Fix Linux process names

[MarcoRossignoli]

Replacement of #35809
I moved changes to new PR for convenience(few line of code) and added a test.
If it's not ok (the solution is of @2E0PGS I only added test) please tell me how I can move commit from another forked branch to mine(so I don't waste time scrubbing on internet), I neved did it.

Fixes https://github.com/dotnet/corefx/issues/34437

/cc @2E0PGS @danmosemsft @krwq @wtgodbe

Output without changes:

~/repos/corefx/artifacts/bin/tests/System.Diagnostics.Process.Tests/netcoreapp-Linux-Debug-x64 ~/repos/corefx/src/System.Diagnostics.Process/tests
    Discovering: System.Diagnostics.Process.Tests (method display = ClassAndMethod, method display options = None)
    Discovered:  System.Diagnostics.Process.Tests (found 1 of 273 test case)
    Starting:    System.Diagnostics.Process.Tests (parallel test collections = on, max threads = 4)
      System.Diagnostics.Tests.ProcessTests.GetProcesses_LongProcessName [FAIL]
        Process named '123456789012345678901234567890' not found
        Expected: True
        Actual:   False
        Stack Trace:
          /home/marco/repos/corefx/src/System.Diagnostics.Process/tests/ProcessTests.Unix.cs(95,0): at System.Diagnostics.Tests.ProcessTests.GetProcesses_LongProcessName()
    Finished:    System.Diagnostics.Process.Tests
  === TEST EXECUTION SUMMARY ===
     System.Diagnostics.Process.Tests  Total: 1, Errors: 0, Failed: 1, Skipped: 0, Time: 0.453s
  ~/repos/corefx/src/System.Diagnostics.Process/tests
  ----- end 16:34:01 ----- exit code 1 ----------------------------------------------------------
  Looking around for any Linux dump...
  ... found no dump in /home/marco/repos/corefx/artifacts/bin/tests/System.Diagnostics.Process.Tests/netcoreapp-Linux-Debug-x64

[2E0PGS]

That doesn't really keep the history/ contribution accurate. Why not make fork my branch and add test then submit as PR to me. I merge and then it updates the PR I have open. I believe.

…

On 15 Apr 2019, 15:46, at 15:46, Marco Rossignoli ***@***.***> wrote: Replacement of #35809 I moved change on new PR for convenience(few line of code) and added a test. If it's not ok (the solution is of @2E0PGS I only added test) please tell me how I can move commit from another forked branch to mine(so I don't waste time scrubbing on internet), I neved did it. Fixes https://github.com/dotnet/corefx/issues/34437 /cc @2E0PGS @danmosemsft @krwq @wtgodbe You can view, comment on, or merge this pull request online at: #36880 -- Commit Summary -- * add test -- File Changes -- M src/System.Diagnostics.Process/src/System/Diagnostics/ProcessManager.Linux.cs (9) M src/System.Diagnostics.Process/tests/ProcessTestBase.cs (14) M src/System.Diagnostics.Process/tests/ProcessTests.Unix.cs (47) -- Patch Links -- https://github.com/dotnet/corefx/pull/36880.patch https://github.com/dotnet/corefx/pull/36880.diff -- You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub: #36880

[danmoseley]

Note that we squash history before committing PR'S to master.

[danmoseley]

Looks good except for minor comments. Thanks!

[2E0PGS]

So are we just going to ignore that I won't show as a contributor despite my contributions? Sure the PR mentions my name but GitHub contributors and blame doesn't. Ideally you shouldn't copy and paste that's not making use of the VCS. Squashing I could of done if required. Then you have one commit from me and one from Marco with a test. Cherry pick would be better than copy paste. History is preserved.

…

On 15 Apr 2019, 17:04, at 17:04, Marco Rossignoli ***@***.***> wrote: MarcoRossignoli commented on this pull request. > @@ -58,6 +58,53 @@ public void GetProcessesByName_RemoteMachineNameUnix_ThrowsPlatformNotSupportedE Assert.Throws<PlatformNotSupportedException>(() => Process.GetProcessesByName(currentProcess.ProcessName, machineName)); } + [Fact] C:\Program Files (x86)\Windows Resource Kits\Tools\sleep.exe on my machine....I think it's installed also on some CI machine...BTW in that case will skip. -- You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub: #36880 (comment)

[krwq]

I agree with @2E0PGS that this should start with his squashed changes and changes on top of that separately - after that we should merge/rebase (not squash) this

[danmoseley]

So long as we end up with a single (non merge) commit in master, any variation is fine.

[MarcoRossignoli]

@2E0PGS feel free to copy paste my code it's not a problem to me at all!Don't worry!
Test is less important than investigation and solution.

[danmoseley]

@MarcoRossignoli you should be able to push your change to the existing PR #35809. @2E0PGS might need to give you rights, give it a try.

[MarcoRossignoli]

Ok

[MarcoRossignoli]

@2E0PGS I need permission to write to your branch

remote: Permission to 2E0PGS/corefx.git denied to MarcoRossignoli.
fatal: unable to access 'https://github.com/2E0PGS/corefx.git/': The requested URL returned error: 403

[danmoseley]

It may also work to cherry-pick the change by @2E0PGS into your own branch (if you want to play more with Git 😄 )

I do not know which person is considered the "author" when we squash to merge. I'm guessing the author of the first commit.

It's fine if authorship works out but as I mentioned, we generally don't pay much attention to that and most folks seem OK with that.

[2E0PGS]

Added Marco's write perms.

push to my new branch https://github.com/2E0PGS/corefx/tree/b/linux-proc-name-char

there is a pr for that now.

[MarcoRossignoli]

@2E0PGS added test, I think you need to login for CLA.

[2E0PGS]

done.