BUG: Fix groupby().any() behavior for timedelta columns with all null valuesfix issue #59712

[40gilad]

This pull request addresses a bug in the groupby().any() method when applied to DataFrames with timedelta columns where all values are null (NaT). Previously, the method incorrectly returned True for groups with all null values in the timedelta column.

Changes Made:

• Updated the _call_cython_op method to correctly handle timedelta columns when computing the any() aggregation.
• Added a new test case to verify that groupby().any() returns False for groups where all timedelta values are null.

Issue Reference:

• See issue #59712 for more details about the bug report.

This fix ensures that the groupby().any() method behaves consistently for timedelta columns, aligning with the behavior observed for other data types.

[rhshadrach]

Thanks for the PR!

• Added a new test case to verify that groupby().any() returns False for groups where all timedelta values are null.

Not seeing a new test. Perhaps you didn't commit it?

[40gilad]

Thanks for the PR!

• Added a new test case to verify that groupby().any() returns False for groups where all timedelta values are null.

Not seeing a new test. Perhaps you didn't commit it?

Yeah. sorry. i did now

[rhshadrach]

@40gilad - in case you didn't notice, tests on the CI are failing. Let me know if you could use an assistance.

[KevsterAmp]

@rhshadrach - can I take over this issue? seems like this PR has gone stale

[rhshadrach]

Agreed at @KevsterAmp - but there was already another open PR on this (#59782). I'd suggest giving that PR a few weeks, and if it goes stale, you're next up. Sound reasonable?

[KevsterAmp]

Sounds good to me. Thanks

[github-actions]

This pull request is stale because it has been open for thirty days with no activity. Please update and respond to this comment if you're still interested in working on this.

[rhshadrach]

The other PR was merged. Closing.