API as a Data Source in FMD Framework

[vpinsight]

Hi team,

We are evaluating the FMD Framework for a Microsoft Fabric-based enterprise ingestion pattern and noticed that API does not appear to be available as a first-class data source type in the metadata-driven configuration.

For source systems where data needs to be retrieved directly from REST APIs instead of file-based landing-zone ingestion, what is the recommended FMD pattern?

Specifically:

1. Is API ingestion intentionally excluded as a native/first-class data source type?
2. Is the expected approach to use the custom notebook functionality to call the API and then land the data into the framework's landing zone / downstream medallion flow?
3. If custom notebooks are the intended pattern, is there a recommended metadata configuration pattern for registering and orchestrating those notebooks within FMD?
4. Are there any examples or roadmap plans for API-based ingestion support?

We are trying to keep the implementation aligned with the framework's intended design rather than bolting on a separate ingestion path outside of FMD.

Thanks.

[edkreuk]

The challenge is with API there are so many different, that's why I implemente the custom Notebook to solve this issue. As long you use the start and end part everything will work fine, you can define how you want to extract the data from the API yourselves.
There is a Stored Proc [integration].[sp_UpsertLandingzoneBronzeSilver] https://github.com/edkreuk/FMD_FRAMEWORK/wiki/Adding-a-new-entity. Which you can use to add the CustomNotebooks, you just have to make sure that there are in the same workspace. The Notebook NB_FMD_PROCESSING_LANDINGZONE_MAIN is handling all the processes.
Can you explain a bit more or that roadmap plans for API-based ingestion support?