You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Yicong Huang (Jira)" <ji...@apache.org> on 2026/08/03 20:36:00 UTC
[jira] [Resolved] (SPARK-58161) [PYTHON] Standardize PySpark UDF row count mismatch error to use RESULT_ROWS_MISMATCH
[ https://issues.apache.org/jira/browse/SPARK-58161?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Yicong Huang resolved SPARK-58161.
----------------------------------
Fix Version/s: 4.3.0
Resolution: Fixed
Issue resolved by pull request 57292
[https://github.com/apache/spark/pull/57292]
> [PYTHON] Standardize PySpark UDF row count mismatch error to use RESULT_ROWS_MISMATCH
> -------------------------------------------------------------------------------------
>
> Key: SPARK-58161
> URL: https://issues.apache.org/jira/browse/SPARK-58161
> Project: Spark
> Issue Type: Improvement
> Components: PySpark
> Affects Versions: 4.3.0
> Reporter: gaurav kumar pandey
> Assignee: gaurav kumar pandey
> Priority: Major
> Labels: pull-request-available
> Fix For: 4.3.0
>
>
> Currently, PySpark Pandas and Arrow UDFs raise a custom SCHEMA_MISMATCH_FOR_PANDAS_UDF error class when the returned result length does not match the input row count.
> To align with Spark's error standardization efforts, we should migrate this to the standard RESULT_ROWS_MISMATCH error class and remove the redundant SCHEMA_MISMATCH_FOR_PANDAS_UDF from error-conditions.json.
> Proposed Changes:
> 1. Migrate UDF scalar and arrow verification logic in worker.py to raise RESULT_ROWS_MISMATCH with input/output length parameters.
> 2. Retain the old error message format ("Result vector from pandas_udf was not the required length...") in the message string for backward-compatibility with user code that asserts on the exception message.
> 3. Update affected unit tests, documentation guidelines (debugging.rst), and clean up error-conditions.json.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org