taozhiyi13 commented on issue #12923: URL: https://github.com/apache/seatunnel/issues/12923#issuecomment-7002821807
@SEZ9 Thanks. My preference is **one complete PR organized into three dependency-ordered commits**. The reason is end-to-end coherence: the API and CDC slices do not provide standalone user-facing value without the Elasticsearch/OpenSearch sink, while one PR allows the complete `MySQL-CDC -> Zeta -> Elasticsearch/OpenSearch` E2E suite to run before anything is merged. Reviewers can still review the PR commit by commit; squash happens only after the complete implementation has been reviewed and approved. The three commits will follow the STIP slices: 1. **Shared API and engine contract** — geometry types, Zeta serialization/restore, and Flink/Spark rejection. 2. **MySQL-CDC source** — CDC-specific schema mapping, Debezium decoding, checkpoint tests, and JDBC compatibility regression tests. 3. **Elasticsearch/OpenSearch sink** — geo mapping and validation, E2E tests, and documentation. If the `dev@` discussion prefers stacked PRs, the same three boundaries can be used. I will post the final mailing-list outcome back here. -- 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]
