Skip to content

New Database: Power BI Semantic Models #1190

Description

@WesleyDieterich

Database

Connection

  • How do you typically connect?
    Connections are established via the XMLA endpoint exposed by Power BI / Fabric workspaces:

    powerbi://api.powerbi.com/v1.0/myorg/<workspace-name>
    

    Optionally specifying a dataset:

    Provider=MSOLAP;
    Data Source=powerbi://api.powerbi.com/v1.0/myorg/<workspace>;
    Initial Catalog=<dataset-name>;
    
  • Protocol:
    XMLA (XML for Analysis over HTTP/SOAP)

  • Node.js driver:
    There is no widely adopted native Node.js driver. Typically accessed via:

    • MSOLAP (OLE DB)
    • ADOMD.NET / AMO (via .NET interop)
  • Auth methods:

    • Microsoft Entra ID (OAuth2 / Azure AD)
    • Interactive login (with MFA)
    • Service Principal (where supported)

Additional info

  • Database documentation link:
    https://learn.microsoft.com/en-us/analysis-services/instances/connection-string-properties-analysis-services

  • Special considerations:

    • Requires Build permission or higher on the dataset

    • XMLA endpoint can be read-only or read-write depending on capacity configuration

    • This is not a traditional relational database:

      • Queries are written in DAX, MDX, or via DMVs
      • No support for standard SQL or information_schema
    • Common tools supporting this connection:

      • SQL Server Management Studio (SSMS)
      • Tabular Editor
      • Excel (live connection)
    • Metadata discovery relies on DMVs instead of relational catalogs

Why this matters

Power BI / Fabric semantic models are increasingly used as a centralized semantic layer in modern data architectures. Supporting XMLA endpoints would allow DBCode users to:

  • Explore semantic models directly from the editor
  • Run DAX/MDX queries without switching tools
  • Inspect metadata (measures, tables, relationships) programmatically
  • Integrate BI-layer logic into developer workflows

Suggestion

This feature could potentially be implemented as support for Microsoft Analysis Services / XMLA endpoints (similar to SSAS or Azure Analysis Services), rather than as a traditional SQL database integration.

Even read-only support (query + metadata exploration) would already provide significant value.

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions