SEPURI-SAI-KRISHNA commented on PR #12595:
URL: https://github.com/apache/seatunnel/pull/12595#issuecomment-6162481680
@SEZ9 @DanielLeens confirmed, and thank you both.
**1. The javadoc wording is in the pushed head, not local.** It went up in
`b3f26aeb9` ("[Fix][Zeta] State the MessageDigest.isEqual javadoc as fact and
scope it to the builds checked", 2026-10-11 15:06:57 UTC), which is the commit
before the `dev` merge that produced the current head `364514f12c`. The file is
`seatunnel-engine/seatunnel-engine-server/src/main/java/org/apache/seatunnel/engine/server/rest/filter/BasicAuthFilter.java`,
javadoc on `credentialMatches`. The two sentences you asked about read:
> On the builds disassembled with `javap -c`, 11.0.32, 17.0.20 and 21.0.12,
the length difference is folded into the accumulator and the loop runs for the
first argument's length, with no early return when the lengths differ. On the
Java 8 builds checked, 8u71 and 8u171, `isEqual` instead compares the two array
lengths with `if_icmpeq` and returns false immediately when they differ, so
there the equality of the byte lengths can still be inferred, though the
leading bytes cannot.
It names no boundary update on purpose, and the 8u sentence is impersonal
because that measurement is not mine.
**2. Syncing `dev`.** `07d6f2235a` is on `dev` and my head predates it, so I
will merge and let `Build` run again.
**3. Credit where it belongs.** #12756 was opened by @JeremyXin at
2026-10-13 12:47 UTC, 47 minutes before I filed #12757 at 13:34 UTC, with the
same `EncoderException` / `Direct buffer memory` / `AllNodesFailedException`
chain. The fix was already in flight; my issue did not prompt it. @nzw921rx
closed #12757 as fixed by it this morning.
**4. The #12732 overlap resolved itself.** NganWave closed #12732 unmerged
at 2026-10-14 04:38 UTC with no comment, so the TRUNCATE-race reading was
neither confirmed nor refuted in public. What landed is the task off-heap
memory change, not a TRUNCATE wait. I have nothing further owed there.
Expected after the sync: `all-connectors-it-2` green, and
`all-connectors-it-6` (`DatabendCDCSinkIT`) possibly still red, which is
independent of this diff and needs only a failed-jobs rerun.
--
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]
