hesam-oxe commented on PR #11077: URL: https://github.com/apache/seatunnel/pull/11077#issuecomment-6144928757
@SEZ9 — thanks for the rebase check. Confirming against current head `8b383c0`: **Local unit-test result:** ran the multi-table sink suite locally — **51 tests green, 0 failures** (`MultiTableSinkWriterTest` 36, `SchemaChangeBroadcast` 11, committer 2, aggregated-committer 2). `spotless:check` on `seatunnel-api` passes. One env note: clean-checkout `mvn -pl seatunnel-api -am test` fails in `seatunnel-config-shade` (it compiles against the packaged shaded jar, so `-am install -DskipTests` has to run first) — pre-existing build ordering, unrelated to this PR. **Checklist:** - **PR11077-F1 (key collisions)** — fixed: `MultiTableSink.java:687-736` (class + identifier key, identity fallback), credential-folding warning in `SeaTunnelSink.java:180-187`, INFO log on alias join (`:235-246`, `:354-365`). - **PR11077-F2 (snapshot/restore duplication)** — fixed: `groupByIdentity` (`MultiTableSinkWriter.java:325-335`), snapshot/prepareCommit/abort/close/flush once per writer under `primaryIdentifier`; restore merges legacy per-alias via `getRestoredState` (`MultiTableSink.java:440-452`). - **PR11077-F3 (schema divergence)** — fixed: fail-fast on `SeaTunnelRowType` mismatch (`MultiTableSink.java:142-175`), documented in both incompatible-changes files. - **PR11077-F4 / F6 (docs)** — fixed: "Multi-Table Writer Sharing" section in `docs/en|zh/developer/sink-connector-development.md`. - **PR11077-F5 (proxyContexts)** — fixed: unconditional `put` per alias (`MultiTableSink.java:251`, `:370`), no `containsValue` skip. - **PR11077-F7 (IOException wrap)** — fixed: writer creation uses explicit get/put (`:225-228`, `:338-349`); `computeIfAbsent` only for non-throwing proxies. - **PR11077-F8 (javadoc)** — fixed: `getDestinationKey` docs at `MultiTableSink.java:475-489`. I'm a full-stack engineer working from frontends down through distributed systems and kernels, so shared-writer correctness (no silent cross-destination routing, no state duplication) is the part I verified most carefully. Let me know if you want any item restructured. — Hesam -- 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]
