pnowojski opened a new pull request, #26051:
URL: https://github.com/apache/flink/pull/26051
## What is the purpose of the change
This PR:
- implements the `RowTimeDeduplicateKeepFirstRowFunction` using watermarks
to avoid retracting previous results
- changes planner to take into account that both proc time and rowtime
deduplicate in keep first row variants are append only
Thanks to that, planner can avoid costly operators like
`SinkUpsertMaterializer` downstream the whole query can be append-only -
removing a need for retracting/upserting results from the output table.
Currently there is no variant for the async state backend.
## Verifying this change
This adds new tests and is also covered by existing ITCases.
## Does this pull request potentially affect one of the following parts:
- Dependencies (does it add or upgrade a dependency): (yes / **no**)
- The public API, i.e., is any changed class annotated with
`@Public(Evolving)`: (yes / **no**)
- The serializers: (yes / **no** / don't know)
- The runtime per-record code paths (performance sensitive): (**yes** / no
/ don't know)
- Anything that affects deployment or recovery: JobManager (and its
components), Checkpointing, Kubernetes/Yarn, ZooKeeper: (yes / **no** / don't
know)
- The S3 file system connector: (yes / **no** / don't know)
## Documentation
- Does this pull request introduce a new feature? (yes / **no**)
- If yes, how is the feature documented? (**not applicable** / docs /
JavaDocs / not documented)
--
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]
