nikolauspschuetz opened a new pull request, #57522:
URL: https://github.com/apache/spark/pull/57522
### What changes were proposed in this pull request?
`DataFrame.sort` / `orderBy` / `sortWithinPartitions` raise
`NOT_EXPECTED_TYPE` for an invalid `ascending` argument with `expected_type` =
`"Column, int or str"`. But `ascending` accepts `bool`, `int`, or `list` —
never `Column` or `str`. The literal was copied from the column-argument error
a few lines above in `_preapare_cols_for_sort`. This corrects it to `"bool, int
or list"`.
### Why are the changes needed?
The message names types `ascending` never accepts (`Column`, `str`) and
omits the ones it does (`bool`, `list`), misleading users about how to fix the
call. The `ascending` parameter is documented as `bool or list`.
### Does this PR introduce _any_ user-facing change?
Yes, the error message changes.
Before:
```
[NOT_EXPECTED_TYPE] Argument `ascending` should be Column, int or str, got
str.
```
After:
```
[NOT_EXPECTED_TYPE] Argument `ascending` should be bool, int or list, got
str.
```
### How was this patch tested?
Added `test_sort_ascending_invalid_type` in
`python/pyspark/sql/tests/test_dataframe.py` asserting the corrected
`expected_type`. Verified against a released `pyspark` session: `df.sort("age",
ascending="asc")` reports `bool, int or list` with the fix (previously `Column,
int or str`); valid `ascending` values (`False`, `[True]`) are unaffected.
### Was this patch authored or co-authored using generative AI tooling?
Generated-by: Claude Code (Opus 4.8)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
