sunchao commented on PR #11276: URL: https://github.com/apache/arrow-rs/pull/11276#issuecomment-5898495373
@etseidl Thanks for pointing that out. Yes, looks like #11157 already covers the Skip/Skip case. The remaining benefit here is avoiding repeated range discovery: we calculate the range to request index bytes, then calculate it again when decoding resumes after those bytes arrive. Keeping the pending range avoids that second walk over the decoded metadata. I’ll narrow this PR to that caching change and rebase after #11157 lands, including invalidation when the new selection masks change. I’ll also rerun the benchmarks against that baseline. -- 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]
