maryannxue commented on a change in pull request #32816:
URL: https://github.com/apache/spark/pull/32816#discussion_r701131503
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/exchange/EnsureRequirements.scala
##########
@@ -254,25 +259,40 @@ case class EnsureRequirements(optimizeOutRepartition:
Boolean = true) extends Ru
}
}
- def apply(plan: SparkPlan): SparkPlan = plan.transformUp {
- case operator @ ShuffleExchangeExec(upper: HashPartitioning, child,
shuffleOrigin)
- if optimizeOutRepartition &&
- (shuffleOrigin == REPARTITION_BY_COL || shuffleOrigin ==
REPARTITION_BY_NUM) =>
- def hasSemanticEqualPartitioning(partitioning: Partitioning): Boolean = {
- partitioning match {
- case lower: HashPartitioning if upper.semanticEquals(lower) => true
- case lower: PartitioningCollection =>
- lower.partitionings.exists(hasSemanticEqualPartitioning)
- case _ => false
+ def apply(plan: SparkPlan): SparkPlan = {
+ val newPlan = plan.transformUp {
+ case operator @ ShuffleExchangeExec(upper: HashPartitioning, child,
shuffleOrigin)
+ if optimizeOutRepartition &&
+ (shuffleOrigin == REPARTITION_BY_COL || shuffleOrigin ==
REPARTITION_BY_NUM) =>
+ def hasSemanticEqualPartitioning(partitioning: Partitioning): Boolean
= {
+ partitioning match {
+ case lower: HashPartitioning if upper.semanticEquals(lower) => true
+ case lower: PartitioningCollection =>
+ lower.partitionings.exists(hasSemanticEqualPartitioning)
+ case _ => false
+ }
+ }
+ if (hasSemanticEqualPartitioning(child.outputPartitioning)) {
+ child
+ } else {
+ operator
}
- }
- if (hasSemanticEqualPartitioning(child.outputPartitioning)) {
- child
- } else {
- operator
- }
- case operator: SparkPlan =>
- ensureDistributionAndOrdering(reorderJoinPredicates(operator))
+ case operator: SparkPlan =>
+ ensureDistributionAndOrdering(reorderJoinPredicates(operator))
+ }
+
+ requiredDistribution match {
Review comment:
Thank you for addressing my earlier comment! But what I meant was an
even better integration by calling:
```
ensureDistributionAndOrdering(children: Seq[SparkPlan],
requiredDistributions: Seq[Distribution], requiredOrderings:
Seq[Seq[SortOrder]])
```
then refactor the original `ensureDistributionAndOrdering(operator:
SparkPlan)` to call the above method too.
--
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]
