SEZ9 commented on PR #12271:
URL: https://github.com/apache/seatunnel/pull/12271#issuecomment-5658760280
Follow-up on the merge commit `0f0e78a4c8`, the only new commit since
`6202ad987` ("Ready to merge"). It carries no independent logic, but since the
merge conflicted in the two files this PR touches, I re-read the merged result
against my three earlier points rather than treating it as a no-op:
- The `getTaskClassLoader()` fail-fast is present, now reading off
`context`, and both `BlockingWorker.run()` and `CooperativeTaskWorker.run()`
still route through it.
- The `BlockingWorker` ordering fix (`startedLatch.countDown()` first,
before the class loader lookup) is present and unchanged.
- The identity-checked cleanup in `recycleClassLoader()`/`taskDone()` is now
expressed through `dev`'s `finishExecution()`/`executionContexts.compute()`
path keyed by `TaskGroupContext.executionId`; the intent (a stale generation
can't clobber a newer one) survives.
- Both regression tests,
`testCooperativeTrackerKeepsGenerationSpecificClassLoader` and
`testStaleGenerationCleanupDoesNotRecycleNewerContext`, are present and updated
for the new `TaskGroupContext` constructor.
Nothing was lost in the merge and I have no new concerns. Still good to
merge from my side — thanks for keeping the branch current.
<!-- streview-comment:1036 -->
--
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]
