shenyu0127 commented on PR #10914: URL: https://github.com/apache/pinot/pull/10914#issuecomment-1601752054
> > Should we consider doing the same trick as primitive data types by tracking a `_hasNull` flag to avoid the extra null handling? That way the behavior is consistent across all data types, and the overall cost should be lower > > The `_hasNull` flag is a not a clean solution and we want to avoid if possible. > > * The `_hasNull` flag needs to be used for primitive data types because the primitive data type sets/priority_queues cannot store NULL. > * The `_hasNull` flag has an issue (if there is NULL, it will always be present in the executor result regardless of the ordering) that needs to be fixed in the upstream handling (DistinctTable). Had an offline chat with @Jackie-Jiang and we aligned to use `_hasNull` mainly for performance (avoid additional null checks in the comparators). -- 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: commits-unsubscr...@pinot.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@pinot.apache.org For additional commands, e-mail: commits-h...@pinot.apache.org
