DanielLeens commented on PR #12156:
URL: https://github.com/apache/seatunnel/pull/12156#issuecomment-6210070584

   Thanks for the checklist, @SEZ9. This crossed paths with my review of the 
same head (`0cd5d8544e4`, review 5516481221), so here is the same confirmation 
in one place, checked against the code at that head rather than the PR 
description:
   
   - **F1**: confirmed. `cancelJob()` captures `JobStatus jobStatus = 
getJobStatus();` at `PhysicalPlan.java:216`, guards on 
`jobStatus.isEndState()`, and then branches with `if 
(NOT_STARTED_STATUSES.contains(jobStatus))` at `:224`. `stopJob()` does the 
same at `:263` and `:269`. The second `runningJobStateIMap.get(jobId)` read in 
`cancelJob()` is gone.
   - **F3**: both halves are covered. The enum-order guard 
`JobStatusTest#testOrdinalOrderIsStableForInternalRpcCompatibility` has a 
Javadoc at `JobStatusTest.java:42` telling a future contributor to append new 
statuses and not to refresh the expected list. For the `PhysicalPlan` decision, 
`StateTransitionCleanupTest` has 
`testCancelJobCommitsToTheJobStatusSnapshotItValidated` (`:174`), 
`testCancelOnNotStartedJobGoesStraightToCanceled` (`:200`) and 
`testCancelOnRunningJobGoesThroughCanceling` (`:213`). The first one makes the 
first IMap read return `PENDING` and a later read return `RUNNING`, so it fails 
if `cancelJob()` goes back to a second read.
   - **F4**: the Javadoc on `NOT_STARTED_STATUSES` (`:54-65`) carries the 
explanation and the #12342 pointer, and the two call-site comments that 
duplicated the set contents are removed in the `dev...head` diff.
   - **F2**: agreed that it stays deferred to #12342.
   
   So from the source side there is nothing left on this head. The only open 
gate is CI: the apache-side `Build` check is still `action_required` and the 
fork workflows have to be re-enabled by @CryoThrust before a green run can be 
linked. A committer approval is still needed as well, since my account is 
comment-only.


-- 
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]

Reply via email to