DanielLeens commented on PR #11206:
URL: https://github.com/apache/seatunnel/pull/11206#issuecomment-6139066019

   Follow-up to the latest review on commit 
`9007896511517e4b874142cfe704102861bbc444`:
   
   - Synced the branch with current `apache/dev` 
(`b0f6feddf9ed0dc1f5d485cfc69d387edfe6ac82`) by merge, not rebase. The only 
conflict was in `MySqlIncrementalSourceFactory`; the resolution keeps both the 
current dev specific-offset guards and this PR's newly-added-table options.
   - Fixed the restore compatibility blocker by changing 
`scan.newly-added-table.enabled` to opt-in (`false` by default) and updating 
the English and Chinese MySQL CDC docs.
   - Added a real runtime-sink capability gate before sink/writer creation. 
`TableSinkFactory` defaults to unsupported, JDBC explicitly opts in, and an 
unsupported sink now fails early with the sink type and table path instead of 
entering generic factory creation.
   - Caught rejected runtime identifiers at the CDC deserializer boundary, 
logging a WARN and skipping only that table instead of failing the whole task. 
Added a negative regression test.
   - Expanded the safe MySQL identifier rule to accept legal digit-leading 
names such as `2024_orders` while still rejecting all-numeric and unsafe names, 
with tests.
   - Added English and Chinese JDBC sink documentation for runtime tables 
discovered by MySQL CDC.
   
   The remaining Medium follow-ups about multi-sink target pinning, an explicit 
retry/idempotency contract, and old-format checkpoint/rollback documentation 
are unchanged in this commit.
   
   Verification performed locally: `spotless:apply` and `spotless:check` passed 
for `seatunnel-api`, `connector-cdc-base`, `connector-cdc-mysql`, and 
`connector-jdbc`; `git diff apache/dev...HEAD --check` passed; the final PR 
diff against current `apache/dev` contains no `pom.xml` changes. Per the 
repository workflow for this task, I did not run local compilation or tests. 
GitHub CI is queued.


-- 
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