SEPURI-SAI-KRISHNA commented on PR #12595: URL: https://github.com/apache/seatunnel/pull/12595#issuecomment-6110495304
Thanks @DanielLeens, and thanks for re-reading the whole PR rather than just the new commit. **Issue 1 taken.** You are right that a source comment should stand on its own: "a reviewer" means nothing to someone opening the file in two years, and the attribution belongs in this thread rather than in the code. It now reads as fact, naming the builds and the bytecode on each side without saying who ran what. While in there I re-audited the rest of that Javadoc against what I had actually run and tightened two more of my own sentences that still claimed too much: "present in every build checked" became "seen in all three builds disassembled here", since the 8u builds were yours and were never checked for that case; and "reads exactly as many bytes as its first argument holds" is now scoped to when it enters the loop, because on the 8u shape a length mismatch reads none at all. Comment-only again: `git diff` has no non-comment line. `BasicAuthFilterTest` 8 tests 0F 0E, spotless clean. **On your CI note, one of the four has a firmer answer than a flake.** You flagged these as inference rather than known-flake claims, so to close that out: - `transform-v2-it-part-1 (8)`, `TestFilterRowKindIT.testFilterRowKindMultiTable`: this is #12116, which names that exact test and records it failing only on Flink legs, matching the container you saw. It was closed on 2026-10-05T16:44, the same minute #12216 merged, and #12216 is the change to `AssertSinkWriter`'s close path that the issue's root-cause analysis points at. `git merge-base --is-ancestor` puts `fea1ac0a4` in `dev` and **not** in this branch, which is 91 behind, so the branch predates the fix and the sync should clear this one rather than reroll it. - `paimon-connector-it (8)`, `PaimonWithS3IT`: #11679 is the open issue in that area, but it documents a job-level hang that never produces a result line and dies at the 180 minute wall, not four 300 second per-test timeouts. Different symptom, so I would not claim they are the same without reading the logs. - `engine-k8s-it` on both legs: agreed, the k3s download failed before any test ran, so there is nothing to attribute. Synced with the latest `dev` in the push above, so `fea1ac0a4` is now in the branch. `BasicAuthFilterTest` is still 8 of 8 on the merged tree and spotless is clean; the merge is a pure one, with my head differing from `dev` in only the three files this PR owns. -- 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]
