[GitHub] [spark] zhengruifeng commented on pull request #42548: [WIP][SPARK-44750][PYTHON][CONNECT] Apply configuration to sparksession during creation

["zhengruifeng (via GitHub)" <gi...@apache.org>]

zhengruifeng commented on PR #42548:
URL: https://github.com/apache/spark/pull/42548#issuecomment-1683449163

   Before this PR, the configurations set in `Session.Builder` completely takes no effect.
   
   We can use `spark.conf.get` to get the `RuntimeConf` and `SparkConf`;
   But `spark.conf.set` only accepts dynamic SQL configurations, and the internal `options` contains `SparkConf` like
   
   ```
   {'spark.sql.pyspark.jvmStacktrace.enabled': 'false', 'spark.connect.execute.reattachable.senderMaxStreamDuration': '1s', 'spark.connect.execute.reattachable.senderMaxStreamSize': '123', 'spark.app.name': 'SparkConnectSessionWithOptionsTest', 'spark.remote': 'sc://localhost', 'spark.master': 'local[4]', 'spark.local.connect': '1', 'spark.connect.grpc.binding.port': '0'}
   ```
   
   These `SparkConf` actually won't take effect, even after this PR.
   
   There are some places in which the `SparkConf` is already used in `Session.Builder`
   
   
   
   https://github.com/apache/spark/blob/4be6cd1cc0a111f20e41ee2fa4dd45ec4296364d/python/pyspark/testing/connectutils.py#L174-L177  @HyukjinKwon 
   
   https://github.com/apache/spark/blob/433fb2af8a3ca239958cb7b006e2924ecfac0d56/python/pyspark/ml/tests/connect/test_connect_classification.py#L35  @WeichenXu123 


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org