Au-Miner opened a new pull request, #28150:
URL: https://github.com/apache/flink/pull/28150
## What is the purpose of the change
Fix incorrect results from Window TVF aggregation when using the RocksDB
state backend. The bug could make multiple window timers fire with the last
buffered key instead of the correct per-key state.
## Brief change log
- Update RecordsWindowBuffer#flush to copy WindowKey before passing it to
RecordsCombiner on the reusable-key path, so timer registration does not retain
a mutable key reference.
- Adjust the RecordsCombiner contract to reflect that callers now pass a
stable window key object.
## Verifying this change
Added RecordsWindowBufferTest#testFlushUsesStableWindowKeys, which asserts
that flushing buffered records keeps distinct keys instead of reusing the last
key for all entries. Also ran RecordsWindowBufferTest and the full
flink-table-runtime module test suite.
## Does this pull request potentially affect one of the following parts:
- Dependencies (does it add or upgrade a dependency): no
- The public API, i.e., is any changed class annotated with
@Public(Evolving): no
- The serializers: no
- The runtime per-record code paths (performance sensitive): no
- Anything that affects deployment or recovery: JobManager (and its
components), Checkpointing, Kubernetes/Yarn, ZooKeeper: no
- The S3 file system connector: no
## Documentation
- Does this pull request introduce a new feature? no
- If yes, how is the feature documented? not 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]
