SEZ9 commented on PR #11673:
URL: https://github.com/apache/seatunnel/pull/11673#issuecomment-5230722619

   Thanks @loustler — this addresses both of my findings, and I'm happy with 
the reasoning as well as the fixes.
   
   **Issue 1.** Accepted, and I'll concede the framing: "shaded jars are the 
runtime surface for every connector" is the right way to think about it, and 
the #11648 `okio`/k8s-e2e precedent is a concrete demonstration that the static 
check alone can't stand in for runtime coverage. The <1% commit-frequency data 
point settles the cost question for me. Good catch on the `hadoop3.version` 
lockstep detail too — agreed that `seatunnel-core/**` catching a *correct* bump 
is incidental protection, not a guarantee, which is exactly why Option A is the 
safer default.
   
   **Issue 2.** The full-prefix mask with the constant pointing back at the pom 
property is what I was after, and re-running all four controls (patched pass, 
unpatched 3.4.3 fail, missing file, non-zip) after narrowing the regex is 
exactly the right discipline — thanks for not skipping that. Ordering fix 
confirmed on my end as well; the guard now runs before `checkLicense.sh` wipes 
`target/`.
   
   **Remaining (non-blocking, per my earlier comment):**
   1. Fold in SEZ9's Issues 1–3 plus my suggestion of a small unit test for the 
masking regex — fine to do here or as a fast follow-up, your call. Of these, 
extending the guard to also verify the uber jar still relocates Jackson is the 
one I'd prioritize, since a regression there would reproduce #11655 from the 
other direction without this CI step firing.
   2. Since this PR touches `seatunnel-shade/**` it now triggers the full 
matrix on itself — please make sure the final head goes green end-to-end before 
merge. That was my one open item and it applies doubly now.
   
   Core fix and guard are correct as they stand — approving once CI is green on 
the final head.
   
   <!-- streview-comment:113 -->


-- 
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