DanielLeens commented on issue #10443: URL: https://github.com/apache/seatunnel/issues/10443#issuecomment-5570389485
Thanks @moneycat957. I rechecked the current `dev` tree at `72acda57042efa3408681dc2c93663910b06b3e8`, the JDBC dialect registration path, recent history, and the complete open-PR set. There is no CockroachDB dialect or connector in the tree and no active CockroachDB PR, so the March claim is no longer an active implementation path. A `connector-jdbc` dialect is the right direction; please do not create a standalone `connector-cockroachdb` module. The existing connector already discovers database-specific behavior through `JdbcDialectFactory`, and the PostgreSQL-family dialects are the appropriate structural reference. For a first PR, keep the scope to the CockroachDB factory/dialect and only the converter or row-mapping pieces whose behavior actually differs from PostgreSQL. In particular, cover dialect selection, the intended native UPSERT SQL, and the supported CockroachDB type mapping with focused unit tests. Do not copy the PostgreSQL package mechanically, add a new driver dependency, or expand into CDC, changefeeds, transaction-retry policy, topology-aware routing, or Cockroach-specific catalog/DDL behavior. Please include a small real integration test using the official single-node CockroachDB image, but add it to the existing `connector-jdbc-e2e` reactor rather than creating a separate E2E module. It should prove actual dialect discovery plus a basic read/write path and the native upsert behavior. Keep the full PR linked here when it is ready; that is how we will track the claim. No label or assignee was changed. -- 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]
