Hyukjin Kwon created SPARK-61036:
------------------------------------
Summary: Clone the session on the server in cloneSession in the
Python client as PySpark does
Key: SPARK-61036
URL: https://issues.apache.org/jira/browse/SPARK-61036
Project: Spark
Issue Type: Bug
Components: Connect
Affects Versions: connect-rust-4.3.0
Reporter: Hyukjin Kwon
In PySpark Connect, {{spark.cloneSession(new_session_id=None)}} clones the
session on the server with a {{CloneSession}} RPC. The new session has a copy
of the server-side state of this one, its SQL configurations, temporary views,
registered functions and catalog state, and the id {{new_session_id}} if one is
given. Changes to either session after that do not show in the other.
The Rust-backed Python client, and the Rust API
{{SparkSession::clone_session}}, return a new session instead, with none of the
state of this one, and ignore {{new_session_id}}. After
{{spark.conf.set("spark.sql.shuffle.partitions", "7")}},
{{spark.cloneSession().conf.get("spark.sql.shuffle.partitions")}} is {{200}},
not {{7}}, and the temporary views of the session are not found. Clone the
session as PySpark does.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
