[
https://issues.apache.org/jira/browse/SPARK-61769?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Hyukjin Kwon updated SPARK-61769:
---------------------------------
Fix Version/s: connect-rust-4.3.0
> Make pyspark.sql.SparkSession PySpark's classic class, the builder of the
> Connect session PySpark Connect's, and the Connect functions of their own
> modules, as in PySpark
> --------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: SPARK-61769
> URL: https://issues.apache.org/jira/browse/SPARK-61769
> Project: Spark
> Issue Type: Improvement
> Components: Connect
> Affects Versions: connect-rust-4.3.0
> Reporter: Hyukjin Kwon
> Priority: Major
> Fix For: connect-rust-4.3.0
>
>
> In the drop-in, pyspark.sql.SparkSession is the native class of the Connect
> session, so isinstance(spark, pyspark.sql.SparkSession) is True and
> SparkSession.builder.getOrCreate() without a URL returns the active session,
> where PySpark 4.3.0 gives False and raises CONNECT_URL_NOT_SET. This makes
> pyspark.sql.SparkSession PySpark's classic class (Apache Spark's
> pyspark/sql/session.py as it is), whose builder makes the session with the
> builder of pyspark.sql.connect.session.SparkSession, which becomes PySpark
> Connect's (no SPARK_REMOTE, SESSION_NEED_CONN_STR_OR_BUILDER,
> remote(location)). It also gives the functions of
> pyspark.sql.connect.functions and the class of cogroup their
> pyspark.sql.connect modules, and ships PySpark's py.typed and typing stubs.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]