uros-b commented on code in PR #56770:
URL: https://github.com/apache/spark/pull/56770#discussion_r3473868614
##########
sql/core/src/test/scala/org/apache/spark/sql/SortMergeAsOfJoinSuite.scala:
##########
@@ -670,4 +670,60 @@ class SortMergeAsOfJoinSuite extends QueryTest
)
}
}
+
+ test("forward join - spill to disk") {
+ withSQLConf(
+ SQLConf.SORT_MERGE_JOIN_EXEC_BUFFER_IN_MEMORY_THRESHOLD.key -> "1",
+ SQLConf.SORT_MERGE_JOIN_EXEC_BUFFER_SPILL_THRESHOLD.key -> "1") {
+ val (df1, df2) = prepareForAsOfJoin()
+ checkAnswer(
+ df1.joinAsOf(
+ df2, df1.col("a"), df2.col("a"), usingColumns = Seq.empty,
+ joinType = "inner", tolerance = null,
+ allowExactMatches = true, direction = "forward"),
+ Seq(
+ Row(1, "x", "a", 1, "v", 1),
+ Row(5, "y", "b", 6, "y", 6)
+ )
+ )
+ }
+ }
+
+ test("nearest join - spill to disk") {
+ withSQLConf(
+ SQLConf.SORT_MERGE_JOIN_EXEC_BUFFER_IN_MEMORY_THRESHOLD.key -> "1",
+ SQLConf.SORT_MERGE_JOIN_EXEC_BUFFER_SPILL_THRESHOLD.key -> "1") {
+ val (df1, df2) = prepareForAsOfJoin()
+ checkAnswer(
+ df1.joinAsOf(
+ df2, df1.col("a"), df2.col("a"), usingColumns = Seq.empty,
+ joinType = "inner", tolerance = null,
+ allowExactMatches = true, direction = "nearest"),
+ Seq(
+ Row(1, "x", "a", 1, "v", 1),
+ Row(5, "y", "b", 6, "y", 6),
+ Row(10, "z", "c", 7, "z", 7)
+ )
+ )
+ }
+ }
+
+ test("left outer join - spill to disk") {
Review Comment:
Uses usingColumns = Seq("b"), yielding exactly one right row per equi-key
group; with the in-memory threshold set to 1, that single row stays in memory
and the spill path is never exercised. The test validates the left-outer
null-fill output (correct) but contradicts its "spill to disk" name. Fix: use
usingColumns = Seq.empty or add a second right row sharing the equi-key to
breach the threshold.
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
