Forum Discussion

ToddChitt's avatar
ToddChitt
Icon for Super User rankSuper User
2 years ago

Save PySpark datafram to Fabric Warehouse via T-SQL type MERGE operation

I am building out a solution that will need to process lots of JSON files. I initially set it up to save the JSON file to a known location and known file name in the Lakehouse, then use a Dataflow Ge...
data engineering
data warehouse
Anonymous's avatar
Anonymous
Not applicable
2 years ago

Hi ToddChitt ,

Thanks for using Fabric Community.
Unfortunately, you cannot directly send data to a Warehouse table in Fabric notebooks

But you can write the data to lake house and use it with three part name in warehouse 


Here is the Session where pawel showed this in our community session , skip to 1:02:05
https://www.youtube.com/watch?v=u7peyrNhYgw

Helpful Docs to refer -
Load data to MS Fabric Warehouse from notebook - Stack Overflow

Updating Fabric warehouse tables through a notebook - Stack Overflow


I hope it might give some ideas to your use case. 

Element115's avatar
Element115
Icon for Memorable Member rankMemorable Member
2 years ago

What are you saying?  Quote: 'But you can write the data to lake house and use it with three part name in warehouse ...'
That it is possible to write to a LH table from the LH SQL endpoint?   I thought that CRUD ops on a LH were only possible from a Notebook using either %%SQL or PySpark. Yes? No?