andygrove commented on issue #2067: URL: https://github.com/apache/datafusion-ballista/issues/2067#issuecomment-4994615135
Fixed in #2070. Root cause: `restrict_scan_to_partition` applied the stage's partition number directly to every scan, but a `UnionExec` concatenates its children's partitions — stage partition `N + k` is the right child's local partition `k`. For partitions past a scan's group count the guard bailed out and left that scan **unrestricted**, free to read its whole table. The fix walks the plan from the stage root and mirrors `UnionExec::execute`'s own partition arithmetic, pinning each scan to the group its local partition reads. The minimal case is now correct at every partition count tried (4, 5, 6, 8, 16, 20, 32), and q2/q5 verify in their original form and join the gate. -- 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]
