Using ruff alongside flake8-noqa

[Apakottur]

I am working on a project where we're using both ruff and flake8 (with the flake8-noqa) plugin.

Consider the following code:

# test.py
text = "Русский"

Running ruff with RUF001 enabled gives us:

ruff check --select RUF001 test.py 
test.py:1:9: RUF001 [*] String contains ambiguous unicode character `Р` (did you mean `P`?)
test.py:1:10: RUF001 [*] String contains ambiguous unicode character `у` (did you mean `y`?)
test.py:1:11: RUF001 [*] String contains ambiguous unicode character `с` (did you mean `c`?)
test.py:1:12: RUF001 [*] String contains ambiguous unicode character `с` (did you mean `c`?)
Found 4 errors.

If I attempt to silence the warning by modifying the file:

# test.py
text = "Русский"  # noqa: RUF001

Then running flake8 with the flake8-noqa plugin gives us:

flake8 test.py
test.py:1:19: NQA102 "# noqa: RUF001" has no matching violations

It would be great if ruff could be disabled on a line with something like # ruff-disable: RUF001 and not # noqa to not clash with flake8 on that matter, but perhaps there is another way to do this (without disabling the ruff rule completely, or for the entire file)?

[charliermarsh]

Yeah there's not a great solution to this right now. We do support the opposite thing, which is that if you use Ruff's RUF100 rule instead of flake8-noqa, you can mark codes that Ruff doesn't know about as "allowed" via the external setting.

[FrostRunner]

It would be great if this rule considered the string to be good if the string is Unicode.

text = u"Русский"

[charliermarsh]

Interestingly, VS Code flags "Русс" but not "Русск" (or anything longer). Maybe they're detecting whether the string contains at least one non-"confusable" Unicode character?

[charliermarsh]

Yes, that's roughly it: microsoft/vscode#143720 (comment)

[charliermarsh]

We should implement that