Retrieve from the database and immediatelly save

[foxale]

Let's say I want to load some table from the database using SQLTableDataSet and automatically save it as a CSV file somewhere in the data/ space. I can see three potential workarounds:

1. Create a custom dataset inheriting after SQLTableDataSet and add saving logic myself
2. Create an empty Kedro node (with no python code) which would save the dataset as its output
3. Maybe play around with transcoding datasets?

What's the Kedro-native way to achieve this?

[oscarvilla]

It have passed a lot of time.
I was working with something similar: taking data from BigQuery through a query and then saving the result in a .csv in bucket.
My approach was like you point in 2.