Hyukjin Kwon created SPARK-61235:
------------------------------------
Summary: Add the Spark Connect resource.profile module and the
DistributedSequenceID and UnresolvedNamedLambdaVariable expressions to the
Python client
Key: SPARK-61235
URL: https://issues.apache.org/jira/browse/SPARK-61235
Project: Spark
Issue Type: Bug
Components: Connect
Affects Versions: connect-rust-4.3.0
Reporter: Hyukjin Kwon
The pyspark-client 4.3.0 packages ship pyspark.sql.connect.resource.profile,
and the official test_connect_column imports DistributedSequenceID and
UnresolvedNamedLambdaVariable from pyspark.sql.connect.expressions. The drop-in
has neither: the import of the module raises ModuleNotFoundError, and the test
module fails at collection with ImportError.
Repro: `from pyspark.sql.connect.resource.profile import ResourceProfile` and
`from pyspark.sql.connect.expressions import DistributedSequenceID`. Official
PySpark 4.3.0 imports both.
The new expressions build their proto as PySpark's do (an unresolved function
distributed_sequence_id, an unresolved named lambda variable), and the lambda
variable names of pyspark.sql.connect.functions count from
UnresolvedNamedLambdaVariable._nextVarNameId, as PySpark's do.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
