tdas commented on a change in pull request #26162: [SPARK-29438][SS] Use
partition ID of StateStoreAwareZipPartitionsRDD for determining partition ID of
state store in stream-stream join
URL: https://github.com/apache/spark/pull/26162#discussion_r368339416
##########
File path:
sql/core/src/test/scala/org/apache/spark/sql/streaming/StreamingJoinSuite.scala
##########
@@ -933,5 +934,57 @@ class StreamingOuterJoinSuite extends StreamTest with
StateStoreMetricsTest with
assert(e.getMessage.toLowerCase(Locale.ROOT)
.contains("the query is using stream-stream outer join with state format
version 1"))
}
-}
+ test("SPARK-29438: ensure UNION doesn't lead stream-stream join to use
shifted partition IDs") {
+ def constructUnionDf(desiredPartitionsForInput1: Int)
+ : (MemoryStream[Int], MemoryStream[Int], MemoryStream[Int], DataFrame)
= {
+ val input1 = MemoryStream[Int](desiredPartitionsForInput1)
+ val df1 = input1.toDF
+ .select(
+ 'value as "key",
+ 'value as "leftValue",
+ 'value as "rightValue")
+ val (input2, df2) = setupStream("left", 2)
+ val (input3, df3) = setupStream("right", 3)
+
+ val joined = df2
+ .join(df3,
+ df2("key") === df3("key") && df2("leftTime") === df3("rightTime"),
+ "inner")
+ .select(df2("key"), 'leftValue, 'rightValue)
+
+ (input1, input2, input3, df1.union(joined))
+ }
+
+ withTempDir { tempDir =>
+ val (input1, input2, input3, unionDf) = constructUnionDf(2)
+
+ testStream(unionDf)(
+ StartStream(checkpointLocation = tempDir.getAbsolutePath),
+ AddData(input1, 11, 12, 13),
+ MultiAddData(input2, 11, 12, 13, 14, 15)(input3, 13, 14, 15, 16, 17),
Review comment:
Why is the adding of data split between AddData and MultiAddData. I think
all the data should be added together with MultiAddData. The point of
MultiAddData was to ensure that the micro-batch scheduler is locked while
adding the data to all the streams so that all the streams pick up the new data
together. Without this locking, different streams can pick the new data in
different batches causing test flakiness. I suggest extended MultiAddData to
take an arbitrary number of inputs (e.g. `def apply[A](inputs:
Map(MemoryStream[A], Seq[A]))` or `def apply[A](inputs: (MemoryStream[A],
Seq[A])*)` and use that.
----------------------------------------------------------------
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]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
