UP045 fixes `Optional[None]` to `None | None`

[dscorbett]

Summary

The fix for non-pep604-annotation-optional (UP045) can produce None | None, the persistently inconvenient exception to PEP 604.

$ cat >up045.py <<'# EOF'
from typing import Optional
foo: Optional[None] = None
# EOF

$ ruff --isolated check up045.py --select UP045 --preview --target-version py310 --fix 
Found 1 error (1 fixed, 0 remaining).

$ cat up045.py
from typing import Optional
foo: None | None = None

$ python up045.py 2>&1 | tail -n 1
TypeError: unsupported operand type(s) for |: 'NoneType' and 'NoneType'

Version

ruff 0.11.13 (5faf72a 2025-06-05)

[chirizxc]

we should ignore it or fix foo: Optional[None] = None => foo: None = None?

[ntBre]

Is there ever a valid reason for Optional[None] or is it a likely mistake on its own? If it's probably a mistake, we should leave it alone in UP045 and possibly consider a separate rule to catch it.

[chirizxc]

Is there ever a valid reason for Optional[None] or is it a likely mistake on its own? If it's probably a mistake, we should leave it alone in UP045 and possibly consider a separate rule to catch it.

Optional[None] => Union[None, None], i don't think that makes sense

[chirizxc]

should this be put a separate rule? or should the fix just do from Optional[None] to None?

[ntBre]

should this be put a separate rule? or should the fix just do from Optional[None] to None?

Yeah that's what I was getting at. In my opinion, we should not transform Optional[None] to None because that's assuming a lot about the user's intention when the original code was likely a separate mistake. So the action items would be:

• Avoid offering an UP045 fix here (again because it's unclear what the user wanted)
• Open a separate rule suggestion for catching Optional[None] so we can discuss that generally