Copilot commented on code in PR #13050:
URL: https://github.com/apache/gluten/pull/13050#discussion_r4585306564
##########
backends-velox/src/test/scala/org/apache/gluten/functions/WindowFunctionsValidateSuite.scala:
##########
@@ -92,4 +105,118 @@ class WindowFunctionsValidateSuite extends
FunctionsValidateSuite {
}
}
}
+
+ test("window partition/order by TIMESTAMP_NTZ goes native") {
+ withTable("ntz_window_data") {
+ createNtzWindowTable()
+
+ runQueryAndCompare(
+ "select k, ts, rank() over (partition by k order by ts) as rnk " +
+ "from ntz_window_data") {
+ checkGlutenPlan[WindowExecTransformer]
+ }
Review Comment:
All added window queries partition by integer `k`, so none exercises the
newly enabled NTZ partition-key path. The suite also sets
`spark.sql.shuffle.partitions` to `1`, leaving multi-partition hashing
untested. Add a Spark-result comparison using `PARTITION BY ts`, repeated and
null keys, and multiple shuffle partitions with adaptive coalescing disabled,
while asserting native window execution.
--
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]