Pandas version checks
-
[x] I have checked that this issue has not already been reported.
-
[x] I have confirmed this bug exists on the latest version of pandas.
-
[x] I have confirmed this bug exists on the main branch of pandas.
Reproducible Example
The failure is visible in GitHub Actions.
Steps
1. Push any branch that is up-to-date with pandas-dev/main
, or open a fresh PR.
2. Observe workflow Code Checks / Docstring validation, typing, and other manual pre-commit hooks.
3. The job stops in step Run docstring validation with the traceback below.
Example failing runs (public logs)
- https://github.com/pandas-dev/pandas/actions/runs/15921431436 ← PR #61718
- https://github.com/pandas-dev/pandas/actions/runs/15886481522 ← another PR on latest main
File ".../site-packages/numpydoc/validate.py", line 234, in name
return ".".join([self.obj.__module__, self.obj.__name__])
AttributeError: 'getset_descriptor' object has no attribute '__module__'. Did you mean: '__reduce__'?
Issue Description
- The docstring-validation job crashes before any pandas code is executed, so all current PRs fail.
- The stack trace originates inside numpydoc/validate.py; no pandas files are involved.
Expected Behavior
The docstring-validation step should complete without errors, allowing the entire CI workflow to finish green.
Installed Versions
Comment From: evgmosme
If maintainers confirm this as a valid issue that needs fixing, I'm happy to submit a PR with the fix.
Comment From: evgmosme
Superseded by PR #61725, which pins numpydoc==1.8.0 and fixes the CI failure. Closing in favour of that solution.