Microsoft Fabric¶

KelpMesh supports Microsoft Fabric via the SQL Analytics endpoint (T-SQL over ODBC).

Configuration¶

warehouse:
  type: fabric
  account: myworkspace.datawarehouse.fabric.microsoft.com
  database: MyLakehouse
  threads: 4

Or use a direct connection string:

warehouse:
  type: fabric
  connection_string: "Driver={ODBC Driver 18 for SQL Server};Server=tcp:myworkspace.datawarehouse.fabric.microsoft.com,1433;Database=MyLakehouse;Encrypt=yes;"

Install dependencies¶

pip install pyodbc azure-identity

Install the ODBC Driver 18 for SQL Server:

Windows: Download from Microsoft
macOS: brew install microsoft/mssql-release/msodbcsql18
Linux: Microsoft Linux instructions

Authentication¶

The Fabric adapter uses DefaultAzureCredential from azure-identity. It tries sources in order:

Environment variables (AZURE_CLIENT_ID, AZURE_CLIENT_SECRET, AZURE_TENANT_ID)
Azure CLI (az login)
Managed Identity (when running in Azure)
Visual Studio / VS Code credentials

For local development, az login is the simplest option.

Permissions¶

GRANT CREATE TABLE TO [your_principal];
GRANT CREATE VIEW TO [your_principal];
GRANT SELECT, INSERT, UPDATE, DELETE ON SCHEMA::dbo TO [your_principal];

Materialization support¶

Materialization	Supported	Notes
`view`	✅	`CREATE VIEW` (T-SQL)
`table`	✅	`SELECT * INTO` (T-SQL — not `CREATE TABLE AS`)
`incremental` (append)	✅	`INSERT INTO`
`incremental` (merge)	✅	T-SQL `MERGE INTO`
`ephemeral`	✅	Inlined as CTE

Incremental merge example¶

-- {{ config(materialized="incremental", unique_key="customer_id", incremental_strategy="merge") }}
SELECT
  customer_id,
  email,
  plan,
  updated_at
FROM raw.customers
WHERE updated_at >= DATEADD(DAY, -7, GETDATE())

KelpMesh generates a T-SQL MERGE:

MERGE INTO [dim_customers] AS target
USING (...) AS source
ON target.[customer_id] = source.[customer_id]
WHEN MATCHED THEN UPDATE SET target.[email] = source.[email], ...
WHEN NOT MATCHED THEN INSERT ([customer_id], [email], ...) VALUES (source.[customer_id], ...);

Note the trailing semicolon — required by T-SQL MERGE syntax.

T-SQL differences¶

Standard SQL	T-SQL equivalent
`CREATE TABLE t AS SELECT ...`	`SELECT * INTO t FROM (...) AS _src`
`LIMIT n`	`SELECT TOP n`
`CREATE OR REPLACE VIEW`	`ALTER VIEW` / `CREATE VIEW`
`NOW()`	`GETDATE()`
`INTERVAL '7' DAY`	`DATEADD(DAY, 7, date)`

KelpMesh handles the CREATE TABLE difference automatically.