sririshindra commented on a change in pull request #28330:
URL: https://github.com/apache/spark/pull/28330#discussion_r416806448
##########
File path:
sql/core/src/test/scala/org/apache/spark/sql/execution/metric/SQLMetricsSuite.scala
##########
@@ -325,34 +325,57 @@ class SQLMetricsSuite extends SharedSparkSession with
SQLMetricsTestUtils
// +- LocalTableScan(nodeId = 7)
Seq((1L, 2L, 5L, false), (2L, 3L, 7L, true)).foreach {
case (nodeId1, nodeId2, nodeId3, enableWholeStage) =>
- val df = df1.join(df2, "key")
- testSparkPlanMetrics(df, 1, Map(
- nodeId1 -> (("ShuffledHashJoin", Map(
- "number of output rows" -> 2L))),
- nodeId2 -> (("Exchange", Map(
- "shuffle records written" -> 2L,
- "records read" -> 2L))),
- nodeId3 -> (("Exchange", Map(
- "shuffle records written" -> 10L,
- "records read" -> 10L)))),
- enableWholeStage
- )
+ val df = df1.join(df2, "key")
+ testSparkPlanMetrics(df, 1, Map(
+ nodeId1 -> (("ShuffledHashJoin", Map(
+ "number of output rows" -> 2L))),
+ nodeId2 -> (("Exchange", Map(
+ "shuffle records written" -> 2L,
+ "records read" -> 2L))),
+ nodeId3 -> (("Exchange", Map(
+ "shuffle records written" -> 10L,
+ "records read" -> 10L)))),
+ enableWholeStage
+ )
}
}
}
+ test("ShuffledHashJoin(left,outer) metrics") {
+ withSQLConf(SQLConf.SHUFFLE_PARTITIONS.key -> "2",
+ SQLConf.PREFER_SORTMERGEJOIN.key -> "false") {
+ val leftDf = Seq((1, "1"), (2, "2")).toDF("key", "value")
+ val rightDf = (1 to 10).map(i => (i, i.toString)).toSeq.toDF("key2",
"value")
+ Seq((0L, "right_outer", leftDf, rightDf, 10L, false),
+ (0L, "left_outer", rightDf, leftDf, 10L, false),
+ (0L, "right_outer", leftDf, rightDf, 10L, true),
+ (0L, "left_outer", rightDf, leftDf, 10L, true),
+ (2L, "left_anti", rightDf, leftDf, 8L, true),
+ (2L, "left_semi", rightDf, leftDf, 2L, true),
+ (1L, "left_anti", rightDf, leftDf, 8L, false),
+ (1L, "left_semi", rightDf, leftDf, 2L, false))
+ .foreach { case (nodeId, joinType, leftDf, rightDf, rows,
enableWholeStage) =>
+ val df = leftDf.hint("shuffle_hash").join(
+ rightDf.hint("shuffle_hash"), $"key" === $"key2", joinType)
+ testSparkPlanMetrics(df, 1, Map(
+ nodeId -> (("ShuffledHashJoin", Map(
+ "number of output rows" -> rows)))),
+ enableWholeStage
+ )
+ }
+ }
+ }
+
test("BroadcastHashJoin(outer) metrics") {
val df1 = Seq((1, "a"), (1, "b"), (4, "c")).toDF("key", "value")
val df2 = Seq((1, "a"), (1, "b"), (2, "c"), (3, "d")).toDF("key2", "value")
// Assume the execution plan is
// ... -> BroadcastHashJoin(nodeId = 0)
- Seq(("left_outer", 0L, 5L, false), ("right_outer", 0L, 6L, false),
- ("left_outer", 1L, 5L, true), ("right_outer", 1L, 6L, true)).foreach {
- case (joinType, nodeId, numRows, enableWholeStage) =>
+ Seq(("left_outer", 0L, 5L, false), ("right_outer", 0L, 6L, false),
("left_outer", 1L, 5L, true),
+ ("right_outer", 1L, 6L, true)).foreach { case (joinType, nodeId,
numRows, enableWholeStage) =>
val df = df1.join(broadcast(df2), $"key" === $"key2", joinType)
testSparkPlanMetrics(df, 2, Map(
- nodeId -> (("BroadcastHashJoin", Map(
- "number of output rows" -> numRows)))),
+ nodeId -> (("BroadcastHashJoin", Map("number of output rows" ->
numRows)))),
Review comment:
Fixed in latest commit
----------------------------------------------------------------
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:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
