ATX AnalyticsTable connects to Microsoft Fabric — Warehouses, Lakehouse SQL analytics endpoints and SQL databases — and to Azure Synapse pools with Microsoft Entra ID, and works on the results locally: a DuckDB copy in the project, pivots and charts on your machine, and personal data masked before an extract is shared.
The Microsoft Fabric Warehouse connection form in ATX AnalyticsTable 1.7, filled in with the example values below. Secrets go into the encrypted vault, never into the project.
After sharing the warehouse with the service principal (Read) in the Fabric portal: GRANT SELECT ON SCHEMA::dbo TO [atx-reporting-sp];
Reporting query — employee e-mails hashed in the Fabric Warehouse: SELECT DATEFROMPARTS(YEAR(j.posting_date), MONTH(j.posting_date), 1) AS month, j.cost_center, CONVERT(varchar(64), HASHBYTES('SHA2_256', LOWER(e.email)), 2) AS employee_key, SUM(j.amount) AS amount FROM dbo.journal_lines AS j JOIN dbo.employees AS e ON e.employee_id = j.employee_id WHERE j.posting_date >= DATEADD(month, -12, CAST(GETDATE() AS date)) GROUP BY DATEFROMPARTS(YEAR(j.posting_date), MONTH(j.posting_date), 1), j.cost_center, CONVERT(varchar(64), HASHBYTES('SHA2_256', LOWER(e.email)), 2) ORDER BY month; — On Synapse, give a query that is loaded page by page its own ORDER BY on output columns — Synapse has no OFFSET … FETCH.
Then reshape locally — a Run SQL step on the project's DuckDB: -- "input" is the table the previous pipeline step produced (the import above). -- This runs on your machine, in the project's DuckDB, not on the database. SELECT region, SUM(revenue) AS revenue, COUNT(DISTINCT customer_key) AS customers FROM input GROUP BY region ORDER BY revenue DESC; — Column names follow the query above; on the local engine the pipeline's current table is registered as input before the script runs.
Fabric workspace and item permissions, and the warehouse's own column and row security, apply to the Entra identity that runs the query. With on-behalf-of turned on, every Entra mode except user name and password runs queries as the signed-in person.
Fabric is a cloud service, so queries need a route to port 1433. The imported copy and everything you build on it work offline on your machine; the licence validates offline.
Access to the Fabric workspace or item (for example Read on the warehouse) and SELECT on what it reads. AnalyticsTable connects to the SQL endpoint directly with Entra ID; nothing is published to a Power BI workspace.
A Lakehouse SQL endpoint and a Warehouse share a host name. The type decides whether writes are attempted — Lakehouse endpoints are read-only.
Yes — dedicated and serverless pools. SQL logins work on Synapse; paged loads use ROW_NUMBER because Synapse has no OFFSET … FETCH.
This site needs JavaScript for the full experience. Every page above is reachable without it.