Github user liancheng commented on a diff in the pull request:
https://github.com/apache/spark/pull/12040#discussion_r57913846
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/Window.scala ---
@@ -937,6 +944,14 @@ private[execution] object AggregateProcessor {
imperatives.toArray,
trackPartitionSize)
}
+
+ // Rewrites the window partition size attribute in all
`SizeBasedWindowFunction`s using executor
+ // side `SizeBasedWindowFunction.n`.
+ def rewriteWindowPartitionSize(expression: Expression): Expression = {
+ expression.transformDown {
+ case f: SizeBasedWindowFunction =>
f.withPartitionSize(SizeBasedWindowFunction.n)
--- End diff --
I'd prefer not to do so since it doesn't bring much performance gain, yet
depends on a not so obvious implicit assumption (partition size attribute
ordinal is always 0) and makes the code harder to grasp.
---
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]
