mrhhsg commented on PR #68279:
URL: https://github.com/apache/doris/pull/68279#issuecomment-6264019215
<!-- autopilot-response -->
Re-checked the full review history at HEAD
`ac69ac763059df4b8d8b0d41cf1878021546ad40` (rebased onto `e0bf4874070b`).
**Previous inline findings — fixed and resolved:**
- Sparse-accounting overcount in `segment_iterator.cpp`: condition ranges
are again built for the whole segment, and the Bloom filter / zone map /
expression zone map stages charge only rows still set in `_row_bitmap`
(`_count_alive_rows`). `DisjointKeyRangesOnlyCountAliveRows` covers disjoint
key ranges (`{1}`, `{5000}`) and asserts `RowsStatsFiltered == 1`.
- Trailing `DROP TABLE` in `point_lookup_zone_map_profile.groovy`: removed;
the tables are only dropped before creation.
**Other top-level comments:** the `hello-stephen` entries are the
contribution template, TPC-H / TPC-DS / ClickBench benchmark payloads and a
coverage report; the remaining entries are review/build commands and CI failure
notices. None contains a code finding.
The current `CHANGES_REQUESTED` decision comes from the `github-actions`
review of 2026-09-20; that reviewer's later review at an earlier head reported
no remaining blocking finding. The only follow-up suggestion there (extra
disjoint/rejecting Bloom-filter and differing page-layout cases) is
non-blocking; the point-lookup fixture has since been rebuilt as a real
single-segment table and the variant Bloom probes now run without
inverted-index pruning.
**Validation at this HEAD (fresh local runs):**
- `query_profile.point_lookup_zone_map_profile` — 1 suite, 0 failures
- `variant_p0/predefine.test_variant_predefine_types_with_indexes_profile` —
1 suite, 0 failures
Could a maintainer take another look and clear the change request if the
remaining concerns are satisfied?
--
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]
