haohuaijin commented on PR #10852: URL: https://github.com/apache/arrow-rs/pull/10852#issuecomment-5441079656
Thanks @hhhizzz, look good to me. This is a good improvement for the first-predicate path. One remaining limitation is that when the selection stays selector-backed, later predicates still fully materialize selectors via `from_filters`. A follow-up can extend capped construction to those predicates(becuase when page index are available and page pruning is effective, datafusion usually provides an existing `RowSelection`). for the regression in Q14/Q50, maybe we can run the benchmark once again to check if this a noisy, because both Q14/Q50 are join-dominated, the pr should not affect so much. -- 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]
