Hyukjin Kwon created SPARK-61628:
------------------------------------
Summary: Match PySpark's SRID and type-string errors, and the bool
arguments of checkpoint and unionByName, in the Python client
Key: SPARK-61628
URL: https://issues.apache.org/jira/browse/SPARK-61628
Project: Spark
Issue Type: Improvement
Components: Connect
Affects Versions: connect-rust-4.3.0
Reporter: Hyukjin Kwon
Three differences from PySpark 4.3.0 that a post-merge review found. (1)
collect, head, first, take, tail, toPandas and isEmpty of a DataFrame whose
schema has an SRID that PySpark does not support raise the server's error,
where PySpark raises its ST_INVALID_SRID_VALUE before the query runs. (2)
StructField, ArrayType and MapType take a type string, where PySpark asserts a
DataType, and a bad type string gives an error message with a Rust Debug dump.
(3) localCheckpoint(eager), checkpoint(eager) and
unionByName(allowMissingColumns) reject the ints, None and objects with
__index__ that PySpark's protobuf bool conversion accepts.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]