adiranjan25 commented on issue #18723: URL: https://github.com/apache/iceberg/issues/18723#issuecomment-6674910697
Hi @Rishavbhattarai, Thanks for the detailed reproducer and for highlighting the distinction between expression equality and the ordering of stored lower/upper bounds. I'd like to help work on a fix for this issue. My plan is to reproduce the behavior in the Java API tests and add regression coverage for `float` and `double` across row evaluation and metrics/manifest pruning, particularly for equality, `IN`, and range predicates involving `-0.0` and `+0.0`. Changing `Comparators` globally could affect the spec-required ordering of stored bounds, so I'll investigate a predicate-specific solution rather than assuming a global comparator change is safe. Before implementing, could a maintainer confirm whether the preferred fix belongs in predicate evaluation/binding or another shared comparison layer? Also, @Rishavbhattarai, please let me know if you're already preparing a patch so we can coordinate and avoid duplicating work. Thanks! -- 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]
