JingsongLi commented on PR #8076: URL: https://github.com/apache/paimon/pull/8076#issuecomment-4610842740
LGTM. I went through the condition rewrite/evaluation path, matched and not-matched filtering, blob column validation, empty-batch handling, and the new test coverage. The current implementation looks reasonable to me. Two minor non-blocking suggestions: 1. It may be worth adding one test for duplicate source rows matching the same target row when the matched condition filters the duplicates down to zero or one output row. This would make the intended cardinality semantics explicit for conditional merge. 2. Column validation is currently mostly delegated to DataFusion at execution time. That is fine, but a small driver-side validation/parse step could make error reporting more predictable, especially for empty input batches. -- 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]
