Hyukjin Kwon created SPARK-61179:
------------------------------------
Summary: Unset a DataFrameReader option of None, and take the
description of Catalog.createExternalTable, in the Python client as PySpark does
Key: SPARK-61179
URL: https://issues.apache.org/jira/browse/SPARK-61179
Project: Spark
Issue Type: Bug
Components: Connect
Affects Versions: connect-rust-4.3.0
Reporter: Hyukjin Kwon
Two arguments of the Python client's readers and catalog differ from PySpark's:
1. PySpark's DataFrameReader.option(key, None), options(key=None) and load(...,
key=None) unset the option: the reader keeps None, and the plan leaves out the
options of None. So spark.read.option("header", "true").option("header",
None).csv(path) reads without a header. The client skips None, so the earlier
value stays and the file is read with a header.
2. PySpark's Catalog.createExternalTable(tableName, path, source, schema,
**options) calls createTable(tableName, path, source, schema, **options), so a
description keyword is the description of the table. The client sends it as an
option "description", so the table has no comment and a "description" property.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
