joe-ucp commented on PR #8877: URL: https://github.com/apache/arrow-rs/pull/8877#issuecomment-3554917562
> sorry @joe-ucp -- I don't understand what problem this PR is solving > > Can you please remind me what the rationale for these changes are? This one is mostly about alignment + safety, not fixing a known broken case in main. Concretely, this PR: - Rewrites nullif to use the same core `Buffer::bitwise_*` APIs we just added for other bitwise logic, instead of its own ad-hoc bitmap math. - Makes the offset / validity bitmap behavior explicit and pushes that into tests (sliced arrays, nested types, null count consistency). - Ensures nullif always builds results with `offset = 0` and correctly aligned buffers, which is the pattern used elsewhere and makes future refactors much safer. The existing tests still pass, and the new tests mainly serve to lock in those layout assumptions so nullif doesn’t become a special case the next time we touch bitwise code. -- 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]
