cloud-fan commented on a change in pull request #31178:
URL: https://github.com/apache/spark/pull/31178#discussion_r560962193
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/window/WindowFunctionFrame.scala
##########
@@ -305,12 +306,22 @@ class UnboundedOffsetWindowFunctionFrame(
inputIterator = input.generateIterator()
// drain the first few rows if offset is larger than one
inputIndex = 0
- while (inputIndex < offset - 1) {
- if (inputIterator.hasNext) inputIterator.next()
- inputIndex += 1
+ if (ignoreNulls) {
+ findNextRowWithNonNullInput()
+ if (nextSelectedRow == EmptyRow) {
+ // Use default values since the offset row whose input value is not
null does not exist.
+ fillDefaultValue(EmptyRow)
+ } else {
+ projection(nextSelectedRow)
+ }
+ } else {
+ while (inputIndex < offset - 1) {
Review comment:
Can we follow
https://github.com/apache/spark/pull/31178/files#diff-09c876759e9b45e198238f7244eba48900058896287bf371acb27de9e25032b8R367
to write the loop?
----------------------------------------------------------------
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]
