For datasources it's just something that is run on the connection before you statement is executed, it doesn't seem to depend on the specific jdbc driver. See here https://github.com/apache/spark/blob/95fc4c56426706546601d339067ce6e3e7f4e03f/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/jdbc/JDBCRDD.scala#L343
But are you looking for something similar before writing? I'm not aware of anything but you can just use foreachPartition to try it out? On Fri, 26 Nov 2021, 04:55 , <trs...@gmail.com> wrote: > Sorry I somehow missed the "Scope" column in the docs, which > explicitly states its for reads only. I don't suppose anyone knows of some > other method I can submit SET statements for write sessions? > > On Fri, Nov 26, 2021 at 12:51 PM <trs...@gmail.com> wrote: > >> Hello, >> >> Regarding JDBC sinks, the docs state: >> https://spark.apache.org/docs/latest/sql-data-sources-jdbc.html >> sessionInitStatement: >> After each database session is opened to the remote DB and before >> starting to read data, this option executes a custom SQL statement (or a >> PL/SQL block). Use this to implement session initialization code. Example: >> option("sessionInitStatement", """BEGIN execute immediate 'alter session >> set "_serial_direct_read"=true'; END;""") >> >> The language suggests this is for reads only? >> >> I'm wondering if I can use this with Google Cloud Spanner to run: >> SET AUTOCOMMIT_DML_MODE = 'PARTITIONED_NON_ATOMIC' >> https://cloud.google.com/spanner/docs/use-oss-jdbc#set_autocommit_dml_mode >> >> I've tried, but I think it's just being ignored, because >> sessionInitStatement is not accepted by the open source spanner JDBC >> driver. Is this somehow handled by Spark independently of the JDBC driver >> or is it sessionInitStatement driver specific? >> >> >>
