Fix g:ycm_roslyn_binary_path being ignored #1721

dkaszews · 2023-12-09T17:54:28Z

Fixes the g:ycm_roslyn_binary_path only being used in prechecks, and the actual server always being started from the ycmd/third_party directory.

Also fixes the :YcmDebugInfo printing the directory containing default OmniSharp executable (and not the executable file itself); now prints full command with arguments used to start the server.

This change is

bstaletic

Functionally, this looks good.
I did leave one comment.

Reviewed 2 of 2 files at r1, all commit messages.
Reviewable status: 0 of 2 LGTMs obtained (waiting on @dkaszews)

ycmd/tests/cs/debug_info_test.py line 207 at r1 (raw file):

  def _RoslynFromUserOption_popen_mock(*args, **kwargs):

This will fail flake8 check.

( *args, **kwargs ) - mind the "floating" parentheses.

Also, I'm too hungry to take a closer look now, but...

if this helper function needs to be a member of the test class, I would prefer to see self, *args, **kwargs.
If it needs not be a member of the test class, let's move it above the test class.

dkaszews

Reviewable status: 0 of 2 LGTMs obtained (waiting on @bstaletic)

ycmd/tests/cs/debug_info_test.py line 207 at r1 (raw file):

Previously, bstaletic (Boris Staletic) wrote…

This will fail flake8 check.

( *args, **kwargs ) - mind the "floating" parentheses.

Also, I'm too hungry to take a closer look now, but...

if this helper function needs to be a member of the test class, I would prefer to see self, *args, **kwargs.
If it needs not be a member of the test class, let's move it above the test class.

It doesn't need to be member of test class, but I would like to keep it next to the test case as it has same hardcoded value, basically a multiline lambda. Maybe @staticmethod then?

I also just had a stupid idea to check out tommorow. We only care about value here and the forward to real Popen is just to keep rest of test of failing. Just throw some custom exception that will only be caught in test and end it early. Should save a lot of time by not actually starting the server and waiting on it.

dkaszews

.

Reviewable status: 0 of 2 LGTMs obtained (waiting on @bstaletic)

bstaletic

Let's see if CI has come to its senses.

Reviewable status: 0 of 2 LGTMs obtained (waiting on @dkaszews)

ycmd/tests/cs/debug_info_test.py line 207 at r1 (raw file):

Previously, dkaszews (Dominik Kaszewski) wrote…

It doesn't need to be member of test class, but I would like to keep it next to the test case as it has same hardcoded value, basically a multiline lambda. Maybe @staticmethod then?

I also just had a stupid idea to check out tommorow. We only care about value here and the forward to real Popen is just to keep rest of test of failing. Just throw some custom exception that will only be caught in test and end it early. Should save a lot of time by not actually starting the server and waiting on it.

@staticmethod sounds good to me.

bstaletic · 2023-12-10T12:44:04Z

Also, we have merged the unicode/regex update, so you should rebase.

dkaszews

Fixed and rebased

Reviewable status: 0 of 2 LGTMs obtained (waiting on @bstaletic)

ycmd/tests/cs/debug_info_test.py line 207 at r1 (raw file):

Previously, bstaletic (Boris Staletic) wrote…

@staticmethod sounds good to me.

Done, the exception idea unfortunatelly did not work, as it was caught by production code I don't have time to investigate.

bstaletic

Reviewed 1 of 1 files at r2, all commit messages.
Reviewable status: 1 of 2 LGTMs obtained (waiting on @dkaszews)

bstaletic

Reviewed 1 of 1 files at r3, all commit messages.
Reviewable status: 0 of 2 LGTMs obtained (and 1 stale) (waiting on @dkaszews)

dkaszews · 2023-12-10T22:30:30Z

Apparently @patch does not work with @staticmethod before Python 3.10 and CI uses 3.8. Wouldn't be an issue if stupid Python had multiline lambdas. Will have to look at alternative solutions.

bstaletic

We can just drop @staticmethod and add a self as the first argument, which it would have anyway. Then just add a comment about it too.

Reviewable status: 0 of 2 LGTMs obtained (and 1 stale) (waiting on @dkaszews)

dkaszews · 2023-12-11T15:47:59Z

That did not work either, was getting errors that instance method was called without self. Found alternative of patching inside test with local function.

puremourning

Reviewed 1 of 1 files at r4, 1 of 1 files at r5, all commit messages.
Reviewable status: 0 of 2 LGTMs obtained (and 1 stale) (waiting on @dkaszews)

ycmd/tests/cs/debug_info_test.py line 212 at r5 (raw file):

    # `@patch` does not play nice with functions defined at class scope
    def _popen_mock( *args, **kwargs ):
      assert_that( args[ 0 ][ 1 ], equal_to( 'my_roslyn.exe' ) )

given that we're using args[0] directly, why not:

def _popen_mock( cmdline, *args, **kwargs ):
    # use cmdline

dkaszews

Applied suggestion

Reviewable status: 0 of 2 LGTMs obtained (and 1 stale) (waiting on @puremourning)

ycmd/tests/cs/debug_info_test.py line 212 at r5 (raw file):

Previously, puremourning (Ben Jackson) wrote…

given that we're using args[0] directly, why not:
def _popen_mock( cmdline, *args, **kwargs ):
    # use cmdline

Fair point, done. I omitted *args because all other arguments are named-only.

puremourning

Reviewed 1 of 1 files at r6, all commit messages.
Reviewable status: 1 of 2 LGTMs obtained (and 1 stale) (waiting on @dkaszews)

dkaszews · 2023-12-13T20:08:55Z

Any clue why this fails only on older Windows?

puremourning · 2023-12-21T16:55:43Z

I have re-run them. Let's see if it was CI flake.

puremourning · 2023-12-21T17:35:47Z

Nope, looks like the test really does just fail on windows:

======================================================================
ERROR: test_GetCompleter_RoslynFromUserOption (ycmd.tests.cs.debug_info_test.DebugInfoTest)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "C:\hostedtoolcache\windows\Python\3.8.10\x64\lib\unittest\mock.py", line 1325, in patched
    return func(*newargs, **newkeywargs)
  File "D:\a\ycmd\ycmd\ycmd\tests\cs\__init__.py", line 70, in Wrapper
    test( test_case_instance, app, *args, **kwargs )
  File "D:\a\ycmd\ycmd\ycmd\tests\cs\debug_info_test.py", line 219, in test_GetCompleter_RoslynFromUserOption
    with WrapOmniSharpServer( app, filepath ):
  File "C:\hostedtoolcache\windows\Python\3.8.10\x64\lib\contextlib.py", line 113, in __enter__
    return next(self.gen)
  File "D:\a\ycmd\ycmd\ycmd\tests\cs\__init__.py", line 108, in WrapOmniSharpServer
    GetDiagnostics( app, filepath )
  File "D:\a\ycmd\ycmd\ycmd\tests\cs\__init__.py", line 99, in GetDiagnostics
    return app.post_json( '/event_notification', event_data ).json
  File "C:\hostedtoolcache\windows\Python\3.8.10\x64\lib\site-packages\webtest\utils.py", line 34, in wrapper
    return self._gen_request(method, url, **kw)
  File "C:\hostedtoolcache\windows\Python\3.8.10\x64\lib\site-packages\webtest\app.py", line 749, in _gen_request
    return self.do_request(req, status=status,
  File "C:\hostedtoolcache\windows\Python\3.8.10\x64\lib\site-packages\webtest\app.py", line 646, in do_request
    self._check_status(status, res)
  File "C:\hostedtoolcache\windows\Python\3.8.10\x64\lib\site-packages\webtest\app.py", line 675, in _check_status
    raise AppError(
webtest.app.AppError: Bad response: 500 Internal Server Error (not 200 OK or 3xx redirect for http://localhost/event_notification)
b'{"exception":{"TYPE":"AssertionError"},"message":"\\nExpected: \'my_roslyn.exe\'\\n     but: was \'-p\'\\n","traceback":"Traceback (most recent call last):\\n  File \\"D:\\\\a\\\\ycmd\\\\ycmd\\\\third_party\\\\bottle\\\\bottle.py\\", line 876, in _handle\\n    return route.call(**args)\\n  File \\"D:\\\\a\\\\ycmd\\\\ycmd\\\\third_party\\\\bottle\\\\bottle.py\\", line 1756, in wrapper\\n    rv = callback(*a, **ka)\\n  File \\"D:\\\\a\\\\ycmd\\\\ycmd\\\\ycmd\\\\handlers.py\\", line 65, in EventNotification\\n    response_data = getattr( _server_state.GetFiletypeCompleter( filetypes ),\\n  File \\"D:\\\\a\\\\ycmd\\\\ycmd\\\\ycmd\\\\completers\\\\cs\\\\cs_completer.py\\", line 262, in OnFileReadyToParse\\n    solutioncompleter._StartServer()\\n  File \\"D:\\\\a\\\\ycmd\\\\ycmd\\\\ycmd\\\\completers\\\\cs\\\\cs_completer.py\\", line 421, in _StartServer\\n    return self._StartServerNoLock()\\n  File \\"D:\\\\a\\\\ycmd\\\\ycmd\\\\ycmd\\\\completers\\\\cs\\\\cs_completer.py\\", line 466, in _StartServerNoLock\\n    self._omnisharp_phandle = utils.SafePopen(\\n  File \\"D:\\\\a\\\\ycmd\\\\ycmd\\\\ycmd\\\\utils.py\\", line 370, in SafePopen\\n    return subprocess.Popen( args, **kwargs )\\n  File \\"C:\\\\hostedtoolcache\\\\windows\\\\Python\\\\3.8.10\\\\x64\\\\lib\\\\unittest\\\\mock.py\\", line 1081, in __call__\\n    return self._mock_call(*args, **kwargs)\\n  File \\"C:\\\\hostedtoolcache\\\\windows\\\\Python\\\\3.8.10\\\\x64\\\\lib\\\\unittest\\\\mock.py\\", line 1085, in _mock_call\\n    return self._execute_mock_call(*args, **kwargs)\\n  File \\"C:\\\\hostedtoolcache\\\\windows\\\\Python\\\\3.8.10\\\\x64\\\\lib\\\\unittest\\\\mock.py\\", line 1155, in _execute_mock_call\\n    return self._mock_wraps(*args, **kwargs)\\n  File \\"D:\\\\a\\\\ycmd\\\\ycmd\\\\ycmd\\\\tests\\\\cs\\\\debug_info_test.py\\", line 212, in _popen_mock\\n    assert_that( cmdline[ 1 ], equal_to( \'my_roslyn.exe\' ) )\\n  File \\"C:\\\\hostedtoolcache\\\\windows\\\\Python\\\\3.8.10\\\\x64\\\\lib\\\\site-packages\\\\hamcrest\\\\core\\\\assert_that.py\\", line 58, in assert_that\\n    _assert_match(actual=actual_or_assertion, matcher=matcher, reason=reason)\\n  File \\"C:\\\\hostedtoolcache\\\\windows\\\\Python\\\\3.8.10\\\\x64\\\\lib\\\\site-packages\\\\hamcrest\\\\core\\\\assert_that.py\\", line 73, in _assert_match\\n    raise AssertionError(description)\\nAssertionError: \\nExpected: \'my_roslyn.exe\'\\n     but: was \'-p\'\\n\\n"}'

nExpected: 'my_roslyn.exe'\n but: was '-p'\n\n"}'

dkaszews · 2023-12-21T17:40:44Z

Could not get through the callstack myself, but now it just looks like it fails because on Windows it runs with my_roslyn.exe ... instead of mono my_roslyn.exe .... Easy enough fix, that's why I asked on gitter whether this test will run on Windows 🤷

puremourning · 2023-12-21T17:47:35Z

yeah, it's a bit cryptic the way the failure is reported!

codecov · 2023-12-21T18:45:15Z

Codecov Report

Merging #1721 (e318b5e) into master (f4cfdb8) will increase coverage by 0.00%.
The diff coverage is 100.00%.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #1721   +/-   ##
=======================================
  Coverage   95.45%   95.45%           
=======================================
  Files          83       83           
  Lines        8136     8143    +7     
  Branches      163      163           
=======================================
+ Hits         7766     7773    +7     
  Misses        320      320           
  Partials       50       50

puremourning

Reviewed 1 of 1 files at r7, all commit messages.
Reviewable status: 0 of 2 LGTMs obtained (and 2 stale) (waiting on @dkaszews)

ycmd/tests/cs/debug_info_test.py line 212 at r7 (raw file):

    # `@patch` does not play nice with functions defined at class scope
    def _popen_mock( cmdline, **kwargs ):
      i = cmdline[ 0 ].endswith( 'mono' )

not a fan of this. I would rather:

exe_idx = 1 if cmdline[ 0 ].endswith( 'mono' ) else 0

Because that says what it is doing, rather than assuming you implicitly know that, and that the True will turn into 1 when used as an index... which is weird.

dkaszews

Reviewable status: 0 of 2 LGTMs obtained (and 2 stale) (waiting on @puremourning)

ycmd/tests/cs/debug_info_test.py line 212 at r7 (raw file):

Previously, puremourning (Ben Jackson) wrote…

not a fan of this. I would rather:
exe_idx = 1 if cmdline[ 0 ].endswith( 'mono' ) else 0
Because that says what it is doing, rather than assuming you implicitly know that, and that the True will turn into 1 when used as an index... which is weird.

Done.

puremourning

Let's squash.

Reviewed 1 of 1 files at r9.
Reviewable status: 1 of 2 LGTMs obtained (and 1 stale) (waiting on @dkaszews)

-- commits line 28 at r9:
let's squash the commits now, LGTM.

puremourning

Reviewed all commit messages.
Reviewable status: 1 of 2 LGTMs obtained (and 1 stale) (waiting on @dkaszews)

dkaszews · 2023-12-23T21:56:34Z

@puremourning Do you need me to squash manually? I'm AFK so might be a bit hard.

The codecov fails seem to be because I rewrote bits of code to be shorter so the line percentage went down 🙄

puremourning · 2023-12-24T13:44:30Z

@Mergifyio squash

Fix g:ycm_roslyn_binary_path being ignored * Add test * Clean test * Fix formatting * Fix UT * Fix UT for Python 3.8 * Make popen mock args more readable * Fix UT on Windows * Explicit index * Explicit index name

mergify · 2023-12-24T13:44:40Z

squash

✅ Pull request squashed successfully

bstaletic

@dkaszews Thanks for the pull request!

Reviewed 1 of 1 files at r4.
Reviewable status: 1 of 2 LGTMs obtained (and 1 stale) (waiting on @dkaszews)

puremourning

Reviewable status: complete! 2 of 2 LGTMs obtained (waiting on @dkaszews)

mergify · 2023-12-27T16:23:54Z

Thanks for sending a PR!

puremourning

Reviewed all commit messages.
Reviewable status: complete! 2 of 2 LGTMs obtained (waiting on @dkaszews)

mergify · 2023-12-27T16:24:01Z

Thanks for sending a PR!

dkaszews force-pushed the fix_roslyn_path branch from d7eca35 to 6ef84d3 Compare December 9, 2023 20:45

bstaletic requested changes Dec 9, 2023

View reviewed changes

dkaszews commented Dec 9, 2023

View reviewed changes

bstaletic requested changes Dec 10, 2023

View reviewed changes

dkaszews force-pushed the fix_roslyn_path branch from 6ef84d3 to 67274ed Compare December 10, 2023 19:35

dkaszews commented Dec 10, 2023

View reviewed changes

dkaszews requested a review from bstaletic December 10, 2023 20:00

dkaszews mentioned this pull request Dec 10, 2023

Add instructions for use with .NET 6.0 ycm-core/YouCompleteMe#4205

Merged

4 tasks

bstaletic approved these changes Dec 10, 2023

View reviewed changes

bstaletic reviewed Dec 11, 2023

View reviewed changes

puremourning requested changes Dec 12, 2023

View reviewed changes

dkaszews commented Dec 12, 2023

View reviewed changes

puremourning approved these changes Dec 12, 2023

View reviewed changes

puremourning requested changes Dec 21, 2023

View reviewed changes

dkaszews commented Dec 21, 2023

View reviewed changes

puremourning approved these changes Dec 22, 2023

View reviewed changes

puremourning reviewed Dec 22, 2023

View reviewed changes

Fix g:ycm_roslyn_binary_path being ignored (ycm-core#1721)

e318b5e

Fix g:ycm_roslyn_binary_path being ignored * Add test * Clean test * Fix formatting * Fix UT * Fix UT for Python 3.8 * Make popen mock args more readable * Fix UT on Windows * Explicit index * Explicit index name

puremourning force-pushed the fix_roslyn_path branch from a6e4146 to e318b5e Compare December 24, 2023 13:44

bstaletic reviewed Dec 26, 2023

View reviewed changes

puremourning added the Ship It! Manual override to merge a PR by maintainer label Dec 27, 2023

puremourning reviewed Dec 27, 2023

View reviewed changes

mergify bot merged commit 80a034d into ycm-core:master Dec 27, 2023
18 checks passed

puremourning reviewed Dec 27, 2023

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Fix g:ycm_roslyn_binary_path being ignored #1721

Fix g:ycm_roslyn_binary_path being ignored #1721

dkaszews commented Dec 9, 2023 •

edited by Valloric

bstaletic left a comment

dkaszews left a comment

dkaszews left a comment

bstaletic left a comment

bstaletic commented Dec 10, 2023

dkaszews left a comment

bstaletic left a comment

bstaletic left a comment

dkaszews commented Dec 10, 2023

bstaletic left a comment

dkaszews commented Dec 11, 2023

puremourning left a comment

dkaszews left a comment

puremourning left a comment

dkaszews commented Dec 13, 2023

puremourning commented Dec 21, 2023

puremourning commented Dec 21, 2023 •

edited

dkaszews commented Dec 21, 2023

puremourning commented Dec 21, 2023

codecov bot commented Dec 21, 2023 •

edited

puremourning left a comment

dkaszews left a comment

puremourning left a comment

puremourning left a comment

dkaszews commented Dec 23, 2023

puremourning commented Dec 24, 2023

mergify bot commented Dec 24, 2023

bstaletic left a comment

puremourning left a comment

mergify bot commented Dec 27, 2023

puremourning left a comment

mergify bot commented Dec 27, 2023

Fix g:ycm_roslyn_binary_path being ignored #1721

Fix g:ycm_roslyn_binary_path being ignored #1721

Conversation

dkaszews commented Dec 9, 2023 • edited by Valloric

bstaletic left a comment

Choose a reason for hiding this comment

dkaszews left a comment

Choose a reason for hiding this comment

dkaszews left a comment

Choose a reason for hiding this comment

bstaletic left a comment

Choose a reason for hiding this comment

bstaletic commented Dec 10, 2023

dkaszews left a comment

Choose a reason for hiding this comment

bstaletic left a comment

Choose a reason for hiding this comment

bstaletic left a comment

Choose a reason for hiding this comment

dkaszews commented Dec 10, 2023

bstaletic left a comment

Choose a reason for hiding this comment

dkaszews commented Dec 11, 2023

puremourning left a comment

Choose a reason for hiding this comment

dkaszews left a comment

Choose a reason for hiding this comment

puremourning left a comment

Choose a reason for hiding this comment

dkaszews commented Dec 13, 2023

puremourning commented Dec 21, 2023

puremourning commented Dec 21, 2023 • edited

dkaszews commented Dec 21, 2023

puremourning commented Dec 21, 2023

codecov bot commented Dec 21, 2023 • edited

Codecov Report

puremourning left a comment

Choose a reason for hiding this comment

dkaszews left a comment

Choose a reason for hiding this comment

puremourning left a comment

Choose a reason for hiding this comment

puremourning left a comment

Choose a reason for hiding this comment

dkaszews commented Dec 23, 2023

puremourning commented Dec 24, 2023

mergify bot commented Dec 24, 2023

✅ Pull request squashed successfully

bstaletic left a comment

Choose a reason for hiding this comment

puremourning left a comment

Choose a reason for hiding this comment

mergify bot commented Dec 27, 2023

puremourning left a comment

Choose a reason for hiding this comment

mergify bot commented Dec 27, 2023

dkaszews commented Dec 9, 2023 •

edited by Valloric

puremourning commented Dec 21, 2023 •

edited

codecov bot commented Dec 21, 2023 •

edited