Tuesday
Hi,
I am using databricks runtime 17.3.x-scala2.13 ,
Tuesday
You can use the language SQL instead of PYTHON as its the supported language for stored procedures.
SQL stored procedures are good for scripting & creating modular SQL based workflows within boundary of Unity Catalog. It's a secure governed way to execute SQL without leaving warehouse.
However, if the workflows need advanced code logic, complex loops or manipulation of the Spark session, you can use databricks notebook running Python spark code as it gives access to the full power of Spark helping you create scalable & complex data pipelines that SQL cannot support.
More details here
Tuesday
You can use the language SQL instead of PYTHON as its the supported language for stored procedures.
SQL stored procedures are good for scripting & creating modular SQL based workflows within boundary of Unity Catalog. It's a secure governed way to execute SQL without leaving warehouse.
However, if the workflows need advanced code logic, complex loops or manipulation of the Spark session, you can use databricks notebook running Python spark code as it gives access to the full power of Spark helping you create scalable & complex data pipelines that SQL cannot support.
More details here
Wednesday
Hi @RGSLCA ,
This video is a bit misleading - look at the comments section. This feature was not released and as of now you can only create stored procedure using an SQL language.
Wednesday
Databricks stored procedures currently support LANGUAGE SQL, not LANGUAGE PYTHON, so that error is expected. Docs show CREATE PROCEDURE supports LANGUAGE SQL only, on DBR 17.0+ with Unity Catalog.
yesterday
Hi
As others have said stored procedures don't currently support Python. You can either create the stored procedure with SQL using Windows functions and describe history or put it into a notebook and not have stored procedures. The video is about experimental features.
Current docs:
Thanks,
Emma