joshrosen-stripe commented on a change in pull request #25164: [SPARK-28375]
Prevent the PullupCorrelatedPredicates optimizer rule from removing predicates
if run multiple times
URL: https://github.com/apache/spark/pull/25164#discussion_r304121011
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/subquery.scala
##########
@@ -275,13 +275,16 @@ object PullupCorrelatedPredicates extends
Rule[LogicalPlan] with PredicateHelper
plan transformExpressions {
case ScalarSubquery(sub, children, exprId) if children.nonEmpty =>
val (newPlan, newCond) = pullOutCorrelatedPredicates(sub, outerPlans)
- ScalarSubquery(newPlan, newCond, exprId)
+ val conds = newCond ++ children.filter(_.isInstanceOf[Predicate])
Review comment:
@gatorsmile
> We want to ensure all these optimizer rules can be run multiple times.
This is critical for adaptive query execution.
Quick clarification question: does Spark 2.4.x perform double-optimization
when adaptive execution is used? Or is that only a characteristic of the new /
expanded adaptive execution features in Spark 3.x? I ask in order to help
determine whether this patch should be a correctness backport for 2.4.x.
----------------------------------------------------------------
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]
