Github user maropu commented on a diff in the pull request:
https://github.com/apache/spark/pull/19781#discussion_r152455868
--- Diff: sql/core/src/test/scala/org/apache/spark/sql/JoinSuite.scala ---
@@ -857,4 +857,29 @@ class JoinSuite extends QueryTest with
SharedSQLContext {
joinQueries.foreach(assertJoinOrdering)
}
+
+ test("SPARK-22445 Respect stream-side child's needCopyResult in
BroadcastHashJoin") {
+ val df1 = Seq((2, 3), (2, 5), (2, 2), (3, 8), (2, 1)).toDF("k", "v1")
+ val df2 = Seq((2, 8), (3, 7), (3, 4), (1, 2)).toDF("k", "v2")
+ val df3 = Seq((1, 1), (3, 2), (4, 3), (5, 1)).toDF("k", "v3")
+
+ withSQLConf(
+ SQLConf.AUTO_BROADCASTJOIN_THRESHOLD.key -> "-1",
+ SQLConf.JOIN_REORDER_ENABLED.key -> "false") {
+ val df = df1.join(df2, "k").join(functions.broadcast(df3), "k")
+ val plan = df.queryExecution.sparkPlan
+
+ // Check if `needCopyResult` in `BroadcastHashJoin` is correct when
smj->bhj
--- End diff --
`q6` also failed when smj->bhj
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
