SEPURI-SAI-KRISHNA commented on PR #13061: URL: https://github.com/apache/seatunnel/pull/13061#issuecomment-6542365751
@chovy-3012 thanks, all three are fixed in the new commit. 1. `convertDataType` now maps `int64` to `BIGINT`. 2. `convertArrowSchemaToJsonArrowSchema` now labels a 64-bit integer column `int64` (`uint64` when unsigned). This one matters more than it looks: the directory namespace returns no schema from `describeTable`, so `getTable` always goes through this fallback. Fixing only 1 still read the column as `INT`. 3. `readBigints` now fails on a duplicated id. Coverage: `LanceCatalogTest.getTable` already creates a `long_col` BIGINT column, but its `toString` comparison never saw the column types, because `TableSchema.toString()` leaves out the columns. It now also asserts `long_col` reads back as `BIGINT` and `integer_col` as `INT`. Reverting either 1 or 2 alone fails it with `expected: <BIGINT> but was: <INT>`. The narrower and unsigned integer types still read as `INT`, as before; I kept this PR to `BIGINT`. @DanielLeens the conflict is resolved by merging `dev`. Only the two `incompatible-changes.md` files conflicted; both entries are kept, and the merged change against `dev` is the same as before. Checks, JDK 17: `connector-lance` 41 tests, 0 failures, with the CI `surefire.jvm.args`; `spotless:check` clean on `connector-lance` and `connector-lance-e2e`. The PR description is updated. -- 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]
