SEZ9 commented on PR #11503: URL: https://github.com/apache/seatunnel/pull/11503#issuecomment-5628670693
Thanks for the F1–F8 status against `f77cdfa677d4`. I can't close findings on the description alone, so here is what I still need before I mark anything resolved: - **F1 (`default.default` hardcode)** — Please point to the hunks in `f77cdfa677d4` where the table path is derived from the split/row-type table IDs and the WARN-and-skip fallback lives, plus the tests that assert the real table path (including the multi-table and ambiguous-identity cases). Once I can see those in the diff I'll close it. - **F2 (unknown-event fallback)** — Keeping this at Medium for now. Please point to the dispatcher/handler changes and the no-op `applySchemaChange` default in the diff, and add the doc note (or test) enumerating which connectors are covered via the shared dispatcher vs. an explicit guard vs. the no-op default. - **F3 (null `changeAfter`)** — Please point to the `requireNonNull` in the constructor and the `IllegalStateException` in `getRestoredTable()` in the diff, and to a test covering the null case if there is one. - **F4 (non-atomic `clear()`+`putAll()`)** — Please point to the synchronized `getHistoryTableChanges()` / `restoreCheckpointHistoryTableChanges()` / `deserialize()` paths in the diff so I can confirm they share the same monitor. - **F5 (widened restore gate)** — Your F5 note appears to be cut off mid-sentence. Could you repost it and confirm (a) that the event reaching transforms/sinks is only emitted on an actual row-type inequality, and (b) which test asserts that a restore with an unchanged schema emits no `RestoreTableSchemaEvent`? - **F6 (dispatcher restore shortcut not verifying the target table)**, **F7 (missing Javadoc on `restoreCheckpointHistoryTableChanges`)**, **F8 (full `CatalogTable` list logged at INFO)** — I don't see status for these. Please state for each whether it's addressed in `f77cdfa677d4` and where (class/method), or why you consider it out of scope. Once the diff locations are on the thread I'll do a final pass and we can move forward. <!-- streview-comment:956 --> -- 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]
