allow-global-unused-variables should respect dummy-variables-rgx

[com6056]

Type of Changes

	Type
✓	🐛 Bug fix

Description

Right now, --allow-global-unused-variables=no doesn't respect --dummy-variables-rgx, I believe this should fix it 🤞

Refs #8174

Closes #8174

[com6056]

Not sure if this change requires news or not, let me know if it does though! I just don't have all the tools installed to regenerate everything but can get that going if needed.

[DanielNoord]

This would indeed require a news entry, if you don't want to install the tools you can probably copy one from a previous PR and see what the syntax looks like there.

[codecov]

Codecov Report

Attention: Patch coverage is 50.00000% with 1 lines in your changes are missing coverage. Please review.

Project coverage is 95.81%. Comparing base (67bfab4) to head (e80d0a5).
Report is 10 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #9570      +/-   ##
==========================================
- Coverage   95.81%   95.81%   -0.01%     
==========================================
  Files         173      173              
  Lines       18825    18827       +2     
==========================================
+ Hits        18038    18039       +1     
- Misses        787      788       +1     

Files	Coverage Δ
pylint/checkers/variables.py	97.28% <50.00%> (-0.07%)	⬇️

[com6056]

This would indeed require a news entry, if you don't want to install the tools you can probably copy one from a previous PR and see what the syntax looks like there.

Done! Just peeked at some other PRs, took a few tries but seems like it is happy now haha, let me know if there's anything else needed before we can get this merged in! 🎉

[Pierre-Sassoulas]

Thank you for opening a PR. I do think we need to think about #8013 before "fixing" this. There's a bad/ "two options in parallel" initial design and we need to restore consistency (and document this better then close #8013 in this MR, if this MR restore consistency).

[github-actions]

🤖 According to the primer, this change has no effect on the checked open source code. 🤖🎉

This comment was generated for commit e80d0a5

[com6056]

Thank you for opening a PR. I do think we need to think about #8013 before "fixing" this. There's a bad/ "two options in parallel" initial design and we need to restore consistency (and document this better then close #8013 in this MR, if this MR restore consistency).

Any idea how long this will take? Could I instead just add a new flag that allows you instead to opt-in to this behavior then if it might take a while to resolve #8013?

[Pierre-Sassoulas]

If you can propose a consistent design it could be fast, otherwise we need to wait for someone else to think this is important and volunteer to think about it. Maybe the current MR is enough and making them mutually exclusive and using only one option at a time is not the way to go. But it should be argumented/documented. Right now I feel we're fixing a particular use case, instead of actually fixing the root issue and are going to throw all this when actually looking into it.

[com6056]

Where in the docs should I add this? And what do you mean by "argumented"? Should I add another CLI flag that enables this?

[Pierre-Sassoulas]

And what do you mean by "argumented"? Should I add another CLI flag that enables this?

Sorry I was thinking in french, I actually meant argued / reasoned that the better solution is to keep both options and not make them mutually exclusive. We would need to document both of the options if that's the case (here

pylint/pylint/checkers/variables.py

Line 1243 in 3c8be8e

"help": "A regular expression matching the name of dummy "

).