[
https://issues.apache.org/jira/browse/SPARK-51206?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Hyukjin Kwon reassigned SPARK-51206:
------------------------------------
Assignee: Haoyu Weng
> Python Data Sources incorrectly imports from Spark Connect
> ----------------------------------------------------------
>
> Key: SPARK-51206
> URL: https://issues.apache.org/jira/browse/SPARK-51206
> Project: Spark
> Issue Type: Bug
> Components: PySpark
> Affects Versions: 4.0.0, 4.1.0, 4.1
> Reporter: Haoyu Weng
> Assignee: Haoyu Weng
> Priority: Major
> Labels: pull-request-available
>
> Python Data Source workers plan_data_source_read and write_into_data_source
> import some arrow/row conversion helpers from pyspark.sql.connect.conversion.
> The imported helper actually doesn't need any Spark Connect dependency.
> However pyspark.sql.connect.conversion checks for Spark Connect dependencies.
> This causes failure when using Python Data Source if any of pandas, grpcio,
> etc. is missing.
> {code:python}
> .venv/lib/python3.9/site-packages/pyspark/sql/worker/plan_data_source_read.py:34:
> in <module> from pyspark.sql.connect.conversion import
> ArrowTableToRowsConversion, LocalDataToArrowConversion
> .venv/lib/python3.9/site-packages/pyspark/sql/connect/conversion.py:20: in
> <module> check_dependencies(__name__)
> .venv/lib/python3.9/site-packages/pyspark/sql/connect/utils.py:37: in
> check_dependencies require_minimum_grpc_version()
> .venv/lib/python3.9/site-packages/pyspark/sql/connect/utils.py:49: in
> require_minimum_grpc_version raise PySparkImportError( E
> pyspark.errors.exceptions.base.PySparkImportError: [PACKAGE_NOT_INSTALLED]
> grpcio >= 1.48.1 must be installed; however, it was not found.{code}
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
