Cython pure python declare

[jtoledo1974]

Type of Changes

	Type

Description

This Pull Requests provides a test case for #8024.
A pure python cython.declare(cython.int... is reported as no-member. These should be ignored.

Refs #8024

Closes #XXXX

[codecov]

Codecov Report

Merging #8025 (416c233) into main (5612134) will increase coverage by 0.01%.
The diff coverage is n/a.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #8025      +/-   ##
==========================================
+ Coverage   95.42%   95.44%   +0.01%     
==========================================
  Files         176      176              
  Lines       18521    18528       +7     
==========================================
+ Hits        17674    17684      +10     
+ Misses        847      844       -3     

Impacted Files	Coverage Δ
pylint/pyreverse/inspector.py	78.94% <0.00%> (ø)
pylint/checkers/variables.py	97.38% <0.00%> (+<0.01%)	⬆️
pylint/checkers/utils.py	95.33% <0.00%> (+<0.01%)	⬆️
pylint/checkers/base/basic_checker.py	97.87% <0.00%> (+0.02%)	⬆️
pylint/testutils/lint_module_test.py	86.15% <0.00%> (+1.53%)	⬆️

[jtoledo1974]

@Pierre-Sassoulas , I took a look at it. Both pure python cython problems get resolved by ignoring the cython module. I took a look to see whether this would be a problem for the cython developers, but they themselves don't seem to do cimports.

Problem is that just setting a default for ignored-imports doesn't cut it because it would be overridden by any user configuration.

The option reading code is complex and I don't know where in the codebase it is that you would like to have the forced ignored-import added. I'll do it if you give me a hint.

[DanielNoord]

We don't test with cython right? So this PR doesn't make a lot of sense in its current state imo.

[jtoledo1974]

We don't test with cython right? So this PR doesn't make a lot of sense in its current state imo.

The point of the matter is that this is not just cython code. It is valid python interpreted code. Cython allows to compile it as well, but it shouldn't be flagged as invalid.

[DanielNoord]

We don't test with cython right? So this PR doesn't make a lot of sense in its current state imo.

The point of the matter is that this is not just cython code. It is valid python interpreted code. Cython allows to compile it as well, but it shouldn't be flagged as invalid.

I don't really understand your point. I was just saying that this test will never run in CI which is why this PR has a successful CI even though there is no fix for the added test.

[jtoledo1974]

We don't test with cython right? So this PR doesn't make a lot of sense in its current state imo.

The point of the matter is that this is not just cython code. It is valid python interpreted code. Cython allows to compile it as well, but it shouldn't be flagged as invalid.

I don't really understand your point. I was just saying that this test will never run in CI which is why this PR has a successful CI even though there is no fix for the added test.

I apologize. I misunderstood and my choice of words was poor.

I was forcing the test to run to make sure that it failed first and then it passed. I would say it was run together with the other tests after a full run, but I can't say for sure. But if CI passes I guess it didn't.

[github-actions]

This PR needs take over because because it has been open 8 weeks with no activity.