XiaoHongbo-Hope opened a new pull request, #9217:
URL: https://github.com/apache/paimon/pull/9217
### Purpose
Return a no-op result when a transformed Ray Dataset contains no rows.
Ray drops the schema of an empty shuffle input, so grouping an empty update
by
`_FIRST_ROW_ID` currently fails with `ArrowInvalid: Invalid sort key column`.
This change adds an internal sentinel to preserve the shuffle schema and
removes
it before writing. It does not pre-count or materialize the Dataset, so the
upstream computation is not executed twice.
### Tests
- Reproduced the failure with a real Ray Dataset filtered to zero rows.
- Verified the update returns `{"num_updated": 0}` and leaves the table
unchanged.
- Ran the row-id update and shared merge regression tests.
--
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]
