kumarUjjawal commented on code in PR #24050:
URL: https://github.com/apache/datafusion/pull/24050#discussion_r3698227662
##########
benchmarks/queries/h2o/window.sql:
##########
@@ -196,3 +196,53 @@ SELECT pk, largest_v2 FROM (
RANK() OVER (PARTITION BY (id3 % 100000) ORDER BY v2 DESC) AS rk_v2
FROM large WHERE v2 IS NOT NULL
) sub_query WHERE rk_v2 <= 2;
+
+-- Window Top-N (DENSE_RANK top-2 per partition, ~100 partitions)
+-- The DENSE_RANK queries below mirror the RANK cardinality sweep above.
+-- DENSE_RANK semantics keep every row whose ORDER BY value is among the
+-- K distinct-smallest values in the partition, so total kept per partition
Review Comment:
nit: should we change the comment to mean greatest value?
--
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]
