yuqi1129 opened a new issue, #14525: URL: https://github.com/apache/gravitino/issues/14525
### Version branch-1.3 (Spark 3.5 connector) ### Describe what's wrong The release branch PostgreSQL Spark 3.5 catalog does not override Spark's write-aware loadTable overload. A write may load metadata using the read path and omit the required MODIFY_TABLE privilege check. ### Error message and/or stacktrace A SELECT_TABLE-only caller can reach INSERT/INSERT OVERWRITE without the expected write authorization failure. ### How to reproduce Use the branch-1.3 PostgreSQL Spark 3.5 connector and execute INSERT INTO or INSERT OVERWRITE as a caller lacking MODIFY_TABLE. The write-aware load must require MODIFY_TABLE before touching the physical table. ### Additional context Tracks https://github.com/datastrato/gravitino-enterprise/issues/3158. Main already has the common guard from PR #12414; this is a targeted branch-1.3 backport. Tests exercise real Spark SQL planning with controlled metadata/table boundaries; deployed REST + PostgreSQL end-to-end verification remains separate. -- 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]
