DanielLeens commented on PR #12592: URL: https://github.com/apache/seatunnel/pull/12592#issuecomment-6411596616
Thanks @SEZ9 for the careful follow-up. The two items you list as open were answered in @hutiefang76's 2026-10-24 comment above, so here are the concrete pointers on the unchanged head `cb6ae01c05481b1085bc3bc3c3f430d74b909fd2` (source reading only, nothing executed locally): **F4 (upgrade note):** `docs/en/introduction/concepts/incompatible-changes.md:127` says a standard DuckDB `TIMESTAMP` source now reports scale 6, so generated DDL declares the precision explicitly, naming MySQL `DATETIME(6)`, PostgreSQL `timestamp(6)`, SQL Server `DATETIME2(6)` and Oracle `TIMESTAMP(6)`, "where it previously relied on the target default", and that existing target tables are not resized. One small wording nuance: the note does not literally write "instead of `DATETIME`", it says the previous behavior relied on the target default, which for MySQL is the unqualified `DATETIME`. If you want the most common case spelled out in those exact words, that is a one-sentence doc tweak, but the information is already there. The zh page also carries the `DATETIME(6)` text. **F2 / F6 (pre-epoch limitations):** both are pinned by `DuckDBSourceAndSinkTest.testPreEpochAliasReadsKeepTheLegacyGetterForDstGapAndCutover` (around line 1255 of the connector-jdbc test file). It covers a 1918-03-31 02:30 wall clock inside the US spring-forward gap, a 1918-06-15 control row, and a 1582-10-10 pre-cutover row, and asserts the legacy-getter results per zone (the US zones move 02:30 to 03:30, the pre-cutover date comes back ten days later). The note at line 129 documents the same two limitations and says reconciling them is deferred. So this is "known and pinned", not "fixed", which matches what you said you can accept. I agree the F5 / F7 `Calendar` hoisting can stay a non-blocking follow-up. My view is unchanged: no source-level or CI blocker on this head (required Build is green). This is a targeted reply, not a new full review; the remaining gates are your final pass and a committer's approval. -- 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]
