Micha Reiser

Hmm that's an interesting one. The rule is working as intended but I can see how it is undesired in this specific case. I would favor the solution where the...

Hey @MusicalNinjaRandInt We're glad that you're having a good time with ruff! What's the error that you see in your CI pipeline? I'm asking because I'm confused that you see...

Thanks, that clarifies things. I wonder if there's a way to detect if a file is stored inside WSL or on the Windows disk.

> A while back, I originally reported this in one PR but it did not get any attention. I'm sorry for this. That was not intentional. I'm happy to take...

That's an interesting thought. We could create a file in the `.ruff_cache` directory once and reuse that one instead of re-creating a tempfile every time ruff runs. The one issue...

I would go with the temp file for simplicity. I expect the performance overhead for creating the tempfile to be marginal when linting many files where ruff needs to read...

> My expectation would be for ruff to catch this as a duplicate import, and force the import simplification. I can see how unifying the imports is desired but it...

```python a = { "key": ( "very very very very very very very very very very very very very long line " ) } ``` Ruff doesn't parenthesize long dict...

@gaborbernat yes, this is a bug (with the lint rule) and tracked in https://github.com/astral-sh/ruff/issues/10041#issuecomment-1955916714

@gaborbernat I'm trying to reproduce your diff, but I'm struggling to understand it. What I understand is that. * The formatter produces the second output (`+`) * The linter fix...