topic Re: How to enable CDF when saveAsTable from pyspark code? in Data Engineering

How to enable CDF when saveAsTable from pyspark code?

drag7ter — Thu, 11 Apr 2024 21:18:44 GMT

I'm running this code in databricks notebook and I want the table from dataframe in catalog were created with CDF enables. When I run the code table hasn't exited yet.

This code doesn't create a table with enables CDF. It doesn't add:

delta.enableChangeDataFeed = true

 df \
.write \
.format("delta") \
.mode("overwrite") \
.option("overwriteSchema", "true") \
.option("delta.enableChangeDataFeed", "true") \
.saveAsTable(my_table_name_in_catalog):

Or it is possible only enable it via Spark SQL, after table already exists?

Re: How to enable CDF when saveAsTable from pyspark code?

raphaelblg — Fri, 12 Apr 2024 15:43:54 GMT

Hello @drag7ter ,

I don't see anything wrong with your approach, check my repro:

Re: How to enable CDF when saveAsTable from pyspark code?

drag7ter — Sat, 13 Apr 2024 11:04:01 GMT

Hi @raphaelblg in your screenshot cdcTest table already exists in unity catalog, or it will be created for the first time, running your code? It is really strange as I'm not able to enable from the pyspark code, only after tables created ALTER TABLES works.

"delta.enableChangeDataFeed", "true"