Github user cloud-fan commented on a diff in the pull request:
https://github.com/apache/spark/pull/12719#discussion_r62791648
--- Diff:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/Optimizer.scala
---
@@ -650,15 +646,15 @@ object FoldablePropagation extends Rule[LogicalPlan] {
if (foldableExprSet.isEmpty) {
plan
- } else if (hasCommandOrUnion(plan)) {
- replaceInSubplan(plan)
} else {
val foldableMap =
AttributeMap(foldableExprSet.toSeq.map(_.asInstanceOf[Alias])
.map(a => (a.toAttribute, a.child)))
- CleanupAliases(plan.transformAllExpressions {
- case a : AttributeReference if foldableMap.contains(a) =>
- Alias(foldableMap.getOrElse(a.toAttribute, a), a.name)()
+ CleanupAliases(plan.transformUp {
+ case p: LogicalPlan if !hasCommandOrUnion(p) =>
p.transformAllExpressions {
+ case a: AttributeReference if foldableMap.contains(a) =>
+ Alias(foldableMap.getOrElse(a.toAttribute, a), a.name)()
+ }
--- End diff --
I'd like to write:
```
var stop = false
plan.transformUp {
case u: Union =>
stop = true
u
case c: Command =>
stop = true
c
case p if !stop => p.transformExpressions {
case a: AttributeReference if foldableMap.contains(a) =>
foldableMap(a) // foldableMap's value shoud be Alias, not its child
}
}
```
if earlier sibling is `Union`, this is suboptimal, but still safe(it's ok
to not apply this optimization for some corner cases)
---
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]
