Copilot commented on code in PR #58776:
URL: https://github.com/apache/spark/pull/58776#discussion_r4005885073
##########
sql/core/src/test/scala/org/apache/spark/sql/SortMergeAsOfJoinSuite.scala:
##########
@@ -647,7 +661,7 @@ class SortMergeAsOfJoinSuite extends QueryTest
SQLConf.SORT_MERGE_JOIN_EXEC_BUFFER_SPILL_THRESHOLD.key -> "1") {
val (df1, df2) = prepareForAsOfJoin()
// No equi-key (bufferAllRight path)
- checkAnswer(
+ checkAnswerAndSpill(
Review Comment:
Could this also wrap the equi-key spill cases in the forward/nearest tests?
The new helper only checks the `bufferAllRight` path (`usingColumns =
Seq.empty`); the later `bufferRightGroup` cases use multiple rows with the same
key and the forced thresholds but still call `checkAnswer`, so a regression in
repeated-group `spillSize` accounting through `clear()`/`close()` would remain
undetected. Applying the same assertion there would cover the second spill path.
--
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]
