urosstan-db opened a new pull request, #59990: URL: https://github.com/apache/spark/pull/59990
### What changes were proposed in this pull request? Remove the unreleased `spark.sql.legacy.oracle.timestampNTZMapping.enabled` configuration. Oracle now has constant `true` defaults for `zonelessTimestampAsNTZ` and `timestampNTZAsWallClock`; explicit JDBC options continue to control type mapping and value conversion separately. Update the JDBC documentation, migration guide, and Oracle tests to use those options. Setting both to `false` restores the pre-4.4 behavior. For JDBC catalog tables, `zonelessTimestampAsNTZ` belongs in the catalog configuration. ### Why are the changes needed? Follow-up to #59671. The two JDBC options cover the compatibility controls previously provided by the session flag, which has never shipped in a Spark release. Removing it also prevents V2 schema resolution and scan construction from deriving different conversion defaults when the session flag changes between those phases. ### Does this PR introduce _any_ user-facing change? Yes, relative to unreleased branches only: the session flag is removed. Oracle's default behavior stays the same, and users can select the old mapping and conversion through the JDBC options. ### How was this patch tested? Passed on JDK 17 with `-Pdocker-integration-tests`: - `sql/testOnly org.apache.spark.sql.jdbc.JDBCSuite -- -z "SPARK-58876"` (6 tests). - `docker-integration-tests/Test/compile`. - Scoped Scala style checks for Catalyst, SQL production/tests, and Docker integration tests. - `git diff --check`. The updated unit tests cover the default and explicit read/write conversion options, including the value shift in a non-UTC JVM zone, and the type-mapping combinations with `preferTimestampNTZ`. Live Oracle integration tests were not run because no Oracle Docker image is installed locally. ### Was this patch authored or co-authored using generative AI tooling? Generated-by: OpenAI Codex (GPT-6) -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
