cshuo opened a new pull request, #19819: URL: https://github.com/apache/hudi/pull/19819
### Describe the issue this Pull Request addresses Closes #19818. `HoodieFlinkEngineContext#mapGroupsByKey` submitted only a lazy stream transformation to its dedicated `ForkJoinPool`. The terminal operation ran after the submitted task returned and the pool was shut down, so `processFunc` normally executed in the common pool instead of honoring the dedicated parallelism. Empty input also attempted to construct a pool with zero parallelism. ### Summary and Changelog - Execute the complete grouped-map stream pipeline, including flattening and collection, inside the dedicated `ForkJoinPool`. - Return empty `HoodieListData` directly when there are no groups. - Add regression coverage for dedicated-pool execution and empty input. ### Impact Flink grouped-map processing now uses the intended isolated pool rather than contending with unrelated common-pool work. There are no public API or configuration changes. ### Risk Level Low. The change only moves the existing terminal stream operations into the already-intended execution scope and adds an empty-input guard. Targeted unit tests verify the pool identity, configured parallelism, results, and empty-input behavior. ### Documentation Update None. ### Contributor's checklist - [ ] Read through [contributor's guide](https://hudi.apache.org/contribute/how-to-contribute) - [x] Enough context is provided in the sections above - [x] Adequate tests were added if applicable -- 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]
