hutiefang76 opened a new pull request, #12557: URL: https://github.com/apache/seatunnel/pull/12557
### Purpose of this pull request DuckDB catalog discovery currently reads `character_maximum_length` for DECIMAL columns. DuckDB reports 38 there for a declared `DECIMAL(10,2)`, so copying an inferred table can silently widen the column. Read `numeric_precision` for DECIMAL and NUMERIC while retaining `character_maximum_length` for other types. ### Does this PR introduce _any_ user-facing change? Yes. A `table_path` source discovering `DECIMAL(10,2)` now reports precision 10 and scale 2, and a table created from that inferred schema keeps `DECIMAL(10,2)`. The English and Chinese DuckDB source docs describe this behavior. ### How was this patch tested? - Reproduced the defect before the fix: `DuckDBCatalogTest` expected precision 10 but observed 38. - JDK 8 and JDK 17: focused `DuckDBCatalogTest` runs passed (8 tests each, no failures, errors, or skips). - JDK 17: `MarkdownTest` passed after the documentation edit. - Scoped `spotless:apply` and `verify` passed; `git diff --check` passed. ### Check list * [x] No new binary or connector registration is introduced. * [x] English and Chinese connector documentation updated. * [x] No incompatible configuration change. -- 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]
