Gerrrr opened a new pull request #11447: URL: https://github.com/apache/kafka/pull/11447
Right now, the `repartition` operator always filters out `null` keys. This behavior is not correct and is a regression compared to the deprecated `through` operator. This patch fixes the issue by filtering `null` keys only for optimizable repartition nodes. First, it removes unnecessary processor node from the topology for `UnoptimizableRepartitionNode`. Second, it only adds the filtering processor only if the node is optimizable. I introduced an `isOptimizable` boolean method for repartition nodes to make this property more obvious in the code. To verify that change, I added a new test to the repartition integration suite that produces pairs with `null` keys and expects them to be present after the repartition. ### Committer Checklist (excluded from commit message) - [ ] Verify design and implementation - [ ] Verify test coverage and CI build status - [ ] Verify documentation (including upgrade notes) -- 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: jira-unsubscr...@kafka.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
