zhuqi-lucas commented on PR #23696: URL: https://github.com/apache/datafusion/pull/23696#issuecomment-5381141892
@adriangb done — this PR is now reshaped per your proposal: - Rebased on #24509 (PR 1) and #24572 (PR 3), both merged. - PR 2 folded in here: `build_row_filter` is reimplemented on top of `prebuild_row_filter_candidates` + `row_filter_from_prebuilt`, so there's a single conjunct-split/order/metric path. - Lazy metric registration: no public field, no semver flag, and the 41 `row_filter_skipped_fully_matched=0` baseline entries are gone. - `test_input_file_name_projection` move reverted. - Single-vec: deferred — alignment is now only two adjacent permutations + a `debug_assert`, and the type change would churn ~40 main-side test assertions. Happy to do as a follow-up. Net: 11 files / +844 −207 → 8 files / +636 −170, with zero churn in unrelated slt baselines. PR description updated. Ready for another look. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
