kita-renji commented on PR #25742: URL: https://github.com/apache/datafusion/pull/25742#issuecomment-5843888219
Thanks for the quick turnaround. I rebuilt at 84ef8f7 and re-ran the checks. Both planning regressions are fixed: `ORDER BY t.x` works for INTERSECT ALL and EXCEPT ALL, including when both sides read the same table, and the `t1.x, t2.x` case returns (1, 2) again. `t1.x, t2.x ... INTERSECT ALL SELECT t3.x, t4.x ...` also plans now, which fails on main with "Ambiguous reference to unqualified field x". Columns aliased to the synthetic row number names don't collide either. The DuckDB differential run (5300 queries) has no error mismatches, and the only result mismatches left are the -0.0 cases tracked in #25766. The slt files, `datafusion-substrait` and the core set operation tests pass locally. The docs, the upgrade guide note and the Substrait cleanup look good to me. I agree the plan shape question should wait for a maintainer. From my side this is good to go once someone weighs in on #12955. -- 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]
