[
https://issues.apache.org/jira/browse/SPARK-57567?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Max Gekk updated SPARK-57567:
-----------------------------
Shepherd: Max Gekk
> Support TimeType in Arrow-based Python/pandas UDF evaluation
> ------------------------------------------------------------
>
> Key: SPARK-57567
> URL: https://issues.apache.org/jira/browse/SPARK-57567
> Project: Spark
> Issue Type: Sub-task
> Components: PySpark
> Affects Versions: 4.1.0
> Reporter: Max Gekk
> Priority: Major
>
> h2. What
> Support {{TimeType}} columns as input/output of Arrow-optimized Python UDFs
> and pandas UDFs.
> h2. Where
> {{ColumnarArrowEvalPythonEvaluatorFactory}}
> (sql/core/.../execution/python/ColumnarArrowEvalPythonEvaluatorFactory.scala,
> around line 312)
> builds a columnar batch from the input rows (via the row-to-columnar
> converter) before the
> Arrow hand-off to the Python worker.
> h2. Relationship to SPARK-54203
> This area is gated at the row-to-columnar converter level by SPARK-54203
> ({{RowToColumnConverter.getConverterForType}} in
> sql/core/src/main/scala/org/apache/spark/sql/execution/Columnar.scala), which
> currently has
> no {{TimeType}} case. Once SPARK-54203 lands, the work below becomes
> possible. It may also
> require TIME support in this component's own layer (Arrow type mapping,
> Parquet/ORC logical
> types, or Variant encoding).
> h2. Acceptance criteria
> * Arrow-optimized Python UDFs and pandas UDFs accept and return TIME columns.
> * Tests added in PySpark (e.g. arrow/test_arrow.py) and the relevant Scala
> evaluator suite.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
