Github user nongli commented on a diff in the pull request:
https://github.com/apache/spark/pull/9959#discussion_r45834786
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/Optimizer.scala
---
@@ -690,7 +690,15 @@ object PushPredicateThroughAggregate extends
Rule[LogicalPlan] with PredicateHel
def apply(plan: LogicalPlan): LogicalPlan = plan transform {
case filter @ Filter(condition,
aggregate @ Aggregate(groupingExpressions, aggregateExpressions,
grandChild)) =>
- val (pushDown, stayUp) =
splitConjunctivePredicates(condition).partition {
+
+ // Create a map of Alias for grouping keys or literals
+ val aliasMap = AttributeMap(aggregateExpressions.collect {
+ case a: Alias if groupingExpressions.contains(a.child) ||
a.child.foldable =>
+ (a.toAttribute, a.child)
+ })
+ val newCond = PushPredicateThroughProject.replaceAlias(condition,
aliasMap)
+
+ val (pushDown, stayUp) =
splitConjunctivePredicates(newCond).partition {
conjunct => conjunct.references subsetOf
AttributeSet(groupingExpressions)
--- End diff --
Can you see if you can make AttributeSet deal with this. It is more general
and more consistent with what I understand the intent of AttributeSet is.
The comment in AttributeSet is clear that it wants AttributeReferences
which Aliases are not. Is that possible?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
