LuciferYang commented on a change in pull request #29434:
URL: https://github.com/apache/spark/pull/29434#discussion_r471064312
##########
File path:
sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/StarJoinCostBasedReorderSuite.scala
##########
@@ -351,6 +351,18 @@ class StarJoinCostBasedReorderSuite extends PlanTest with
StatsEstimationTestBas
.join(t5.join(t6, Inner, Some(nameToAttr("t5_c2") ===
nameToAttr("t6_c2"))), Inner,
Some(nameToAttr("d2_c2") === nameToAttr("t5_c1")))
.select(outputsOf(d1, t3, t4, f1, d2, t5, t6, d3, t1, t2): _*)
+ } else {
Review comment:
Maybe we can add a Manual sort like
```
while (k <= lev - k) {
val oneSideCandidates =
existingLevels(k).values.toSeq.sortBy(_.itemIds.head)
for (i <- oneSideCandidates.indices) {
val oneSidePlan = oneSideCandidates(i)
val otherSideCandidates = if (k == lev - k) {
// Both sides of a join are at the same level, no need to repeat
for previous ones.
oneSideCandidates.drop(i)
} else {
existingLevels(lev - k).values.toSeq.sortBy(_.itemIds.head)
}
..
```
to ensure the candidates plan input order, but the results tend to be 2.13
----------------------------------------------------------------
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]
