andygrove commented on PR #4950: URL: https://github.com/apache/datafusion-comet/pull/4950#issuecomment-5840423031
Pushed 7c8e995fe. @sunchao, the sketch coverage you flagged is now scoped: only the 24 queries that read invalid UTF-8 run without the native scan, and the rest of `hll.sql`, `thetasketch.sql` and `tuplesketch.sql` runs with Comet. Details are on the [original thread](https://github.com/apache/datafusion-comet/pull/4950#discussion_r4109075518). The rest of this update: - Merged `main` twice. `CometNativeScan` conflicted with #5868 (direct Variant projection). Both sides' checks are kept, and the helper's doc no longer claims a requested Variant never reaches the scan. The user guide conflicted with #6168, and the 4.2 rows now say the SQL suite runs nightly. - `4.2.0.diff` picks up #6066, which drops `spark.comet.memoryOverhead`. I did not port #5868's `ResolveDefaultColumnsSuite` edit, because that test passes unchanged on 4.2 with Comet: the scan falls back on the Variant column's existence default. `run-spark-4.2-tests` is still applied, so the Spark 4.2 SQL shards on this push are the check on the patched harness. -- 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]
