andygrove commented on PR #6835: URL: https://github.com/apache/datafusion-comet/pull/6835#issuecomment-6113962214
Agreed, please go ahead and narrow #6810 to the NOT IN case so it can land first. Keep field 11 there and I'll move `streamed_sort_orders` to 12 when I rebase this onto it. I ran your test against this branch after merging main and got the same result, 75 rows against Spark's 37, so this shouldn't go in before the sort does. Once #6810 is in I'll rebase and add your NOT IN test, an aggregate on the key above the join, with the `SortedInputExec` comment saying that the declared order is visible above the join, a RightOuter with BuildLeft case, multiset row comparisons instead of `toSet`, and planner tests for `with_declared_order` and the fallback. I'll get a Spark SQL run before queueing. -- 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]
