SEZ9 commented on PR #11885: URL: https://github.com/apache/seatunnel/pull/11885#issuecomment-5628412228
Thanks @li3zhi4 — the pointers at `befad345c` are exactly what I needed. On F1/F6: `shouldPreserveStopOffsetThroughJavaSerializationRoundTrip` (lines 34-47), the `roundTrip(...)` helper (lines 64-73) using real `ObjectOutputStream`/`ObjectInputStream`, and `shouldDefaultStopOffsetToNullOnLegacySerializedState` (lines 49-62) in `IncrementalPhaseStateTest` are the right shape for what I flagged. Your comment appears to have been cut off right after "`DefaultSerializer` → `Ser`" — could you post the rest of that paragraph? I'll then verify the test and the absence of a custom `PendingSplitsStateSerializer` / `getEnumeratorStateSerializer` override against the diff before marking F1/F6 resolved. On the rest, I see the thread already points at the `resolvedStopOffset == null` guard at `IncrementalSplitAssigner.java:329-343` and `resolveLatestStopOffsetWithRetry` for F2/F5, the `MysqlCDCStopModeSpecificIT` assertion for F3, and the doc locations for F4/F7/F8. I'll check each of those against the diff myself rather than ask for them again, and will follow up if anything looks off. Once I have the tail of the F1/F6 note, I'll do a final pass. <!-- streview-comment:953 --> -- 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]
