Azure Management Data Factory Client

Warnings

• breaking Major breaking changes occurred between versions 8.x and 9.x. The client constructor signatures changed, authentication now primarily uses `azure.identity` credentials, and `msrestazure` specific patterns are largely replaced. Migration to `azure.identity.DefaultAzureCredential` is mandatory.
  Fix: Rewrite client instantiation to `DataFactoryManagementClient(credential, subscription_id)` and use `from azure.identity import DefaultAzureCredential` for authentication. Refer to Azure SDK migration guides for full details.
• breaking Version 9.0.0 and above of `azure-mgmt-datafactory` require Python 3.9 or higher. Installations on older Python versions will fail due to `Requires-Python` metadata.
  Fix: Upgrade your Python environment to 3.9 or newer before installing or running `azure-mgmt-datafactory` version 9 or higher.
• gotcha Authentication issues are common. `DefaultAzureCredential` attempts various authentication methods (environment variables, managed identity, Azure CLI, etc.). Incorrect setup will lead to `CredentialUnavailableError` or permission errors.
  Fix: Ensure relevant environment variables (`AZURE_CLIENT_ID`, `AZURE_TENANT_ID`, `AZURE_CLIENT_SECRET`) are set for service principals, or you are logged in via `az login` for user authentication, or other supported authentication methods are correctly configured for your environment.
• gotcha Many management operations require the Azure Resource Group name and other resource-specific names (e.g., factory name, pipeline name) as direct parameters. Forgetting or mistyping these can lead to `ResourceNotFound` or `404` errors, even with correct authentication.
  Fix: Always double-check that required resource group names and other resource identifiers are correctly provided to the client methods. Use `os.environ.get()` to load them reliably for robust scripts.

Install

• pip install azure-mgmt-datafactory azure-identity Install core library and authentication

Imports

• DataFactoryManagementClient

  from azure.mgmt.datafactory import DataFactoryManagementClient

  The top-level package import is the standard and recommended way for Azure SDKs since version 2.x, avoiding deep package paths.
• DefaultAzureCredential

  from azure.identity import DefaultAzureCredential

Quickstart

This quickstart demonstrates how to authenticate using `DefaultAzureCredential` and initialize the `DataFactoryManagementClient`. It then performs a basic operation to list Data Factories within a specified resource group. Ensure `AZURE_SUBSCRIPTION_ID` and `AZURE_RESOURCE_GROUP` are set in your environment variables for successful execution.

import os
from azure.identity import DefaultAzureCredential
from azure.mgmt.datafactory import DataFactoryManagementClient

# --- Authentication --- 
# Set environment variables for authentication:
# AZURE_CLIENT_ID, AZURE_CLIENT_SECRET, AZURE_TENANT_ID (for Service Principal)
# or log in via Azure CLI: az login
# or use managed identity, etc.

# Retrieve subscription ID from environment variable
subscription_id = os.environ.get("AZURE_SUBSCRIPTION_ID", "")
if not subscription_id:
    print("Please set the AZURE_SUBSCRIPTION_ID environment variable.")
    exit(1)

# Create a credential object using DefaultAzureCredential
# This attempts to authenticate via various methods (environment variables, managed identity, Azure CLI, etc.)
credential = DefaultAzureCredential()

# Create a DataFactoryManagementClient
data_factory_client = DataFactoryManagementClient(credential, subscription_id)

# --- Example Operation: List Data Factories in a Resource Group ---

resource_group_name = os.environ.get("AZURE_RESOURCE_GROUP", "") # Replace with your resource group or set env var

if not resource_group_name:
    print("Please set the AZURE_RESOURCE_GROUP environment variable or provide a default.")
    exit(1)

print(f"Listing data factories in resource group '{resource_group_name}':")
try:
    factories = data_factory_client.factories.list_by_resource_group(resource_group_name)
    found_factories = False
    for factory in factories:
        print(f"- Factory Name: {factory.name}, Location: {factory.location}")
        found_factories = True
    if not found_factories:
        print("No data factories found in this resource group.")
except Exception as e:
    print(f"Error listing data factories: {e}")
    print("Make sure the resource group exists and your credential has 'Contributor' or 'Data Factory Contributor' role.")

print("Quickstart complete.")