Yicong Huang created SPARK-58157:
------------------------------------
Summary: Refactor SQL_TRANSFORM_WITH_STATE_PYTHON_ROW_UDF
Key: SPARK-58157
URL: https://issues.apache.org/jira/browse/SPARK-58157
Project: Spark
Issue Type: Sub-task
Components: PySpark
Affects Versions: 5.0.0
Reporter: Yicong Huang
Refactor SQL_TRANSFORM_WITH_STATE_PYTHON_ROW_UDF so that the worker uses the
plain ArrowStreamSerializer for pure Arrow stream I/O, moving the per-eval-type
logic (regrouping rows by grouping key into Row objects and converting result
Rows back to Arrow) from TransformWithStateInPySparkRowSerializer into
read_udfs() in worker.py. The serializer class is kept for now since
TransformWithStateInPySparkRowInitStateSerializer (init-state variant, tracked
separately) still subclasses it. Part of SPARK-55388.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
