ErikBPF commented on PR #5903: URL: https://github.com/apache/datafusion-comet/pull/5903#issuecomment-6387045230
Final head is `72cd19892`, rebased onto `5a50bd6f2`. Still four files, +83/-6. Changed since `8c5700c5a`, after re-running RED on apollo against the delivered revision: - `withDistinctMergeMemoryPressure` is restored in the grouped test. Without it the query passes with `.asNullable` removed, because the default plan's shuffle-boundary cast hides the mismatch. It is there for the plan, not the memory budget. - The spill assertion and the two DataFusion settings stay out; neither stage spills on this fixture at 4 MiB, 2 MiB or 400 KiB. - Comment and quote corrections in the test and the SQL fixtures. Validation on `72cd19892` (apollo, Spark 4.1.3, JDK 17): compile + spotless clean; grouped regression 1/1; `collect_list` 2/2 and `collect_set` 8/8; full `CometAggregateSuite` 163 passed, 0 failed, 2 ignored. With `.asNullable` removed the grouped test and both `collect_list` fixtures fail with `column types must match schema types`; the ungrouped `collect_set` cases pass either way and are coverage, not a gate. PR body updated. -- 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]
