DanielLeens commented on PR #11577: URL: https://github.com/apache/seatunnel/pull/11577#issuecomment-5385688134
Thanks @SEZ9 for the deep pass on the current head (`0a25d27e5051`). Issue 1 (DDL built by string-concatenating CDC-supplied column names into `ALTER TABLE ... ADD COLUMN`) is a serious one if confirmed — an unvalidated identifier reaching a BigQuery multi-statement query job is exactly the kind of injection surface that should block merge on its own, regardless of the basis for my earlier approval. I'm treating this as the current blocking item rather than my July 29 approval. I'll fold this together with the writer-schema-refresh propagation race (Issue 4) and the unsupported-DDL crash-loop concern (Issue 5) into a full re-review of the schema-evolution path once a new commit lands — both of those also touch restore/consistency behavior that matters for this connector. One thing I want to flag so we don't lose track of it: this round doesn't repeat your 2026-08-01 possible-API-change concern on `AbstractBigQuerySinkWriter`. Could you confirm whether that one is now considered resolved, or still open alongside the new findings 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]
