Does pep8speaks follow #noqa flags?

[fmaussion]

We are in the process of moving from stickler-ci to pep8speaks, but we get false alarms from lines annotated with a #noqa (see this PR).

Is this expected? Is there a way to avoid these? Thanks a lot!

[OrkoHunter]

Hey @fmaussion! Thank you for reaching out to me about the issue.

I wasn't aware of #noqa until now. PEP8Speaks uses pycodestyle as the linter. So, the error messages are not up to me.

I found some discussion about implementing it in pycodestyle, but if I am not wrong, it still remains a missing feature.

So, yes this is expected as of now, and the only way to avoid it is to put E741 in the ignores list of .pep8speaks.yml

[fmaussion]

Thanks for the info! Maybe we'll keep this issue open for future reference.

[OrkoHunter]

Hi @fmaussion !

I can see that E741 has been added to the ignore list in .pep8speaks.yml. Does @pep8speaks still complain about any #noqa errors other than E741?

I would also like to let you know that there is no need to maintain two copies of pycodestyle/flake8 settings in .pep8speaks.yml and setup.cfg. You can go ahead and remove pycodestyle: section from .pep8speaks.yml and it will read the [flake8] section of setup.cfg file. See #95 and pandas-dev/pandas#25605

[OrkoHunter]

Hi @fmaussion !

PEP 8 Speaks can now support #noqa. After #115, projects can choose between pycodestyle and flake8 as their linter. The scanner.linter option .pep8speaks.yml should be set to flake8.

Example .pep8speaks.yml -

scanner:
    diff_only: True
    linter: flake8

flake8:  # Skip this if you have [flake8] in your setup.cfg file
    ignore:
        - EXXX
        - WXXX

Complete configuration details is in the README. I am going to close this issue. Let me know if you have any questions/comments!