Possible limitations about an user defined scalar function in a Synapse dedicated SQL pool

Question

Hi,

for a Synapse Analytics solution I need to implement some controls about data to read from a staging area in order to write them to a dwh.

So, I think to create the necessary user defined scalar function in T-SQL language, but I'd like to know which are the possible limitations respect to a SQL Server on-premise data platform.

Reading this article, it seems that the scalar functions are fully supported: https://learn.microsoft.com/en-us/azure/synapse-analytics/sql/overview-features

Any helps to me, please? Thanks

Answer 1

@pmscorca Thanks for using Microsoft Q&A forum and posting your query.

Scalar User-defined functions are fully supported in Synapse Analytics Dedicated SQL, and you can create them using T-SQL language. These functions can be used to implement controls about data to read from a staging area in order to write them to a data warehouse.

Regarding the possible limitations of user-defined scalar functions in Synapse Analytics, there are no significant differences compared to SQL Server on-premises data platforms.

Looks like the limitations or restrictions called out in this document applies to On-Prem SQL, Azure SQL & Azure Synapse Dedicated SQL: SQL UDF - Limitations and restrictions

Here is the latest information specific to Azure Synapse analytics documentation: CREATE FUNCTION (Azure Synapse Analytics and Microsoft Fabric)

Hope this info helps.

---

Please don’t forget to Accept Answer and Yes for "was this answer helpful" wherever the information provided helps you, this can be beneficial to other community members.