dlt.destinations.impl.fabric.configuration
Configuration for Fabric Warehouse destination - extends Synapse configuration with COPY INTO support
FabricCredentials Objects
@configspec(init=False)
class FabricCredentials(AzureServicePrincipalCredentials)
Credentials for Microsoft Fabric Warehouse with Service Principal authentication.
Fabric Warehouse requires Azure AD Service Principal authentication. Inherits from AzureServicePrincipalCredentials for Service Principal fields and automatic fallback to DefaultAzureCredential.
drivername
SQLAlchemy driver name for SQL Server/Fabric.
host
Fabric Warehouse host (e.g., abc12345-6789-def0-1234-56789abcdef0.datawarehouse.fabric.microsoft.com)
port
Database port (default: 1433)
database
Fabric Warehouse database name
connect_timeout
Connection timeout in seconds (default: 15)
azure_storage_account_name
Not used for Fabric Warehouse credentials (only staging credentials need this)
on_partial
def on_partial() -> None
Enable fallback to DefaultAzureCredential if explicit credentials not provided.
get_odbc_dsn_dict
def get_odbc_dsn_dict() -> Dict[str, Any]
Build ODBC DSN dictionary with Fabric-specific settings.
to_odbc_dsn
def to_odbc_dsn() -> str
Build ODBC connection string for pyodbc.
to_native_credentials
def to_native_credentials() -> Optional[Any]
Return credentials in a format suitable for the native driver/library.
FabricClientConfiguration Objects
@configspec
class FabricClientConfiguration(DestinationClientDwhWithStagingConfiguration)
Configuration for Fabric Warehouse destination with staging and collation support.
Uses FabricCredentials for Service Principal authentication. Supports OneLake/Lakehouse or Azure Blob Storage staging with COPY INTO for efficient data loading.
Example usage with OneLake/Lakehouse staging (recommended): fabric( credentials={ "host": "abc12345-6789-def0-1234-56789abcdef0.datawarehouse.fabric.microsoft.com", "database": "mydb", "tenant_id": "your-tenant-id", "client_id": "your-client-id", "client_secret": "your-client-secret", }, staging=filesystem(
IMPORTANT: Must use workspace GUID and lakehouse GUID (not names)
Format: abfss://<workspace_guid>@onelake.dfs.fabric.microsoft.com/<lakehouse_guid>/Files
bucket_url="abfss://12345678-1234-1234-1234-123456789012@onelake.dfs.fabric.microsoft.com/87654321-4321-4321-4321-210987654321/Files",
IMPORTANT: Must specify Service Principal credentials (same as warehouse)
credentials={ "azure_storage_account_name": "onelake", "azure_account_host": "onelake.blob.fabric.microsoft.com", "azure_tenant_id": "your-tenant-id", "azure_client_id": "your-client-id", "azure_client_secret": "your-client-secret", }, ), collation="Latin1_General_100_BIN2_UTF8", )
Note: The bucket_url must use GUIDs for both workspace and lakehouse, not their display names. You can find these GUIDs in the Fabric portal workspace/lakehouse URLs.
Example usage with Azure Blob Storage staging: fabric( credentials={ "host": "abc12345-6789-def0-1234-56789abcdef0.datawarehouse.fabric.microsoft.com", "database": "mydb", "tenant_id": "your-tenant-id", "client_id": "your-client-id", "client_secret": "your-client-secret", }, staging=filesystem( bucket_url="az://your-container", credentials={ "azure_storage_account_name": "your-account-name", "azure_storage_account_key": "your-account-key", }, ), collation="Latin1_General_100_BIN2_UTF8", )
destination_type
type: ignore[misc]
collation
Database collation to use for text columns.
Note: Fabric Warehouse does not support table indexing. Storage is automatically managed by the system.
create_indexes
Whether primary_key and unique column hints are applied.
has_case_sensitive_identifiers
Whether identifiers (table/column names) are case-sensitive. Depends on database collation.
__config_gen_annotations__
Database collation for varchar columns. Fabric supports:
- Latin1_General_100_BIN2_UTF8 (default, case-sensitive)
- Latin1_General_100_CI_AS_KS_WS_SC_UTF8 (case-insensitive)
Both have UTF-8 encoding. LongAsMax=yes is automatically configured.
data_location
def data_location() -> str
Returns host:port.
fingerprint
def fingerprint() -> str
Returns a fingerprint of the physical Fabric location.