gatorsmile commented on a change in pull request #25258: [SPARK-19712][SQL]
Move subquery rewrite to beginning of optimizer
URL: https://github.com/apache/spark/pull/25258#discussion_r310313096
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/Optimizer.scala
##########
@@ -152,10 +152,10 @@ abstract class Optimizer(sessionCatalog: SessionCatalog)
Batch("LocalRelation early", fixedPoint,
ConvertToLocalRelation,
PropagateEmptyRelation) ::
- Batch("Pullup Correlated Expressions", Once,
- PullupCorrelatedPredicates) ::
Batch("Subquery", FixedPoint(1),
- OptimizeSubqueries) ::
+ OptimizeSubqueries,
+ PullupCorrelatedPredicates,
+ RewritePredicateSubquery) ::
Review comment:
Our CBO is based on the pattern matching. We only support InnerLike joins.
Is that possible the pattern matched before but it does not matched after we
convert subquery to join? RewritePredicateSubquery adds LeftSemi or LeftAnti
joins into the original plan.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
