created_at stringlengths 20 20 | test_patch stringlengths 226 26.3k | issue_numbers listlengths 1 1 | instance_id stringlengths 26 28 | repo stringclasses 1
value | patch stringlengths 441 162k | base_commit stringlengths 40 40 | problem_statement stringlengths 24 3.57k | version stringlengths 3 4 | pull_number int64 51 1.41k | hints_text stringlengths 0 9.55k | environment_setup_commit stringlengths 40 40 |
|---|---|---|---|---|---|---|---|---|---|---|---|
2021-01-05T21:40:46Z | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
--- a/.github/workflows/CI.yml
+++ b/.github/workflows/CI.yml
@@ -57,7 +57,7 @@ jobs:
- { os: 'ubuntu-latest', target: 'arm-unknown-linux-gnueabihf', cross: true }
- { os: 'ubuntu-latest', target: 'arm-unknown-linux-musleabihf', cro... | [
"187"
] | GuillaumeGomez__sysinfo-385 | GuillaumeGomez/sysinfo | diff --git a/src/linux/system.rs b/src/linux/system.rs
--- a/src/linux/system.rs
+++ b/src/linux/system.rs
@@ -503,6 +503,14 @@ impl SystemExt for System {
fn get_users(&self) -> &[User] {
&self.users
}
+
+ fn get_name(&self) -> Option<String> {
+ get_system_info("NAME=")
+ }
+
+ fn g... | 5341d2e5229a5d81a6acf1bb1f0d43bb15d8e48c | Add system version and/or OS name
| 0.15 | 385 | The [`whoami` crate](https://crates.io/crates/whoami) has been really useful for this. It offers:
* `whoami::platform()` - a serialisable enum for platform, such as `Linux`, `MacOS`, `Windows `, etc
* `whoami::os()` - a string representation of the OS version, such as `Mac OS X 10.15.3 19D76`
I tend to prefer imple... | 8c2b5a0583404120f1a910d9af32f40fd1dd9d08 |
2020-12-27T22:17:38Z | diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
--- a/.github/workflows/CI.yml
+++ b/.github/workflows/CI.yml
@@ -101,7 +101,13 @@ jobs:
toolchain: ${{ matrix.toolchain }}
override: true
- - name: Execute tests
+ - name: Execute tests (not mac)
run: cargo test... | [
"375"
] | GuillaumeGomez__sysinfo-377 | GuillaumeGomez/sysinfo | diff --git a/Cargo.toml b/Cargo.toml
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -14,7 +14,7 @@ build = "build.rs"
[dependencies]
cfg-if = "0.1"
-rayon = "^1.0"
+rayon = "^1.5"
doc-comment = "0.3"
once_cell = "1.0"
diff --git a/Cargo.toml b/Cargo.toml
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -25,6 +25,12 @@ ntapi = "0.3"
... | 2a2e91cf8cd9319b1dbaa8fb3af889743a06519c | get_disks() doesn't return all mounted disks on Mac OS
I've just started learning Rust. I'm on a 2013 MacBook Air running macOS Sierra and rustc 1.48.0
I modified the sysinfo example to just print out the disks in the system. My boot disk is an external USB 3.0 SSD. (The internal SSD has been missing since I got the... | 0.15 | 377 | That's intriguing. I'll need to check what's going there.
> That's intriguing. I'll need to check what's going there.
Thanks Guillaume. I updated the report by fixing a problem in the code that was mixing 0.3 and 0.15 versions of the example code and tested under Windows 10, which has a different problem.
In utils.r... | 8c2b5a0583404120f1a910d9af32f40fd1dd9d08 |
2020-10-02T14:52:22Z | diff --git a/tests/process.rs b/tests/process.rs
--- a/tests/process.rs
+++ b/tests/process.rs
@@ -104,3 +104,25 @@ fn test_process_disk_usage() {
p.disk_usage().written_bytes
);
}
+
+#[test]
+fn cpu_usage_is_not_nan() {
+ let mut system = sysinfo::System::new();
+ system.refresh_processes();
+
+ ... | [
"366"
] | GuillaumeGomez__sysinfo-367 | GuillaumeGomez/sysinfo | diff --git a/Cargo.toml b/Cargo.toml
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -1,6 +1,6 @@
[package]
name = "sysinfo"
-version = "0.15.2"
+version = "0.15.3"
authors = ["Guillaume Gomez <guillaume1.gomez@gmail.com>"]
description = "Library to get system information such as processes, processors, disks, components and... | f57031a38b0d527958a58605682c52e262f3f017 | Process cpu_usage() returns NaN in some cases
Hello,
I'm using `sysinfo` on version `0.15.2` on Linux mint 19.
`cargo -V` outputs `cargo 1.46.0 (149022b1d 2020-07-17)`.
`rustc -V` outputs `rustc 1.46.0 (04488afe3 2020-08-24)`.
When `system.refresh_process(pid)` is called too often, the cpu_usage() of this process... | 0.15 | 367 | This is indeed where the bug is coming from. However, I'm not too aware on how to compare floats and check if they're equal. If you have any directions, it'd be awesome! (You can also send a PR if you want to go faster :wink: ).
I appreciate the quick reply :)
there are some crates that handle this, like [float_cmp](... | 8c2b5a0583404120f1a910d9af32f40fd1dd9d08 |
2020-06-01T11:28:42Z | diff --git a/src/sysinfo.rs b/src/sysinfo.rs
--- a/src/sysinfo.rs
+++ b/src/sysinfo.rs
@@ -92,7 +92,7 @@ cfg_if! {
#[cfg(test)]
const MIN_USERS: usize = 1;
- } else if #[cfg(unix)] {
+ } else if #[cfg(any(target_os = "linux", target_os = "android"))] {
mod linux;
use linux as... | [
"319"
] | GuillaumeGomez__sysinfo-322 | GuillaumeGomez/sysinfo | diff --git a/src/sysinfo.rs b/src/sysinfo.rs
--- a/src/sysinfo.rs
+++ b/src/sysinfo.rs
@@ -151,7 +151,7 @@ mod utils;
/// let s = System::new_all();
/// ```
pub fn set_open_files_limit(mut _new_limit: isize) -> bool {
- #[cfg(all(not(target_os = "macos"), unix))]
+ #[cfg(any(target_os = "linux", target_os = "a... | 98fb1e1d4fc5187b9701464ebd6674245879bea0 | Fails on FreeBSD: cannot find value `CLOCK_BOOTTIME` in crate `libc`
```
error[E0425]: cannot find value `CLOCK_BOOTTIME` in crate `libc`
--> /wrkdirs/usr/ports/shells/starship/work/starship-0.41.3/cargo-crates/sysinfo-0.14.2/src/linux/system.rs:121:43
|
121 | if unsafe { libc::clock_gettime(libc::CLOCK_... | 0.14 | 322 | I'm surprised that FreeBSD is considered as a unix target... I'm sorry but I have no machine with a FreeBSD installed. So I have two leads to solve this issue:
1. Understand which condition failed, allowing FreeBSD to use unix code.
2. Fix the code directly by using the matching constant on FreeBSD.
Practically e... | 98fb1e1d4fc5187b9701464ebd6674245879bea0 |
2020-04-15T13:12:29Z | diff --git a/src/linux/disk.rs b/src/linux/disk.rs
--- a/src/linux/disk.rs
+++ b/src/linux/disk.rs
@@ -148,3 +72,156 @@ impl DiskExt for Disk {
}
}
}
+
+fn new_disk(name: &OsStr, mount_point: &Path, file_system: &[u8]) -> Disk {
+ let mount_point_cpath = utils::to_cpath(mount_point);
+ let type_ = ... | [
"300"
] | GuillaumeGomez__sysinfo-302 | GuillaumeGomez/sysinfo | diff --git a/Cargo.toml b/Cargo.toml
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -1,6 +1,6 @@
[package]
name = "sysinfo"
-version = "0.13.2"
+version = "0.13.3"
authors = ["Guillaume Gomez <guillaume1.gomez@gmail.com>"]
description = "Library to get system information such as processes, processors, disks, components and... | a6228731cb3464cfc72c3475b2ab0e10448262a0 | test_disks fail in Fedora buildsystem
```rust
---- test_disks stdout ----
thread 'test_disks' panicked at 'assertion failed: s.get_disks().len() > 0', tests/disk_list.rs:14:5
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
failures:
test_disks
test result: FAILED. 0 passed; 1 fai... | 0.13 | 302 | You didn't have this issue before I assume?
On 0.12.0 no, but now I upgraded to 0.13.2 and have it :)
I suspect #297, which is weird considering that it was supposed to handle more disk kinds... Can you show what your `/proc/mounts` file contains please?
```
tmpfs /proc tmpfs rw,seclabel,relatime 0 0
proc /proc proc ... | a6228731cb3464cfc72c3475b2ab0e10448262a0 |
2020-03-19T10:58:00Z | diff --git a/src/sysinfo.rs b/src/sysinfo.rs
--- a/src/sysinfo.rs
+++ b/src/sysinfo.rs
@@ -316,3 +316,18 @@ mod test {
assert!(s.get_users().len() >= MIN_USERS);
}
}
+
+// Used to check that System is Send and Sync.
+#[cfg(doctest)]
+doc_comment!(
+ "
+```
+fn is_send<T: Send>() {}
+is_send::<sysinfo:... | [
"282"
] | GuillaumeGomez__sysinfo-283 | GuillaumeGomez/sysinfo | diff --git a/src/unknown/component.rs b/src/unknown/component.rs
--- a/src/unknown/component.rs
+++ b/src/unknown/component.rs
@@ -26,6 +26,5 @@ impl ComponentExt for Component {
""
}
- fn refresh(&mut self) {
- }
+ fn refresh(&mut self) {}
}
diff --git a/src/unknown/system.rs b/src/unknown/sy... | 90049ce2c712284ee93379e4e5f9c03da41b9ee2 | `System` implements `Send` on Linux, but not on Windows
As the title says, the `System` struct implements `Send` on Linux, but not on Windows.
Take a look at the implemented traits:
https://docs.rs/sysinfo/0.11.7/x86_64-pc-windows-msvc/sysinfo/struct.System.html
https://docs.rs/sysinfo/0.11.7/sysinfo/struct.Sy... | 0.11 | 283 | It shouldn't implement `Send` on any platform. Sending a PR. | 90049ce2c712284ee93379e4e5f9c03da41b9ee2 |
2020-02-10T21:08:36Z | diff --git a/tests/process.rs b/tests/process.rs
--- a/tests/process.rs
+++ b/tests/process.rs
@@ -6,12 +6,12 @@
extern crate sysinfo;
+#[cfg(not(windows))]
+use sysinfo::ProcessExt;
+use sysinfo::SystemExt;
+
#[test]
fn test_process() {
- #[cfg(not(windows))]
- use sysinfo::ProcessExt;
- use sysinfo::S... | [
"261"
] | GuillaumeGomez__sysinfo-262 | GuillaumeGomez/sysinfo | diff --git a/src/traits.rs b/src/traits.rs
--- a/src/traits.rs
+++ b/src/traits.rs
@@ -538,7 +538,7 @@ pub trait SystemExt: Sized {
fn refresh_processes(&mut self);
/// Refresh *only* the process corresponding to `pid`. Returns `false` if the process doesn't
- /// exist or isn't listed.
+ /// exist. I... | d7bfc1e62c91c5f34e9e444bb083fa67716f3051 | Different behavior on window for refresh_process
Hi,
I have a use case where I just ran a new child process, and then I call system.refresh_process(pid), and try to get the process info.
Things work fine on linux and mac, however on windows, the process is not found.
I suspect windows does not add a new proces... | 0.11 | 262 | Great catch and thanks for the precise report! I'll check what's going on. | 90049ce2c712284ee93379e4e5f9c03da41b9ee2 |
2020-01-26T22:19:43Z | diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -15,7 +15,7 @@ Support the following platforms:
* Linux
* Raspberry
* Android
- * Mac OSX
+ * macOS
* Windows
It also compiles for Android but never been tested on it.
diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -... | [
"215"
] | GuillaumeGomez__sysinfo-245 | GuillaumeGomez/sysinfo | diff --git a/Cargo.toml b/Cargo.toml
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -16,17 +16,15 @@ build = "build.rs"
cfg-if = "0.1"
rayon = "^1.0"
doc-comment = "0.3"
+once_cell = "1.0"
[target.'cfg(windows)'.dependencies]
-winapi = { version = "0.3", features = ["fileapi", "handleapi", "ioapiset", "minwindef", "pdh", "... | 4ae1791d21f84f911ca8a77e3ebc19996b7de808 | Feature Request: support retrieve CPU number and load info
Thanks for providing the awesome library for retrieving system information. But some information cannot be retrieved by this crate, like CPU number and CPU average load. (So I must use another crate like https://docs.rs/sys-info/0.5.8/sys_info/index.html).
I... | 0.10 | 245 | Ah indeed. I'll check if it's available as well on windows and OSX, otherwise I'll have a bit more work to do for them. | 4ae1791d21f84f911ca8a77e3ebc19996b7de808 |
2019-06-24T15:31:47Z | diff --git a/src/system.rs b/src/system.rs
--- a/src/system.rs
+++ b/src/system.rs
@@ -25,14 +25,15 @@ mod tests {
#[test]
fn test_refresh_process() {
let mut sys = System::new();
- assert!(sys.refresh_process(utils::get_current_pid()));
+ assert!(sys.refresh_process(utils::get_current_... | [
"182"
] | GuillaumeGomez__sysinfo-183 | GuillaumeGomez/sysinfo | diff --git a/Cargo.toml b/Cargo.toml
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -1,6 +1,6 @@
[package]
name = "sysinfo"
-version = "0.8.6"
+version = "0.9.0"
authors = ["Guillaume Gomez <guillaume1.gomez@gmail.com>"]
description = "Library to handle processes"
diff --git a/src/utils.rs b/src/utils.rs
--- a/src/utils.rs... | 5744eb9221c99229f38375d776ee681032ca4f86 | Make get_current_pid() return Result to prevent panic on unknown platform
| 0.8 | 183 | 5744eb9221c99229f38375d776ee681032ca4f86 | |
2019-06-22T11:19:14Z | diff --git a/src/sysinfo.rs b/src/sysinfo.rs
--- a/src/sysinfo.rs
+++ b/src/sysinfo.rs
@@ -54,7 +54,6 @@ extern crate cfg_if;
extern crate libc;
extern crate rayon;
-#[cfg(test)]
#[macro_use]
extern crate doc_comment;
| [
"76"
] | GuillaumeGomez__sysinfo-178 | GuillaumeGomez/sysinfo | diff --git a/.travis.yml b/.travis.yml
--- a/.travis.yml
+++ b/.travis.yml
@@ -55,7 +55,7 @@ script:
- rustc --version
- sysctl -a | grep mem
- if [[ "$TRAVIS_RUST_VERSION" == "nightly" ]]; then
- rustup component add clippy-preview && cargo clippy;
+ (rustup component add clippy-preview && cargo cli... | bf0691f89ac36af1418a522591cc012ce584d0aa | `sysinfo::System::new()` slow?
A simple program creating a `sysinfo::System` takes several times longer to run (after `cargo build --release`) than existing unix utilities from the procps suite (~40ms instead of ~14ms):
```
extern crate sysinfo;
use sysinfo::SystemExt;
fn main()
{
let mut system = sysinfo::S... | 0.8 | 178 | Even more than reasonable: it sounds great! I'll gladly accept any patch improving the speed of this crate.
A first step on improving this issue is done in #88. I'll try to check how to reduce computation times.
[Here's the flamegraph of the benchmark.](https://rawgit.com/antoyo/d0169a8bbe8204091f261719b9a93c9d/raw/c0f... | 5744eb9221c99229f38375d776ee681032ca4f86 |
2019-02-28T23:34:55Z | diff --git a/.travis.yml b/.travis.yml
--- a/.travis.yml
+++ b/.travis.yml
@@ -40,14 +40,22 @@ matrix:
script:
- rustc --version
- sysctl -a | grep mem
- - cargo install clippy || touch clippy_failed
+ - if [ "$TRAVIS_RUST_VERSION" == "nightly" ]; then
+ rustup component add clippy-preview && cargo clippy;
... | [
"160"
] | GuillaumeGomez__sysinfo-161 | GuillaumeGomez/sysinfo | diff --git a/Cargo.toml b/Cargo.toml
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -13,9 +13,9 @@ categories = ["filesystem", "os::macos-apis", "os::unix-apis", "os::windows-apis
build = "build.rs"
[dependencies]
-libc = "^0.2"
cfg-if = "0.1"
rayon = "^1.0"
+libc = "0.2"
[target.'cfg(windows)'.dependencies]
winapi = {... | 8cff91770cafe71f388cf8a798fc871509520c1e | Memory usage seems inaccurate
I'm writing a improved version of top, and when listing the memory usage of individual processes, I noticed that it doesn't line up with with what's reported by top or activity monitor. Is there any reason that this wouldn't be the case? I'm running MacOS Mojave.
]
mod tests {
- use ::{System, SystemExt};
+ use ::{ProcessExt, System, SystemExt};
use ::utils;
#[test]
diff --git a/src/system.rs b/src/system.rs
--- a/src/system.rs
+++ b/src/system.rs
... | [
"148"
] | GuillaumeGomez__sysinfo-150 | GuillaumeGomez/sysinfo | diff --git a/Cargo.toml b/Cargo.toml
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -1,6 +1,6 @@
[package]
name = "sysinfo"
-version = "0.6.2"
+version = "0.7.0"
authors = ["Guillaume Gomez <guillaume1.gomez@gmail.com>"]
description = "Library to handle processes"
diff --git a/src/windows/process.rs b/src/windows/process.r... | 95d0ba8a80c99f2c6ca67dd3801192dd15e4d51d | `Process` is does not implement `Send + Sync` on windows
Hey,
on Linux the `Process` struct implements `Send + Sync`, but not on Windows.
We have the following [bug report](https://github.com/paritytech/substrate/issues/1253).
After looking into the `Process` code for Windows, I think that `HANDLE` is the problem... | 0.6 | 150 | It's part of the problem, yes. However, `Process` cannot be used safely between threads. I'll make another check for linux but I think it shouldn't be `Send+Sync` either. I need it internally so I can perform faster computations but that's it.
To fix your issue, you'll have to wrap it inside a `Mutex` I guess. I'll ... | 95d0ba8a80c99f2c6ca67dd3801192dd15e4d51d |
2017-02-17T10:54:44Z | diff --git a/.travis.yml b/.travis.yml
--- a/.travis.yml
+++ b/.travis.yml
@@ -16,6 +16,7 @@ script:
- cargo install clippy || touch clippy_failed
- RUST_BACKTRACE=1 cargo build
- if [ ! -f clippy_failed ]; then cargo clippy; fi
+ - RUST_BACKTRACE=1 cargo test
- cd examples
- RUST_BACKTRACE=1 cargo buil... | [
"50"
] | GuillaumeGomez__sysinfo-51 | GuillaumeGomez/sysinfo | diff --git a/Cargo.toml b/Cargo.toml
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -1,6 +1,6 @@
[package]
name = "sysinfo"
-version = "0.2.7"
+version = "0.2.8"
authors = ["Guillaume Gomez <guillaume1.gomez@gmail.com>"]
description = "Library to handle processes"
diff --git a/src/linux/disk.rs b/src/linux/disk.rs
--- a/sr... | 37829411d4186264c24dfa96f15522655d5060bb | panic when get linux disk info
The test code is very simple:
```
extern crate sysinfo;
use sysinfo::{System, SystemExt};
fn main() {
let s = System::new();
println!("total memory: {}", s.get_total_memory());
println!("total cpu cores: {}", s.get_processor_list().len());
}
```
the panic backtr... | 0.2 | 51 | My os is ubuntu12.04
```
$ cat /proc/mounts | grep "/dev/sda"
/dev/sda6 / ext4 rw,relatime,errors=remount-ro,data=ordered 0 0
/dev/sda1 /boot ext4 rw,relatime,data=ordered 0 0
/dev/sda6 /var/lib/docker/devicemapper ext4 rw,relatime,errors=remount-ro,data=ordered 0 0
```
| 37829411d4186264c24dfa96f15522655d5060bb |
2022-02-02T20:15:17Z | diff --git a/tests/process.rs b/tests/process.rs
--- a/tests/process.rs
+++ b/tests/process.rs
@@ -254,7 +254,7 @@ fn test_process_times() {
std::process::Command::new("waitfor")
.arg("/t")
.arg("3")
- .arg("CwdSignal")
+ .arg("ProcessTimes")
.stdout... | [
"686"
] | GuillaumeGomez__sysinfo-687 | GuillaumeGomez/sysinfo | diff --git a/src/common.rs b/src/common.rs
--- a/src/common.rs
+++ b/src/common.rs
@@ -46,6 +46,9 @@ macro_rules! pid_decl {
Self(v)
}
}
+ #[allow(clippy::from_over_into)]
+ // This Into implementation is required otherwise it seems you can't do `pid.into()`
+ ... | 8a13566c75b44d51b7da3fc58901e863e72a9e26 | [Linux] Issue with process removal logic after calling `refresh_process()`
I'm observing an issue with the following scenario:
1) Call `refresh_processes()`
2) Call `refresh_process()` on a running process P
3) Kill the process P
4) Call `refresh_processes()`
5) Call `processes()` to get the list of running proc... | 0.23 | 687 | This is definitely a bug. I'll add a test with this scenario to ensure the killed process doesn't show up. | 6f1c70092ee030535cc2939960baf41ce331a4b2 |
2022-01-17T17:26:04Z | diff --git a/tests/process.rs b/tests/process.rs
--- a/tests/process.rs
+++ b/tests/process.rs
@@ -245,6 +245,8 @@ fn cpu_usage_is_not_nan() {
#[test]
fn test_process_times() {
+ use std::time::{SystemTime, UNIX_EPOCH};
+
if !sysinfo::System::IS_SUPPORTED || cfg!(feature = "apple-sandbox") {
return... | [
"680"
] | GuillaumeGomez__sysinfo-681 | GuillaumeGomez/sysinfo | diff --git a/src/linux/process.rs b/src/linux/process.rs
--- a/src/linux/process.rs
+++ b/src/linux/process.rs
@@ -10,7 +10,7 @@ use std::str::FromStr;
use libc::{gid_t, kill, uid_t};
-use crate::sys::system::REMAINING_FILES;
+use crate::sys::system::{SystemInfo, REMAINING_FILES};
use crate::sys::utils::{get_all_... | c1a6c6f2c438d7cf4c947f555df3d6d408ef0fe4 | On Linux `ProcessExt::start_time()` does not return the time since the epoch as documented
It seems to just return `(22) starttime` from `/proc/[pid]/stat`, see `man proc`: "The time the process started after system boot."
`info.process(get_current_pid().unwrap()).unwrap().start_time()` is much smaller than the expe... | 0.22 | 681 | It's pretty bad in any case. Interested into sending a PR? Otherwise I'll try to fix it as soon as possible. | c1a6c6f2c438d7cf4c947f555df3d6d408ef0fe4 |
2022-01-14T22:08:03Z | diff --git /dev/null b/tests/code_checkers/docs.rs
new file mode 100644
--- /dev/null
+++ b/tests/code_checkers/docs.rs
@@ -0,0 +1,135 @@
+// Take a look at the license at the top of the repository in the LICENSE file.
+
+use super::utils::{show_error, TestResult};
+use std::ffi::OsStr;
+use std::path::Path;
+
+fn to_c... | [
"592"
] | GuillaumeGomez__sysinfo-679 | GuillaumeGomez/sysinfo | diff --git a/src/apple/macos/component.rs b/src/apple/macos/component.rs
--- a/src/apple/macos/component.rs
+++ b/src/apple/macos/component.rs
@@ -18,7 +18,7 @@ pub(crate) const COMPONENTS_TEMPERATURE_IDS: &[(&str, &[i8])] = &[
("Battery", &['T' as i8, 'B' as i8, '0' as i8, 'T' as i8]), // Battery "TB0T"
];
-p... | b1d66813b7a5f35832302e5ed3d4d85ab94e9464 | Add check to ensure that types are using common md files and not manual doc comments
For example `System` or `Process`.
| 0.22 | 679 | c1a6c6f2c438d7cf4c947f555df3d6d408ef0fe4 | |
2022-01-14T16:21:58Z | diff --git a/src/lib.rs b/src/lib.rs
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -128,6 +128,52 @@ pub fn set_open_files_limit(mut _new_limit: isize) -> bool {
}
}
+// FIXME: Can be removed once negative trait bounds are supported.
+#[cfg(doctest)]
+mod doctest {
+ /// Check that `Process` doesn't implement `Clone`... | [
"609"
] | GuillaumeGomez__sysinfo-678 | GuillaumeGomez/sysinfo | diff --git a/src/apple/app_store/process.rs b/src/apple/app_store/process.rs
--- a/src/apple/app_store/process.rs
+++ b/src/apple/app_store/process.rs
@@ -5,7 +5,6 @@ use std::path::Path;
use crate::{DiskUsage, Pid, ProcessExt, ProcessStatus, Signal};
#[doc = include_str!("../../../md_doc/process.md")]
-#[derive(Cl... | 46333182d9a86be3b27adb2e6f34cd75f3620f9c | Ensure that `Clone` isn't implemented on `Process`
Can be done by adding a `compile_fail` test in a doc comment.
| 0.22 | 678 | c1a6c6f2c438d7cf4c947f555df3d6d408ef0fe4 | |
2021-12-08T11:08:48Z | diff --git /dev/null b/.cirrus.yml
new file mode 100644
--- /dev/null
+++ b/.cirrus.yml
@@ -0,0 +1,35 @@
+task:
+ name: rust 1.54 on freebsd 11
+ freebsd_instance:
+ image: freebsd-11-4-release-amd64
+ setup_script:
+ - pkg install -y curl
+ - curl https://sh.rustup.rs -sSf --output rustup.sh
+ - sh rust... | [
"622"
] | GuillaumeGomez__sysinfo-627 | GuillaumeGomez/sysinfo | diff --git a/CHANGELOG.md b/CHANGELOG.md
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,3 +1,7 @@
+# 0.21.2
+
+ * Unsupported targets: fix build.
+
# 0.21.1
* Linux: Process CPU usage cannot go above maximum value (number of CPUs * 100) anymore.
diff --git a/Cargo.toml b/Cargo.toml
--- a/Cargo.toml
+++ b/Cargo.toml
@... | 3e6d919394fd281ffd0e8484cd9020fa0462dc9d | Taking rootfs as a disk by mistake
In some kinds of Linux systems, there is a virtual type of filesystem named `rootfs`. However, sysinfo cannot recognize it correctly and takes it as the real disk.
This is the version of centos:
```
[VM-9-75-centos ~]$ cat /proc/version
Linux version 3.10.0-1062.18.1.el7.x86... | 0.21 | 627 | 3e6d919394fd281ffd0e8484cd9020fa0462dc9d | |
2021-11-23T19:49:11Z | diff --git a/.cirrus.yml b/.cirrus.yml
--- a/.cirrus.yml
+++ b/.cirrus.yml
@@ -12,14 +12,14 @@ task:
test_script:
- . $HOME/.cargo/env
- cargo check
- - cargo check --example simple
- - cargo test
- cargo clippy -- -D warnings
+ - cargo check --example simple
+ - FREEBSD_CI=1 cargo test -j... | [
"433"
] | GuillaumeGomez__sysinfo-620 | GuillaumeGomez/sysinfo | diff --git a/.cirrus.yml b/.cirrus.yml
--- a/.cirrus.yml
+++ b/.cirrus.yml
@@ -1,7 +1,7 @@
task:
- name: rust 1.54 on freebsd 11
+ name: rust 1.54 on freebsd 13
freebsd_instance:
- image: freebsd-11-4-release-amd64
+ image: freebsd-13-0-release-amd64
setup_script:
- pkg install -y curl
- curl h... | a2c2a7779336384b726a73567002de19574ec843 | Add FreeBSD support?
As FreeBSD is still used on lots of places it might be nice to support it.
I don't think you want to add this but are you open to PRs to support FreeBSD?
| 0.21 | 620 | I gave it a try a long time ago but had issues finding some information. If you feel like giving it a try, you're very welcome to do so!
Ah that's a bummer, i've always been interested in freebsd but never found any reason to use it so this might be a nice way to learn some new things.
Good luck then. :wink:
A gener... | 3e6d919394fd281ffd0e8484cd9020fa0462dc9d |
2021-06-28T13:42:01Z | diff --git a/src/lib.rs b/src/lib.rs
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -341,10 +341,11 @@ mod test {
if System::IS_SUPPORTED {
assert!(!s.processors().is_empty());
+ // In case we are running inside a VM, it's possible to not have a physical core, only
+ // logical ones,... | [
"519"
] | GuillaumeGomez__sysinfo-529 | GuillaumeGomez/sysinfo | diff --git a/src/linux/processor.rs b/src/linux/processor.rs
--- a/src/linux/processor.rs
+++ b/src/linux/processor.rs
@@ -263,10 +263,8 @@ pub fn get_cpu_frequency(cpu_core_index: usize) -> u64 {
pub fn get_physical_core_count() -> Option<usize> {
let mut s = String::new();
- if File::open("/proc/cpuinfo")
... | 3aa83dc410ebe2f6d5063a095104ea6c6394507f | test_physical_core_numbers fails on armv7hl aarch64 and ppc64le
```
---- test_physical_core_numbers stdout ----
thread 'test_physical_core_numbers' panicked at 'assertion failed: count.unwrap() > 0', tests/processor.rs:36:9
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
failures:
... | 0.18 | 529 | For this one, please show me the content of `/proc/cpuinfo`. Also, why are you running the tests to build a package?
Fedora policy is to always run the tests at the end of the build to validate that the package will actually work. This isn't specific to rust, it's done for all packages that have tests available.
arm... | 3aa83dc410ebe2f6d5063a095104ea6c6394507f |
2021-06-11T08:30:22Z | diff --git a/src/system.rs b/src/system.rs
--- a/src/system.rs
+++ b/src/system.rs
@@ -89,4 +89,15 @@ mod tests {
assert!(!hostname.contains('\u{0}'))
}
}
+
+ #[test]
+ fn check_uptime() {
+ let sys = System::new();
+ let uptime = sys.get_uptime();
+ if System::IS_S... | [
"508"
] | GuillaumeGomez__sysinfo-509 | GuillaumeGomez/sysinfo | diff --git a/src/linux/system.rs b/src/linux/system.rs
--- a/src/linux/system.rs
+++ b/src/linux/system.rs
@@ -121,7 +121,6 @@ pub struct System {
components: Vec<Component>,
disks: Vec<Disk>,
networks: Networks,
- uptime: u64,
users: Vec<User>,
boot_time: u64,
}
diff --git a/src/linux/syst... | 5c69175f028a5608093d31597b93031d165774c5 | Uptime: To cache, or not to cache?
I've been looking into FreeBSD support this week (#433), and while trying to implement `SystemExt::get_uptime`, I found an inconsistency.
Linux uses whatever value was cached during the most recent refresh.
https://github.com/GuillaumeGomez/sysinfo/blob/f8574e459d0b1d65ed649cd08... | 0.18 | 509 | Linux should be updated to look like the others. Gonna send a PR which will also enforce this behaviour with a test (and thanks for working on freebsd!). | 3aa83dc410ebe2f6d5063a095104ea6c6394507f |
2021-06-06T19:49:13Z | diff --git a/src/lib.rs b/src/lib.rs
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -343,8 +343,14 @@ mod test {
if System::IS_SUPPORTED {
assert!(!s.get_processors().is_empty());
+ assert!(
+ s.get_physical_core_count()
+ .expect("failed to get number of physi... | [
"502"
] | GuillaumeGomez__sysinfo-503 | GuillaumeGomez/sysinfo | diff --git a/examples/src/simple.rs b/examples/src/simple.rs
--- a/examples/src/simple.rs
+++ b/examples/src/simple.rs
@@ -166,6 +166,13 @@ fn interpret_input(input: &str, sys: &mut System) -> bool {
"processors" => {
// Note: you should refresh a few times before using this, so that usage statist... | 05427d4e26b83be87f64865b3c15c9663e5d2e62 | Issue getting processor counts on Windows
I was wondering why the physical number of CPUs would show up greater than the number of CPU cores on Windows 10?
https://github.com/mikemadden42/inv/blob/master/src/main.rs#L33
```
# Window 10 21H1
Number of cores: 2
Number of physical processors: 3
```
```
Get-... | 0.18 | 503 | I get that information there: https://github.com/GuillaumeGomez/sysinfo/blob/master/src/windows/tools.rs#L44-L61. Surprisingly, I initiate the vec with "nb processors + 1" but I don't put an extra one into it. This is weird...
My bad, the physical core count is actually there: https://github.com/GuillaumeGomez/sysinfo/... | 3aa83dc410ebe2f6d5063a095104ea6c6394507f |
2021-05-12T19:50:07Z | diff --git a/src/linux/network.rs b/src/linux/network.rs
--- a/src/linux/network.rs
+++ b/src/linux/network.rs
@@ -248,3 +274,51 @@ impl NetworkExt for NetworkData {
self.tx_errors
}
}
+
+#[cfg(test)]
+mod test {
+ use super::refresh_networks_list_from_sysfs;
+ use std::collections::HashMap;
+ u... | [
"479"
] | GuillaumeGomez__sysinfo-481 | GuillaumeGomez/sysinfo | diff --git a/src/linux/network.rs b/src/linux/network.rs
--- a/src/linux/network.rs
+++ b/src/linux/network.rs
@@ -9,7 +9,7 @@ use std::io::Read;
use std::path::Path;
use crate::{NetworkExt, NetworksExt, NetworksIter};
-use std::collections::HashMap;
+use std::collections::{hash_map, HashMap};
/// Network interf... | 190926a97e50c2abf1b155ea59abba1c23f4ebf0 | [Linux] get_networks keeps returning removed interfaces
Using `get_networks` with calls to `refresh_networks_list`, new network interfaces are picked up, however they are still returned when they are removed from the system.
Small program to reproduce:
```
use std::thread::sleep;
use std::time::Duration;
use ... | 0.17 | 481 | I only gave a quick look, but I think the `self.interface` HashMap needs to be cleared in the beginning of `refresh_networks_list` https://github.com/GuillaumeGomez/sysinfo/blob/master/src/linux/network.rs#L76
Thanks for the issue! The simpler fix here would be to do it in two passes: we first we check if all our netwo... | 190926a97e50c2abf1b155ea59abba1c23f4ebf0 |
2021-03-17T13:54:47Z | diff --git a/src/lib.rs b/src/lib.rs
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -208,6 +208,29 @@ mod test {
assert!(s.get_users().len() >= MIN_USERS);
}
+ #[test]
+ fn check_uid_gid() {
+ let mut s = System::new();
+ assert!(s.get_users().is_empty());
+ s.refresh_users_list();
+ ... | [
"167"
] | GuillaumeGomez__sysinfo-449 | GuillaumeGomez/sysinfo | diff --git a/src/apple/users.rs b/src/apple/users.rs
--- a/src/apple/users.rs
+++ b/src/apple/users.rs
@@ -4,7 +4,10 @@
// Copyright (c) 2020 Guillaume Gomez
//
-use crate::User;
+use crate::{
+ common::{Gid, Uid},
+ User,
+};
use crate::sys::utils;
use libc::{c_char, endpwent, getgrgid, getgrouplist, get... | 58b31e3a5d2821fb0483125220331b3afab5cd11 | Add user id and group id
| 0.16 | 449 | 58b31e3a5d2821fb0483125220331b3afab5cd11 | |
2021-01-26T15:07:08Z | diff --git a/src/sysinfo.rs b/src/sysinfo.rs
--- a/src/sysinfo.rs
+++ b/src/sysinfo.rs
@@ -185,11 +185,11 @@ pub fn set_open_files_limit(mut _new_limit: isize) -> bool {
#[cfg(test)]
mod test {
- use super::*;
+ use crate::*;
#[test]
fn check_memory_usage() {
- let mut s = ::System::new();
... | [
"403"
] | GuillaumeGomez__sysinfo-417 | GuillaumeGomez/sysinfo | diff --git a/Cargo.toml b/Cargo.toml
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -11,6 +11,7 @@ readme = "README.md"
categories = ["filesystem", "os", "api-bindings"]
build = "build.rs"
+edition = "2018"
[dependencies]
cfg-if = "1.0"
diff --git a/src/apple/component.rs b/src/apple/component.rs
--- a/src/apple/componen... | 8c2b5a0583404120f1a910d9af32f40fd1dd9d08 | Switch to 2018 edition
| 0.15 | 417 | 8c2b5a0583404120f1a910d9af32f40fd1dd9d08 | |
2021-01-25T20:57:53Z | diff --git a/src/sysinfo.rs b/src/sysinfo.rs
--- a/src/sysinfo.rs
+++ b/src/sysinfo.rs
@@ -222,8 +223,18 @@ mod test {
if MIN_USERS > 0 {
let s = System::new();
assert!(!s.get_name().expect("Failed to get system name").is_empty());
+
+ cfg_if! {
+ if #[cfg(no... | [
"413"
] | GuillaumeGomez__sysinfo-414 | GuillaumeGomez/sysinfo | diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -91,9 +91,10 @@ for (pid, process) in sys.get_processes() {
}
// Display system information:
-println!("System name: {:?}", sys.get_name());
-println!("System version: {:?}", sys.get_version());
-println!("System host name: {:?}", sys.get_... | 639f6afc6dc8808572853b721fa634eefd06b489 | `get_version()` returns kernel version on MacOS and iOS
`get_verison()` will return the compiled kernel version of MacOS and not the "product" version of MacOS/iOS. This is a problem because while the kernel version is still needed we currently have no way to get the more recognizable system version for system informa... | 0.15 | 414 | 8c2b5a0583404120f1a910d9af32f40fd1dd9d08 | |
2021-01-09T18:09:21Z | diff --git a/src/sysinfo.rs b/src/sysinfo.rs
--- a/src/sysinfo.rs
+++ b/src/sysinfo.rs
@@ -97,6 +97,9 @@ cfg_if! {
} else if #[cfg(any(target_os = "linux", target_os = "android"))] {
mod linux;
use linux as sys;
+ // Can remove once `slice_internals` is stabilized
+ // https://doc.r... | [
"390"
] | GuillaumeGomez__sysinfo-393 | GuillaumeGomez/sysinfo | diff --git a/Cargo.toml b/Cargo.toml
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -31,6 +31,9 @@ core-foundation-sys = "0.8"
[target.'cfg(any(target_os = "macos", target_os = "ios"))'.build-dependencies]
cc = "1.0"
+[target.'cfg(any(target_os = "linux", target_os = "android"))'.dependencies]
+memchr = "2.3"
+
[lib]
name ... | d162ddd4e2f5ebf1139779b3c57304b3542bbc7f | Add `hostname` support
It would be *really* nice for this crate to support getting a system's hostname to get the network name of the host.
e.g. equivalent to `/bin/hostname`'s output.
Sorry if I missed this somewhere in the docs!
| 0.15 | 393 | `sysinfo` doesn't provide this information yet. From what I can see, on unix-like systems, it just needs to call the `gethostname` function. On windows it's [getcomputernamea](https://docs.microsoft.com/fr-fr/windows/win32/api/winbase/nf-winbase-getcomputernamea). Interested into sending a PR maybe? | 8c2b5a0583404120f1a910d9af32f40fd1dd9d08 |
2024-12-04T00:06:06Z | diff --git a/benches/basic.rs b/benches/basic.rs
--- a/benches/basic.rs
+++ b/benches/basic.rs
@@ -127,6 +127,6 @@ fn bench_refresh_users_list(b: &mut test::Bencher) {
let mut users = sysinfo::Users::new_with_refreshed_list();
b.iter(move || {
- users.refresh_list();
+ users.refresh();
})... | [
"1405"
] | GuillaumeGomez__sysinfo-1411 | GuillaumeGomez/sysinfo | diff --git a/examples/simple.rs b/examples/simple.rs
--- a/examples/simple.rs
+++ b/examples/simple.rs
@@ -163,7 +163,7 @@ fn interpret_input(
}
"refresh_users" => {
writeln!(&mut io::stdout(), "Refreshing user list...");
- users.refresh_list();
+ users.refresh();
... | ee9d6da5bb7794be56acd100406fd1b0200bced5 | Remove `refresh_list` methods
It's not very coherent with how processes are refreshed. And now with https://github.com/GuillaumeGomez/sysinfo/pull/1404, it makes even less sense to have that. Better just refresh everything and keep the current items.
| 0.32 | 1,411 | ee9d6da5bb7794be56acd100406fd1b0200bced5 | |
2024-11-22T21:12:44Z | diff --git a/Cargo.toml b/Cargo.toml
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -115,6 +115,7 @@ tempfile = "3.9"
serde_json = "1.0" # Used in documentation tests.
bstr = "1.9.0"
tempfile = "3.9"
+itertools = "0.13.0"
[[example]]
name = "simple"
diff --git a/tests/disk.rs b/tests/disk.rs
--- a/tests/disk.rs
+++ b/test... | [
"1375"
] | GuillaumeGomez__sysinfo-1387 | GuillaumeGomez/sysinfo | diff --git a/Cargo.toml b/Cargo.toml
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -103,7 +103,7 @@ ntapi = { version = "0.4", optional = true }
windows = { version = ">=0.54, <=0.57", optional = true }
[target.'cfg(not(any(target_os = "unknown", target_arch = "wasm32")))'.dependencies]
-libc = "^0.2.165"
+libc = "^0.2.164"... | 5721f848f7e9b5d883e4ff7f88bd0b297d4ee05c | Selectively refreshing disks
When running `Disks::new_with_refreshed_list()` on android devices, I often see warnings (from SELinux?) along the lines of:
```
type=1400 audit(0.0:339351): avc: denied { search } for name="block" dev="tmpfs" ino=12
```
I've done some spelunking, and it appears they are a side-e... | 0.32 | 1,387 | Seems a bit weird to want to have disks but not their path or even their kind. I think the best course would be to instead detect when the `canonicalize` call is actually needed and only call it if so.
> Seems a bit weird to want to have disks but not their path or even their kind.
I was thinking those *would* be pr... | ee9d6da5bb7794be56acd100406fd1b0200bced5 |
2024-10-07T21:44:33Z | diff --git a/src/common/system.rs b/src/common/system.rs
--- a/src/common/system.rs
+++ b/src/common/system.rs
@@ -2462,7 +2462,7 @@ mod test {
#[test]
fn check_cpu_arch() {
- assert_eq!(System::cpu_arch().is_some(), IS_SUPPORTED_SYSTEM);
+ assert!(!System::cpu_arch().is_empty());
}
... | [
"1356"
] | GuillaumeGomez__sysinfo-1358 | GuillaumeGomez/sysinfo | diff --git a/src/common/system.rs b/src/common/system.rs
--- a/src/common/system.rs
+++ b/src/common/system.rs
@@ -777,8 +777,8 @@ impl System {
///
/// println!("CPU Architecture: {:?}", System::cpu_arch());
/// ```
- pub fn cpu_arch() -> Option<String> {
- SystemInner::cpu_arch()
+ pub fn ... | e022ae4fd1d27c2d7159cad3e7018fc08e5b822d | Use std::env::consts::ARCH for System::cpu_arch
The `std::env::consts::ARCH` constant can be used instead of computing the value for every call to the function.
| 0.32 | 1,358 | Interested into sending a PR?
> Interested into sending a PR?
Should the function fallback to `std::env::consts::ARCH` if `SystemInner::cpu_arch` returns `None`?
Sounds good to me! | ee9d6da5bb7794be56acd100406fd1b0200bced5 |
2024-10-03T21:06:38Z | diff --git a/benches/basic.rs b/benches/basic.rs
--- a/benches/basic.rs
+++ b/benches/basic.rs
@@ -33,9 +33,9 @@ fn bench_refresh_all(b: &mut test::Bencher) {
fn bench_refresh_processes(b: &mut test::Bencher) {
let mut s = sysinfo::System::new();
- s.refresh_processes(sysinfo::ProcessesToUpdate::All); // to ... | [
"1340"
] | GuillaumeGomez__sysinfo-1353 | GuillaumeGomez/sysinfo | diff --git a/examples/simple.rs b/examples/simple.rs
--- a/examples/simple.rs
+++ b/examples/simple.rs
@@ -408,7 +408,7 @@ fn interpret_input(
.take(1)
.next()
{
- if sys.refresh_processes(sysinfo::ProcessesToUpdate::Some(&[pid])) != 0 {
+ ... | 6f1d3822765b3ba3b9fe774d8d2e3798c31fce5d | Add some way to remove a closed process
I'm attaching to specific individual processes and want to know when they are closed, so I use
```rust
if system.refresh_processes_specifics(ProcessesToUpdate::Some(&[pid]), ProcessRefreshKind::new()) == 0 {
...
}
```
to check for that situation. Unfortunately as state... | 0.31 | 1,353 | Interesting: I was thinking about adding a new `remove_dead_processes` boolean field in `ProcessProcessKind` or as new parameter. Not clear yet. I suppose it would fix your case?
Yeah absolutely, as long as it doesn't wastefully have to scan the entire process list or so.
Sadly it kinda still needs to iterate processes... | 6f1d3822765b3ba3b9fe774d8d2e3798c31fce5d |
2024-07-24T20:46:11Z | diff --git a/tests/process.rs b/tests/process.rs
--- a/tests/process.rs
+++ b/tests/process.rs
@@ -813,3 +813,47 @@ fn test_parent_change() {
// We kill the child to clean up.
child.kill();
}
+
+// We want to ensure that if `System::refresh_process*` methods are called
+// one after the other, it won't impac... | [
"1299"
] | GuillaumeGomez__sysinfo-1324 | GuillaumeGomez/sysinfo | diff --git a/src/common/system.rs b/src/common/system.rs
--- a/src/common/system.rs
+++ b/src/common/system.rs
@@ -367,6 +367,9 @@ impl System {
/// exist (it will **NOT** be removed from the processes if it doesn't exist anymore). If it
/// isn't listed yet, it'll be added.
///
+ /// ⚠️ If you need t... | 296bbf02359b11eef28fc344832760b654215a22 | Refreshing multiple processes on macos leads to invalid cpu usage
I have a piece of code that looks like this:
```rust
let mut sys = sysinfo::System::new();
let process_refresh_kind = ProcessRefreshKind::everything().without_disk_usage().without_environ();
sys.refresh_process_specifics(pid_a, process_refresh_kind... | 0.30 | 1,324 | For this kind of operations, you likely want to use [refresh_pids_specifics](https://docs.rs/sysinfo/latest/sysinfo/struct.System.html#method.refresh_pids_specifics). Although, in this case it should still work.
Indeed, I didn't know about this API, and it fixes the issue. I'm leaving this issue open since there is sti... | 296bbf02359b11eef28fc344832760b654215a22 |
2024-04-10T14:17:27Z | diff --git /dev/null b/tests/components.rs
new file mode 100644
--- /dev/null
+++ b/tests/components.rs
@@ -0,0 +1,17 @@
+// Take a look at the license at the top of the repository in the LICENSE file.
+
+#[test]
+fn test_components() {
+ use std::env::var;
+
+ let mut c = sysinfo::Components::new();
+ assert!... | [
"1245"
] | GuillaumeGomez__sysinfo-1249 | GuillaumeGomez/sysinfo | diff --git a/src/unix/linux/component.rs b/src/unix/linux/component.rs
--- a/src/unix/linux/component.rs
+++ b/src/unix/linux/component.rs
@@ -374,16 +374,15 @@ impl ComponentsInner {
continue;
};
let entry = entry.path();
- if !file_type.is_dir()
- ... | 93f9b823c1a5cd56f271e48cf879ede312914c64 | Temperatures on Linux unavailable
A commit on April 7, 2024 causes temperatures on Linux to not be available. The `temperature` command in `examples/simple.rs` no longer prints anything on commit [87c627c1ff07c6a43060c397bcb401e5d72d33e0](https://github.com/GuillaumeGomez/sysinfo/commit/87c627c1ff07c6a43060c397bcb401e... | 0.30 | 1,249 | 296bbf02359b11eef28fc344832760b654215a22 | |
2024-01-04T18:06:14Z | diff --git a/.cirrus.yml b/.cirrus.yml
--- a/.cirrus.yml
+++ b/.cirrus.yml
@@ -37,14 +37,14 @@ task:
- FREEBSD_CI=1 cargo test --lib -j1 -- --ignored
task:
- name: rust 1.70 on mac m1
+ name: rust 1.74 on mac m1
macos_instance:
image: ghcr.io/cirruslabs/macos-monterey-base:latest
setup_script:
... | [
"1190"
] | GuillaumeGomez__sysinfo-1196 | GuillaumeGomez/sysinfo | diff --git a/.cirrus.yml b/.cirrus.yml
--- a/.cirrus.yml
+++ b/.cirrus.yml
@@ -1,10 +1,10 @@
task:
- name: rust 1.70 on freebsd 13
+ name: rust 1.74 on freebsd 13
freebsd_instance:
image: freebsd-13-1-release-amd64
setup_script:
- curl https://sh.rustup.rs -sSf --output rustup.sh
- - sh rustup.sh ... | 622cd8a1b3a60983c15273e82bb9fdf9e9fb6177 | Process List incomplete on long names
**Describe the bug**
This is on 0.30.3. The issue is that the stat file on Linux is parsed under the assumption that it contains valid UTF-8. This is not necessarily the case for two reasons:
1. Process names don't need to be valid UTF-8 at all on Linux. They can contain invalid ... | 0.30 | 1,196 | Thanks for the detailed description! Using `OsStr` is the best solution here I think.
I see that you can get a process/thread name with `pthread_getname_np` or by reading the proc `comm` file. Not sure if it's worth it though. | 296bbf02359b11eef28fc344832760b654215a22 |
2023-12-17T22:57:54Z | diff --git a/src/lib.rs b/src/lib.rs
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -365,34 +365,33 @@ mod test {
#[test]
fn check_system_info() {
- let s = System::new();
-
// We don't want to test on unsupported systems.
if IS_SUPPORTED {
- assert!(!s.name().expect("Failed to get... | [
"1138"
] | GuillaumeGomez__sysinfo-1171 | GuillaumeGomez/sysinfo | diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -54,10 +54,10 @@ println!("total swap : {} bytes", sys.total_swap());
println!("used swap : {} bytes", sys.used_swap());
// Display system information:
-println!("System name: {:?}", sys.name());
-println!("System kernel version: ... | 3cc9cc814d21b7977a0e7bb8260a7b64529e6a1b | Extract "standalone" methods from `System`
If they don't need information stored in `System`, they shouldn't be a method of it.
| 0.29 | 1,171 | 3cc9cc814d21b7977a0e7bb8260a7b64529e6a1b | |
2023-12-17T17:21:49Z | diff --git a/tests/process.rs b/tests/process.rs
--- a/tests/process.rs
+++ b/tests/process.rs
@@ -474,7 +474,7 @@ fn test_wait_child() {
std::process::Command::new("waitfor")
.arg("/t")
.arg("300")
- .arg("RefreshProcess")
+ .arg("WaitChild")
.stdou... | [
"1133"
] | GuillaumeGomez__sysinfo-1170 | GuillaumeGomez/sysinfo | diff --git a/src/common.rs b/src/common.rs
--- a/src/common.rs
+++ b/src/common.rs
@@ -265,7 +265,65 @@ impl System {
/// s.refresh_processes_specifics(ProcessRefreshKind::new());
/// ```
pub fn refresh_processes_specifics(&mut self, refresh_kind: ProcessRefreshKind) {
- self.inner.refresh_process... | 05d5c2cbfe823a792333841547d5c251db99fb1b | Allow to refresh multiple PIDs at once (so it can be multithreaded)
Adding methods called `refresh_pids` and `refresh_pids_specifics`.
| 0.29 | 1,170 | I took a look at this and on linux, I need to figure out if the `/proc/[PID]/task/` reading is really necessary. I couldn't find a reason why it was actually useful but I'll continue to investigate.
More explanations about the `task` folder [here](https://github.com/htop-dev/htop/issues/1341).
Not a problem, the tasks ... | 3cc9cc814d21b7977a0e7bb8260a7b64529e6a1b |
2023-11-29T16:35:14Z | diff --git a/tests/process.rs b/tests/process.rs
--- a/tests/process.rs
+++ b/tests/process.rs
@@ -34,7 +34,7 @@ fn test_cwd() {
if let Some(p) = p {
assert_eq!(p.pid(), pid);
- assert_eq!(p.cwd(), std::env::current_dir().unwrap());
+ assert_eq!(p.cwd().unwrap(), &std::env::current_dir().u... | [
"1137"
] | GuillaumeGomez__sysinfo-1162 | GuillaumeGomez/sysinfo | diff --git a/src/c_interface.rs b/src/c_interface.rs
--- a/src/c_interface.rs
+++ b/src/c_interface.rs
@@ -462,7 +462,7 @@ pub extern "C" fn sysinfo_process_executable_path(process: CProcess) -> RString
assert!(!process.is_null());
let process = process as *const Process;
unsafe {
- if let Some(p)... | 8737fa06afbf879062551d77ed60b0e10c250464 | Make `Process::root` return an `Option`
It can be a non-available information and as such, should return an `Option` rather than an empty `Path`. (For example it's empty on mac)
| 0.29 | 1,162 | 3cc9cc814d21b7977a0e7bb8260a7b64529e6a1b | |
2023-11-29T15:55:10Z | diff --git a/tests/process.rs b/tests/process.rs
--- a/tests/process.rs
+++ b/tests/process.rs
@@ -2,28 +2,6 @@
use sysinfo::{Pid, ProcessRefreshKind, System, UpdateKind};
-#[test]
-fn test_process() {
- let mut s = System::new();
- assert_eq!(s.processes().len(), 0);
- s.refresh_processes();
- if !sys... | [
"1141"
] | GuillaumeGomez__sysinfo-1161 | GuillaumeGomez/sysinfo | diff --git a/src/common.rs b/src/common.rs
--- a/src/common.rs
+++ b/src/common.rs
@@ -221,13 +221,14 @@ impl System {
/// It does the same as:
///
/// ```no_run
- /// # use sysinfo::{ProcessRefreshKind, System};
+ /// # use sysinfo::{ProcessRefreshKind, System, UpdateKind};
/// # let mut syst... | 9fb48e65376a6b479e8f134d38c6b38068436804 | Should `exe` be included in default process retrieval (and removed from `ProcessRefreshKind`)?
It seems to be a basic information that everyone might want all the time.
| 0.29 | 1,161 | Another solution would be to just remove `refresh_process[es]`. | 3cc9cc814d21b7977a0e7bb8260a7b64529e6a1b |
2023-10-14T21:57:41Z | diff --git a/src/unix/apple/macos/process.rs b/src/unix/apple/macos/process.rs
--- a/src/unix/apple/macos/process.rs
+++ b/src/unix/apple/macos/process.rs
@@ -699,19 +687,3 @@ fn parse_command_line<T: Deref<Target = str> + Borrow<str>>(cmd: &[T]) -> Vec<St
}
command
}
-
-#[cfg(test)]
-mod test {
- use sup... | [
"1094"
] | GuillaumeGomez__sysinfo-1097 | GuillaumeGomez/sysinfo | diff --git a/src/unix/apple/macos/process.rs b/src/unix/apple/macos/process.rs
--- a/src/unix/apple/macos/process.rs
+++ b/src/unix/apple/macos/process.rs
@@ -1,6 +1,5 @@
// Take a look at the license at the top of the repository in the LICENSE file.
-use std::ffi::CStr;
use std::mem::{self, MaybeUninit};
use std:... | 12bb072a2dca91e271b3da45008bb162d258b02b | macos process.root() is returning the PATH env var
**Describe the bug**
On MacOS in nushell, I'm writing a small process information command to inspect nushell memory on demand. The code is working great on Windows but on MacOS, process.root() is returning the system PATH instead of the root folder where the executabl... | 0.29 | 1,097 | What is:
> the system PATH
? Shod me the path you expect and the one you get, might be better to understand for me.
Sorry for not being clearer. The system PATH is probably mischaracterized but it's supposed to mean the environment variable named PATH on unix type systems.
In this image you can see the path I ... | 3cc9cc814d21b7977a0e7bb8260a7b64529e6a1b |
2023-09-29T14:40:58Z | diff --git a/tests/process.rs b/tests/process.rs
--- a/tests/process.rs
+++ b/tests/process.rs
@@ -659,3 +659,23 @@ fn test_process_creds() {
true
}));
}
+
+// Regression test for <https://github.com/GuillaumeGomez/sysinfo/issues/1084>
+#[test]
+fn test_process_memory_refresh() {
+ if !sysinfo::IS_SUP... | [
"1084"
] | GuillaumeGomez__sysinfo-1085 | GuillaumeGomez/sysinfo | diff --git a/src/windows/process.rs b/src/windows/process.rs
--- a/src/windows/process.rs
+++ b/src/windows/process.rs
@@ -489,6 +489,7 @@ impl ProcessInner {
if refresh_kind.disk_usage() {
update_disk_usage(self);
}
+ update_memory(self);
self.run_time = now.saturating_su... | a728002ac8bc33f0d868776ad4044b2931d3a7b2 | process memory/virtual_memory on windows is only set on second refresh
Tested on sysinfo master (f6c1a25aa22ff323087b5e35c5014d1d81f7fe7d).
Here is a code showing the issue:
```rust
use sysinfo::{Pid, PidExt, ProcessExt, ProcessRefreshKind, System, SystemExt};
fn main() {
let mut args = std::env::args();... | 0.29 | 1,085 | 3cc9cc814d21b7977a0e7bb8260a7b64529e6a1b | |
2023-09-14T20:59:50Z | diff --git a/src/lib.rs b/src/lib.rs
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -22,52 +22,24 @@ cfg_if::cfg_if! {
#[cfg(test)]
pub(crate) const MIN_USERS: usize = 0;
- } else if #[cfg(any(target_os = "macos", target_os = "ios"))] {
- mod apple;
- use apple as sys;
- pub(crate) mod... | [
"1038"
] | GuillaumeGomez__sysinfo-1062 | GuillaumeGomez/sysinfo | diff --git a/src/macros.rs b/src/macros.rs
--- a/src/macros.rs
+++ b/src/macros.rs
@@ -60,7 +60,7 @@ macro_rules! declare_signals {
#[cfg(all(unix, not(feature = "unknown-ci")))]
macro_rules! retry_eintr {
(set_to_0 => $($t:tt)+) => {{
- let errno = crate::libc_errno();
+ let errno = crate::unix::l... | ffd95c1a84cee358f3a73411f38599357d323619 | Move unix-like targets into a unix sub-folder to have better folder hierarchy
FreeBSD, linux and macOS should be moved into it alongside `src/users.rs` and `src/network_helper_nix.rs`.
| 0.29 | 1,062 | 3cc9cc814d21b7977a0e7bb8260a7b64529e6a1b | |
2023-07-21T16:30:34Z | diff --git a/src/lib.rs b/src/lib.rs
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -275,7 +275,7 @@ mod test {
.all(|(_, proc_)| proc_.cpu_usage() == 0.0));
// Wait a bit to update CPU usage values
- std::thread::sleep(std::time::Duration::from_millis(100));
+ std::thread::sleep(System::MIN... | [
"881"
] | GuillaumeGomez__sysinfo-1017 | GuillaumeGomez/sysinfo | diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -97,9 +97,9 @@ loop {
for cpu in sys.cpus() {
print!("{}% ", cpu.cpu_usage());
}
- // Sleeping for 500 ms to let time for the system to run for long
+ // Sleeping to let time for the system to run for long
// enough to h... | ab2d268ca4bad60be367a4893161b520955aae4a | Process data inconsistency between `refresh_processes_specifics` and `refresh_process_specifics`
Thanks for all your work on sysinfo!
I have found a way to get inconsistent data (or at least inconsistent process names) regarding a given process:
* `system.refresh_processes_specifics()` (without any PID given) uses ... | 0.29 | 1,017 | First, thanks for the issue and the very complete explanations. Both `refresh_process` and `refresh_processes` should gather coherent data. So if you're motivated to write a fix, it's very welcome!
OK, I'll try to have a look at it.
Which path do you think would make most sense?
* using `NtQuerySystemInformation` t... | 3cc9cc814d21b7977a0e7bb8260a7b64529e6a1b |
2023-05-05T14:53:18Z | diff --git a/tests/process.rs b/tests/process.rs
--- a/tests/process.rs
+++ b/tests/process.rs
@@ -615,3 +615,47 @@ fn test_process_cpu_usage() {
assert!(process.cpu_usage() <= max_usage);
}
}
+
+#[test]
+fn test_process_creds() {
+ use sysinfo::{ProcessExt, System, SystemExt};
+
+ if !sysinfo::Sys... | [
"970"
] | GuillaumeGomez__sysinfo-977 | GuillaumeGomez/sysinfo | diff --git a/src/apple/app_store/process.rs b/src/apple/app_store/process.rs
--- a/src/apple/app_store/process.rs
+++ b/src/apple/app_store/process.rs
@@ -76,10 +76,18 @@ impl ProcessExt for Process {
None
}
+ fn effective_user_id(&self) -> Option<&Uid> {
+ None
+ }
+
fn group_id(&self... | 41367e7f05b6fcafee328db5ba5859a00b2e2156 | Wrong userid for process sometimes returned on Linux
**Describe the bug**
sysinfo version 0.28.3, platform is Amazon Linux 2 (also reproed on Ubuntu)
`sysinfo` always reports a userid of `root` on Linux for processes with the `dumpable` attribute set to a value other than `1`. This is because it attempts to get t... | 0.28 | 977 | Thanks for the detailed report. Interested into sending a fix maybe?
So turns out on Linux, `sysinfo` gives you the effective UID and GID only, leaving behind real, saved and filesystem IDs. However, on FreeBSD it gives you real IDs instead. :( I actually don't know what the last ones are even so I'd ignore them (until... | 41367e7f05b6fcafee328db5ba5859a00b2e2156 |
2023-01-17T12:46:36Z | diff --git a/src/system.rs b/src/system.rs
--- a/src/system.rs
+++ b/src/system.rs
@@ -164,7 +164,7 @@ mod tests {
for pid in &pids {
sys.refresh_process_specifics(*pid, ProcessRefreshKind::new().with_cpu());
}
- // To ensure that linux doesn't give too high numbers... | [
"905"
] | GuillaumeGomez__sysinfo-924 | GuillaumeGomez/sysinfo | diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -1,4 +1,4 @@
-# sysinfo ![img_github_ci] [![][img_crates]][crates] [![][img_doc]][doc]
+# sysinfo [![][img_crates]][crates] [![][img_doc]][doc]
`sysinfo` is a crate used to get a system's information.
diff --git a/README.md b/README.md
--- a/REA... | 2e28eea78967c43e8a8edbb6a61b7cdb988007ce | process.cpu_usage() is 0
When I use the following code to get the cpu usage of each process, it returns 0, both on windwos and mac
```
use sysinfo::{NetworkExt, NetworksExt, ProcessExt, System, SystemExt};
fn main() {
let mut sys = System::new_all();
sys.refresh_all();
for (pid, process) in sys.pro... | 0.27 | 924 | It's something I'm working on in https://github.com/GuillaumeGomez/sysinfo/pull/903. I need to add a timeout to not refresh the system time too often.
> It's something I'm working on in #903. I need to add a timeout to not refresh the system time too often.
Looking forward to updating as soon as possible.
Start of the... | 2e28eea78967c43e8a8edbb6a61b7cdb988007ce |
2023-01-16T16:45:43Z | diff --git a/src/system.rs b/src/system.rs
--- a/src/system.rs
+++ b/src/system.rs
@@ -129,10 +129,9 @@ mod tests {
return;
}
- let mut sys = System::new();
- sys.refresh_processes_specifics(ProcessRefreshKind::new().with_cpu());
- sys.refresh_cpu();
+ let mut sys = S... | [
"918"
] | GuillaumeGomez__sysinfo-919 | GuillaumeGomez/sysinfo | diff --git a/src/apple/cpu.rs b/src/apple/cpu.rs
--- a/src/apple/cpu.rs
+++ b/src/apple/cpu.rs
@@ -207,7 +207,7 @@ pub(crate) fn init_cpus(
let frequency = if refresh_kind.frequency() {
get_cpu_frequency()
} else {
- 0
+ global_cpu.frequency
};
unsafe {
diff --git a/src/appl... | a02d518033cf8712a13bce82cc09b1b8dda4006a | oddly high CPU usage values on macOS systems after 0.27.4
I was recently updating some stuff to 0.27.4, when I noticed that there were some process CPU usages that were fine before were way too high now, up in the thousands, which seems really weird. I first thought I was perhaps not updating correctly, or that I forgo... | 0.27 | 919 | The problem before `0.27.4` was that the system time (used to compute CPU usage) was recomputed every time `refresh_process` was called. So for each process updated through this method, the CPU usage was 0 or very close to 0, which was wrong. So I added a minimum update time (which is stored in [this constant](https://... | 2e28eea78967c43e8a8edbb6a61b7cdb988007ce |
2022-12-07T10:58:49Z | diff --git a/tests/code_checkers/docs.rs b/tests/code_checkers/docs.rs
--- a/tests/code_checkers/docs.rs
+++ b/tests/code_checkers/docs.rs
@@ -38,8 +38,7 @@ fn check_md_doc_path(p: &Path, md_line: &str, ty_line: &str) -> bool {
show_error(
p,
&format!(
- ... | [
"886"
] | GuillaumeGomez__sysinfo-887 | GuillaumeGomez/sysinfo | diff --git a/examples/simple.rs b/examples/simple.rs
--- a/examples/simple.rs
+++ b/examples/simple.rs
@@ -342,11 +342,7 @@ fn interpret_input(input: &str, sys: &mut System) -> bool {
let minutes = uptime / 60;
writeln!(
&mut io::stdout(),
- "{} days {} hours {}... | abe8e369d0cee1cf85ddc2d864da875c6312d524 | linux: environ is truncated for long environ
```
$ wc -c /proc/882252/environ
30417 /proc/882252/environ
```
environ for this process is truncated approximately half way through
| 0.26 | 887 | suspicious size matching truncation: https://github.com/GuillaumeGomez/sysinfo/blob/master/src/linux/process.rs#L623
Nice catch! It should indeed add a size check when reading the file. | abe8e369d0cee1cf85ddc2d864da875c6312d524 |
2022-12-01T12:55:37Z | diff --git a/src/common.rs b/src/common.rs
--- a/src/common.rs
+++ b/src/common.rs
@@ -962,4 +991,23 @@ mod tests {
fn check_display_impl_process_status() {
println!("{} {:?}", ProcessStatus::Parked, ProcessStatus::Idle);
}
+
+ // This test exists to ensure that the `TryFrom<usize>` and `FromStr` ... | [
"883"
] | GuillaumeGomez__sysinfo-884 | GuillaumeGomez/sysinfo | diff --git a/src/common.rs b/src/common.rs
--- a/src/common.rs
+++ b/src/common.rs
@@ -2,7 +2,7 @@
use crate::{NetworkData, Networks, NetworksExt, UserExt};
-use std::convert::From;
+use std::convert::{From, TryFrom};
use std::fmt;
use std::str::FromStr;
diff --git a/src/common.rs b/src/common.rs
--- a/src/com... | 8be83e1cfbc75e8594b14332a8786978540ff6ff | Uid::from()
Hey guys!
Is Uid::from() method is implemented yet? Cause right now Pid::from() forking fine, but Uid::from() required Uid object anyway.
| 0.26 | 884 | No it's not as no one asked for it. I'll add it today.
Thanks are lot!
Same for Group please
Noted! :laughing: I'll actually go through the newtypes to be sure I don't forget one.
Great, thanks! | abe8e369d0cee1cf85ddc2d864da875c6312d524 |
2022-11-23T15:28:42Z | diff --git a/tests/process.rs b/tests/process.rs
--- a/tests/process.rs
+++ b/tests/process.rs
@@ -293,6 +293,17 @@ fn test_process_times() {
}
}
+// Checks that `session_id` is working.
+#[test]
+fn test_process_session_id() {
+ if !sysinfo::System::IS_SUPPORTED || cfg!(feature = "apple-sandbox") {
+ ... | [
"783"
] | GuillaumeGomez__sysinfo-878 | GuillaumeGomez/sysinfo | diff --git a/md_doc/pid.md b/md_doc/pid.md
--- a/md_doc/pid.md
+++ b/md_doc/pid.md
@@ -1,4 +1,4 @@
-Process id
+Process ID.
Can be used as an integer type by simple casting. For example:
diff --git a/src/apple/app_store/process.rs b/src/apple/app_store/process.rs
--- a/src/apple/app_store/process.rs
+++ b/src/appl... | 0818da38903a7597cb315a1a2bb5392fe0d2e336 | [Feature Request] Process session id
A session id is used to group processes within a session.
For mac/bsd/linix it can be obtained from the [getsid](https://docs.rs/libc/0.2.124/i686-unknown-linux-gnu/libc/fn.getsid.html) system call
For windows it can be found in the [SYSTEM_PROCESS_INFORMATION](https://microso... | 0.26 | 878 | For windows, we can wrap it into an `Option` like we do for the other process IDs. Interested into sending a PR? | abe8e369d0cee1cf85ddc2d864da875c6312d524 |
2022-09-25T00:36:37Z | diff --git a/src/lib.rs b/src/lib.rs
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -328,9 +328,10 @@ mod test {
#[test]
fn check_system_info() {
+ let s = System::new();
+
// We don't want to test on unsupported systems.
if System::IS_SUPPORTED {
- let s = System::new();
... | [
"843"
] | GuillaumeGomez__sysinfo-844 | GuillaumeGomez/sysinfo | diff --git a/src/apple/system.rs b/src/apple/system.rs
--- a/src/apple/system.rs
+++ b/src/apple/system.rs
@@ -605,6 +605,10 @@ impl SystemExt for System {
}
}
}
+
+ fn distribution_id(&self) -> String {
+ std::env::consts::OS.to_owned()
+ }
}
impl Default for System {
diff -... | d5ad6dff70a2e019f8f7ff83e34775c8a20d94bb | Machine-parseable equivalent to System::name()
Hi,
[os-release](https://www.freedesktop.org/software/systemd/man/os-release.html) documents `NAME=` as a human-presentable name of a distro, and `ID=` as a machine-readable identifier. `System::name()` currently relies on `os-release.NAME`, which makes it less than ideal... | 0.26 | 844 | It can be added. The question as always is: "what should it return for other OSes?". On linux you have different distributions so it makes sense, but I think it's the only case. For windows you only have windows, same for macOS. For FreeBSD it could make sense too?
Personally I think it can be the OS name if /etc/os-re... | abe8e369d0cee1cf85ddc2d864da875c6312d524 |
2022-09-03T12:49:44Z | diff --git a/src/lib.rs b/src/lib.rs
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -469,4 +469,19 @@ mod test {
(new_total - total).abs()
);
}
+
+ // We ensure that the `Process` cmd information is retrieved as expected.
+ #[test]
+ fn check_cmd_line() {
+ if !System::IS_SUPPORTED {
+ ... | [
"833"
] | GuillaumeGomez__sysinfo-835 | GuillaumeGomez/sysinfo | diff --git a/src/linux/process.rs b/src/linux/process.rs
--- a/src/linux/process.rs
+++ b/src/linux/process.rs
@@ -353,6 +353,7 @@ fn retrieve_all_new_process_info(
if refresh_kind.user() {
refresh_user_group_ids(&mut p, &mut tmp);
+ tmp.pop();
}
if proc_list.pid.0 != 0 {
diff --git a... | c7046eb4e189afcfb140057caa581d835e1cc51d | Obtaining linux command line requires ProcessRefreshKind::with_user() to be set
It seems retrieving process user information must be turned on when loading processes in linux for the command line to be retrieved.
### Expected
Refreshing a new System with ProcessRefreshKind::new() loads the process command line data... | 0.26 | 835 | This is intriguing. Thanks for opening this issue! I'll try to take a look in the next days but don't hesitate to check before if you want. | abe8e369d0cee1cf85ddc2d864da875c6312d524 |
2022-06-10T12:16:28Z | diff --git a/src/lib.rs b/src/lib.rs
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -444,4 +444,23 @@ mod test {
assert_ne!(proc_.frequency(), 0);
}
}
+
+ // In case `Process::updated` is misused, `System::refresh_processes` might remove them
+ // so this test ensures that it doesn't happen.
+ ... | [
"767"
] | GuillaumeGomez__sysinfo-768 | GuillaumeGomez/sysinfo | diff --git a/src/apple/macos/process.rs b/src/apple/macos/process.rs
--- a/src/apple/macos/process.rs
+++ b/src/apple/macos/process.rs
@@ -30,7 +30,7 @@ pub struct Process {
old_stime: u64,
start_time: u64,
run_time: u64,
- updated: bool,
+ pub(crate) updated: bool,
cpu_usage: f32,
user_i... | ad49264fca17c996547c173e4d8ef792c95b16d5 | macOS: it's necessary to call `refresh_processes` twice to get them all
Using `sysinfo 0.24.2` (also seen on `0.23.13`), with the following code:
```rust
use sysinfo::{System, SystemExt};
fn main() {
let mut s = System::new_all();
s.refresh_processes();
dbg!(s.processes().len());
s.refresh_pr... | 0.24 | 768 | Thanks for the report. I'll try to take a look when I have some time.
After some more investigation, the `System::new_all()` gets them all, but then the first `.refresh_processes()` truncate the list and the second fills it again: I added `println!` calls in `refresh_processes_specifics()`:
```
---- System::new_all... | ad49264fca17c996547c173e4d8ef792c95b16d5 |
2022-06-07T09:09:59Z | diff --git a/src/apple/macos/process.rs b/src/apple/macos/process.rs
--- a/src/apple/macos/process.rs
+++ b/src/apple/macos/process.rs
@@ -615,3 +625,19 @@ fn parse_command_line<T: Deref<Target = str> + Borrow<str>>(cmd: &[T]) -> Vec<St
}
command
}
+
+#[cfg(test)]
+mod test {
+ use super::*;
+
+ #[test... | [
"760"
] | GuillaumeGomez__sysinfo-763 | GuillaumeGomez/sysinfo | diff --git a/src/apple/macos/process.rs b/src/apple/macos/process.rs
--- a/src/apple/macos/process.rs
+++ b/src/apple/macos/process.rs
@@ -247,12 +247,14 @@ pub(crate) fn compute_cpu_usage(
p.updated = true;
}*/
+#[inline]
pub(crate) fn has_been_updated(p: &mut Process) -> bool {
let old = p.updated;
... | f1557bb62eb1b6135fe2e5e7b6f406f84dac3691 | Fix path panic
The following lines in `src/apple/macos/process.rs` would be buggy:
```
fn do_something(env: &str, root: &mut PathBuf, check: &mut bool) {
if *check && env.starts_with("PATH=") {
*check = false;
*root = Path::new(&env[6..]).to_path_... | 0.24 | 763 | Interested into sending a fix? | ad49264fca17c996547c173e4d8ef792c95b16d5 |
2022-06-03T22:53:19Z | diff --git a/src/apple/processor.rs b/src/apple/cpu.rs
--- a/src/apple/processor.rs
+++ b/src/apple/cpu.rs
@@ -316,18 +314,18 @@ mod test {
let sys = System::new_with_specifics(
crate::RefreshKind::new().with_cpu(CpuRefreshKind::new().with_cpu_usage()),
);
- let processors = sys.pr... | [
"736"
] | GuillaumeGomez__sysinfo-754 | GuillaumeGomez/sysinfo | diff --git a/Cargo.toml b/Cargo.toml
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -3,7 +3,7 @@ name = "sysinfo"
version = "0.23.15"
authors = ["Guillaume Gomez <guillaume1.gomez@gmail.com>"]
-description = "Library to get system information such as processes, processors, disks, components and networks"
+description = "Libr... | 6f1c70092ee030535cc2939960baf41ce331a4b2 | Rename `processors` method into `cpus` and `Processor` type into `Cpu`
`ProcessorExt` into `CpuExt` too.
| 0.23 | 754 | 6f1c70092ee030535cc2939960baf41ce331a4b2 | |
2022-05-20T16:09:59Z | diff --git a/tests/process.rs b/tests/process.rs
--- a/tests/process.rs
+++ b/tests/process.rs
@@ -375,7 +375,7 @@ fn test_refresh_process() {
// Let's give some time to the system to clean up...
std::thread::sleep(std::time::Duration::from_secs(1));
- s.refresh_process(pid);
+ assert!(!s.refresh_proc... | [
"741"
] | GuillaumeGomez__sysinfo-743 | GuillaumeGomez/sysinfo | diff --git a/src/apple/macos/process.rs b/src/apple/macos/process.rs
--- a/src/apple/macos/process.rs
+++ b/src/apple/macos/process.rs
@@ -302,6 +302,14 @@ unsafe fn get_task_info(pid: Pid) -> libc::proc_taskinfo {
task_info
}
+fn check_if_pid_is_alive(pid: Pid) -> bool {
+ unsafe { kill(pid.0, 0) == 0 }
+ ... | 13182f50f11646ee958d78a042539491eff6fbdd | refresh_process_specifics won't return false after the process has died
Reproducer:
```rust
#[test]
fn test_repro() {
let mut sys = System::new();
let mut cmd = std::process::Command::new("cat").spawn().unwrap();
let cmd_pid = (cmd.id() as i32).into();
sys.refresh_process_specifics(cmd_pid, P... | 0.23 | 743 | This [test](https://github.com/GuillaumeGomez/sysinfo/blob/master/tests/process.rs#L341-L381) checks exactly that. Generally, the system doesn't remove processes right away. So maybe wait a bit before checking it again?
What could be nice would be to update the documentation about this behaviour though so users aren... | 6f1c70092ee030535cc2939960baf41ce331a4b2 |
2022-03-29T15:35:01Z | diff --git a/tests/process.rs b/tests/process.rs
--- a/tests/process.rs
+++ b/tests/process.rs
@@ -322,6 +322,9 @@ fn test_refresh_processes() {
s.refresh_processes();
assert!(s.process(pid).is_some());
+ // Check that the process name is not empty.
+ assert!(!s.process(pid).unwrap().name().is_empty()... | [
"711"
] | GuillaumeGomez__sysinfo-712 | GuillaumeGomez/sysinfo | diff --git a/src/windows/process.rs b/src/windows/process.rs
--- a/src/windows/process.rs
+++ b/src/windows/process.rs
@@ -178,10 +178,7 @@ unsafe fn get_exe(process_handler: HANDLE, h_mod: *mut c_void) -> PathBuf {
impl Process {
pub(crate) fn new_from_pid(pid: Pid, now: u64) -> Option<Process> {
unsafe... | bcaba0720be731ba78b4585a3a0578145eb5aed4 | Windows: missing process name with refresh_process_specifics(pid, ..)
On Windows, `System::refresh_process_specifics()` and `System::refresh_processes_specifics()` are NOT equivalent. The first one fails to retrieve the process name.
The following produce an empty name.
```rust
let mut system = System::default();
... | 0.23 | 712 | 6f1c70092ee030535cc2939960baf41ce331a4b2 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.