" or \"Mbh\", dtype must be \"hist\".\n t Requested only if dtype=\"nbcap\". A numeric specifying the total number of\n capture occasions in the experiment.\n t0 A numeric used for model M0 or for an Mh model other than Chao’s lower\n bound model : Models are fitted considering only the frequencies of units cap-\n tured 1 to t0 times. By default t0=t.\n m A character string indicating the model to fit, either \"M0\"=M0 model, \"Mt\"=Mt\n model, \"Mh\"=Mh model, \"Mth\"=Mth model, \"Mb\"=Mb model, \"Mbh\"=Mbh\n model.\n h A character string (\"LB\", \"Chao\", \"Poisson\", \"Darroch\" or \"Gamma\") or a nu-\n merical R function specifying the form of the column(s) for heterogeneity in the\n design matrix.\n \"LB\" and \"Chao\" represents Chao’s lower bound model (the default),\n \"Poisson\" represents the function f (k) = thetak − 1,\n \"Darroch\" represents the function f (k) = k 2 /2,\n and \"Gamma\" represents the function f (k) = − log(theta + k) + log(theta)\n where k is the number of captures and theta is a parameter specified with the\n argument theta.\n If an R function is given, it must be the implementation of any convex mathe-\n matical function f (k) (it must have only one argument).\n h.control A list of elements to control the heterogeneous part of the model, if any.\n For a Poisson or Gamma heterogeneous model:\n theta The value of the parameter theta in f (k) = thetak − 1 for the Poisson\n model (the default value is 2) and in f (k) = − log(theta + k) + log(theta)\n for the Gamma model (the default value is 3.5).\n For the Chao’s lower bound Mth model:\n neg If this option is set to TRUE (the default), negative eta parameters in Chao’s\n lower bound models are set to zero.\n ... Further arguments to be passed to glm or print.default.\n x An object, produced by the closedp.bc function, to print.\nDetails\n For the Mt model:\n When t=2, closedp.bc returns the Petersen estimator with Chapman’s (1951) bias correction and\n the bias corrected standard error estimator of Seber (1970) and Wittes (1972). For t>2, closedp.bc\n implements the bias correction of Rivest and Levesque (2001). The estimate for N and its variance\n are calculated by solving an estimating equation as proposed in Seber (1982), not by fitting a Poisson\n regression. This approach works for large values of t.\n For other models:\n The bias correction is done through frequency modifications in Poisson regression as described in\n Rivest and Levesque (2001). The variances calculated with the modified frequencies are less biased\n than the standard ones, but they can overestimate the mean squared errors, especially when the data\n is sparse.\n This function works with fairly large data set, except if an \"Mth\" model is requested. In this case,\n only heterogeneity of the form \"LB\", \"Chao\", \"Poisson\" with theta=2 or \"Darroch\" is accepted.\nValue\n n The number of captured units\n t The total number of capture occasions in the data matrix X.\n t0 For models M0 and Mh only: the value of the argument t0 used in the compu-\n tations.\n results A table containing, for the fitted model:\n abundance: the estimated population size,\n stderr: the standard error of the estimated population size,\n infoFit: a numerical code giving information about error or warnings encoun-\n tered when fitting the model (see Rcapture-package for details).\n glm.warn Only if the corrected population size estimation was obtained with glm: A vector\n of character strings. If the glm function generates one or more warnings when\n fitting the model, a copy of these warnings are stored in glm.warn. NULL if glm\n did not produce any warnings.\n neg.eta For Chao’s lower bound model Mth only: the position of the eta parameters set\n to zero in the loglinear parameter vector, if any.\nNote\n This function uses the glm function of the stats package, except for models Mt and Mh Chao’s\n lower bound for which exact calculation is performed.\nAuthor(s)\n <NAME> <<EMAIL>> and <NAME>\nReferences\n Baillargeon, S. and <NAME>. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\n <NAME>. (1951) Some properties of the hypergeometric distribution with applications to\n zoological sample censuses. University of California Publications in Statistics, 1(7), 131-160.\n <NAME>. and <NAME>. (2001) Improved loglinear model estimators of abundance in capture-\n recapture experiments. Canadian Journal of Statistics, 29, 555-572.\n <NAME>. (1970) The effects of trap response on tag recapture estimates. Biometrics, 26, 13-22.\n <NAME>. (1982) The Estimation of Animal Abundance and Related Parameters, 2nd edition.\n New York: Macmillan.\n <NAME>. (1972) On the bias and estimated variance of Chapman’s two-sample capture-recapture\n population estimate. Biometrics, 28, 592-597.\nSee Also\n closedp\nExamples\n # Third primary period of mvole data set\n period3 <- mvole[, 11:15]\n closedp.bc(period3, m = \"Mh\", h = \"Darroch\")\n closedp.bc(period3, m = \"Mh\", h = \"Gamma\", h.control = list(theta = 3.5))\n # BBS2001 data set\n closedp.bc(BBS2001, dfreq = TRUE, dtype = \"nbcap\", t = 50, t0 = 20,\n m = \"Mh\", h = \"Gamma\", h.control = list(theta = 3.5))\n # Seber (1982) p.107\n # When there are 2 capture occasions, only models M0 and Mt can be fitted\n X <- matrix(c(1,1,167,1,0,781,0,1,254), byrow = TRUE, ncol = 3)\n closedp.bc(X, dfreq = TRUE, m = \"M0\")\n closedp.bc(X, dfreq = TRUE, m = \"Mt\")\n closedp.custom Customization of a Loglinear Model for Closed Populations Capture-\n Recapture Experiments\nDescription\n As of version 1.2-0 of Rcapture, these functions are deprecated, but kept for back compatibility.\n Please use closedpCI.t instead.\n The closedp.mX function fits a loglinear model given a design matrix mX. The closedp.h function\n fits Mh or Mth models for which the form of the column for heterogeneity in the design matrix is\n determined by the user.\nUsage\n closedp.mX(X, dfreq=FALSE, mX, mname=\"Customized model\")\n closedp.h(X, dfreq=FALSE, m=\"Mh\", h=\"Poisson\", a=2)\n ## S3 method for class 'closedp.custom'\n print(x, ...)\n ## S3 method for class 'closedp.custom'\n boxplot(x, ...)\nArguments\n X The matrix of the observed capture histories (see Rcapture-package for a de-\n scription of the accepted formats).\n dfreq A logical. By default FALSE, which means that X has one row per unit. If\n TRUE, it indicates that the matrix X contains frequencies in its last column.\n mX The design matrix of the loglinear model. In this matrix, the order of the capture\n histories is as defined in the histpos.t function.\n mname A character string specifying the name of the customized model.\n m A character string indicating the model to fit, either \"Mh\"=Mh model or \"Mth\"=Mth\n model\n h The character string \"Poisson\" (2k − 1) or a numerical function taking only\n one argument specifying the form of the column for heterogeneity in the design\n matrix.\n a The value of the exponent’s base for a Poisson model.\n x An object, produced by the closedp.mX or the closedp.h function, to print or\n to plot.\n ... Further arguments passed to or from other methods.\nDetails\n An intercept is added to the model. Therefore, the mX matrix must not contain a column of ones.\n The abundance estimation is calculated as the number of captured units plus the exponential of the\n intercept. Therefore, these functions are not suited for models with a behavioral effect.\n In closedp.h, the argument h cannot take the value \"Chao\" or \"Darroch\" because these models are\n already fitted by the closedp function.\n The boxplot.closedp.custom function produces a boxplot of the pearson residuals of the cus-\n tomized model.\nValue\n n The number of captured units\n results A table containing the estimated population size and its standard error, the de-\n viance, the number of degrees of freedom and the Akaike’s information crite-\n rion.\n glm The ’glm’ object obtained from fitting the model.\nNote\n These functions use the glm function of the stats package.\nAuthor(s)\n <NAME> <<EMAIL>> and <NAME>\nReferences\n <NAME>. and <NAME>. (2007) Applications and extensions of Chao’s moment estimator for\n the size of a closed population. Biometrics, 63(4), 999–1006.\nSee Also\n closedp, closedp.Mtb\nExamples\n # HIV data set\n mat <- histpos.t(4)\n mX2 <- cbind(mat, mat[, 1] * mat[, 2])\n closedp.mX(HIV, dfreq = TRUE, mX = mX2)\n # Third primary period of mvole data set\n period3 <- mvole[, 11:15]\n psi <- function(x) { -log(3.5 + x) + log(3.5) }\n closedp.h(period3, h = psi)\n closedp.Mtb Closed Population Capture-Recapture Model Mtb\nDescription\n This function fits model Mtb for closed populations in capture-recapture experiments.\nUsage\n closedp.Mtb(X, dfreq=FALSE, method = \"BFGS\", ...)\n ## S3 method for class 'closedp.Mtb'\n print(x, ...)\nArguments\n X The matrix of the observed capture histories (see Rcapture-package for a de-\n scription of the accepted formats).\n dfreq A logical. By default FALSE, which means that X has one row per unit. If TRUE,\n it indicates that the matrix X contains frequencies in its last column.\n method The method to be used by optim. The default is \"BFGS\".\n ... Further arguments to be passed to optim or print.default.\n x An object, produced by the closedp.Mtb function, to print.\nDetails\n The Mtb model is non-linear. It is fitted with the optim function instead of the glm function.\n Therefore, the abundance estimate can be unstable.\n For the model to be identifiable, the parameters are constrained in the following way: logit(ci ) =\n logit(pi ) + b for i in 2, . . . , l.\nValue\n n The number of captured units\n t The total number of capture occasions in the data matrix X.\n results A table containing, for the fitted model:\n abundance: the estimated population size,\n stderr: the standard error of the estimated population size,\n deviance: the model’s deviance,\n df: the number of degrees of freedom,\n AIC: the Akaike’s information criterion,\n BIC: the bayesian information criterion,\n infoFit: a numerical code giving information about error or warnings encoun-\n tered when fitting the model (see Rcapture-package for details).\n optim The output produced by optim from fitting the model.\n optim.warn A vector of character strings. If the optim function generates one or more warn-\n ings when fitting the model, a copy of these warnings are stored in optim.warn.\n NULL if optim did not produce any warnings.\n parMtb Capture-recapture parameters estimates for model Mtb : the abundance N, p1 to\n pt , the probabilities of first capture for each capture occasion, and c2 to ct , the\n recapture probabilities for each capture occasion.\nAuthor(s)\n <NAME> <<EMAIL>> and <NAME>\nReferences\n Baillargeon, S. and Rivest, L.P. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\nSee Also\n closedp, closedpCI.t\nExamples\n # hare data set\n closedp.Mtb(hare)\n ## Example producing an unstable estimate\n # Fourth primary period of mvole data set\n period4 <- mvole[, 16:20]\n closedp.Mtb(period4)\n closedpCI Customization of a Loglinear Model and Confidence Interval for\n Abundance Estimation in Closed Population Capture-Recapture Ex-\n periments\nDescription\n The closedpCI.t and closedpCI.0 functions fit a loglinear model specified by the user and com-\n pute a confidence interval for the abundance estimation. For a normal heterogeneous model, a\n log-transformed confidence interval (Chao 1987) is produced. For any other model, the multino-\n mial profile likelihood confidence interval (Cormack 1992) is produced.\n The model is identified with the argument m or mX. For heterogeneous models, the form of the het-\n erogeneity is specified with the arguments h and h.control. If h is given with mX, heterogeneity is\n added in mX.\n These functions extend closedp.t and closedp.0 as they broaden the range of models one can fit\n and they compute confidence intervals. Unlike the closedp functions, it fits only one model at a\n time.\nUsage\n closedpCI.t(X, dfreq=FALSE, m=c(\"M0\",\"Mt\",\"Mh\",\"Mth\"), mX=NULL,\n h=NULL, h.control=list(), mname=NULL, alpha=0.05,\n fmaxSupCL=3, ...)\n closedpCI.0(X, dfreq=FALSE, dtype=c(\"hist\",\"nbcap\"), t=NULL, t0=NULL,\n m=c(\"M0\",\"Mh\"), mX=NULL, h=NULL, h.control=list(),\n mname=NULL, alpha=0.05, fmaxSupCL=3, ...)\n plotCI(x.closedpCI, main=\"Profile Likelihood Confidence Interval\", ...)\n ## S3 method for class 'closedpCI'\n print(x, ...)\n ## S3 method for class 'closedpCI'\n boxplot(x, main=\"Boxplots of Pearson Residuals\", ...)\n ## S3 method for class 'closedpCI'\n plot(x, main=\"Scatterplot of Pearson Residuals\", ...)\nArguments\n X The matrix of the observed capture histories (see Rcapture-package for a de-\n scription of the accepted formats).\n dfreq A logical. By default FALSE, which means that X has one row per unit. If TRUE,\n it indicates that the matrix X contains frequencies in its last column.\n dtype A characters string, either \"hist\" or \"nbcap\", to specify the type of data.\n \"hist\", the default, means that X contains complete observed capture histories.\n \"nbcap\" means that X contains numbers of captures (see Rcapture-package\n for details on data formats).\n t Requested only if dtype=\"nbcap\". A numeric specifying the total number of\n capture occasions in the experiment. For closedpCI.0, the value t=Inf is ac-\n cepted. It indicates that captures occur in continuous time (see Rcapture-package).\n t0 A numeric. Models are fitted considering only the frequencies of units captured\n 1 to t0 times. By default, if t is not equal to Inf, t0=t. When t=Inf, the default\n value of t0 is the larger number of captures observed.\n m A character string indicating the model to fit. For closedpCI.0 it can be either\n \"M0\"=M0 model or \"Mh\"=Mh model. For closedpCI.t it can also be \"Mt\"=Mt\n model or \"Mth\"=Mth model.\n mX The design matrix of the loglinear model. By default, the design matrix is built\n based on the m argument. If a mX argument is given, it must be a matrix (or an\n object that can be coerced to a matrix by as.matrix). The order of the capture\n histories in the mX matrix must be as defined in the histpos.t or histpos.0\n function.\n For the closedpCI.t function only, mX can also be an object of class \"formula\"\n or a hierarchical loglinear model name. The details of a formula specification\n are given under Details and the details of a hierarchical loglinear model name\n specification are given in the closedpMS.t documentation.\n An intercept is always added to the model. Therefore, mX must not contain a\n column of ones if it is a matrix. Also, if the argument h is not NULL, one or more\n columns for heterogeneity are added to the design matrix.\n h A character string (\"LB\", \"Chao\", \"Poisson\", \"Darroch\", \"Gamma\" or \"Nor-\n mal\") or a numerical R function specifying the form of the column(s) for hetero-\n geneity in the design matrix.\n \"LB\" and \"Chao\" represents Chao’s lower bound model (the default),\n \"Poisson\" represents the function f (k) = thetak − 1,\n \"Darroch\" represents the function f (k) = k 2 /2,\n and \"Gamma\" represents the function f (k) = − log(theta + k) + log(theta)\n where k is the number of captures and theta is a parameter specified with the\n h.control element theta.\n \"Normal\" represents heterogeneity modeled with a normal mixing distribution,\n as presented in Rivest (2011).\n If an R function is given, it must be the implementation of any convex mathe-\n matical function f (k) (it must have only one argument).\n h.control A list of elements to control the heterogeneous part of the model, if any.\n For a Poisson or Gamma heterogeneous model:\n theta The value of the parameter theta in f (k) = thetak − 1 for the Poisson\n model (the default value is 2) and in f (k) = − log(theta + k) + log(theta)\n for the Gamma model (the default value is 3.5).\n For a Chao’s lower bound heterogeneous model:\n neg If this option is set to TRUE (the default), negative eta parameters in Chao’s\n lower bound models are set to zero.\n For a Normal heterogeneous model:\n initcoef Initial values for the loglinear coefficients given to optim. The de-\n fault values are taken from Darroch’s model.\n initsig Initial value for the sigma parameter of the normal mixing distribution.\n The default is 0.2.\n method The method to be used by optim. The default is \"BFGS\".\n mname A character string specifying the name of the customized model. By default, it\n is derived from the arguments specifying the model.\n alpha A confidence interval with confidence level 1-alpha is constructed. The value\n of alpha must be between 0 and 1; the default is 0.05.\n fmaxSupCL A numeric (by default 3). The upper end point of the interval to be searched\n by uniroot to find the upper bound of the multinomial profile likelihood con-\n fidence interval (Cormack 1992) is defined by fmaxSupCL×N̂ . In this prod-\n uct, fmaxSupCL is a factor multiplying N̂ , the population size estimated by\n Poisson regression. If the upper bound obtained is equal the upper end point\n fmaxSupCL×N̂ , the element SupCL of the output value CI will begin with the\n symbol >. In that case, one could try to increase the value of fmaxSupCL to\n find the upper bound of the multinomial profile likelihood confidence interval\n instead of a minimum value for this upper bound.\n ... Further arguments to be passed to glm, optim, print.default, plot.default\n or boxplot.default.\n x.closedpCI An object, produced by a closedpCI function, to produce a plot of the multino-\n mial profile likelihood for N .\n main A main title for the plot\n x An object, produced by a closedpCI function, to print or to plot.\nDetails\n The closedpCI.t function fits models using the frequencies of the observable capture histories\n (vector of size 2t − 1), whereas closedpCI.0 uses the number of units capture i times, for i =\n 1, . . . , t (vector of size t). Thus, closedpCI.0 can be used with data sets larger than those for\n closedpCI.t, but it cannot fit models with a temporal effect. See Rcapture-package for more\n details about the distinction between .t and .0 functions.\n These functions do not work for closed population models featuring a behavioral effect, such as Mb\n and Mbh. The abundance estimation is calculated as the number of captured units plus the expo-\n nential of the Poisson regression intercept. However, models with a behavioral effect can by fitted\n with closedp.t (Mb and Mbh), closedp.Mtb and closedp.bc.\n CHAO’S LOWER BOUND MODELS\n Chao’s lower bound models estimate a lower bound for the abundance. Rivest (2011) presents a\n generalized loglinear model underlying this estimator. To test whether a certain model for hetero-\n geneity is adequate, one can conduct a likelihood ratio test by subtracting the deviance of a Chao’s\n lower bound model to the deviance of the heterogeneous model under study. The two models should\n have the same mX argument. Under the null hypothesis of equivalence between the two models, the\n difference of deviances follows a chi-square distribution with degrees of freedom equal to the dif-\n ference between the models’ degrees of freedom.\n A Chao’s lower bound model contains t − 2 parameters, called eta parameters, for the heterogene-\n ity. These parameters should theoretically be greater or equal to zero (see Rivest and Baillargeon\n (2007)). When the element neg of the argument h.control is set to TRUE (the default), negative\n eta parameters are set to zero (to do so, columns are removed from the design matrix of the model).\n Degrees of freedom of Chao’s lower bound model increase when eta parameters are set to zero.\n ARGUMENT mX : FORMULA SPECIFICATION\n For the closedpCI.t function, mX can be an object of class \"formula\". The only accepted vari-\n ables in this formula are c1 to ct. The variable ci represents a capture indicator (1 for a capture,\n 0 otherwise) for the ith capture occasions. Also, the formula must not contain a response variable\n since it is only used to construct the design matrix of the model. For example, if t=3, the Mt model\n is fitted if mX = ~ . or mX = ~ c1 + c2 + c3. The symbol . in this formula is a shortcut for c1 + c2 +\n ... + ct. Formula mX arguments facilitate the addition of interactions between capture occasions in\n the model. For example, if t=3, the Mt model with an interaction between the first and the second\n capture occasion is fitted if mX = ~ . + c1:c2. See formula for more details of allowed formulae.\n PLOT METHODS AND FUNCTIONS\n The boxplot.closedpCI function produces a boxplot of the Pearson residuals of the customized\n model.\n The plot.closedpCI function traces the scatterplot of the Pearson residuals in terms of fi (number\n of units captured i times) for the customized model.\n The plotCI function produces a plot of the multinomial profile likelihood for N . The value of N\n maximizing the profile likelihood and the bounds of the confidence interval are identified.\nValue\n n The number of captured units\n t The total number of capture occasions in the data matrix X. When captures oc-\n cur in continuous time (input argument t=Inf), this output value is the larger\n number of captures observed.\n t0 For closedpCI.0 only: the value of the argument t0 used in the computations.\n results A table containing, for the fitted model:\n abundance: the estimated population size,\n stderr: the standard error of the estimated population size,\n deviance: the model’s deviance,\n df: the number of degrees of freedom,\n AIC: the Akaike’s information criterion,\n BIC: the bayesian information criterion,\n infoFit: a numerical code giving information about error or warnings encoun-\n tered when fitting the model (see Rcapture-package for details).\n bias Not produced for normal heterogeneous models (h=\"Normal\"): The asymptotic\n bias of the estimated population size.\n fit The ’glm’ object obtained from fitting the model except for normal heteroge-\n neous models (h=\"Normal\"). These models are not fitted with glm. For them,\n fit is a list with the following elements:\n parameters: the matrix of parameters (loglinear coefficients + sigma parame-\n ter) estimates with their standard errors,\n varcov: the estimated variance-covariance matrix of the estimated parameters,\n y: the y vector used to fit the model,\n fitted.values: the model fitted values,\n initparam: the initial values for the parameters (loglinear coefficients + sigma\n parameter) used by optim,\n optim: the output produced by optim.\n fit.warn A vector of character strings. If the glm or optim (for normal heterogeneous\n models) function generates one or more warnings when fitting the model, a copy\n of these warnings are stored in fit.warn. NULL if no warnings occured.\n neg.eta For Chao’s lower bound models only: the position of the eta parameters set to\n zero in the loglinear parameter vector, if any.\n CI Not produced for normal heterogeneous models (h=\"Normal\"): A table contain-\n ing the abundance estimation and its multinomial profile likelihood confidence\n interval. The last column of the table, named infoCI, contains a numerical code\n giving information about error or warnings encountered when calculating the\n confidence interval. Here is a description of the meaning of this numerical code:\n 0 no error or warning occured;\n -1 an error occured while calculating the multinomial abundance estimation;\n 4 a warning occured while calculating the multinomial abundance estimation;\n 5 a warning occured while calculating the lower bound of the multinomial pro-\n file likelihood confidence interval;\n 6 a warning occured while calculating the upper bound of the multinomial pro-\n file likelihood confidence interval.\n CI.err Not produced for normal heterogeneous models (h=\"Normal\"): If an error oc-\n cured while calculating the multinomial abundance, a copy of the error message\n is stored in CI.err. NULL if no error occured.\n CI.warn Not produced for normal heterogeneous models (h=\"Normal\"): If one or more\n warnings occur while calculating the multinomial abundance estimation or the\n profile likelihood confidence interval, a copy of these warnings are stored in\n CI.warn. NULL if no warnings occured.\n alpha 1-the confidence level of the interval.\n N.CI Not produced for normal heterogeneous models (h=\"Normal\"): The x-coordinates\n for plot.closedpCI.t.\n loglik.CI Not produced for normal heterogeneous models (h=\"Normal\"): The y-coordinates\n for plot.closedpCI.t.\nNote\n For normal heterogeneous models, the closedpCI functions use optim from the stats package.\n Otherwise, models are fitted with glm and the code to compute the multinomial profile likelihood\n confidence interval calls the functions glm, optimize and uniroot, all from the stats package.\nAuthor(s)\n <NAME> <<EMAIL>> and <NAME>\nReferences\n <NAME>. and <NAME>. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\n <NAME>. (1987) Estimating the population size for capture-recapture data with unequal catchability.\n Biometrics, 43(4), 783–791.\n <NAME>. (1992) Interval estimation for mark-recapture studies of closed populations. Bio-\n metrics, 48, 567–576.\n <NAME>. (2011) A lower bound model for multiple record systems estimation with heterogeneous\n catchability. The International Journal of Biostatistics, 7(1), Article 23.\n <NAME>. and <NAME>. (2007) Applications and extensions of Chao’s moment estimator for\n the size of a closed population. Biometrics, 63(4), 999–1006.\nSee Also\n closedp, closedp.Mtb\nExamples\n # hare data set\n CI <- closedpCI.t(hare, m = \"Mth\", h = \"Poisson\", h.control = list(theta = 2))\n CI\n plotCI(CI)\n # HIV data set\n mat <- histpos.t(4)\n mX2 <- cbind(mat, mat[, 1] * mat[ ,2])\n closedpCI.t(HIV, dfreq = TRUE, mX = mX2, mname = \"Mt interaction 1,2\")\n # which can be obtained more conveniently with\n closedpCI.t(HIV, dfreq = TRUE, mX = ~ . + c1:c2, mname = \"Mt interaction 1,2\")\n # BBS2001 data set\n CI0 <- closedpCI.0(BBS2001, dfreq = TRUE, dtype = \"nbcap\", t = 50, t0 = 20,\n m = \"Mh\", h = \"Gamma\", h.control = list(theta = 3.5))\n CI0\n plot(CI0)\n plotCI(CI0)\n ### As an alternative to a gamma model, one can fit a negative Poisson model.\n ### It is appropriate in experiments where very small capture probabilities\n ### are likely. It can lead to very large estimators of abundance.\n # Third primary period of mvole data set\n period3 <- mvole[, 11:15]\n psi <- function(x) { 0.5^x - 1 }\n closedpCI.t(period3, m = \"Mh\", h = psi)\n ### Example of normal heterogeneous models\n ### diabetes data of Bruno et al. (1994)\n histpos <- histpos.t(4)\n diabetes <- cbind(histpos, c(58,157,18,104,46,650,12,709,14,20,7,74,8,182,10))\n # chosen interaction set I in Rivest (2011)\n closedpCI.t(X = diabetes, dfreq = TRUE, mX = ~ . + c1:c3 + c2:c4 + c3:c4,\n h = \"Normal\", mname = \"Mth normal with I\")\n ### Example of captures in continuous time\n # Illegal immigrants data set\n closedpCI.0(ill, dtype = \"nbcap\", dfreq = TRUE, t = Inf, m = \"Mh\", h = \"LB\")\n closedpMS Hierarchical Loglinear Model Selection in Closed Population\n Capture-Recapture Experiments\nDescription\n The closedpMS.t function fits every possible hierarchical loglinear models for a given closed pop-\n ulation capture-recapture data set, under the constraints set by the given maxorder and forced\n arguments. Parameters for heterogeneity in capture probabilites among units can be added to the\n models.\n The getAllModels function lists every possible hierarchical loglinear models for a certain number\n of capture occasions t, under the constraints set by the given maxorder and forced arguments.\nUsage\n closedpMS.t(X, dfreq = FALSE, h = NULL, h.control = list(),\n maxorder = t - 1, forced = 1:t, stopiflong = TRUE, ...)\n ## S3 method for class 'closedpMS'\n print(x, ...)\n ## S3 method for class 'closedpMS'\n plot(x, main=\"Models comparison based on BIC\", omitOutliers = TRUE, ...)\n getAllModels(t, maxorder = t - 1, forced = 1:t, stopiflong = TRUE)\nArguments\n X The matrix of the observed capture histories (see Rcapture-package for a de-\n scription of the accepted formats).\n dfreq A logical. By default FALSE, which means that X has one row per unit. If TRUE,\n it indicates that the matrix X contains frequencies in its last column.\n h A character string (\"LB\", \"Chao\", \"Poisson\", \"Darroch\", \"Gamma\" or \"Nor-\n mal\") or a numerical R function specifying the form of the column(s) for hetero-\n geneity in the design matrix.\n \"LB\" and \"Chao\" represents Chao’s lower bound model (the default),\n \"Poisson\" represents the function f (k) = thetak − 1,\n \"Darroch\" represents the function f (k) = k 2 /2,\n and \"Gamma\" represents the function f (k) = − log(theta + k) + log(theta)\n where k is the number of captures and theta is a parameter specified with the\n h.control element theta.\n \"Normal\" represents heterogeneity modeled with a normal mixing distribution,\n as presented in Rivest (2011).\n If an R function is given, it must be the implementation of any convex mathe-\n matical function f (k) (it must have only one argument).\n h.control A list of elements to control the heterogeneous part of the model, if any.\n For a Poisson or Gamma heterogeneous model:\n theta The value of the parameter theta in f (k) = thetak − 1 for the Poisson\n model (the default value is 2) and in f (k) = − log(theta + k) + log(theta)\n for the Gamma model (the default value is 3.5).\n For a Chao’s lower bound heterogeneous model:\n neg If this option is set to TRUE (the default), negative eta parameters in Chao’s\n lower bound models are set to zero.\n For a Normal heterogeneous model:\n initsig Initial value for the sigma parameter of the normal mixing distribution.\n The default is 0.2.\n method The method to be used by optim. The default is \"BFGS\".\n maxorder A numeric specifying the higher order accepted for the terms in the models. It\n can take a value between 1 and t - 1 (the default).\n forced A vector of the terms forced in the model (by default every first order terms).\n NULL if no terms are forced in the model.\n stopiflong A logical indicating whether the function execution should be stopped when\n the number of possibles models is larger than 10 000, that is when the function\n might be long to run (defaut TRUE).\n ... Further arguments to be passed to glm, optim or print.default.\n x An object, produced by the closedpMS.t function, to print or to plot.\n main A main title for the plot.\n omitOutliers A logical. If TRUE (the default), models with an outlier abundance or BIC value\n are removed from the plot. A value is considered an outlier if it is smaller than\n the first quartile minus 1.5 times the interquartile range, or larger than the third\n quartile plus 1.5 times the interquartile range.\n t A numeric specifying the number of capture occasions in the experiments. It is\n deduced from the data set in the closedpMS.t function. It must be an integer\n between 1 and 9 inclusively.\nDetails\n HIERARCHICAL LOGLINEAR MODEL NAME SYNTAX\n First, a model’s term is written using numbers between 1 and 9 to represent the capture occasions it\n includes (ex.: 134 represents the three-way interaction c1:c3:c4). This syntax limits the maximal\n number of capture occation to 9. This is not a problem since from 6 capture occasions upwards, the\n number of hierarchical models becomes very large and difficult to manage.\n A hierarchical model name is a list of the model’s terms at the top of the hierarchies in the model.\n These terms are separated by commas, without spaces. They are surronded by brackets. For exam-\n ple, \"[123,34,5]\" is the name of the model\n ~ 1 + c1 + c2 + c3 + c4 + c5 + c1:c2 + c1:c3 + c2:c3 + c3:c4 + c1:c2:c3.\nValue\n getAllModels returns a caracter vector with the models names.\n closedpMS.t returns a list with the following elements:\n n The number of captured units\n t The total number of capture occasions in the data matrix X. When captures oc-\n cur in continuous time (input argument t=Inf), this output value is the larger\n number of captures observed.\n results A table containing, for every fitted model:\n abundance the estimated population size,\n stderr the standard error of the estimated population size,\n deviance the model’s deviance,\n df the number of degrees of freedom,\n AIC the Akaike’s information criterion,\n BIC the bayesian information criterion,\n infoFit a numerical code giving information about error or warnings encoun-\n tered when fitting the model (see Rcapture-package for details).\n bias A vector, the asymptotic bias of the estimated population size for every fitted\n model.\n fit.err A list of character string vectors. If an error occurs while fitting a model (with\n glm or optim), a copy of the error message is stored in fit.err$mname where\n mname is the name of the hierarchical loglinear model. A NULL list element\n means that no error occured for the considered model.\n fit.warn A list of character string vectors. If warnings are generated while fitting a model\n (with glm or optim), a copy of these warnings are stored in fit.warn$mname\n where mname is a the name of the hierarchical model. A NULL list element means\n that no warnings were generated for the considered model.\n neg.eta For Chao’s lower bound models only: the position of the eta parameters set to\n zero in the loglinear parameter vector, if any.\nNote\n For normal heterogeneous models, closedpMS.t uses optim from the stats package. Otherwise,\n models are fitted with glm, also from the stats package.\nAuthor(s)\n <NAME> <<EMAIL>> and <NAME>\nReferences\n <NAME>. and <NAME>. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\n <NAME>. (1987) Estimating the population size for capture-recapture data with unequal catchability.\n Biometrics, 43(4), 783–791.\n <NAME>. (1992) Interval estimation for mark-recapture studies of closed populations. Bio-\n metrics, 48, 567–576.\n <NAME>. (2011) A lower bound model for multiple record systems estimation with heterogeneous\n catchability. The International Journal of Biostatistics, 7(1), Article 23.\n <NAME>. and <NAME>. (2007) Applications and extensions of Chao’s moment estimator for\n the size of a closed population. Biometrics, 63(4), 999–1006.\nSee Also\n closedp, closedp.Mtb\nExamples\n # The lesbian data set contains 4 capture occasions.\n # By default, closedpMB.t fits the 113 following models:\n getAllModels(4)\n closedpMS.t(lesbian, dfreq = TRUE)\n # We could reduce the number of models by omitting\n # those with triple interactions.\n closedpMS.t(lesbian, dfreq = TRUE, maxorder = 2)\n # Models with heterogeneity fits better.\n Darr <- closedpMS.t(lesbian, dfreq = TRUE, h = \"Darroch\")\n Darr\n # The plot method allows the visualization of the results\n # from models fitted by closedpMS.t().\n plot(Darr)\n # According to the BIC, the best heterogeneous Darroch model\n # for this data set contains the double interactions 12, 13, 14.\n # Here is the profile likelihood confidence interval for the\n # abundance estimation from this model.\n closedpCI.t(lesbian, dfreq = TRUE, mX = \"[12,13,14]\", h = \"Darroch\")\n descriptive Descriptive Statistics\nDescription\n This function produces descriptive statistics for capture-recapture data.\nUsage\n descriptive(X, dfreq=FALSE, dtype=c(\"hist\",\"nbcap\"), t=NULL)\n ## S3 method for class 'descriptive'\n print(x, ...)\n ## S3 method for class 'descriptive'\n plot(x, main=\"Exploratory Heterogeneity Graph\", ...)\nArguments\n X The matrix of the observed capture histories (see Rcapture-package for a de-\n scription of the accepted formats).\n dfreq A logical. By default FALSE, which means that X has one row per unit. If\n TRUE, it indicates that the matrix X contains frequencies in its last column.\n dtype A characters string, either \"hist\" or \"nbcap\", to specify the type of data. \"hist\",\n the default, means that X contains complete observed capture histories. \"nbcap\"\n means that X contains numbers of captures (see Rcapture-package for details\n on data formats).\n t Requested only if dtype=\"nbcap\". A numeric specifying the total number of\n capture occasions in the experiment. The value t=Inf is accepted. It indicates\n that captures occur in continuous time. (see Rcapture-package).\n x An object, produced by the descriptive function, to print or to plot.\n main A main title for the plot\n ... Further arguments to be passed to methods (see print.default and plot.default).\nDetails\n The plot.descriptive function produces an exploratory heterogeneity graph. In the absence of\n heterogeneity, the relation(s) presented in the graph should be almost linear. Convex functions\n indicate heterogeneity (Baillargeon and Rivest (2007)).\nValue\n n The total number of captured units.\n base.freq A table of basic descriptive statistics. For i = 1, . . . , t :\n fi: the number of units captured i times\n ui: the number of units captured for the first time on occasion i\n vi: the number of units captured for the last time on occasion i\n ni: the number of units captured on occasion i\n If dtype=\"nbcap\", only the the number of units captured i times, fi, can be\n computed.\n m.array Only if dtype=\"hist\": A t by t matrix containing recapture statistics for units\n released on occasion i, for i = 1, . . . , t :\n ni: the number of units released on occasion i;\n c2: the number of units released on occasion i and recaptured for the first time\n on occasion 2;\n ...\n ct: the number of units released on occasion i and recaptured for the first time\n on occasion t;\n not recapt: the number of units released on occasion i and never recaptured\n again.\n The m.array matrix is used in Mark (see White and Burnham 1999) and MSurge\n (see Choquet, Reboulet, Pradel, Gimenez, and Lebreton 2004) to fit open popu-\n lation models to the data.\n call The function call (object of class \"call\").\nAuthor(s)\n <NAME> <<EMAIL>> and <NAME>\nReferences\n <NAME>. and <NAME>. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\n <NAME>., <NAME>., <NAME>., <NAME>. and <NAME>. (2004). M-Surge: New\n Software Specifically Designed for Multistate Capture-Recapture Models. Animal Biodiversity and\n Conservation, 27.1, 207–215.\n <NAME>. (1980) A Poissonness Plot, The American Statistician, 34, 146–149\n <NAME>. (1986) Exponential family mixture models (with least-squares estimators). Annals\n of Statistics, 14, 124–137.\n <NAME>. (2008) Why a time effect often has a limited impact on capture-recapture estimates in\n closed populations. Canadian Journal of Statistics, 36(1), 75–84.\n <NAME>. and <NAME>. (1999) Program Mark: Survival Estimation from Populations of\n Marked Animals. Bird Study, 46 (Supplement), 120–138. The software can be downloaded here:\n http://www.phidot.org/software/mark/.\nExamples\n # hare data set\n hare.desc <- descriptive(hare)\n hare.desc\n plot(hare.desc)\n # cbird data set\n cbird.desc1 <- descriptive(cbird, dfreq = TRUE, dtype = \"nbcap\", t = 11)\n plot(cbird.desc1)\n # To illustrate the option t = Inf.\n cbird.desc2 <- descriptive(cbird, dfreq = TRUE, dtype = \"nbcap\", t = Inf)\n plot(cbird.desc2)\n # The y coordinate has changed.\n duck Eider Duck Data\nDescription\n This data set contains capture-recapture data for eider ducks.\nUsage\n duck\nFormat\n 63 by 7 numeric matrix, with the following columns:\n p1, p2, p3, p4, p5, p6 Capture histories from six periods\n freq Observed frequencies for each capture history\nDetails\n The data points are extracted from a 25-year study by Coulson (1984). The capture periods are six\n consecutive years : years 19-24. This data set is analyzed in Cormack (1989).\n Each row of this data set represents an observed capture history followed by its frequency.\nSource\n <NAME>. (1984) The population dynamics of the Eider Duck Somateria mollissima and evi-\n dence of extensive non breeding by adults ducks. Ibis, 126, 525–543.\nReferences\n <NAME>. and <NAME>. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\n <NAME>. (1989) Loglinear models for capture-recapture. Biometrics, 45, 395–413.\nExamples\n op.m1 <- openp(duck, dfreq = TRUE)\n op.m1$model.fit\n # The pvalue of the goodness of fit test based on the deviance is\n 1 - pchisq(op.m1$model.fit[1, 1], df = 49)\n plot(op.m1)\n # The residual plot shows a large residual for the 13 individuals\n # captured all the times. We redo the analysis without them.\n keep2 <- rowSums(histpos.t(6)) != 6\n op.m2 <- openp(duck, dfreq = TRUE, keep = keep2)\n op.m2$model.fit\n 1 - pchisq(op.m2$model.fit[1, 1],df = 48)\n # The fit is still not satisfactory.\n plot(op.m2)\n # The residual plot has the convex shape characteristic of\n # heterogeneity in the capture probabilities. We also remove the\n # individuals caught at 5 periods out of 6.\n keep3 <- rowSums(histpos.t(6)) < 5\n op.m3 <- openp(duck, dfreq = TRUE, keep = keep3)\n op.m3$model.fit\n 1 - pchisq(op.m3$model.fit[1, 1], df = 42)\n # The fit is better but there is still heterogeneity in the data.\n # To investigate whether the capture probabilities are homogeneous,\n # one can fit a model with equal capture probabilities.\n op.m4 <- openp(duck, dfreq = TRUE, m = \"ep\", keep = keep3)\n op.m4$model.fit\n # It gives a much larger deviance; this model is not considered anymore.\n # We now investigate models for the growth rate N[i+1]/N[i] of this\n # population using the multivariate normal distribution for the\n # abundance estimates. The growth rates and their standard errors are\n growth <- op.m3$N[3:5]/op.m3$N[2:4]\n partial <- matrix(c(-op.m3$N[3]/op.m3$N[2]^2, 1/op.m3$N[2], 0, 0,\n 0, -op.m3$N[4]/op.m3$N[3]^2, 1/op.m3$N[3], 0,\n 0, 0, -op.m3$N[5]/op.m3$N[4]^2, 1/op.m3$N[4]), 3, 4, byrow = TRUE)\n sig <- partial %*% op.m3$cov[9:12, 9:12] %*% t(partial)\n cbind(estimate = growth, stderr = sqrt(diag(sig)))\n # An estimate for the common growth rate is\n siginv <- solve(sig)\n growth.e <- t(rep(1, 3)) %*% siginv %*% growth/(t(rep(1, 3)) %*% siginv %*% rep(1, 3))\n se <- 1/sqrt(t(rep(1, 3)) %*% siginv %*% rep(1, 3))\n cbind(estimate = growth.e, stderr = se)\n # A chi-square statistic for testing the equality of the growth rates\n # and its pvalue\n chisq2 <- t(growth - as.vector(growth.e)) %*% siginv %*% (growth - as.vector(growth.e))\n c(chi2stat = chisq2, pvalue = 1 - pchisq(chisq2, df = 2))\n # The hypothesis of a common growth rate is rejected\n hare Snowshoe Hare Data\nDescription\n This data set contains capture-recapture data for snowshoe hares.\nUsage\n hare\nFormat\n 68 by 6 numeric matrix, with the following columns:\n c1, c2, c3, c4, c5, c6 Capture histories from the six capture occasions\nDetails\n This data set is analyzed in Cormack (1989) and Agresti (1994).\n Each row of this data set represents the capture history of one animal.\nReferences\n Agresti, A. (1994) Simple capture-recapture models permitting unequal catchability and variable\n sampling effort. Biometrics, 50, 494–500.\n <NAME>. and <NAME>. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\n <NAME>. (1989) Loglinear models for capture-recapture. Biometrics, 45, 395–413.\nExamples\n desc <- descriptive(hare)\n plot(desc)\n # The fi plot shows that the two animals caught on all occasions create\n # some heterogeneity in the capture probabilities.\n closedp(hare)\n # The best fitting model Mth Poisson2(N = 81.1, s.e. = 5.7) has an AIC of 146.\n closedpCI.t(hare, m = \"Mth\", h = \"Poisson\", h.control = list(theta = 2))\n # One can compare the fit of this model with that obtained by removing the\n # 2 hares caught 6 times. This can be done by adding a column to the design\n # matrix for Mt taking the value 1 for the capture history (1,1,1,1,1,1).\n col <- rep(0, 2^6-1)\n mat <- histpos.t(6)\n col[rowSums(mat) == 6] <- 1\n closedpCI.t(hare, mX = cbind(mat, col), mname = \"Mt without 111111\")\n # This gives N = 76.8 (s.e. = 3.9) with an AIC of 146.\n histpos Possible Capture Histories\nDescription\n histpos.t builds the matrix of observable capture histories in terms of captures and misses for a\n capture-recapture experiment with t capture occasions.\n histpos.0 builds the matrix of observable capture histories in terms of number of captures for\n each primary period of a robust design with vt capture occasions. For closed populations, vt=t and\n histpos.0 simply returns t:1.\nUsage\n histpos.t(t)\n histpos.0(vt)\nArguments\n t The number of capture occasions.\n vt A vector containing the numbers of capture occasions for each primary sampling\n period of a robust design. The length of this vector equals the number of primary\n sampling periods (noted I).\nValue\n histpos.t gives a 2t − 1 by t matrix with rows representing capture histories. This matrix contains\n only zeros and ones.\n histpos.0 gives a (t1 +1)×(t2 +1)×. . .×(tI +1)−1 by I matrix with rows representing possible\n capture histories in terms of number of captures. This matrix elements are integers between 0 and\n ti inclusively.\nNote\n The histpos.t function is called by descriptive, closedp, closedp.bc, closedp.Mtb, openp,\n robustd.t.\n The histpos.0 function is called by robustd.0.\nAuthor(s)\n <NAME> <<EMAIL>> and <NAME>\nReferences\n <NAME>. and Rivest, L.P. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\nSee Also\n periodhist, descriptive\nExamples\n histpos.t(5)\n histpos.0(5)\n histpos.0(rep(5, 3))\n HIV Epidemiological capture-recapture Data on HIV\nDescription\n Epidemiological capture-recapture data on HIV from four reporting centers in Rome, Italy.\nUsage\n HIV\nFormat\n 15 by 5 numeric matrix, with the following columns:\n c1, c2, c3, c4 Capture histories from the four capture occasions\n freq Observed frequencies for each capture history\nDetails\n The capture histories were obtained by linking the records of the four reporting centers.\n Each row of this data set represents an observed capture history followed by its frequency.\nSource\n <NAME>., <NAME>. and <NAME>. (1994) Capture-recapture to estimate the size of the\n population with human immunodeficiency virus type 1 infection. Epidemiology, 5, 410–414.\nReferences\n Baillargeon, S. and Rivest, L.P. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\nExamples\n desc <- descriptive(HIV, dfreq=TRUE)\n desc\n # 1774 out of 1896 individuals (94%) appear on one list only.\n plot(desc)\n # The fi plot is linear showing that heterogeneity is not a problem.\n # Models with a time (or list) effect and possible pairwise dependencies\n # between lists will be considered.\n cp.m1 <- closedpCI.t(HIV, dfreq = TRUE, mX = ~ (c1+c2+c3+c4)^2,\n mname = \"Mt double interaction\")\n cp.m1\n # The model fits well. Let's find out which interactions are important.\n summary(cp.m1$fit)$coefficients\n # Eliminating the non significant interactions stepwise shows that only\n # the [1,2] interaction is important.\n closedpCI.t(HIV, dfreq = TRUE, mX = ~ . + c1:c2, mname = \"Mt interaction 1,2\")\n ill Illegal immigrants in the Netherlands\nDescription\n This data set contains the frequency fi that an illegal immigrant who is not effectively expelled is\n apprehended exactly i times.\nUsage\n ill\nFormat\n 6 by 2 numeric matrix, with the following columns:\n nbcap Numbers of captures, i.e. number of times an illegal immigrant is apprehended by the police\n freq Observed frequencies for each number of captures\nDetails\n Bohning and Schon (2005) presented this data set as follows:\n The illegal immigrant data are from van der Heijden et al. (2003) and refer to count data on illegal\n immigrants in four large cities in the Netherlands who could not be effectively expelled from the\n country. Illegal immigrants who are apprehended by the police often cannot be effectively expelled\n because either they refuse to disclose their nationality or their home country does not cooperate in\n receiving them back. In such cases they will be asked by the police to leave the country, but it is\n unlikely that they will abide by this request. The data were collected by the police and date back to\n the year 1995.\nSource\n <NAME>, P.G.M., <NAME>. and <NAME>. (2003) Estimating the size of a\n criminal population from police records using the truncated Poisson regression model. Statistica\n Neerlandica, 57(3), 289-304.\nReferences\n <NAME>. and <NAME>. (2005) Nonparametric Maximum Likelihood Estimation of Population\n Size Based on the Counting Distribution. Journal of the Royal Statistical Society: Series C (Applied\n Statistics), 54(4), 721-737.\nExamples\n sdesc <- descriptive(ill, dtype = \"nbcap\", dfreq = TRUE, t = Inf)\n plot(sdesc)\n # A mixture model looks appropriate\n closedp.0(ill, dtype = \"nbcap\", dfreq = TRUE, t = Inf)\n # We can try to fit a normal mixture model:\n closedpCI.0(ill, dtype = \"nbcap\", dfreq = TRUE, t = Inf, m = \"Mh\", h = \"Normal\")\n # We get an estimate similar to the Mh Gamma3.5 estimate.\n # Estimates are highly variable and it seems difficult to come up with a\n # definitive answer. The lower bound estimate is useful in this context.\n closedpCI.0(ill, dtype = \"nbcap\", dfreq = TRUE, t = Inf, m = \"Mh\", h = \"LB\")\n # Considering the lower limit of a 95% confidence estimate for the lower bound,\n # there should be at least 8 000 illegal immigrants in the Netherlands.\n # Less than 25% have been caught.\n lesbian Epidemiological capture-recapture Data on the lesbian population\nDescription\n Epidemiological capture-recapture data on the lesbian population from four four organisations that\n serve the lesbian and gay population of Allegheny County, Pennsylvania, United States and maintain\n large mailing lists.\nUsage\n lesbian\nFormat\n 15 by 5 numeric matrix, with the following columns:\n A, B, C, D Capture histories from the four organisations mailling lists\n freq Observed frequencies for each capture history\nDetails\n The capture histories were obtained by matching the names on the four mailling lists.\n Each row of this data set represents an observed capture history followed by its frequency.\nSource\n <NAME>., <NAME>., <NAME>., <NAME>. (2003) Estimating the lesbian population:\n a capture-recapture approach. Journal of Epidemiology and Community Health , 57(3), 207–209.\nExamples\n closedp(lesbian, dfreq = TRUE)\n # According to the BIC, the best model is Mth Darroch.\n # Let's see if adding interactions between capture\n # histories to the model could improve the model's fit.\n closedpMS.t(lesbian, dfreq = TRUE, h = \"Darroch\")\n # According to the BIC, the best heterogeneous Darroch model\n # contains the double interactions 12, 13, 14.\n mvole Robust Design Data for Adult Male Meadow Voles\nDescription\n This data set contains robust design capture history data for adult male meadow voles.\nUsage\n mvole\nFormat\n 171 by 30 numeric matrix, with the following columns:\n c11, c12, c13, c14, c15 Capture histories from the 5 capture occasions within primary period 1\n c21, c22, c23, c24, c25 Capture histories from the 5 capture occasions within primary period 2\n c31, c32, c33, c34, c35 Capture histories from the 5 capture occasions within primary period 3\n c41, c42, c43, c44, c45 Capture histories from the 5 capture occasions within primary period 4\n c51, c52, c53, c54, c55 Capture histories from the 5 capture occasions within primary period 5\n c61, c62, c63, c64, c65 Capture histories from the 5 capture occasions within primary period 6\nDetails\n The data set is extracted from Table 19.1 of Williams, Nichols and Conroy (2002). The capture\n occasions represent five consecutive days of trapping each month from June to December 1981 at\n Patuxent Wildlife Research Center, Laurel, Maryland.\n Each row of this data set represents the capture history of one animal.\nNote\n In this data set, ten animals are in fact not released after capture. These trap deaths are not identified.\nSource\n <NAME>., <NAME>., and <NAME>. (2002) Analysis and Management of Animal Pop-\n ulations, San Diego: Academic Press.\nReferences\n <NAME>. and <NAME>. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\nExamples\n # This example deals only with the first four primary periods of the data set.\n mvole4 <- mvole[, 1:20]\n # First, a between primary period Jolly-Seber analysis is obtained.\n mvole4.pp <- periodhist(mvole4, vt = rep(5,4))\n op.m1 <- openp(mvole4.pp, dfreq = TRUE)\n plot(op.m1)\n # There is one large residual, removing the corresponding capture history\n # from the analysis does not change the results. The model fits well.\n keep2 <- residuals(op.m1$glm, type = \"pearson\") < 4\n op.m2 <- openp(mvole4.pp, dfreq = TRUE, keep = keep2)\n op.m2$model.fit\n # To find a suitable model within each primary period, the function closedp.t\n # can be used repeatedly. Heterogeneity is detected in all periods except\n # the second one where the data collection was perturbed (the last capture\n # occasion doesn't have any new capture and is taken out of the analysis).\n # In a robust design, we use Mh models for all primary periods bearing in\n # mind the questionable fit in the second one. Since there is no time effect\n # within primary periods, we use the function robustd.0 to fit the model.\n ### The following command might take a few minutes to run.\n rd.m1 <- robustd.0(mvole4[, -10], vt = c(5, 4, 5, 5), vm = \"Mh\", vh = \"Chao\")\n rd.m1$model.fit\n rd.m1$emig.fit\n # The test for temporary immigration is not significant meaning that capture\n # probabilities estimated with the Jolly-Seber model are not different from\n # those estimated with the individual closed population models. The\n # differences, on the logit scale, of the Jolly-Seber minus the closed\n # population models capture probabilities are\n rd.m1$emig.param\n # Even in period 2, where the closed population model does not fit well, the\n # difference on the logit scale is non significant (estimate=.56, s.e.=1.13).\n # The following command allows to fit a robust design that does not specify\n # any model for the second period.\n ### The following command might take a few minutes to run.\n rd.m3 <- robustd.0(mvole4[, -10], vt = c(5, 4, 5, 5),\n vm = c(\"Mh\", \"none\", \"Mh\", \"Mh\"), vh = \"Chao\")\n # With Darroch's model, the closed population estimates of the capture\n # probabilities are significantly smaller than those obtained from the\n # Jolly-Seber model. This cannot be interpreted as indicating temporary\n # immigration. This suggests that Darroch's model is not appropriate within\n # primary sessions.\n # The smallest AIC is obtained with the Poisson model, with parameter a=1.5\n # within sessions.\n rd.m4 <- robustd.0(mvole4[, -10], vt = c(5, 4, 5, 5), vm = \"Mh\",\n vh = \"Poisson\", vtheta = 1.5)\n # The estimators of the demographic parameters obtained with the robust design\n # are similar to those obtained with the Jolly-Seber model applied to the\n # between primary period data.\n cbind(op.m1$survivals, rd.m4$survivals)\n cbind(op.m1$N, rd.m4$N)\n cbind(op.m1$birth, rd.m4$birth)\n cbind(op.m1$Ntot, rd.m4$Ntot)\n openp Loglinear Models for Open Population Capture-Recapture Experi-\n ments\nDescription\n This function computes various demographic parameters using a loglinear model for open popula-\n tions in capture-recapture experiments.\nUsage\n openp(X, dfreq=FALSE, m=c(\"up\",\"ep\"), neg=TRUE, keep=rep(TRUE,2^I-1))\n ## S3 method for class 'openp'\n print(x, ...)\n ## S3 method for class 'openp'\n plot(x, main=\"Scatterplot of Pearson Residuals\", ...)\nArguments\n X The matrix of the observed capture histories (see Rcapture-package for a de-\n scription of the accepted formats).\n dfreq A logical. By default FALSE, which means that X has one row per unit. If\n TRUE, it indicates that the matrix X contains frequencies in its last column.\n m This argument is a character string taking the value \"up\" (up = unconstrained\n probabilities) or \"ep\" (ep = equal probabilities). If m is set to \"up\" (the default),\n no constraint is set on the loglinear parameters. Therefore some loglinear and\n demographic parameters are not estimable. On the other hand, when m takes the\n value \"ep\", the capture probabilities are set to a common value; this enables the\n estimation of all the parameters.\n keep This option is useful to fit the model on a subset of the possible capture histories.\n keep is a logical vector of length 2I − 1 taking the value TRUE for a history\n kept and FALSE for a history put aside. In this vector, the order of the capture\n histories is as defined in the histpos.t function. By default, every capture\n history is kept.\n neg If this option is set to TRUE, relevant negative gamma parameters are set to zero.\n This insures that the estimated survival probabilities belong to [0, 1] and that the\n births are positive.\n x An object, produced by the openp function, to print or to plot.\n main A main title for the plot\n ... Further arguments to be passed to methods (see print.default and plot.default).\nDetails\n The function openp generates statistics to test the presence of a trap effect.\n The plot.openp function produces a scatterplot of the Pearson residuals of the model versus the\n frequencies of capture.\n If the data matrix X was obtained through the periodhist function, the dfreq argument must be\n set to TRUE.\n Standard errors are calculated by linearization.\nValue\n n The number of captured units\n model.fit A table containing the deviance, degrees of freedom and AIC of the fitted model.\n trap.fit A table containing, for the models with an added trap effect, the deviance, de-\n grees of freedom and AIC.\n trap.param The estimated trap effect parameters and their standard errors. For m=\"up\", the\n I − 3 first rows of trap.param are estimations of the differences logit(capture\n probability after a capture)- logit(capture probability after a miss) for periods 3\n to I − 1. The last row gives a pooled estimate of these differences calculated\n under the assumption that they are homogeneous.\n capture.prob The estimated capture probabilities per period and their standard errors.\n survivals The estimated survival probabilities between periods and their standard errors.\n N The estimated population sizes per period and their standard errors.\n birth The estimated number of new arrivals in the population between periods and\n their standard errors.\n Ntot The estimated total number of units who ever inhabited the survey area and its\n standard error.\n glm The ’glm’ object obtained from fitting the loglinear model\n loglin.param The loglinear model parameters estimations and their standard errors, calculated\n by the glm function.\n u.vector The Ui statistics, useful for the survival probabilities calculation, and their stan-\n dard errors\n v.vector The Vi statistics, useful for the population sizes estimation, and their standard\n errors\n cov The covariance matrix of all the demographic parameters estimates.\n neg The position of the gamma parameters set to zero in the loglinear parameter\n vector.\nNote\n If your data contains more than one capture occasion within primary periods, use the periodhist\n function to create the input data matrix X needed by the openp function.\n This function uses the glm function of the stats package.\nAuthor(s)\n <NAME> <<EMAIL>> and <NAME>\nReferences\n <NAME> <NAME>. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\n <NAME>. and <NAME>. (2004) Loglinear models for the robust design in mark-recapture exper-\n iments. Biometrics, 60, 100–107.\nSee Also\n closedp, periodhist, robustd\nExamples\n # duck data set\n op.m1 <- openp(duck, dfreq = TRUE)\n plot(op.m1)\n # To remove the capture history 111111.\n keep2 <- apply(histpos.t(6), 1, sum) != 6\n op.m2 <- openp(duck, dfreq = TRUE, keep = keep2)\n op.m2\n # To remove the capture histories with 5 captures or more\n keep3 <- apply(histpos.t(6), 1, sum) < 5\n op.m3 <- openp(duck, dfreq = TRUE, keep = keep3)\n op.m3\n # mvole data set aggregated per primary period\n mvole.op <- periodhist(mvole, vt = rep(5, 6))\n openp(mvole.op, dfreq = TRUE)\n periodhist Merge capture occasions\nDescription\n This function produces a reduced matrix of capture histories from a complete one by merging\n together some capture occasions.\n It can also be used to change the format of a capture-recapture data set with complete capture\n histories: it transforms a data set with one row per captured unit to a data set with one row per\n capture history followed by its frequency.\nUsage\n periodhist(X, dfreq=FALSE, vt, drop=TRUE)\nArguments\n X The matrix of the observed capture histories (see Rcapture-package for a de-\n scription of the accepted formats).\n dfreq A logical. By default FALSE, which means that X has one row per unit. If\n TRUE, it indicates that the matrix X contains frequencies in its last column.\n vt A vector containing the numbers of capture occasions for each pooled capture\n occasions. The length of this vector equals the number of capture occasions in\n the reduced matrix (noted I).\n drop A logical, by default TRUE, meaning that in the output matrix the unobserved\n capture histories (frequency of 0) are omitted. To keep them, drop must be set\n to FALSE.\nDetails\n This function is useful when using an open population model to analyze a robust design data set. It\n can be used to reduce the data set to one observation per primary period. A one represents a unit\n caught at least once during the period and a zero a unit never caught.\n It is also useful for experiments with a large number of capture occasions but a limited number of\n catches, especially when there is no capture on some occasions. In such cases, one can pool together\n some capture occasions with the periodhist function.\n If a data matrix produced by the periodhist function is given as an argument to a Rcapture\n function, the dfreq argument must be set to TRUE.\nValue\n A 2I − 1 by I + 1 matrix of all the pooled capture histories, with their observed frequencies. If\n drop=TRUE (the default), the number of rows of this matrix is in fact 2I − 1 minus the number of\n unobserved capture histories.\nAuthor(s)\n <NAME> <<EMAIL>> and <NAME>\nReferences\n Baillargeon, S. and Rivest, L.P. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\nSee Also\n openp\nExamples\n # mvole data set aggregated per primary period\n mvole.op <- periodhist(mvole, vt = rep(5, 6))\n openp(mvole.op, dfreq = TRUE)\n profileCI Profile Likelihood Confidence Interval for Abundance Estimation in\n Closed Populations Capture-Recapture Experiments\nDescription\n As of version 1.2-0 of Rcapture, this function is deprecated, but kept for back compatibility. Please\n use closedpCI.t instead.\n The profileCI function computes the multinomial profile likelihood for the abundance of some\n closed population capture-recapture models.\nUsage\n profileCI(X, dfreq=FALSE, m=\"M0\", h=\"Chao\", a=2, mX=NULL,\n mname=\"Customized model\", neg=TRUE, alpha=0.05)\n ## S3 method for class 'profileCI'\n print(x, ...)\nArguments\n X The table of the observed capture histories in one of the two accepted formats.\n In the default format, it has one row per unit captured in the experiment. In this\n case, the number of columns in the table represents the number of capture occa-\n sions in the experiment (noted t). In the alternative format, it contains one row\n per capture history followed by its frequency. In that case, X has t+1 columns.\n The first t columns of X, identifying the capture histories, must contain only\n zeros and ones. The number one indicates a capture.\n dfreq This argument specifies the format of the data matrix X. By default, it is set to\n \"FALSE\", which means that X has one row per unit. If it is set to TRUE, then\n the matrix X contains frequencies in its last column.\n m A character string identifying the model, either \"M0\"=M0 model, \"Mt\"=Mt\n model, \"Mh\"=Mh model or \"Mth\"=Mth model.\n h A character string (\"Chao\" = Chao model, \"Poisson\" = 2k − 1), \"Darroch\" =\n (k 2 )/2 or a numerical function taking only one argument specifying the form of\n the column for heterogeneity in the design matrix.\n a The value of the exponent’s base for a Poisson model.\n mX The design matrix of the loglinear model. In this matrix, the order of the capture\n histories is as defined in the histpos.t function.\n mname A character string specifying the name of the customized model.\n neg If this option is set to TRUE, negative eta parameters in Chao models are set to\n zero.\n alpha A confidence interval with confidence level 1-alpha is constructed. The value of\n alpha must be between 0 and 1; the default is 0.05.\n x An object, produced by the profileCI function, to print.\n ... Further arguments passed to or from other methods.\nDetails\n This function does not work for closed population models featuring a behavioral effect, such as Mb\n and Mbh.\nValue\n This function produces a plot of the multinomial profile likelihood for N. The value of N maximiz-\n ing the profile likelihood and the bounds of the confidence interval are identified. It also produces\n the following objects :\n n The number of captured units\n results A table containing the abundance estimation and its confidence interval.\n alpha 1-the confidence level of the interval.\nAuthor(s)\n <NAME> <<EMAIL>> and <NAME>\nReferences\n Baillargeon, S. and Rivest, L.P. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\n <NAME>. (1992) Interval estimation for mark-recapture studies of closed populations. Bio-\n metrics, 48, 567–576.\nSee Also\n closedp, closedp.mX, closedp.h\nExamples\n # hare data set\n profileCI(hare, m = \"Mth\", h = \"Poisson\", a = 2)\n # HIV data set\n mat <- histpos.t(4)\n mX2 <- cbind(mat, mat[, 1] * mat[, 2])\n profileCI(hare, m = \"Mh\", h = \"Chao\")\n rbvole Robust Design Data for Red-Back Voles\nDescription\n This data set contains robust design capture history data for red-back voles.\nUsage\n rbvole\nFormat\n 66 by 19 numeric matrix, with the following columns:\n c11, c12, c13 Capture histories from the three capture occasions within primary period 1\n c21, c22, c23 Capture histories from the three capture occasions within primary period 2\n c31, c32, c33 Capture histories from the three capture occasions within primary period 3\n c41, c42, c43 Capture histories from the three capture occasions within primary period 4\n c51, c52, c53 Capture histories from the three capture occasions within primary period 5\n c61, c62, c63 Capture histories from the three capture occasions within primary period 6\n freq Observed frequencies for each capture history\nDetails\n Data collection was carried out by Etcheverry and al.. The capture occasions represent three con-\n secutive days of trapping in May 1999, July 1999, August 1999, May 2000, July 2000 and August\n 2000 in the Duchenier conservation area in southeastern Quebec. This data set is analyzed in Rivest\n and Daigle (2004).\n Each row of this data set represents an observed capture history followed by its frequency.\nReferences\n <NAME>. and <NAME>. (2004) Loglinear models for the robust design in mark-recapture exper-\n iments. Biometrics, 60, 100–107.\nExamples\n # According to Rivest and Daigle (2004), a good robust design model\n # for this data set is formed of an Mth Chao model for each period.\n # This model can be fitted as follows.\n ### memory.limit(size = 2000)\n ### rd <- robustd.t(rbvole, dfreq = TRUE, vt = rep(3, 6), vm = \"Mth\", vh = \"Chao\")\n # WARNING : Because the data has 18 capture occasions, fitting this\n # model uses a lot of memory. Its runtime is several minutes long.\n robustd Loglinear Models for Capture-Recapture Experiments Carried Out\n According to a Robust Design\nDescription\n These functions compute various demographic parameters and capture probabilities per period us-\n ing loglinear robust design models in capture-recapture experiments.\n robustd.t and robustd.0 fit the model using different response variable. robustd.t uses the\n frequencies of the observable capture histories in terms of capture success or failure for each cap-\n ture occasions of each primary period. robustd.0 uses the frequencies of the observable capture\n histories in terms of number of captures per primary period.\nUsage\n robustd.t(X, dfreq = FALSE, vt, vm = \"M0\", vh = list(\"Chao\"),\n vtheta = 2, neg = TRUE)\n robustd.0(X, dfreq = FALSE, vt, vm = \"M0\", vh = list(\"Chao\"),\n vtheta = 2, neg = TRUE)\n ## S3 method for class 'robustd'\n print(x, ...)\nArguments\n X The matrix of the observed capture histories (see Rcapture-package for a de-\n scription of the accepted formats).\n dfreq A logical. By default FALSE, which means that X has one row per unit. If\n TRUE, it indicates that the matrix X contains frequencies in its last column.\n vt A vector containing the numbers of capture occasions for each primary sampling\n period. The length of this vector equals the number of primary sampling periods\n (noted I).\n vm A vector indicating the closed population model for each primary period. The\n elements of vm can be \"none\"=no model, \"M0\"=M0 model, \"Mt\"=Mt model,\n \"Mh\"=Mh model or \"Mth\"=Mth model for robustd.t. For robustd.0, models\n with time effect are not allowed. So vm can than be \"none\"=no model, \"M0\"=M0\n model, or \"Mh\"=Mh model. The ’no model’ cannot be selected for the first or\n the last period. If a single character string is given for vm, the corresponding\n model is used for all periods. The default value for vm is \"M0\" for all period.\n vh A list indicating, for each primary period with a heterogeneity model, the form\n of the columns for heterogeneity in the design matrix. The elements of vh can\n be \"Chao\", \"Poisson\", \"Darroch\" or any numerical R function created before-\n hand by the user. \"Chao\" represents Chao’s model, \"Poisson\" represents the\n function f (k) = thetak − 1, where k is the number of captures, \"Darroch\"\n represents the function f (k) = k 2 /2, and \"Gamma\" represents the function\n f (k) = − log(theta + k) + log(theta). If an R function is given, it is the\n implementation of any convex mathematical function f (k). It has only one ar-\n gument. For the Poisson or Gamma model, the parameter theta is specified in\n the argument vtheta.\n vtheta A vector indicating, for each primary period with a Poisson or Gamma hetero-\n geneity model, the value of the parameter. If vtheta consists of a single number,\n this number is used for all the periods with a Poisson or Gamma model. The de-\n fault value for vtheta is 2 for all the periods with a Poisson or Gamma model.\n neg If this option is set to TRUE, negative gamma parameters and negative eta pa-\n rameters in Chao’s models are set to zero. This insures that the estimated sur-\n vival probabilities belong to [0, 1] and that the births are positive.\n x An object, produced by the robustd.t or the robustd.0 function, to print.\n ... Further arguments to be passed to print.default.\nDetails\n These functions also generate statistics to test the presence of temporary emigration.\n The Poisson regression used to fit a robust design model has one entry for each possible capture his-\n tory, including those that are unobserved. The size of the dependent vector is therefore 2sum(vt) −1\n for robustd.t. Models with a large sum(vt) are hard to fit with robustd.t. robustd.0 uses a more\n parsimonious coding for the capture histories and can fit larger models.\n Standard errors are calculated by linearization.\nValue\n n The number of captured units\n models A vector of length I identifying the closed population models chosen for each\n period.\n model.fit A table containing the deviance, degrees of freedom and AIC of the fitted model.\n emig.fit A table containing, for the model with an added temporary emigration effect,\n the deviance, the degrees of freedom and the Akaike’s information criterion.\n emig.param The estimated temporary emigration parameters and their standards errors. The\n I − 1 first rows are estimations of the differences logit(pb ) − logit(pw ) for peri-\n ods 2 to I − 1 (pb represents a between primary period and pw a within primary\n period estimate of the capture probability). The last row gives a pooled estimate\n of these differences calculated under the assumption that they are homogeneous.\n Negative estimates are associated with a temporary emigration.\n capture.prob The estimated capture probabilities per period and their standard errors.\n survivals The estimated survival probabilities between periods and their standard errors.\n N The estimated population sizes per period and their standard errors.\n birth The estimated number of new arrivals in the population between periods and\n their standard errors.\n Ntot The estimated total number of units who ever inhabited the survey area and its\n standard error.\n loglin.param The loglinear model parameters estimations and their standard errors, calculated\n by the glm function.\n u.vector The Ui statistics, useful for the survival probabilities calculation, and their stan-\n dard errors\n v.vector The Vi statistics, useful for the population sizes estimation, and their standard\n errors\n cov The covariance matrix of all the demographic parameters estimates.\n neg The position of the gamma and eta parameters set to zero in the loglinear pa-\n rameter vector.\nNote\n This function uses the glm function of the stats package.\nAuthor(s)\n <NAME> <<EMAIL>> and <NAME>\nReferences\n <NAME>. and <NAME>. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\n <NAME>. and <NAME>. (2004) Loglinear models for the robust design in mark-recapture exper-\n iments. Biometrics, 60, 100–107.\nSee Also\n closedp, openp\nExamples\n # The mvole data set contains a total of 30 capture occasions (the\n # tenth capture occasion doesn't have any new capture and is taken\n # out of the analysis). This number being large, we can only use\n # the robustd.0 function to fit a robust design model.\n # Not run :\n # robustd.0(mvole[, -10], vt = c(5, 4, rep(5, 4)), vm = \"Mh\", vh = \"Poisson\", vtheta = 1.5)\n # Should take a few seconds to run.\n # Not run:\n # robustd.t(mvole[, -10], vt = c(5, 4, rep(5, 4)), vm = 'Mh', vh = 'Poisson', vtheta = 1.5)\n # Should fail.\n # Considering only the first 3 periods of the data set, we can use the\n # robustd.t function to fit a model with a temporal effect.\n robustd.t(mvole[, c(1:9, 11:15)], vt = c(5, 4, 5), vm = \"Mth\", vh = \"Poisson\", vtheta = 1.5)\n uifit Closed Population Models Fit Statistics Concerning the ui\nDescription\n This function produces fit statistics concerning the ui , i.e. the numbers of first captures on each\n capture occasion, for closed population models. It also forecasts, for some models, the number of\n additional units that would be captured if the experiment was continued for five more occasions.\nUsage\n uifit(x.closedp)\nArguments\n x.closedp An object produced by the closedp.t function.\nValue\n predicted The observed and predicted values of the ui frequencies, i = 1, . . . , t, for each\n model in closedp. For models M0, Mh Poisson2, Mh Darroch and Mb, pre-\n dicted values are also extrapolated for i = t + 1, . . . , t + 5.\n fit.stat Chi-square fit statistics for each model in closedp ( i=1 predictedi ).\n day.first.capt\n The mean and variance of the day of first capture, calculated with the observed\n and predicted ui , for each model in closedp.t.\nAuthor(s)\n <NAME> <<EMAIL>> and <NAME>\nReferences\n Baillargeon, S. and Rivest, L.P. (2007) Rcapture: Loglinear models for capture-recapture in R.\n Journal of Statistical Software, 19(5), doi:10.18637/jss.v019.i05.\nSee Also\n closedp\nExamples\n # Third primary period of mvole data set\n period3 <- mvole[, 11:15]\n cp <- closedp(period3)\n uifit(cp)"}}},{"rowIdx":59189,"cells":{"project":{"kind":"string","value":"github.com/lilien1010/taskq/v3"},"source":{"kind":"string","value":"go"},"language":{"kind":"string","value":"Go"},"content":{"kind":"string","value":"README\n [¶](#section-readme)\n---\n\n### Golang asynchronous task/job queue with Redis, SQS, IronMQ, and in-memory backends\n\n\n[](https://pkg.go.dev/github.com/vmihailenco/taskq/v3)\n[](https://taskq.uptrace.dev/)\n[](https://discord.gg/rWtp5Aj)\n\nDocumentation\n [¶](#section-documentation)\n---\n\n[Rendered for](https://go.dev/about#build-context)\n\nlinux/amd64 windows/amd64 darwin/amd64 js/wasm\n\n \n### Overview [¶](#pkg-overview)\n\nPackage taskq implements task/job queue with Redis, SQS, IronMQ, and in-memory backends.\n\nExample (CustomRateLimit) [¶](#example-package-CustomRateLimit)\n```\npackage main\n\nimport (\n\t\"context\"\n\t\"fmt\"\n\t\"time\"\n\n\t\"github.com/vmihailenco/taskq/v3\"\n\t\"github.com/vmihailenco/taskq/v3/memqueue\"\n)\n\ntype RateLimitError string\n\nfunc (e RateLimitError) Error() string {\n\treturn string(e)\n}\n\nfunc (RateLimitError) Delay() time.Duration {\n\treturn 3 * time.Second\n}\n\nfunc main() {\n\tstart := time.Now()\n\tq := memqueue.NewQueue(&taskq.QueueOptions{\n\t\tName: \"test\",\n\t})\n\ttask := taskq.RegisterTask(&taskq.TaskOptions{\n\t\tName: \"Example_customRateLimit\",\n\t\tHandler: func() error {\n\t\t\tfmt.Println(\"retried in\", timeSince(start))\n\t\t\treturn RateLimitError(\"calm down\")\n\t\t},\n\t\tRetryLimit: 2,\n\t\tMinBackoff: time.Millisecond,\n\t})\n\n\tctx := context.Background()\n\tq.Add(task.WithArgs(ctx))\n\n\t// Wait for all messages to be processed.\n\t_ = q.Close()\n\n}\n```\n```\nOutput:\n\nretried in 0s retried in 3s\n```\nShare Format\nRun\n\nExample (MessageDelay) [¶](#example-package-MessageDelay)\n```\nstart := time.Now()\nq := memqueue.NewQueue(&taskq.QueueOptions{\n\tName: \"test\",\n})\ntask := taskq.RegisterTask(&taskq.TaskOptions{\n\tName: \"Example_messageDelay\",\n\tHandler: func() {\n\t\tfmt.Println(\"processed with delay\", timeSince(start))\n\t},\n})\n\nctx := context.Background()\nmsg := task.WithArgs(ctx)\nmsg.Delay = time.Second _ = q.Add(msg)\n\n// Wait for all messages to be processed.\n_ = q.Close()\n```\n```\nOutput:\n\nprocessed with delay 1s\n```\nExample (Once) [¶](#example-package-Once)\n```\nq := memqueue.NewQueue(&taskq.QueueOptions{\n\tName: \"test\",\n\tRedis: redisRing(),\n\tRateLimit: redis_rate.PerSecond(1),\n})\ntask := taskq.RegisterTask(&taskq.TaskOptions{\n\tName: \"Example_once\",\n\tHandler: func(name string) {\n\t\tfmt.Println(\"hello\", name)\n\t},\n})\n\nctx := context.Background()\nfor i := 0; i < 10; i++ {\n\tmsg := task.WithArgs(ctx, \"world\")\n\t// Call once in a second.\n\tmsg.OnceInPeriod(time.Second)\n\n\t_ = q.Add(msg)\n}\n\n// Wait for all messages to be processed.\n_ = q.Close()\n```\n```\nOutput:\n\nhello world\n```\nExample (RateLimit) [¶](#example-package-RateLimit)\n```\nstart := time.Now()\nq := memqueue.NewQueue(&taskq.QueueOptions{\n\tName: \"test\",\n\tRedis: redisRing(),\n\tRateLimit: redis_rate.PerSecond(1),\n})\ntask := taskq.RegisterTask(&taskq.TaskOptions{\n\tName: \"Example_rateLimit\",\n\tHandler: func() {},\n})\n\nconst n = 5\n\nctx := context.Background()\nfor i := 0; i < n; i++ {\n\t_ = q.Add(task.WithArgs(ctx))\n}\n\n// Wait for all messages to be processed.\n_ = q.Close()\n\nfmt.Printf(\"%d msg/s\", timeSinceCeil(start)/time.Second/n)\n```\n```\nOutput:\n\n1 msg/s\n```\nExample (RetryOnError) [¶](#example-package-RetryOnError)\n```\nstart := time.Now()\nq := memqueue.NewQueue(&taskq.QueueOptions{\n\tName: \"test\",\n})\ntask := taskq.RegisterTask(&taskq.TaskOptions{\n\tName: \"Example_retryOnError\",\n\tHandler: func() error {\n\t\tfmt.Println(\"retried in\", timeSince(start))\n\t\treturn errors.New(\"fake error\")\n\t},\n\tRetryLimit: 3,\n\tMinBackoff: time.Second,\n})\n\nctx := context.Background()\nq.Add(task.WithArgs(ctx))\n\n// Wait for all messages to be processed.\n_ = q.Close()\n```\n```\nOutput:\n\nretried in 0s retried in 1s retried in 3s\n```\n### Index [¶](#pkg-index)\n\n* [Variables](#pkg-variables)\n* [func SetLogger(logger *log.Logger)](#SetLogger)\n* [func SetUnknownTaskOptions(opt *TaskOptions)](#SetUnknownTaskOptions)\n* [func Version() string](#Version)\n* [type Consumer](#Consumer)\n* + [func NewConsumer(q Queue) *Consumer](#NewConsumer)\n\t+ [func StartConsumer(ctx context.Context, q Queue) *Consumer](#StartConsumer)\n* + [func (c *Consumer) Add(msg *Message) error](#Consumer.Add)\n\t+ [func (c *Consumer) AddHook(hook ConsumerHook)](#Consumer.AddHook)\n\t+ [func (c *Consumer) Len() int](#Consumer.Len)\n\t+ [func (c *Consumer) Options() *QueueOptions](#Consumer.Options)\n\t+ [func (c *Consumer) Process(msg *Message) error](#Consumer.Process)\n\t+ [func (c *Consumer) ProcessAll(ctx context.Context) error](#Consumer.ProcessAll)\n\t+ [func (c *Consumer) ProcessOne(ctx context.Context) error](#Consumer.ProcessOne)\n\t+ [func (c *Consumer) Purge() error](#Consumer.Purge)\n\t+ [func (c *Consumer) Put(msg *Message)](#Consumer.Put)\n\t+ [func (c *Consumer) Queue() Queue](#Consumer.Queue)\n\t+ [func (c *Consumer) Start(ctx context.Context) error](#Consumer.Start)\n\t+ [func (c *Consumer) Stats() *ConsumerStats](#Consumer.Stats)\n\t+ [func (c *Consumer) Stop() error](#Consumer.Stop)\n\t+ [func (c *Consumer) StopTimeout(timeout time.Duration) error](#Consumer.StopTimeout)\n\t+ [func (c *Consumer) String() string](#Consumer.String)\n* [type ConsumerHook](#ConsumerHook)\n* [type ConsumerStats](#ConsumerStats)\n* [type Delayer](#Delayer)\n* [type Factory](#Factory)\n* [type Handler](#Handler)\n* + [func NewHandler(fn interface{}) Handler](#NewHandler)\n* [type HandlerFunc](#HandlerFunc)\n* + [func (fn HandlerFunc) HandleMessage(msg *Message) error](#HandlerFunc.HandleMessage)\n* [type Message](#Message)\n* + [func NewMessage(ctx context.Context, args ...interface{}) *Message](#NewMessage)\n* + [func (m *Message) MarshalArgs() ([]byte, error)](#Message.MarshalArgs)\n\t+ [func (m *Message) MarshalBinary() ([]byte, error)](#Message.MarshalBinary)\n\t+ [func (m *Message) OnceInPeriod(period time.Duration, args ...interface{})](#Message.OnceInPeriod)\n\t+ [func (m *Message) OnceWithDelay(delay time.Duration)](#Message.OnceWithDelay)\n\t+ [func (m *Message) OnceWithSchedule(tm time.Time)](#Message.OnceWithSchedule)\n\t+ [func (m *Message) SetDelay(delay time.Duration)](#Message.SetDelay)\n\t+ [func (m *Message) String() string](#Message.String)\n\t+ [func (m *Message) UnmarshalBinary(b []byte) error](#Message.UnmarshalBinary)\n* [type ProcessMessageEvent](#ProcessMessageEvent)\n* [type Queue](#Queue)\n* [type QueueConsumer](#QueueConsumer)\n* [type QueueOptions](#QueueOptions)\n* + [func (opt *QueueOptions) Init()](#QueueOptions.Init)\n* [type Redis](#Redis)\n* [type Storage](#Storage)\n* + [func NewLocalStorage() Storage](#NewLocalStorage)\n* [type Task](#Task)\n* + [func RegisterTask(opt *TaskOptions) *Task](#RegisterTask)\n* + [func (t *Task) HandleMessage(msg *Message) error](#Task.HandleMessage)\n\t+ [func (t *Task) Name() string](#Task.Name)\n\t+ [func (t *Task) Options() *TaskOptions](#Task.Options)\n\t+ [func (t *Task) String() string](#Task.String)\n\t+ [func (t *Task) WithArgs(ctx context.Context, args ...interface{}) *Message](#Task.WithArgs)\n* [type TaskMap](#TaskMap)\n* + [func (r *TaskMap) Get(name string) *Task](#TaskMap.Get)\n\t+ [func (r *TaskMap) HandleMessage(msg *Message) error](#TaskMap.HandleMessage)\n\t+ [func (r *TaskMap) Range(fn func(name string, task *Task) bool)](#TaskMap.Range)\n\t+ [func (r *TaskMap) Register(opt *TaskOptions) (*Task, error)](#TaskMap.Register)\n\t+ [func (r *TaskMap) Reset()](#TaskMap.Reset)\n\t+ [func (r *TaskMap) Unregister(task *Task)](#TaskMap.Unregister)\n* [type TaskOptions](#TaskOptions)\n\n#### Examples [¶](#pkg-examples)\n\n* [Package (CustomRateLimit)](#example-package-CustomRateLimit)\n* [Package (MessageDelay)](#example-package-MessageDelay)\n* [Package (Once)](#example-package-Once)\n* [Package (RateLimit)](#example-package-RateLimit)\n* [Package (RetryOnError)](#example-package-RetryOnError)\n\n### Constants [¶](#pkg-constants)\n\nThis section is empty.\n\n### Variables [¶](#pkg-variables)\n\n```\nvar ErrAsyncTask = [errors](/errors).[New](/errors#New)(\"taskq: async task\")\n```\n\n```\nvar ErrDuplicate = [errors](/errors).[New](/errors#New)(\"taskq: message with such name already exists\")\n```\nErrDuplicate is returned when adding duplicate message to the queue.\n\n### Functions [¶](#pkg-functions)\n\n#### \nfunc [SetLogger](https://github.com/vmihailenco/taskq/blob/v3.2.9/taskq.go#L18) [¶](#SetLogger)\n```\nfunc SetLogger(logger *[log](/log).[Logger](/log#Logger))\n```\n#### \nfunc [SetUnknownTaskOptions](https://github.com/vmihailenco/taskq/blob/v3.2.9/task.go#L17) [¶](#SetUnknownTaskOptions)\n```\nfunc SetUnknownTaskOptions(opt *[TaskOptions](#TaskOptions))\n```\n#### \nfunc [Version](https://github.com/vmihailenco/taskq/blob/v3.2.9/version.go#L4) [¶](#Version)\n\nadded in v3.2.5\n```\nfunc Version() [string](/builtin#string)\n```\nVersion is the current release version.\n\n### Types [¶](#pkg-types)\n\n#### \ntype [Consumer](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L51) [¶](#Consumer)\n```\ntype Consumer struct {\n\t// contains filtered or unexported fields\n}\n```\nConsumer reserves messages from the queue, processes them,\nand then either releases or deletes messages from the queue.\n\n#### \nfunc [NewConsumer](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L82) [¶](#NewConsumer)\n```\nfunc NewConsumer(q [Queue](#Queue)) *[Consumer](#Consumer)\n```\nNewConsumer creates new Consumer for the queue using provided processing options.\n\n#### \nfunc [StartConsumer](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L100) [¶](#StartConsumer)\n```\nfunc StartConsumer(ctx [context](/context).[Context](/context#Context), q [Queue](#Queue)) *[Consumer](#Consumer)\n```\nStartConsumer creates new QueueConsumer and starts it.\n\n#### \nfunc (*Consumer) [Add](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L143) [¶](#Consumer.Add)\n```\nfunc (c *[Consumer](#Consumer)) Add(msg *[Message](#Message)) [error](/builtin#error)\n```\n#### \nfunc (*Consumer) [AddHook](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L109) [¶](#Consumer.AddHook)\n```\nfunc (c *[Consumer](#Consumer)) AddHook(hook [ConsumerHook](#ConsumerHook))\n```\nAddHook adds a hook into message processing.\n\n#### \nfunc (*Consumer) [Len](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L121) [¶](#Consumer.Len)\n```\nfunc (c *[Consumer](#Consumer)) Len() [int](/builtin#int)\n```\n#### \nfunc (*Consumer) [Options](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L117) [¶](#Consumer.Options)\n```\nfunc (c *[Consumer](#Consumer)) Options() *[QueueOptions](#QueueOptions)\n```\n#### \nfunc (*Consumer) [Process](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L511) [¶](#Consumer.Process)\n```\nfunc (c *[Consumer](#Consumer)) Process(msg *[Message](#Message)) [error](/builtin#error)\n```\nProcess is low-level API to process message bypassing the internal queue.\n\n#### \nfunc (*Consumer) [ProcessAll](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L314) [¶](#Consumer.ProcessAll)\n```\nfunc (c *[Consumer](#Consumer)) ProcessAll(ctx [context](/context).[Context](/context#Context)) [error](/builtin#error)\n```\nProcessAll starts workers to process messages in the queue and then stops them when all messages are processed.\n\n#### \nfunc (*Consumer) [ProcessOne](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L342) [¶](#Consumer.ProcessOne)\n```\nfunc (c *[Consumer](#Consumer)) ProcessOne(ctx [context](/context).[Context](/context#Context)) [error](/builtin#error)\n```\nProcessOne processes at most one message in the queue.\n\n#### \nfunc (*Consumer) [Purge](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L646) [¶](#Consumer.Purge)\n```\nfunc (c *[Consumer](#Consumer)) Purge() [error](/builtin#error)\n```\nPurge discards messages from the internal queue.\n\n#### \nfunc (*Consumer) [Put](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L591) [¶](#Consumer.Put)\n```\nfunc (c *[Consumer](#Consumer)) Put(msg *[Message](#Message))\n```\n#### \nfunc (*Consumer) [Queue](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L113) [¶](#Consumer.Queue)\n```\nfunc (c *[Consumer](#Consumer)) Queue() [Queue](#Queue)\n```\n#### \nfunc (*Consumer) [Start](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L150) [¶](#Consumer.Start)\n```\nfunc (c *[Consumer](#Consumer)) Start(ctx [context](/context).[Context](/context#Context)) [error](/builtin#error)\n```\nStart starts consuming messages in the queue.\n\n#### \nfunc (*Consumer) [Stats](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L126) [¶](#Consumer.Stats)\n```\nfunc (c *[Consumer](#Consumer)) Stats() *[ConsumerStats](#ConsumerStats)\n```\nStats returns processor stats.\n\n#### \nfunc (*Consumer) [Stop](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L196) [¶](#Consumer.Stop)\n```\nfunc (c *[Consumer](#Consumer)) Stop() [error](/builtin#error)\n```\nStop is StopTimeout with 30 seconds timeout.\n\n#### \nfunc (*Consumer) [StopTimeout](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L202) [¶](#Consumer.StopTimeout)\n```\nfunc (c *[Consumer](#Consumer)) StopTimeout(timeout [time](/time).[Duration](/time#Duration)) [error](/builtin#error)\n```\nStopTimeout waits workers for timeout duration to finish processing current messages and stops workers.\n\n#### \nfunc (*Consumer) [String](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L748) [¶](#Consumer.String)\n```\nfunc (c *[Consumer](#Consumer)) String() [string](/builtin#string)\n```\n#### \ntype [ConsumerHook](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L664) [¶](#ConsumerHook)\n```\ntype ConsumerHook interface {\n BeforeProcessMessage(*[ProcessMessageEvent](#ProcessMessageEvent)) [error](/builtin#error)\n AfterProcessMessage(*[ProcessMessageEvent](#ProcessMessageEvent)) [error](/builtin#error)\n}\n```\n#### \ntype [ConsumerStats](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L26) [¶](#ConsumerStats)\n```\ntype ConsumerStats struct {\n NumWorker [uint32](/builtin#uint32)\n NumFetcher [uint32](/builtin#uint32)\n\n BufferSize [uint32](/builtin#uint32)\n Buffered [uint32](/builtin#uint32)\n\n InFlight [uint32](/builtin#uint32)\n Processed [uint32](/builtin#uint32)\n Retries [uint32](/builtin#uint32)\n Fails [uint32](/builtin#uint32)\n Timing [time](/time).[Duration](/time#Duration)\n}\n```\n#### \ntype [Delayer](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L22) [¶](#Delayer)\n```\ntype Delayer interface {\n Delay() [time](/time).[Duration](/time#Duration)\n}\n```\n#### \ntype [Factory](https://github.com/vmihailenco/taskq/blob/v3.2.9/taskq.go#L24) [¶](#Factory)\n```\ntype Factory interface {\n RegisterQueue(*[QueueOptions](#QueueOptions)) [Queue](#Queue)\n Range(func([Queue](#Queue)) [bool](/builtin#bool))\n StartConsumers([context](/context).[Context](/context#Context)) [error](/builtin#error)\n StopConsumers() [error](/builtin#error)\n Close() [error](/builtin#error)\n}\n```\nFactory is an interface that abstracts creation of new queues.\nIt is implemented in subpackages memqueue, azsqs, and ironmq.\n\n#### \ntype [Handler](https://github.com/vmihailenco/taskq/blob/v3.2.9/handler.go#L18) [¶](#Handler)\n```\ntype Handler interface {\n HandleMessage(msg *[Message](#Message)) [error](/builtin#error)\n}\n```\nHandler is an interface for processing messages.\n\n#### \nfunc [NewHandler](https://github.com/vmihailenco/taskq/blob/v3.2.9/handler.go#L38) [¶](#NewHandler)\n```\nfunc NewHandler(fn interface{}) [Handler](#Handler)\n```\n#### \ntype [HandlerFunc](https://github.com/vmihailenco/taskq/blob/v3.2.9/handler.go#L22) [¶](#HandlerFunc)\n```\ntype HandlerFunc func(*[Message](#Message)) [error](/builtin#error)\n```\n#### \nfunc (HandlerFunc) [HandleMessage](https://github.com/vmihailenco/taskq/blob/v3.2.9/handler.go#L24) [¶](#HandlerFunc.HandleMessage)\n```\nfunc (fn [HandlerFunc](#HandlerFunc)) HandleMessage(msg *[Message](#Message)) [error](/builtin#error)\n```\n#### \ntype [Message](https://github.com/vmihailenco/taskq/blob/v3.2.9/message.go#L22) [¶](#Message)\n```\ntype Message struct {\n Ctx [context](/context).[Context](/context#Context) `msgpack:\"-\"`\n\n // SQS/IronMQ message id.\n\tID [string](/builtin#string) `msgpack:\"1,omitempty,alias:ID\"`\n\n // Optional name for the message. Messages with the same name\n\t// are processed only once.\n\tName [string](/builtin#string) `msgpack:\"-\"`\n\n // Delay specifies the duration the queue must wait\n\t// before executing the message.\n\tDelay [time](/time).[Duration](/time#Duration) `msgpack:\"-\"`\n\n // Args passed to the handler.\n\tArgs []interface{} `msgpack:\"-\"`\n\n\t// Binary representation of the args.\n ArgsCompression [string](/builtin#string) `msgpack:\"2,omitempty,alias:ArgsCompression\"`\n ArgsBin [][byte](/builtin#byte) `msgpack:\"3,alias:ArgsBin\"`\n\n // SQS/IronMQ reservation id that is used to release/delete the message.\n\tReservationID [string](/builtin#string) `msgpack:\"-\"`\n\n // The number of times the message has been reserved or released.\n\tReservedCount [int](/builtin#int) `msgpack:\"4,omitempty,alias:ReservedCount\"`\n\n TaskName [string](/builtin#string) `msgpack:\"5,alias:TaskName\"`\n Err [error](/builtin#error) `msgpack:\"-\"`\n\t// contains filtered or unexported fields\n}\n```\nMessage is used to create and retrieve messages from a queue.\n\n#### \nfunc [NewMessage](https://github.com/vmihailenco/taskq/blob/v3.2.9/message.go#L56) [¶](#NewMessage)\n```\nfunc NewMessage(ctx [context](/context).[Context](/context#Context), args ...interface{}) *[Message](#Message)\n```\n#### \nfunc (*Message) [MarshalArgs](https://github.com/vmihailenco/taskq/blob/v3.2.9/message.go#L120) [¶](#Message.MarshalArgs)\n```\nfunc (m *[Message](#Message)) MarshalArgs() ([][byte](/builtin#byte), [error](/builtin#error))\n```\n#### \nfunc (*Message) [MarshalBinary](https://github.com/vmihailenco/taskq/blob/v3.2.9/message.go#L143) [¶](#Message.MarshalBinary)\n```\nfunc (m *[Message](#Message)) MarshalBinary() ([][byte](/builtin#byte), [error](/builtin#error))\n```\n#### \nfunc (*Message) [OnceInPeriod](https://github.com/vmihailenco/taskq/blob/v3.2.9/message.go#L76) [¶](#Message.OnceInPeriod)\n```\nfunc (m *[Message](#Message)) OnceInPeriod(period [time](/time).[Duration](/time#Duration), args ...interface{})\n```\nOnceInPeriod uses the period and the args to generate such a message name that message with such args is added to the queue once in a given period.\nIf args are not provided then message args are used instead.\n\n#### \nfunc (*Message) [OnceWithDelay](https://github.com/vmihailenco/taskq/blob/v3.2.9/message.go#L81) [¶](#Message.OnceWithDelay)\n```\nfunc (m *[Message](#Message)) OnceWithDelay(delay [time](/time).[Duration](/time#Duration))\n```\n#### \nfunc (*Message) [OnceWithSchedule](https://github.com/vmihailenco/taskq/blob/v3.2.9/message.go#L88) [¶](#Message.OnceWithSchedule)\n```\nfunc (m *[Message](#Message)) OnceWithSchedule(tm [time](/time).[Time](/time#Time))\n```\n#### \nfunc (*Message) [SetDelay](https://github.com/vmihailenco/taskq/blob/v3.2.9/message.go#L69) [¶](#Message.SetDelay)\n```\nfunc (m *[Message](#Message)) SetDelay(delay [time](/time).[Duration](/time#Duration))\n```\nSetDelay sets the message delay.\n\n#### \nfunc (*Message) [String](https://github.com/vmihailenco/taskq/blob/v3.2.9/message.go#L63) [¶](#Message.String)\n```\nfunc (m *[Message](#Message)) String() [string](/builtin#string)\n```\n#### \nfunc (*Message) [UnmarshalBinary](https://github.com/vmihailenco/taskq/blob/v3.2.9/message.go#L175) [¶](#Message.UnmarshalBinary)\n```\nfunc (m *[Message](#Message)) UnmarshalBinary(b [][byte](/builtin#byte)) [error](/builtin#error)\n```\n#### \ntype [ProcessMessageEvent](https://github.com/vmihailenco/taskq/blob/v3.2.9/consumer.go#L657) [¶](#ProcessMessageEvent)\n```\ntype ProcessMessageEvent struct {\n Message *[Message](#Message)\n StartTime [time](/time).[Time](/time#Time)\n\n Stash map[interface{}]interface{}\n}\n```\n#### \ntype [Queue](https://github.com/vmihailenco/taskq/blob/v3.2.9/queue.go#L140) [¶](#Queue)\n```\ntype Queue interface {\n\t[fmt](/fmt).[Stringer](/fmt#Stringer)\n Name() [string](/builtin#string)\n Options() *[QueueOptions](#QueueOptions)\n Consumer() [QueueConsumer](#QueueConsumer)\n\n Len() ([int](/builtin#int), [error](/builtin#error))\n Add(msg *[Message](#Message)) [error](/builtin#error)\n ReserveN(ctx [context](/context).[Context](/context#Context), n [int](/builtin#int), waitTimeout [time](/time).[Duration](/time#Duration)) ([][Message](#Message), [error](/builtin#error))\n Release(msg *[Message](#Message)) [error](/builtin#error)\n Delete(msg *[Message](#Message)) [error](/builtin#error)\n Purge() [error](/builtin#error)\n Close() [error](/builtin#error)\n CloseTimeout(timeout [time](/time).[Duration](/time#Duration)) [error](/builtin#error)\n}\n```\n#### \ntype [QueueConsumer](https://github.com/vmihailenco/taskq/blob/v3.2.9/queue.go#L158) [¶](#QueueConsumer)\n\nadded in v3.2.6\n```\ntype QueueConsumer interface {\n\t// AddHook adds a hook into message processing.\n AddHook(hook [ConsumerHook](#ConsumerHook))\n Queue() [Queue](#Queue)\n Options() *[QueueOptions](#QueueOptions)\n Len() [int](/builtin#int)\n\t// Stats returns processor stats.\n Stats() *[ConsumerStats](#ConsumerStats)\n Add(msg *[Message](#Message)) [error](/builtin#error)\n // Start starts consuming messages in the queue.\n\tStart(ctx [context](/context).[Context](/context#Context)) [error](/builtin#error)\n // Stop is StopTimeout with 30 seconds timeout.\n\tStop() [error](/builtin#error)\n // StopTimeout waits workers for timeout duration to finish processing current\n\t// messages and stops workers.\n\tStopTimeout(timeout [time](/time).[Duration](/time#Duration)) [error](/builtin#error)\n // ProcessAll starts workers to process messages in the queue and then stops\n\t// them when all messages are processed.\n\tProcessAll(ctx [context](/context).[Context](/context#Context)) [error](/builtin#error)\n // ProcessOne processes at most one message in the queue.\n\tProcessOne(ctx [context](/context).[Context](/context#Context)) [error](/builtin#error)\n\t// Process is low-level API to process message bypassing the internal queue.\n Process(msg *[Message](#Message)) [error](/builtin#error)\n Put(msg *[Message](#Message))\n\t// Purge discards messages from the internal queue.\n Purge() [error](/builtin#error)\n String() [string](/builtin#string)\n}\n```\nQueueConsumer reserves messages from the queue, processes them,\nand then either releases or deletes messages from the queue.\n\n#### \ntype [QueueOptions](https://github.com/vmihailenco/taskq/blob/v3.2.9/queue.go#L12) [¶](#QueueOptions)\n```\ntype QueueOptions struct {\n // Queue name.\n\tName [string](/builtin#string)\n\n // Minimum number of goroutines processing messages.\n\t// Default is 1.\n\tMinNumWorker [int32](/builtin#int32)\n // Maximum number of goroutines processing messages.\n\t// Default is 32 * number of CPUs.\n\tMaxNumWorker [int32](/builtin#int32)\n // Global limit of concurrently running workers across all servers.\n\t// Overrides MaxNumWorker.\n\tWorkerLimit [int32](/builtin#int32)\n // Maximum number of goroutines fetching messages.\n\t// Default is 8 * number of CPUs.\n\tMaxNumFetcher [int32](/builtin#int32)\n\n // Number of messages reserved by a fetcher in the queue in one request.\n\t// Default is 10 messages.\n\tReservationSize [int](/builtin#int)\n // Time after which the reserved message is returned to the queue.\n\t// Default is 5 minutes.\n\tReservationTimeout [time](/time).[Duration](/time#Duration)\n // Time that a long polling receive call waits for a message to become\n\t// available before returning an empty response.\n\t// Default is 10 seconds.\n\tWaitTimeout [time](/time).[Duration](/time#Duration)\n // Size of the buffer where reserved messages are stored.\n\t// Default is the same as ReservationSize.\n\tBufferSize [int](/builtin#int)\n\n // Number of consecutive failures after which queue processing is paused.\n\t// Default is 100 failures.\n\tPauseErrorsThreshold [int](/builtin#int)\n\n // Processing rate limit.\n\tRateLimit redis_rate.Limit\n\n // Optional rate limiter. The default is to use Redis.\n\tRateLimiter *redis_rate.Limiter\n\n // Redis client that is used for storing metadata.\n\tRedis [Redis](#Redis)\n\n // Optional storage interface. The default is to use Redis.\n\tStorage [Storage](#Storage)\n\n // Optional message handler. The default is the global Tasks registry.\n\tHandler [Handler](#Handler)\n\n // ConsumerIdleTimeout Time after which the consumer need to be deleted.\n\t// Default is 6 hour\n\tConsumerIdleTimeout [time](/time).[Duration](/time#Duration)\n\n // SchedulerBackoffTime is the time of backoff for the scheduler(\n\t// Scheduler was designed to clean zombie Consumer and requeue pending msgs, and so on.\n\t// Default is randomly between 1~1.5s\n\t// We can change it to a bigger value so that it won't slowdown the redis when using redis queue.\n\t// It will be between SchedulerBackoffTime and SchedulerBackoffTime+250ms.\n\tSchedulerBackoffTime [time](/time).[Duration](/time#Duration)\n\t// contains filtered or unexported fields\n}\n```\n#### \nfunc (*QueueOptions) [Init](https://github.com/vmihailenco/taskq/blob/v3.2.9/queue.go#L76) [¶](#QueueOptions.Init)\n```\nfunc (opt *[QueueOptions](#QueueOptions)) Init()\n```\n#### \ntype [Redis](https://github.com/vmihailenco/taskq/blob/v3.2.9/taskq.go#L32) [¶](#Redis)\n```\ntype Redis interface {\n Del(ctx [context](/context).[Context](/context#Context), keys ...[string](/builtin#string)) *redis.IntCmd\n SetNX(ctx [context](/context).[Context](/context#Context), key [string](/builtin#string), value interface{}, expiration [time](/time).[Duration](/time#Duration)) *redis.BoolCmd\n Pipelined(ctx [context](/context).[Context](/context#Context), fn func(pipe redis.Pipeliner) [error](/builtin#error)) ([]redis.Cmder, [error](/builtin#error))\n\n\t// Eval Required by redislock\n Eval(ctx [context](/context).[Context](/context#Context), script [string](/builtin#string), keys [][string](/builtin#string), args ...interface{}) *redis.Cmd\n EvalSha(ctx [context](/context).[Context](/context#Context), sha1 [string](/builtin#string), keys [][string](/builtin#string), args ...interface{}) *redis.Cmd\n ScriptExists(ctx [context](/context).[Context](/context#Context), scripts ...[string](/builtin#string)) *redis.BoolSliceCmd\n ScriptLoad(ctx [context](/context).[Context](/context#Context), script [string](/builtin#string)) *redis.StringCmd\n}\n```\n#### \ntype [Storage](https://github.com/vmihailenco/taskq/blob/v3.2.9/storage.go#L11) [¶](#Storage)\n```\ntype Storage interface {\n Exists(ctx [context](/context).[Context](/context#Context), key [string](/builtin#string)) [bool](/builtin#bool)\n}\n```\n#### \nfunc [NewLocalStorage](https://github.com/vmihailenco/taskq/blob/v3.2.9/storage.go#L25) [¶](#NewLocalStorage)\n\nadded in v3.2.9\n```\nfunc NewLocalStorage() [Storage](#Storage)\n```\n#### \ntype [Task](https://github.com/vmihailenco/taskq/blob/v3.2.9/task.go#L77) [¶](#Task)\n```\ntype Task struct {\n\t// contains filtered or unexported fields\n}\n```\n#### \nfunc [RegisterTask](https://github.com/vmihailenco/taskq/blob/v3.2.9/task.go#L84) [¶](#RegisterTask)\n```\nfunc RegisterTask(opt *[TaskOptions](#TaskOptions)) *[Task](#Task)\n```\n#### \nfunc (*Task) [HandleMessage](https://github.com/vmihailenco/taskq/blob/v3.2.9/task.go#L105) [¶](#Task.HandleMessage)\n```\nfunc (t *[Task](#Task)) HandleMessage(msg *[Message](#Message)) [error](/builtin#error)\n```\n#### \nfunc (*Task) [Name](https://github.com/vmihailenco/taskq/blob/v3.2.9/task.go#L93) [¶](#Task.Name)\n```\nfunc (t *[Task](#Task)) Name() [string](/builtin#string)\n```\n#### \nfunc (*Task) [Options](https://github.com/vmihailenco/taskq/blob/v3.2.9/task.go#L101) [¶](#Task.Options)\n```\nfunc (t *[Task](#Task)) Options() *[TaskOptions](#TaskOptions)\n```\n#### \nfunc (*Task) [String](https://github.com/vmihailenco/taskq/blob/v3.2.9/task.go#L97) [¶](#Task.String)\n```\nfunc (t *[Task](#Task)) String() [string](/builtin#string)\n```\n#### \nfunc (*Task) [WithArgs](https://github.com/vmihailenco/taskq/blob/v3.2.9/task.go#L115) [¶](#Task.WithArgs)\n```\nfunc (t *[Task](#Task)) WithArgs(ctx [context](/context).[Context](/context#Context), args ...interface{}) *[Message](#Message)\n```\n#### \ntype [TaskMap](https://github.com/vmihailenco/taskq/blob/v3.2.9/registry.go#L11) [¶](#TaskMap)\n```\ntype TaskMap struct {\n\t// contains filtered or unexported fields\n}\n```\n```\nvar Tasks [TaskMap](#TaskMap)\n```\n#### \nfunc (*TaskMap) [Get](https://github.com/vmihailenco/taskq/blob/v3.2.9/registry.go#L15) [¶](#TaskMap.Get)\n```\nfunc (r *[TaskMap](#TaskMap)) Get(name [string](/builtin#string)) *[Task](#Task)\n```\n#### \nfunc (*TaskMap) [HandleMessage](https://github.com/vmihailenco/taskq/blob/v3.2.9/registry.go#L59) [¶](#TaskMap.HandleMessage)\n```\nfunc (r *[TaskMap](#TaskMap)) HandleMessage(msg *[Message](#Message)) [error](/builtin#error)\n```\n#### \nfunc (*TaskMap) [Range](https://github.com/vmihailenco/taskq/blob/v3.2.9/registry.go#L53) [¶](#TaskMap.Range)\n```\nfunc (r *[TaskMap](#TaskMap)) Range(fn func(name [string](/builtin#string), task *[Task](#Task)) [bool](/builtin#bool))\n```\n#### \nfunc (*TaskMap) [Register](https://github.com/vmihailenco/taskq/blob/v3.2.9/registry.go#L25) [¶](#TaskMap.Register)\n```\nfunc (r *[TaskMap](#TaskMap)) Register(opt *[TaskOptions](#TaskOptions)) (*[Task](#Task), [error](/builtin#error))\n```\n#### \nfunc (*TaskMap) [Reset](https://github.com/vmihailenco/taskq/blob/v3.2.9/registry.go#L49) [¶](#TaskMap.Reset)\n```\nfunc (r *[TaskMap](#TaskMap)) Reset()\n```\n#### \nfunc (*TaskMap) [Unregister](https://github.com/vmihailenco/taskq/blob/v3.2.9/registry.go#L45) [¶](#TaskMap.Unregister)\n```\nfunc (r *[TaskMap](#TaskMap)) Unregister(task *[Task](#Task))\n```\n#### \ntype [TaskOptions](https://github.com/vmihailenco/taskq/blob/v3.2.9/task.go#L22) [¶](#TaskOptions)\n```\ntype TaskOptions struct {\n // Task name.\n\tName [string](/builtin#string)\n\n // Function called to process a message.\n\t// There are three permitted types of signature:\n\t// 1. A zero-argument function\n\t// 2. A function whose arguments are assignable in type from those which are passed in the message\n\t// 3. A function which takes a single `*Message` argument\n\t// The handler function may also optionally take a Context as a first argument and may optionally return an error.\n\t// If the handler takes a Context, when it is invoked it will be passed the same Context as that which was passed to\n\t// `StartConsumer`. If the handler returns a non-nil error the message processing will fail and will be retried/.\n\tHandler interface{}\n // Function called to process failed message after the specified number of retries have all failed.\n\t// The FallbackHandler accepts the same types of function as the Handler.\n\tFallbackHandler interface{}\n\n // Optional function used by Consumer with defer statement\n\t// to recover from panics.\n\tDeferFunc func()\n\n // Number of tries/releases after which the message fails permanently\n\t// and is deleted.\n\t// Default is 64 retries.\n\tRetryLimit [int](/builtin#int)\n // Minimum backoff time between retries.\n\t// Default is 30 seconds.\n\tMinBackoff [time](/time).[Duration](/time#Duration)\n // Maximum backoff time between retries.\n\t// Default is 30 minutes.\n\tMaxBackoff [time](/time).[Duration](/time#Duration)\n\t// contains filtered or unexported fields\n}\n```"}}},{"rowIdx":59190,"cells":{"project":{"kind":"string","value":"cencrne"},"source":{"kind":"string","value":"cran"},"language":{"kind":"string","value":"R"},"content":{"kind":"string","value":"Package ‘cencrne’\n January 9, 2023\nTitle Consistent Estimation of the Number of Communities via\n Regularized Network Embedding\nVersion 1.0.0\nDescription The network analysis plays an important role in numerous application domains includ-\n ing biomedicine.\n Estimation of the number of communities is a fundamental and critical issue in network analy-\n sis. Most existing studies assume that the number of communities is known a pri-\n ori, or lack of rigorous theoretical guarantee on the estimation consistency. This method pro-\n poses a regularized network embedding model to simultaneously estimate the community struc-\n ture and the number of communities in a unified formulation.\n The proposed model equips network embedding with a novel composite regulariza-\n tion term, which pushes the embedding vector towards its center and collapses similar commu-\n nity centers with each other. A rigorous theoretical analysis is conducted, establishing asymp-\n totic consistency in terms of community detection and estimation of the number of communities.\n Reference:\n <NAME>., <NAME>. and <NAME>. (2022). ``Consistent Estimation of the Number of Communi-\n ties via Regularized Network Embedding''. Biometrics, <doi:10.1111/biom.13815>.\nLicense GPL-2\nEncoding UTF-8\nImports MASS, Matrix\nLazyData true\nLazyLoad yes\nRoxygenNote 7.1.2\nDepends R (>= 3.5.0)\nSuggests knitr, rmarkdown\nVignetteBuilder knitr, rmarkdown\nNeedsCompilation no\nAuthor <NAME> [aut, cre] (<https://orcid.org/0000-0002-8061-9940>),\n <NAME> [aut],\n <NAME> [aut]\nMaintainer <NAME> <<EMAIL>>\nRepository CRAN\nDate/Publication 2023-01-09 10:00:05 UTC\nR topics documented:\nevaluatio... 2\nexample.dat... 3\ngen.in... 3\ngenelambda.ob... 5\nnetwork.comm.nu... 6\n evaluation Consistent Estimation of the Number of Communities via Regularized\n Network Embedding.\nDescription\n The evaluation function for Consistent Estimation of the Number of Communities via Regularized\n Network Embedding.\nUsage\n evaluation(Z.hat, Z.true, cluster.matrix.hat, cluster.matrix.true,\n P.true, Theta.true, K.hat=4, K.true=4)\nArguments\n Z.hat A n * r matrix, the estimated embedding vectors corresponding to n nodes.\n Z.true A n * r matrix, the true embedding vectors corresponding to n nodes.\n cluster.matrix.hat\n A n * n estimated membership matrix, whose (i,j)-element is 1, if nodes i and j\n are estimated to belong to the same community, and 0, otherwise.\n cluster.matrix.true\n A n * n true membership matrix, whose (i,j)-element is 1, if nodes i and j belong\n to the same community, and 0, otherwise.\n P.true A n * n true probability matrix.\n Theta.true A n * n true matrix: Z.true %*% t(Z.true).\n K.hat The true number of communities.\n K.true The estimated number of communities.\nValue\n A vector including five evaluation index. prop. 1: the estimated and actual number of communities\n are equal; 0: not equal.\nAuthor(s)\n <NAME>.\n example.data Some example data\nDescription\n Some example data\nFormat\n A list including: A: An observed n * n adjacency matrix of undirected graph, n=360. K.true: The\n estimated number of communities. Z.true: A n * r matrix, the true embedding vectors corresponding\n to n nodes, n=360, r=5. B.true: A n * r matrix, the true community centers corresponding to n\n nodes. P.true: A n * n true probability matrix. Theta.true: A n * n true matrix: Z.true %*%\n t(Z.true). cluster.matrix.true: A n * n true membership matrix, whose (i,j)-element is 1, if nodes i\n and j belong to the same community, and 0, otherwise.\nSource\n Simulated data\nExamples\n data(example.data)\n gen.int Consistent Estimation of the Number of Communities via Regularized\n Network Embedding.\nDescription\n The function generating the initial values.\nUsage\n gen.int(A, R=8, K.max0=8,rand.seed=123,\n lambda3=0, a=3, kappa=1, alpha=1,\n eps = 1e-2, niter = 20, niter.Z=5)\nArguments\n A An observed n * n adjacency matrix of undirected graph.\n R Int, the relatively large dimension of embedding vectors given in advance.\n K.max0 The relatively large upper bound of the number of communities given in advance\n to generate initial values of B.\n rand.seed The random seed of generating initial value.\n lambda3 A float value, the tuning parameter for sparsity of Z.\n a A float value, regularization parameter in MCP, the default setting is 3.\n kappa A float value, the penalty parameter in ADMM algorithm, the default setting is\n 1.\n alpha A float value, the step size of coordinate descent algorithm updating Z, the de-\n fault setting is 1.\n eps A float value, algorithm termination threshold.\n niter Int, maximum number of cycles of the overall ADMM algorithm.\n niter.Z Int, maximum number of cycles of coordinate descent algorithm updating Z.\nValue\n A list including all estimated parameters and the BIC values with all choices of given tuning param-\n eters, and the selected optional parameters. Opt_Z: A n * r matrix, the estimated embedding vectors\n corresponding to n nodes; Opt_B: A n * r matrix, the estimated community centers corresponding\n to n nodes; Opt_K: Int, the estimated number of communities; Opt_member: A n-dimensional vec-\n tor, describing the membership of n nodes; Opt_cluster.matrix: A n * n membership matrix, whose\n (i,j)-element is 1, if nodes i and j belong to the same community, and 0, otherwise.\nAuthor(s)\n <NAME>.\nReferences\n <NAME>., <NAME>., <NAME>. and <NAME>. (2022). Consistent Estimation of the Number of Commu-\n nities via Regularized Network Embedding.\nExamples\n library(cencrne)\n data(example.data)\n A = example.data$A\n K.true = example.data$K.true\n Z.true = example.data$Z.true\n B.true = example.data$B.true\n P.true = example.data$P.true\n Theta.true = example.data$Theta.true\n cluster.matrix.true = example.data$cluster.matrix.true\n n = dim(A)[1]\n sample.index.n = rbind(combn(n,2),1:(n*(n-1)/2))\n int.list = gen.int(A)\n Z.int = int.list$Z.int\n B.int = int.list$B.int\n genelambda.obo Generate tuning parameters\nDescription\n Generating a sequence of the tuning parameters (lambda1, lambda2, and lambda3).\nUsage\n genelambda.obo(nlambda1=10,lambda1_max=1,lambda1_min=0.05,\n nlambda2=10,lambda2_max=1,lambda2_min=0.01,\n nlambda3=10,lambda3_max=5,lambda3_min=0.5)\nArguments\n nlambda1 The numbers of lambda 1.\n lambda1_max The maximum values of lambda 1.\n lambda1_min The minimum values of lambda 1.\n nlambda2 The numbers of lambda 2.\n lambda2_max The maximum values of lambda 2.\n lambda2_min The minimum values of lambda 2.\n nlambda3 The numbers of lambda 3.\n lambda3_max The maximum values of lambda 3.\n lambda3_min The minimum values of lambda 3.\nValue\n A sequence of the tuning parameters (lambda1, lambda2, and lambda3).\nAuthor(s)\n <NAME>\nExamples\n lambda <- genelambda.obo(nlambda1=5,lambda1_max=0.5,lambda1_min=0.1, nlambda2=15,lambda2_max=1.5,\n lambda2_min=0.1, nlambda3=10,lambda3_max=3.5,lambda3_min=0.5)\n lambda\n network.comm.num Consistent Estimation of the Number of Communities via Regularized\n Network Embedding.\nDescription\n The main function for Consistent Estimation of the Number of Communities via Regularized Net-\n work Embedding.\nUsage\n network.comm.num(A, sample.index.n, lambda, Z.int, B.int,\n a=3, kappa=1, alpha=1, eps=5e-2, niter=20,\n niter.Z=5, update.B=\"ADMM\",local.oppro=FALSE, merge.all=TRUE,\n ad.BIC=FALSE, Fully.Connected=TRUE, trace=FALSE,\n line.search=TRUE, ad.BIC.B=FALSE)\nArguments\n A An observed n * n adjacency matrix of undirected graph.\n sample.index.n A 3 * (n*(n-1)/2) matrix, all pairs of integers from 1 to n.\n lambda A list, the sequences of the tuning parameters (λ1 , λ2 , and λ3 ).\n Z.int A n * r matrix, the initial values of embedding vectors corresponding to n nodes.\n B.int A n * r matrix, the initial values of community centers corresponding to n nodes.\n a A float value, regularization parameter in MCP, the default setting is 3.\n kappa A float value, the penalty parameter in ADMM algorithm, the default setting is\n 1.\n alpha A float value, the step size of coordinate descent algorithm updating Z, the de-\n fault setting is 1.\n eps A float value, algorithm termination threshold.\n niter Int, maximum number of cycles of the overall ADMM algorithm.\n niter.Z Int, maximum number of cycles of coordinate descent algorithm updating Z.\n update.B The optimization algorithm updating B, which can be selected \"ADMM\" (de-\n fault) and \"AMA\".\n local.oppro The logical variable, whether to use local approximations when updating Z, the\n default setting is F.\n merge.all Whether to merge pairs of nodes indirectly connected (but without the direct\n edge) in the estimated community membership matrix.\n ad.BIC Whether to use the adjusted BIC, the default setting is F.\n Fully.Connected\n Whether to use the all pairs (i,j) in fusion penalty, the default setting is T. If F, the\n pairs (i,j) in fusion penalty will be determined by the observed n * n adjacency\n matrix A.\n trace Whether to output the intermediate process of the algorithm.\n line.search Linear search or not, the default setting is T.\n ad.BIC.B Whether the BIC criterion contains terms involving the B matrix, the default\n setting is F.\nValue\n A list including all estimated parameters and the BIC values with all choices of given tuning param-\n eters, and the selected optional parameters. Opt_Z: A n * r matrix, the estimated embedding vectors\n corresponding to n nodes; Opt_B: A n * r matrix, the estimated community centers corresponding\n to n nodes; Opt_K: Int, the estimated number of communities; Opt_member: A n-dimensional vec-\n tor, describing the membership of n nodes; Opt_cluster.matrix: A n * n membership matrix, whose\n (i,j)-element is 1, if nodes i and j belong to the same community, and 0, otherwise.\nAuthor(s)\n <NAME>, <NAME>, <NAME>. Maintainer: <NAME> <EMAIL>.\nReferences\n <NAME>., <NAME>. and <NAME>. (2022). Consistent Estimation of the Number of Communities via\n Regularized Network Embedding.\nExamples\n library(cencrne)\n data(example.data)\n A = example.data$A\n K.true = example.data$K.true\n Z.true = example.data$Z.true\n B.true = example.data$B.true\n P.true = example.data$P.true\n Theta.true = example.data$Theta.true\n cluster.matrix.true = example.data$cluster.matrix.true\n n = dim(A)[1]\n lam.max = 3\n lam.min = 0.5\n lam1.s = 2/log(n)\n lam2.s = sqrt(8*log(n)/n)\n lam3.s = 1/8/log(n)/sqrt(n)\n lambda = genelambda.obo(nlambda1=3,lambda1_max=lam.max*lam1.s,lambda1_min=lam.min*lam1.s,\n nlambda2=10,lambda2_max=lam.max*lam2.s,lambda2_min=lam.min*lam2.s,\n nlambda3=1,lambda3_max=lam.max*lam3.s,lambda3_min=lam.min*lam3.s)\n sample.index.n = rbind(combn(n,2),1:(n*(n-1)/2))\n int.list = gen.int(A)\n Z.int = int.list$Z.int\n B.int = int.list$B.int\n res = network.comm.num(A, sample.index.n, lambda, Z.int, B.int)\n K.hat = res$Opt_K # the estimated number of communities\n Z.hat = res$Opt_Z # the estimated embedding vectors corresponding to n nodes\n cluster.matrix.hat = res$Opt_cluster.matrix # the n * n estimated membership matrix\n evaluation(Z.hat, Z.true, cluster.matrix.hat, cluster.matrix.true,\n P.true, Theta.true, K.hat, K.true)"}}},{"rowIdx":59191,"cells":{"project":{"kind":"string","value":"hep"},"source":{"kind":"string","value":"ctan"},"language":{"kind":"string","value":"TeX"},"content":{"kind":"string","value":"# The hep-reference package+\nFootnote †: This document corresponds to hep-reference v1.2.\n\nAdjustments of standard references, footnote, and citations\n\n<NAME>\n\n<EMAIL>\n\n2023/07/01\n\n###### Abstract\n\nThe hep-reference package applies some light convenience modifications to the reference, citation and footnote macros after loading standard classes improving on the default LaTeX behaviour.\n\n## 1 Introduction\n\nThe hyperref package is loaded and the links are hidden.\n\n\\cref References are extended with the cleveref package [1], which allows to e.g. just type \\cref(\\_key) in order to write 'figure 1'. Furthermore, the cleveref package allows to reference multiple objects within one \\creff(\\_key_,_key_2).\n\n\\cite Citations are adjusted to not start on a new line in order to avoid the repeated use of \\cite(\\_key_).\n\n\\ref References are also adjusted to not start on a new line.\n\n\\eqref Footnotes are adjusted to swallow white space before the footnote mark and at the beginning of \\subref the footnote text. Additionally the reference back to the text where they are called from.\n\n\\footnote\n\n## References\n\n* [1] <NAME>. 'The cleveref package: Intelligent cross-referencing' (2006). CTAN: cleveref. url: dr-qubit.org/cleveref."}}},{"rowIdx":59192,"cells":{"project":{"kind":"string","value":"bigmds"},"source":{"kind":"string","value":"cran"},"language":{"kind":"string","value":"R"},"content":{"kind":"string","value":"Package ‘bigmds’\n October 12, 2022\nTitle Multidimensional Scaling for Big Data\nVersion 2.0.1\nDescription MDS is a statistic tool for reduction of dimensionality, using as input a distance\n matrix of dimensions n × n. When n is large, classical algorithms suffer from\n computational problems and MDS configuration can not be obtained.\n With this package, we address these problems by means of three algorithms:\n - Divide-and-conquer MDS proposed by <NAME>. and <NAME> (2021)\n <arXiv:2007.11919>.\n - Interpolation MDS, also proposed by <NAME>. and <NAME> (2021)\n <arXiv:2007.11919>, which uses Gower's interpolation formula as described in\n <NAME>. and <NAME> (1995).\n - Fast MDS, which is an implementation of the algorithm proposed by\n <NAME>, <NAME>, and <NAME> (2006).\n The main idea of these algorithms is based on partitioning the data set into small\n pieces, where classical methods can work. In order to align all the solutions,\n Procrustes formula is used as described in Borg, I. and <NAME> (2005).\nLicense MIT + file LICENSE\nEncoding UTF-8\nRoxygenNote 7.1.1\nImports stats, parallel\nSuggests testthat\nURL https://github.com/pachoning/bigmds\nBugReports https://github.com/pachoning/bigmds/issues\nNeedsCompilation no\nAuthor <NAME> [aut, cre]\n (<https://orcid.org/0000-0001-9518-4874>),\n <NAME> [aut] (<https://orcid.org/0000-0003-3933-4852>)\nMaintainer <NAME> <<EMAIL>>\nRepository CRAN\nDate/Publication 2021-10-05 10:00:02 UTC\nR topics documented:\ndivide_conquer_md... 2\nfast_md... 3\ninterpolation_md... 5\n divide_conquer_mds Divide-and-conquer MDS\nDescription\n Roughly speaking, a large data set, x, of size n is divided into parts, then classical MDS is performed\n over every part and, finally, the partial configurations are combined so that all the points lie on the\n same coordinate system with the aim to obtain a global MDS configuration.\nUsage\n divide_conquer_mds(x, l, c_points, r, n_cores = 1, dist_fn = stats::dist, ...)\nArguments\n x A matrix with n points (rows) and k variables (columns).\n l The size for which classical MDS can be computed efficiently (using cmdscale\n function). It means that if ¯l is the limit size for which classical MDS is applica-\n ble, then l≤ ¯l.\n c_points Number of points used to align the MDS solutions obtained by the division of x\n into p data subsets. Recommended value: 2·r.\n r Number of principal coordinates to be extracted.\n n_cores Number of cores wanted to use to run the algorithm.\n dist_fn Distance function used to compute the distance between the rows.\n ... Further arguments passed to dist_fn function.\nDetails\n The divide-and-conquer MDS starts dividing the n points into p partitions: the first partition con-\n tains l points and the others contain l-c_points points. Therefore, p = 1+(n−l)/(l-c_points).\n The partitions are created at random.\n Once the partitions are created, c_points different random points are taken from the first partition\n and concatenated to the other partitions After that, classical MDS is applied to each partition, with\n target low dimensional configuration r.\n Since all the partitions share c_points points with the first one, Procrustes can be applied in order\n to align all the configurations. Finally, all the configurations are concatenated in order to obtain a\n global MDS configuration.\nValue\n Returns a list containing the following elements:\n points A matrix that consists of n points (rows) and r variables (columns) corresponding to the\n principal coordinates. Since a dimensionality reduction is performed, r<< k\n Pp\n eigen The first r largest eigenvalues: λ̄i , i ∈ {1, . . . , r}, where λ̄i = 1/p j=1 λji /nj , being λji\n the i − th eigenvalue from partition j and nj the size of the partition j.\n GOF A numeric vector of length 2.\n The first element corresponds to 1/n j=1 nj Gj1 , where Gj1 = i=1 λji / i=1 |λji |.\n The second element corresponds to 1/n j=1 nj Gj2 where Gj2 = i=1 λji / i=1 max(λji , 0).\nReferences\n <NAME>. and <NAME> (2021). Multidimensional Scaling for Big Data. https://\n arxiv.org/abs/2007.11919.\n <NAME>. and <NAME> (2005). Modern Multidimensional Scaling: Theory and Applications.\n Springer.\nExamples\n set.seed(42)\n x <- matrix(data = rnorm(4*10000), nrow = 10000) %*% diag(c(9, 4, 1, 1))\n mds <- divide_conquer_mds(x = x, l = 200, c_points = 2*2, r = 2, n_cores = 1, dist_fn = stats::dist)\n head(mds$points)\n mds$eigen\n mds$GOF\n points <- mds$points\n plot(x[1:10, 1],\n x[1:10, 2],\n xlim = range(c(x[1:10,1],points[1:10,1])),\n ylim = range(c(x[1:10,2], points[1:10,2])),\n pch = 19,\n col = \"green\")\n text(x[1:10, 1], x[1:10, 2], labels=1:10)\n points(points[1:10, 1], points[1:10, 2], pch = 19, col = \"orange\")\n text(points[1:10, 1], points[1:10, 2], labels=1:10)\n abline(v = 0, lwd=3, lty=2)\n abline(h = 0, lwd=3, lty=2)\n fast_mds Fast MDS\nDescription\n Fast MDS uses recursive programming in combination with a divide and conquer strategy in order\n to obtain an MDS configuration for a given large data set x.\nUsage\n fast_mds(x, l, s_points, r, n_cores = 1, dist_fn = stats::dist, ...)\nArguments\n x A matrix with n individuals (rows) and k variables (columns).\n l The size for which classical MDS can be computed efficiently (using cmdscale\n function). It means that if ¯l is the limit size for which classical MDS is applica-\n ble, then l≤ ¯l.\n s_points Number of points used to align the MDS solutions obtained by the division of x\n into p submatrices. Recommended value: 2·r.\n r Number of principal coordinates to be extracted.\n n_cores Number of cores wanted to use to run the algorithm.\n dist_fn Distance function used to compute the distance between the rows.\n ... Further arguments passed to dist_fn function.\nDetails\n Fast MDS randomly divides the whole sample data set, x, of size n into p =l/s_points data\n subsets, where l ≤ ¯l being ¯l the limit size for which classical MDS is applicable. Each one of\n the p data subsets has size ñ = n/p. If ñ ≤ l then classical MDS is applied to each data subset.\n Otherwise, fast MDS is recursively applied. In either case, a final MDS configuration is obtained\n for each data subset.\n In order to align all the partial solutions, a small subset of size s_points is randomly selected\n from each data subset. They are joined to form an alignment set, over which classical MDS is\n performed giving rise to an alignment configuration. Every data subset shares s_points points\n with the alignment set. Therefore every MDS configuration can be aligned with the alignment\n configuration using a Procrustes transformation.\nValue\n Returns a list containing the following elements:\n points A matrix that consists of n individuals (rows) and r variables (columns) corresponding to\n the principal coordinates. Since we are performing a dimensionality reduction, r<< k\n Pp\n eigen The first r largest eigenvalues: λ̄i , i ∈ {1, . . . , r}, where λ̄i = 1/p j=1 λji /nj , being λji\n the i − th eigenvalue from partition j and nj the size of the partition j.\n GOF A numeric vector of length 2.\n The first element corresponds to 1/n j=1 nj G1 , where G1 = i=1 λi / i=1 |λi |.\n The second element corresponds to 1/n j=1 nj Gj2 where Gj2 = i=1 λji / i=1 max(λji , 0).\nReferences\n <NAME>. and <NAME> (2021). Multidimensional Scaling for Big Data. https://\n arxiv.org/abs/2007.11919.\n <NAME>., <NAME>, <NAME>, and W.Wang (2006). A fast approximation to multidimensional\n scaling. In Proceedings of the ECCV Workshop on Computation Intensive Methods for Computer\n Vision (CIMCV).\n <NAME>. and <NAME> (2005). Modern Multidimensional Scaling: Theory and Applications.\n Springer.\nExamples\n set.seed(42)\n x <- matrix(data = rnorm(4*10000), nrow = 10000) %*% diag(c(9, 4, 1, 1))\n mds <- fast_mds(x = x, l = 200, s_points = 2*2, r = 2, n_cores = 1, dist_fn = stats::dist)\n head(mds$points)\n mds$eigen\n mds$GOF\n points <- mds$points\n plot(x[1:10, 1],\n x[1:10, 2],\n xlim = range(c(x[1:10,1],points[1:10,1])),\n ylim = range(c(x[1:10,2], points[1:10,2])),\n pch = 19,\n col = \"green\")\n text(x[1:10, 1], x[1:10, 2], labels=1:10)\n points(points[1:10, 1], points[1:10, 2], pch = 19, col = \"orange\")\n text(points[1:10, 1], points[1:10, 2], labels=1:10)\n abline(v = 0, lwd=3, lty=2)\n abline(h = 0, lwd=3, lty=2)\n interpolation_mds Interpolation MDS\nDescription\n Given that the size of the data set is too large, this algorithm consists of taking a random sample\n from it of size l ≤ ¯l, being ¯l the limit size for which classical MDS is applicable, to perform\n classical MDS to it, and to extend the obtained results to the rest of the data set by using Gower’s\n interpolation formula, which allows to add a new set of points to an existing MDS configuration.\nUsage\n interpolation_mds(x, l, r, n_cores = 1, dist_fn = stats::dist, ...)\nArguments\n x A matrix with n individuals (rows) and k variables (columns).\n l The size for which classical MDS can be computed efficiently (using cmdscale\n function). It means that if ¯l is the limit size for which classical MDS is applica-\n ble, then l≤ ¯l.\n r Number of principal coordinates to be extracted.\n n_cores Number of cores wanted to use to run the algorithm.\n dist_fn Distance function used to compute the distance between the rows.\n ... Further arguments passed to dist_fn function.\nDetails\n Gower’s interpolation formula is the central piece of this algorithm since it allows to add a new set\n of points to an existing MDS configuration so that the new one has the same coordinate system.\n Given the matrix x with n points (rows) and and k variables (columns), a first data subsets (based\n on a random sample) of size l is taken and it is used to compute a MDS configuration.\n The remaining part of x is divided into p = (n−l)/l data subsets (randomly). For every data\n subset, it is obtained a MDS configuration by means of Gower’s interpolation formula and the first\n MDS configuration obtained previously. Every MDS configuration is appended to the existing one\n so that, at the end of the process, a global MDS configuration for x is obtained.\nValue\n Returns a list containing the following elements:\n points A matrix that consists of n individuals (rows) and r variables (columns) corresponding to\n the principal coordinates. Since we are performing a dimensionality reduction, r<< k\n eigen The first r largest eigenvalues: λi , i ∈ {1, . . . , r}, where each λi is obtained from applying\n classical MDS to the first data subset.\n GOF A numeric vector of length 2.\n Pr Pn−1\n The first element corresponds to i=1 λi / i=1 |λi |.\n Pr Pn−1\n The second element corresponds to i=1 λi / i=1 max(λi , 0).\nReferences\n <NAME>. and <NAME> (2021). Multidimensional Scaling for Big Data. https://\n arxiv.org/abs/2007.11919.\n <NAME>. and <NAME> (1995). Biplots, Volume 54. CRC Press.\n <NAME>. and <NAME> (2005). Modern Multidimensional Scaling: Theory and Applications.\n Springer.\nExamples\n set.seed(42)\n x <- matrix(data = rnorm(4*10000), nrow = 10000) %*% diag(c(9, 4, 1, 1))\n mds <- interpolation_mds(x = x, l = 200, r = 2, n_cores = 1, dist_fn = stats::dist)\n head(mds$points)\n mds$eigen\n mds$GOF\n points <- mds$points\n plot(x[1:10, 1],\n x[1:10, 2],\n xlim = range(c(x[1:10,1],points[1:10,1])),\n ylim = range(c(x[1:10,2], points[1:10,2])),\n pch = 19,\n col = \"green\")\n text(x[1:10, 1], x[1:10, 2], labels=1:10)\n points(points[1:10, 1], points[1:10, 2], pch = 19, col = \"orange\")\n text(points[1:10, 1], points[1:10, 2], labels=1:10)\n abline(v = 0, lwd=3, lty=2)\n abline(h = 0, lwd=3, lty=2)"}}},{"rowIdx":59193,"cells":{"project":{"kind":"string","value":"dissolver"},"source":{"kind":"string","value":"hex"},"language":{"kind":"string","value":"Erlang"},"content":{"kind":"string","value":"Dissolver\n===\n\n### \n\n NOTE: This is a wip repo. So be warned there maybe be bugs.\n\nThis project is a fork of <https://github.com/elixirdrops/kerosene>. \nI thought to take it over because it does not look as if its being activly developed and I wanted more features.\n\nTODO: \n\n* [x] Lazy query - Instead of pagination calling Repo.all it will return an Ecto query. This is useful for subqueries where you will passying the query to something like a preload.\n* [x] Custom themes - Now you can pass a module as the source of your theme.\n* [ ] Refactor all the test so that the modules only exposed required interfaces.\n* [ ] The way this lib queries for total counts is a bit odd since it's trying to account for groub_by and multi sourced froms. I'm going to see if we can't make this cleaner.\n* [ ] Refactor namespace of functions. General clean up of internal methods.\n\n---\n\n[](https://coveralls.io/github/MorphicPro/dissolver?branch=master)\n\nPagination for Ecto and Phoenix.\n\n Installation\n---\n\nadd Dissolver to your mix.exs dependencies:\n```\ndef deps do\n [\n {:dissolver, \"~> 0.9.4\"}\n ]\nend\n```\nNext provide Dissolver your Repo module via the config.\nAdd the following to your config:\n```\n...\nconfig :dissolver,\n repo: MyApp.Repo\n per_page: 2\n\nimport_config \"#{Mix.env()}.exs\"\n```\nFor more information about the configuration options look at the [Configurations](#module-configuration) section\n\nNow you are ready to start using Dissolver.\n\n Usage\n---\n\nStart paginating your queries\n```\ndef index(conn, params) do\n {products, paginator} =\n Product\n |> Product.with_lowest_price\n |> Dissolver.paginate(params)\n\n render(conn, \"index.html\", products: products, paginator: paginator)\nend\n```\nAdd the view helper to your view\n```\ndefmodule MyApp.ProductView do\n use MyApp.Web, :view\n import Dissolver.HTML end\n```\nGenerate the links using the view helper in your template\n```\n<%= paginate @conn, @paginator %>\n```\nImporting [`Dissolver.HTML`](Dissolver.HTML.html) provides your template access to [`Dissolver.HTML.paginate/3`](Dissolver.HTML.html#paginate/3) as the prior example shows.\nThe [`Dissolver.HTML.paginate/3`](Dissolver.HTML.html#paginate/3) can take a host of options to aid the theme from how many links to show (`window: integer`) to what the link labels should read.\n\nBy default the theme used to generate the html is the [`Dissolver.HTML.Simple`](Dissolver.HTML.Simple.html) theme.\nIt will only provide the very basic prev|next buttons. For more theme options, including providing your own, read the following [Configurations](#module-configuration)\n\n Configuration\n---\n\nThis module uses the following that can be set as globals in your `config/config.ex` configurations\n\n* `:repo` - **Required** Your app's Ecto Repo\n* `:theme` (default: [`Dissolver.HTML.Simple`](Dissolver.HTML.Simple.html)) - A module that implements the [`Dissolver.HTML.Theme`](Dissolver.HTML.Theme.html) behavior\n There are a few pre defiend theme modules found in `dessolver/html/`\n\n\t+ [`Dissolver.HTML.Simple`](Dissolver.HTML.Simple.html) - This is the *default* with only previous | next links\n\t+ [`Dissolver.HTML.Bootstrap`](Dissolver.HTML.Bootstrap.html) - [A Bootstrap 4 theme](https://getbootstrap.com/)\n\t+ [`Dissolver.HTML.Foundation`](Dissolver.HTML.Foundation.html) - [A Foundation theme](https://get.foundation/)\n\t+ [`Dissolver.HTML.Materialize`](Dissolver.HTML.Materialize.html) - [A Materialize theme](https://materializecss.com/)\n\t+ [`Dissolver.HTML.Semantic`](Dissolver.HTML.Semantic.html) - [A Semantic UI theme](https://semantic-ui.com/)\n\t+ [`Dissolver.HTML.Tailwind`](Dissolver.HTML.Tailwind.html) - [A Tailwind CSS theme](https://tailwindcss.com/)\n* `:per_page` (default: 20) - The global per page setting\n* `:max_page` - The limit of pages allow to navigate regardless of total pages found\n This option is ignored if not provided and defaults to total pages found in the query.\n* `:lazy` (default: false) - This option if enabled will result in all [`Dissolver.paginate/3`](Dissolver.html#paginate/3) calls\n return an Ecto.Query rather than call Repo.all. This is useful for when you need to paginate\n on an association via a preload. TODO: provide example.\n\n JSON API Support.\n---\n```\ndefmodule MyApp.ProductView do\n use MyApp.Web, :view\n import Dissolver.JSON\n\n def render(\"index.json\", %{products: products, dissolver: dissolver, conn: conn}) do\n %{data: render_many(products, MyApp.ProductView, \"product.json\"),\n pagination: paginate(conn, dissolver)}\n end\n\n def render(\"product.json\", %{product: product}) do\n %{id: product.id,\n name: product.name,\n description: product.description,\n price: product.price}\n end end\n```\n Lazy Example.\n---\n\nSay you have a tag that has many posts and you want to paginate the post as they relate to a given tag. \n\nHeres an example of using the lazy option so that Dissolver.paginate returns a query rather than a result. \n```\ndef get_post_for_tag!(tag_name, params \\ %{}) do\n total_count =\n from(t in \"tags\",\n join: pt in \"post_tags\",\n on: pt.tag_id == t.id,\n where: t.name == ^tag_name,\n select: count()\n )\n |> Repo.one()\n\n {posts_query, paginator} =\n from(p in Post, order_by: [desc: :inserted_at], preload: [:tags])\n |> Dissolver.paginate(params, total_count: total_count, lazy: true)\n\n tag =\n from(t in Tag, where: t.name == ^tag_name, preload: [posts: ^posts_query])\n |> Repo.one!()\n\n {tag, paginator}\n end\n```\nYou will notice that in this case I also have to supply total_count since Dissolver.paginate does not currently have a way to scope total_count of a given tag since they query is applied after Dissolver.paginate was called. So for this use case we just supply the total count up front. \nFor my controller it looks just like most.\n```\ndef show_post(conn, %{\"tag\" => tag} = params) do\n {tag, paginator} = Blog.get_post_for_tag!(tag, params)\n render(conn, \"show_posts.html\", tag: tag, paginator: paginator)\nend\n```\nYou can also send in options to paginate helper look at the docs for more details.\n\n Contributing\n---\n\nIf you can start by writing an issue ticket. \nThen if you like feel free to fork and submit a PR for review. \n\n Acknowledgement\n---\n\nI would like to Thank\n\n* Matt (@mgwidmann)\n* <NAME> (@drewolson)\n* <NAME> (@amatsuda)\n\n License\n---\n\nPlease take a look at LICENSE.md\n\n[Link to this section](#summary)\n Summary\n===\n\n[Functions](#functions)\n---\n\n[paginate(query, params, opts \\\\ [])](#paginate/3)\n\n[Link to this section](#functions)\nFunctions\n===\n\nDissolver.HTML\n===\n\nAn Html helper to render the pagination links,\n\nStart by importing the [`Dissolver.HTML`](#content) in your view module.\n```\ndefmodule MyApp.ProductView do\n use MyApp.Web, :view\n import Dissolver.HTML end\n```\nnow you have the [`paginate/2`](#paginate/2) view helper in your template file.\n```\n<%= paginate @conn, @paginator %>\n```\nWhere `@page` is a `%Dissolver.Paginator{}` struct returned from [`Dissolver.paginate/2`](Dissolver.html#paginate/2).\n\n`paginate` helper takes keyword list of `options` and `params`.\n <%= paginate @conn, @page, window: 5, next_label: \">>\", previous_label: \"<<\", first: true, last: true, first_label: \"First\", last_label: \"Last\" %[Link to this section](#summary)\n Summary\n===\n\n[Functions](#functions)\n---\n\n[paginate(paginator, route_helper, opts \\\\ [])](#paginate/3)\n\nGenerates the HTML pagination links for a given page returned by Dissolver.\n\n[Link to this section](#functions)\nFunctions\n===\n\nDissolver.HTML.Bootstrap\n===\n\nThis is a theme to support Bootstrap4\n<https://getbootstrap.com>/\n\nDissolver.HTML.Foundation\n===\n\nThis is a theme to support Foundation\n<https://get.foundation>/\n\nDissolver.HTML.Materialize\n===\n\nThis is a theme to support Materialize\n<https://materializecss.com>/\n\nDissolver.HTML.Semantic\n===\n\nThis is a theme to support Semantic UI.\n<https://semantic-ui.com>/\n\nDissolver.HTML.Simple\n===\n\nThis is a simple nav theme\n\nDissolver.HTML.Tailwind\n===\n\nThis is a theme to support Tailwind css.\n<https://tailwindcss.com>/\n\nDissolver.HTML.Theme behaviour\n===\n\nThis is a behavior for implementing a custom Dissolver.HTML.Theme\n\n[Link to this section](#summary)\n Summary\n===\n\n[Callbacks](#callbacks)\n---\n\n[generate_links(any, arg2)](#c:generate_links/2)\n\n[Link to this section](#callbacks)\nCallbacks\n===\n\nDissolver.JSON\n===\n\nJSON helpers to render the pagination links in json format.\nimport the [`Dissolver.JSON`](#content) in your view module.\n```\ndefmodule MyApp.ProductView do\n use MyApp.Web, :view\n import Dissolver.JSON\n\n def render(\"index.json\", %{conn: conn, products: products, paginator: paginator}) do\n %{data: render_many(products, MyApp.ProductView, \"product.json\"),\n pagination: paginate(conn, paginator)}\n end end\n```\nWhere `paginator` is a `%Dissolver.Paginator{}` struct returned from [`Dissolver.paginate/2`](Dissolver.html#paginate/2).\n\n`paginate` helper takes keyword list of `options`.\n paginate(paginator, window: 5, next_label: \">>\", previous_label: \"<<\", first: true, last: true, first_label: \"First\", last_label: \"Last\")\n\n[Link to this section](#summary)\n Summary\n===\n\n[Functions](#functions)\n---\n\n[paginate(conn, paginator, opts \\\\ [])](#paginate/3)\n\n[render_page_list(page_list)](#render_page_list/1)\n\n[Link to this section](#functions)\nFunctions\n===\n\nDissolver.Paginator\n===\n\nThis module is responsible for building the struct used for navigating to the respective pages of a given query\n\nThe HTML and JSON view helper will call into this module calling the [`paginate/3`](#paginate/3) passing the Plug.Conn, along with a Hydrated version of this struct.\n\n[Link to this section](#summary)\n Summary\n===\n\n[Types](#types)\n---\n\n[t()](#t:t/0)\n\n[Functions](#functions)\n---\n\n[build_options(opts)](#build_options/1)\n\n[paginate(conn, paginator, opts \\\\ [])](#paginate/3)\n\n[Link to this section](#types)\nTypes\n===\n\n[Link to this section](#functions)\nFunctions\n==="}}},{"rowIdx":59194,"cells":{"project":{"kind":"string","value":"tapkee"},"source":{"kind":"string","value":"cran"},"language":{"kind":"string","value":"R"},"content":{"kind":"string","value":"Package ‘tapkee’\n October 14, 2022\nType Package\nTitle Wrapper for 'tapkee' Dimension Reduction Library\nVersion 1.2\nDate 2020-12-20\nAuthor <NAME>\nMaintainer <NAME> <<EMAIL>>\nDescription Wrapper for using 'tapkee' command line utility,\n it allows to run it from inside R and catch the results for further analysis and plotting.\n 'Tapkee' is a program for fast dimension reduction, see 'package?tapkee' and <http:\n //tapkee.lisitsyn.me/>\n for installation and other details.\nSystemRequirements 'tapkee' (http://tapkee.lisitsyn.me/)\nSuggests scatterplot3d, rgl, R.rsp\nVignetteBuilder R.rsp\nLicense GPL (>= 2)\nLazyLoad yes\nNeedsCompilation no\nRepository CRAN\nDate/Publication 2021-01-04 12:00:02 UTC\nR topics documented:\ntapkee-packag... 2\nGen.dr.dat... 3\nTapke... 4\n tapkee-package ’tapkee’ command line utility installation\nDescription\n Here is the description of how to install ’tapkee’ utility on different operation systems.\nGeneral instructions\n Download Executablle files for macOS, Windows and Linux are available here: https://github.com/lisitsyn/tapkee/releases/ta\nSpecific instructions for Linux\n Downloaded binary: Linux binary is a 64 bit version which works on Ubuntu 16.04 LTS and\n likely will work on other systems. Depends on ‘ldd tapkee‘, install dependencies (you might\n need to install at least \"libarpack2\"). Copy ’tapkee’ binary to where system will find it (‘echo\n $PATH‘).\nSpecific instructions for macOS\n 1) In Terminal.app, run:\n $ echo $PATH\n 2) Copy ’tapkee’ into one of mentioned directories, e.g.:\n $ cp tapkee /usr/local/bin\n 3) Run:\n $ tapkee -h\n If you see the list of ’tapkee’ options, everything is OK. If not, you need to check (4) and (5).\n 4) If you have the message similar to:\n $ dyld: Library not loaded: ...\n install eigen and arpack with Homebrew (install Homebrew first, google how to do it):\n $ brew install arpack && brew install eigen\n then run\n $ tapkee -h\n again.\n 5) You might also run:\n $ echo $DYLD_LIBRARY_PATH\n to see where installed libraries should be located.\nSpecific instructions for Windows\n 1) Download the ’tapkee-win.exe’ executable which works under Windows 10. Please also make\n sure that you installed \"Microsoft Visual C++ Redistributable 32-bit\" (google the name to\n obtain link) to get required DLLs.\n 2) Find where did you place the executable file. Open command prompt and direct it to this folder,\n for example (if the ’tapkee-win.exe’ is in \"C:\\Users\\Me\\Downloads\" directory):\n > c: > cd \"C:\\Users\\Me\\Downloads\"\n Then check is the executable works. Type\n > tapkee-win.exe -h\n Long message with options should appear. If not, you are either not in the proper place (double\n check where is executable file and then \"cd\" to there), or there are some other problems (e.g.,\n required DLLs are not installed properly).\n Next, find the place where to install ’tapkee-win.exe’. Best is to start the command prompt\n window and run there:\n > echo\n Install the executable (and change its name to just ’tapkee.exe’) into one of folders which\n are in the list. For example, if \"C:\\Some dir\" is in the list (entries there are separated with\n semicolon), run:\n > copy tapkee-win.exe \"C:\\Some dir\\tapkee.exe\"\n You might also try to install everything in the R working folder (to find it, run ’getwd()’ inside\n R) but this is less usable.\n 3) Now open new command prompt window and run:\n > tapkee -h\n If you see the list of ’tapkee’ options, everything is OK. If not, you need to check (1) and (2)\n again.\n Gen.dr.data Generates 3D data\nDescription\n Generates typical 3D dimension reduction data\nUsage\n Gen.dr.data(type, N=1000)\nArguments\n type one of \"swissroll\", \"scurve\" (S-curve), \"helix\", \"ssphere\" (severed sphere)\n N number of data points\nDetails\n ’Gen.dr.data()’ generates some frequently used 3D data. Formulas taken partly from ’tapkee’\n ’borsch’ script and partly from Python ’scikit-learn’.\nAuthor(s)\n <NAME>\nExamples\n # generate four variants of 3D data\n SC <- Gen.dr.data(\"scurve\")\n SR <- Gen.dr.data(\"swissroll\")\n HX <- Gen.dr.data(\"helix\")\n SS <- Gen.dr.data(\"ssphere\")\n # plot them (requires packages 'scatterplot3d' and 'rgl')\n if (requireNamespace(\"rgl\", quietly = TRUE)) {\n COL <- colorRampPalette(c(\"green\", \"orange\"))(1000)\n scatterplot3d::scatterplot3d(SC, color=COL, pch=20, cex.symbols=1.4)\n } else {\n cat(\"Please install 'rgl' package to see the plot\")\n }\n if (requireNamespace(\"rgl\", quietly = TRUE)) {\n rgl::plot3d(SR, col=rainbow(1100))\n } else {\n cat(\"Please install 'rgl' package to see the plot\")\n }\n if (requireNamespace(\"rgl\", quietly = TRUE)) {\n rgl::plot3d(HX, col=rainbow(1100))\n } else {\n cat(\"Please install 'rgl' package to see the plot\")\n }\n if (requireNamespace(\"rgl\", quietly = TRUE)) {\n rgl::plot3d(SS, col=rainbow(1100))\n } else {\n cat(\"Please install 'rgl' package to see the plot\")\n }\n Tapkee Tapkee wrapper\nDescription\n R wrapper for the ’tapkee’ dimension reduction library\nUsage\n Tapkee(data, method=\"pca\", td=2, verbose=FALSE, add=\"\", prefix=\"Dim\", rm=TRUE)\nArguments\n data R numerical matrix or data frame (will be converted into matrix)\n method ’tapkee’ method, run \"system(’tapkee -h’)\" for the list, default is \"pca\"\n td Number of dimensions to output, default is 2\n verbose If TRUE, ’tapkee’ is verbose, defalut is FALSE\n add ’tapkee’ additional arguments as character string: see \"system(’tapkee -h’)\"\n prefix Variable name prefix in the resulted data frame, default is \"Dim\"\n rm Remove temp files (but temp folder will be removed anyway in the end of R\n session), default is TRUE\nDetails\n Interface (wrapper) for the ’tapkee’, flexible and efficient C++ template library for dimension re-\n duction. ’tapkee’ is extremely fast comparing with other DR tools.\n For methods used in ’tapkee’, run ’vignette(tapkee_methods)’.\n Users should install ’tapkee’ independently from author Web site (https://github.com/lisitsyn/tapkee)\n or associated GitHub (https://github.com/lisitsyn/tapkee). Run ’package?tapkee’ or help(\"tapkee-\n package\") for details related with your operation system. If ’tapkee’ is not installed, Tapkee() will\n fail gracefully and output the input data with warning.\n Please note that \"[warning] The neighborhood graph is not connected\" message in most cases means\n that ’tapkee’ run was unsuccessful. As a result, Tapkee() might return the matrix of NaN’s. One of\n possible workarounds is to specify the higher number of neigbors (’-k’ option, default is 10). See\n below for the example.\n Note that the wrapper catches only one (main) type of ’tapkee’ utility outputs. For other possible\n output types (see ’tapkee -h’ for explanation), run ’tapkee’ without wrapper.\nValue\n Data frame with number of columns equal to number of dimensions given and \"prefix\" column\n names prefixes.\nAuthor(s)\n <NAME>\nReferences\n <NAME> and <NAME> and <NAME>. Tapkee: An Efficient Di-\n mension Reduction Library. Journal of Machine Learning Research, 14: 2355-2359, 2013.\nSee Also\n tapkee-package\nExamples\n ## 'tapkee' vs. R base functions\n system.time(Tapkee(iris[, -5], method=\"mds\"))\n system.time(cmdscale(dist(iris[, -5])))\n ## How to use 'add' option\n plot(Tapkee(iris[, -5], \"isomap\", add=\"-k 47\"), col=iris[, 5])\n ## 'tapkee' methods as of March 2019:\n TM <- c(\n \"lle\", # 1) locally_linear_embedding (lle),\n \"npe\", # 2) neighborhood_preserving_embedding (npe),\n \"ltsa\", # 3) local_tangent_space_alignment (ltsa),\n \"lltsa\", # 4) linear_local_tangent_space_alignment (lltsa),\n \"hlle\", # 5) hessian_locally_linear_embedding (hlle),\n \"la\", # 6) laplacian_eigenmaps (la),\n \"lpp\", # 7) locality_preserving_projections (lpp),\n \"dm\", # 8) diffusion_map (dm),\n \"isomap\", # 9) isomap (isomap),\n \"l-isomap\", # 10) landmark_isomap (l-isomap),\n \"mds\", # 11) multidimensional_scaling (mds),\n \"l-mds\", # 12) landmark_multidimensional_scaling (l-mds),\n \"spe\", # 13) stochastic_proximity_embedding (spe),\n \"kpca\", # 14) kernel_pca (kpca),\n \"pca\", # 15) pca (pca),\n \"ra\", # 16) random_projection (ra),\n \"fa\", # 17) factor_analysis (fa),\n \"t-sne\", # 18) t-stochastic_neighborhood_embedding (t-sne),\n \"ms\") # 19) manifold_sculpting (ms)\n ## Iris example\n oldpar <- par(mfrow=c(4, 5), mar=c(1, 1, 3, 1), xaxt=\"n\", yaxt=\"n\")\n for (n in c(1:18)) {\n plot(Tapkee(iris[, -5], method=TM[n], add=\"-k 50\"),\n col=iris[, 5], pch=20, main=TM[n], xlab=\"\", ylab=\"\")\n }\n plot(iris[, 1:2], col=iris[, 5], pch=20, main=\"iris[, 1:2]\", xlab=\"\", ylab=\"\")\n par(oldpar)\n ## Generate typical 3D data\n SR <- Gen.dr.data(\"swissroll\")\n SC <- Gen.dr.data(\"scurve\")\n HX <- Gen.dr.data(\"helix\")\n SS <- Gen.dr.data(\"ssphere\")\n ## This will separate colors better\n COL <- rainbow(1100)[1:1000]\n ## 3D plot (if no 'scatterplot3d' package, plots XY axes)\n T3D <- function(dat, title, col=COL) {\n if (requireNamespace(\"scatterplot3d\", quietly = TRUE)) {\n scatterplot3d::scatterplot3d(dat, color=col, main=title, pch=20, xlab=\"\", ylab=\"\", zlab=\"\",\n axis=FALSE, tick.marks=FALSE, label.tick.marks=FALSE, mar=c(1, 1, 3, 1))\n } else {\n plot(dat[, 1:2], col=col, main=paste(title, \"(2D projection)\"), pch=20)\n warning(\"Please install 'scatterplot3d' package to see the 3D plot\")\n }}\n ## Swiss Roll\n oldpar <- par(mfrow=c(4, 5))\n T3D(SR, title=\"Swiss Roll\")\n for (n in 1:18) plot(Tapkee(SR, method=TM[n]), col=COL, pch=20, main=TM[n],\n xlab=\"\", ylab=\"\", xaxt=\"n\", yaxt=\"n\")\n par(oldpar)\n ## S-Curve\n oldpar <- par(mfrow=c(4, 5))\n T3D(SC, title=\"S-Curve\")\n for (n in 1:18) plot(Tapkee(SC, method=TM[n]), col=COL, pch=20, main=TM[n],\n xlab=\"\", ylab=\"\", xaxt=\"n\", yaxt=\"n\")\n par(oldpar)\n ## Helix\n oldpar <- par(mfrow=c(4, 5))\n T3D(HX, title=\"Helix\")\n for (n in 1:18) plot(Tapkee(HX, method=TM[n]), col=COL, pch=20,\n main=TM[n], xlab=\"\", ylab=\"\", xaxt=\"n\", yaxt=\"n\")\n par(oldpar)\n ## Severed Sphere\n oldpar <- par(mfrow=c(4, 5))\n T3D(SS, title=\"Severed Sphere\", col=rainbow(nrow(SS)))\n for (n in 1:18) plot(Tapkee(SS, method=TM[n]), col=rainbow(nrow(SS)), pch=20,\n main=TM[n], xlab=\"\", ylab=\"\", xaxt=\"n\", yaxt=\"n\")\n par(oldpar)"}}},{"rowIdx":59195,"cells":{"project":{"kind":"string","value":"@next-auth/upstash-redis-adapter"},"source":{"kind":"string","value":"npm"},"language":{"kind":"string","value":"JavaScript"},"content":{"kind":"string","value":"### \n**Upstash Redis Adapter** - NextAuth.js\n\n Open Source. Full Stack. Own Your Data.\n \n\nOverview\n---\n\nThis is the Upstash Redis adapter for [`next-auth`](https://authjs.dev). This package can only be used in conjunction with the primary `next-auth` and `@upstash/redis` packages. It is not a standalone package.\n\nGetting Started\n---\n\n1. Install `next-auth` and `@next-auth/upstash-redis-adapter` as well as `@upstash/redis` via NPM.\n```\nnpm install next-auth @next-auth/upstash-redis-adapter @upstash/redis\n```\n2. Add the following code to your `pages/api/[...nextauth].js` next-auth configuration object.\n```\nimport NextAuth from \"next-auth\"\nimport { UpstashRedisAdapter } from \"@next-auth/upstash-adapter\"\nimport { Redis } from \"@upstash/redis\"\n\nconst redis = new Redis({\n url:\"UPSTASH_REDIS_REST_URL\",\n token:\"UPSTASH_REDIS_REST_TOKEN\",\n})\n\n// For more information on each option (and a full list of options) go to\n// https://authjs.dev/reference/configuration/auth-options export default NextAuth({\n ...\n adapter: UpstashRedisAdapter(redis)\n ...\n})\n```\nUsing Multiple Apps with a Single Upstash Redis Instance\n---\n\nThe Upstash free-tier allows for only one Redis instance. If you have multiple Next-Auth connected apps using this instance, you need different key prefixes for every app.\n\nYou can change the prefixes by passing an `options` object as the second argument to the adapter factory function.\n\nThe default values for this object are:\n```\nconst defaultOptions = {\n baseKeyPrefix: \"\",\n accountKeyPrefix: \"user:account:\",\n accountByUserIdPrefix: \"user:account:by-user-id:\",\n emailKeyPrefix: \"user:email:\",\n sessionKeyPrefix: \"user:session:\",\n sessionByUserIdKeyPrefix: \"user:session:by-user-id:\",\n userKeyPrefix: \"user:\",\n verificationTokenKeyPrefix: \"user:token:\",\n}\n```\nUsually changing the `baseKeyPrefix` should be enough for this scenario, but for more custom setups, you can also change the prefixes of every single key.\n\nExample:\n```\nexport default NextAuth({\n ...\n adapter: UpstashRedisAdapter(redis, {baseKeyPrefix: \"app2:\"})\n ...\n})\n```\nContributing\n---\n\nWe're open to all community contributions! If you'd like to contribute in any way, please read our [Contributing Guide](https://github.com/nextauthjs/.github/blob/main/CONTRIBUTING.md).\n\nLicense\n---\n\nISC\nReadme\n---\n\n### Keywords\n\n* next-auth\n* next.js\n* oauth\n* upstash\n* redis"}}},{"rowIdx":59196,"cells":{"project":{"kind":"string","value":"kiwi.js"},"source":{"kind":"string","value":"npm"},"language":{"kind":"string","value":"JavaScript"},"content":{"kind":"string","value":"THIS REPOSITORY IS NO LONGER MAINTAINED\n===\n\nLooking for an alternative? Please have a look at [lume/kiwi](https://github.com/lume/kiwi).\n\nIf you are interested in maintaining this repository (and taking ownership of it), please reach out to me [here](mailto:<EMAIL>).\n\nkiwi.js\n===\n\nKiwi.js is a fast TypeScript implementation of the Cassowary constraint solving algorithm, based on the seminal Cassowary paper.\nOriginally created by <NAME>, it was redesigned from the ground up to be lightweight, fast and easy to maintain. View the [benchmarks](#benchmarks) to see how it compares to [Cassowary.js](https://github.com/slightlyoff/cassowary.js).\n\nIndex\n---\n\n* [THIS REPOSITORY IS NO LONGER MAINTAINED](#this-repository-is-no-longer-maintained)\n* [kiwi.js](#kiwijs)\n\t+ [Index](#index)\n\t+ [Getting started](#getting-started)\n\t+ [Documentation](#documentation)\n\t+ [Benchmarks](#benchmarks)\n\t+ [Tests](#tests)\n\t+ [Contribute](#contribute)\n\nGetting started\n---\n\nInstall using NPM:\n```\nnpm install kiwi.js\n```\nThe following example creates a solver which automatically calculates the width:\n```\nimport * as kiwi from 'kiwi.js';\nimport { equal } from 'assert';\n\n// Create a solver var solver = new kiwi.Solver();\n\n// Create edit variables var left = new kiwi.Variable();\nvar width = new kiwi.Variable();\nsolver.addEditVariable(left, kiwi.Strength.strong);\nsolver.addEditVariable(width, kiwi.Strength.strong);\nsolver.suggestValue(left, 100);\nsolver.suggestValue(width, 400);\n\n// Create and add a constraint var right = new kiwi.Variable();\nsolver.addConstraint(new kiwi.Constraint(new kiwi.Expression([-1, right], left, width), kiwi.Operator.Eq));\n\n// Solve the constraints solver.updateVariables();\nequal(right.value(), 500);\n```\nDocumentation\n---\n\n* [API Reference](https://github.com/IjzerenHein/kiwi.js/blob/HEAD/docs/Kiwi.md)\n\nBenchmarks\n---\n\nTo run the benchmark in the browser, [just visit this page](https://rawgit.com/IjzerenHein/kiwi/master/bench/index.html).\n\nTo run the benchmark locally using nodejs, *clone or download this repository* and execute the following steps:\n```\nnpm install npm run bench\n```\nTests\n---\n\nTo run the tests in the browser, [just visit this page](https://rawgit.com/IjzerenHein/kiwi/master/test/index.html).\n\nTo run the tests locally using nodejs, *clone or download this repository* and execute the following steps:\n```\nnpm install npm run build && npm run test\n```\nContribute\n---\n\nIf you like this project and want to support it, show some love and give it a star.\nReadme\n---\n\n### Keywords\n\n* cassowary\n* kiwi\n* kiwi.js\n* constraint solver\n* autolayout"}}},{"rowIdx":59197,"cells":{"project":{"kind":"string","value":"projmgr"},"source":{"kind":"string","value":"cran"},"language":{"kind":"string","value":"R"},"content":{"kind":"string","value":"Package ‘projmgr’\n October 14, 2022\nTitle Task Tracking and Project Management with GitHub\nVersion 0.1.0\nDescription Provides programmatic access to 'GitHub' API with a\n focus on project management. Key functionality includes\n setting up issues and milestones from R objects or 'YAML' configurations,\n querying outstanding or completed tasks, and generating progress updates\n in tables, charts, and RMarkdown reports. Useful for those using 'GitHub' in personal,\n professional, or academic settings with an emphasis on streamlining\n the workflow of data analysis projects.\nLicense MIT + file LICENSE\nURL https://github.com/emilyriederer/projmgr\nBugReports https://github.com/emilyriederer/projmgr/issues\nDepends R (>= 3.1.2)\nImports gh, magrittr\nSuggests clipr, curl, dplyr, ggplot2, knitr, purrr, reprex, rmarkdown,\n testthat, tidyr, yaml, htmltools, httr, covr\nEncoding UTF-8\nLanguage en-US\nLazyData true\nRoxygenNote 6.1.1\nNeedsCompilation no\nAuthor <NAME> [cre, aut]\nMaintainer <NAME> <<EMAIL>>\nRepository CRAN\nDate/Publication 2019-08-05 15:40:06 UTC\nR topics documented:\nbrowse_doc... 3\nbrowse_issue... 3\nbrowse_milestone... 4\nbrowse_rep... 5\ncheck_credential... 5\ncheck_interne... 6\ncheck_rate_limi... 6\ncreate_repo_re... 7\nget_issue... 8\nget_issue_comment... 9\nget_issue_event... 10\nget_milestone... 10\nget_repo_label... 11\nhel... 12\nlistcol_extrac... 13\nlistcol_filte... 14\nlistcol_pivo... 15\nparse_issue... 16\nparse_issue_comment... 17\nparse_issue_event... 17\nparse_milestone... 18\nparse_repo_label... 19\npost_issu... 20\npost_mileston... 21\npost_pla... 21\npost_tod... 22\nprojmg... 23\nread_pla... 23\nread_tod... 24\nreport_discussio... 25\nreport_pla... 26\nreport_progres... 27\nreport_taskboar... 28\nreport_tod... 29\ntaskboard_helper... 30\ntemplate_yam... 32\nviz_gant... 33\nviz_taskboar... 34\nviz_waterfal... 35\n browse_docs View GitHub API documentation\nDescription\n Opens browser to relevant parts of GitHub API documentation to learn more about field definitions\n and formatting. Inspired by similar browse_ functions included in the usethis package.\nUsage\n browse_docs(action = c(\"get\", \"post\"), object = c(\"milestone\", \"issue\",\n \"issue event\", \"issue comment\", \"repo labels\"))\nArguments\n action Character string denoting action you wish to complete: \"get\" (list existing) or\n \"post\" (creating new)\n object Character string denoting object on wish you want to apply an action. Supports\n \"milestone\", \"issue\", \"issue event\"\nValue\n Returns URL in non-interactive session or launches browser to docs in interactive session\nExamples\n ## Not run:\n browse_docs('get', 'milestone')\n ## End(Not run)\n browse_issues Browse issues for given GitHub repo\nDescription\n Opens browser to GitHub issues for a given repo. Inspired by similar browse_ functions included\n in the usethis package.\nUsage\n browse_issues(repo_ref, number = \"\")\nArguments\n repo_ref Repository reference as created by create_repo_ref()\n number Optional argument of issue number, if opening page for specific issue is desired\nValue\n Returns URL in non-interactive session or launches browser to docs in interactive session\nExamples\n ## Not run:\n my_repo <- create_repo_ref(\"repo_owner\", \"repo\")\n browse_issues(my_repo)\n ## End(Not run)\n browse_milestones Browse milestones for given GitHub repo\nDescription\n Opens browser to GitHub milestones for a given repo. Inspired by similar browse_ functions in-\n cluded in the usethis package.\nUsage\n browse_milestones(repo_ref, number = \"\")\nArguments\n repo_ref Repository reference as created by create_repo_ref()\n number Optional argument of milestone ID, if opening page for specific milestone is\n desired\nValue\n Returns URL in non-interactive session or launches browser to docs in interactive session\nExamples\n ## Not run:\n my_repo <- create_repo_ref(\"repo_owner\", \"repo\")\n browse_milestones(my_repo)\n ## End(Not run)\n browse_repo Browse a given GitHub repo\nDescription\n Opens browser to a given GitHub repo. Inspired by similar browse_ functions included in the\n usethis package.\nUsage\n browse_repo(repo_ref)\nArguments\n repo_ref Repository reference as created by create_repo_ref()\nValue\n Returns URL in non-interactive session or launches browser to docs in interactive session\nExamples\n ## Not run:\n my_repo <- create_repo_ref(\"repo_owner\", \"repo\")\n browse_repo(my_repo)\n ## End(Not run)\n check_credentials Check for valid credentials and repo permissions\nDescription\n Check for valid credentials and repo permissions\nUsage\n check_credentials(ref)\nArguments\n ref Any repository reference being used. Repository information is stripped out and\n only authentication credentials are validated.\nValue\n Prints GitHub username as determined by credentials (if valid) and repo-level permissions (if any),\n else throws 401 Unauthorized error.\nExamples\n ## Not run:\n experigit <- create_repo_ref('emilyriederer', 'experigit')\n check_authentication(experigit)\n ## End(Not run)\n check_internet Check internet connection (re-export of curl::has_internet())\nDescription\n Basic wrapper around curl::has_internet()\nUsage\n check_internet()\nValue\n Returns TRUE is connected to internet and false otherwise\nSee Also\n Other check: check_rate_limit\nExamples\n ## Not run:\n check_internet()\n ## End(Not run)\n check_rate_limit Find requests remaining and reset time\nDescription\n Source: copied from httr vignette \"Best practices for API packages\" by <NAME>\nUsage\n check_rate_limit(ref)\nArguments\n ref Any repository reference being used. Repository information is stripped out and\n only authentication credentials are used to determine the rate limit.\nValue\n Informative message on requests remaining and reset time\nSee Also\n Other check: check_internet\nExamples\n ## Not run:\n experigit <- create_repo_ref('emilyriederer', 'experigit')\n check_rate_limit(experigit)\n ## End(Not run)\n create_repo_ref Create reference to a GitHub repository\nDescription\n This function constructs a list of needed information to send API calls to a specific GitHub repos-\n itory. Specifically, it stores information on the repository’s name and owner, the type (whether or\n not Enterprise GitHub), and potentially credentials to authenticate.\nUsage\n create_repo_ref(repo_owner, repo_name, is_enterprise = FALSE,\n hostname = \"\", identifier = \"\")\nArguments\n repo_owner Repository owner’s username or GitHub Organization name\n repo_name Repository name\n is_enterprise Boolean denoting whether or not working with Enterprise GitHub.Defaults to\n FALSE\n hostname Host URL stub for Enterprise repositories (e.g. \"mycorp.github.com\")\n identifier Ideally should be left blank and defaults to using GITHUB_PAT or GITHUB_ENT_PAT\n environment variables as Personal Access Tokens. If identifier, this is as-\n sumed to be an alternative name of the environment variable to use for your\n Personal Access Token\nDetails\n Note that this package can be used for GET requests on public repositories without any authentica-\n tion (resulting in a lower rate limit.) To do this, simply pass any string into identifier that is not an\n environment variable already defined for your system (e.g. accessible through Sys.getenv(\"MY_VAR\"))\nValue\n List of repository reference information and credentials\nExamples\n ## Not run:\n myrepo <- create_repo_ref('emilyriederer', 'myrepo')\n ## End(Not run)\n get_issues Get issues from GitHub repository\nDescription\n A single issue can be obtained by identification number of number is passed through ...s. In this\n case, all other query parameters will be ignored.\nUsage\n get_issues(ref, limit = 1000, ...)\nArguments\n ref Repository reference (list) created by create_repo_ref()\n limit Number of records to return, passed directly to gh documentation. Defaults to\n 1000 and provides message if number of records returned equals the limit\n ... Additional user-defined query parameters. Use browse_docs() to learn more.\nValue\n Content of GET request as list\nSee Also\n Other issues: get_issue_comments, get_issue_events, parse_issue_comments, parse_issue_events,\n parse_issues, post_issue, report_discussion, report_progress, viz_waterfall\nExamples\n ## Not run:\n myrepo <- create_repo_ref('emilyriederer', 'myrepo')\n issues_res <- get_issues(myrepo)\n issues <- parse_issues(issues_res)\n ## End(Not run)\n get_issue_comments Get comments for a specific issue from GitHub repository\nDescription\n In addition to information returned by GitHub API, appends field \"number\" for the issue number to\n which the returned comments correspond.\nUsage\n get_issue_comments(ref, number, ...)\nArguments\n ref Repository reference (list) created by create_repo_ref()\n number Number of issue\n ... Additional user-defined query parameters. Use browse_docs() to learn more.\nValue\n Content of GET request as list\nSee Also\n Other issues: get_issue_events, get_issues, parse_issue_comments, parse_issue_events,\n parse_issues, post_issue, report_discussion, report_progress, viz_waterfall\n Other comments: parse_issue_comments, report_discussion\nExamples\n ## Not run:\n myrepo <- create_repo_ref('emilyriederer', 'myrepo')\n comments_res <- get_issue_comments(myrepo, number = 1)\n comments <- parse_issue_comments(comments_res)\n ## End(Not run)\n get_issue_events Get events for a specific issue from GitHub repository\nDescription\n In addition to information returned by GitHub API, appends field \"number\" for the issue number to\n which the returned events correspond.\nUsage\n get_issue_events(ref, number)\nArguments\n ref Repository reference (list) created by create_repo_ref()\n number Number of issue\nValue\n Content of GET request as list\nSee Also\n Other issues: get_issue_comments, get_issues, parse_issue_comments, parse_issue_events,\n parse_issues, post_issue, report_discussion, report_progress, viz_waterfall\n Other events: parse_issue_events\nExamples\n ## Not run:\n myrepo <- create_repo_ref('emilyriederer', 'myrepo')\n events_res <- get_issue_events(myrepo, number = 1)\n events <- parse_issue_events(events_res)\n ## End(Not run)\n get_milestones Get milestones from GitHub repository\nDescription\n A single milestone can be obtained by identification number of number is passed through ...s. In\n this case, all other query parameters will be ignored.\nUsage\n get_milestones(ref, ...)\nArguments\n ref Repository reference (list) created by create_repo_ref()\n ... Additional user-defined query parameters. Use browse_docs() to learn more.\nValue\n Content of GET request as list\nSee Also\n Other milestones: parse_milestones, post_milestone\nExamples\n ## Not run:\n myrepo <- create_repo_ref(\"emilyriederer\", \"myrepo\")\n milestones_res <- get_milestones(myrepo)\n milestones <- parse_milestones(milestones_res)\n ## End(Not run)\n get_repo_labels Get all labels for a repository\nDescription\n Get all labels for a repository\nUsage\n get_repo_labels(ref)\nArguments\n ref Repository reference (list) created by create_repo_ref()\nValue\n Content of GET request as list\nSee Also\n Other labels: parse_repo_labels\nExamples\n ## Not run:\n labels_res <- get_repo_labels(my_repo)\n labels <- parse_repo_labels(labels_res)\n ## End(Not run)\n help Learn about optional fields for related get_ functions\nDescription\n The help family of functions lists the optional query parameters available for each of the related\n get_ functions. When no optional arguments are available, a blank character vector is returned.\nUsage\n help_get_issues()\n help_get_issue_events()\n help_get_issue_comments()\n help_get_milestones()\n help_get_repo_label()\n help_post_issue()\n help_post_milestone()\nDetails\n For more details on these parameters, please use the browse_docs() function to navigate to the\n appropriate part of the GitHub API documentation.\nValue\n Character string of optional field names\nExamples\n help_get_issues()\n help_get_milestones()\n listcol_extract Extract new dataframe column from list-column matching pattern\nDescription\n Creaes a new column in your dataframe based on a subset of list-column values following a certain\n patten. For example, this is useful if you have labels you always apply to a repository with a set\n structure, e.g. key-value pairs like \"priority:high\", \"priority:medium\", and \"priority:low\" or other\n structures like \"engagement-team\", \"teaching-team\", etc. This function could create a new variable\n (e.g. \"priority\", \"team\") with the values encoded within the labels.\nUsage\n listcol_extract(data, col_name, regex, new_col_name = NULL,\n keep_regex = FALSE)\nArguments\n data Dataframe containing a list column (e.g. an issues dataframe)\n col_name Character string containing column name of list column (e.g. labels_name or\n assignees_login)\n regex Character string of regular expression to identify list items of interest (e.g. \"^priority:\",\n \"^(bug|feature)$\")\n new_col_name Optional name of new column. Otherwise regex is used, stripped of any leading\n or trailing punctuation\n keep_regex Optional logical denoting whether to keep regex part of matched item in value.\n Defaults to FALSE\nDetails\n This function works only if each observatino contains at most one instance of a given patterns.\n When multiple labels match the same pattern, one is returned at random.\nValue\n Dataframe with new column taking values extracted from list column\nExamples\n ## Not run:\n issues <- get_issues(repo)\n issues_df <- parse_issues(issues)\n listcol_extract(issues_df, \"labels_name\", \"-team$\")\n ## End(Not run)\n listcol_filter Filter dataframe by list-column elements\nDescription\n Some outputs of the get_ and parse_ functions contain list-columns (e.g. the labels column in the\n issues dataframe). This is an efficient way to represent the provided information, but may make\n certain information seem slightly inaccessible. This function allows users to filter list columns by\n the presence of one or more values or a regular expression.\nUsage\n listcol_filter(data, col_name, matches, is_regex = FALSE)\nArguments\n data Dataframe containing a list column (e.g. an issues dataframe)\n col_name Character string containing column name of list column (e.g. labels_name or\n assignees_login)\n matches A character vector containing a regular expression or one or more exact-match\n values. An observation will be kept in the returned data if any of the\n is_regex Logical to indicate whether charalcter indicates a regular expression or specific\n values\nValue\n Dataframe containing only rows in which list-column contains element matching provided criteria\nExamples\n ## Not run:\n issues <- get_issues(repo)\n issues_df <- parse_issues(issues)\n # keep observation containing a label of either \"bug\" or \"feature\"\n listcol_filter(issues_df, col_name = \"labels_name\", matches = c(\"bug\", \"feature\"))\n # keep observation containing a label that starts with \"region\"\n listcol_filter(issues_df, col_name = \"labels_name\", matches = \"^region:\", is_regex = TRUE)\n ## End(Not run)\n listcol_pivot Pivot list-column elements to indicator variables\nDescription\n Some outputs of the get_ and parse_ functions contain list-columns (e.g. the labels column in\n the issues dataframe). This is an efficient way to represent the provided information, but may\n make certain information seem slightly inaccessible. This function allows users to \"pivot\" these\n list columns and instead create a seperate indicator variable to represent the presence or absence of\n matches within the list column.\nUsage\n listcol_pivot(data, col_name, regex = \".\", transform_fx = identity,\n delete_orig = FALSE)\nArguments\n data Dataframe containing a list column (e.g. an issues dataframe)\n col_name Character string containing column name of list column (e.g. labels_name or\n assignees_login)\n regex Character string of regular expression to identify list items of interest (e.g. \"^priority:\",\n \"^(bug|feature)$\")\n transform_fx Function to transform label name before converting to column (e.g. sub(\":\",\n \"_\"))\n delete_orig Logical denoting whether or not to delete original list column provided by col_name\nDetails\n For example, if a repository tags issues with \"priority:high\", \"priority:medium\", and \"priority:low\"\n along with other labels, this function could be used to create separate \"high\", \"medium\", and\n \"low\" columns to denote different issue severities. This could be done with listcol_pivot(data,\n \"labels_name\", \"^priority:\", function(x) sub(\"^priority:\", \"\"))\nValue\n Dataframe additional logical columns denoting absence / presence of specified list-column elements\nExamples\n ## Not run:\n issues <- get_issues(repo)\n issues_df <- parse_issues(issues)\n listcol_pivot(issues_df,\n col_name = \"labels_name\",\n regex = \"^priority:\",\n transform_fx = function(x) paste0(\"label_\",x),\n delete_orig = TRUE)\n ## End(Not run)\n parse_issues Parse issues overview from get_issues\nDescription\n Parse issues overview from get_issues\nUsage\n parse_issues(res)\nArguments\n res List returned by corresponding get_ function\nValue\n data.frame with one record / issue\nSee Also\n Other issues: get_issue_comments, get_issue_events, get_issues, parse_issue_comments,\n parse_issue_events, post_issue, report_discussion, report_progress, viz_waterfall\nExamples\n ## Not run:\n myrepo <- create_repo_reference('emilyriederer', 'myrepo')\n issues_res <- get_issues(myrepo)\n issues <- parse_issues(issues_res)\n ## End(Not run)\n parse_issue_comments Parse issue comments from get_issues_comments\nDescription\n Parse issue comments from get_issues_comments\nUsage\n parse_issue_comments(res)\nArguments\n res List returned by corresponding get_ function\nValue\n Dataframe with one record / issue-comment\nSee Also\n Other issues: get_issue_comments, get_issue_events, get_issues, parse_issue_events,\n parse_issues, post_issue, report_discussion, report_progress, viz_waterfall\n Other comments: get_issue_comments, report_discussion\nExamples\n ## Not run:\n myrepo <- create_repo_ref('emilyriederer', 'myrepo')\n comments_res <- get_issue_comments(myrepo, number = 1)\n comments <- parse_issue_comments(comments_res)\n ## End(Not run)\n parse_issue_events Parse issue events from get_issues_events\nDescription\n This function convert list output returned by get into a dataframe. Due to the diverse fields for\n different types of events, many fields in the dataframe may be NA.\nUsage\n parse_issue_events(res)\nArguments\n res List returned by corresponding get_ function\nDetails\n Currently, the following event types are unsupported (with regard to processing all of their fields)\n due to their additional bulk and limited utility with respect to this packages functionality. Please file\n an issue if you disagree:\n • \"(removed_from/moved_columns_in/added_to)_project\"Since this package has limited value\n with GitHub projects\n • \"converted_note_to_issue\"Since issue lineage is not a key concern\n • \"head_ref_(deleted/restored)\"Since future support for pull requests would likely be handled\n separately\n • \"merged\"Same justification as head_ref\n • \"review_(requested/dismissed/request_removed)Same justification as head_ref\nValue\n Dataframe with one record / issue-event\nSee Also\n Other issues: get_issue_comments, get_issue_events, get_issues, parse_issue_comments,\n parse_issues, post_issue, report_discussion, report_progress, viz_waterfall\n Other events: get_issue_events\nExamples\n ## Not run:\n myrepo <- create_repo_ref('emilyriederer', 'myrepo')\n events_res <- get_issue_events(myrepo, number = 1)\n events <- parse_issue_events(events_res)\n ## End(Not run)\n parse_milestones Parse milestones from get_milestones\nDescription\n Parse milestones from get_milestones\nUsage\n parse_milestones(res)\nArguments\n res List returned by corresponding get_ function\nValue\n Dataframe with one record / milestone\nSee Also\n Other milestones: get_milestones, post_milestone\nExamples\n ## Not run:\n myrepo <- create_repo_ref(\"emilyriederer\", \"myrepo\")\n milestones_res <- get_milestones(myrepo)\n milestones <- parse_milestones(milestones_res)\n ## End(Not run)\n parse_repo_labels Parse labels from get_repo_labels\nDescription\n Parse labels from get_repo_labels\nUsage\n parse_repo_labels(res)\nArguments\n res List returned by corresponding get_ function\nValue\n Dataframe with one record / label\nSee Also\n Other labels: get_repo_labels\nExamples\n ## Not run:\n labels_res <- get_repo_labels(my_repo)\n labels <- parse_repo_labels(labels_res)\n ## End(Not run)\n post_issue Post issue to GitHub repository\nDescription\n Post issue to GitHub repository\nUsage\n post_issue(ref, title, ..., distinct = TRUE)\nArguments\n ref Repository reference (list) created by create_repo_ref()\n title Issue title (required)\n ... Additional user-defined body parameters. Use browse_docs() to learn more.\n distinct Logical value to denote whether issues with the same title as a current open issue\n should be allowed\nValue\n Number (identifier) of posted issue\nSee Also\n Other issues: get_issue_comments, get_issue_events, get_issues, parse_issue_comments,\n parse_issue_events, parse_issues, report_discussion, report_progress, viz_waterfall\nExamples\n ## Not run:\n myrepo <- create_repo_ref('emilyriederer', 'myrepo')\n post_issue(myrepo,\n title = 'this is the issue title',\n body = 'this is the issue body',\n labels = c('priority:high', 'bug'))\n ## End(Not run)\n ## Not run:\n # can be used in conjunction with reprex pkg\n # example assumes code for reprex is on clipboard\n reprex::reprex(venue = \"gh\")\n post_issue(myrepo,\n title = \"something is broken\",\n body = paste( clipr::read_clip(), collapse = \"\\n\") )\n ## End(Not run)\n post_milestone Post milestone to GitHub repository\nDescription\n Post milestone to GitHub repository\nUsage\n post_milestone(ref, title, ...)\nArguments\n ref Repository reference (list) created by create_repo_ref()\n title Milestone title (required)\n ... Additional user-defined body parameters. Use browse_docs() to learn more.\nValue\n Number (identifier) of posted milestone\nSee Also\n Other milestones: get_milestones, parse_milestones\nExamples\n ## Not run:\n myrepo <- create_repo_ref('emilyriederer', 'myrepo')\n post_milestone(myrepo,\n title = 'this is the title of the milestone',\n description = 'this is the long and detailed description',\n due_on = '2018-12-31T12:59:59z')\n ## End(Not run)\n post_plan Post plan (milestones + issues) to GitHub repository\nDescription\n Post custom plans (i.e. create milestones and issues) based on yaml read in by read_plan. Please\n see the \"Building Custom Plans\" vignette for details.\nUsage\n post_plan(ref, plan, distinct = TRUE)\nArguments\n ref Repository reference (list) created by create_repo_ref()\n plan Plan list as read with read_plan()\n distinct Logical value to denote whether issues with the same title as a current open issue\n should be allowed. Passed to get_issues()\nValue\n Dataframe with numbers (identifiers) of posted milestones and issues and issue title\nSee Also\n Other plans and todos: post_todo, read_plan, read_todo, report_plan, report_todo, template_yaml\nExamples\n ## Not run:\n # This example uses example file included in pkg\n # You should be able to run example as-is after creating your own repo reference\n file_path <- system.file(\"extdata\", \"plan.yml\", package = \"projmgr\", mustWork = TRUE)\n my_plan <- read_plan(file_path)\n post_plan(ref, my_plan)\n ## End(Not run)\n post_todo Post to-do list (issues) to GitHub repository\nDescription\n Post custom to-do lists (i.e. issues) based on yaml read in by read_todo. Please see the \"Building\n Custom Plans\" vignette for details.\nUsage\n post_todo(ref, todo, distinct = TRUE)\nArguments\n ref Repository reference (list) created by create_repo_ref()\n todo To-do R list structure as read with read_todo()\n distinct Logical value to denote whether issues with the same title as a current open issue\n should be allowed. Passed to get_issues()\nDetails\n Currently has know bug in that cannot be used to introduce new labels.\nValue\n Number (identifier) of posted issue\nSee Also\n Other plans and todos: post_plan, read_plan, read_todo, report_plan, report_todo, template_yaml\nExamples\n ## Not run:\n # This example uses example file included in pkg\n # You should be able to run example as-is after creating your own repo reference\n file_path <- system.file(\"extdata\", \"todo.yml\", package = \"projmgr\", mustWork = TRUE)\n my_todo <- read_todo(file_path)\n post_todo(ref, my_todo)\n ## End(Not run)\n projmgr projmgr package\nDescription\n A quick and easy wrapper for working with GitHub and other project management tools\nDetails\n See the README on GitHub\n read_plan Read plan from YAML\nDescription\n This function reads a carefully constructed YAML file representing a project plan (of milestones\n and issues). YAML is converted into an R list structure which can then be passed to post_plan()\n to build infrastructure for your repository.\nUsage\n read_plan(input)\nArguments\n input Either filepath to YAML file or character string. Assumes filepath if ends in\n \".yml\" and assumes string otherwise.\nDetails\n Please see the \"Building Custom Plans\" vignette for more details.\nValue\n List containing plan compatible with post_plan() or post_todo()\nSee Also\n Other plans and todos: post_plan, post_todo, read_todo, report_plan, report_todo, template_yaml\nExamples\n ## Not run:\n # This example uses example file included in pkg\n # You should be able to run example as-is after creating your own repo reference\n file_path <- system.file(\"extdata\", \"plan.yml\", package = \"projmgr\", mustWork = TRUE)\n my_plan <- read_plan(file_path)\n post_plan(ref, my_plan)\n ## End(Not run)\n read_todo Read to-do list from YAML\nDescription\n This function reads a carefully constructed YAML file representing a to-do list (of issues). YAML\n is converted into an R list structure which can then be passed to post_todo() to build infrastructure\n for your repository.\nUsage\n read_todo(input)\nArguments\n input Either filepath to YAML file or character string. Assumes filepath if ends in\n \".yml\" and assumes string otherwise.\nDetails\n Please see the \"Building Custom Plans\" vignette for more details.\nValue\n List containing plan compatible with post_plan() or post_todo()\nSee Also\n Other plans and todos: post_plan, post_todo, read_plan, report_plan, report_todo, template_yaml\nExamples\n ## Not run:\n # This example uses example file included in pkg\n # You should be able to run example as-is after creating your own repo reference\n file_path <- system.file(\"extdata\", \"todo.yml\", package = \"projmgr\", mustWork = TRUE)\n my_todo <- read_todo(file_path)\n post_todo(ref, my_todo)\n ## End(Not run)\n report_discussion Print issue comments in RMarkdown friendly way\nDescription\n Interprets dataframe or tibble of issues by breaking apart milestones and listing each issue title as\n open or closed, and uses HTML to format results in a highly readable and attractive way. Resulting\n object returned is a character vector of HTML code with the added class of 'knit_asis' so that\n when included in an RMarkdown document knitting to HTML, the results will be correctly rendered\n as HTML.\nUsage\n report_discussion(comments, issue = NA, link_url = TRUE)\nArguments\n comments Dataframe or tibble of comments for a single issue, as returned by get_issue_comments()\n issue Optional dataframe or tibble of issues, as returned by get_issues(). If pro-\n vided, output includes issue-level data such as the title, initial description, cre-\n ation date, etc.\n link_url Boolean. Whether or not to provide link to each item, as provided by url column\n in dataset\nDetails\n HTML output is wrapped in a <div> of class ’report_disccusion’ for custom CSS styling.\nValue\n Returns character string of HTML with class attribute to be correctly shown \"as-is\" in RMarkdown\nSee Also\n Other issues: get_issue_comments, get_issue_events, get_issues, parse_issue_comments,\n parse_issue_events, parse_issues, post_issue, report_progress, viz_waterfall\n Other comments: get_issue_comments, parse_issue_comments\nExamples\n ## Not run:\n # the following could be run in RMarkdown\n repo <- create_repo_ref(\"emilyriederer\", \"projmgr\")\n issue <- get_issues(repo, number = 15)\n issue_df <- parse_issues(issue)\n comments <- get_issue_comments(repo, number = 15)\n comments_df <- parse_issue_comments(comments)\n report_discussion(issue_df, comments_df)\n ## End(Not run)\n report_plan Print plan in RMarkdown friendly way\nDescription\n Interprets list representation of plan, using HTML to format results in a highly readable and at-\n tractive way. Resulting object returned is a character vector of HTML code with the added class\n of 'knit_asis' so that when included in an RMarkdown document knitting to HTML, the results\n will be correctly rendered as HTML.\nUsage\n report_plan(plan, show_ratio = TRUE)\nArguments\n plan List of project plan, as returned by read_plan()\n show_ratio Boolean. Whether or not to report (# Closed Items / # Total Items) for each\n group as a ratio\nDetails\n The resulting HTML unordered list (<ul>) is tagged with class ’report_plan’ for custom CSS styling.\nValue\n Returns character string of HTML with class attribute to be correctly shown \"as-is\" in RMarkdown\nSee Also\n Other plans and todos: post_plan, post_todo, read_plan, read_todo, report_todo, template_yaml\nExamples\n ## Not run:\n # the following could be run in RMarkdown\n plan_path <- system.file(\"extdata\", \"plan-ex.yml\", package = \"projmgr\", mustWork = TRUE)\n my_plan <- read_plan(plan_path)\n report_plan(my_plan)\n ## End(Not run)\n report_progress Print issue-milestone progress in RMarkdown friendly way\nDescription\n Interprets dataframe or tibble of items (e.g. issues) by breaking apart groups (e.g. milestones),\n listing each item title as open or closed, and using HTML to format results in a highly readable and\n attractive way. Resulting object returned is a character vector of HTML code with the added class\n of 'knit_asis' so that when included in an RMarkdown document knitting to HTML, the results\n will be correctly rendered as HTML.\nUsage\n report_progress(issues, group_var = \"milestone_title\", link_url = TRUE,\n show_ratio = TRUE, show_pct = TRUE)\nArguments\n issues Dataframe or tibble of issues and milestones, as returned by get_issues() and\n parse_issues()\n group_var Character string variable name by which to group issues. Defaults to \"milestone_title\"\n link_url Boolean. Whether or not to provide link to each item, as provided by url column\n in dataset\n show_ratio Boolean. Whether or not to report (# Closed Items / # Total Items) for each\n group as a ratio\n show_pct Boolean. Whether or not to report (# Closed Items / # Total Items) for each\n group as a percent\nDetails\n The resulting HTML unordered list (<ul>) is tagged with class ’report_progress’ for custom CSS\n styling.\n Items without a related group are put into an \"Ungrouped\" category. Filter these out before using\n this function if you wish to only show items that are in a group.\nValue\n Returns character string of HTML with class attribute to be correctly shown \"as-is\" in RMarkdown\nSee Also\n Other issues: get_issue_comments, get_issue_events, get_issues, parse_issue_comments,\n parse_issue_events, parse_issues, post_issue, report_discussion, viz_waterfall\nExamples\n ## Not run:\n repo <- create_repo_ref(\"emilyriederer\", \"projmgr\")\n issues <- get_issues(repo, state = 'all')\n issues_df <- parse_issues(issues)\n report_progress(issues_df)\n ## End(Not run)\n report_taskboard Report HTML-based task board of item status\nDescription\n Produces three column task board showing any relevant objects (typically issues or milestones) as\n \"Not Started\", \"In Progress\", or \"Done\".\nUsage\n report_taskboard(data, in_progress_when, include_link = FALSE,\n hover = FALSE, colors = c(\"#f0e442\", \"#56b4e9\", \"#009e73\"))\nArguments\n data Dataset, such as those representing issues or milestones (i.e. from parse_issues()).\n Must have state variable.\n in_progress_when\n Function with parameter data that returns Boolean vector. Generally, one of the\n taskboard helper functions. See ?taskboard_helpers for details.\n include_link Boolean whether or not to include links back to GitHub\n hover Boolean whether or not tasks should be animated to slightly enlarge on hover\n colors Character vector of hex colors for not started, in progress, and complete tasks\n (respectively)\nDetails\n The following logic is used to determine the status of each issue:\n • Done: Items with a state of \"closed\"\n • In Progress: Custom logic via in_progress_when. See ?taskboard_helpers for details.\n • Not Started: Default case for items neither In Progress or Closed\nValue\n Returns character string of HTML/CSS with class attribute to be correctly shown \"as-is\" in RMark-\n down\nExamples\n ## Not run:\n # in RMarkdown\n ```{r}\n issues <- get_issues(myrepo, milestone = 1)\n issues_df <- parse_issues(issues)\n report_taskboard(issues_df, in_progress_when = is_labeled_with('in-progress'))\n ````\n ## End(Not run)\n report_todo Print to-do lists in RMarkdown friendly way\nDescription\n Interprets list representation of to-do list, using HTML to format results in a highly readable and\n attractive way. Resulting object returned is a character vector of HTML code with the added class\n of 'knit_asis' so that when included in an RMarkdown document knitting to HTML, the results\n will be correctly rendered as HTML.\nUsage\n report_todo(todo, show_ratio = TRUE)\nArguments\n todo List of to-do list, as returned by read_todo()\n show_ratio Boolean. Whether or not to report (# Closed Items / # Total Items) for each\n group as a ratio\nDetails\n The resulting HTML unordered list (<ul>) is tagged with class ’report_todo’ for custom CSS\n styling.\nValue\n Returns character string of HTML with class attribute to be correctly shown \"as-is\" in RMarkdown\nSee Also\n Other plans and todos: post_plan, post_todo, read_plan, read_todo, report_plan, template_yaml\nExamples\n ## Not run:\n # the following could be run in RMarkdown\n todo_path <- system.file(\"extdata\", \"todo-ex.yml\", package = \"projmgr\", mustWork = TRUE)\n my_todo <- read_todo(todo_path)\n report_todo(my_todo)\n ## End(Not run)\n taskboard_helpers Tag \"in-progress\" items for taskboard visualization\nDescription\n The viz_taskboard() function creates a three-column layout of entities that are not started, in\n progress, or done. Objects are classified as done when they have a state of \"closed\". Object are\n classified as \"To-Do\" when they are neither \"Closed\" or \"In Progress\". However, what constistutes\n \"In Progress\" is user and project dependent. Thus, these functions let users specify what they mean.\nUsage\n is_labeled()\n is_labeled_with(label, any = TRUE)\n is_assigned()\n is_assigned_to(login, any = TRUE)\n is_in_a_milestone()\n is_in_milestone(number)\n is_created_before(created_date)\n is_part_closed()\n is_due()\n is_due_before(due_date)\n has_n_commits(events, n = 1)\nArguments\n label Label name(s) as character vector\n any When the supplied vector has more than one value, should the result return\n TRUE if any of those values are present in the dataset (logical OR)\n login User login(s) as character vector\n number Milestone number\n created_date Date as character in \"YYYY-MM-DD\" format\n due_date Date as character in \"YYYY-MM-DD\" format\n events Dataframe containing events for each issue in data\n n Minimum of commits required to be considered in progress\nDetails\n General options:\n • is_created_before: Was created before a user-specified data (as \"YYYY-MM-DD\" character\n string)\n Issue-specific options:\n • is_labeled_with: User-specified label (as character string) exists\n • is_assigned: Has been assigned to anyone\n • is_assigned_to: Has been assigned to specific user-specified login (as character string)\n • is_in_a_milestone: Has been put into any milestone\n • is_in_milestone: Has been put into a specific milestone\n Milestone-specific options:\n • is_part_closed: Has any of its issues closed\n • is_due: Has a due date\n • is_due_before: Has a due data by or before a user-specified date (as \"YYYY-MM-DD\" char-\n acter string)\nValue\n Function to be passed as in_progress_when argument in viz_taskboard()\nExamples\n ## Not run:\n viz_taskboard(issues, in_progress_when = is_labeled_with('in-progress'))\n viz_taskboard(milestones, in_progress_when = is_created_before('2018-12-31'))\n viz_taskboard(issues, in_progress_when = is_in_milestone())\n report_taskboard(issues, in_progress_when = is_labeled_with('in-progress'))\n report_taskboard(milestones, in_progress_when = is_created_before('2018-12-31'))\n report_taskboard(issues, in_progress_when = is_in_milestone())\n ## End(Not run)\n template_yaml Print YAML template to console\nDescription\n Prints YAML templates for either a plan or to-do list to the console as an example for developing\n your own custom plans and to-do lists. Inspired by similar template_ functions included in the\n pkgdown package.\nUsage\n template_yaml(template = c(\"plan\", \"todo\"))\nArguments\n template One of \"plan\" or \"todo\" denoting template desired\nDetails\n Note that depending on the console, text editor, and settings you are using, the template may or\n may not preserve the necessary whitespace shown in the output. If you copy-paste the template for\n modification, ensure that it still adheres to traditional YAML indentation.\nValue\n Prints template to console\nSee Also\n Other plans and todos: post_plan, post_todo, read_plan, read_todo, report_plan, report_todo\nExamples\n template_yaml('plan')\n template_yaml('todo')\n viz_gantt Visualize Gantt-style chart of planned or actual time to completion\nDescription\n Produces plot with one vertical bar from the specified start variable’s value to the end variable’s\n value. Common uses would be to visualize time-to-completion for issues gotten by (get_issues\n and parse_issues) or milestones. Bars are colored by duration with longer bars as a darker shade\n of blue, and start/completion is denoted by points at the ends of the bars.\nUsage\n viz_gantt(data, start = \"created_at\", end = \"closed_at\",\n str_wrap_width = 30)\nArguments\n data Dataset, such as those representing issues or milestones (i.e. parse_issues()\n or parse_milestones()). Must have unique title variable and variables to\n specify for start and end\n start Unquoted variable name denoting issue start date\n end Unquoted variable name denoting issue end date\n str_wrap_width Number of characters before text of issue title begins to wrap\nDetails\n By default, the start date is the issue’s created_at date, and the end date is the issue’s closed_at\n date. However, either of these can be altered via the start and end parameters since these dates\n might not be reflective of the true timeframe (e.g. if issues are posted well in advance of work\n beginning.)\n Unfinished tasks (where the value of the end variable is NA) are colored grey and do not have dots\n on their bars. Unstarted tasks are dropped because user intent is ambiguous in that case.\nValue\n ggplot object\nSee Also\n viz_linked\nExamples\n ## Not run:\n issues <- get_issues(myrepo, state = \"closed\")\n issues_df <- parse_issues(issues)\n viz_gantt(issues_df)\n ## End(Not run)\n viz_taskboard Visualize Agile-style task board of item status\nDescription\n Produces three column task board showing any relevant objects (typically issues or milestones) as\n \"Not Started\", \"In Progress\", or \"Done\".\nUsage\n viz_taskboard(data, in_progress_when, str_wrap_width = 30,\n text_size = 3)\nArguments\n data Dataset, such as those representing issues or milestones (i.e. from parse_issues()).\n Must have state variable.\n in_progress_when\n Function with parameter data that returns Boolean vector. Generally, one of the\n taskboard helper functions. See ?taskboard_helpers for details.\n str_wrap_width Number of characters before text of issue title begins to wrap\n text_size Text size\nDetails\n The following logic is used to determine the status of each issue:\n • Done: Items with a state of \"closed\"\n • In Progress: Custom logic via in_progress_when. See ?taskboard_helpers for details.\n • Not Started: Default case for items neither In Progress or Closed\nValue\n ggplot object\nSee Also\n viz_linked\nExamples\n ## Not run:\n issues <- get_issues(myrepo, milestone = 1)\n issues_df <- parse_issues(issues)\n viz_taskboard(issues_df, in_progress_when = is_labeled_with('in-progress'))\n viz_taskboard(issues_df, in_progress_when = is_in_a_milestone())\n ## End(Not run)\n viz_waterfall Visualize waterfall of opened, closed, and pending items over time-\n frame\nDescription\n Creates a four-bar waterfall diagram. Within the specified timeframe, shows initial, newly opened,\n newly closed, and final open counts. Works with either issues or milestones, as obtained by the get\n and parse functions.\nUsage\n viz_waterfall(data, start_date, end_date, start = \"created_at\",\n end = \"closed_at\")\nArguments\n data Dataset, such as those representing issues or milestones (i.e. parse_issues()\n or parse_milestones()). Must have state variable and variables to specify\n for start and end\n start_date Character string in ’YYYY-MM-DD’ form for first date to be considered (inclu-\n sive)\n end_date Character string in ’YYYY-MM-DD’ form for last date to be considered (inclu-\n sive)\n start Unquoted variable name denoting issue start date\n end Unquoted variable name denoting issue end date\nDetails\n The following logic is used to classify issues:\n • Initial: start < start_date and (end > start_date or state == 'open')\n • Open: start >= start_date and start <= end_date\n • Closed: end >= start_date and end <= end_date\n • Final: start < end_date and (end > end_date or state == 'open')\n The exact accuracy of the logic depends on filtering that has already been done to the dataset. Think\n carefully about the population you wish to represent when getting your data.\nValue\n ggplot object\nSee Also\n Other issues: get_issue_comments, get_issue_events, get_issues, parse_issue_comments,\n parse_issue_events, parse_issues, post_issue, report_discussion, report_progress\nExamples\n ## Not run:\n viz_waterfall(milestones, '2017-01-01', '2017-03-31')\n ## End(Not run)"}}},{"rowIdx":59198,"cells":{"project":{"kind":"string","value":"resampledata"},"source":{"kind":"string","value":"cran"},"language":{"kind":"string","value":"R"},"content":{"kind":"string","value":"Package ‘resampledata’\n October 14, 2022\nType Package\nTitle Data Sets for Mathematical Statistics with Resampling in R\nVersion 0.3.1\nSuggests resample (>= 0.4)\nMaintainer <NAME> <<EMAIL>>\nDescription Package of data sets from ``Mathematical Statistics\n with Resampling in R'' (1st Ed. 2011, 2nd Ed. 2018) by <NAME> and <NAME>.\nDepends R (>= 2.10)\nLicense CC0\nLazyData TRUE\nRoxygenNote 6.1.1\nURL https://github.com/rudeboybert/resampledata\nBugReports https://github.com/rudeboybert/resampledata/issues\nNeedsCompilation no\nAuthor <NAME> [aut],\n <NAME> [aut],\n <NAME> [aut, cre]\nRepository CRAN\nDate/Publication 2019-08-30 04:20:10 UTC\nR topics documented:\nAlelage... 3\nBanglades... 3\nBatters200... 4\nBeerwing... 4\nBookPrice... 5\nBushmea... 5\nBushmeat... 6\nCamera... 6\nCereal... 7\nChallenge... 7\nChiMarathonMe... 8\ncorrExercise... 8\ncorrExercise... 9\nCuckoo... 9\nDiving201... 10\nFatalitie... 10\nFishMercur... 11\nFlightDelay... 11\nGirls200... 12\nGrocerie... 12\nGSS200... 13\nGSS200... 13\nIceCrea... 14\nILBoy... 14\nIlliterac... 15\nLotter... 15\nmanatee... 16\nMathAnxiet... 16\nMaunalo... 17\nMnGroundwate... 17\nMobileAd... 18\nNasda... 18\nNBA161... 19\nNCBirths200... 19\nOlympics201... 20\nPhillies200... 20\nPitchers200... 21\nQuake... 21\nQuetza... 22\nRangersTwins201... 22\nReadin... 23\nRecidivis... 23\nresampledat... 24\nSalarie... 24\nSat200... 25\nServic... 25\nSkateboar... 26\nSkating201... 26\nSpruc... 27\nStarcraf... 27\nTitani... 28\nTurbin... 28\nT... 29\nTXBirths200... 29\nVerizo... 30\nVolleyball200... 30\nwafer... 31\nWalley... 31\nWatertabl... 32\n Alelager Calories and alcohol content for ales and lagers.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Alelager\nFormat\n A data frame with X rows and Y variables:\n •\n Bangladesh Bangladesh\nDescription\n The Bangladesh data is provided solely for illustrative purposes and to enable statistical analysis.\n Full data are available from the British Geological Survey web site: http://www.bgs.ac.uk/\n arsenic/bphase2/datadownload.htm.\nUsage\n Bangladesh\nFormat\n A data frame with X rows and Y variables:\n •\nDetails\n A data set containing X. The variables are as follows:\n Batters2005 Batters2005\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Batters2005\nFormat\n A data frame with X rows and Y variables:\n •\n Beerwings Beer and hot wings consumption.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Beerwings\nFormat\n A data frame with X rows and Y variables:\n •\n BookPrices Price of hardcover textbooks from mathematics and the natural sci-\n ences and the social sciences.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n BookPrices\nFormat\n A data frame with X rows and Y variables:\n •\n Bushmeat Bushmeat in Ghana.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Bushmeat\nFormat\n A data frame with X rows and Y variables:\n •\n Bushmeat2 Bushmeat in Ghana.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Bushmeat2\nFormat\n A data frame with X rows and Y variables:\n •\n Cameras Prices of a sample of point-and-shoot digital cameras.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Cameras\nFormat\n A data frame with X rows and Y variables:\n •\n Cereals Information on various cereals.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Cereals\nFormat\n A data frame with X rows and Y variables:\n •\n Challenger Data on 23 Challenger flights.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Challenger\nFormat\n A data frame with X rows and Y variables:\n •\n ChiMarathonMen Data on Marathon times.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n ChiMarathonMen\nFormat\n A data frame with X rows and Y variables:\n •\n corrExerciseA Correlation Exercise A p.294\nDescription\n A data set containing X. The variables are as follows:\nUsage\n corrExerciseA\nFormat\n A data frame with X rows and Y variables:\n •\ncorrExerciseB 9\n corrExerciseB Correlation Exercise B p.294\nDescription\n A data set containing X. The variables are as follows:\nUsage\n corrExerciseB\nFormat\n A data frame with X rows and Y variables:\n •\n Cuckoos Cuckoos\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Cuckoos\nFormat\n A data frame with X rows and Y variables:\n •\n Diving2017 Diving times in 2017\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Diving2017\nFormat\n A data frame with X rows and Y variables:\n •\n Fatalities Random sample of 100 driver fatalities in 2009 in Pennsylvania.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Fatalities\nFormat\n A data frame with X rows and Y variables:\n •\n FishMercury Mercury levels (parts per million) for 30 fish caught in lakes in Min-\n nesota.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n FishMercury\nFormat\n A data frame with X rows and Y variables:\n •\n FlightDelays Information on 4029 United and American airlines departures from\n LGA during May and June 2009.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n FlightDelays\nFormat\n A data frame with X rows and Y variables:\n •\n Girls2004 Random sample of 40 baby girls born in Alaska and 40 baby girls born\n in Wyoming.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Girls2004\nFormat\n A data frame with X rows and Y variables:\n •\n Groceries Groceries.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Groceries\nFormat\n A data frame with X rows and Y variables:\n •\n GSS2002 Results from 2002 General Society Survey.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n GSS2002\nFormat\n A data frame with X rows and Y variables:\n •\n GSS2006 Results from 2006 General Society Survey.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n GSS2006\nFormat\n A data frame with X rows and Y variables:\n •\n IceCream Calorie information for a sample of brands of chocolate and vanilla\n ice cream.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n IceCream\nFormat\n A data frame with X rows and Y variables:\n •\n ILBoys IL Boys\nDescription\n A data set containing X. The variables are as follows:\nUsage\n ILBoys\nFormat\n A data frame with X rows and Y variables:\n •\n Illiteracy Data on a sample of countries where female illiteracy is more that 5\n percent.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Illiteracy\nFormat\n A data frame with X rows and Y variables:\n •\n Lottery Winning numbers for the daily games from May 5, 2010 through Au-\n gust 15, 2010.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Lottery\nFormat\n A data frame with X rows and Y variables:\n •\n manatees manatees\nDescription\n A data set containing X. The variables are as follows:\nUsage\n manatees\nFormat\n A data frame with X rows and Y variables:\n •\n MathAnxiety Math Anxiety\nDescription\n A data set containing X. The variables are as follows:\nUsage\n MathAnxiety\nFormat\n A data frame with X rows and Y variables:\n •\n Maunaloa Data on average CO2 levels (ppm) for the month of May from 1990 to\n 2010.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Maunaloa\nFormat\n A data frame with X rows and Y variables:\n •\n MnGroundwater Measurements on water quality of 895 randomly selected wells in Min-\n nesota.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n MnGroundwater\nFormat\n A data frame with X rows and Y variables:\n •\n MobileAds Mobile Ads.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n MobileAds\nFormat\n A data frame with X rows and Y variables:\n •\n Nasdaq NASDAQ Data.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Nasdaq\nFormat\n A data frame with X rows and Y variables:\n •\n NBA1617 NBA 2016-2017 data.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n NBA1617\nFormat\n A data frame with X rows and Y variables:\n •\n NCBirths2004 Random sample of 1009 babies born in North Carolina during 2004.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n NCBirths2004\nFormat\n A data frame with X rows and Y variables:\n •\n Olympics2012 2012 Olympics Data.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Olympics2012\nFormat\n A data frame with X rows and Y variables:\n •\n Phillies2009 Data from the 2009 season for the baseball team of the Philadelphia\n Phillies.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Phillies2009\nFormat\n A data frame with X rows and Y variables:\n •\n Pitchers2005 Pitchers2005\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Pitchers2005\nFormat\n A data frame with X rows and Y variables:\n •\n Quakes Time between earthquakes (in days).\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Quakes\nFormat\n A data frame with X rows and Y variables:\n •\n Quetzal Quetzal.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Quetzal\nFormat\n A data frame with X rows and Y variables:\n •\n RangersTwins2016 Rangers/Twins 2016 Baseball Data.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n RangersTwins2016\nFormat\n A data frame with X rows and Y variables:\n •\n Reading Children’s reading abilities.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Reading\nFormat\n A data frame with X rows and Y variables:\n •\n Recidivism Recidivism data.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Recidivism\nFormat\n A data frame with X rows and Y variables:\n •\n resampledata resampledata: Data Sets for Mathematical Statistics with Resampling\n in R\nDescription\n Package of data sets from \"Mathematical Statistics with Resampling in R\" (1st Ed. 2011, 2nd\n Ed. 2018) by <NAME> and <NAME>. Note the Bangladesh data is provided solely\n for illustrative purposes and to enable statistical analysis. Full data are available from the British\n Geological Survey web site http://www.bgs.ac.uk/arsenic/bangladesh/.\nData sets\n Type data(package=\"resampledata\") to view all datasets.\nReferences\n Hesterberg, Tim (2014), What Teachers Should Know about the Bootstrap: Resampling in the\n Undergraduate Statistics Curriculum, http://arxiv.org/abs/1411.5279.\n Chihara, <NAME>., Hesterberg, <NAME>. (1st Ed. 2011, 2nd Ed. 2018). Mathematical Statistics with\n Resampling and R. Hoboken, NJ: John Wiley & Sons.\nExamples\n library(resampledata)\n data(FlightDelays)\n Salaries Salaries\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Salaries\nFormat\n A data frame with X rows and Y variables:\n •\n Sat2008 Sat2008\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Sat2008\nFormat\n A data frame with X rows and Y variables:\n •\n Service Service times (in minutes) for 174 customers at a college snack bar.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Service\nFormat\n A data frame with X rows and Y variables:\n •\n Skateboard Skateboarding data\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Skateboard\nFormat\n A data frame with X rows and Y variables:\n •\n Skating2010 Scores from the short program and free skate for men’s figure skating\n in the 2010 Olympics.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Skating2010\nFormat\n A data frame with X rows and Y variables:\n •\n Spruce Study of factors affecting the growth of the black spruce.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Spruce\nFormat\n A data frame with X rows and Y variables:\n •\n Starcraft Starcraft\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Starcraft\nFormat\n A data frame with X rows and Y variables:\n •\n Titanic Data on male passengers on the Titanic.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Titanic\nFormat\n A data frame with X rows and Y variables:\n •\n Turbine Wind Speeds (m/s) from Carleton College Turbine.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Turbine\nFormat\n A data frame with X rows and Y variables:\n •\n TV TV\nDescription\n A data set containing X. The variables are as follows:\nUsage\n TV\nFormat\n A data frame with X rows and Y variables:\n •\n TXBirths2004 Random sample of 1587 babies born in Texas in 2004.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n TXBirths2004\nFormat\n A data frame with X rows and Y variables:\n •\n Verizon Random sample of repair times for 1664 ILEC and 23 CLEC cus-\n tomers.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Verizon\nFormat\n A data frame with X rows and Y variables:\n •\n Volleyball2009 Volleyball2009\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Volleyball2009\nFormat\n A data frame with X rows and Y variables:\n •\n wafers Wafers\nDescription\n A data set containing X. The variables are as follows:\nUsage\n wafers\nFormat\n A data frame with X rows and Y variables:\n •\n Walleye Length and weight measurements for a sample of 60 walleye caught in\n Minnesota lakes.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Walleye\nFormat\n A data frame with X rows and Y variables:\n •\n Watertable Relationship between seedling growth and water table depth for a\n sample of seedlings.\nDescription\n A data set containing X. The variables are as follows:\nUsage\n Watertable\nFormat\n A data frame with X rows and Y variables:\n •"}}},{"rowIdx":59199,"cells":{"project":{"kind":"string","value":"growthcleanr"},"source":{"kind":"string","value":"cran"},"language":{"kind":"string","value":"R"},"content":{"kind":"string","value":"Package ‘growthcleanr’\n March 3, 2023\nType Package\nTitle Data Cleaner for Anthropometric Measurements\nVersion 2.1.1\nDescription Identifies implausible anthropometric (e.g., height,\n weight) measurements in irregularly spaced longitudinal datasets, such as those from elec-\n tronic health records.\nURL https://carriedaymont.github.io/growthcleanr/index.html,\n https://github.com/carriedaymont/growthcleanr\nBugReports https://github.com/carriedaymont/growthcleanr/issues\nImports R.utils (>= 2.11.0), data.table (>= 1.13.0), tidyr (>= 1.1.0),\n plyr (>= 1.8.6), dplyr (>= 1.0.1), foreach (>= 1.5.0),\n doParallel (>= 1.0.15), labelled (>= 2.5.0), magrittr (>= 1.5)\nDepends R (>= 2.10)\nLicense MIT + file LICENSE\nEncoding UTF-8\nLazyData true\nSuggests argparser (>= 0.6), bit64 (>= 4.0.2), knitr (>= 1.29),\n rmarkdown (>= 2.3), testthat (>= 2.3.2)\nRoxygenNote 7.2.3\nNeedsCompilation no\nAuthor <NAME> [ctb, cre],\n <NAME> [aut],\n <NAME> [aut],\n <NAME> [aut],\n <NAME> [ctb],\n <NAME> [ctb],\n <NAME> [ctb],\n <NAME> [ctb],\n <NAME> [ctb],\n <NAME> [ctb],\n <NAME> [ctb],\n <NAME> [ctb],\n <NAME> [ctb],\n <NAME> [ctb],\n <NAME> [ctb],\n <NAME> [ctb]\nMaintainer <NAME> <<EMAIL>>\nRepository CRAN\nDate/Publication 2023-03-03 10:20:02 UTC\nR topics documented:\nacf_answer... 3\nadjustcarryforwar... 4\nbmianthr... 6\nCDCref_... 7\ncleangrowt... 7\newm... 10\next_bmi... 12\ngrowth_cdc_ex... 14\nlenanthr... 14\nlongwid... 15\nnhanes-reference-median... 16\nread_anthr... 17\nrecode_se... 17\nsd_media... 18\nsimple_bm... 19\nsplitinpu... 20\nsyngrowt... 21\ntanner_ht_ve... 22\ntanner_ht_vel_with_2s... 22\ntestac... 23\ntest_syngrowth_sas_output_compar... 24\ntest_syngrowth_wid... 24\nweianthr... 24\nwho_ht_maxve... 25\nwho_ht_maxvel_2s... 25\nwho_ht_vel_2s... 25\nwho_ht_vel_3s... 26\n acf_answers Answers for adjustcarryforward\nDescription\n Determines what should absolutely be reincluded or definitely excluded for a given dataset, already\n run through cleangrowth.\nUsage\n acf_answers(\n subjid,\n param,\n agedays,\n sex,\n measurement,\n orig.exclude,\n sd.recenter = NA,\n ewma.exp = -1.5,\n ref.data.path = \"\",\n quietly = TRUE\n )\nArguments\n subjid Vector of unique identifiers for each subject in the database.\n param Vector identifying each measurement, may be ’WEIGHTKG’, ’HEIGHTCM’,\n or ’LENGTHCM’ ’HEIGHTCM’ vs. ’LENGTHCM’ only affects z-score cal-\n culations between ages 24 to 35 months (730 to 1095 days). All linear mea-\n surements below 731 days of life (age 0-23 months) are interpreted as supine\n length, and all linear measurements above 1095 days of life (age 36+ months)\n are interpreted as standing height. Note: at the moment, all LENGTHCM will\n be converted to HEIGHTCM. In the future, the algorithm will be updated to\n consider this difference.\n agedays Numeric vector containing the age in days at each measurement.\n sex Vector identifying the gender of the subject, may be ’M’, ’m’, or 0 for males,\n vs. ’F’, ’f’ or 1 for females.\n measurement Numeric vector containing the actual measurement data. Weight must be in\n kilograms (kg), and linear measurements (height vs. length) in centimeters (cm).\n orig.exclude Vector of exclusion assessment results from cleangrowth()\n sd.recenter Data frame or table with median SD-scores per day of life\n ewma.exp Exponent to use for weighting measurements in the exponentially weighted\n moving average calculations. Defaults to -1.5. This exponent should be neg-\n ative in order to weight growth measurements closer to the measurement being\n evaluated more strongly. Exponents that are further from zero (e.g. -3) will in-\n crease the relative influence of measurements close in time to the measurement\n being evaluated compared to using the default exponent.\n ref.data.path Path to reference data. If not supplied, the year 2000 Centers for Disease Control\n (CDC) reference data will be used.\n quietly Determines if function messages are to be displayed and if log files (parallel\n only) are to be generated. Defaults to TRUE.\nValue\n A data frame, containing an index \"n\" of rows, corresponding to the original order of the input\n vectors, and \"acf_answers\", containing the answers on whether a height value should be kept or\n excluded (returns \"Definitely Exclude\", \"Definitely Include\", or \"Unknown\" for height values, NA\n for weight values).\n adjustcarryforward adjustcarryforward adjustcarryforward Uses absolute height ve-\n locity to identify values excluded as carried forward values for rein-\n clusion.\nDescription\n adjustcarryforward adjustcarryforward Uses absolute height velocity to identify values excluded\n as carried forward values for reinclusion.\nUsage\n adjustcarryforward(\n subjid,\n param,\n agedays,\n sex,\n measurement,\n orig.exclude,\n exclude_opt = 0,\n sd.recenter = NA,\n ewma.exp = -1.5,\n ref.data.path = \"\",\n quietly = TRUE,\n minfactor = 0.5,\n maxfactor = 2,\n banddiff = 3,\n banddiff_plus = 5.5,\n min_ht.exp_under = 2,\n min_ht.exp_over = 0,\n max_ht.exp_under = 0.33,\n max_ht.exp_over = 1.5\n )\nArguments\n subjid Vector of unique identifiers for each subject in the database.\n param Vector identifying each measurement, may be ’WEIGHTKG’, ’HEIGHTCM’,\n or ’LENGTHCM’ ’HEIGHTCM’ vs. ’LENGTHCM’ only affects z-score cal-\n culations between ages 24 to 35 months (730 to 1095 days). All linear mea-\n surements below 731 days of life (age 0-23 months) are interpreted as supine\n length, and all linear measurements above 1095 days of life (age 36+ months)\n are interpreted as standing height. Note: at the moment, all LENGTHCM will\n be converted to HEIGHTCM. In the future, the algorithm will be updated to\n consider this difference.\n agedays Numeric vector containing the age in days at each measurement.\n sex Vector identifying the gender of the subject, may be ’M’, ’m’, or 0 for males,\n vs. ’F’, ’f’ or 1 for females.\n measurement Numeric vector containing the actual measurement data. Weight must be in\n kilograms (kg), and linear measurements (height vs. length) in centimeters (cm).\n orig.exclude Vector of exclusion assessment results from cleangrowth()\n exclude_opt Number from 0 to 3 indicating which option to use to handle strings of carried-\n forwards: 0. no change.\n 1. when deciding to exclude values, if we have a string of carried forwards,\n drop the most deviant value, and all CFs in the same string, and move on as\n normal.\n 2. when deciding to exclude values, if the most deviant in a string of carried\n forwards is flagged, check all the CFs in that string from 1:N. Exclude all\n after the first that is flagged for exclusion when comparing to the Include\n before and after. Do not remove things designated as include.\n 3. when deciding to exclude values, if the most deviant in a string of carried\n forwards is flagged, check all the CFs in that string from 1:N. Exclude all\n after the first that is flagged for exclusion when comparing to the Include\n before and after. Make sure remove things designated as include.\n sd.recenter Data frame or table with median SD-scores per day of life\n ewma.exp Exponent to use for weighting measurements in the exponentially weighted\n moving average calculations. Defaults to -1.5. This exponent should be neg-\n ative in order to weight growth measurements closer to the measurement being\n evaluated more strongly. Exponents that are further from zero (e.g. -3) will in-\n crease the relative influence of measurements close in time to the measurement\n being evaluated compared to using the default exponent.\n ref.data.path Path to reference data. If not supplied, the year 2000 Centers for Disease Control\n (CDC) reference data will be used.\n quietly Determines if function messages are to be displayed and if log files (parallel\n only) are to be generated. Defaults to TRUE.\n minfactor Sweep variable for computing mindiff.next.ht in 15f, default 0.5\n maxfactor Sweep variable for computing maxdiff.next.ht in 15f, default 2\n banddiff Sweep variable for computing mindiff.next.ht in 15f, default 3\n banddiff_plus Sweep variable for computing maxdiff.next.ht in 15, default 5.5\n min_ht.exp_under\n Sweep variable for computing ht.exp in 15f, default 2\n min_ht.exp_over\n Sweep variable for computing ht.exp in 15f, default 0\n max_ht.exp_under\n Sweep variable for computing ht.exp in 15f, default 0.33\n max_ht.exp_over\n Sweep variable for computing ht.exp in 15f, default 1.5\nValue\n Re-evaluated exclusion assessments based on height velocity.\nExamples\n # Run on a small subset of given data\n df <- as.data.frame(syngrowth)\n df <- df[df$subjid %in% unique(df[, \"subjid\"])[1:5], ]\n clean_df <- cbind(df,\n \"gcr_result\" = cleangrowth(df$subjid,\n df$param,\n df$agedays,\n df$sex,\n df$measurement))\n # Adjust carry forward values in cleaned data\n adj_clean <- adjustcarryforward(subjid = clean_df$subjid,\n param = clean_df$param,\n agedays = clean_df$agedays,\n sex = clean_df$sex,\n measurement = clean_df$measurement,\n orig.exclude = clean_df$gcr_result)\n bmianthro BMI Anthro\nDescription\n Part of default CDC-derived tables\nDetails\n Contains BMI data for calculating BMI\nbmianthro.txt.gz\n Used in function cleangrowth()\nCDCref_d 7\n CDCref_d CDC BMI reference data\nDescription\n Used for extended BMIz computation\nCDCref_d.csv.gz\n Used for extended BMI computation\n cleangrowth Clean growth measurements\nDescription\n Clean growth measurements\nUsage\n cleangrowth(\n subjid,\n param,\n agedays,\n sex,\n measurement,\n recover.unit.error = FALSE,\n sd.extreme = 25,\n z.extreme = 25,\n lt3.exclude.mode = \"default\",\n height.tolerance.cm = 2.5,\n error.load.mincount = 2,\n error.load.threshold = 0.5,\n sd.recenter = NA,\n sdmedian.filename = \"\",\n sdrecentered.filename = \"\",\n include.carryforward = FALSE,\n ewma.exp = -1.5,\n ref.data.path = \"\",\n log.path = NA,\n parallel = FALSE,\n num.batches = NA,\n quietly = TRUE,\n adult_cutpoint = 20,\n weight_cap = Inf,\n adult_columns_filename = \"\"\n )\nArguments\n subjid Vector of unique identifiers for each subject in the database.\n param Vector identifying each measurement, may be ’WEIGHTKG’, ’WEIGHTLBS’,\n ’HEIGHTCM’, ’HEIGHTIN’, or ’LENGTHCM’ ’HEIGHTCM’/’HEIGHTIN’\n vs. ’LENGTHCM’ only affects z-score calculations between ages 24 to 35\n months (730 to 1095 days). All linear measurements below 731 days of life\n (age 0-23 months) are interpreted as supine length, and all linear measurements\n above 1095 days of life (age 36+ months) are interpreted as standing height.\n Note: at the moment, all LENGTHCM will be converted to HEIGHTCM. In\n the future, the algorithm will be updated to consider this difference. Addition-\n ally, imperial ’HEIGHTIN’ and ’WEIGHTLBS’ measurements are converted to\n metric during algorithm calculations.\n agedays Numeric vector containing the age in days at each measurement.\n sex Vector identifying the gender of the subject, may be ’M’, ’m’, or 0 for males,\n vs. ’F’, ’f’ or 1 for females.\n measurement Numeric vector containing the actual measurement data. Weight must be in\n kilograms (kg), and linear measurements (height vs. length) in centimeters (cm).\n recover.unit.error\n Indicates whether the cleaning algorithm should attempt to identify unit errors\n (I.e. inches vs. cm, lbs vs. kg). If unit errors are identified, the value will be\n corrected and retained within the cleaning algorithm as a valid measurement.\n Defaults to FALSE.\n sd.extreme Measurements more than sd.extreme standard deviations from the mean (either\n above or below) will be flagged as invalid. Defaults to 25.\n z.extreme Measurements with an absolute z-score greater than z.extreme will be flagged\n as invalid. Defaults to 25.\n lt3.exclude.mode\n Determines type of exclusion procedure to use for 1 or 2 measurements of one\n type without matching same ageday measurements for the other parameter. Op-\n tions include \"default\" (standard growthcleanr approach), and \"flag.both\" (in\n case of two measurements of one type without matching values for the other\n parameter, flag both for exclusion if beyond threshold)\n height.tolerance.cm\n maximum decrease in height tolerated for sequential measurements\n error.load.mincount\n minimum count of exclusions on parameter before considering excluding all\n measurements. Defaults to 2.\n error.load.threshold\n threshold of percentage of excluded measurement count to included measure-\n ment count that must be exceeded before excluding all measurements of either\n parameter. Defaults to 0.5.\n sd.recenter specifies how to recenter medians. May be a data frame or table w/median SD-\n scores per day of life by gender and parameter, or \"NHANES\" or \"derive\" as a\n character vector.\n • If sd.recenter is specified as a data set, use the data set\n • If sd.recenter is specified as \"nhanes\", use NHANES reference medians\n • If sd.recenter is specified as \"derive\", derive from input\n • If sd.recenter is not specified or NA:\n – If the input set has at least 5,000 observations, derive medians from\n input\n – If the input set has fewer than 5,000 observations, use NHANES\n If specifying a data set, columns must include param, sex, agedays, and sd.median\n (referred to elsewhere as \"modified Z-score\"), and those medians will be used\n for recentering. A summary of how the NHANES reference medians were de-\n rived is available in README.md. Defaults to NA.\n sdmedian.filename\n Name of file to save sd.median data calculated on the input dataset to as CSV.\n Defaults to \"\", for which this data will not be saved. Use for extracting medians\n for parallel processing scenarios other than the built-in parallel option.\n sdrecentered.filename\n Name of file to save re-centered data to as CSV. Defaults to \"\", for which this\n data will not be saved. Useful for post-processing and debugging.\n include.carryforward\n Determines whether Carry-Forward values are kept in the output. Defaults to\n False.\n ewma.exp Exponent to use for weighting measurements in the exponentially weighted\n moving average calculations. Defaults to -1.5. This exponent should be neg-\n ative in order to weight growth measurements closer to the measurement being\n evaluated more strongly. Exponents that are further from zero (e.g. -3) will in-\n crease the relative influence of measurements close in time to the measurement\n being evaluated compared to using the default exponent.\n ref.data.path Path to reference data. If not supplied, the year 2000 Centers for Disease Control\n (CDC) reference data will be used.\n log.path Path to log file output when running in parallel (non-quiet mode). Default is\n NA. A new directory will be created if necessary. Set to NA to disable log files.\n parallel Determines if function runs in parallel. Defaults to FALSE.\n num.batches Specify the number of batches to run in parallel. Only applies if parallel is set\n to TRUE. Defaults to the number of workers returned by the getDoParWorkers\n function in the foreach package.\n quietly Determines if function messages are to be displayed and if log files (parallel\n only) are to be generated. Defaults to TRUE\n adult_cutpoint Number between 18 and 20, describing ages when the pediatric algorithm should\n not be applied (< adult_cutpoint), and the adult algorithm should apply (>=\n adult_cutpoint). Numbers outside this range will be changed to the closest num-\n ber within the range. Defaults to 20.\n weight_cap Positive number, describing a weight cap in kg (rounded to the nearest .1, +/- .1)\n within the adult dataset. If there is no weight cap, set to Inf. Defaults to Inf.\n adult_columns_filename\n Name of file to save original adult data, with additional output columns to as\n CSV. Defaults to \"\", for which this data will not be saved. Useful for post-\n analysis. For more information on this output, please see README.\nValue\n Vector of exclusion codes for each of the input measurements.\n Possible values for each code are:\n • ’Include’, ’Unit-Error-High’, ’Unit-Error-Low’, ’Swapped-Measurements’, ’Missing’,\n • ’Exclude-Carried-Forward’, ’Exclude-SD-Cutoff’, ’Exclude-EWMA-Extreme’, ’Exclude-EWMA-\n Extreme-Pair’,\n • ’Exclude-Extraneous-Same-Day’,\n • ’Exclude-EWMA-8’, ’Exclude-EWMA-9’, ’Exclude-EWMA-10’, ’Exclude-EWMA-11’, ’Exclude-\n EWMA-12’, ’Exclude-EWMA-13’, ’Exclude-EWMA-14’,\n • ’Exclude-Min-Height-Change’, ’Exclude-Max-Height-Change’,\n • ’Exclude-Pair-Delta-17’, ’Exclude-Pair-Delta-18’, ’Exclude-Pair-Delta-19’,\n • ’Exclude-Single-Outlier’, ’Exclude-Too-Many-Errors’, ’Exclude-Too-Many-Errors-Other-Parameter’\nExamples\n # Run calculation using a small subset of given data\n df_stats <- as.data.frame(syngrowth)\n df_stats <- df_stats[df_stats$subjid %in% unique(df_stats[, \"subjid\"])[1:5], ]\n clean_stats <-cleangrowth(subjid = df_stats$subjid,\n param = df_stats$param,\n agedays = df_stats$agedays,\n sex = df_stats$sex,\n measurement = df_stats$measurement)\n # Once processed you can filter data based on result value\n df_stats <- cbind(df_stats, \"clean_result\" = clean_stats)\n clean_df_stats <- df_stats[df_stats$clean_result == \"Include\",]\n # Parallel processing: run using 2 cores and batches\n clean_stats <- cleangrowth(subjid = df_stats$subjid,\n param = df_stats$param,\n agedays = df_stats$agedays,\n sex = df_stats$sex,\n measurement = df_stats$measurement,\n parallel = TRUE,\n num.batches = 2)\n ewma Exponentially Weighted Moving Average (EWMA)\nDescription\n ewma calculates the exponentially weighted moving average (EWMA) for a set of numeric observa-\n tions over time.\nUsage\n ewma(agedays, z, ewma.exp, ewma.adjacent = TRUE)\nArguments\n agedays Vector of age in days for each z score (potentially transformed to adjust weight-\n ing).\n z Input vector of numeric z-score data.\n ewma.exp Exponent to use for weighting.\n ewma.adjacent Specify whether EWMA values excluding adjacent measurements should be cal-\n culated. Defaults to TRUE.\nValue\n Data frame with 3 variables:\n • The first variable (ewma.all) contains the EWMA at observation time excluding only the actual\n observation for that time point.\n • The second variable (ewma.before) contains the EWMA for each observation excluding both\n the actual observation and the immediate prior observation.\n • The third variable (ewma.after) contains the EWMA for each observation excluding both the\n actual observation and the subsequent observation.\nExamples\n # Run on 1 subject, 1 type of parameter\n df_stats <- as.data.frame(syngrowth)\n df_stats <- df_stats[df_stats$subjid == df_stats$subjid[1] &\n df_stats$param == \"HEIGHTCM\", ]\n # Get the uncentered z-scores\n measurement_to_z <- read_anthro(cdc.only = TRUE)\n sd <- measurement_to_z(df_stats$param,\n df_stats$agedays,\n df_stats$sex,\n df_stats$measurement,\n TRUE)\n # Calculate exponentially weighted moving average\n e_df <- ewma(df_stats$agedays, sd, ewma.exp = -1.5)\n ext_bmiz Calculate extended BMI measures\nDescription\n ext_bmiz Calculates the sigma (scale parameter for the half-normal distribution), extended BMI\n percentile, extended BMIz, and the CDC LMS Z-scores for weight, height, and BMI for children\n between 2 and 19.9 years of age. Note that for BMIs <= 95th percentile of the CDC growth charts,\n the extended values for BMI are equal to the LMS values. The extended values differ only for\n children who have a BMI > 95th percentile.\nUsage\n ext_bmiz(\n data,\n age = \"agem\",\n wt = \"wt\",\n ht = \"ht\",\n bmi = \"bmi\",\n adjust.integer.age = TRUE,\n ref.data.path = \"\"\n )\nArguments\n data Input data frame or data table\n age Name of input column containing subject age in months in quotes, default \"agem\"\n wt Name of input column containing weight (kg) value in quotes, default \"wt\"\n ht Name of input column containing height (cm) value in quotes, default \"ht\"\n bmi Name of input column containing calculated BMI in quotes, default \"bmi\"\n adjust.integer.age\n If age inputs are all integer, add 0.5 if TRUE; default TRUE\n ref.data.path Path to directory containing reference data\nDetails\n This function should produce output equivalent to the SAS macro provided at https://www.cdc.gov/nccdphp/dnpao/growthcha\n The macro was updated in December, 2022, according to the findings of the NCHS report avail-\n able at https://dx.doi.org/10.15620/cdc:121711. This function has been updated to match it as of\n growthcleanr v2.1.0.\n The extended BMIz is the inverse cumulative distribution function (CDF) of the extended BMI\n percentile. If the extended percentile is very close to 100, the qnorm function in R produces an\n infinite value. This occurs only if the the extended BMI percentile is > 99.99999999999999. This\n occurs infrequently, such as a 48-month-old with a BMI > 39, and it is likely that these BMIs\n represent data entry errors. For these cases, extended BMIz is set to 8.21, a value that is slightly\n greater than the largest value that can be calculated.\n See the README.md file for descriptions of the output columns generated by this function.\n data must have columns for at least age, sex, weight, height, and bmi.\n age should be coded in months, using the most precise values available. To convert to months from\n age in years, multiply by 12. To convert to months from age in days, divide by 30.4375 (365.25 /\n 12).\n sex is coded as 1, boys, Boys, b, B, males, Males, m, or M for male subjects or 2, girls, Girls, g, G,\n females, Females, f, or F for female subjects. Note that this is different from cleangrowth, which\n uses 0 (Male) and 1 (Female).\n wt should be in kilograms.\n ht should be in centimeters.\n Specify the input data parameter names for age, wt, ht, bmi using quotation marks. See example\n below.\n If the parameter adjust.integer.age is TRUE (the default), 0.5 will be added to all age if all input\n values are integers. Set to FALSE to disable.\n By default, the reference data file CDCref_d.csv, made available at https://www.cdc.gov/nccdphp/dnpao/growthcharts/resour\n is included in this package for convenience. If you are developing this package, use ref.data.path\n to adjust the path to this file from your working directory if necessary.\nValue\n Expanded data frame containing computed BMI values\nExamples\n # Run on a small subset of given data\n df <- as.data.frame(syngrowth)\n df <- df[df$subjid %in% unique(df[, \"subjid\"])[1:5], ]\n df <- cbind(df,\n \"gcr_result\" = cleangrowth(df$subjid,\n df$param,\n df$agedays,\n df$sex,\n df$measurement))\n df_wide <- longwide(df) # convert to wide format for ext_bmiz\n df_wide_bmi <- simple_bmi(df_wide) # compute simple BMI\n # Calling the function with default column names\n df_bmiz <- ext_bmiz(df_wide_bmi)\n # Specifying different column names; note that quotes are used\n dfc <- simple_bmi(df_wide)\n colnames(dfc)[colnames(dfc) %in% c(\"agem\", \"wt\", \"ht\")] <-\n c(\"agemos\", \"weightkg\", \"heightcm\")\n df_bmiz <- ext_bmiz(dfc, age=\"agemos\", wt=\"weightkg\", ht=\"heightcm\")\n # Disabling conversion of all-integer age in months to (age + 0.5)\n dfc <- simple_bmi(df_wide)\n df_bmiz <- ext_bmiz(dfc, adjust.integer.age=FALSE)\n growth_cdc_ext CDC Growth Percentile Table\nDescription\n Part of default CDC-derived tables\nDetails\n Contains percentiles for various ages, gender, and weights, pre-calculated by CDC\ngrowthfile_cdc_ext.csv.gz\n Used in function cleangrowth()\n lenanthro Length to Age Table\nDescription\n Part of default CDC-derived tables\nDetails\n Contains percentiles for various ages, gender, and weights, pre-calculated by CDC\nlenanthro.txt.gz\n Used in function cleangrowth()\n longwide Transform data in growthcleanr format into wide structure for BMI\n calculation\nDescription\n longwide transforms data from long to wide format. Ideal for transforming output from growth-\n cleanr::cleangrowth() into a format suitable for growthcleanr::ext_bmiz().\nUsage\n longwide(\n long_df,\n id = \"id\",\n subjid = \"subjid\",\n sex = \"sex\",\n agedays = \"agedays\",\n param = \"param\",\n measurement = \"measurement\",\n gcr_result = \"gcr_result\",\n include_all = FALSE,\n inclusion_types = c(\"Include\"),\n extra_cols = NULL,\n keep_unmatched_data = FALSE\n )\nArguments\n long_df A data frame to be transformed. Expects columns: id, subjid, sex, agedays,\n param, measurement, and gcr_result.\n id name of observation ID column\n subjid name of subject ID column\n sex name of sex descriptor column\n agedays name of age (in days) descriptor column\n param name of parameter column to identify each type of measurement\n measurement name of measurement column containing the actual measurement data\n gcr_result name of column of results from growthcleanr::cleangrowth()\n include_all Determines whether the function keeps all exclusion codes. If TRUE, all ex-\n clusion types are kept and the inclusion_types argument is ignored. Defaults to\n FALSE.\n inclusion_types\n Vector indicating which exclusion codes from the cleaning algorithm should\n be included in the data, given that include_all is FALSE. For all options, see\n growthcleanr::cleangrowth(). Defaults to c(\"Include\").\n extra_cols Vector of additional columns to include in the output. If a column C1 differs\n on agedays matched height and weight values, then include separate ht_C1 and\n wt_C1 columns as well as a match_C1 column that gives booleans indicating\n where ht_C1 and wt_C1 are the same. If the agedays matched height and weight\n columns are identical, then only include a single version of C1. Defaults to\n empty vector (not keeping any additional columns).\n keep_unmatched_data\n boolean indicating whether to keep height/weight observations that do not have\n a matching weight/height on that day\nValue\n Returns a data frame transformed from long to wide. Includes only values flagged with indicated\n inclusion types. Potentially includes additional columns if arguments are passed to extra_cols. For\n each subject, heights without corresponding weights for a given age (and vice versa) will be dropped\n unless keep_unmatched_data is set to TRUE.\nExamples\n # Run on a small subset of given data\n df <- as.data.frame(syngrowth)\n df <- df[df$subjid %in% unique(df[, \"subjid\"])[1:5], ]\n df <- cbind(df,\n \"gcr_result\" = cleangrowth(df$subjid,\n df$param,\n df$agedays,\n df$sex,\n df$measurement))\n # Convert to wide format\n wide_df <- longwide(df)\n # Include all inclusion types\n wide_df <- longwide(df, include_all = TRUE)\n # Specify all inclusion codes\n wide_df <- longwide(df, inclusion_types = c(\"Include\", \"Exclude-Carried-Forward\"))\n nhanes-reference-medians\n NHANES reference medians\nDescription\n Contains reference median values for default recentering, derived from NHANES years 2009-2018\nnhanes-reference-medians.csv.gz\n Used in function cleangrowth()\n read_anthro Function to calculate z-scores and csd-scores based on anthro tables.\nDescription\n Function to calculate z-scores and csd-scores based on anthro tables.\nUsage\n read_anthro(path = \"\", cdc.only = FALSE)\nArguments\n path Path to supplied reference anthro data. Defaults to package anthro tables.\n cdc.only Whether or not only CDC data should be used. Defaults to false.\nValue\n Function for calculating BMI based on measurement, age in days, sex, and measurement value.\nExamples\n # Return calculating function with all defaults\n afunc <- read_anthro()\n # Return calculating function while specifying a path and using only CDC data\n afunc <- read_anthro(path = system.file(\"extdata\", package = \"growthcleanr\"),\n cdc.only = TRUE)\n recode_sex Recode binary sex variable for compatibility\nDescription\n recode_sex recodes a binary sex variable for a given source column in a data frame or data table.\n Useful in transforming output from growthcleanr::cleangrowth() into a format suitable for growth-\n cleanr::ext_bmiz().\nUsage\n recode_sex(\n input_data,\n sourcecol = \"sex\",\n sourcem = \"0\",\n sourcef = \"1\",\n targetcol = \"sex_recoded\",\n targetm = 1L,\n targetf = 2L\n )\nArguments\n input_data a data frame or data table to be transformed. Expects a source column containing\n a binary sex variable.\n sourcecol name of sex descriptor column. Defaults to \"sex\"\n sourcem variable indicating \"male\" sex in input data. Defaults to \"0\"\n sourcef variable indicating \"female\" sex in input data. Defaults to \"1\"\n targetcol desired name of recoded sex descriptor column. Defaults to \"sex_recoded\"\n targetm desired name of recoded sex variable indicating \"male\" sex in output data. De-\n faults to 1\n targetf desired name of recoded sex variable indicating \"female\" sex in output data.\n Defaults to 2\nValue\n Returns a data table with recoded sex variables.\nExamples\n # Run on given data\n df <- as.data.frame(syngrowth)\n # Run with all defaults\n df_r <- recode_sex(df)\n # Specify different targets\n df_rt <- recode_sex(df, targetcol = \"sexr\", targetm = \"Male\", targetf = \"Female\")\n # Specify different inputs\n df_ri <- recode_sex(df_rt, sourcecol = \"sexr\", sourcem = \"Male\", sourcef = \"Female\")\n sd_median Calculate median SD score by age for each parameter.\nDescription\n Calculate median SD score by age for each parameter.\nUsage\n sd_median(param, sex, agedays, sd.orig)\nArguments\n param Vector identifying each measurement, may be ’WEIGHTKG’, or ’HEIGHTCM’.\n sex Vector identifying the gender of the subject, may be ’M’, ’m’, or 0 for males,\n vs. ’F’, ’f’ or 1 for females.\n agedays Numeric vector containing the age in days at each measurement.\n sd.orig Vector of previously calculated standard deviation (SD) scores for each mea-\n surement before re-centering.\nValue\n Table of data with median SD-scores per day of life by gender and parameter.\nExamples\n # Run on 1 subject\n df_stats <- as.data.frame(syngrowth)\n df_stats <- df_stats[df_stats$subjid == df_stats$subjid[1], ]\n # Get the original standard deviations\n measurement_to_z <- read_anthro(cdc.only = TRUE)\n sd.orig <- measurement_to_z(df_stats$param,\n df_stats$agedays,\n df_stats$sex,\n df_stats$measurement,\n TRUE)\n # Calculate median standard deviations\n sd.m <- sd_median(df_stats$param,\n df_stats$sex,\n df_stats$agedays,\n sd.orig)\n simple_bmi Compute BMI using standard formula\nDescription\n simple_bmi Computes BMI using standard formula. Assumes input compatible with output from\n longwide().\nUsage\n simple_bmi(wide_df, wtcol = \"wt\", htcol = \"ht\")\nArguments\n wide_df A data frame or data table containing heights and weights in wide format, e.g.,\n after transformation with longwide()\n wtcol name of observation height value column, default ’wt’\n htcol name of subject weight value column, default ’ht’\nValue\n Returns a data table with the added column \"bmi\"\nExamples\n # Simple usage\n # Run on a small subset of given data\n df <- as.data.frame(syngrowth)\n df <- df[df$subjid %in% unique(df[, \"subjid\"])[1:5], ]\n df <- cbind(df,\n \"gcr_result\" = cleangrowth(df$subjid,\n df$param,\n df$agedays,\n df$sex,\n df$measurement))\n # Convert to wide format\n wide_df <- longwide(df)\n wide_df_with_bmi <- simple_bmi(wide_df)\n # Specifying different column names; note that quotes are used\n colnames(wide_df)[colnames(wide_df) %in% c(\"wt\", \"ht\")] <-\n c(\"weight\", \"height\")\n wide_df_with_bmi <- simple_bmi(wide_df, wtcol = \"weight\", htcol = \"height\")\n splitinput Split input data into multiple files\nDescription\n splitinput Splits input based on keepcol specified, yielding csv files each with at least the min-\n inum number of rows that are written and saved separately (except for the last split file written,\n which may be smaller). Allows splitting input data while ensuring all records for each individual\n subject will stay together in one file. Pads split filenames with zeros out to five digits for consistency,\n assuming < 100,000 file count result.\nUsage\n splitinput(\n df,\n fname = deparse(substitute(df)),\n fdir = NA,\n min_nrow = 10000,\n keepcol = \"subjid\"\n )\nArguments\n df data frame to split\n fname new name for each of the split files to start with\n fdir directory to put each of the split files (use \".\" for working directory). Must be\n changed from default (NA), which will trigger error.\n min_nrow minimum number of rows for each split file (default 10000)\n keepcol the column name (default \"subjid\") to use to keep records with the same values\n together in the same single split file\nValue\n the count number referring to the last split file written\nExamples\n # Run on given data\n df <- as.data.frame(syngrowth)\n # Run with all defaults (specifying directory)\n splitinput(df, fdir = tempdir())\n # Specifying the name, directory and minimum row size\n splitinput(df, fname = \"syngrowth\", fdir = tempdir(), min_nrow = 5000)\n # Specifying a different subject ID column\n colnames(df)[colnames(df) == \"subjid\"] <- \"sub_id\"\n splitinput(df, fdir = tempdir(), keepcol = \"sub_id\")\n syngrowth syngrowth\nDescription\n A synthetic set of measurements from ~3,500 subjects generated using Synthea, with measurement\n errors for testing with growthcleanr. Contains both pediatric and adult data.\nUsage\n syngrowth\nFormat\n A data frame with six variables: id, subjid, sex, agedays, param, and measurement\nDetails\n Example electronic health record (heightcm, weightkg) data.\n tanner_ht_vel Tanner Growth Velocity Table\nDescription\n Part of default CDC-derived tables\nDetails\n Contains velocities for growth pre-calculated by CDC\ntanner_ht_vel.csv.gz\n Used in function cleangrowth()\n tanner_ht_vel_with_2sd\n Tanner Growth Velocity Table with (2σ)\nDescription\n Part of default CDC-derived tables\nDetails\n Contains velocities for growth pre-calculated by CDC, including those 2 standard deviations away.\ntanner_ht_vel_with_2sd.csv.gz\n Used in function acf_answers()\n testacf Function to test adjust carried forward\nDescription\n The goal of this script is to consider the height values that growthcleanr excludes as “carried for-\n ward” for potential re-inclusion by using a reverse absolute height velocity check based on step 15\n of the Daymont et al. algorithm\nUsage\n testacf(\n infile,\n seed = 7,\n searchtype = \"random\",\n grid.length = 9,\n writeout = FALSE,\n outfile = paste0(\"test_adjustcarryforward_\", format(Sys.time(),\n \"%m-%d-%Y_%H-%M-%S\")),\n quietly = FALSE,\n param = \"none\",\n debug = FALSE,\n maxrecs = 0,\n exclude_opt = 0,\n add_answers = TRUE\n )\nArguments\n infile Input data frame/data table, cleaned by cleangrowth(), with columns as de-\n scribed in main README.md\n seed Numeric random seed, used only when performing random search\n searchtype Type of search to perform: random (default), line-grid, full-grid\n grid.length Number of steps in grid to search\n writeout Write output to file? Default FALSE.\n outfile \"Output file name, default ’test_adjustcarrforward_DATE_TIME’, where DATE\n is the current system date and time\"\n quietly Verbose progress info\n param \"none\", or data frame to specify which parameters to run full search on, and\n values to use if not, used only when performing full-grid search\n debug Produce extra data files for debugging\n maxrecs Limit to specified # subjects, default 0 (no limit)\n exclude_opt Type of exclusion method for carried forward strings, 0 to 3. See adjustcarry-\n forward documentation for more information\n add_answers TRUE or FALSE, indicating whether or not to add answers (definely include/exclude)\n for the given dataset. Defaults to TRUE\nValue\n A list containing: testacf_res: data frame with adjustcarryforward results for each run, params: a\n data frame containing parameter values for each run. debug_filtered_data: data frame with original\n data, returned if debug TRUE\n test_syngrowth_sas_output_compare\n CDC SAS BMI Output\nDescription\n Contains results of CDC SAS macro for calculating BMI values.\ntest_syngrowth_sas_output_compare.csv.gz\n Used to test function ext_bmiz()\n test_syngrowth_wide CDC SAS BMI Input\nDescription\n Contains input data for CDC SAS macro for calculating BMI values.\ntest_syngrowth_wide.csv.gz\n Used to test function ext_bmiz()\n weianthro Weight Anthro Table\nDescription\n Part of default CDC-derived tables\nDetails\n Contains median and standard deviation for weight by age and gender\nweianthro.csv.gz\n Used in function cleangrowth()\n who_ht_maxvel WHO Maximum Height Velocity for (3σ)\nDescription\n Part of default WHO-derived tables\nDetails\n Contains three standard deviations for the World Health Organization values of maximum height\n velocities.\nwho_ht_maxvel_3sd.csv.gz\n Used in function cleangrowth()\n who_ht_maxvel_2sd WHO Maximum Height Velocity for (2σ)\nDescription\n Part of default WHO-derived tables\nDetails\n Contains two standard deviations for the World Health Organization values of maximum height\n velocities.\nwho_ht_maxvel_2sd.csv.gz\n Used in function acf_answers()\n who_ht_vel_2sd WHO Height Velocity for (2σ)\nDescription\n Part of default WHO-derived tables\nDetails\n Contains two standard deviations for the World Health Organization values of height velocities.\nwho_ht_vel_2sd.csv.gz\n Used in function acf_answers()\n who_ht_vel_3sd WHO Height Velocity for (3σ)\nDescription\n Part of default WHO-derived tables\nDetails\n Contains three standard deviations for the World Health Organization values of height velocities.\nwho_ht_vel_3sd.csv.gz\n Used in function cleangrowth()"}}}],"truncated":false,"partial":false},"paginationData":{"pageIndex":591,"numItemsPerPage":100,"numTotalItems":59733,"offset":59100,"length":100}},"jwt":"eyJhbGciOiJFZERTQSIsImtpZCI6IjVHZDBvd0g5MTM2eDZjc1FvbE1zcktNWUZoRVFoUm5PVVVybEpjOUhaUEEifQ.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sImlhdCI6MTc4NjE5OTM3NSwianRpIjoiOTkzY2VlZDQtYWE1MC00MGU5LTgwMDUtOTE5NWU2YzgyYWY1Iiwic3ViIjoiL2RhdGFzZXRzL1NpdmlsVGFyYW0vc3RhcmNvZGVyMi1kb2N1bWVudGF0aW9uIiwiZXhwIjoxNzg2MjAyOTc1LCJpc3MiOiJodHRwczovL2h1Z2dpbmdmYWNlLmNvIn0.z7YDgViVOnhuaufEU43BCw602jwit2thMtPR44rUUnxZQI7gmYFu2M5AX6-MAGYf2uTh2NuVIWdoC_EAPsNVBw","displayUrls":true,"splitSizeSummaries":[{"config":"default","split":"train","numRows":59733,"numBytesParquet":1850870873}]},"discussionsStats":{"closed":0,"open":2,"total":2},"fullWidth":true,"hasGatedAccess":true,"hasFullAccess":true,"isEmbedded":false,"savedQueries":{"community":[{"views":[{"key":"default/train","displayName":"train","viewName":"train"}],"sql":"-- The SQL console is powered by DuckDB WASM and runs entirely in the browser.\n-- Get started by typing a query or selecting a view from the options below.\nSELECT *\nFROM train\nWHERE source = 'free_programming_book'\n AND lower(content) LIKE '%python%'\n AND LOWER(content) LIKE '%algorithms%';","title":"Python Algorithms in Free Books","createdAt":"2025-04-03T17:17:56.697Z","slug":"RUf2lH9","private":false,"justification":"Retrieves entries related to both Python and algorithms from a specific source, providing a filtered dataset but lacking broader analytical depth.","viewNames":["train"]},{"views":[{"key":"default/train","displayName":"train","viewName":"train"}],"sql":"-- The SQL console is powered by DuckDB WASM and runs entirely in the browser.\n-- Get started by typing a query or selecting a view from the options below.\nSELECT *\nFROM train\nWHERE source = 'free_programming_book'\n AND language = 'Python'\nLIMIT 100;","title":"Python Code from Free Books","createdAt":"2025-04-03T11:27:46.424Z","slug":"WGn_NVw","private":false,"justification":"This query retrieves a limited number of records related to Python programming books, providing basic filtering but minimal analytical insight.","viewNames":["train"]}],"user":[]}}">
Search is not available for this dataset
project stringlengths 1 235 | source stringclasses 16
values | language stringclasses 48
values | content stringlengths 909 64.8M |
|---|---|---|---|
en_wikibooks_org_wiki_Non-Programmer%27s_Tutorial_for_Python_3 | free_programming_book | Python | Non-Programmer's Tutorial for Python 3 is a featured book on Wikibooks because it contains substantial content, it is well-formatted, and the Wikibooks community has decided to feature it on the main page or in other places. Please continue to improve it and thanks for the great work so far! You can edit its advertisem... |
peakRAM | cran | R | Package ‘peakRAM’
October 14, 2022
Title Monitor the Total and Peak RAM Used by an Expression or Function
Version 1.0.2
URL http://github.com/tpq/peakRAM
BugReports http://github.com/tpq/peakRAM/issues
Description When working with big data sets, RAM conservation is critically... |
iced-nrepl | hex | Erlang | iced nREPL
===
[nREPL](https://github.com/nrepl/nREPL) middleware to support [vim-iced](https://github.com/liquidz/vim-iced).
[](https://circleci.com/gh/liquidz/iced-nrepl)
[
---
### Azure Custom Providers Module for Go
[](https://pkg.go.dev/github.com/Azure/azure-sdk-for-go/sdk/resourcemanager/customproviders/armcustomproviders)
The `... |
fractaldim | cran | R | Package ‘fractaldim’
October 13, 2022
Version 0.8-5
Date 2021-10-05
Title Estimation of Fractal Dimensions
Author <NAME> <<EMAIL>>,
<NAME> <<EMAIL>>,
<NAME> <<EMAIL>>
Maintainer <NAME> <<EMAIL>>
Depends R (>= 2.11.0), abind
Suggests wavelets, pcaPP, RandomFields, s... |
ImpShrinkage | cran | R | Package ‘ImpShrinkage’
June 21, 2023
Type Package
Title Improved Shrinkage Estimations for Multiple Linear Regression
Version 1.0.0
Maintainer <NAME> <<EMAIL>>
Description
A variety of improved shrinkage estimators in the area of statistical analysis: unrestricted; re... |
github.com/uli-go/xz | go | Go | README
[¶](#section-readme)
---
### Package xz
This Go language package supports the reading and writing of xz compressed streams. It includes also a gxz command for compressing and decompressing data. The package is completely written in Go and doesn't have any dependency on any C code.
The package is currently un... |
TAQMNGR | cran | R | Package ‘TAQMNGR’
October 12, 2022
Version 2018.5-1
Date 2018-05-12
Title Manage Tick-by-Tick Transaction Data
Author <NAME>, <NAME>, <NAME> and 'gzstream' authors.
Maintainer <NAME> <<EMAIL>>
Depends R (>= 3.0-2)
LinkingTo Rcpp (>= 0.11.0)
Imports
Suggests
SystemRequirement... |
jagshelper | cran | R | Package ‘jagshelper’
November 8, 2022
Type Package
Title Extracting and Visualizing Output from 'jagsUI'
Version 0.1.11
Date 2022-11-08
Author <NAME> [aut, cre]
Maintainer <NAME> <<EMAIL>>
Description Tools are provided to streamline Bayesian analyses in 'JAGS' using
the '... |
greergan-validator | npm | JavaScript | **node-validator is a library of string validation, filtering and sanitization methods.**
To install node-validator, use [npm](http://github.com/isaacs/npm):
```
$ npm install validator
```
To use the library in the browser, include `validator-min.js`
Example
---
```
var check = require('validator').check, sanitiz... |
iIneq | cran | R | Package ‘iIneq’
October 13, 2022
Type Package
Title Computing Individual Components of the Gini and the Theil
Indices
Version 1.0.2
Date 2021-01-11
Author <NAME>
Maintainer <NAME> <<EMAIL>>
Description Computes individual contributions to the overall Gini and
Theil'... |
proxytranslation | readthedoc | Unknown | Translation Proxy Online
Documentation
Release latest
Aug 16, 2023
Contents 1.1 Feature... 2 1.2 White-labe... 3 1.3 Pricing - plan... 3 1.4 Pricing - grandfathered pay-as-you-g... 3 2.1 Registration & Logi... 5 2.2 Setting up Your First Projec... 5 2.3 Running a Discover...... |
TSSVM | cran | R | Package ‘TSSVM’
December 2, 2022
Type Package
Title Time Series Forecasting using SVM Model
Version 0.1.0
Depends R (>= 2.3.1), e1071,forecast
Description Implementation and forecasting univariate time series data using the Support Vector Ma-
chine model. Support Vector Ma... |
honeydew | hex | Erlang | Honeydew 💪🏻🍈
===
[](https://travis-ci.org/koudelka/honeydew)
[](https://hex.pm/packages/honeydew)
Honeydew (["Honey, do!"](http://en.wiktionary.org/wiki/honey_do_list)) is a pl... |
jsonpath-rw | readthedoc | CoffeeScript | jsonpath-rw stable documentation
[jsonpath-rw](index.html#document-index)
---
Python JSONPath RW[¶](#python-jsonpath-rw)
===
<https://github.com/kennknowles/python-jsonpath-rwThis library provides a robust and significantly extended implementation of JSONPath for Python. It is tested with Python 2.6, 2.7, 3.2, 3.3.... |
soundcloud2000 | ruby | Ruby | soundcloud2000
===
The next generation SoundCloud client. Without all these stupid CSS files. Runs on OSX and Linux.
 for creating, applying and inspecting UPS patches.
* For creating patches, start with the create function.
* For applying patches, start with the patch_file function.
* For inspecting patches, use the functions in the parse module.
Modules
---
applyFunctions for a... |
iterpc | cran | R | Package ‘iterpc’
October 13, 2022
Type Package
Title Efficient Iterator for Permutations and Combinations
Version 0.4.2
Date 2020-01-08
Author <NAME> [aut, cre]
Maintainer <NAME> <<EMAIL>>
Description Iterator for generating permutations and combinations. They
can be eith... |
logger_backend_sqlite | hex | Erlang | API Reference
===
Modules
---
[LoggerBackendSqlite](LoggerBackendSqlite.html)
Logger backend for saving log data into a Ecto Repo.
LoggerBackendSqlite
===
Logger backend for saving log data into a Ecto Repo.
Usage
---
{:ok, _} = Logger.add_backend(LoggerBackendSqlite)
require Logger Logger.info "hello, world"
... |
aws-sdk-rum | rust | Rust | Crate aws_sdk_rum
===
**Please Note: The SDK is currently in Developer Preview and is intended strictly for feedback purposes only. Do not use this SDK for production workloads.**
With Amazon CloudWatch RUM, you can perform real-user monitoring to collect client-side data about your web application performance from a... |
nbval | readthedoc | Python | nbval 0.10.0 documentation
[nbval](#)
---
IPython Notebook Validation for py.test - Documentation[](#IPython-Notebook-Validation-for-py.test---Documentation)
===
One of the powerful uses of the IPython notebook is for documentation purposes, here we use a notebook to demonstrate the behaviour and usage of the IPyt... |
lqmm | cran | R | Package ‘lqmm’
October 13, 2022
Type Package
Title Linear Quantile Mixed Models
Version 1.5.8
Date 2022-04-05
Author <NAME>
Maintainer <NAME> <<EMAIL>>
Depends R (>= 3.0.0)
Imports stats, utils, nlme (>= 3.1-124), SparseGrid
Description Functions to fit quantile regression mo... |
babylonjs_gaurav | npm | JavaScript | Babylon.js
===
Getting started? Play directly with the Babylon.js API using our [playground](https://playground.babylonjs.com/). It also contains a lot of samples to learn how to use it.
**Any questions?** Here is our official [forum](https://forum.babylonjs.com/).
CDN
---
* <https://cdn.babylonjs.com/babylon.js>
*... |
hgvs | readthedoc | SQL | HGVS 1.4.0
[HGVS](index.html#document-index)
---
hgvs[¶](#hgvs)
===
hgvs is a Python package to parse, format, validate, normalize, and map biological sequence variants according to recommendations of the Human Genome Variation Society. Documentation at
<https://hgvs.readthedocs.io/[Source](https://github.com/bioco... |
triangulation | cran | R | Package ‘triangulation’
October 14, 2022
Type Package
Title Determine Position of Observer
Version 0.5.0
Date 2016-10-22
Description Measuring angles between points in a landscape is much easier
than measuring distances. When the location of three points is known the
... |
piar | cran | R | Package ‘piar’
August 10, 2023
Title Price Index Aggregation
Version 0.5.0
Description
Most price indexes are made with a two-step procedure, where period-over-period elemental in-
dexes are first calculated for a collection of elemental aggregates at each point ... |
cablab | readthedoc | Python | ESDC User Guide
[ESDC](index.html#document-index)
---
Earth System Data Cube (ESDC)[¶](#earth-system-data-cube-esdc)
===
The Earth System Data Cube (ESDC) is a multi-variate data set of essential Earth System variables on a common grid and sharing a common data model. The ESDC also comprises a Data Anal... |
SQLAlchemy-Utils | readthedoc | Markdown | This part of the documentation covers the installation of SQLAlchemy-Utils.
## Supported platforms¶
SQLAlchemy-Utils is currently tested against the following Python platforms.
* cPython 3.6
* cPython 3.7
* cPython 3.8
* cPython 3.9
* cPython 3.10
* cPython 3.11
## Installing an official release¶
You can inst... |
ahoy_email | ruby | Ruby | Ahoy Email
===
First-party email analytics for Rails
:fire: For web and native app analytics, check out [Ahoy](https://github.com/ankane/ahoy)
:bullettrain_side: To manage unsubscribes, check out [Mailkick](https://github.com/ankane/mailkick)
[ — Method
```
_bezier(t, p0, p1, p2, p3) -> Any
```
Plot a Bezier curve with `t` in `[0,1]` and fixed points `p0`, `p1`, `p2`, `p3`.
[`Escher._curve_idx`](#Escher._curve_idx-Tuple{Any, Any}) — Method
```
_curve_idx(segment, fraction) -> Any
```
Helper function to appro... |
SemiCompRisks | cran | R | Package ‘SemiCompRisks’
October 12, 2022
Type Package
Title Hierarchical Models for Parametric and Semi-Parametric Analyses
of Semi-Competing Risks Data
Version 3.4
Date 2021-2-2
Author <NAME>, <NAME>, <NAME>, and <NAME>
Maintainer <NAME> <<EMAIL>>
Description Hierarchica... |
bmabey-email_spec | ruby | Ruby | Email Spec
===
A collection of RSpec matchers and Cucumber steps to make testing emails go smoothly.
Setup
---
```
script/plugin install git://github.com/bmabey/email-spec.git
```
### Gem Setup
```
gem install bmabey-email_spec
# config/environments/test.rb config.gem 'bmabey-email_spec', :lib => 'email_spec'
```
##... |
clusteredinterference | cran | R | Package ‘clusteredinterference’
October 12, 2022
Type Package
Date 2019-03-16
Title Causal Effects from Observational Studies with Clustered
Interference
Version 1.0.1
Author <NAME> [aut, cre] (<https://orcid.org/0000-0003-1787-4735>),
<NAME> [ctb]
Maintainer <NAME... |
inherited_resources | ruby | Ruby | Notice
---
Inherited Resources is no longer actively maintained by the original author and has been transferred to the ActiveAdmin organization for maintenance. New feature requests are not encouraged.
If you are not already using Inherited Resources we suggest instead using Rails'
`respond_with` feature alongside th... |
create-next-app-testt | npm | JavaScript | Create Next App
===
The easiest way to get started with Next.js is by using `create-next-app`. This CLI tool enables you to quickly start building a new Next.js application, with everything set up for you. You can create a new app using the default Next.js template, or by using one of the [official Next.js examples](h... |
CCNStatusItem | cocoapods | Objective-C | ccnstatusitem
===
Welcome to the documentation for the **CCNStatusItem** library, which is designed to simplify the implementation of status bar items in macOS applications.
* Easy integration of status bar items into macOS applications
* Customizable status bar icons
* Support for status bar menus
* ... |
ROI | cran | R | Package ‘ROI.plugin.alabama’
July 7, 2023
Version 1.0-2
Title 'alabama' Plug-in for the 'R' Optimization Infrastructure
Description Enhances the R Optimization Infrastructure ('ROI') package
with the 'alabama' solver for solving nonlinear optimization problems.
Imports... |
ardl | cran | R | Package ‘ardl.nardl’
January 6, 2023
Type Package
Title Linear and Nonlinear Autoregressive Distributed Lag Models:
General-to-Specific Approach
Version 1.2.3
Date 2023-01-04
Description Estimate the linear and nonlinear autoregressive distributed lag (ARDL & NARDL) mo... |
pgmm | cran | R | Package ‘pgmm’
September 25, 2023
Type Package
Title Parsimonious Gaussian Mixture Models
Version 1.2.7
Date 2023-09-25
Author <NAME> [aut, cre] (<https://orcid.org/0000-0002-2482-523X>),
<NAME> [aut],
<NAME> [ctb],
<NAME> [aut],
<NAME> [aut],
<NAM... |
evinf | cran | R | Package ‘evinf’
September 20, 2023
Type Package
Title Inference with Extreme Value Inflated Count Data
Version 0.8.7
Description Allows users to model and draw inferences from extreme value in-
flated count data, and to evaluate these models and compare to non extreme-va... |
oauth2_github | hex | Erlang | OAuth2 GitHub
===
> A GitHub OAuth2 Provider for Elixir
[](https://travis-ci.org/chrislaskey/oauth2_github)
OAuth2 GitHub is convenience library built on top of [`oauth2`](https://hex.pm/packages/oauth2). It adds GitHub specific functio... |
github.com/jcuga/golongpoll | go | Go | README
[¶](#section-readme)
---
### golongpoll build workflow [codecov](https://codecov.io/gh/jcuga/golongpoll) [GoDoc](https://godoc.org/github.com/jcuga/golongpoll) [Go Report Card](https://goreportcard.com/report/jcuga/golongpoll)
Documentation
[¶](#section-documentation)
---
### Index [¶](#pkg-index)
* [typ... |
Test-Simple | cpan | Perl | Changes for version 1.302195 - 2023-04-28
---
* Fix done_testing(0) producing 2 plans and an incorrect message
[ Show lessShow more ]
Documentation
---
[Test::Tutorial](/dist/Test-Simple/view/lib/Test/Tutorial.pod)
A tutorial about writing really basic tests
[Test2::Transition](/dist/Test-Simple/view/lib/Test2/... |
GITHUB_hellogcc_100-gdb-tips.zip_unzipped_100个gdb小技巧（v1.0）.pdf | free_programming_book | Unknown | Table of Contents 1.
2.
i. gdb ii. gdb iii.
iv.
v.
3.
i.
4.
i. ASCII ii.
5.
i.
ii.
iii.
iv.
v.
vi.
vii.
viii.
ix.
6.
i.
ii.
iii.
iv.
v.
vi.
vii.
viii.
7.
i.
ii.
iii.
iv.
8. Catchpoint i. catchpoint ii. forkcatchpoint iii. vforkcatchpoint iv. execcatchpoint v. catchpoint vi. ptracecatchpointanti-debugging
9.
i. STL ii.
... |
faraday | ruby | Ruby | [Faraday](https://lostisland.github.io/faraday)
===
[](https://rubygems.org/gems/faraday)
[](https://github.com/lostisland/faraday/actions?query=workflow%3ACI)
[![GitHub Discussions](ht... |
godot-ko | readthedoc | Markdown | # Shading reference
# Shading language
# Shading language¶
Godot uses a shading language similar to GLSL ES 3.0. Most datatypes and functions are supported, and the few remaining ones will likely be added over time.
If you are already familiar with GLSL, the Godot Shader Migration Guide is a resource that will help... |
cfy-vue-puzzle-vcode | npm | JavaScript | vue-puzzle-vcode
===
基于开源的vue-puzzle-vcode改造的，实现了图形后台生成，验证也是走的后端。
###
安装
```
npm install cfy-vue-puzzle-vcode --save
```
###
使用
```
import Vcode from "cfy-vue-puzzle-vcode";
<Vcode
:show="isShow"
@onSuccess="onSuccess"
/>
<Vcode
:show="isVerificationShow"
:imgs="puzzleImgList"... |
raulexposito_com_the-little-go-book-en-castellano_html | free_programming_book | Markdown | # Sobre este Libro
El Pequeño Libro de Go (The Little Go Book) posee una licencia Attribution-NonCommercial-ShareAlike 4.0 Internacional. No deberías haber pagado por este libro.
Eres libre de copiar, distribuir, modificar o mostrar el libro. Sin embargo, te pido que siempre me atribuyas la autoría del libro a mí, <N... |
Entity_Framework_Code_First_Succinctly.pdf | free_programming_book | Unknown | 1 1
Entity Framework Code First Succinctly By
<NAME> by <NAME> 2
Copyright 2014 by Syncfusion Inc.
2501 Aerial Center Parkway Suite 200 Morrisville, NC 27560 USA
All rights reserved.
I mportant licensing information. Please read.
This book is available for free download from www.syncfusion.com on completion of a regis... |
Flask-Cors | readthedoc | Markdown | A Flask extension for handling Cross Origin Resource Sharing (CORS), making cross-origin AJAX possible.
This package has a simple philosophy: when you want to enable CORS, you wish to enable it for all use cases on a domain. This means no mucking around with different allowed headers, methods, etc.
By default, submis... |
hbmem | cran | R | Package ‘hbmem’
August 22, 2023
Type Package
Title Hierarchical Bayesian Analysis of Recognition Memory
Version 0.3-4
Date 2023-08-21
Author <NAME>
Maintainer <NAME> <<EMAIL>>
Depends R (>= 1.8.0), methods
Description Contains functions for fitting hierarchical versions of
... |
morph | cran | R | Package ‘morph’
February 14, 2023
Type Package
Title 3D Segmentation of Voxels into Morphologic Classes
Version 1.1.0
Date 2023-02-14
Maintainer <NAME> <<EMAIL>>
Description Automatically segments a 3D array of voxels into mutually exclusive morphological ele-
ments. Thi... |
www_opennet_ru_docs_RUS_perl-maslov | free_programming_book | Unknown | # Введение в Perl
В книге приводятся начальные сведения по новому языку программирования Perl. Данный язык получил широкое распространение в связи с развитием компьютерной сети Интернет. Все примеры в книге проверены для Perl версии 5.003 операционной системы Unix FreeBSD 2.1.0.
Для программистов, системных администр... |
ed25519 | rust | Rust | Crate ed25519
===
RustCrypto: Ed25519
---
---
### Overview [¶](#pkg-overview)
SPDX-License-Identifier: BSD-3-Clause Copyright (c) 2022, Unikraft GmbH and The KraftKit Authors.
Licensed under the BSD-3-Clause License (the "License").
You may not use this file except in compliance with the License.
### Index ... |
RadialVisGadgets | cran | R | Package ‘RadialVisGadgets’
October 12, 2022
Type Package
Title Interactive Gadgets for Radial Visualization Approaches
Version 0.2.0
URL https://github.com/jmatute/RadialShinyGadgets
Imports import, ggplot2, tidyr, dplyr, miniUI, shiny, shinyjs, caret,
rlang, shinyscreensh... |
goji.io | go | Go | README
[¶](#section-readme)
---
### Goji
[](https://godoc.org/goji.io) [](https://travis-ci.org/goji/goji)
Goji is a HTTP request multiplexer, similar to [`net/http.ServeMux`](https://golang.org/pkg/net/h... |
momento | hex | Erlang | momento v0.1.2
API Reference
===
* [Modules](#modules)
Modules
===
[Momento](Momento.html)
Momento is an Elixir port of [Moment.js](https://github.com/moment/moment) for the purpose of parsing, validating,
manipulating, and formatting dates
[Momento.Add](Momento.Add.html)
This module holds all the various [`add/3`... |
glib | rust | Rust | Crate glib
===
Rust GLib and GObject bindings
---
**Rust** bindings and wrappers for GLib, part of gtk-rs-core.
GLib **2.56** is the lowest supported version for the underlying library.
This library contains bindings to GLib and GObject types and APIs as well as common building blocks used in both handmade and mach... |
MMAC | cran | R | Package ‘MMAC’
October 12, 2022
Type Package
Title Data for Mathematical Modeling and Applied Calculus
Version 0.1.2
Date 2018-7-2
Depends R (>= 2.10)
Suggests mosaic, mosaicCalc, manipulate
Author
<NAME> <<EMAIL>>, <NAME> <<EMAIL>>
Maintainer <NAME> <<EMAIL>>
Descriptio... |
stoppingrule | cran | R | Package ‘stoppingrule’
September 2, 2023
Type Package
Title Create and Evaluate Stopping Rules for Safety Monitoring
Version 0.3.0
Author <NAME> [aut,cre], <NAME> [aut]
Maintainer <NAME> <<EMAIL>>
Description Provides functions for creating, displaying, and evaluating stoppin... |
demodelr | cran | R | Package ‘demodelr’
October 13, 2022
Type Package
Title Simulating Differential Equations with Data
Version 1.0.1
Depends R (>= 4.1.0)
Description Designed to support the visualization, numerical computation,
qualitative analysis, model-data fusion, and stochastic simulat... |
plantuml_encoding | rust | Rust | Crate plantuml_encoding
===
Encoding and decoding text plantuml diagrams to facilitate communication of them through URL.
### Overview
Consider the next plain text plantuml diagram:
```
@startuml PUML -> RUST: HELLO
@enduml
```
It can be encoded to `0IO0sVz0StHXSdHrRMmAK5LDJ20jFY1ILLDKEY18HKnCJo0AG6LkP7LjR000` and w... |
@shardlabs/ethereumjs-vm | npm | JavaScript | SYNOPSIS
===
Implements Ethereum's VM in Javascript.
####
Fork Support
The VM currently supports the following hardfork rules:
* `Byzantium`
* `Constantinople`
* `Petersburg` (default)
* `Istanbul`
* `MuirGlacier` (only `mainnet` and `ropsten`)
If you are still looking for a [Spurious Dragon](https://eips.ethereu... |
C.pdf | free_programming_book | Unknown | C C
Notes for Professionals Notes for Professionals 300+ pages of professional hints and tricks GoalKicker.com Free Programming Books Disclaimer
This is an unocial free book created for educational purposes and is not aliated with ocial C group(s) or company(s).
All trademarks and registered trademarks are the property... |
js-sys | rust | Rust | Crate js_sys
===
Bindings to JavaScript’s standard, built-in objects, including their methods and properties.
This does *not* include any Web, Node, or any other JS environment APIs. Only the things that are guaranteed to exist in the global scope by the ECMAScript standard.
https://developer.mozilla.org/en-US/docs/... |
topepo_github_io_caret_index_html | free_programming_book | Unknown | # The caret Package
Date: 2019-03-27
Categories:
Tags:
# The
`caret` Package
2019-03-27
# 1 Introduction
The `caret` package (short for Classification And REgression Training) is a set of functions that attempt to streamline the process for creating predictive models. The package contains tools for:
* data ... |
cPCG | cran | R | Package ‘cPCG’
October 12, 2022
Type Package
Title Efficient and Customized Preconditioned Conjugate Gradient
Method for Solving System of Linear Equations
Version 1.0
Date 2018-12-30
Author <NAME>
Maintainer <NAME> <<EMAIL>>
Description Solves system of linear equatio... |
passerine | readthedoc | Python | Passerine ORM 1.4.0 documentation
[Passerine ORM](index.html#document-index)
---
Passerine[¶](#passerine)
===
| Author: | <NAME> <[<EMAIL>](mailto:j<EMAIL>)> |
A generic object relational mapper (ORM) and data abstraction layer (DAL) primarily designed for NoSQL databases.
You can **install Passerine** by running... |
timber_ecto | hex | Erlang | Timber.Ecto
===
Timber integration for Ecto.
Timber can hook into Ecto's Telemetry events to gather information about queries including the text of the query and the time it took to execute. This information is then logged as a [`Timber.Events.SQLQueryEvent`](https://hexdocs.pm/timber/3.1.0/Timber.Events.SQLQueryEven... |
php-cache | readthedoc | PHP | This library provides a PSR-6 compliant bridge between Doctrine and a Cache Pool. The bridge implements the
interface. This is useful for projects that require an implementation of
, but when you still want to use a PSR-6 implementation.
## Installation¶
```
composer require cache/psr-6-doctrine-bridge
```
## U... |
spaczz | readthedoc | Python | spaczz documentation
[spaczz](#)
---
spaczz: Fuzzy matching and more for spaCy[¶](#spaczz-fuzzy-matching-and-more-for-spacy)
===
Reference[¶](#reference)
---
* [spaczz.matcher](#module-spaczz.matcher)
* [spaczz.pipeline](#module-spaczz.pipeline)
* [spaczz.registry](#module-spaczz.registry)
* [spaczz.customattrs](#... |
time | cran | R | Package ‘time.slots’
October 14, 2022
Title Display Data in a Weekly Calendar View
Version 0.2.0
Description Generate weekly timetables as a ggplot2 layer. Add
informative timeslots with elements such as title, key-value pairs, or
colour to reveal trends.
License MIT... |
owls | readthedoc | Unknown | # AD Leo
Date: 2021-06-02
Categories:
Tags:
AD Leo# Search exo.mast Search SIMBAD Date S err 2/6/2021 8:39:32 8.660 0.361 1/12/2022 10:42:27 9.671 0.404 TESS Light Curve#
# EV Lac
Date: 2021-11-09
Categories:
Tags:
EV Lac# Search exo.mast Search SIMBAD Date S err 9/11/2021 9:26:27 9.614 0.405 9/13/2022 9:34:17... |
github.com/go-vgo/robotgo | go | Go | README
[¶](#section-readme)
---
### Robotgo
[](https://github.com/go-vgo/robotgo/commits/master)
[](https://circleci.com/gh/go-vgo/robotgo)
[
**ReFrame 4.1.1**
---
Welcome to ReFrame[¶](#welcome-to-reframe)
===
ReFrame is a powerful framework for writing system regression tests and benchmarks, specifically targeted to HPC systems.
The goal of the framework is to abstract away the complexity of the interactions wi... |
nerodia | readthedoc | Python | Nerodia 0.5.0 documentation
[Nerodia](index.html#document-index)
---
---
Nerodia[¶](#nerodia)
===
Nerodia is a Python port of the Watir ruby gem. <https://github.com/watir/watirInstallation[¶](#installation)
---
### Supported Python Versions[¶](#supported-python-versions)
* Python 2.7
* Python 3.4+
### Installi... |
github.com/gomodule/redigo/redis | go | Go | None
Documentation
[¶](#section-documentation)
---
### Overview [¶](#pkg-overview)
* [Connections](#hdr-Connections)
* [Executing Commands](#hdr-Executing_Commands)
* [Pipelining](#hdr-Pipelining)
* [Concurrency](#hdr-Concurrency)
* [Publish and Subscribe](#hdr-Publish_and_Subscribe)
* [Reply Helpers](#hdr-Reply_... |
rainbow-code | npm | JavaScript | Rainbow
===
Rainbow is a code syntax highlighting library written in Javascript.
It was designed to be lightweight (~2.5kb), easy to use, and extendable.
It is completely themable via CSS.
Demo
---
You can see rainbow in action at <http://rainbowco.de>.
You can also build/download custom packages from there.
Con... |
rebird | cran | R | Package ‘rebird’
October 14, 2022
Version 1.3.0
Title R Client for the eBird Database of Bird Observations
Description A programmatic client for the eBird database
(<https://ebird.org/home>), including functions for searching for bird
observations by geographic loc... |
rebus | cran | R | Package ‘rebus.datetimes’
November 3, 2022
Title Date and Time Extensions for the 'rebus' Package
Version 0.0-2
Date 2022-10-16
Author <NAME> [aut, cre]
Maintainer <NAME> <<EMAIL>>
Description Build regular expressions piece by piece using human readable code.
This packag... |
GITHUB_raaminz_training.zip_unzipped_6_-_Http_Status_Codes_And_Exception_Handling.pdf | free_programming_book | Unknown | Http Status Codes
& Exception Handling
Presented by <NAME> Twitter: @zare88r
HTTP Status Codes
Http Status header
HTTP Status Codes
1xx
100 continue
101 Switching Protocols
2xx
200 OK
Redirection 3xx
303 SEE OTHER
HTTP Status Codes
4
400 Bad Request
401 Unauthorized
403 Forbidden
5
https://httpstatuses.com/... |
EMpeaksR | cran | R | Package ‘EMpeaksR’
March 29, 2023
Type Package
Title Conducting the Peak Fitting Based on the EM Algorithm
Version 0.3.1
Description The peak fitting of spectral data is performed by using the frame work of EM algo-
rithm. We adapted the EM algorithm for the peak fittin... |
@mattleff/react-native-track-player | npm | JavaScript | ### 📢 The Android side of RN track player is currently being rewritten with Kotlin and [KotlinAudio](https://github.com/DoubleSymmetry/KotlinAudio). Help us test! [More information here.](https://github.com/DoubleSymmetry/react-native-track-player/discussions/1264)
---
A fully-fledged audio module created for music ... |
kraftkit.sh/unikraft | go | Go | None
Documentation
[¶](#section-documentation)
---
### Overview [¶](#pkg-overview)
SPDX-License-Identifier: BSD-3-Clause Copyright (c) 2022, Unikraft GmbH and The KraftKit Authors.
Licensed under the BSD-3-Clause License (the "License").
You may not use this file except in compliance with the License.
SPDX-Licen... |
github.com/nicksherron/bashhub-server | go | Go | README
[¶](#section-readme)
---
### bashhub-server
[](https://goreportcard.com/report/github.com/nicksherron/bashhub-server) [](https://travis-ci.o... |
SNSchart | cran | R | Package ‘SNSchart’
October 12, 2022
Type Package
Title Sequential Normal Scores in Statistical Process Management
Version 1.4.0
Author <NAME> [aut],
<NAME> [aut, cre],
<NAME> [ctb]
Maintainer <NAME> <<EMAIL>>
Description The methods discussed in this package are ne... |
Rcapture | cran | R | Package ‘Rcapture’
October 12, 2022
Type Package
Title Loglinear Models for Capture-Recapture Experiments
Version 1.4-4
Date 2022-05-03
Description Estimation of abundance and other demographic parameters for closed
populations, open populations and the robust design in ... |
github.com/lilien1010/taskq/v3 | go | Go | README
[¶](#section-readme)
---
### Golang asynchronous task/job queue with Redis, SQS, IronMQ, and in-memory backends
[](https://pkg.go.dev/github.com/vm... |
cencrne | cran | R | Package ‘cencrne’
January 9, 2023
Title Consistent Estimation of the Number of Communities via
Regularized Network Embedding
Version 1.0.0
Description The network analysis plays an important role in numerous application domains includ-
ing biomedicine.
Estim... |
hep | ctan | TeX | # The hep-reference package+
Footnote †: This document corresponds to hep-reference v1.2.
Adjustments of standard references, footnote, and citations
<NAME>
<EMAIL>
2023/07/01
###### Abstract
The hep-reference package applies some light convenience modifications to the reference, citation and footnote macros afte... |
bigmds | cran | R | Package ‘bigmds’
October 12, 2022
Title Multidimensional Scaling for Big Data
Version 2.0.1
Description MDS is a statistic tool for reduction of dimensionality, using as input a distance
matrix of dimensions n × n. When n is large, classical algorithms suffer from
c... |
dissolver | hex | Erlang | Dissolver
===
###
NOTE: This is a wip repo. So be warned there maybe be bugs.
This project is a fork of <https://github.com/elixirdrops/kerosene>.
I thought to take it over because it does not look as if its being activly developed and I wanted more features.
TODO:
* [x] Lazy query - Instead of pagination cal... |
tapkee | cran | R | Package ‘tapkee’
October 14, 2022
Type Package
Title Wrapper for 'tapkee' Dimension Reduction Library
Version 1.2
Date 2020-12-20
Author <NAME>
Maintainer <NAME> <<EMAIL>>
Description Wrapper for using 'tapkee' command line utility,
it allows to run it from inside R and c... |
@next-auth/upstash-redis-adapter | npm | JavaScript | ###
**Upstash Redis Adapter** - NextAuth.js
Open Source. Full Stack. Own Your Data.
Overview
---
This is the Upstash Redis adapter for [`next-auth`](https://authjs.dev). This package can only be used in conjunction with the primary `next-auth` and `@upstash/redis` packages. It is not a standalone package.
Getti... |
kiwi.js | npm | JavaScript | THIS REPOSITORY IS NO LONGER MAINTAINED
===
Looking for an alternative? Please have a look at [lume/kiwi](https://github.com/lume/kiwi).
If you are interested in maintaining this repository (and taking ownership of it), please reach out to me [here](mailto:<EMAIL>).
kiwi.js
===
Kiwi.js is a fast TypeScript implemen... |
projmgr | cran | R | Package ‘projmgr’
October 14, 2022
Title Task Tracking and Project Management with GitHub
Version 0.1.0
Description Provides programmatic access to 'GitHub' API with a
focus on project management. Key functionality includes
setting up issues and milestones from R ... |
resampledata | cran | R | Package ‘resampledata’
October 14, 2022
Type Package
Title Data Sets for Mathematical Statistics with Resampling in R
Version 0.3.1
Suggests resample (>= 0.4)
Maintainer <NAME> <<EMAIL>>
Description Package of data sets from ``Mathematical Statistics
with Resampling in R... |
growthcleanr | cran | R | Package ‘growthcleanr’
March 3, 2023
Type Package
Title Data Cleaner for Anthropometric Measurements
Version 2.1.1
Description Identifies implausible anthropometric (e.g., height,
weight) measurements in irregularly spaced longitudinal datasets, such as those from elec-... |
Subsets and Splits
Python Algorithms in Free Books
Retrieves entries related to both Python and algorithms from a specific source, providing a filtered dataset but lacking broader analytical depth.
Python Code from Free Books
This query retrieves a limited number of records related to Python programming books, providing basic filtering but minimal analytical insight.