SEZ9 commented on PR #12443: URL: https://github.com/apache/seatunnel/pull/12443#issuecomment-5882647787
Thanks for pushing this through the remaining rounds — I've gone over the discussion since `ddca53b7a`. On the previously raised points: - **F3 / F2 (key semantics + docs):** The note in `docs/en/engines/zeta/rest-api-v2.md` describing that keys which fail HOCON path parsing are kept literally, while keys that do parse still go through the existing `->`/quote/whitespace normalization, is exactly the clarification I was after. Could you confirm the same note is present in the `docs/zh` counterpart as well? I only see the English file confirmed in the thread. - **F1 / F5 (decryptConfig and the other `Map`→`Config` entry points):** The thread states that all four production choke points are now guarded and covered by tests. That resolves my main concern, but please point me to the specific test(s) that push a regex key such as `^t_nova_.*$` through the `ConfigShadeUtils.decryptConfig()` path so I can double-check the round trip. - **F7 (title):** Given the final diff spans `seatunnel-api` and `seatunnel-core-starter` in addition to engine-server, `[Fix][Core]` is appropriate — withdrawing that nit. **Remaining ask before merge:** the current Build run on `ddca53b7a` is reported as failing across a number of connector/engine IT jobs that don't overlap with the changed files, and the branch is reported to be well behind `dev`. Rather than reasoning about flakiness from the outside, please merge latest `dev` into this branch and let CI re-run. Once that comes back green and the two confirmations above are in, I'm ready to approve. <!-- streview-comment:1393 --> -- 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]
