jorgecarleitao edited a comment on pull request #8630: URL: https://github.com/apache/arrow/pull/8630#issuecomment-727175118
I have rebased against latest master and further improved the code. IMO this is now ready to review: benchmarks are now better for all cases where we pre-compute the filter (1.1-3x) except for 1 case (very low selectivity) where it is 1.5x worse. For a single filter case, I know how to improve it: we need to not pre-compute the filter and instead use an iterator. I can continue to investigate this, but IMO this is an improvement to the code base and performance. ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org