Best practice to substitute udf scalar value functions

[MaxBronder]

In our current onprem data warehouse we have many scalar value functions with complex physics calculations.

Because scalar value functions are not supported I'm wondering what is the best practice now to substitute them.

Converting them to a multi statement tvf could be a solution, but they are also not supported.

Is this limitation only for preview or will there still be no support for Scalar and Mulit tvfs by the date of GA?

And if so, what would you recommend on how the replicate this functionality?

KR

[Anonymous · Accepted solution]

Hi @MaxBronder  ,
Thanks for using Fabric Community.

A quick suggestion is to use Python to implement those Physics calculations. A notebook can be used to

• Read the source table(s) ( from Lakehouse  or Warehouse).
• Compute the functions as needed.
• Save the results in a lakehouse, together with a row identifier.
• After that, SQL queries can join the original data with the newly created Lakehouse table and collect the results.

The support for scalar UDF or TVF is not in Fabric warehouse general availability scope. Scalar UDF is on our radar for post-general availability planning. We don't have a specific ETA to share at this moment.

Appreciate if you could share the feedback on our feedback channel. Which would be open for the user community to upvote & comment on. This allows our product teams to effectively prioritize your request against our existing feature backlog and gives insight into the potential impact of implementing the suggested feature.

Hope this helps. Please let me know if you have any further queries.

[Anonymous]

Hi @MaxBronder ,
We haven’t heard from you on the last response and was just checking back to see if your query has been resolved. Please let us know if you have any further queries.

[Anonymous]

Hi @MaxBronder ,

Glad that your query got resolved. Please continue using Fabric Community for any help regarding your queries.