sunchao commented on PR #6427: URL: https://github.com/apache/datafusion-comet/pull/6427#issuecomment-6553033467
@viirya Thanks for pointing out the gap in the plan assertions. Addressed in 0a7550dac641. Both Parquet queries now check the whole executed plan, formatted EXPLAIN, and every SQL UI graph node description. They also check the size of the full DOT graph. This covers older Spark versions that omit descriptions from DOT. I renamed the suite to `CometExpressionDisplaySuite` in both CI workflows and removed the literal lookup, the constructor-value assertions, and the scalar-subquery test. The throwing-literal test still catches an attempt to format or evaluate the payload. I also documented the partition-pruning case next to the alias limitation. When the Bloom probe references only partition columns, the scan’s `PartitionFilters` can print the full literal in formatted EXPLAIN and the SQL UI. All five tests passed locally on Spark 3.5.9 and 4.1.3 against GitHub’s proposed merge `b46d452f1650` with its matching native library. The PR head also passed Spark 4.1.3 test compilation, Spotless, and both workflow suite-registration checks. -- 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]
