etseidl commented on PR #11262: URL: https://github.com/apache/arrow-rs/pull/11262#issuecomment-6020868360
> Thanks for spotting! I added [645d147](https://github.com/apache/arrow-rs/commit/645d14750c771b30d6840482cec45e6f140267f0) which implements the suggested fix, but happy to drop it if you'd rather that gets merged separately. Thanks @HippoBaro, I think it's fine to roll that up with this PR. I need to look at the benches...it's a bit surprising to me that none of the reader benches are showing the dramatic slowdown that the roundtrip plain case does. Guess I need to do more profiling. I wonder if for the nullable-but-no-nulls case we could happy path a check of the encoded size and see if it matches the expected size and skip the def level decoding. Maybe a cheat and see if the def levels are a single run of max_def_level 🤷 -- 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]
