Uh oh!
There was an error while loading. Please reload this page.
[fix](case) fix insert_group_commit_into_max_filter_ratio - #64402
Conversation
hello-stephen
commented
Jun 11, 2026
Thank you for your contribution to Apache Doris. Please clearly describe your PR:
|
mymeiyi
commented
Jun 11, 2026
run buildall |
There was a problem hiding this comment.
Pull request overview
This PR updates the regression test insert_group_commit_into_max_filter_ratio to use enable_strict_cast in the JDBC insert section and adjusts the expected minimum row count for the async visibility wait.
Changes:
- Explicitly sets
enable_strict_castin the JDBCconnectblock (toggle false → true → false around a cast-failure insert). - Replaces a strict-insert toggle (
enable_insert_strict = true) withenable_strict_cast = truefor the cast failure scenario. - Lowers the
get_row_count_with_retry(...)threshold from 8 to 7.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
mymeiyi
commented
Jun 11, 2026
/review |
There was a problem hiding this comment.
Review summary for PR 64402
Overall opinion: no additional blocking issue found beyond the existing inline threads already opened on this PR. The change is a focused regression-test fix: it makes the cast-failure case use enable_strict_cast and aligns the async wait threshold with the seven rows recorded in insert_group_commit_into_max_filter_ratio.out.
Critical checkpoint conclusions:
- Goal/test proof: The goal appears to be fixing this regression case strict-cast/session-variable behavior and expected async row count. The changed expected count is consistent with the generated .out rows: 1, 2, 3, 4, 5, 7, and 9. I did not run the regression suite locally; GitHub-side checks visible now pass except TeamCity compile is still queued.
- Scope: Small and focused; one regression test file, no production path changes.
- Concurrency, lifecycle, config, compatibility, persistence, protocol: Not applicable; no runtime code, config item, storage format, FE-BE protocol, or persistence change is introduced.
- Parallel paths: Stream-load and http-stream sections in the same suite remain unchanged, and their row-count checks match the existing expected output.
- Test correctness: Existing review threads already cover two robustness gaps: the strict-cast try/catch should fail when no exception occurs, and get_row_count_with_retry should assert after retries. I did not add duplicate inline comments for those already-known issues. I found no separate new test-output mismatch.
- Observability/performance: Not applicable for this test-only change; no meaningful runtime observability or performance impact.
- User focus: review_focus.txt contains no additional user-provided focus points.
PR approved by at least one committer and no changes requested. |
PR approved by anyone and no changes requested. |
Uh oh!
There was an error while loading. Please reload this page.
No description provided.