added
stringdate 2025-04-01 04:05:38
2025-04-01 07:14:06
| created
timestamp[us]date 2001-10-09 16:19:16
2025-01-01 03:51:31
| id
stringlengths 4
10
| metadata
dict | source
stringclasses 2
values | text
stringlengths 0
1.61M
|
|---|---|---|---|---|---|
2025-04-01T04:35:26.710688
| 2024-06-16T12:40:43
|
2355741914
|
{
"authors": [
"ratmaiden",
"sbcarp"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10599",
"repo": "sbcarp/TopazVideoPauser",
"url": "https://github.com/sbcarp/TopazVideoPauser/issues/5"
}
|
gharchive/issue
|
Does not pause the video render and the tray icon is a yellow lock
I've tried pausing the render, but it won't stop nor freeze the UI at all. I've tried multiple times. I'm still using the 2.2.0 version of Topaz though, maybe that's the problem? Can't update, because it's on my PC from work and they've frozen any app installs for some odd reason.
Hi, 2.2.0 is little too old to support, but if it's an esay fix I'll do it, what is the process name for your Topaz program? You may find it in Task Manager -> Details -> check for process has topaz keyword or anything related
Hi, thanks for replying. Its process name is "Topaz Video Enhance AI.exe" I
hope it works, but if it won't, it's fine. No pressure.
On Wed, 19 Jun 2024, 1:52 am Han Fang, @.***> wrote:
Hi, 2.2.0 is little too old to support, but if it's an esay fix I'll do
it, what is the process name for your Topaz program? You may find it in
Task Manager -> Details -> check for process has topaz keyword or anything
related
—
Reply to this email directly, view it on GitHub
https://github.com/sbcarp/TopazVideoPauser/issues/5#issuecomment-2176657224,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/AYYL75ZN5STPFEMPPMGUVKLZIBXWVAVCNFSM6AAAAABJMSFETKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNZWGY2TOMRSGQ
.
You are receiving this because you authored the thread.Message ID:
@.***>
Okay, I added the process name to the app, you can download the latest version to try it, hope it works!
Hey man, you see the thing is, it still doesn't work. However, it's fine. It's not your fault, it's just the shitty company policy's fault. I'll probably just use my home PC for enhancing videos. Thanks for trying, at least.
https://github.com/sbcarp/TopazVideoPauser/assets/103858167/876cb7c2-08c2-49e4-9da5-22e275cef969
Thanks for the video.
I see what's going on here, Topaz Video 2.x doesn't spawn ffmpeg process for video tasks, but doing all the work in it's own process. I think there's a way to fix it without modifying current project structure.
New version is published, welcome to try.
Hey man, great news. IT WORKS!!! Great job, and thanks. I really appreciate it!
|
2025-04-01T04:35:26.740122
| 2017-07-27T21:06:15
|
246169200
|
{
"authors": [
"dwijnand"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10601",
"repo": "sbt/sbt",
"url": "https://github.com/sbt/sbt/issues/3379"
}
|
gharchive/issue
|
Re-enable (neo-)sbt-scalafmt
Now that v1.9 has been release cross-built for sbt 1.
[x] https://github.com/sbt/io/pull/71
[x] https://github.com/sbt/util/pull/118
[x] https://github.com/sbt/librarymanagement/pull/158
[ ] https://github.com/sbt/zinc/pull/392
[ ] https://github.com/sbt/sbt/pull/3418
|
2025-04-01T04:35:26.745605
| 2019-05-01T16:06:26
|
439219396
|
{
"authors": [
"BillyAutrey",
"eed3si9n"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10602",
"repo": "sbt/sbt",
"url": "https://github.com/sbt/sbt/issues/4642"
}
|
gharchive/issue
|
Feature Request: SettingKey to ignore test failures
[x] I've read the CONTRIBUTING guidelines
Steps
Run sbt test in any project with a test that runs, but fails assertions.
Problem
Test failures will throw an sbt.TestsFailedException when tests are run. This is caused by TestResultLogger, specifically here.
The result of this behavior is that Jenkins pipelines who execute testing as a step in a larger build pipeline can fail before the pipeline completes. This exits SBT with a non-zero exit code. I have encountered users who are willing to continue pipeline execution if a single test fails in the code base, but also who want to halt pipeline execution if a test fails to compile. This would be resolved with an option to ignore test execution failures without throwing a TestsFailedException.
Expectation
Similar to Maven's <testFailureIgnore>true</testFailureIgnore> option in Surefire, it would be useful to have a boolean SettingKey (maybe testFailureIgnore?).
If the setting is false, SBT behaves as it does now. This is the default setting.
If the setting is set to true:
When sbt test is run and all tests succeed, no change in behavior.
When sbt test is run and a test is run, but fails assertions, no exceptions are thrown. Exit 0.
When sbt test is run, but test compilation fails, exit with an error. No change here.
Notes
This is the current workaround to achieve this behavior.
sbt version: 1.2.x
https://www.scala-sbt.org/1.x/docs/Tasks.html#Handling+Failure might be used to create a testFailOk task.
|
2025-04-01T04:35:26.751644
| 2020-08-20T14:55:18
|
682783565
|
{
"authors": [
"SethTisue",
"eed3si9n",
"eugeniyk",
"rfan-debug"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10603",
"repo": "sbt/sbt",
"url": "https://github.com/sbt/sbt/issues/5785"
}
|
gharchive/issue
|
Concurrency issue while compiling compiler bridge on a fresh machine?
steps
Recently I ran into this while sending https://github.com/sbt/sbt/pull/5767/files.
If you change the sbt/src/sbt-test/plugins/dotty-sandwich/test to
> compile
> compile
and wipe out Travis CI cache you might be able to reproduce what I saw.
problem
The problem you come across may be somewhat arbitrary but one example is "Cannot redefine component"
[info] Running plugins/dotty-sandwich
[info] [info] welcome to sbt 1.4.0-SNAPSHOT (AdoptOpenJDK Java 11.0.1)
[info] [info] loading settings for project global-plugins from metals.sbt ...
[info] [info] loading global plugins from /tmp/sbt_d7daa57/global/plugins
[info] [info] loading settings for project sbt_d7daa57-build from plugins.sbt ...
[info] [info] loading project definition from /tmp/sbt_d7daa57/project
[info] [info] loading settings for project root from build.sbt ...
[info] [info] set current project to root (in build file:/tmp/sbt_d7daa57/)
[info] [info] compiling 1 Scala source to /tmp/sbt_d7daa57/foo-core/target/scala-2.13/classes ...
[info] [info] compiling 1 Scala source to /tmp/sbt_d7daa57/bar-core/target/scala-0.23/classes ...
[info] [info] Non-compiled module 'compiler-bridge_2.13' for Scala 2.13.4-bin-aeee8f0. Compiling...
[info] [info] done compiling
[info] [info] compiling 1 Scala source to /tmp/sbt_d7daa57/bar-app/target/scala-2.13/classes ...
[info] [info] Non-compiled module 'compiler-bridge_2.13' for Scala 2.13.4-bin-aeee8f0. Compiling...
[info] [info] Compilation completed in 18.539s.
[info] [info] Compilation completed in 24.223s.
[info] [info] Attempting to fetch org.scala-sbt:compiler-bridge_2.13:1.4.0-M10.
[info] [error] ## Exception when compiling 1 sources to /tmp/sbt_d7daa57/foo-core/target/scala-2.13/classes
[info] [error] sbt.internal.scriptedtest.ScriptedLauncher$3$1$1$1$1$1: sbt.internal.scriptedtest.ScriptedLauncher$3$1$1$1$foo: sbt.internal.scriptedtest.ScriptedLauncher$3$1$1$1$foo: sbt.internal.scriptedtest.ScriptedLauncher$3$1$1$1$foo: Cannot redefine component. ID: sbt.internal.scriptedtest.ScriptedLauncher$1@73f23e8e, files: /tmp/sbt_159a0475/org.scala-sbt-compiler-bridge_2.13-1.4.0-M10-bin_2.13.4-bin-aeee8f0__55.0.jar
[info] [error]
[info] [error]
[info] [info] done compiling
[info] [error] sbt.internal.scriptedtest.ScriptedLauncher$3$1$1$1$1$1: sbt.internal.scriptedtest.ScriptedLauncher$3$1$1$1$foo: sbt.internal.scriptedtest.ScriptedLauncher$3$1$1$1$foo: sbt.internal.scriptedtest.ScriptedLauncher$3$1$1$1$foo: Cannot redefine component. ID: sbt.internal.scriptedtest.ScriptedLauncher$1@73f23e8e, files: /tmp/sbt_159a0475/org.scala-sbt-compiler-bridge_2.13-1.4.0-M10-bin_2.13.4-bin-aeee8f0__55.0.jar
[info] [error] Caused by: sbt.internal.scriptedtest.ScriptedLauncher$3$1$1$1$foo: sbt.internal.scriptedtest.ScriptedLauncher$3$1$1$1$foo: sbt.internal.scriptedtest.ScriptedLauncher$3$1$1$1$foo: Cannot redefine component. ID: sbt.internal.scriptedtest.ScriptedLauncher$1@73f23e8e, files: /tmp/sbt_159a0475/org.scala-sbt-compiler-bridge_2.13-1.4.0-M10-bin_2.13.4-bin-aeee8f0__55.0.jar
[info] [error] (fooCore / Compile / compileIncremental) sbt.internal.scriptedtest.ScriptedLauncher$3$1$1$1$1$1: sbt.internal.scriptedtest.ScriptedLauncher$3$1$1$1$foo: sbt.internal.scriptedtest.ScriptedLauncher$3$1$1$1$foo: sbt.internal.scriptedtest.ScriptedLauncher$3$1$1$1$foo: Cannot redefine component. ID: sbt.internal.scriptedtest.ScriptedLauncher$1@73f23e8e, files: /tmp/sbt_159a0475/org.scala-sbt-compiler-bridge_2.13-1.4.0-M10-bin_2.13.4-bin-aeee8f0__55.0.jar
[error] x plugins/dotty-sandwich
[error] Cause of test exception: {line 1} Command failed: compile failed
expectation
Some form of concurrency control and logging around this.
Keep having this daily:
sbt 1.4.6, scala 2.12.x
I am also having the same issue:
sbt 1.3.13, scala 2.12.x, open-jdk 11.
I am kinda not sure if the issue came from the jdk 11 upgrade. We used to use JDK8 and there was no such an issue.
but has it ever happened with the current release (sbt 1.5.4)?
My datapoint is: after switching back to openjdk8, this issue does not happen any more.
My setting: sbt 1.3.13, scala 2.12.13, open-jdk8
Yea. According to https://github.com/sbt/sbt/commit/3c955abea626141b5e959a2075f2d58e116f93df, the earliest fixed version is 1.5.0, so please try the current stable first.
|
2025-04-01T04:35:26.753359
| 2020-10-08T16:38:29
|
717499424
|
{
"authors": [
"eatkins"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10604",
"repo": "sbt/sbt",
"url": "https://github.com/sbt/sbt/pull/5960"
}
|
gharchive/pull-request
|
Set SimpleTerminal properties based on system props
When running sbt 1.4.0 with -Dsbt.ci=true and -Dsbt.color=true, there is
no color output. This was because in this scenario, a SimpleTerminal was
used and isAnsiSupported and isColorEnabled were hardcoded to false
rather than reading the values from the system properties.
Fixes #5958
|
2025-04-01T04:35:26.758713
| 2023-09-06T05:29:31
|
1883252969
|
{
"authors": [
"rin-st",
"technophile-04"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10605",
"repo": "scaffold-eth/scaffold-eth-2",
"url": "https://github.com/scaffold-eth/scaffold-eth-2/pull/522"
}
|
gharchive/pull-request
|
fix: test file name
Description
Renamed test file
Additional Information
[x] I have read the contributing docs (if this is your first contribution)
[x] This is not a duplicate of any existing pull request
Your ENS/address:
0xrinat
Ohh lol tysm Rinat !
|
2025-04-01T04:35:26.779045
| 2015-10-20T19:23:56
|
112442092
|
{
"authors": [
"gzm0",
"sjrd"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10606",
"repo": "scala-js/scala-js",
"url": "https://github.com/scala-js/scala-js/issues/1961"
}
|
gharchive/issue
|
Honor call-site @inline and @noinline
The feature was introduced in Scala/JVM in https://github.com/scala/scala/pull/4801
Can we move this post-v1?
We should at least figure out now how we are going to represent that information in the IR. Could they be two additional flags in the ApplyFlags of Apply* nodes? Currently the ApplyFlags are all semantically binding, whereas @inline and @noinline would be optimization hints.
Ah, OK fair enough.
Thinking more about this, I'm confident we can use the ApplyFlags for this purpose, so delaying to post-v1.0.0.
I've started looking into this:
Changes in the IR/optimizer seem trivial.
It seems we have forgotten flags on the New tree. So we cannot do this for class inlining.
|
2025-04-01T04:35:26.783110
| 2015-10-30T09:50:36
|
114235296
|
{
"authors": [
"sjrd"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10607",
"repo": "scala-js/scala-js",
"url": "https://github.com/scala-js/scala-js/issues/1990"
}
|
gharchive/issue
|
Node.js 4's console does not deduplicate % anymore -> our test breaks
Failing build log: https://scala-webapps.epfl.ch/jenkins/job/scalajs-task-worker/66687/console
We need to remove the hack at https://github.com/scala-js/scala-js/blob/6e3255a6c0e502c317b9d14a645dcb091a3407aa/js-envs/src/main/scala/org/scalajs/jsenv/nodejs/NodeJSEnv.scala#L276-L296 to get back a normal behavior.
I don't know whether we can have a "smart hack" that duplicates %'s on old Nodes, and doesn't on recent ones.
We could use process.version, and test whether it returns something older than "v4.something", i.e., process.version starts with v0., v1., v2. or v3..
Actually, it seems to have been fixed in v2.1.0 already: https://github.com/nodejs/node/commit/214d02040e1c46b8cac47451bf2f0e2fc15a3e0f
Starting with v2.1.0, console.log will not consider %% an escaped % if console.log is given only 1 argument. Apparently this behavior is consistent with browsers, including Firefox.
So we should try and use process.version to detect old versions (below v2.1.0), and only apply our hack for those.
|
2025-04-01T04:35:26.791922
| 2024-06-11T12:41:45
|
2346332909
|
{
"authors": [
"sjrd",
"tanishiking"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10608",
"repo": "scala-js/scala-js",
"url": "https://github.com/scala-js/scala-js/issues/4997"
}
|
gharchive/issue
|
Allow for linktime conditional branching
Part of https://github.com/scala-js/scala-js/issues/4991
Motivation
The motivation behind this feature was to allow switching the implementation based on whether the target is Wasm running in the browser (WasmJS) or a standalone Wasm application. For example, with regular expressions, WasmJS could reuse the JavaScript Regex implementation, while a standalone Wasm application would use a pure-Scala Regex implementation.
While the original motivation is from Wasm use cases, the feature itself is self-contained and can be used for optimization purposes in the JavaScript backend as well.
How to
The main idea is to introducing the new IR node LInkTimeIf like:
To put into writing something we discussed offline at some point: I believe the easiest and cleanest way to do this is with a new IR node that would look like
case class LinkTimeIf(cond: LinkTimeCondition, thenp: Tree, elsep: Tree) extends Tree
Then, our 3 main link-time passes process it as follows:
the Analyzer, for the reachability analysis, deals with it during the Infos builder. Since the Infos builder walks the IR trees to build Infos, it can resolve the LinkTimeCondition and only recurse in the appropriate branch
the OptimizerCore would trivially replace the LinkTimeIf with either thenp or elsep, allowing further optimizations
the backends would do the same, if any LinkTimeIf survives until then (normally, only if the optimizer is disabled)
The precise nature of LinkTimeCondition remains to be determined, but it would fairly limited. Definitely not a full Tree.
Originally posted by @sjrd in https://github.com/scala-js/scala-js/issues/4991#issuecomment-2154680386
Where the LinkTimeCondition should be like same as scala-native's LinktimeCondition
How to: frontend
Introduce @LinkTime (or @resolvedAtLinktime like SN) annotation, only the annotated values (functions?) can be used as part of LinkTimeCondition
if expression will be translated to LinkTimeIf instead of If, when the condition is only composed of @LinkTime annotated value, Literals, and BinaryOps.
For example, the if will be translated to LinkTimeIf because the condition is composed of @LinkTime annotated booleans + BinaryOp. (if it's mixed with the runtime-value, we should fail linking).
@LinkTime
def foo: Boolean = ...
@LinkTime
def bar: Boolean = ...
if (foo && bar) { ... }
I'm not sure about magically turning an if into a LinkTimeIf depending on some non-local information. That would be quite brittle, as there would no compile-time reporting if something that should be a LinkTimeIf falls back to a regular If.
I believe the more Scala.js way to do this would be a dedicated primitive in scala.scalajs.LinkingInfo. Something that we would write
LinkingInfo.linkTimeIf(LinkingInfo.isWebAssembly) {
ifTrue
} {
ifFalse
}
|
2025-04-01T04:35:26.806462
| 2019-05-16T10:00:40
|
444864665
|
{
"authors": [
"amitjoy",
"retronym"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10609",
"repo": "scala/jardiff",
"url": "https://github.com/scala/jardiff/issues/32"
}
|
gharchive/issue
|
Exclude files from diff calculation
Is there any possibility to exclude specific files from the diff calculation? I think that would really be an interesting feature.
One possibility is to use -g to persist the diffs to an on-disk Git repository, and then use Git to browse subsets of the changelog.
I'd also be open to extending the command line options of jardiff itself to support this use case directly (it would be faster to exclude before diffing).
I've added this feature to 0.3.0 in #36
|
2025-04-01T04:35:26.827777
| 2022-06-07T01:20:40
|
1262604190
|
{
"authors": [
"Philippus",
"devosalain"
],
"license": "apache-2.0",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10610",
"repo": "scala/scala-swing",
"url": "https://github.com/scala/scala-swing/issues/164"
}
|
gharchive/issue
|
which jdk version is adviced ?
I have openjdk versions 8,11,12,13,14,15,16,17 installed.
But which one would you advice to use ?
I think you can follow this: https://docs.scala-lang.org/overviews/jdk-compatibility/overview.html
This library is tested on JDK 8, 11 and 17: https://github.com/scala/scala-swing/blob/work/.github/workflows/ci.yml#L13
|
2025-04-01T04:35:26.892971
| 2015-10-19T08:12:26
|
112094635
|
{
"authors": [
"xeno-by"
],
"license": "bsd-3-clause",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10611",
"repo": "scalameta/scalameta",
"url": "https://github.com/scalameta/scalameta/issues/270"
}
|
gharchive/issue
|
support for SAMs
As brought up by @darkdimius, we should probably have special treatment for lambdas that are used in SAM positions. This needs to be thought through and, if necessary, implemented.
Semantic API has evolved significantly since this issue has been opened. Instead of trying to marry syntactic and semantic representations of code, while providing support for a full range of language features, we are doing baby steps, adding features as necessary for our roadmap.
As a result, we're limiting the surface of our work to actionable items only - either bugs in existing code or feature requests that are warranted by the roadmap. This ticket is neither, so I'm going to close it.
|
2025-04-01T04:35:26.897840
| 2014-10-29T20:26:18
|
47199108
|
{
"authors": [
"dlwh",
"mikeunwin",
"quentinms"
],
"license": "apache-2.0",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10612",
"repo": "scalanlp/breeze",
"url": "https://github.com/scalanlp/breeze/issues/320"
}
|
gharchive/issue
|
Creating CSCMatrix from SparseVector throws ArrayIndexOutOfBondException when the active size is a power of 2
I have the following snippet
for (fill ← 0 until 2000) {
val vb = new VectorBuilder[Int](421337)
for (i ← 0 to fill) {
vb.add(i,i)
}
val sv = vb.toSparseVector
try{
sv.asCSCMatrix()
} catch {
case t: Throwable => println("It crashed for: "+sv.activeSize)
}
}
Which gives me this output:
It crashed for: 1
It crashed for: 2
It crashed for: 4
It crashed for: 8
It crashed for: 16
It crashed for: 32
It crashed for: 64
It crashed for: 128
It crashed for: 256
It crashed for: 512
It crashed for: 1024
I tried looking at the source but I'm not familiar enough with it to understand what causes this.
I am using scala 2.10.4 and Breeze 0.10
The complete error is:
Exception in thread "main" java.lang.ArrayIndexOutOfBoundsException: 1
at breeze.linalg.SparseVector$mcI$sp$$anonfun$4.apply$mcII$sp(SparseVector.scala:159)
at breeze.linalg.SparseVector$mcI$sp$$anonfun$4.apply(SparseVector.scala:158)
at breeze.linalg.SparseVector$mcI$sp$$anonfun$4.apply(SparseVector.scala:158)
at scala.Array$.tabulate(Array.scala:331)
at breeze.linalg.SparseVector$mcI$sp.asCSCMatrix$mcI$sp(SparseVector.scala:158)
at XXX.tests$$anonfun$1.apply(tests.scala:17)
at XXX.tests$$anonfun$1.apply(tests.scala:7)
at scala.collection.immutable.Range.foreach(Range.scala:141)
at XXX.tests$delayedInit$body.apply(tests.scala:7)
at scala.Function0$class.apply$mcV$sp(Function0.scala:40)
at scala.runtime.AbstractFunction0.apply$mcV$sp(AbstractFunction0.scala:12)
at scala.App$$anonfun$main$1.apply(App.scala:71)
at scala.App$$anonfun$main$1.apply(App.scala:71)
at scala.collection.immutable.List.foreach(List.scala:318)
at scala.collection.generic.TraversableForwarder$class.foreach(TraversableForwarder.scala:32)
at scala.App$class.main(App.scala:71)
at XXX.tests$.main(tests.scala:5)
at XXX.tests.main(tests.scala)
Hi all, did this ever get patched?
nope. @gabeos ??
On Mon, May 18, 2015 at 10:47 AM, mikeunwin<EMAIL_ADDRESS>wrote:
Hi all, did this ever get patched?
—
Reply to this email directly or view it on GitHub
https://github.com/scalanlp/breeze/issues/320#issuecomment-103146818.
apparently already fixed...
|
2025-04-01T04:35:26.899695
| 2018-07-17T13:08:39
|
341911981
|
{
"authors": [
"jdegoes",
"tuleism"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10613",
"repo": "scalaz/scalaz-zio",
"url": "https://github.com/scalaz/scalaz-zio/pull/116"
}
|
gharchive/pull-request
|
Simplify name for IOApp, IOQueue and IORef
Fixes #114.
@tuleism Looks great, thank you! Can you resolve the conflicts?
Done!
@tuleism One more minor conflict. Sorry for the rash of PRs being merged!
Done :smirk_cat:
@tuleism Looks great. Thank you! 😄
|
2025-04-01T04:35:26.929433
| 2021-08-09T09:34:06
|
963812667
|
{
"authors": [
"adriengibrat"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10614",
"repo": "scaleway/scaleway-ui",
"url": "https://github.com/scaleway/scaleway-ui/pull/618"
}
|
gharchive/pull-request
|
chore: remove Box component typing hacks
Summary
Improve Box component typings
& use snapshots classname generation based on style content to reduce false positive snaphosts test failure.
Type
Enhancement
Summarise concisely:
A visual testing was done locally with loki to ensure there was no regression.
What is expected?
No change at all, only component typings refactor.
Review c1ba7c0, the other commit only contains snapshot updates.
No significant improvement in replacing a hack by another, but the emotion snapshot config is intersting.
Closing this PR in favor of https://github.com/scaleway/scaleway-ui/pull/641
|
2025-04-01T04:35:26.991181
| 2023-04-11T08:55:12
|
1662043369
|
{
"authors": [
"codecov-commenter",
"tomaz-s1"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10615",
"repo": "scalyr/dataset-go",
"url": "https://github.com/scalyr/dataset-go/pull/3"
}
|
gharchive/pull-request
|
Add status badges to the readme
This pull request adds GHA workflow and codecov.io status badge to the readme.
It also removes codecov token from the workflow - this token is not needed anymore for integration now that this repo is public.
I also pushed a change so code quality workflow now also triggers for pull requests and also runs on a daily basis as part of a cron trigger.
@martin-majlis-s1 code coverage step is failing, not sure why - https://github.com/scalyr/dataset-go/actions/runs/4665731080/jobs/8259472292?pr=3
Codecov Report
:exclamation: No coverage uploaded for pull request base (main@161c71b). Click here to learn what that means.
The diff coverage is n/a.
@@ Coverage Diff @@
## main #3 +/- ##
=======================================
Coverage ? 60.93%
=======================================
Files ? 9
Lines ? 901
Branches ? 0
=======================================
Hits ? 549
Misses ? 300
Partials ? 52
:mega: We’re building smart automated test selection to slash your CI/CD build times. Learn more
Per our discussion on Slack - I think it's fine to run it on push to any branch, but only run it for prs against main branch (which is the most common workflow).
I don't think it should run twice though. We use similar triggers in other repos and it only runs once.
|
2025-04-01T04:35:27.000797
| 2023-04-26T13:06:04
|
1684996413
|
{
"authors": [
"bunufi",
"scambra"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10616",
"repo": "scambra/devise_invitable",
"url": "https://github.com/scambra/devise_invitable/issues/889"
}
|
gharchive/issue
|
When is the next planned release?
I have noticed that turbo issue, when validation errors are not displayed in "set password" form, is fixed with: https://github.com/scambra/devise_invitable/commit/87228c1feefa0fea1d74f82fe7867111fddc7f91
This is great and is exactly what we need for our app!
Currently we are requiring devise_invitable via git in the gemfile and would like to move to regular releases.
When the next release is planned?
There is really no rush and I have set up a github notification for this. I just think that this will be very relevant for other stacks with this gem and hotwire.
Thanks for creating and maintaining this gem!
2.0.8 released
Amazing! Thanks a lot for such a quick response and release.
|
2025-04-01T04:35:27.027262
| 2022-11-28T14:41:08
|
1466550119
|
{
"authors": [
"alexandereneroth",
"mJarsater"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10617",
"repo": "scania-digital-design-system/sdds",
"url": "https://github.com/scania-digital-design-system/sdds/pull/659"
}
|
gharchive/pull-request
|
docs(canvas): made canvas the default tab
Describe pull-request
Removed the parameter setting docs as the default tab.
Solving issue
Fixes: AB#2729
How to test
Go to storybook link below
Check in a couple of random stories.
Make sure canvas is the default tab.
Screenshots
Additional context
Also made canvas the first tab, which i think makes sence since its the default.
|
2025-04-01T04:35:27.056987
| 2023-09-13T22:21:31
|
1895342767
|
{
"authors": [
"Garfield100",
"chkothe"
],
"license": "BSD-2-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10619",
"repo": "sccn/xdf",
"url": "https://github.com/sccn/xdf/issues/64"
}
|
gharchive/issue
|
Mandatory XML content
Hello again,
In the StreamHeader chunk section of the specification, the given XML example is described as "[...] the minimal XML content as populated by an early client [...]". It contains quite a few tags.
In contrast, the streams in the minimal.xdf example file have fewer tags.
The following tags are shared between the two sources:
name
type
channel_count
nominal_srate
channel_format
created_at
uid
These additional tags, not found in minimal.xdf, are included in the spec example:
source id
version
session_id
hostname
minimal.xdf also has the desc tag, which is not in the spec. Although it is empty, it is there.
The situation is similar with the StreamFooter XML with tags like writer, clock_offsets, and measured_srate.
So I ask: which tags are truly mandatory for an XDF file and which are optional?
And why are clock offsets included again in the footer when they already have their own chunks?
Hi Enrico, the XDF spec follows the usual convention of using the "must" and "should" keywords to indicate what's mandatory vs advised but not required, and therefore only the fields where the spec says they must be present, are formally mandatory (those are channel_count, nominal_srate, and channel_format), since those are the fields needed to decode the binary time-series payload. As such, the mandatory core subset of XDF is a bare multi-stream time-series container where all content meta-data is optional.
However, there are a 3 additional fields where the spec says they should be present for actual interpretability of the data, which are name, type, and desc (btw note that desc is actually both in the spec text and in the example snippet), and much of the practical value of XDF comes from the content of those fields. For that reason, applications that load XDF files will probably complain or error out if those three fields are not present, since most applications do something that depends on the content type ("modality") of the stream. This is enabled by additional modality-specific appendix pages of the spec that describe pre-defined values for type (the stream content-type), and pre-defined sub-fields for desc according to that content-type. XDF is open-ended / extensible with regards to the kinds of data modalities that may be stored in it, and most practical applications will only write or interpret the subset of desc that is relevant to the application's domain -- for example, typical EEG software will not interpret (and will therefore ignore) a stream containing video data and vice versa. Only very few programs (multi-modal data analysis software) would interpret the majority of XDF fields.
The minimal.xdf example file is mainly for the convenience of those who want to test their importers and was created using a real-world client (I believe) after a few years of XDF having been out "in the wild", which is likely why a few (technically non-mandatory) fields like created_at and uid were not stripped out of the file, even though that renders the file not strictly minimal (that said, generally, XDF files may contain any number of additional fields, for forward compatibility, and importers are not supposed to choke on those, so that's why this file is still a useful test case, even though it is not the single source of truth about what's mandatory or not).
Let's talk about the StreamFooter chunk. That chunk is there for two purposes: a) to indicate that the file was completed/transmitted in full and not somehow truncated (e.g. when a recording crashed), and b) as a reserved area where current applications, or future revisions of the XDF spec, may store additional information that enables faster loading of, or seeking into, the data. At this time the spec does not consider anything there mandatory, because the initial version of the spec is intentionally lean to help adoption, and if you are writing an application that generates XDF files, you should be able to safely write an empty chunk there (if an importer chokes on an empty StreamFooter chunk, that's probably a bug and worth fixing*). The reason why the spec has no mandatory fields in this chunk is that XDF content is intended to be fully interpretable even if the file was truncated at the end (if your recording crashes after 3h, you should have 3h worth of usable data).
Hope that helps! And let me know if you have additional questions.
*: of course the developers of those importers are free to disagree with me here.
In retrospect "must" vs "should" should have been more obvious to me 😅
Thank you for your swift answer, I have no further questions.
No problem! It's still a fair question, because many XDF files probably contain most or all fields in the <info> example in the spec, and some applications might try to eagerly read some or all of those and fail of one is not present (that's again a thing that should be fixed in the application's importer). That said, we've been maintaining a piece of software (NeuroPype) that definitely doesn't write anything other than those 6 fields, and we haven't yet had a problem with any XDF files created by that in now 8 years. On the footer chunk I should say that, if the footer chunk is present, then the <info> tag may be expected to be there (even if empty) by applications that look at the footer, just because of the way those applications tend to be written. That's one thing that I think could be clarified in the spec, but OTOH we'd also be hesitant to make it stricter on a whim and invalidate some existing and otherwise perfectly compliant files.
|
2025-04-01T04:35:27.061998
| 2023-03-24T22:32:51
|
1640135951
|
{
"authors": [
"mileslucas"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10620",
"repo": "scexao-org/vampires_dpp",
"url": "https://github.com/scexao-org/vampires_dpp/issues/4"
}
|
gharchive/issue
|
Skip intermediate data products by default
Right now every step of the pipeline is written to take a path to a FITS file in and produce a new FITS file and return the new path. This leads to a lot of data copying, which can cause significant ballooning of working directory sizes because you essentially get a 1:1 copy of the raw data for the following steps
calibration
frame selection
registration
This can easily take a ~200 GB raw dataset and make it take >1 TB of space.
The frame selection and registration already have a somewhat decoupled interface between measuring the metrics/offsets and modifying the data. My path forward is going to look at how to modify the pipeline to only use these files unless a user specifically requests the intermediate FITS files.
Initially I was hesitant to do this by default because it would potentially slow down repeated reductions of the pipeline. Let's look at that case, though. In the case that we already have metrics and offsets measured, we only need to "remake" frame selected or registered data if we want to change our collapse method or something earlier in the pipeline. For most reruns the changes are after this step so we don't really need the registered or selected data on disk. For cases where the selection or registration need redone the files would have to be remade no matter what, so it's time lost either way.
This change would also be a good opportunity for the FastPDI workflow to change so that Wollaston states aren't split into separate files. That was kind of a mental idea to begin with in terms of data volume. In essence, all of the frame selection and registration is going to get moved into the collapse part of the pipeline. So, as long as I have a CSV for the metrics and the offsets, I can take an input cube and create a collapsed frame. For the Wollaston prism, this means I can measure the offsets for each beam and then from the same calibrated file produce two collapsed files.
|
2025-04-01T04:35:27.074019
| 2018-05-22T12:47:05
|
325278075
|
{
"authors": [
"schalkneethling"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10621",
"repo": "schalkneethling/project-calavera",
"url": "https://github.com/schalkneethling/project-calavera/pull/9"
}
|
gharchive/pull-request
|
docs(readme): Add getting started docs
Add documentation for installing and using Porject Calavera
fix #8
:tada: This PR is included in version 1.2.1 :tada:
The release is available on:
npm package (@latest dist-tag)
GitHub release
Your semantic-release bot :package::rocket:
|
2025-04-01T04:35:27.074858
| 2017-03-02T13:36:25
|
211392627
|
{
"authors": [
"jvhaarst",
"mschatz"
],
"license": "apache-2.0",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10622",
"repo": "schatzlab/genomescope",
"url": "https://github.com/schatzlab/genomescope/pull/2"
}
|
gharchive/pull-request
|
Add k value used to plots.
I also put the units on a separate line, so they are not in front of the next value, which might be confusing (it was for me at first).
Looks good, thank you!
|
2025-04-01T04:35:27.098265
| 2024-10-27T05:12:22
|
2616323192
|
{
"authors": [
"Retropikzel",
"lassik"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10623",
"repo": "scheme-containers/monorepo",
"url": "https://github.com/scheme-containers/monorepo/pull/45"
}
|
gharchive/pull-request
|
Build latest script
Script similar to the build-heads script. Also added comment about how to use on that script too.
It only builds lates if it's an integer. This is made under the assumption that the latest directory should be the main version number of the series. For example 2, for 2.N or 2.N.N.
If for example directory named "2" is found this script will build tags schemers/implementation:latest and schemers/implementation:2.
It makes updating little easier and is one more step towards automating the whole thing. :)
LGTM.
The scripts are getting complex enough that we should probably switch to Gauche. Since this shell script is working code, let's merge it.
|
2025-04-01T04:35:27.104992
| 2023-02-03T07:27:13
|
1569356498
|
{
"authors": [
"p1ranha",
"schlac"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10624",
"repo": "schlac/hacs-integration-discovergy",
"url": "https://github.com/schlac/hacs-integration-discovergy/issues/18"
}
|
gharchive/issue
|
Integration fails to start with HA 2023.2.1
During startup following error occurs:
Logger: homeassistant.components.sensor
Source: custom_components/discovergy/sensor.py:256
Integration: Sensor (documentation, issues)
First occurred: 8:13:43 AM (1 occurrences)
Last logged: 8:13:43 AM
Error while setting up discovergy platform for sensor
Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/helpers/entity_platform.py", line 293, in _async_setup_platform
await asyncio.shield(task)
File "/config/custom_components/discovergy/sensor.py", line 228, in async_setup_entry
DiscovergySensor(
File "/config/custom_components/discovergy/sensor.py", line 256, in init
self.precision = precision
AttributeError: can't set attribute 'precision'
HA: 2023.2.1
Integration version: 0.4.1
Updating to latest beta, solves the problem though.
Thanks for reporting, but this is a duplicate of #16
With HA 2023.2 please use either <= v0.3.0 or >= v0.4.3.
|
2025-04-01T04:35:27.141940
| 2012-04-30T18:42:09
|
4355469
|
{
"authors": [
"Blandph88",
"apextemple",
"emptyflask",
"schneems",
"xanview"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10625",
"repo": "schneems/wicked",
"url": "https://github.com/schneems/wicked/issues/7"
}
|
gharchive/issue
|
Wizard paths with object IDs
I'm not sure how to use Wicked with paths like:
/controller/:id/wizard_step
The situation is that I want to create a new product:
/products/new
Save a new object and redirect to the wizard to populate its data:
/products/1/step_1
/products/1/step_2
etc.
I'm still confused with this, I have this:
def update
steps_logic
@activation.attributes = current_params.to_h
render_wizard(@activation, activation_id: @activation.id)
end
In my logs I see this:
15:23:32 log.1 | 2016-05-30 15:23:32.134454 I [51931:70239350409860] (203.1ms) ActivationsController -- Completed #update -- {:controller=>"ActivationsController", :action=>"update", :params=>{"utf8"=>"✓", "authenticity_token"=>"uQff3b5EzMSU1SqiI3LHHu36sNsdfJqLKhIHawazUC3g79mAuCrJDeC/z8kGoUAotlTrzaBBVK964Kf3kN/6TA==", "activation"=>{"timebox_serial"=>"unactivated1", "terms"=>"1"}, "commit"=>"Next", "id"=>"validate"}, :format=>"HTML", :method=>"PUT", :path=>"/activations/validate", :status=>302, :view_runtime=>0.0, :status_message=>"Found"}
15:23:32 log.1 | 2016-05-30 15:23:32.289774 I [51931:70239350409860] (101.2ms) ActivationsController -- Completed #show -- {:controller=>"ActivationsController", :action=>"show", :params=>{"id"=>"user"}, :format=>"HTML", :method=>"GET", :path=>"/activations/user", :status=>302, :view_runtime=>0.0, :status_message=>"Found"}
15:23:32 log.1 | 2016-05-30 15:23:32.656819 I [51931:70239350409860] (316.8ms) ActivationsController -- Completed #show -- {:controller=>"ActivationsController", :action=>"show", :params=>{"id"=>"timeline"}, :format=>"HTML", :method=>"GET", :path=>"/activations/timeline", :status=>200, :view_runtime=>247.19, :status_message=>"OK"}
So it looks like render_wizard does the correct thing and updates the activation, but then it redirects to the show action without passing activation_id, how do I get it to pass activation_id?
If you did your routes right, you won't need to manually pass the ID through. render_wizard does not pass arguments to redirect like that.
If your route is something like activations/:activation_id/products/<step_name> then the activation_id will be inferred by Rails automatically.
Thank you for your reply. My routes is just "resources :activations"
I ended up doing this instead:
def update
...
session[:activation_id] = @activation.id if @activation
...
end
And when the activation is successful:
def show
...
session.delete(:activation_id)
...
end
Sounds good, just be wary not to put any info in the session that you don't mind losing. If you imagine they'll be going through the wizard all in one go, then you're probably fine. If you think they might stop, email themselves a link to the URL they are on and resume, you might want to make sure the ID is in the URL.
Hi Richard,
The wicked wizard is exactly what I'm looking for, so thanks for creating.
I'm still having bit of trouble though and haven't cracked it yet.
I'm trying to create a teampage object.
Currently receiving the following error:
undefined local variable or method `wizard_path' for #TeamBuildingTeampagesController:0x007fc8242dc4b0
I've got the following:
team_building_teampage_steps_controller.rb
class TeamBuildingTeampageStepsController < ApplicationController
include Wicked::Wizard
steps :team_building_team_types, .....
def show
@team_building_teampage = TeamBuildingTeampage.find(params[:team_building_teampage_id])
render_wizard
end
def update
@team_building_teampage = TeamBuildingTeampage.find(params[:team_building_teampage_id])
@team_building_teampage.update_attributes(params[:team_building_teampage])
render_wizard @team_building_teampage
end
team_building_teampages_controller.rb
class TeamBuildingTeampagesController < ApplicationController
before_action :authenticate_user!, except: [:index, :show]
def create
@team_building_teampage = current_user.team_building_teampages.create
if @team_building_teampage.save
redirect_to wizard_path
else
render :new
end
end
private
def team_building_teampage_params
params.require(:team_building_teampage).permit(:title,.....)
end
end
config/routes.rb
resources :team_building_teampage_steps
post "/team_building_teampages/team_building_teampage_id/team_building_teampage_steps/team_building_team_types", to: "team_building_teampage_steps#show", as: "wizard_path"
and this is the form_for I've got in each of my step views:
<%= form_for (@team_building_teampage, :url => wizard_path, :method => :put) do |f| %>
......
<% end %>
Also receiving this error when I try to go to the views directly:
Couldn't find TeamBuildingTeampage with 'id'=
def show
@team_building_teampage = TeamBuildingTeampage.find(params[:team_building_teampage_id])
render_wizard
end
Bit of a mess...
@Blandph88 did you find a solution? I'm also having a similar problem. The first step is working but when I want to get to the second step, I'm getting the following error: Couldn't find Product withoun an ID
Ended up following Ben Awad’s YouTube tutorial for Airbnb clone where he covers making a multi step form. Look it up. 👌🏻
|
2025-04-01T04:35:27.178986
| 2022-07-27T17:28:38
|
1319871413
|
{
"authors": [
"HJZollner",
"mkwbritton"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10627",
"repo": "schorschinho/osprey",
"url": "https://github.com/schorschinho/osprey/issues/475"
}
|
gharchive/issue
|
[BUG] -io_loadspec_twix Crash When Importing 2+ Datasets - osp_LoadTwix - Mark Britton
Describe the bug
Bug occurs while importing a set of HERMES .dat files in the MATLAB command line (rather than using the Osprey GUI). osp_LoadTwix fails on the second data file with this error message:
`Index in position 2 exceeds array bounds.
Error in mapVBVD>evalMDH (line 782)
mdh.lMeasUID = data_uint32(:,2); % 5 : 8
Error in mapVBVD (line 423)
[mdh, mask] = evalMDH( mdh_blob, version ); % this is quasi-instant (< 1s) :-)
Error in io_loadspec_twix (line 34)
twix_obj=mapVBVD(filename);
Error in osp_LoadTwix (line 66)
raw = io_loadspec_twix(MRSCont.files{metab_ll,kk});
Error in OspreyLoad (line 181)
[MRSCont] = osp_LoadTwix(MRSCont);`
Have not been able to reproduce this bug with other data sets. I have tested the data files in different orders, and it only occurs when more than one file is imported simultaneously. Order does not seem to matter.
To Reproduce
Generate an Osprey jobfile with more than one .dat file and the following options:
seqType = 'HERMES';
editTarget = {'GABA', 'GSH', 'EtOH'};
opts.saveLCM = 1;
opts.savejMRUI = 0;
opts.saveVendor = 0;
opts.savePDF = 1;
opts.fit.method = 'Osprey';
opts.fit.includeMetabs = {'default'};
opts.fit.style = 'Separate';
opts.fit.range = [0.2 4.2];
opts.fit.rangeWater = [2.0 7.4];
opts.fit.bLineKnotSpace = 0.4;
opts.fit.fitMM = 1;
opts.fit.coMM3 = '1to1GABA';
opts.fit.FWHMcoMM3 = 14;
opts.SpecReg = 'RobSpecReg';
Set up jobfile with [MRSCont] = OspreyJob('Jobfile.m')
Import data using OspreyLoad. Error message occurs at second .dat file.
Expected behavior
Expected behavior is importation of 2+ Twix files with no error.
Screenshots
Computer configuration (please complete the following information):
Mac OS
Monterey 12.4
2021b
Osprey 2.3.0
Hi @mkwbritton,
Thanks for reaching out! Could you show me how the files cell array is defined in your jobfile and how the struct entry looks like after running OspreyJob?
Best,
Helge
Hi Helge,
No problem!
Here is the struct:
struct with fields:
flags: [1×1 struct]
opts: [1×1 struct]
ospFolder: '/Users/mark.britton/Desktop/Software/osprey-July2022'
files: {1×2 cell}
files_mm: {}
files_ref: {}
files_w: {}
files_mm_ref: {}
files_nii: {1×2 cell}
files_seg: {}
files_sense: {}
outputFolder: '/Users/mark.britton/Desktop/DESA/MRS_Output'
colormap: [1×1 struct]
loadedJob: 'DESA_Individual.m'
ver: [1×1 struct]
outputFile: 'DESA_Individual.mat'
nDatasets: 2```
Here is the files cell array:
```files = {'/Users/mark.britton/Desktop/STUDY/SUB-8102/ses-01/mrs/sub-0000_ses-01_run-02_acq-hermes.dat',...
'/Users/mark.britton/Desktop/STUDY/SUB-0000/ses-01/mrs/sub-0000_ses-01_run-01_acq-hermes.dat'};```
Just to clarify. You can load a single file, but it crashes as soon as you add a second file to the list?
I'll write you on slack to avoid posting details of the study publicly.
The issue is related to a corrupted TWIX file and not the loader itself.
|
2025-04-01T04:35:27.201507
| 2015-10-28T18:19:31
|
113888669
|
{
"authors": [
"Maigard",
"sbotman"
],
"license": "apache-2.0",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10628",
"repo": "schubergphilis/nagios",
"url": "https://github.com/schubergphilis/nagios/pull/451"
}
|
gharchive/pull-request
|
Fix inheritance options
I was thinking about the inheritance options, and the only way I could think about the directive modifiers was to store them and add them back in in the when they are rendered in the definition, and to skip any objects named "null" or "*".
Just took a quick peek at this and looks good... But want to do some more testing before merging :)
I fixed the contact_name, but when I was adding some testing, I discovered that there is still a problem where multiple contacts could be specified for a service and they will both be added to the directive. This causes nagios to treat it as a contact instead of disabling the inherited option. I'll have to think about it a bit to fix that one.
This is a bit more involved than I was hoping. I am going to close this request and split it out into two different ones.
|
2025-04-01T04:35:27.248059
| 2024-11-30T17:03:36
|
2707666078
|
{
"authors": [
"LucaCappelletti94"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10629",
"repo": "scikit-fingerprints/scikit-fingerprints",
"url": "https://github.com/scikit-fingerprints/scikit-fingerprints/pull/334"
}
|
gharchive/pull-request
|
Added fixed version of map4
Changes
Resolved issue #303. Here follows the TSNEs of the older and newer versions - the improvement in quality is quite clear.
Checklist before requesting a review
[x] Docstrings added/updated in public functions and classes
[ ] Tests added, reasonable test coverage (at least ~90%, make test-coverage)
[x] Updated tests
[ ] Run coverage
[ ] Sphinx docs added/updated and rendered properly (make docs and see docs/_build/index.html)
I am still unable to run coverage or any other such part. After another day with a colleague trying to make it work, I focused on doing the PR. Hopefully, at some point, I'll manage to set it up correctly.
Good PR in general, few changes requested.
In general, I think that we should not use mhfp library, but instead take the necessary parts of MHFPEncoder (https://github.com/reymond-group/mhfp/blob/master/mhfp/encoder.py) and create such class here, at the bottom of map.py file. We need only a few methods, and it can also probably be optimized.
Yeah, arguably that org does not have the greatest track record in maintaining software considering that this refactoring was primarily started by me discovering their software didn't work. Better fold that in.
Also, we should support the count variant. It's quite trivial, only requires addition instead of setting in the .fold() method of MHFPEncoder: https://github.com/reymond-group/mhfp/blob/master/mhfp/encoder.py#L343. Note that this requires int type, rather than np.uint8 in many places, this is why we use it currently.
Ok, I'll try.
|
2025-04-01T04:35:27.294937
| 2024-04-09T14:51:33
|
2233637909
|
{
"authors": [
"LukasDrews97",
"codecov-commenter",
"mirand863"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10630",
"repo": "scikit-learn-contrib/hiclass",
"url": "https://github.com/scikit-learn-contrib/hiclass/pull/119"
}
|
gharchive/pull-request
|
Add probabilistic classification to hiclass
Add probabilistic classification via calibration to hiclass using the following methods:
Platt Scaling
Isotonic Regression
Beta calibration
(Inductive/Cross) Venn-ABERS calibration
Hi @LukasDrews97,
Just a quick request from someone from France. Would it be possible to add a threshold to remove labels that have low probability?
Codecov Report
Attention: Patch coverage is 90.29703% with 98 lines in your changes missing coverage. Please review.
Project coverage is 93.94%. Comparing base (4595264) to head (db15547).
Report is 4 commits behind head on main.
Files with missing lines
Patch %
Lines
hiclass/_calibration/VennAbersCalibrator.py
79.62%
43 Missing :warning:
hiclass/HierarchicalClassifier.py
85.84%
16 Missing :warning:
hiclass/LocalClassifierPerLevel.py
89.33%
8 Missing :warning:
hiclass/metrics.py
95.45%
8 Missing :warning:
hiclass/_calibration/Calibrator.py
91.89%
6 Missing :warning:
hiclass/Pipeline.py
44.44%
5 Missing :warning:
hiclass/LocalClassifierPerNode.py
95.78%
4 Missing :warning:
hiclass/_calibration/BetaCalibrator.py
87.87%
4 Missing :warning:
...iclass/probability_combiner/ProbabilityCombiner.py
88.88%
3 Missing :warning:
hiclass/LocalClassifierPerParentNode.py
98.76%
1 Missing :warning:
Additional details and impacted files
@@ Coverage Diff @@
## main #119 +/- ##
==========================================
- Coverage 96.68% 93.94% -2.74%
==========================================
Files 13 28 +15
Lines 1268 2297 +1029
==========================================
+ Hits 1226 2158 +932
- Misses 42 139 +97
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
|
2025-04-01T04:35:27.435897
| 2019-06-17T16:30:47
|
457029944
|
{
"authors": [
"jacobmontiel",
"smastelini"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10631",
"repo": "scikit-multiflow/scikit-multiflow",
"url": "https://github.com/scikit-multiflow/scikit-multiflow/pull/119"
}
|
gharchive/pull-request
|
Fixing division by zero and dealing with numerical imprecision on Regression and Multi-target Regression HT
Greetings @jacobmontiel,
I'm happy to finally be able to contribute again to scikit-multiflow. It's been a long time!
I've fixed some minor bugs when calculating the standard deviation and comparing split merits. These changes encompass both the Regression HT (R-HT) and Multi-target Regression HT (MTR-HT) algorithms.
The way the variance and standard deviation are calculated in the regression HT algorithms can lead to numerical imprecision. This fact is discussed in this link. Considering that is not easy to change the HTs to add a more efficient online variance calculation algorithm, I added a simple check to verify whether the resulting variance is negative (in trees/variance_reduction_split_criterion).
Following, I will list the implemented changes.
Changes proposed in this pull request:
Fixed division by zero in the split attempt for both R-HT and MTR-HT (Addition of a conditional clause to check whether best_suggestion.merit > 0).
Added check for negative variance values before calculating the standard deviation for the R-HT.
Checklist
[x] Code complies with PEP-8 and is consistent with the framework.
[x] Code is properly documented.
[x] Tests are included for new functionality or updated accordingly.
[x] Travis CI build passes with no errors.
[x] Test Coverage is maintained (threshold is -0.2%).
[x] Files changed (update, add, delete) are in the PR's scope (no extra files are included).
Hello @smastelini and welcome back :-)
Thanks for submitting this fix. Could you elaborate on this?
Considering that is not easy to change the HTs to add a more efficient online variance calculation algorithm, I added a simple check to verify whether the resulting variance is negative...
Hello @jacobmontiel,
By not easy, I'm referring to the way the trees store necessary statistics for making the splits. The link I sent in the previous message suggests a more robust and stable way of calculating variances. Nonetheless, this approach presupposes a recursive variance calculation strategy, which does not reflect the way the Hoeffding Tree algorithms work:
Firstly, variance estimations need to be made in any splitting point (not only considering all the data)
Moreover, the statistics stored by the new presented algorithm do not have the addictive properties of the current implementation of the Hoeffding Trees (combining partial statistics -- summing and subtracting). In fact, the strategy of storing the sum of y and y^2 (as the current algorithms do) works very much alike the ClusStream algorithm (and its microclusters).
For these reasons, I made the proposed changes just to avoid generating warnings due to numeric inconsistencies. Looking at the MOA implementation, there is not seem to be any special treatment for these cases, too.
What do you think?
Thanks for the clarification @smastelini
I think the safety check is a straightforward solution to avoid negative values, considering the limitation that you mentioned in the context of the current implementation of HT.
|
2025-04-01T04:35:27.438640
| 2022-08-27T03:46:00
|
1352930844
|
{
"authors": [
"EmmaRenauld",
"arnaudbore"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10632",
"repo": "scilus/tractometer_website",
"url": "https://github.com/scilus/tractometer_website/pull/1"
}
|
gharchive/pull-request
|
First PR
Copying files from scil + changing content for tractometer pages.
A lot of tweeking still needed. Nearly all links should be updated.
Things that would be nice:
Make subtitles bigger
Improve general style: banners and other. Main page.
Hyperlinks colors! (They color on hover but base color is not a visible choice)
Fix figcaption to make it the same size as the image.
@arnaudbore , It is more or less good when tested locally. Could serve as base, we will improve the style later.
Look very nice appart from the missing logo and the banner. We should try to push this asap. Once is done ... it's done :+1:
|
2025-04-01T04:35:27.439843
| 2018-10-08T07:42:35
|
367669958
|
{
"authors": [
"kormat",
"scrye"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10633",
"repo": "scionproto/scion",
"url": "https://github.com/scionproto/scion/issues/1931"
}
|
gharchive/issue
|
SD.go: sciond doesn't filter expired paths
If sciond has old paths in its pathdb, it will hand them out to clients even if they've already expired.
Fixed by https://github.com/scionproto/scion/pull/1944.
|
2025-04-01T04:35:27.449778
| 2019-04-10T08:40:51
|
431376125
|
{
"authors": [
"HBuenoo",
"rgommers"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10634",
"repo": "scipy/scipy",
"url": "https://github.com/scipy/scipy/issues/10043"
}
|
gharchive/issue
|
Trouble importing scipy.spacial.transform
I do have scipy installed, but when I need it in another package, I cant get over this error, says cant find module from scipy
ModuleNotFoundError Traceback (most recent call last)
in
4 import astropy.units as u
5 from astropy.time import Time
----> 6 from gala import coordinates as gc
7 from astropy.table import QTable, Column
~/anaconda3/lib/python3.7/site-packages/gala-1.0.dev2456-py3.7-linux-x86_64.egg/gala/init.py in
24 if not ASTROPY_SETUP:
25 from . import coordinates
---> 26 from . import dynamics
27 from . import integrate
28 from . import potential
~/anaconda3/lib/python3.7/site-packages/gala-1.0.dev2456-py3.7-linux-x86_64.egg/gala/dynamics/init.py in
1 from .core import PhaseSpacePosition, CartesianPhaseSpacePosition
2 from .orbit import Orbit, CartesianOrbit
----> 3 from .analyticactionangle import *
4 from .actionangle import *
5 from .nonlinear import *
~/anaconda3/lib/python3.7/site-packages/gala-1.0.dev2456-py3.7-linux-x86_64.egg/gala/dynamics/analyticactionangle.py in
11
12 # Project
---> 13 from ..potential import (Hamiltonian, PotentialBase,
14 IsochronePotential, HarmonicOscillatorPotential)
15 from ..util import atleast_2d
~/anaconda3/lib/python3.7/site-packages/gala-1.0.dev2456-py3.7-linux-x86_64.egg/gala/potential/init.py in
----> 1 from .potential import *
2 from .hamiltonian import *
3 from .frame import *
4 from .scf import SCFPotential
~/anaconda3/lib/python3.7/site-packages/gala-1.0.dev2456-py3.7-linux-x86_64.egg/gala/potential/potential/init.py in
----> 1 from .core import *
2 from .cpotential import *
3 from .ccompositepotential import *
4 from .builtin import *
5 from .io import *
~/anaconda3/lib/python3.7/site-packages/gala-1.0.dev2456-py3.7-linux-x86_64.egg/gala/potential/potential/core.py in
10 import astropy.units as u
11 from astropy.utils import isiterable
---> 12 from scipy.spatial.transform import Rotation
13
14 # Project
ModuleNotFoundError: No module named 'scipy.spatial.transform'
Rotation was added in the latest release (1.2.0). You didn't say what scipy version you have installed, but most likely you just need to upgrade.
wow...that was simple hahah
thank you so much!
|
2025-04-01T04:35:27.452598
| 2022-02-10T10:59:46
|
1129842637
|
{
"authors": [
"rgommers",
"tupui"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10635",
"repo": "scipy/scipy",
"url": "https://github.com/scipy/scipy/issues/15568"
}
|
gharchive/issue
|
BENCH/CI: Benchmark timeout
Benchmarks are reliably timing out half way (at a different point).
This is the first commit which is failing in the history. But there is no link with the changes.
https://github.com/scipy/scipy/commit/3134ab622b09d5d9bfd34c7aabbf8b92a8497723
Can't sleep, so having a look at the daily failure emails. Benchmarks job was last passing on main on Feb 4th and failing on Feb 8th:
This was the time span of the ASV 0.5 and 0.5.1 releases. Apparently something else broke in 0.5 (see gh-15536) that was fixed in 0.5.1 - but not this timeout issue.
This is probably the same as Pandas is seeing, as reported in https://github.com/airspeed-velocity/asv/issues/1026. Let's keep an eye on that, and pin in the meantime.
|
2025-04-01T04:35:27.457556
| 2016-02-07T00:56:56
|
131918347
|
{
"authors": [
"ev-br",
"josef-pkt",
"rgommers"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10636",
"repo": "scipy/scipy",
"url": "https://github.com/scipy/scipy/issues/5817"
}
|
gharchive/issue
|
BUG: skew, kurtosis return np.nan instead of "propagate"
Initial guess
I thought "propagate" just calculates with nans in the arrays, like np.sum, so we get the correct shape with just some extra nans.
Instead we get a scalar nan back
https://github.com/scipy/scipy/blame/v0.17.0/scipy/stats/stats.py#L1319
possible candidate for
https://github.com/statsmodels/statsmodels/pull/2800
>>> aa = np.arange(5 * 4.).reshape(5,4)
>>> aa[0,1] = np.nan
>>> np.sum(aa, 0)
array([ 40., nan, 50., 55.])
>>> from scipy import stats
>>> stats.kurtosis(aa, axis=0)
array([-1.3, nan, -1.3, -1.3])
>>> stats.skew(aa, axis=0)
array([ 0., nan, 0., 0.])
>>> scipy.__version__
'0.15.1'
The actual problem for the statsmodels case seems to be scalar versus array return
continued example
1-d versus 2d inputs
>>> stats.skew(aa[0], axis=0)
nan
>>> stats.skew(aa[1], axis=0)
0.0
>>> stats.skew(aa[:1], axis=1)
array([ nan])
>>> stats.skew(aa[1:2], axis=1)
array([ 0.])
>>> np.sum(aa[1], axis=0)
22.0
>>> np.sum(aa[1:2], axis=1)
array([ 22.])
>>> np.sum(aa[0], axis=0)
nan
>>> np.sum(aa[:1], axis=1)
array([ nan])
That's indeed a regression that needs fixing. I thought we had not merged one PR yet with the functions that suffered from this, but apparently we missed one.
This seems to be a pattern that has been added to several reduce functions
https://github.com/scipy/scipy/commit/73c4f620bb6602632f6539ca3764a5d8583172b8
I think statsmodels is also using normaltest, skewtest and kurtosistest, but in general we are not supposed to have nans in the calculated data, so there are no broken test cases.
I'll change the milestone to 0.17.1 because this functionality is new in 0.17.0 so it could be good to have it fixed in the 0.17.x series.
|
2025-04-01T04:35:27.465876
| 2016-03-10T03:55:23
|
139777151
|
{
"authors": [
"Eric89GXL",
"e-q",
"endolith"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10637",
"repo": "scipy/scipy",
"url": "https://github.com/scipy/scipy/issues/5954"
}
|
gharchive/issue
|
Complex conjugates in correlate()
I don't know when this was introduced, but there's a discrepancy in handling of complex conjugates in scipy 0.17.0:
scipy.signal.correlate(1j, 2j) (-2+0j)
scipy.signal.correlate([1j], [2j]) array([ 2.+0.j])
scipy.signal.correlate([[1j]], [[2j]]) array([[ 2.+0.j]])
scipy.signal.correlate2d([[1j]], [[2j]]) array([[-2.+0.j]])
numpy.correlate([1j], [2j]) array([ 2.+0.j])
Wikipedia's definition of cross-correlation is:
which matches Wolfram's:
so this would mean conjugating the 1j into -1j and then multiplying it to get +2 as the correct output.
but then correlate(1j, 2) should equal 1j.conj() * 2 = -1j * 2 = -2j, which none of these agree with:
scipy.signal.correlate(1j, 2) 2j
scipy.signal.correlate([1j], [2]) array([ 0.+2.j])
scipy.signal.correlate([[1j]], [[2]]) array([[ 0.+2.j]])
scipy.signal.correlate2d([[1j]], [[2]]) array([[ 0.+2.j]])
numpy.correlate([1j], [2]) array([ 0.+2.j])
Can someone run similar things in matlab and see which way it calculates?
MATLAB r2015b
> xcorr([1j, 0], [2j, 0])
ans =
0 2 0
> xcorr([1j, 0], [2, 0])
ans =
Column 1
0 + 0i
Column 2
0 + 2i
Column 3
0 + 0i
> xcorr([1, 0], [2j, 0])
ans =
Column 1
0 + 0i
Column 2
0 - 2i
Column 3
0 + 0i
>> conv([1j, 0], [2j, 0])
ans =
-2 0 0
>> conv([1j, 0], [2, 0])
ans =
Column 1
0 + 2i
Column 2
0 + 0i
Column 3
0 + 0i
>> conv([1, 0], [2j, 0])
ans =
Column 1
0 + 2i
Column 2
0 + 0i
Column 3
0 + 0i
So we can either change everything to match book definitions and become inconsistent with other software, or just change correlate 0d and correlate2d to make them consistent internally, and add Notes explaining the discrepancy
Any explanation from Octave or MATLAB docs about their convention?
They both define what they're doing, but don't give a reason why they swapped which input gets conjugated.
http://www.mathworks.com/help/signal/ref/xcorr.html#bubr0in
http://www.mathworks.com/help/signal/ref/xcorr2.html#buwgc15
I looked for refs and found one source that conjugates the second input:
http://web.eecs.umich.edu/~fessler/course/451/l/pdf/c2.pdf
And another that conjugates first:
http://www.ee.ic.ac.uk/hp/staff/dmb/courses/E1Fourier/00800_Correlation.pdf
but most book refs I found don't mention the conjugation at all, they deal only with real signals.
So I think it's fine to leave it as is, but fix scipy.signal.correlate 0d and correlate2d to be consistent, and explain the convention in the docstrings.
Agreed, it seems like they should at least be internally consistent.
Actually now that I look at this again, more carefully, all of the definitions I can find are equivalent, they just have minus or plus signs making them look different. So correlate is correct for ND inputs, incorrect for 0d inputs, and correlate2d is incorrect.
The complex conjugate exists to make sure the output is positive when the inputs have the same sign and the same imaginary sign. So correlate(3, 4) should be positive, and correlate(3j, 4j) should also be positive.
|
2025-04-01T04:35:27.483598
| 2018-02-28T11:44:19
|
300993320
|
{
"authors": [
"cdipaolo",
"ilayn",
"jomsdev",
"kdhingra307",
"rgommers"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10638",
"repo": "scipy/scipy",
"url": "https://github.com/scipy/scipy/issues/8498"
}
|
gharchive/issue
|
Query Regarding threshold variable in clarkson_woodruff_transform
Abstract:
There is a threshold variable 0.1 in the clarkson_woodruff_transform demo and in the documentation, which according to my overall understanding should not be there because it has never been used in the calculations as such.
Details:
I found clarkson_woodruff_transform through GSOC18 and started going through the research paper as well as the code present in the scipy.
There was a threshold variable, which confused me because we are setting threshold and sketch_size on our own(it's because the sense in which threshold is mentioned is confusing). Because According to my understanding of the whole algorithm, threshold depends on sketch_size and no relation has been calculated for those.
from scipy import linalg
n_rows, n_columns, sketch_n_rows = (2000, 100, 100)
threshold = 0.1
tmp = np.random.normal(0, 0.1, n_rows*n_columns)
A = np.reshape(tmp, (n_rows, n_columns))
sketch = linalg.clarkson_woodruff_transform(A, sketch_n_rows)
sketch.shape
(100, 100)
normA = linalg.norm(A)
norm_sketch = linalg.norm(sketch)
Can anyone please help clearing this confusion??
also this example will not pass the test case, written for clarkson transform
for that alpha has to be reduced from 0.1 to around 0.01 (somewhat near to that)
if u need that test case, i can share (i didnt share because it was very big)
pinging @jomsdev
thanks for pinging me @ilayn
@kdhingra307 I am not actually at home, I will replay tomorrow :)
@jomsdev thanks
no issues
Hello @kdhingra307 I am glad to see that you are interested in contributing to improve the randomized numerical linear algebra methods in scipy.
If I am not wrong you are talking about the tests for the Clarkson transform (CWT). As you may have seen in the paper, there are two related terms in the transformation: the size of the sketch and the error.
In a real world application you would have to first calculate the size of the sketch given the error bound that you want to have. However, for testing it, we decided to fix the size of the matrices, and put a threshold that is big enough to don't fail (knowing that our matrices are random.normal with mu=0, sigma=0.01) if the transformation is performed correctly.
# Big dense matrix dimensions
n_matrix_rows = 2000
n_matrix_columns = 100
# Sketch matrix dimensions
n_sketch_rows = 100
# Error threshold
threshold = 0.1
The API does not let you yet get the right size given a desired error, neither the error given a size of the sketch but it is something that would be nice.
Not sure if this is what you were asking for, let me know if I can help you clarifying anything else.
@jomsdev that's what I wanted to know
because it was little misleading initially and in the docs(not in test cases), the threshold variable needs to be changed a little bit (based on my calculations)
It was written in the docs about threshold and sketch size but I didn't find any code to find good sketch size (that's why I put this query)
should I add on to this, because I feel its important to find the right sketch size
@kdhingra307 it would be a good contribution.
I have a proposal for this GSoC for implementing Blendenpik and one of the sub tasks is to implement a method for this.
If you want to learn more about sketching errors, scikit has two implementatios of random projections.
In it's documentation you can find some useful links for the The Johnson-Lindenstrauss lemma
And other interesting contribution could be having two different sketch methods implemented the CWT (adapted to sparse matrices) and the Fast Johnson-Lindenstrauss trasnformation for dense ones. Then our algorithms (low rank, square solvers etc) could take advantage of each algorithm.
@jomsdev Are u mentoring GSoC project for Sketches?? Because I got to know about Clarkson Woodruff transformation from there only.
thanks for scikit-learn link, I will go through those.
I did read the research paper about Johnson-Lindenstrauss transformation but shouldn't we sparsify clarkson woodruf transform first??
I am not mentoring but applying as student too (I am currently finishing my master in discrete mathematics so this was my last chance for participating to GSoC :D)
As you said, one of the main advantages of CWT is that you could compute the sketch in nnz(A), where nnz is the number of non zeroes entries in the matrix. CWT was my first contribution to Scipy and given that I wasn't sure about how much support sicpy has for sparse matrix I decided to go step by step.
There is no problem in having both methods: CWT and FJLT implemented only for dense matrices and then improve the support of CWT for sparse matrices or do it in inverse order, first have CWT for sparse matrices and then implement the FJLT.
@jomsdev same here, I am in the final year too :P
I think, we should go by implementation in the inverse order, because then for FJLT we don't have to sparsify too. We can then have direct sparsify implementation of FJLT.
plus also I think, if we implement algorithm for dense matrix first, then it would be redo of some of the work later on.
What do you say?
Well, FJLT can be used for dense and CWT for dense and sparse. This is how it should look like after everything is implemented. From there, I do not see a a huge advantage about having the sparse version first because even without it we could implement Blendenpik and other algorithms on top of it but not taking advantage of sparsity.
But yes, I don't see any problem in having CWT for dense and sparse before FJLT.
Let's go with the sparsity of CWT first and finish this to a good point, then we should go for FJLT.
@kdhingra307 time for writing the proposals, do you still being interested in RNLA? I am glad to help in finding something interesting/useful.
In fact, we may end up doing something complementary :)
Let me know if you need something.
@jomsdev yaa I am, I have studied whole CWT and majority of FJLT in the meantime.
I was not available last week due to a minor surgery.😅😅
We will surely,
Of course, I will start adding stuff now, related to CWT and its sparsity.
I have created a pr with #8579
Also, I will also share my proposal and proposed workflow during the GSoC timeline
@cdipaolo, @jomsdev just checking, can we close this issue or is there something of interest left here?
@rgommers Yes I believe this issue can be closed since the reference to a threshold in tests has been removed in #9884. Also, we provide some guidance on what an actual threshold should be in the docstring now.
That said I think it would be reasonable looking forward to add methods into scipy.linalg or scipy.sparse.linalg which use sketching under the hood to perform routine numerical linear algebra tasks like solving least squares problems approximately [1] or finding approximate low rank factorizations [1,2]. If/when these methods do get added (which is also worthy of discussion) it will be very important to include correct tolerance/threshold parameters. This could require some relatively extensive empirical investigation to get right, since many upper bounds in the theory don't give correct constant factors (and sometimes even asymptotics) as far as I'm aware. A recent paper by Dobriban and Liu [3] looks like it might be useful here though.
That said I think it would be reasonable looking forward to add methods into scipy.linalg or scipy.sparse.linalg which use sketching under the hood to perform routine numerical linear algebra tasks like solving least squares problems approximately [1] or finding approximate low rank factorizations [1,2].
@cdipaolo yes I think so.
If/when these methods do get added (which is also worthy of discussion) it will be very important to include correct tolerance/threshold parameters. This could require some relatively extensive empirical investigation to get right, since many upper bounds in the theory don't give correct constant factors (and sometimes even asymptotics) as far as I'm aware. A recent paper by Dobriban and Liu [3] looks like it might be useful here though.
What would also be very useful is an overview of what is available. E.g. there's some randomized low-rank approximation methods in scipy.linalg.interpolative, and there's https://ristretto.readthedocs.io/en/latest/ which includes some sketching methods. It's pretty hard for non-experts to figure out what they need here, or the relative performance of the options.
closing since this issue was addressed
|
2025-04-01T04:35:27.490949
| 2018-10-16T13:00:47
|
370600354
|
{
"authors": [
"ev-br",
"ilayn",
"kstacer",
"ludcila",
"rgommers"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10639",
"repo": "scipy/scipy",
"url": "https://github.com/scipy/scipy/issues/9381"
}
|
gharchive/issue
|
linregress should warn user in 2x2 array case
The heuristic for splitting an array into x and y inside linregress is ambiguous for 2x2 arrays. I discovered this ambiguity when I called linregress with a 2x2 array where x and y were oriented as columns (a common case when working with pandas dataframes) and got an unexpected result.
Reproducing code example:
Sample code to reproduce the problem:
#A simple numpy array that is 2x2 where each row is the x, y values.
from scipy.stats import linregress
import numpy as np
arr = np.array([[30,31], [50,50]])
slope, intercept, r_value, p_value, std_err = linregress(arr)
#slope = 1.0
# intercept =20.0
# r_value = 1.0
# p_value = 0.0
#std_err = 0.0
#Visualize with plot
import matplotlib.pyplot as plt
x = arr[:,0]
y = arr[:,1]
plt.plot(x, y, 'o', label='original data')
plt.plot(x, intercept + slope*x, 'r', label='fitted line')
plt.legend()
plt.show()
print(r_value**2)
#Tranposing the array gives the expected result:
slope, intercept, r_value, p_value, std_err = linregress(arr.T)
#slope = 1.0
#intercept = 1.0
#r_value = 1.0
#p_value = 0.0
#std_err = 0.0
x = arr[:,0]
y = arr[:,1]
plt.plot(x, y, 'o', label='original data')
plt.plot(x, intercept + slope*x, 'r', label='fitted line')
plt.legend()
plt.show()
print(r_value**2)
Desired Resolution:
In this case, no error message is detected; however, the ambiguity of the 2x2 case invites some additional documentation and perhaps a RunTime warning message when this corner case is detected. At the least, a clear description of this heuristic in the documentation is desired explaining how x and y are created internally.
Scipy/Numpy/Python version information:
import sys, scipy, numpy; print(scipy.__version__, numpy.__version__, sys.version_info)
('1.0.0', '1.13.3', sys.version_info(major=2, minor=7, micro=14, releaselevel='final', serial=0))
I think it's best to add a note to the documentation, either in the Notes section, or under Examples. Pull request welcome, I'd say.
Hi, I would like to take this up to get some practice :grinning:
I noticed that there is already a Notes section which is not written directly in the docstring, but appended somewhere else. In order to keep things simple, do you think it would be enough to expand the Parameters description:
Two sets of measurements. Both arrays should have the same length.
If only x is given (and y=None), then it must be a two-dimensional
array where one dimension has length 2. The two sets of measurements
are then found by splitting the array along the length-2 dimension.
with
In the case where y=None and x is a 2x2 array, the splitting will be
done along the rows, resulting in the sets x[0] and x[1].
?
@ludcila Thanks for taking the time. The situation is a bit awkward indeed. The extra part comes from this part
https://github.com/scipy/scipy/blob/701b9ac6c451e68a0a5659a6186c1f4d847a0be1/scipy/stats/mstats_basic.py#L59-L65
Then further down the file there is the part
https://github.com/scipy/scipy/blob/701b9ac6c451e68a0a5659a6186c1f4d847a0be1/scipy/stats/mstats_basic.py#L840-L841
I think this was meant as a generic machinery however only lingress is using it hence I think we can take it back to its original docstring by cleaning these up.
Hi @ilayn, I can take care of putting the notes in the docstring and cleaning that up. Do you think what I wrote is enough to clarify the 2x2 case? If so, should it be added to the parameters section or to the notes? :smiley:
Fixed by gh-9423, so closing. Thanks all!
Do you think what I wrote is enough to clarify the 2x2 case?
yes, it was clear:)
|
2025-04-01T04:35:27.507522
| 2020-04-20T14:35:25
|
603287130
|
{
"authors": [
"Kai-Striega",
"WarrenWeckesser",
"ilayn",
"mdhaber",
"rgommers"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10640",
"repo": "scipy/scipy",
"url": "https://github.com/scipy/scipy/pull/11902"
}
|
gharchive/pull-request
|
BUG: linalg: test_heequb failure
Investigate/fix failure of test_heequb failure caused by gh-11664.
@Kai-Striega @ilayn feel free to push here directly.
@ilayn it seems that you understood the issue, but I don't, so I wanted to rule out the possibility that the addition of the gejsv tests are interfering somehow with the tests of heequb (workspace issues?). Then I'd confirm that removing the gejsv wrappers altogether fixes the failure.
@ilayn I was wondering where test_heequb came from? heequb is for improving the condition number of the matrix, but the matrix in this test is singular, so the condition number is always going to be infinite. Is the output of heequb meaningful? Why not a more direct test that confirms that the equilibration reduces the condition number of the matrix?
It's not condition number but equilibration numbers similar to matrix_balance. Only deals with matrix entries no algebraic properties are checked. Think of it as trying to match the row and column order of magnitudes before certain operations.
Yeah, we've worked on equilibration together before. I guess I was distracted by the documentation mentioning that:
The scale factors S are computed by the BIN algorithm (see references) so that the scaled matrix B with elements B(i,j) = S(i)*A(i,j)*S(j) has a condition number within a factor N of the smallest possible condition number over all possible diagonal scalings.
But presumably this BIN algorithm does a well-defined job even for singular matrices. OK.
Still, where did the test come from? Is it possible that this example is pathological such that slight changes in the execution of the algorithm could lead to very different results? Is there a way to confirm whether the different scalings s are equally valid?
For instance, if we left out the subdiagonal, then the matrix would be nonsingular and we could just check its condition number rather than the values of the scalings. Before equilibration, its condition number is 512; afterwards, we would assert that it must be less than 10 (because it must be within a factor N of the smallest possible, which is 1 in this case).
This is a LAPACK version problem. As I probably written somewhere LAPACK maintainers fixed a bug in this routine. That's why I branched off as pre3.7 and post3.7 but there is some other version in between or after I guess.
The routine itself is well-defined and has been around for very-long time so I am certain that this is not related to any numerical issues.
Found the link https://github.com/scipy/scipy/pull/10741#issuecomment-531984271
Found the link #10741 (comment)
Thanks; I had read that before. Maybe I don't understand how that version issue is showing up as a result of simply adding a wrapper for a LAPACK routine. Is the version of LAPACK installed during CI dependent on what routines are wrapped? (That sounds crazy to me.)
Probably our CI boxes received an update about their resources.
But the test is not failing for other PRs that ran recently (after we started noticing the issue) and before we merged gh-11664...
@ilayn test_heequb passed when I removed only the tests of gejsv.
I suggest that I add the tests back individually to see if the problem comes from a particular test or not.
Does that sound useful or does it still seem like a version issue? AFAICT this is the same version of LAPACK that is being used in PRs that are failing, like gh-11905 (3.7.1-4ubuntu1?).
It is indeed surprising. Because there are seemingly no relations between the two. So I agree let's do it like that. In the meantime I'll check the versions
test_gejsv_NAG is OK.
test_gejsv_NAG is OK.
@mdhaber test_heequb seems to also be failing on the TravisCI build for #11890. There it is only failing for the python-dbg build, perhaps that has something to do with it?
@Kai-Striega Well, yes, test_heequb has only ever failed on that build. It failed only on that build in gh-11664, and since gh-11664 was merged with master, now master fails on that build. Since your branch in gh-11890 is merged with master before running CI, I would expect it to fail on that build in your PR, too.
Surprisingly, merely running test_gejsv_general is what is causing test_heequb to fail. I'm almost done figuring out which part of test_gejsv_general is causing the interference.
I thought that it had failed on previous runs with non debug builds, therefore it only failing on debug builds would be significant. Apologies for the noise.
If the problem is jobt you can safely ignore it. That is an experimental feature and also I'm really surprised that LAPACK maintainers let that in. There is no reason for us to test it and both reference and MKL docs strictly warn against jobt=1 for now.
Are you seeing the results of the Travis build in CI above? I do not see Travis in the list.
Probably having a bad day I've seen it before happening
Currently Travis-CI is in the list, reporting a failure. It's strange that it is not (or was not) there when you looked, but I think Travis is having issues today. I have a PR over in numpy that took almost two hours to run, and the UI in the numpy page still shows the test as Pending, but if you click the Details link, it shows that the tests passed a while ago. Refreshing the page doesn't help. Go figure.
If you go back to old PRs (pre 10k) you'll see some of them look like they are still running. They're not of course but once it fails to send the trigger to the github side it will stay like that forever
OK, I'm really confused as to why this failed, because when I removed these parameterizations (rather than skipping the tests), it passed. Let the record show that with these modifications:
test_heequb fails. Now I'm going to remove the parameterizations again to see if it passes (as I think it did before).
OK, can somebody tell me why the current changes:
Pass and the changes above didn't?
This is exactly what happened to me in the heequb I think we are looking at the wrong place. This most probably is related to heequb and nothing to do with gejsv. Let's get the NAG example and be done with this. It's not worth of your time.
The question is what to do about the failing test in master. Shall we just merge this PR? Whatever is going on, it seems to solve the problem with minimal changes to the tests.
I would skip the heequb test instead. The ?gejsv test is seemingly stable across the machines.
@ilayn OK, I think this is ready. What's the long-term plan - always skip test_heequb, or should I open an issue?
I would skip the heequb test instead. The ?gejsv test is seemingly stable across the machines.
Not completely sure this comment was addressed, but I'll merge this anyway since the failures are quite annoying. Please send a follow up PR if another tweak was desired.
Thanks @mdhaber, all!
This is still failing on multiple CI jobs regularly. Example from gh-12269:
_________________________________ test_heequb __________________________________
[gw2] linux -- Python 3.7.5 /home/travis/build/scipy/scipy/builds/venv/bin/python
scipy/linalg/tests/test_lapack.py:1759: in test_heequb
assert_allclose(np.log2(s), [0., -1.]*2 + [0.] + [-4]*5)
E AssertionError:
E Not equal to tolerance rtol=1e-07, atol=0
E
E Mismatched elements: 7 / 10 (70%)
E Max absolute difference: 5.
E Max relative difference: 1.25
E x: array([ 0., 0., 0., 0., 0., -9., -9., -9., -9., -9.])
E y: array([ 0., -1., 0., -1., 0., -4., -4., -4., -4., -4.])
A = array([[ 2.+0.j, 0.+1.j, 0.+0.j, 0.+0.j, 0.+0.j, 0.+0.j,
0.+0.j, 0.+0.j, 0.+0.j, ... 0.+0.j, 0.+0.j, 0.+0.j, 0.+0.j, 0.+0.j, 0.+0.j,
0.+0.j, 0.+0.j, 0.+0.j, 1002.+0.j]])
amax = 1002.0
info = 0
s = array([1. , 1. , 1. , 1. , 1. ,
0.00195312, 0.00195312, 0.00195312, 0.00195312, 0.00195312])
scond = 0.001953125
Adding an xfail in gh-12276
|
2025-04-01T04:35:27.511522
| 2022-11-13T22:02:52
|
1447106570
|
{
"authors": [
"doronbehar",
"rgommers"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10641",
"repo": "scipy/scipy",
"url": "https://github.com/scipy/scipy/pull/17416"
}
|
gharchive/pull-request
|
BLD: fix a lot of configuration warnings by using fs.copyfile
Warnings looked like (~100x):
WARNING: custom_target '_lib_pxd' has more than one output! Using the first one.
WARNING: custom_target '_cython_tree' has more than one output! Using the first one.
WARNING: custom_target '_dummy_init_cyoptimize' has more than one output! Using the first one.
Note that fs.copyfile is new in Meson 0.64.0, so this cannot be backported.
When I tried this branch it had some config warnings left,
Yes, the remaining ones are for the _generate_pyx.py scripts in 3 submodules, those can't be changed - we need an upstream improvement there for supporting custom_target with multiple outputs better. There's an open issue for it. Will revisit once the more important stuff is done.
Now updating to v1.10.0, I still get warnings like these:
WARNING: custom_target 'cython_linalg' has more than one output! Using the first one.
WARNING: custom_target 'cython_special' has more than one output! Using the first one.
WARNING: custom_target '_stats_gen_pyx' has more than one output! Using the first one.
And these lines repeat themselves, as before.
@doronbehar yes there's some left - about 50% of the number of warnings that were present before this PR. The ones that are left are these three, they will need a much more elaborate approach to fix, which is not a priority right now.
|
2025-04-01T04:35:27.514536
| 2024-04-08T11:38:17
|
2230966416
|
{
"authors": [
"t-om5",
"yuanx749"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10642",
"repo": "scipy/scipy",
"url": "https://github.com/scipy/scipy/pull/20414"
}
|
gharchive/pull-request
|
BUG: fix jensenshannon returns NaN for some similar distributions
Closes #20083
The jensenshannon function returns NaN values for some distributions close to each other by taking the square root of a negative number. The negative numbers appear in the return of the rel_entr function.
To fix this, use the equivalent kl_div function instead of rel_entr because it always returns non-negative values.
This is possible because the extra parameters on the kl_div formula that make the values non-negative will be canceled out in the sum.
Though I suggested this workaround in https://github.com/scipy/scipy/issues/20083#issuecomment-1980946457, I am not sure whether this is a reasonable solution.
In the example of the original issue #20083, the second jsd is:
second jsd: 0.0001306777430188037
After this fix, the result becomes:
second jsd: 0.0001788337319063894
Seems the results are unstable due to the small numbers. Thoughts from the maintainers would be appreciated.
|
2025-04-01T04:35:27.524988
| 2021-12-10T16:26:47
|
1077020091
|
{
"authors": [
"WesleiRamos",
"c-smile",
"pravic"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10643",
"repo": "sciter-sdk/rust-sciter",
"url": "https://github.com/sciter-sdk/rust-sciter/issues/125"
}
|
gharchive/issue
|
How to inject code in all windows?
I'm making an application where the user can create windows using their own html files, but it would require the windows to have a default behavior, so I developed a simple code that could be manually added by the user, but if there was a way to inject that code would make it a lot easier, I searched and found RuntimeOptions::InitScript, I tried calling a native function and nothing happened.
There's an answer here: https://sciter.com/forums/topic/how-to-inject-code-in-all-windows/
I tried calling a native function and nothing happened.
Could you show the code? It's not clear yet.
Could you show the code? It's not clear yet.
This is what i tried (still doesn't work)
#[macro_use]
extern crate sciter;
struct Handler;
impl Handler {
fn calc_sum(&self, a: i32, b: i32) {
print!("{} + {} = {}", a, b, a + b);
}
}
impl sciter::EventHandler for Handler {
dispatch_script_call! {
fn calc_sum(i32, i32);
}
}
fn main() {
sciter::set_options(sciter::RuntimeOptions::DebugMode(true)).unwrap();
sciter::set_options(sciter::RuntimeOptions::InitScript("
Window.this.xcall('calc_sum', 1, 2);
")).unwrap();
let mut frame = sciter::Window::new();
frame.event_handler(Handler);
frame.load_html(b"<p>Hello</p><script>Window.this.xcall('calc_sum', 1, 5);</script>", None);
frame.run_app();
}
I believe, it's too early to call. I wonder if timers work in this case.
I believe, it's too early to call. I wonder if timers work in this case.
I tried several possible ways and nothing works, my app doesn't need to call native functions too early, I just need to create some functions to manipulate the window.
extern crate sciter;
fn main() {
sciter::set_options(sciter::RuntimeOptions::DebugMode(true)).unwrap();
sciter::set_options(sciter::RuntimeOptions::InitScript("setTimeout(() => console.log('Hello World1'), 5000)")).unwrap();
let mut frame = sciter::Window::new();
frame.load_html(b"<p>:)</p><script>setTimeout(() => console.log('Hello World2'), 5000)</script>", None);
frame.run_app();
}
cc @c-smile
@WesleiRamos Try to ask on http://sciter.com/forums/ - I think, it's a Sciter related issue.
@WesleiRamos Try to ask on http://sciter.com/forums/ - I think, it's a Sciter related issue.
InitScript should work in <IP_ADDRESS>, and see: https://sciter.com/forums/topic/how-to-inject-code-in-all-windows/
It's working now, thanks guys
|
2025-04-01T04:35:27.530851
| 2019-08-12T18:35:38
|
479790144
|
{
"authors": [
"sclevine",
"zachgersh"
],
"license": "apache-2.0",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10644",
"repo": "sclevine/spec",
"url": "https://github.com/sclevine/spec/pull/13"
}
|
gharchive/pull-request
|
updates travis to newer go versions
Hey from DigitalOcean @sclevine!
I had been looking for a project to use this on and I finally found a good fit in:
https://github.com/digitalocean/doctl/pull/549
Any pointers on organization with the library let me know.
Otherwise:
Wanted to drop 1.9 which is deprecated and bring in 1.12 (should be fine).
Looks good. Thanks!
You can use spec.Run anywhere where you'd normally use subtests. If you want to use the report package and put tests from multiple files in the same suite, the new Suite feature is helpful.
That said, as of Go 1.10, go test can output JSON. I'm thinking about deprecating the reporter interface entirely, as there are dedicated reporting tools for go test -json now (e.g., https://github.com/gotestyourself/gotestsum).
|
2025-04-01T04:35:27.537250
| 2022-02-09T11:14:42
|
1128401528
|
{
"authors": [
"phracek",
"pkubatrh"
],
"license": "apache-2.0",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10645",
"repo": "sclorg/container-common-scripts",
"url": "https://github.com/sclorg/container-common-scripts/pull/230"
}
|
gharchive/pull-request
|
Add support for generating Dockerfile.c9s
This pull request adds support for generation Dockefile.c9s files.
And specify distgen configuration file.
Signed-off-by: Petr "Stone" Hracek<EMAIL_ADDRESS>
[test]
The configuration file for C9S already exist here: https://github.com/devexp-db/distgen/blob/main/distgen/distconf/centos-stream-9-x86_64.yaml
[test]
LGTM
[test]
[test]
[test]
[test]
[test]
Fedora test failed here:
```bash
08:36:11 out: [PASSED] for 'hw' test_run_hw_application
08:36:11 out: [PASSED] for 'hw' test_incremental_build
08:36:11 out: [PASSED] for 'hw' test_build_express_webapp
08:36:11 out: [PASSED] for 'clients' express
08:36:11 out: [FAILED] for 'clients' pino
08:36:11 out: [PASSED] for 'clients' prom-client
08:36:11 out: [PASSED] for 'binary' test_run_binary_application
CentOS 7 test failed here:
1m[INFO] Testing variable presence during `docker exec`[0m
Value /opt/rh/rh-nodejs14/root/usr/share/man is missing from variable MANPATH
MANPATH=/opt/rh/rh-python27/root/usr/share/man:/opt/rh/python27/root/usr/share/man:/opt/rh/httpd24/root/usr/share/man:/opt/rh/rh-nodejs14/root/usr/share/man
[1m[INFO] Checking if all scl variables are defined in Dockerfile[0m
RHEL7 test failed here:
08:50:02 out: [PASSED] for 'app' test_scl_usage
08:50:02 out: [PASSED] for 'app' test_connection
08:50:02 out: [PASSED] for 'app' test_docker_run_usage
08:50:02 out: [FAILED] for 'app' test_scl_variables_in_dockerfile
08:50:02 out: [PASSED] for 'app' test_npm_functionality
08:50:02 out: [PASSED] for 'app' test_npm_cache_cleared
08:39:51 out:
08:39:51 out: [INFO] Running test test_scl_variables_in_dockerfile .......
08:39:51 out:
08:39:51 out: Testing npm availability in Dockerfile
08:39:51 out:
Sending build context to Docker daemon 2.048 kB
08:39:51 out: Step 1/2 : FROM rhscl/nodejs-14-rhel7:1
08:39:51 out: ---> 5ac5d773b0df
08:39:51 out: Step 2/2 : RUN command -v npm | grep "^/opt/rh"
08:39:51 out: ---> Running in 984da1c20b29
08:39:51 out:
08:39:51 out: /opt/rh/rh-nodejs14/root/usr/bin/npm
08:39:51 out: ---> a226d6d877e0
08:39:51 out: Removing intermediate container 984da1c20b29
08:39:51 out: Successfully built a226d6d877e0
08:39:51 out:
08:39:51 out: [INFO] Testing variable presence during `docker exec`...
08:39:51 out:
08:39:52 out: Value /opt/rh/rh-nodejs14/root/usr/share/man is missing from variable MANPATH
08:39:52 out: MANPATH=/opt/rh/rh-nodejs14/root/usr/share/man
08:39:52 out: S2I image 'rhscl/nodejs-14-rhel7:1' test FAILED (exit code: 1)
|
2025-04-01T04:35:27.586171
| 2019-12-23T06:14:07
|
541594850
|
{
"authors": [
"SkyWriter",
"chessai"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10646",
"repo": "scotty-web/scotty",
"url": "https://github.com/scotty-web/scotty/pull/259"
}
|
gharchive/pull-request
|
Handle adjacent slashes in the path as one
A PR that [hopefully] fixes the https://github.com/scotty-web/scotty/issues/189
LGTM! Thanks!
Thank you for the wonderful library!
|
2025-04-01T04:35:27.594655
| 2020-05-14T21:25:01
|
618545301
|
{
"authors": [
"noviluni"
],
"license": "bsd-3-clause",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10647",
"repo": "scrapinghub/dateparser",
"url": "https://github.com/scrapinghub/dateparser/issues/680"
}
|
gharchive/issue
|
dateparser.parse raises unexpected errors in some edge cases
I ran the dateparser.parse method against all the strings included in the Big List of Naughty Strings and I found 4 edge cases where dateparser raises an error.
Big number string:
dateparser.parse('2.2250738585072011e-308')
OverflowError: Python int too large to convert to C int
Subscript or superscript numbers:
dateparser.parse('⁰⁴⁵')
ValueError: invalid literal for int() with base 10: '⁰⁴⁵'
dateparser.parse('₀₁₂')
ValueError: invalid literal for int() with base 10: '₀₁₂'
dateparser.parse('⁰⁴⁵₀₁₂')
ValueError: invalid literal for int() with base 10: '⁰⁴⁵₀₁₂'
I know these are edge cases, but it should be fixed as this package can be used to parse user inputs and it could be a security flaw.
Note that search_dates() seems not to be affected.
P.S: I'm happy to see that a lot of strings would have failed before merging this: https://github.com/scrapinghub/dateparser/pull/622 and now they work as expected.
After checking this further, I have:
submitted a patch for the superscript and subscript numbers: https://github.com/scrapinghub/dateparser/pull/684
created a new ticket for the first example: https://github.com/scrapinghub/dateparser/issues/685
To understand the first case, it's necessary to understand that what is happening is that it is translated to '2.2250738585072011 hour-308'.
I think we can close this issue after merging the PR.
As both (https://github.com/scrapinghub/dateparser/pull/686 and https://github.com/scrapinghub/dateparser/pull/684) were merged we can close this issue :tada:
|
2025-04-01T04:35:27.599061
| 2015-09-30T18:22:20
|
109150532
|
{
"authors": [
"iwxfer",
"kmike"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10648",
"repo": "scrapinghub/splash",
"url": "https://github.com/scrapinghub/splash/issues/310"
}
|
gharchive/issue
|
Is is possible to take a screen shot with splash?
I would like to take a screen shot to a webste. Is it possible with this piece of awesome code?
Hey @iwxfer,
Yes, sure - it started as a screenshoting server. Check render.png, render.jpeg or render.json endpoints, or Splash Scripts + splash:png() or splash:jpeg() methods if you need something more custom.
|
2025-04-01T04:35:27.659704
| 2023-12-13T06:55:12
|
2039073030
|
{
"authors": [
"georgehao",
"zimpha"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10649",
"repo": "scroll-tech/scroll",
"url": "https://github.com/scroll-tech/scroll/pull/1046"
}
|
gharchive/pull-request
|
feat(contract): change contract unit test coverage to codecov
Purpose or design rationale of this PR
Describe your change. Make sure to answer these three questions: What does this PR do? Why does it do it? How does it do it?
change the contract unit test coverage tool
PR title
Your PR title must follow conventional commits (as we are doing squash merge for each PR), so it must start with one of the following types:
[ ] build: Changes that affect the build system or external dependencies (example scopes: yarn, eslint, typescript)
[ ] ci: Changes to our CI configuration files and scripts (example scopes: vercel, github, cypress)
[ ] docs: Documentation-only changes
[x] feat: A new feature
[ ] fix: A bug fix
[ ] perf: A code change that improves performance
[ ] refactor: A code change that doesn't fix a bug, or add a feature, or improves performance
[ ] style: Changes that do not affect the meaning of the code (white-space, formatting, missing semi-colons, etc)
[ ] test: Adding missing tests or correcting existing tests
Deployment tag versioning
Has tag in common/version.go been updated or have you added bump-version label to this PR?
[x] No, this PR doesn't involve a new deployment, git tag, docker image tag
[ ] Yes
Breaking change label
Does this PR have the breaking-change label?
[x] No, this PR is not a breaking change
[ ] Yes
lcov --remove can change to lcov --rc branch_coverage=1 --remove, which will preserve the branch hitting data.
lcov --remove can change to lcov --rc branch_coverage=1 --remove, which will preserve the branch hitting data.
updated
|
2025-04-01T04:35:27.687358
| 2023-03-07T07:53:33
|
1612914639
|
{
"authors": [
"LucaMarconato",
"kevinyamauchi",
"melonora"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10650",
"repo": "scverse/napari-spatialdata",
"url": "https://github.com/scverse/napari-spatialdata/issues/42"
}
|
gharchive/issue
|
Speeding up napari-spatialdata
Hey everyone! I just wanted to make an issue to discuss ways we can try to improve the performance of napari-spatial data
for large data. I've listed a few ideas/suggestions off the top of my head below to get things started.
[easy] Profiling loading of data. I think we need to do some profilng to figure out exactly where the bottlenecks are
[easy] The initial drawing of Shapes is slow because they need to be meshed (i.e., turned into triangles). Last year there was a fix that improved triangulation performance that requires the triangle library (see https://github.com/napari/napari/pull/3867 - ~100x speedup for large numbers of shapes). We can include triangle as a dependency and it should "just work".
[medium] I am guessing generating the colormaps for the points and shapes is somewhat slow because we have to loop over the values (e.g., cluster, expression) to set the color. We could try numba to speed that up.
[medium] switch back to using points for the circles Shapes: https://github.com/scverse/napari-spatialdata/issues/37 (We should reach out to Lorenzo Gaifas - he was working on points sizing recently)
[hard] async is coming along in napari. We can start transitioning to async and tiled rendering(e.g., for large 2D images), which should speed somethings up. Joel Lüthi is probably the best person to talk to about this.
[medium] stop using AnnData for storing the information to show the annotation. It's easy to do but quite some work because it requires a big refactoring. https://github.com/scverse/napari-spatialdata/issues/53
[medium] load the data for annotations on demand from the user. As above, easy to do but requires some preliminary refactoring. https://github.com/scverse/napari-spatialdata/issues/54
Polygons:
[medium] better extraction of data from the GeoDataFrame and better transfer of this data to self._viewer.add_shapes() https://github.com/scverse/napari-spatialdata/issues/56
Hey everyone! I just wanted to make an issue to discuss ways we can try to improve the performance of napari-spatial data for large data. I've listed a few ideas/suggestions off the top of my head below to get things started.
[easy] Profiling loading of data. I think we need to do some profilng to figure out exactly where the bottlenecks are
[easy] The initial drawing of Shapes is slow because they need to be meshed (i.e., turned into triangles). Last year there was a fix that improved triangulation performance that requires the triangle library (see Faster 2D shape layer creation napari/napari#3867 - ~100x speedup for large numbers of shapes). We can include triangle as a dependency and it should "just work".
[medium] I am guessing generating the colormaps for the points and shapes is somewhat slow because we have to loop over the values (e.g., cluster, expression) to set the color. We could try numba to speed that up.
[medium] switch back to using points for the circles Shapes: Use napari Points instead of napari Ellipses when SpatialData Shapes are actually circles #37 (We should reach out to Lorenzo Gaifas - he was working on points sizing recently)
[hard] async is coming along in napari. We can start transitioning to async and tiled rendering(e.g., for large 2D images), which should speed somethings up. Joel Lüthi is probably the best person to talk to about this.
Point regarding initial drawing being slow should be fixed when #5555 is merged. It avoids the drawing of the "filling" so that the triangles do not have to be calculated on the fly.
Tomorrow I will discuss exposing the rdp epsilon parameter in the community meeting. It should be ready for merge this week.
Cool, thanks @melonora !
|
2025-04-01T04:35:27.688858
| 2024-06-15T12:01:09
|
2354792076
|
{
"authors": [
"LucaMarconato"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10651",
"repo": "scverse/spatialdata-io",
"url": "https://github.com/scverse/spatialdata-io/pull/158"
}
|
gharchive/pull-request
|
fix bin_size parsing visium hd
Closes #157
Tests fail because we need to cut a new release in spatialdata, but all good.
The CI status is red because we need to make a release in spatialdata.
|
2025-04-01T04:35:27.690412
| 2024-08-08T15:34:04
|
2456104377
|
{
"authors": [
"fruch",
"pehala"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10652",
"repo": "scylladb/argus",
"url": "https://github.com/scylladb/argus/issues/429"
}
|
gharchive/issue
|
Add ability to hide not planned tests
A fair share of tests are not planned for specific release (~190 for 2024.2) and as such are not planned, however, they still show at the list, which creates visual clutter and can be severely confusing for a newcomer (like me).
I propose adding an option to hide not planned tests and groups that contain only not planned test either as user option or a general button on a site
@pehala
you are referring to "not planned" ones ?
Yes, sorry for the confusion, I updated name and the description
|
2025-04-01T04:35:27.697747
| 2023-05-09T04:19:15
|
1701297482
|
{
"authors": [
"bhalevy",
"fruch"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10653",
"repo": "scylladb/scylla-ccm",
"url": "https://github.com/scylladb/scylla-ccm/pull/452"
}
|
gharchive/pull-request
|
node: run_cqlsh: add timeout parameter
Set by default to 10 minutes to prevent blocking in communicate forever.
Blocked run_cqlsh was seen in https://jenkins.scylladb.com/view/master/job/scylla-master/job/dtest-daily-release/244/testReport/cqlsh_tests.cqlsh_copy_tests/TestCqlshCopy/Run_Dtest_Parallel_Cloud_Machines___FullDtest___full_split000___test_copy_to_with_child_process_crashing/
https://jenkins.scylladb.com/view/master/job/scylla-master/job/dtest-daily-release/244/artifact/logs-full.release.000/dtest-gw3.log
13:13:59,598 751 cqlsh_tests.cqlsh_copy_tests DEBUG cqlsh_copy_tests.py :1525 | test_copy_to_with_child_process_crashing: Exporting to csv file: /tmp/tmpqskw5279 with {"exit_range": {"start": 0, "end":<PHONE_NUMBER>000000000}}
...
13:58:50,244 751 errors ERROR conftest.py :203 | test_copy_to_with_child_process_crashing: test failed:
Damn seems like there's an issue with the ccm dependencies I'll need to look into
It worked :)
https://jenkins.scylladb.com/view/master/job/scylla-master/job/dtest-daily-release/249/testReport/cqlsh_tests.cqlsh_copy_tests/TestCqlshCopy/Run_Dtest_Parallel_Cloud_Machines___FullDtest___full_split000___test_copy_to_with_child_process_crashing/
Not that it helps in finding out why cqlsh timed out, but at least the command timed out and not the whole test.
Further investigation in https://github.com/scylladb/scylla-cqlsh/issues/37
|
2025-04-01T04:35:27.700396
| 2024-03-19T02:56:47
|
2193896480
|
{
"authors": [
"dkropachev",
"zey1996"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10654",
"repo": "scylladb/scylla-cdc-go",
"url": "https://github.com/scylladb/scylla-cdc-go/pull/18"
}
|
gharchive/pull-request
|
fix: reader don't exit in some cases.
I notice cdc can't change to new generation in some case.
I think it's a bug.
https://github.com/zey1996/scylla-cdc-go/blob/e99bcf197b49a2603acc317875f5c6d62db9a0e2/stream_batch.go#L160
when we close reader, sbr.interruptCh be writed a struct. but ,if sbr.reachedEndOfTheGeneration(wnd.begin) return true.
the reader will never be close.
I notice cdc can't change to new generation in some case. I think it's a bug.
Do you have any steps how to reproduce it, logs maybe ?
when we close reader, sbr.interruptCh be writed a struct. but ,if sbr.reachedEndOfTheGeneration(wnd.begin) return true. the reader will never be close.
Could you please clarify this statement, I don't see any pathway for it to happen.
|
2025-04-01T04:35:27.765637
| 2015-01-19T10:01:30
|
54744983
|
{
"authors": [
"sdeleuze"
],
"license": "apache-2.0",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10655",
"repo": "sdeleuze/rxweb",
"url": "https://github.com/sdeleuze/rxweb/issues/2"
}
|
gharchive/issue
|
Use spring-web converters
The main issue to fix is that they use InputStream / OutputStream in their API.
Maybe we could implement something like Netty ByteBufInputStream and ByteBufOutputStream for java.nio.ByteBuffer.
To be moved in the spring-rxweb repository.
|
2025-04-01T04:35:27.796622
| 2019-10-30T17:38:08
|
514857987
|
{
"authors": [
"cc-a",
"finswimmer"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10656",
"repo": "sdispater/poetry",
"url": "https://github.com/sdispater/poetry/pull/1520"
}
|
gharchive/pull-request
|
Flake8 config
Quite speculative but this PR provides the flake8 configuration file I used whilst working on poetry. It's compatible with black so should be a useful addition for people who use flake8 in their IDE's as standard.
Hello @cc-a ,
thanks a lot for your contribution! In the meantime @sdispater has included a .flake8 which includes your suggestions and something more. This is why I close the PR.
fin swimmer
|
2025-04-01T04:35:27.806156
| 2016-06-02T09:20:50
|
158095072
|
{
"authors": [
"noumar",
"sdogruyol"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10657",
"repo": "sdogruyol/kemal",
"url": "https://github.com/sdogruyol/kemal/pull/152"
}
|
gharchive/pull-request
|
Fix SSL context
Fixes the recently moved SSL context for servers.
Travis fails on latest as the change is still only in Crystal nightly.
@sdogruyol, Travis fails on latest as the change is still only in Crystal nightly.
Hey @noumar , thanks for the pull request.
Let's wait till this gets published in Crystal latest 👍
Closing in favour of #162
|
2025-04-01T04:35:27.808937
| 2019-11-06T12:24:56
|
518438610
|
{
"authors": [
"manang",
"peaceiris"
],
"license": "CC0-1.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10658",
"repo": "sdras/awesome-actions",
"url": "https://github.com/sdras/awesome-actions/issues/249"
}
|
gharchive/issue
|
remove asset or release
Hi,
I want to know if it is a way to remove the assets or the release during the action job.
I need to update the release, but I don't want to remove the release.
I want to overwrite the pre existing assets.
Thanks
Please use GitHub Actions - GitHub Community Forum. This repository is not suitable for questions.
|
2025-04-01T04:35:27.810310
| 2016-09-30T13:47:07
|
180307979
|
{
"authors": [
"joewandy",
"justinjjvanderhooft"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10659",
"repo": "sdrogers/ms2ldaviz",
"url": "https://github.com/sdrogers/ms2ldaviz/issues/24"
}
|
gharchive/issue
|
Correlate positive and negative mode alpha vectors
Not sure how to visualize it (?), but it would be a cool picture for the manuscript if we see any of the combinations that I passed on earlier showing up - and perhaps some novel ones?
Justin said low priority, nice to have but maybe don't bother if no time.
Whatever it is, seems like we're not doing it.
|
2025-04-01T04:35:27.816175
| 2018-02-06T07:02:30
|
294652362
|
{
"authors": [
"Zhiyuan-Amos",
"eugenepeh",
"vivekscl"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10660",
"repo": "se-edu/addressbook-level4",
"url": "https://github.com/se-edu/addressbook-level4/pull/820"
}
|
gharchive/pull-request
|
Email regex validation does not accept + #743
Fixes #743
Do note that the domain name has to be 2 characters long so that it does not start and end with dots or hyphens.
On a side note, the updated email regex just allows the use of more characters to solve the issue. Should a more restrictive validation be used instead?
For instance this regex,
^[\\w!#$%&’*+/=?`{|}~^-]+(?:\\.[\\w!#$%&’*+/=?`{|}~^-]+)*@(?:[a-zA-Z0-9-]+\\.)+[a-zA-Z]{2,6}$
sets a predefined length for the top-level domain name (which would be edu in<EMAIL_ADDRESS>and restricts trailing and consecutive dots.
Just a quick thought, is the regex too confusing for students to understand? :O @eugenepeh any thoughts? :)
Just a quick thought, is the regex too confusing for students to understand? :O @eugenepeh any thoughts? :)
Not just for students, its a bit too confusing for me to understand too hahahaha, I will need to refer to https://docs.oracle.com/javase/8/docs/api/java/util/regex/Pattern.html.
Plus I am not sure if regex was taught in the current batch students' syllabus. I wonder if we should make a regex util or something that breaks down and combine regex string for readability.
Yup @vivekscl you will need to split up the regex into smaller parts, maybe something like:
String DOMAIN_REGEX = ...
String LOCAL_PART_REGEX = ...
String EMAIL_VALIDATION_REGEX = DOMAIN REGEX + "@" + LOCAL_PART_REGEX;
Feel free to split it up further as you deem fit :)
For the link, I believed that it was best to get into the details immediately, but then again, the overview does explain the RFC 5322 format. Will change it accordingly then.
Oh and for the email constraints message, when in full screen, there is no need to scroll up, down, left or right. But it seems in other laptops it's too long. I'll try to make it consistent with the other constraint messages in AB4 and try to make it more succinct then.
The comment says Represents a Person's phone number.... I'll change it to email in this PR itself as it seems too trivial to make it into a new issue.
|
2025-04-01T04:35:27.821789
| 2019-06-13T10:11:31
|
455653528
|
{
"authors": [
"julianbreiteneicher",
"vulder"
],
"license": "BSD-2-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10661",
"repo": "se-passau/VaRA-Tool-Suite",
"url": "https://github.com/se-passau/VaRA-Tool-Suite/pull/78"
}
|
gharchive/pull-request
|
Adds support to restrict plots to a case study
resolves se-passau/VaRA#403
Here is an overview of what got changed by this pull request:
Complexity increasing per file
==============================
- varats/plots/plots.py 2
See the complete overview on Codacy
hmm, this is not related to the change
do you have wrongly named results in your result folder?
A right. Yes I do. That's why a recent pull request of mine solved a similar issue by skipping files that don't match the regex
tmux-tmux-da359269cb_92c5e562-40fc-46a0-ac0f-4ad9613a3ed1.yaml
project-binary-commithash-UUID.yaml
if the binary is not their, you did not update all your old file names
Hmm, now it generates a graph, but it doesn't contain anything:
vara-gen-graph -r results/gzip/ -p gzip -c paper_configs/small_gzip/cm.txt.cmap interaction_plot
do you have results?
The yaml file is 34MB as usual
this plot tracks changes over time, you have to supply more than one file
Here is an overview of what got changed by this pull request:
Complexity increasing per file
==============================
- varats/plots/plots.py 2
See the complete overview on Codacy
All right, I was holding it wrong. :)
Seems to work now! :)
|
2025-04-01T04:35:27.828612
| 2022-03-10T04:52:44
|
1164751308
|
{
"authors": [
"noahjade",
"rainasong"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10662",
"repo": "se701team3/Forgettable",
"url": "https://github.com/se701team3/Forgettable/pull/90"
}
|
gharchive/pull-request
|
NavBar Component
Created the NavBar component as desired from #57
Note this is just the UI implementation, I will create a new ticket to address and PR to address the routing to different pages :)
Pictured here in lightmode:
I also messed with the colour files a bit to set up for implementing darkmode in the future using this tutorial. But happy to consolidate my implementation with whatever ends up happening with the darkmode :)
SVGs were changed so that I could apply filters to them.
Looks amazing!!
|
2025-04-01T04:35:27.841963
| 2016-08-06T23:09:48
|
169770979
|
{
"authors": [
"adriangb",
"jethrogb",
"seanmonstar"
],
"license": "apache-2.0",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10663",
"repo": "seanmonstar/httparse",
"url": "https://github.com/seanmonstar/httparse/issues/18"
}
|
gharchive/issue
|
Improve API
I would like to build a function
fn read_headers<'a,'b,R: BufRead>(
stream: &mut R,
buf: &'b mut Vec<u8>,
headers: &'a mut [Header<'b>]
) -> Result<Request<'a,'b>,E>
that reads as much from stream into buf as necessary to get a Complete return from Request::parse. This turns out to not be trivial and require lots of extra allocations and work.
Here's what I came up with:
fn read_headers<'a,'b,R: BufRead>(clnt: &mut R, buf: &'b mut Vec<u8>, headers: &'a mut [Header<'b>]) -> Result<Request<'a,'b>,String> {
fn extend_and_parse<R: BufRead>(clnt: &mut R, headers: &mut [Header]) -> Result<Vec<u8>,String> {
let mut buf=Vec::<u8>::new();
let len=headers.len();
loop {
let buf_orig_len=buf.len();
let additional_len={
let additional=try!(clnt.fill_buf().map_err(|e|e.to_string()));
buf.extend_from_slice(additional);
additional.len()
};
let mut headers=Vec::with_capacity(len);
headers.resize(len,httparse::EMPTY_HEADER);
let mut req=Request::new(&mut headers);
match req.parse(&buf) {
Ok(httparse::Status::Complete(n)) => {
clnt.consume(n-buf_orig_len);
break
},
Ok(httparse::Status::Partial) => {
clnt.consume(additional_len);
}
Err(e) => return Err(format!("HTTP parse error {:?}",e)),
};
}
Ok(buf)
}
let result=extend_and_parse(clnt,headers);
result.map(move|nb|{
::core::mem::replace(buf,nb);
let mut req=Request::new(headers);
req.parse(buf);
req
})
}
The main issues are having to allocate a new array of temporary headers for every iteration, and having to parse the succesful result twice.
I think this is partially Rust's fault, but also partially httparse for having a not so great API. For example, the lifetime of everything is fixed upon Request creation, so that a parse failure doesn't release the borrow of the input buffer.
I agree that the lifetime situation feels difficult. If I could fix anything, I'd fix the feeling of trying to use httparse.
As for your specific example, I don't believe allocationg a Vec of headers is necessary, unless you need to at runtime decide how many is the maximum allowed.
As for your specific example, I don't believe allocationg a Vec of headers is necessary, unless you need to at runtime decide how many is the maximum allowed.
You need a buffer that's as big as the input slice headers. You can't do this statically until we get integer generics.
Sure. Who provides the headers slice? I'm actually quite curious at this design, because I haven't experienced it both in sync hyper and async hyper, and I haven't seen others use it that way in other mini http implementations.
For reference, this is how hyper 0.9 would keep reading into a buffer synchronously and try to parse again: https://github.com/hyperium/hyper/blob/0.9.x/src/http/h1.rs#L869-L899
I'm actually quite curious at this design, because I haven't experienced it both in sync hyper and async hyper, and I haven't seen others use it that way in other mini http implementations.
For reference, this is how hyper 0.9 would keep reading into a buffer synchronously and try to parse again: https://github.com/hyperium/hyper/blob/0.9.x/src/http/h1.rs#L869-L899
This looks like it basically runs the same loop as me above, except with a fixed max headers, which means you're exactly avoiding the issues I'm running into. I'd like the caller to specify the max headers as well as avoid large additional allocations (either on the heap or stack) during the parsing.
Been thinking about this, what if Headers were not an array of slices, but array of offsets? It'd remove the lifetimes, but require you to slice the buffer yourself with the offsets to get the name and value.
Another option is to make the parse function to update the lifetime of the Request. This might require an unsafe transmute, which would require that parse ensures all of the fields of Request, include the headers array, are reset, to prevent any dangling references.
What about merging new and parse into one associated function?
fn parse(headers: &'h mut [Header<'b>], buf: &'b [u8]) -> Result<(usize,Request<'h, 'b>)>
You'd need to change the Status enum as well...
That could be done, but it still ties the lifetime of the Header slice to the buf, and it makes the Status trickier, as you said.
I think a reset() method can be added to Request, that will empty all the fields and reset the Header slice, and update it's lifetime to allow a new buffer to be used.
I wrote a little Python wrapper for this library. From the perspective of writing a Python wrapper with PyO3 the lifetime on Headers is problematic.
If you're using the flow of "try to parse, if not wait for Python to call me with more data" you have to re-allocate headers every time because you can't anything with a Header in it (including the buffer) to Python because of the lifetime.
The lifetime requirement on the data is not a problem, PyO3 let's you access the underlaying memory without copying stuff.
|
2025-04-01T04:35:27.843896
| 2020-08-10T19:05:37
|
676350454
|
{
"authors": [
"eaufavor",
"seanmonstar"
],
"license": "apache-2.0",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10664",
"repo": "seanmonstar/httparse",
"url": "https://github.com/seanmonstar/httparse/pull/78"
}
|
gharchive/pull-request
|
Replace deprecated 'try!' with '?'
https://doc.rust-lang.org/std/macro.try.html
Deprecated since 1.39.0: use the ? operator instead
We can't make this change, because then it stops building on Rust versions before ? became stable.
|
2025-04-01T04:35:27.858680
| 2015-06-28T21:41:54
|
91645265
|
{
"authors": [
"kramer",
"sunilalmond"
],
"license": "apache-2.0",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10665",
"repo": "searchbox-io/Jest",
"url": "https://github.com/searchbox-io/Jest/pull/218"
}
|
gharchive/pull-request
|
Added new fields - index, type and score for Hit
Added a new test for score.
Thank you for your contribution @sunilalmond !
|
2025-04-01T04:35:27.874385
| 2021-12-03T17:35:30
|
1070812182
|
{
"authors": [
"twelch"
],
"license": "BSD-3-Clause",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10666",
"repo": "seasketch/geoprocessing",
"url": "https://github.com/seasketch/geoprocessing/issues/102"
}
|
gharchive/issue
|
GP function result caching not working properly
It's not clear caching is working properly for GP functions.
The cache related code doesn't seem to be well documented or have much testing in place.
Initial approach:
Start with manually testing by creating a Postman test suite or similar.
Expose the issue and fix as needed. Backfill missing unit tests.
Stretch goal:
Consider writing a black box e2e test script that runs for a given deployment and verifies cache scenarios, or an e2e test suite in the repo that mocks some of the calls
Ideas:
A node CLI script that lets you specify a function URL, type (sync/async), and a geojson sketch payload
run an initial unique request, verify cache miss, then run again, and verify cache hit, and results are equal.
This is baby steps and can grow into parsing project manifest, automating function discovery, and adding a UI to explore all functions for a project.
The browser in-memory cache (LRU) should also be checked that it's working properly.
|
2025-04-01T04:35:27.880093
| 2023-02-03T02:37:09
|
1569114520
|
{
"authors": [
"CLAassistant",
"luky116",
"springrain",
"wuxiangege"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10667",
"repo": "seata/seata-go-samples",
"url": "https://github.com/seata/seata-go-samples/pull/4"
}
|
gharchive/pull-request
|
add seata at_zorm tests
add seata at_zorm tests
Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you all sign our Contributor License Agreement before we can accept your contribution.1 out of 2 committers have signed the CLA.:white_check_mark: springrain:x: wuxianwuxian seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.You have signed the CLA already but the status is still pending? Let us recheck it.
merge
merge
辛苦处理下冲突
辛苦处理下冲突
好的,已处理
|
2025-04-01T04:35:27.894068
| 2022-08-18T01:58:51
|
1342471682
|
{
"authors": [
"PangXing",
"codecov-commenter"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10668",
"repo": "seata/seata-go",
"url": "https://github.com/seata/seata-go/pull/230"
}
|
gharchive/pull-request
|
bug: fix infinite loop of asyncCallback
What this PR does:
Fix infinite loop of asyncCallback
Which issue(s) this PR fixes:
Fixes #
Special notes for your reviewer:
Does this PR introduce a user-facing change?:
Codecov Report
Merging #230 (eed8447) into master (fd2c058) will decrease coverage by 16.01%.
The diff coverage is 70.00%.
@@ Coverage Diff @@
## master #230 +/- ##
===========================================
- Coverage 63.56% 47.54% -16.02%
===========================================
Files 49 56 +7
Lines 2083 2810 +727
===========================================
+ Hits 1324 1336 +12
- Misses 695 1409 +714
- Partials 64 65 +1
Impacted Files
Coverage Δ
pkg/protocol/message/request_message.go
92.30% <ø> (ø)
pkg/remoting/getty/getty_client.go
94.64% <40.00%> (-5.36%)
:arrow_down:
pkg/protocol/codec/global_begin_request_codec.go
100.00% <100.00%> (ø)
pkg/tm/global_transaction.go
94.54% <100.00%> (ø)
pkg/datasource/sql/driver.go
3.92% <0.00%> (ø)
pkg/datasource/sql/db.go
0.00% <0.00%> (ø)
pkg/datasource/sql/at.go
4.13% <0.00%> (ø)
pkg/datasource/sql/connector.go
0.00% <0.00%> (ø)
pkg/datasource/sql/conn.go
0.00% <0.00%> (ø)
pkg/datasource/sql/stmt.go
0.00% <0.00%> (ø)
... and 1 more
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.
|
2025-04-01T04:35:27.925113
| 2021-04-15T17:53:53
|
859114588
|
{
"authors": [
"holly-r-li",
"neel-a-mehta"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10669",
"repo": "seattle-uat/civiform",
"url": "https://github.com/seattle-uat/civiform/issues/801"
}
|
gharchive/issue
|
Make Tailwind Playground demo for Currency question type
The task
Please create a Tailwind playground demo for this feature:
UX task: #559
Story: #450
Tracker: tiny.cc/cvf-ux-tracker
Background
On tiny.cc/cvf-flow, we've created some wireframes for what a few key pages of our app will look like. We will eventually be building these pages using the Tailwind CSS library.
But as an intermediary step, we need someone to envision what the final page would look like and how it would be implemented using Tailwind.
So check out that board with wireframes, and use the Tailwind Playground to create a demo of what this page should look like.
Submit
Reply to this issue with the URL for your Tailwind Playground demo for this page, such as https://play.tailwindcss.com/wMKlKdzVhW.
@AWBaumann was working on this one: https://play.tailwindcss.com/traAFbmqFG
Stylized here https://play.tailwindcss.com/Wh89ehQdUm?file=css
Great, thanks!
|
2025-04-01T04:35:28.017891
| 2020-06-05T17:50:06
|
631796777
|
{
"authors": [
"JohanBendz",
"jeroendew"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10670",
"repo": "sebbebebbe/com.philips.hue.zigbee",
"url": "https://github.com/sebbebebbe/com.philips.hue.zigbee/issues/194"
}
|
gharchive/issue
|
GU10 and E27 White Ambiance with BT, not recognized
Today I got the GU10 Ambiance with BT but it's not recognized.
Could somebody add these to the app?
Product ID: LGT002
Service ID: 268
Profile ID: 260
Edit:
And the E27 White Ambiance with BT isn't recognized either.
Product ID: LTA001
Service ID: 268
Profile ID: 260
Thanks,
Jeroen
Hi @jeroendew, I think the first product should be LTG002, correct?
If so, both products are already added to the test version of the app. Waiting for more testers to verify that everything works just fine before I publish it to everyone.
Fell free to try it out, I would love to get feedback on any issues :)
https://homey.app/a/com.philips.hue.zigbee/test/
Hello @JohanBendz ,
Sorry, yes it was the LTG002. I've installed the test version and now the lamps are recognized.
Great, thanks for letting me know! :)
|
2025-04-01T04:35:28.081811
| 2023-07-11T23:04:09
|
1799883929
|
{
"authors": [
"nicksuper"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10671",
"repo": "secretflow/scql",
"url": "https://github.com/secretflow/scql/pull/106"
}
|
gharchive/pull-request
|
Update scql_config_manual.po
Update scql_config_manual.po by yanghui
I have read the CLA Document and I hereby sign the CLA
I have read the CLA Document and I hereby sign the CLA
|
2025-04-01T04:35:28.085168
| 2021-04-24T09:14:37
|
866701414
|
{
"authors": [
"hectorkambow",
"munubi254"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10672",
"repo": "section-engineering-education/engineering-education",
"url": "https://github.com/section-engineering-education/engineering-education/issues/2143"
}
|
gharchive/issue
|
Getting started with Kubernetes Clusters
Proposed title of the article
Getting started with Kubernetes Clusters
Introduction paragraph (2-3 paragraphs):
A Kubernetes Cluster is a dynamic system that places and manages containers, grouped in pods, running on nodes, along with all the interconnections and communication channels working to perform a certain operation. In this article, we will first look at how to install Kubernetes on a Cloud Provider and locally in our machines. We will also look at how to use the Kubernetes Client(kubectl) to check our clusters and a few cluster components available such as the Kubernetes Proxy, Kubernetes UI, and Kubernetes DNS.
Key takeaways:
We will get an overview of:
Installing Kubernetes on a Cloud Provider and locally in our machines
The Kubernetes Client(kubectl)
Kubernetes Cluster components
References:
N/A
very cool - approved @munubi254
|
2025-04-01T04:35:28.090032
| 2021-07-21T10:03:10
|
949540660
|
{
"authors": [
"WanjaMIKE",
"ahmadmardeni1",
"hectorkambow",
"salvador02"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10673",
"repo": "section-engineering-education/engineering-education",
"url": "https://github.com/section-engineering-education/engineering-education/issues/2989"
}
|
gharchive/issue
|
Connect-Flash, Bootstrap, Passport and Express-sessions on the go
Proposed title of article
Connect-Flash, Bootstrap, Passport and Express-sessions on the go
Introduction paragraph (2-3 paragraphs):
Connect-flash is node module that allows developers to send flash messages to the user interface efficiently. Connect Flash usually works with express. Express sessions are used in managing sessions in an application. Session handling is a vital part in every web application.
Demonstrating session handling requires an authentication system and that where we will need passport. Passport is an authentication middle ware used to manage authentication in an application. Bootstrap is a CSS framework used for building responsive application.
This article aims at combining all the modules and demonstrate their functionality in a single application. We will build a login system that is based on passport authentication, manage our sessions using express sessions. And show flash messages using connect-flash with bootstrap styling.
Key takeaways:
Session handling in an authentication application
Styling connect-flash messages using bootstrap
Authentication with passport
References:
Please list links to any published content/research that you intend to use to support/guide this article.
Good afternoon and thank you for submitting your topic suggestion.
Your topic form has been entered into our queue and should be reviewed (for approval) as soon as a content moderator is finished reviewing the ones in the queue before it. @salvador02
@salvador02
Sounds like a helpful topic - lets please be sure it add value beyond what is in the official docs and what is covered in other blog sites.
Please be sure to double check that it does not overlap with any existing EngEd articles or incoming topic suggestions (if you haven't already) to avoid any potential article closure, please reference any relevant EngEd articles in yours. - Approved
Hello @hectorkambow , I have finished writing the article. Would you please open the issue.
Closed #2989 via #3599
|
2025-04-01T04:35:28.102752
| 2021-09-29T18:42:53
|
1011327077
|
{
"authors": [
"ahmadmardeni1",
"hectorkambow",
"njerikaren"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10674",
"repo": "section-engineering-education/engineering-education",
"url": "https://github.com/section-engineering-education/engineering-education/issues/4079"
}
|
gharchive/issue
|
[Android] Getting started with android application over the air updates
Topic Suggestion
Pre-submission advice
By following all our pre-submission advice and reviewing our Resources folder, you will maximise your chances of your topic is approved.
We ask that you please be patient as our team works through approving and publishing all articles/tutorials in a timely manner.
Allow 1-3 days for a topic to be reviewed and/or approved - allow 3-7 days for articles to be reviewed and/or published.
Be sure to visit our Resources Page for tools, resources, and example articles that will help you propose and write a successful article.
Please ensure that you have only one open issue + linked pull request at a time. This will ensure that we complete the article in a timely manner from inception to publishing.)
We tend to stray away or tend not to publish reviews/comparisons of commercial product offerings.
Proposal Submission
Getting started with the android application over the air updates
Your title should be descriptive of the article/tutorial. Be Specific.
Use keyword research to improve your article's SEO.
Proposed article introduction
Android devices in the field can receive and install over-the-air (OTA) updates to the system, application software, and time zone rules. In addition, OTA makes it easy to notify users of new application updates every time there is a release.
In the article, I will explain how to implement OTA in an android application. The application created in the article will be hosted in Github other than playstore, when we update the .apk file on Github, then the users will be notified of the new app release and prompted to download the newer version.
For example, if the article is based on machine learning, use the following: [Machine learning] Introduction to Machine Learning.
If the article is based on developing an Android application, use the following: [Android] Developing Apps using Android.
ONLY viable topics to pick from:
API
Artificial Intelligence
Blockchain
Containers
Edge Computing
Languages
Machine Learning
Networking
Node.js
Security
Key takeaways
What are the 3-5 most important things the reader should understand or be able to do after reading this article?
Use this area to get your ideas down on the bulk of your article or tutorial.
Integrating OTA in an android application.
Android application versioning
Article quality
How is your article unique? Tell us what makes your approach different from similar articles that have been published on the same topic?
Is yours more in-depth? Does it cover additional topics? Do you provide handy tips or anecdotal advice?
References
Please list links to any published content/research that you intend to use to support/guide this article.
Conclusion
Finally, remove the Pre-Submission advice section and all our blockquoted notes as you fill in the form before you submit.
We look forward to reviewing your topic suggestion.
Templates to use as guides
How To Guide Template
Software Review Template
Tutorial Template
Good afternoon and thank you for submitting your topic suggestion.
Your topic form has been entered into our queue and should be reviewed (for approval) as soon as a content moderator is finished reviewing the ones in the queue before it.
Sounds like a helpful topic - let's please be sure it adds value beyond what is in any official docs and/or what is covered in other blog sites. (the articles should go beyond a basic explanation - and it is always best to reference any EngEd article and build upon it). @njerikaren
Please be attentive to grammar/readability and make sure that you put your article through a thorough editing review prior to submitting it for final approval. (There are some great free tools that we reference in EngEd resources.)
ANY ARTICLE SUBMITTED WITH GLARING ERRORS WILL BE IMMEDIATELY CLOSED.
Please be sure to double-check that it does not overlap with any existing EngEd articles, articles on other blog sites, or any incoming EngEd topic suggestions (if you haven't already) to avoid any potential article closure, please reference any relevant EngEd articles in yours. - Approved
Just wanted to follow up on this topic, as we will be clearing up the queue where possible.
Just and FYI I will be closing this topic form just to help clear up the queue where possible - can be REOPENED at anytime when ready. 👍
|
2025-04-01T04:35:28.123730
| 2022-01-21T09:07:09
|
1110223373
|
{
"authors": [
"hectorkambow",
"katungi",
"marienjus",
"srishilesh"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10675",
"repo": "section-engineering-education/engineering-education",
"url": "https://github.com/section-engineering-education/engineering-education/pull/6166"
}
|
gharchive/pull-request
|
Managing React State with Hookstate
Article checklist
Attention
Our Peer Reviewer and Content Moderator teams do NOT provide any revisions services. All revisions and edits should be completed on your own forked repo (as to not take up room in the queue) in order for our team to review them all in a timely manner.
ANY ARTICLE SUBMITTED WITH GLARING ERRORS WILL BE IMMEDIATELY CLOSED.
As a rule of thumb - please be sure to only submit articles (pull requests) that are fully polished and ready to be published. Be sure to go through our resources documents for extra and 3rd party (vetted) resources to help improve overall technical writing.
Pre-submission checks
Please ensure that you have only one open topic suggestion (issue) + in-review article (linked pull request) at a time. This will ensure that we complete the review process in a timely manner from inception to publishing
If you have not already, please go over our Resources Page for more tips and tricks on how to improve your overall technical writing so reviews are swifter, increase the chance of a payout and provide more value to our readers.
For first-time contributors and for more details on our submission guidelines, see our Contributing Guidelines.
Formatting and structure checks
[x] Have you used the correct folder and file structure? - see Contribution Guidelines
[x] Is your article properly formatted in Markdown? - see Github Markdown guide
[x] Have you used our preferred heading structure? - H3 (###) and up
[x] Does your article follow any of the suggested structure formats? - see example formats
[x] Is your article over 750 words? - Tool: Wordcounter.net
[x] Is the article you are submitting an in-depth and unique article? Does it go beyond what is in the official docs and what is covered in other blog sites. See these articles as examples.
[x] Does your article provide enough value and detail about your topic? - Articles should be clear, accurate, and fully explained.
[x] Can your article be understood by beginners? - Assume the audience is smart but has no prior exposure to the common terminology in your article.
[x] Have you included a hero image that is 600x400 pixels, under 300KB in size, and have the copyright to use it? - Tools: Pixlr Image Editor for resizing and Unsplash for Creative Commons images
Grammar & plagiarism checks
[x] Have you spellchecked and grammar checked your article? - Tools: Write&Improve - ProWritingAid - Grammarly
[x] Please place your article through a 3rd party plagiarism checker. We suggest using Quetext, this tool is free and has a daily limit. We typically accept articles with 10% or less.
[x] Have you checked your article for readability? - Tool: Hemmingway
[x] Have you added sources for quotes and images that aren't yours?
Technical checks
[x] Are your code snippets properly formatted for syntax highlighting - see Syntax guide
[x] Have you checked your code runs correctly and you've highlighted all necessary dependencies for installation?
[x] Are the software programs and packages you're highlighting in your article up to date, using current versions and not deprecated?
Topic suggestion this closes
Remove the backticks and add the issue number below to link and close your Topic Suggestion (issue) when your article has been published (PR has been merged). See this video for more details.
This closes #issue-number
Finally, delete the article checklist notes in blockquotes and submit your PR. We look forward to reviewing your article.
This fixes #3187
Hey @marienjus .
We are happy to have you in the program. Thank you for submitting this article for review. As a react developer, my interest is definitely piqued.
A member of the review team will jump on the PR shortly. In the meantime, please fill in the PR checkboxes as required.
Cheers,
Daniel
I believe this was for a tryout.
Just and FYI - we are very close to filling all the new peer reviewer positions. (as we only have a few open).
But we can always use this as a reference for next time when we start adding more.
Hey @marienjus ,
Can you link this PR to an issue?
On a quick glance, I find this topic already covered here. We have several articles on the same topic.
How is your article different? I guess you've just used a different example for explaining the same concept.
Sorry, I had just mentioned the issue number under comments. Hey Hookstate and React Query are different libraries. React Query is more used for server state and caching IMO.
Sorry, I had just mentioned the issue number under comments. Hey Hookstate and React Query are different libraries. React Query is more used for server state and caching IMO.
Thanks for the clarification. Looking forward to read your article.
Good afternoon and thank you for submitting your topic to the EngEd program and jumping on your revisions.
As a rule of thumb - we can only accept articles (pull requests) that are fully polished and complete articles - articles that are ready to be published.
All revisions and edits should be completed on your own forked repo (as to not take up room in the queue).
Once edits are complete we can re-consider re-opening this
Hey @marienjus ,
You may create a new PR with all the required changes.
|
2025-04-01T04:35:28.137030
| 2024-12-21T05:07:57
|
2753765225
|
{
"authors": [
"RoyalOughtness",
"nihil-admirari",
"spaceoden"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10676",
"repo": "secureblue/secureblue",
"url": "https://github.com/secureblue/secureblue/pull/714"
}
|
gharchive/pull-request
|
feat: yafti: harden ramdisk
#697
modifying fstab in yafti is a bit scary :smile:
we should test this thoroughly, maybe even keep it in PR until we can run it through @girlpocalypse's new integration testing action
Mount points can be modified with systemd mount units without the need to touch fstab. See e.g. https://github.com/konstruktoid/hardening/blob/master/config/tmp.mount.
|
2025-04-01T04:35:28.138654
| 2023-06-11T20:59:05
|
1751601906
|
{
"authors": [
"darioalessandro"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10677",
"repo": "security-union/zoom-rs",
"url": "https://github.com/security-union/zoom-rs/issues/57"
}
|
gharchive/issue
|
(Tauri) Limit MacOS min version to 13.4
Is your feature request related to a problem? Please describe.
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
Describe the solution you'd like
Set MacOS min version to 13.4 using the tauri plist key: LSMinimumSystemVersion
Lol, MacOS is
|
2025-04-01T04:35:28.165453
| 2018-05-01T06:58:48
|
319132153
|
{
"authors": [
"markdalgleish",
"ryanseddon",
"seek-oss-ci"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10678",
"repo": "seek-oss/html-sketchapp-cli",
"url": "https://github.com/seek-oss/html-sketchapp-cli/pull/39"
}
|
gharchive/pull-request
|
feat(middleware): Add concept of middleware for calling html-sketchapp APIs
Since I landed brainly/html-sketchapp#90 with resizingCintraints api I've been trying to think of a good way to expose calling APIs from here.
So I've come up a kind of middleware concept with this as something that works but happy to iterate on this.
Using it
html-sketchapp-cli --symbol-middleware 'src/middleware.js'
cli.js then requires this based on the presence of this flag and passes it through to snapshotSymbols which then calls it while looping over each layer.
Where it passes through a bunch of stuff the most important one being the layer so I can call layer.setResizingConstraint() on it and a few of the constructors to check instanceof types and change resizingContraints based on that.
Example middleware
module.exports = (args) => {
const { layer, SVG, RESIZING_CONSTRAINTS } = args;
const { TOP, LEFT, WIDTH, HEIGHT } = RESIZING_CONSTRAINTS;
if(layer instanceof SVG) {
layer.setResizingConstraint(TOP, LEFT, WIDTH, HEIGHT);
} else {
layer.setResizingConstraint(LEFT, TOP);
}
};
Just a couple questions, which I apply to every potential feature:
Since we're trying to be a bit more opinionated in this tool, is this something we could be opinionated about? Does it even need to be configurable at all? Why/why not?
If it needs to be configurable, could it be locally configurable, via data attributes, etc?
Since we're trying to be a bit more opinionated in this tool, is this something we could be opinionated about? Does it even need to be configurable at all? Why/why not?
It definitely could be but was trying not to enforce my use case on other people who might use this tool.
If it needs to be configurable, could it be locally configurable, via data attributes, etc?
It can in some parts and I have some code (not pushed) that uses data-sketch-resizing-constraint attributes where possible. But for somethings like extracting out svg backgrounds and creating actual elements for pseudo-elements would require a hook like I've done here to modify the sketch layer.
So this could be both, use the data attribute if possible otherwise you can fallback to the middleware concept or whatever it ends up being.
So this is what I had when using data attributes:
<Alert type="success">
<Title data-sketch-resizing-constraint="top left">Success Alert</Title>
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut laboreet dolore magna aliqua.
</Alert>
Some elements I can apply this others I can't like the text below <Title>. This component also has a background svg applied that i inject at run time, see brainly/html-sketchapp#81.
So I think we'd need both cases.
Had a closer look at this, and I think this makes sense to me if I view it as an escape hatch. Long term, I feel like resizing constraints should be handled for you, but I can see the need for this as an interim solution.
For this to be approved, I think we need a new snapshot test that sets some resizing constraints, and some documentation in the readme.
Also, do you think we could add support for functions as config values too, not just file paths, so you can inline the middleware directly in html-sketchapp.config.js?
Also, do you think we could add support for functions as config values too, not just file paths, so you can inline the middleware directly in html-sketchapp.config.js?
Heh, I originally had it as an inline function but thought people wouldn't like that. Will add support for both a string on inline function.
@markdalgleish any more feedback?
Just a heads up that I've upgraded html-sketchapp in master so I could also fix the install command in v3, so this PR will need to be updated.
Donions
:tada: This PR is included in version 0.5.0 :tada:
The release is available on:
npm package (@latest dist-tag)
GitHub release
Your semantic-release bot :package::rocket:
@ryanseddon Thanks so much for the PR, and thanks for your patience on this one. Really appreciate your contribution here 👏
|
2025-04-01T04:35:28.211339
| 2024-05-26T10:12:49
|
2317622092
|
{
"authors": [
"Piumal1999",
"anjula-sack",
"dilankavishka"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10679",
"repo": "sef-global/scholarx-frontend",
"url": "https://github.com/sef-global/scholarx-frontend/issues/105"
}
|
gharchive/issue
|
Create a separate component for mentor card (horizontal view)
Is your feature request related to a problem? Please describe.
In the admin's mentee profile view, we are currently displaying the mentor details like this:
Currently, it is built inside the MentorProfile component, rather than as a separate component.
Describe the solution you'd like
Make it a separate component. Consider other pages where this component can be reused and adjust the design accordingly.
Describe alternatives you've considered
N/A
Additional context
Bug introduced with #103
Shall i work on this issue please? @Piumal1999
Any updates on this? @dilankavishka
can i get some more time to do this work please? @anjula-sack
Any updates @dilankavishka
Yeah anjula. i'm working with this.. maybe i am able do finish this before this saturday
Are you working on this? @dilankavishka
Yeah.
|
2025-04-01T04:35:28.215304
| 2021-07-16T14:51:24
|
946364327
|
{
"authors": [
"anjula-sack",
"kumuditha-udayanga"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10680",
"repo": "sef-global/scholarx-frontend",
"url": "https://github.com/sef-global/scholarx-frontend/issues/204"
}
|
gharchive/issue
|
Prevent showing 'Manage' button to the mentor in "MENTEE_APPLICATION" state
Describe the bug
MENTEE_APPLICATION period showing a manage button for mentors (in my programs)
To Reproduce
Add a condition to prevent showing the button.
Expected behaviour
A clear and concise description of what you expected to happen.
'Manage' button shouldn't appear to mentors in the "MENTEE_APPLICATION" state.
Desktop (please complete the following information):
OS: MacOS BigSur
Browser : Safari
Version : 14.0.1
@anjula-sack, can you assign me for this issue
Go ahead
|
2025-04-01T04:35:28.218208
| 2016-11-15T02:15:06
|
189281216
|
{
"authors": [
"CannedGeo",
"ShiangYong"
],
"license": "apache-2.0",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10681",
"repo": "seg/2016-ml-contest",
"url": "https://github.com/seg/2016-ml-contest/issues/15"
}
|
gharchive/issue
|
PE regression?
Hi all- Has anyone tried generating a PE for the wells that do not have it (Alexander D and Kimzey A)?
I have made some attempts using from sklearn.svm import SVR and playing around with the different models. Do you think this would lead to a valid answer? Are there better regression techniques to use?
I started a repo for PE regression that includes a notebook I have been playing with. To call it sloppy is probably an understatement, but its there as a work in progress.
If I can come up with a solution that's worth further discussion, maybe this can be looked at closer. Will close.
Hi CannedGeo, in my submission I used a convolutional neural network to do PE regression. Comparing my classifier using imputed PE via regression against just using mean value of PE, I obtained higher F1 scores with the regression PE.
ar4's submission also confirmed this.
Hi @ShiangYong, thanks for the input. It definitely seems more helpful to use a regression model for the PE instead of mean. I also had decent LOGO results using gradient boost regressor. Good luck!
|
2025-04-01T04:35:28.228015
| 2017-03-29T22:07:58
|
218023859
|
{
"authors": [
"Jaimal",
"codecov-io",
"hankim813"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10682",
"repo": "segment-integrations/analytics.js-integration-twitter-ads",
"url": "https://github.com/segment-integrations/analytics.js-integration-twitter-ads/pull/12"
}
|
gharchive/pull-request
|
Fire singleTags AS WELL AS PageView when Universal Website Tag ID
Hey Guys,
So this fix is for Turner, the logic has it that EITHER the PageView OR the singleTag were fired. Turner wants to use both.
https://segment.atlassian.net/browse/EPD-2462
@hankim813 @sperand-io
Codecov Report
Merging #12 into master will not change coverage.
The diff coverage is 100%.
@@ Coverage Diff @@
## master #12 +/- ##
=====================================
Coverage 100% 100%
=====================================
Files 1 1
Lines 99 99
=====================================
Hits 99 99
Impacted Files
Coverage Δ
lib/index.js
100% <100%> (ø)
:arrow_up:
Continue to review full report at Codecov.
Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update ef919c9...01c156f. Read the comment docs.
@hankim813 Added the unit test
after deploy merge: https://github.com/segmentio/site-docs/pull/2193
|
2025-04-01T04:35:28.253958
| 2021-11-19T16:49:21
|
1058727109
|
{
"authors": [
"brandongregoryscott",
"vojtatom"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10683",
"repo": "segmentio/evergreen",
"url": "https://github.com/segmentio/evergreen/issues/1365"
}
|
gharchive/issue
|
Empty divs in evergree-portal-container using Tooltip
Hello, I think I have encountered similar issue to this one https://github.com/segmentio/evergreen/issues/329 using the Tooltip component.
I'm using version 6.6.1
React:
<Tooltip content="Layers" >
<IconButton icon={LayersIcon} appearance="minimal" onClick={() => toggleMenu(Menu.Layers)} />
</Tooltip>
Looks like tooltip creates one extra div on hover, see:
Before hover:
On hover:
After mouse leave:
I suppose https://github.com/segmentio/evergreen/pull/330 did not fix the problem or am I missing something?
Thank you for such a great library guys. 😊
This should be resolved v6.13.1
|
2025-04-01T04:35:28.271560
| 2024-12-10T19:22:00
|
2730994806
|
{
"authors": [
"TravisHRabble",
"pwseg"
],
"license": "CC-BY-4.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10684",
"repo": "segmentio/segment-docs",
"url": "https://github.com/segmentio/segment-docs/pull/7309"
}
|
gharchive/pull-request
|
Rabble AI Destination Documentation
Proposed changes
This is the initial commit for Rabble AI Destination Documentation
Merge timing
ASAP once approved
Related issues (optional)
just one change for me
Went ahead and committed this so we can deploy today.
|
2025-04-01T04:35:28.275107
| 2018-01-31T00:31:05
|
292989151
|
{
"authors": [
"jocgir",
"metmajer"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10685",
"repo": "segmentio/terraform-docs",
"url": "https://github.com/segmentio/terraform-docs/pull/45"
}
|
gharchive/pull-request
|
Extend options
Hi,
I added several new options to terraform-docs in order to be able to export results in various format YAML, HCL, XML in addition to JSON and MD. It is also possible to select only part of the data (inputs or outputs).
We can also inject variables files and even output results coming either from terraform output -json or outfile from terraform plan -out outfile (working with terraform 0.11.2).
I invite you to try it and don't hesitate to communicate with me if you have any comment or question.
Regards,
Jo
P.S. Our binaries are available in https://github.com/coveo/terraform-docs/releases
Hi @metmajer,
Sure, I will check how to split the whole thing and have a look to the new features and project structure.
Regards,
Jo
@jocgir I really admire the work you've put into your fork. However, I feel it's too bulky and I would prefer to see it split up into smaller feature branches that we can better handle. I would see the various output formats as a very useful addition to this project and the new architecture should make it easy for you to incorporate your changes if you are interested.
I'll, therefore, close this pull request, but I am looking forward to seeing some of your work in the upstream repository. Thanks!
|
2025-04-01T04:35:28.289113
| 2021-11-28T19:33:14
|
1065416028
|
{
"authors": [
"Bintaaaa",
"seikosantana"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10686",
"repo": "seikosantana/gampah",
"url": "https://github.com/seikosantana/gampah/issues/6"
}
|
gharchive/issue
|
where is the navigation back button? sometimes it goes to getStartedPage or blank page.
I changed the navigation, have you received my pull?
I investigated this and seems like the issue is caused by the CustomToolbarWidget using pop while the profile widget is not navigated using push. We replaced named and it changes the tree, not appending into it so pop leads to the root tree.
https://github.com/seikosantana/gampah/blob/bf51551769c4d1a4814ced895bfb5931260a881c/lib/ui/pages/page_profile.dart#L26-L28
https://github.com/seikosantana/gampah/blob/bf51551769c4d1a4814ced895bfb5931260a881c/lib/ui/widgets/widget_toolbar.dart#L5-L20
https://github.com/seikosantana/gampah/blob/bf51551769c4d1a4814ced895bfb5931260a881c/lib/ui/pages/page_home.dart#L68-L70
Other possible cause
https://stackoverflow.com/questions/53723294/flutter-navigator-popcontext-returning-a-black-screen
if you have accepted my pull and you have run it should not need to be changed again
It still happens on my end
Weird. No longer happening.
if it happens again please do a "flutter clean" in your terminal
Noted.
|
2025-04-01T04:35:28.341717
| 2021-06-10T15:00:51
|
917445527
|
{
"authors": [
"kevin-brown",
"sphuber"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10687",
"repo": "select2/select2",
"url": "https://github.com/select2/select2/issues/6048"
}
|
gharchive/issue
|
Allow the selection of the same value multiple times
The current behaviour is that if a value is already selected, if it is selected again from the drop down menu, it will be deselected instead of being a tag being added another time to the selected values. I searched online and the documentation for mention of this functionality but it seems this feature does not exist. How could this be implemented? I am using select2 v4.0.13.
Not sure if this feature request requires an MVE but here goes: https://jsfiddle.net/45fmnoyz/
As you can see, I can only select A once but I would like it to be possible for A to be selectable multiple times.
Duplicate of https://github.com/select2/select2/issues/4472. Going to reopen that one and let it go through a stale cycle to see if there's still enough interest.
|
2025-04-01T04:35:28.355118
| 2022-09-10T15:16:22
|
1368650499
|
{
"authors": [
"admizh",
"amuthansakthivel",
"asolntsev"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10688",
"repo": "selenide/selenide-appium",
"url": "https://github.com/selenide/selenide-appium/issues/76"
}
|
gharchive/issue
|
Support for methods like scrollTo, $(byText()) in selenide-appium
I am trying to use selenide-appium for mobile automation needs. It would be great to have features like scrollTo, finding element by Text in appium. For now these utilities work in selenide-selenium.
@amuthansakthivel Yes, I agree.
I know there is an built-in API for scrolling in Android, but what about iOS?
I am not sure it's possible to find elements by text in Android. But I know people usually find elements by "accessibility id". It's possible now.
Hi @asolntsev
Scrolling is actually painful stuff in android and ios, especially apps built using ionic. Please find the below logic which I used to scroll to an element.
We cannot rely on isDisplayed and isEnabled methods from appium.
Please let me know if you need any other info
static void scrollForMobile(By by) {
String previousPageSource = "";
while (isElementNotEnabled(by) && isNotEndOfPage(previousPageSource)) {
previousPageSource = DriverManager.getDriver().getPageSource();
performScroll();
}
}
private static boolean isNotEndOfPage(String previousPageSource) {
return !previousPageSource.equals(DriverManager.getDriver().getPageSource());
}
private static boolean isElementNotEnabled(By by) {
List<WebElement> elements = DriverManager.getDriver().findElements(by);
if (isAndroid()) {
return elements.isEmpty();
}
/*
Written this logic below to find if the element is actually visible in the screen
In ios, even though the element is not in screen findElements returns list with size 1
I cannot apply this logic to android as there is no visible attribute found for android.
*/
if (!elements.isEmpty()) {
return elements.get(0).getAttribute("visible").equalsIgnoreCase("false");
}
return true;
}
private static void performScroll() {
Dimensions dimensions = Dimensions.getInstance();
if (isAndroid()) performScrollForAndroid(dimensions);
else performScrollForIOS(dimensions);
}
private static void performScrollForAndroid(Dimensions dimensions) {
new AndroidTouchAction((PerformsTouchActions) DriverManager.getDriver())
.press(PointOption.point(dimensions.getCenter(), dimensions.getScrollStart()))
.waitAction(WaitOptions.waitOptions(Duration.ofSeconds(2)))
.moveTo(PointOption.point(dimensions.getCenter(), dimensions.getScrollEnd()))
.release()
.perform();
}
private static void performScrollForIOS(Dimensions dimensions) {
new IOSTouchAction((PerformsTouchActions) DriverManager.getDriver())
.press(PointOption.point(dimensions.getCenter(), dimensions.getScrollStart()))
.waitAction(WaitOptions.waitOptions(Duration.ofSeconds(2)))
.moveTo(PointOption.point(dimensions.getCenter(), dimensions.getScrollEnd()))
.release()
.perform();
}
@Setter
@Getter
public class Dimensions {
private static Dimensions dimensions;
private int scrollStart;
private int scrollEnd;
private int center;
private int deviceHeight;
private Dimensions() {
Dimension dimension = DriverManager.getDriver().manage().window().getSize();
this.scrollStart = (int) (dimension.getHeight() * 0.5);
this.scrollEnd = (int) (dimension.getHeight() * 0.2);
this.center = (int) (dimension.getWidth() * 0.5);
this.deviceHeight = dimension.getHeight();
}
public static Dimensions getInstance() {
if (Objects.isNull(dimensions)) {
dimensions = new Dimensions();
}
return dimensions;
}
@asolntsev Moreover, I have an youtube channel called testing mini bytes and I am going to make a video series on Selenid, its advantages and how to effectively use it. More power to you for creating this amazing library.
For now, I have created a selenide practice project and diving deep into the selenide code and tests code.
https://github.com/amuthansakthivel/SelenidePractice
@asolntsev - I have not tested this but simple google search is returning this result.
JavascriptExecutor executor = (JavascriptExecutor) appiumDriver;
HashMap<String, Object> scrollobj = new HashMap<String, Object>();
scrollobj.put("predicateString", "value == '" + predicateStringValueText + "'");
scrollobj.put("direction", "down");
executor.executeScript("mobile: scroll", scrollobj);
yes, in one Android project I used this code for scrolling:
protected fun scroll(selector: ScrollBy): SelenideElement {
val container = "new UiSelector().scrollable(true)"
val script = "new UiScrollable($container).scrollIntoView(${selector.uiSelector})"
return `$`(MobileBy.AndroidUIAutomator(script)).`as`(selector.uiSelector).shouldBe(visible)
}
@amuthansakthivel and @asolntsev
Yes, current Appium touch actions is migrated to W3C standards. Not sure for iOS, but for Android scrolling like this: https://stackoverflow.com/questions/71037969/how-to-use-actions-or-do-scroll-in-app-in-java
private final static PointerInput FINGER = new PointerInput(TOUCH, "finger");
public static void verticalSwipe(boolean toEnd, int swipeTimeout) {
AppiumDriver appiumDriver = extractAppiumDriver();
Dimension dimension = appiumDriver.manage().window().getSize();
if (toEnd) {
Point start = new Point((int) (dimension.width * 0.5), (int) (dimension.height * 0.9));
Point end = new Point((int) (dimension.width * 0.2), (int) (dimension.height * 0.1));
W3cActions.doSwipe(appiumDriver, start, end, swipeTimeout);
} else {
Point start = new Point((int) (dimension.width * 0.2), (int) (dimension.height * 0.2));
Point end = new Point((int) (dimension.width * 0.5), (int) (dimension.height * 0.8));
W3cActions.doSwipe(appiumDriver, start, end, swipeTimeout);
}
}
public static void doSwipe(AppiumDriver driver, Point start, Point end, int duration) {
Sequence swipe = new Sequence(FINGER, 1)
.addAction(FINGER.createPointerMove(ofMillis(0), viewport(), start.getX(), start.getY()))
.addAction(FINGER.createPointerDown(LEFT.asArg()))
.addAction(FINGER.createPointerMove(ofMillis(duration), viewport(), end.getX(), end.getY()))
.addAction(FINGER.createPointerUp(LEFT.asArg()));
driver.perform(Collections.singletonList(swipe));
}
Hi @admizh - I have already created a PR to add scroll to command with almost similar implementation. Please check - https://github.com/selenide/selenide-appium/pull/80
|
2025-04-01T04:35:28.392961
| 2020-01-30T19:12:32
|
557694468
|
{
"authors": [
"rjyounes",
"sa-bpelakh"
],
"license": "CC-BY-4.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10689",
"repo": "semanticarts/gist",
"url": "https://github.com/semanticarts/gist/issues/164"
}
|
gharchive/issue
|
Remove Visio files since out of date
Does this also apply to the PDFs and PNGs in the Documentation directory?
@rjyounes I don't know how many users looked at the giant PDF rather than loading into Protege and poking around. Technically we have embedded docs, though logical grouping and smart layout is not inherent.
Need to remove.
|
2025-04-01T04:35:28.405500
| 2016-08-04T14:22:32
|
169386624
|
{
"authors": [
"megastef",
"otisg",
"rocketraman"
],
"license": "apache-2.0",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10690",
"repo": "sematext/sematext-agent-docker",
"url": "https://github.com/sematext/sematext-agent-docker/issues/13"
}
|
gharchive/issue
|
On logsene service 503, container stopped with exit code 2
Saw this from the :dev container:
sematext-agent-docker-1: Container stopped with exit code 2
2016-08-04T14:19:03.835216077Z at emitNone (events.js:91:20)
The container automatically restarted. Is there a way to see the logs of the stopped container so I can provide the full stack trace?
Actually we have six of these over six nodes, and all of them are throwing this error within a few minutes of each other.
The "latest" Image (1.30.15) should be updated to support you JSON logs https://github.com/sematext/sematext-agent-docker/issues/12. Does it work?
Feel free to post stack trace of "dev" image.
I don't know how to get the stack trace. By the time the error above happens, the container has already restarted, and the logs of the exited container are no longer accessible via Docker Cloud. I do see a lot of these in the logs of the running containers though -- perhaps it is related:
sematext-agent-docker-5 | 2016-08-04T14:47:31.361141497Z 2016-08-04T14:47:30.951Z - error: Error in logsene-js: source=logsene, message=Logsene status code:503, httpStatus=503, httpBody=, url=https://logsene-receiver.sematext.com/0b792730-a746-44f6-8d8b-efcb443a84ae/_bulk
By the time the error above happens, the container has already restarted, and the logs of the exited container are no longer accessible via Docker Cloud.
Shouldn't you be able to see the logs in Logsene?
Shouldn't you be able to see the logs in Logsene?
Apparently no. The closest I get is this event from dockercloud/events:latest:
sending event: {"status":"die","id":"7681056fd9938b1cec6fec4a2ffb2b7baf3eb6cb52395428f76f43ee996a930a","from":"sematext/sematext-agent-docker:dev","time":1470321899834976922,"exitcode":"2"}
@otisg by default SDA logs are not logged to Logsene anymore.
Ah, sorry, I didn't realize this was about the SDA container. But one can enable SDA log shipping to Logsene, so @rocketraman could use that and capture those logs, too.
@otisg @megastef If configured to send logs to logsene, will the SDA container capture the logs of error conditions that prevent logging to logsene, and send them once it comes back up?
Also: the last time the SDA container exited with code 2 corresponds with the last time I see this error in the logs:
sematext-agent-docker-5 | 2016-08-04T14:47:31.361141497Z 2016-08-04T14:47:30.951Z - error: Error in logsene-js: source=logsene, message=Logsene status code:503, httpStatus=503, httpBody=, url=https://logsene-receiver.sematext.com/0b792730-a746-44f6-8d8b-efcb443a84ae/_bulk
The reported error is probably not the reason for a crash. Only "uncaught errors" with the message to contact support are critical / lead to exit. HTTP 503, is normally not critcal, logs get then buffered, maybe too long and we face a memory issue then? Do you use SPM and know SDA memory usage?
@rocketraman Did you try released version 1.30.15 / latest?
@rocketraman you can get stats about logging (shipped, failed, restransmit) using -e ENABLE_LOGSENE_STATS=true
If you start SDA without "-d" it should run in foreground in the terminal and you could copy error messages from console. You might want to use -e SPM_LOG_LEVEL=debug (so we get more surrounding messages).
Feel free to share full log via e-mail<EMAIL_ADDRESS>- if you don't like to post it here.
Do you use SPM and know SDA memory usage?
SPM reports SDA container memory is relatively consistent, way below limit, and no memory allocation failures.
@rocketraman Did you try released version 1.30.15 / latest?
I haven't yet. I was seeing if I could help debug this issue first.
The error has stopped happening since the 503's stopped occurring. So not sure adding debug logging now will tell us anything useful, unless you can simulate the 503's for my account on the server-side, and we can see if the error happens again.
I was able to reproduce this error. @rocketraman thank you for reporting!
Awesome, is this in dev already? Can you post here when this is in latest?
Testing v1.30.16 right now.
RE: storing failed logs over restarts. If you want to make sure that logs are shipped after restarts, then use
-v /tmp:/logsene-log-buffer
/tmp on host stores then logs, in case of a network or service outage. Docker Agent deletes this files after successful transmission. Taken from https://github.com/sematext/sematext-agent-docker#configuration-parameters
@rocketraman latest build for 1.30.16 is available. https://hub.docker.com/r/sematext/sematext-agent-docker/builds/. Again, thank you for reporting this issue. We use a service returning 503 for tests :)
|
2025-04-01T04:35:28.415118
| 2022-07-27T13:19:17
|
1319555479
|
{
"authors": [
"lsroman"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10691",
"repo": "semrush/intergalactic",
"url": "https://github.com/semrush/intergalactic/pull/322"
}
|
gharchive/pull-request
|
TEST preview site for pull request
What changed?
🦄...
Definition of Done
[ ] Dependencies checked (if applicable)
[ ] TS types updated (if applicable)
[ ] CHANGELOG.md updated (if applicable)
[ ] Complete unit testing (if applicable)
[ ] Complete screenshot testing (if applicable)
[ ] Complete UX review (if applicable)
[ ] Documentation design guide updated (if any)
[ ] Documentation examples updated (if any)
[ ] Documentation API updated (if any)
[ ] Add link to design (if any)
[ ] No major bugs pending
[ ] Complete self code review
Reviewer checklist
[ ] Does the code work? Does it perform its intended function and the logic is correct?
[ ] Is the code easily understandable?
[ ] Verify the design is implemented as per the design requirements
[ ] Verify the names used in the programs/methods/functions convey the intent and all functions commented
[ ] Verify the unit tests are testing the code to perform the intended functionality
[ ] Verify the unit tests cover the positive and negative scenarios
[ ] Verify the screenshot tests are added and are they comprehensive
[ ] Verify documentation (design/api/example) has been added/updated
/gcbrun
|
2025-04-01T04:35:28.421435
| 2018-10-09T14:58:37
|
368254950
|
{
"authors": [
"SendGridDX",
"spencerbyw",
"thinkingserious"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10692",
"repo": "sendgrid/docs",
"url": "https://github.com/sendgrid/docs/pull/4101"
}
|
gharchive/pull-request
|
Fix various typos in the documentation
Description of the change:
Fixes for a couple wording issues I found while reading through the docs.
Reason for the change:
Fix wording.
Link to original source:
N/A
Thank you for your submission, we really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.You have signed the CLA already but the status is still pending? Let us recheck it.
Hello @spencerbyw,Thanks again for the PR! It's HACKTOBERFEST! We want to show our appreciation by sending you some special Hacktoberfest swag. If you have not already, could you please fill out this form so we can send it to you? Thanks! Team SendGrid DX
Hello @spencerbyw,Thanks again for the PR! We noticed that you have earned 1 out of the 5 points needed to receive glorious SendGrid Hacktoberfest swag. Please take a moment to checkout this link to find more issues to get you past the required threshold. Also, please be sure that you have officially registered here. Thank you and Happy Hacktobering! Team SendGrid DX
Hello @spencerbyw,Thanks again for the PR! You have earned 1 out of the 5 points needed to receive glorious SendGrid Hacktoberfest swag. Please take a moment to checkout this link to find more issues to get you past the required threshold or to simply continue the celebration. Also, please be sure you have officially registered with us here by November 1, 2018 to qualify. If you have any questions you can email us at<EMAIL_ADDRESS> Thank you and Happy Hacktobering! Team SendGrid
|
2025-04-01T04:35:28.430939
| 2020-11-25T18:07:47
|
751024057
|
{
"authors": [
"jarscr",
"thinkingserious"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10693",
"repo": "sendgrid/sendgrid-php",
"url": "https://github.com/sendgrid/sendgrid-php/issues/1023"
}
|
gharchive/issue
|
Domain International Extension
Issue Summary
Hi, I have issue sending email from domain with international extension, I don't find one solution for this.
Thanks
Code Snippet
<EMAIL_ADDRESS>, 'Name');
<EMAIL_ADDRESS>, 'Name');
$mail->addTo('me@gmail.com);
Exception/Log
"$emailAddress" must be a valid email address. Got<EMAIL_ADDRESS>
Technical details:
sendgrid-php version: 7.9.1
php version: 7.3
Hello @jarscr,
Thanks for submitting a GitHub issue! We are very sorry that you are running into this problem. In order to better serve you, as this does not present itself as a library specific issue, we would like to ask that you reach out to our support team at https://support.sendgrid.com.
Thank you!
SendGrid DX Team
|
2025-04-01T04:35:28.436593
| 2015-05-22T08:41:58
|
79357507
|
{
"authors": [
"Znow",
"senny"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10694",
"repo": "senny/pdfjs_viewer-rails",
"url": "https://github.com/senny/pdfjs_viewer-rails/issues/4"
}
|
gharchive/issue
|
Unexpected server response - Amazon
Heya
Getting this error when trying to retrieve from Amazon:
PDF.js v1.0.907 (build: e9072ac)
Message: Unexpected server response (0) while retrieving PDF "https://xxxxx.s3.amazonaws.com/uploads/document/document/3/xxxxx.pdf".
Duplicate of #12. The linked issue also contains required steps to solve the issue.
|
2025-04-01T04:35:28.439380
| 2015-07-23T18:13:48
|
96872422
|
{
"authors": [
"shakycode"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10695",
"repo": "sensu-plugins/sensu-plugin",
"url": "https://github.com/sensu-plugins/sensu-plugin/issues/94"
}
|
gharchive/issue
|
only handling every 180 occurrences
I have the latest sensu-core installed and I've installed the sensu-plugin gem using the omnibus package.
Sensu is up and running no problem, but I'm running into major issues with handlers. I've setup both the standard ruby mailer.rb from the community plugins, as well as the pagerduty handler.
When I add the handlers to a check that I know is failing in the logs I see this:
{\"id\":\"a1e807d2-425f-4891-b291-ed5bda5b6105\",\"client\":{\"name\":\"mfp\",\"address\":\"<IP_ADDRESS>\",\"subscriptions\":[\"test\"],\"version\":\"0.20.0\",\"timestamp\":1437673191},\"check\":{\"command\":\"/etc/sensu/plugins/check-disk.sh -d /dev/md1 -w 80 -c 90\",\"interval\":10,\"handlers\":[\"default\",\"pagerduty\",\"mailer\"],\"occurrences\":3,\"subscribers\":[\"test\"],\"name\":\"disk\",\"issued\":1437673279,\"executed\":1437673201,\"duration\":0.01,\"output\":\"CRITICAL - /dev/md1 space used=94% | '/dev/md1 usage'=94%;80;90;\\n\",\"status\":2,\"history\":[\"2\",\"2\",\"2\",\"2\",\"2\",\"2\",\"2\",\"2\",\"2\",\"2\",\"2\",\"2\",\"2\",\"2\",\"2\",\"2\",\"2\",\"2\",\"2\",\"2\",\"2\"],\"total_state_change\":0},\"occurrences\":513,\"action\":\"create\"}"]}
{"timestamp":"2015-07-23T12:41:20.873279-0500","level":"info","message":"handler output","handler":{"type":"pipe","command":"/etc/sensu/handlers/pagerduty.rb","api_key":"2e24c101ef754ff6add39aaaffa846c9","name":"pagerduty"},"output":["only handling every 180 occurrences: ws/disk\n"]}
{"timestamp":"2015-07-23T12:41:20.873561-0500","level":"info","message":"handler output","handler":{"type":"pipe","command":"/etc/sensu/handlers/pagerduty.rb","api_key":"2e24c101ef754ff6add39aaaffa846c9","name":"pagerduty"},"output":["only handling every 180 occurrences: wwv/disk\n"]}
{"timestamp":"2015-07-23T12:41:21.055407-0500","level":"info","message":"handler output","handler":{"type":"pipe","command":"/etc/sensu/handlers/pagerduty.rb","api_key":"2e24c101ef754ff6add39aaaffa846c9","name":"pagerduty"},"output":["only handling every 180 occurrences: mfp/disk\n"]}````
So what it looks like is the handler is firing but being filtered by the sensu-handler.rb gem and not creating emails or pagerduty events.
Here's the offending line
`only handling every 180 occurrences`
Does anyone know what the problem is? Is this a bug? Do I need to override something? I have a ton of occurrences in sensu for this particular check (check_disk) so it should be firing to pagerduty and the mailer.
If I need to move this to serverfault I will, but this seems like a bug in sensu-handler.rb and I'm not sure how to troubleshoot this.
After talking with the dev team in IRC it looks like the sensu-plugin needed the api host declarative in config.json. This resolved my issue and was hotpatched yesterday, docs need to be updated to note this need.
|
2025-04-01T04:35:28.442483
| 2016-07-16T07:38:29
|
165917394
|
{
"authors": [
"CVTJNII",
"cwjohnston",
"fessyfoo"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10696",
"repo": "sensu-plugins/sensu-plugin",
"url": "https://github.com/sensu-plugins/sensu-plugin/pull/142"
}
|
gharchive/pull-request
|
sensu-handler: filter_repeated pass resolve events
resolution events should not be filtered by filter_repeated
On-call personnel may not receive resolve notifications due to this
filtering method. I believe this is a bad user experience which may lead
some people to mistrust Sensu as a monitoring and alerting system. I
think this is detrimental to our community
-- cwjohnston
https://github.com/sensu-plugins/sensu-plugin/pull/136#issuecomment-232738538
note, this is not in opposition to the goal of moving filtering out of sensu-handler.rb #136 it just fixes an issue prior to moving filtering out as well.
this PR is very similar to https://github.com/sensu-plugins/sensu-plugin/pull/103#issuecomment-158998616
and will suffer the same problem:
Even if your sequence of events didn't meet the Warning or Critical occurrence threshold necessary to invoke the handler, the handler will process the subsequent OK.
For example
0, 0, 0, 1, 0, 0, 0, 0
> May not trigger a notification because of the warning, but it'll always send the OK.
however taking that as requirement, then that's a very subtle thing and [built into core filtering](https://github.com/sensu/sensu-extensions/issues/10) should consider it.
I'm going to close this as the filter_repeated method is deprecated and I believe the occurrences extension provides an improved implementation using event watermarks.
:+1: for merging for those of us who will still be running this code for a while. We're running Sensu in Prod and changes take time, plus alert handling and filtering is changing way to rapidly to deploy so we're going to keep running the older handlers until the dust settles.
|
2025-04-01T04:35:28.448291
| 2018-03-19T23:01:42
|
306669427
|
{
"authors": [
"mercul3s"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10697",
"repo": "sensu/sensu-docs-site",
"url": "https://github.com/sensu/sensu-docs-site/pull/275"
}
|
gharchive/pull-request
|
RBAC - read-only user guide and reference doc refactoring
Closes #245
Closes #267
@palourde I think I've addressed all your comments, ready for another round of review.
|
2025-04-01T04:35:28.455565
| 2017-02-11T20:36:52
|
207006547
|
{
"authors": [
"agoddard",
"cwjohnston",
"dkinon",
"portertech"
],
"license": "mit",
"license_source": "bigquery",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10698",
"repo": "sensu/sensu",
"url": "https://github.com/sensu/sensu/issues/1551"
}
|
gharchive/issue
|
Sensu Stashes Disappear
When I stash both client and check specific alerts with no expiration via uchiwa, some time later (seems to be arbitrary but always within1-2 days) the stashes just disappear and alerts start flowing again. So far I haven't been unable to correlate this to any specific log entries.
Expected Behavior
Sensu stashes with no expiration should last forever.
Current Behavior
Sensu stashes with no expiration only last from 1-2 days before disappearing.
Possible Solution
No ideas.
Steps to Reproduce (for bugs)
These steps are
In uchiwa, stash a client or client check with no expiration.
Wait 1-2 days
Watch for stashes to disappear
Context
Using stashes the standard way to keep alerts notifications from spewing (using slack plugin). So far I've only tested/reproduced this issue on stashes with no expiration so I'm not certain this would affect stashes with a defined expiration.
Your Environment
Sensu version used: 0.26.5-2 (rpm from http://repositories.sensuapp.org/yum/el/7/x86_64/)
Active Sensu extensions: None
Operating System and version: CentOS Linux release 7.2.1511
Transport and runtime versions: RabbitMQ 3.6.5-1 (rpm from https://www.rabbitmq.com/releases/rabbitmq-server/v3.6.5/) on Erlang R16B-03.18 (rpm from epel7) and Redis 2.8.19-2 (rpm from epel7)
For confirmation, the sensu api does report the stashes correctly when they are created:
# curl -s http://localhost:4567/stashes | jq [ { "path": "silence/research.XXXXX.XXX", "content": { "reason": "Disable Alerts", "source": "uchiwa", "timestamp":<PHONE_NUMBER> }, "expire": -1 }, { "path": "silence/qa.XXXXX.XXX", "content": { "reason": "Disable Alerts", "source": "uchiwa", "timestamp":<PHONE_NUMBER> }, "expire": -1 } ]
I put a script in place to notice when the stashes disappear and may have found the cause of this issue. My script checks once a minute and alerted at 2017-02-14T09:53:01, there was nothing interesting in the sensu-api logs but in the sensu server logs I found these unexpected logs occurring within the previous minute:
{"timestamp":"2017-02-14T09:52:16.473590+0000","level":"warn","message":"resigning as leader"}
{"timestamp":"2017-02-14T09:52:26.474388+0000","level":"info","message":"i am now the leader"}```
As you can see a brief leader election takes place which may look completely normal except that there is no other sensu-server running, I only run a single instance (and I reconfirmed this in my environment just to be 100% sure). What could be causing this phantom leader election and does a leader election cause stashes to be cleared from wherever they are stored?
@dkinon from my perspective, both disappearing stashes and unexpected leader election are consistent with the Sensu data store being wiped out (e.g. Redis flushdb or similar. Is some scheduled or automated process performing maintenance on your Redis server?
@cwjohnston thanks for the perspective, I don't have any redis maintenance jobs that I know of but I'll comb through the environment to see if I find anything. In the mean time here are the redis logs surrounding the last occurrence:
[7720] 14 Feb 09:49:20.073 * Background saving started by pid 638
[638] 14 Feb 09:49:20.078 * DB saved on disk
[638] 14 Feb 09:49:20.078 * RDB: 0 MB of memory used by copy-on-write
[7720] 14 Feb 09:49:20.174 * Background saving terminated with success
[7720] 14 Feb 09:52:12.789 * DB saved on disk
[7720] 14 Feb 09:52:13.808 * DB saved on disk
[7720] 14 Feb 09:57:14.045 * 10 changes in 300 seconds. Saving...
[7720] 14 Feb 09:57:14.045 * Background saving started by pid 1864
[1864] 14 Feb 09:57:14.049 * DB saved on disk
[1864] 14 Feb 09:57:14.050 * RDB: 0 MB of memory used by copy-on-write
[7720] 14 Feb 09:57:14.146 * Background saving terminated with success```
The 2 successive `DB saved on disk` entries right before the stashes are lost doesn't seem normal (based on the other saves I'm seeing in the logs) but I don't see any evidence of an error or a flush. I'm using the redis-2.8.19-2.el7.x86_64 rpm from epel7. I don't have much experience with access redis data, is there an easy way for me to watch the stashes programmatically in redis?
@dkinon Redis has a MONITOR command, which streams the commands being executed on your Redis database.
thanks @cwjohnston, I've got a script logging the output of redis-cli monitor and am waiting for another occurrence.
Hi @dkinon have you seen this issue again after setting up the monitoring?
Closing this issue.
|
2025-04-01T04:35:28.462523
| 2018-06-15T08:58:30
|
332696680
|
{
"authors": [
"azupanc",
"devisperessutti",
"fvivian",
"nilshempelmann"
],
"license": "MIT",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10699",
"repo": "sentinel-hub/eo-learn",
"url": "https://github.com/sentinel-hub/eo-learn/issues/2"
}
|
gharchive/issue
|
TRUE-COLOR-S2-L1C layer configuration
I was trying to run the example land-cover-map over Switzerland on my local machine but ran into a problem:
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-20-54b437ee3056> in <module>()
8 results = workflow.execute({input_task:{'bbox':bbox, 'time_interval':time_interval},
9 export_val_sh:{'filename':f'CH_data/valid_count-L1C/{tiff_name}'},
---> 10 save:{'eopatch_folder':patch_name}
11 })
~\Anaconda3\lib\site-packages\eolearn\core\eoworkflow.py in execute(self, input_args)
280 input_args = {WorkflowResult.get_key(k): v for k, v in input_args.items()} if input_args else {}
281
--> 282 _, intermediate_results = self._execute_tasks(input_args=input_args, outdegs=outdegs)
283
284 return WorkflowResult(intermediate_results)
~\Anaconda3\lib\site-packages\eolearn\core\eoworkflow.py in _execute_tasks(self, input_args, outdegs)
303 result = self._execute_task(input_args=input_args,
304 intermediate_results=intermediate_results,
--> 305 task_id=t_id)
306
307 intermediate_results[t_id] = result
~\Anaconda3\lib\site-packages\eolearn\core\eoworkflow.py in _execute_task(self, input_args, intermediate_results, task_id)
331 inputs = tuple(intermediate_results[t_dep] for t_dep in self.deps[task_id])
332 LOGGER.debug("Computing %s(*%s, **%s)", str(task), str(inputs), str(kw_inputs))
--> 333 return task(*inputs, **kw_inputs)
334
335 def _relax_dependencies(self, *, intermediate_results, out_degrees, current_task_id):
~\Anaconda3\lib\site-packages\eolearn\core\eotask.py in __call__(self, *eopatches, **kwargs)
39
40 def __call__(self, *eopatches, **kwargs):
---> 41 return self.execute(*eopatches, **kwargs)
42
43
~\Anaconda3\lib\site-packages\eolearn\mask\cloud_mask.py in execute(self, eopatch)
246 # Raise error if last channel dimension is less than required
247 if new_data.shape[-1] < len(self.classifier.band_idxs):
--> 248 raise ValueError("Data field has less than the required 10 bands")
249
250 # Compute cloud mask and add as feature to EOPatch
ValueError: Data field has less than the required 10 bands
This message was shown when running the cell [20] in eo-learn/examples/land-cover-map/2_eopatch-L1C.ipynb with some minor tweaks.
Now I know it might be an issue due to my own adaptation to the region of Switzerland. So I would like to try run the notebook exactly as shown in this repository. Therefore I need to set up the layer in the configuartor. What is the specification of the TRUE-COLOR-S2-L1C layer you mention in the 2_eopatch-L1C.ipynb?
Thanks for pointing it out. I've added a README (https://github.com/sentinel-hub/eo-learn/tree/master/examples) where instructions on how to add new configuration to your SH account using a predefined template that has all layers used in the examples are given -- including the TRUE-COLOR-S2-L1C and BANDS-S2CLOUDLESS.
Based on your error I assume you have changed the layer name in the following line
add_clm = AddCloudMaskTask(cloud_classifier,
'BANDS-S2CLOUDLESS',
cm_size_y='160m',
cm_size_x='160m',
cmask_field='CLM',
cprobs_field='CLP',
instance_id=INSTANCE_ID)
If you did, I suggest you to change it back to BANDS-S2CLOUDLESS, since the cloud detector needs the 10 bands in order to make the cloud classification.
Thank you very much, this helped to actually be able to use your notebooks for testing. Unfortunately, now I get the following error from the same cell:
DownloadFailedException: Failed to download with HTTPError:
400 Client Error: Bad Request for url: https://services.sentinel-hub.com/ogc/wcs/0f6c0cd7-002d-4460-8017-4d78502d7d35?SERVICE=wcs&BBOX=370230.5261411405%2C5085303.344972428%2C380225.31836121203%2C5095400.767924464&FORMAT=image%2Ftiff%3Bdepth%3D32f&CRS=EPSG%3A32633&RESX=160m&RESY=160m&COVERAGE=TRUE_COLOR&REQUEST=GetCoverage&TIME=2017-01-01T10%3A04%3A07%2F2017-01-01T10%3A04%3A07&MAXCC=80.0&ShowLogo=False&Transparent=False&EvalScript=cmV0dXJuIFtCMDEsQjAyLEIwNCxCMDUsQjA4LEI4QSxCMDksQjEwLEIxMSxCMTJd
Server response: "Layer TRUE_COLOR not found"
I did not change the notebooks in any way, so I am sure to use the proper version. Now I don't know why it asks for the TRUE_COLOR layer, since the layers specified are TRUE-COLOR-S2-L1C and BANDS-S2CLOUDLESS. The configuration is as explained in the README.
If it is of any help: The error occured after the cell was running for several minutes. Network traffic was on full load, so I guess it was actually downloading some data.
OK, I see now where the problem is. The AddCloudMaskTask actually doesn't create a SentinelHub request using the 'BANDS-S2CLOUDLESS' layer but does this by using eval script (simple script that returns the 10 necessary bands). Unfortunately, the OGC standard requires the LAYER parameter always to be specified although it has no impact on what kind of data is returned (10 bands in this case). The AddCloudMaskTask at the moment sets the layer to 'TRUE_COLOR' when it makes the request (the layer name is hardocded). If this layer doesn't exist then the above error is issued.
This clearly needs to be fixed in the AddCloudMaskTask, but until then please use the updated version of example notebook. The issue can be circumvented by downloading the 10 bands in another task. If the data is already available the cloud detector doesn't download the data again.
@fvivian the issue should be solved now. Please update your version of eo-learn and run the examples again.
Thank you for the explanation and quick fix. The example runs without any issues now.
@fvivian Sorry for opening the issue again.
There is still/again a :
Server response: "Layer BANDS-S2-L1C not found"
Was running your notebook
Is therer a way to query the Layer names?
Hi @nilshempelmann
Make sure the layer configuration is set up properly as for these requirements.
Login to the Sentinel Hub configurator and add a new configuration using the Python scripts template. Then update your instance ID and the script should run.
Got it. Thanks !
|
2025-04-01T04:35:28.467984
| 2023-12-26T15:20:44
|
2056500273
|
{
"authors": [
"docktermj"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10700",
"repo": "senzing-garage/evaluate",
"url": "https://github.com/senzing-garage/evaluate/issues/45"
}
|
gharchive/issue
|
Update repository after transfer
See https://github.com/senzing-garage/garage-roadmap/issues/3
Pull requests:
https://github.com/senzing-garage/docker-xterm/pull/267
|
2025-04-01T04:35:28.468962
| 2023-12-22T19:12:25
|
2054421866
|
{
"authors": [
"docktermj"
],
"license": "Apache-2.0",
"license_source": "github-api",
"license_type": "permissive",
"provenance": "gharchive-dolma-0004.json.gz:10701",
"repo": "senzing-garage/ibm-openshift-guide",
"url": "https://github.com/senzing-garage/ibm-openshift-guide/issues/30"
}
|
gharchive/issue
|
Update repository after transfer
See https://github.com/senzing-garage/garage-roadmap/issues/3
Pull requests:
https://github.com/senzing-garage/ibm-openshift-guide/pull/31
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.