AngersZhuuuu commented on a change in pull request #31402:
URL: https://github.com/apache/spark/pull/31402#discussion_r568480394
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala
##########
@@ -3170,7 +3170,9 @@ class Analyzer(override val catalogManager:
CatalogManager)
WindowExpression(wf, s.copy(frameSpecification = wf.frame))
case we @ WindowExpression(e, s @ WindowSpecDefinition(_, o,
UnspecifiedFrame))
if e.resolved =>
- val frame = if (o.nonEmpty) {
+ val frame = if (e.isInstanceOf[RankLike]) {
+ SpecifiedWindowFrame(RowFrame, UnboundedPreceding, CurrentRow)
Review comment:
> what happens if order spec is present? do we force to use row frame?
In PostgresSQL, for all AggregationWindowFunction, it has a default frame,
but user can write all kind of frame and finally it just use default frame. So
in current PR, We can replace user-defined frame to default frame in
`ResolveWindowFrame`, how about current change.
----------------------------------------------------------------
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
