How can I import modules for a Python Azure Function?

Azure Functions for Python provides several methods to import and use external modules. While early versions had limitations, modern Azure Functions offers better support for dependency management.

Method 1: Using requirements.txt (Recommended)

The most straightforward approach is to create a requirements.txt file in your function app root directory −

# requirements.txt
requests==2.28.2
pandas==1.5.3
numpy==1.24.3

Azure Functions will automatically install these dependencies during deployment. Then import them normally in your function code −

import azure.functions as func
import requests
import pandas as pd

def main(req: func.HttpRequest) -> func.HttpResponse:
    # Use your imported modules
    response = requests.get('https://api.example.com/data')
    df = pd.DataFrame(response.json())
    
    return func.HttpResponse(f"Processed {len(df)} records")

Method 2: Using Virtual Environment (Manual Setup)

For local development or manual deployment, you can use a virtual environment approach −

Steps to Set Up

1. Create your Python script in Azure Functions

2. Open Kudu console and navigate to your script location

3. Create a virtual environment: python -m venv myvenv

4. Activate the environment: myvenv\Scripts\activate.bat (Windows) or source myvenv/bin/activate (Linux)

5. Install dependencies: pip install package_name

Then modify your function to include the virtual environment path −

import sys
import os.path

# Add virtual environment to Python path
venv_path = os.path.abspath(os.path.join(os.path.dirname(__file__), 'myvenv/Lib/site-packages'))
sys.path.append(venv_path)

# Now import your modules
import requests
import pandas as pd

Method 3: Bundling Modules Directly

For small modules or when you need specific control, you can upload modules directly to your function directory −

import sys
import os

# Add current directory to path for local modules
current_dir = os.path.dirname(__file__)
sys.path.insert(0, current_dir)

# Import your local module
import my_custom_module

def main(req):
    result = my_custom_module.process_data()
    return result

Comparison

Conclusion

Use requirements.txt for standard Python packages as it provides automatic dependency management. For custom modules or complex setups, consider virtual environments or direct bundling approaches.