Getting Started

The code below uses the ibis library to connect to the DataArts Fabric data lake, execute data queries, and convert the results into the basic syntax of the DataFrame format.

The example is for reference only. Modify it as needed.

For more detailed usage of Ibis, refer to official Ibis documentation.

import ibis  # Import Ibis dependencies.

con = ibis.fabric.connect(  # Call the DataArts Fabric backend connection to establish a connection.
    endpoint=FABRIC_ENDPOINT,  # Specify the service region. For details, see Regions and Endpoints.
    endpoint_id=FABRIC_ENDPOINT_ID,  # Querying an Endpoint ID
    domain=FABRIC_DOMAIN,  # Tenant name
    user=FABRIC_USER,  # IAM username
    password=FABRIC_PASS,  # IAM password
    access_key=ACCESS_KEY,  # Obtaining an AK/SK
    secret_key=SECRET_KEY,
    security_token=SECURITY_TOKEN,  # Obtaining a Temporary AK/SK
    project_id=FABRIC_PROJECT_ID,  # How Do I Obtain a Project ID?
    catalog_name=IBIS_TEST_FABRIC_CATELOG,  # Connect to a specified catalog.
    workspace_id=FABRIC_WORKSPACE_ID, # Obtaining a Workspace ID
    lakeformation_instance_id=IBIS_TEST_FABRIC_LAKEFORMATION_INSTANCE_ID, # LakeFormation instance ID. For details, see Mapping Between DataArts Fabric SQL and LakeFormation Data Types.
    obs_directory_base=OBS_DIRECTORY_BASE,  # Storage path for UDFs in OBS.
    obs_bucket_name=OBS_BUCKET_NAME,  # OBS bucket name.
    obs_server=OBS_SERVER,  # OBS access address. For details, see Endpoints and Domain Names.
    verify=VERIFY,  # Whether to verify the server's SSL certificate. The value is of the Boolean type.
)
t = con.table("table_name", database="db")  # Retrieve table information by connecting to the backend and create a table object.
t.select("cola")  # Query table columns.
df = t.execute()  # Convert DataFrame to SQL, send it to the backend for execution, and return the result in Pandas DataFrame format.