LuciferYang commented on a change in pull request #29434:
URL: https://github.com/apache/spark/pull/29434#discussion_r474761400
##########
File path:
sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/optimizer/StarJoinCostBasedReorderSuite.scala
##########
@@ -350,7 +350,7 @@ class StarJoinCostBasedReorderSuite extends PlanTest with
StatsEstimationTestBas
Some(nameToAttr("d3_c2") === nameToAttr("t1_c1")))
.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): _*)
+ .select(outputsOf(d1, t3, t4, f1, d3, d2, t5, t6, t1, t2): _*)
Review comment:
@srowen @cloud-fan From the test results above, the output of the
current `CostBasedReorder` rule is already non-deterministic when there are
multiple candidate plans with the same cost and the input order is different,
but I think both results are equally correct.
If you agree, I'll create a new JIRA to describe this problem and leave a
TODO to make the rule deterministic later.
----------------------------------------------------------------
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]
