DanielLeens commented on issue #10593: URL: https://github.com/apache/seatunnel/issues/10593#issuecomment-5731111187
Thanks @rucciva. I rechecked current `dev` at `1325a44b2f4013198fdb2c12f9df345de1cde4bc`: PostgreSQL CDC discovery still selects only `TABLE_TYPE = 'BASE TABLE'`, and the approximate row-count path still selects only `relkind = 'r'`. I also found no open PR covering this parent-partition root cause, so it is not fixed in `dev`. Please do not revive a closed branch blindly. The issue is still assigned to @misi1987107, so first coordinate here to avoid parallel implementations; if that work is no longer active, link the earlier PR so we can preserve any valid review history and start a new focused PR. The first change needs to cover the whole parent-table contract, not just widen one predicate: 1. discover a configured partitioned parent without admitting unrelated tables; 2. define a safe snapshot split and row-count strategy for the parent and its children rather than assuming the parent's `reltuples` represents child rows; and 3. add an end-to-end regression that configures the parent, verifies initial rows from child partitions, and verifies a later child-partition change. The existing child-table wildcard/list workaround remains useful, but it is not a first-class parent-table implementation. Once the earlier PR reference and ownership are clear, a focused proposal with those acceptance tests will be ready for review. -- 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]
