SEZ9 commented on PR #11077:
URL: https://github.com/apache/seatunnel/pull/11077#issuecomment-5673682895

   Thanks for the sync, @hesam-oxe. Rebasing onto `dev` (`bd09f6d9`) to pick up 
the MinIO mirror fix makes sense, and I agree the `minio/minio` Docker Hub 
`404` on the previous head looks like infrastructure rather than this change. 
I'll wait for the re-run on the new head before drawing any conclusions.
   
   One thing to flag so expectations are clear: the new head 
`ef2bb0955f30bc108ef96d30e9c4adfe1f294279` is a `dev`-sync merge on top of 
`a9d846469b` with no new commits of this branch's own, so the earlier review 
findings are unchanged. Still open from the previous round:
   
   - **PR11077-F1 / F2 (HIGH)** — destination-key collisions routing one 
table's rows through another sink's writer, and the snapshot fan-out plus 
restore-time union duplicating shared-writer state N times on recovery. These 
are the blockers; a fix or a written argument for why the current behavior is 
safe would unblock this.
   - **PR11077-F3, F5, F7 (MEDIUM)** — shared writer built from an arbitrary 
alias's sink/CatalogTable with no schema/config divergence guard; 
`proxyContexts` only registering the first alias per destination (plus the 
O(n²) `containsValue` cost); `IOException` from `createWriter`/`restoreWriter` 
wrapped in an unchecked `RuntimeException` inside `computeIfAbsent`.
   - **PR11077-F4, F6 (MEDIUM, docs)** — `getPhysicalDestinationIdentifier()` 
and the writer-sharing behavior aren't described in `docs/`, and the changed 
`restoreWriter` contract (merged state from all aliased identifiers in one 
call) isn't documented for connector implementers.
   - **PR11077-F8 (LOW)** — missing param/return tags on the 
`getDestinationKey` Javadoc.
   
   If you've already addressed some of these locally, just push and add a short 
note mapping commits to finding IDs and I'll re-review promptly. If you 
disagree with any of them, a reply on the individual finding is equally welcome 
— happy to discuss.
   
   <!-- streview-comment:1051 -->


-- 
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