Azure Functions with Python

Azure functions

• Azure Functions allow you to run Python scripts and functions in Azure without having to create infrastructure.
• Azure Functions provide an alternative to setting up infrastructure-heavy tools like DataBricks, VM instances, and others.
• Azure Functions streamline the deployment and execution of Python code by managing the underlying infrastructure.

Azure Function Concepts

• Triggers: Determine when a function runs (e.g., on a schedule, file addition, or an event).
• Input Bindings: Connect functions to data sources or objects (e.g., fetching a file from a storage account).
• Output Bindings: Send function results to other services or destinations (e.g., writing data to a storage account file).

Local Development Setup

• Install the Azure Function Core Tools for local execution.
• Create a new Azure Functions project using func init --python <project_name>.
• Create and activate a Python virtual environment using python -m venv <env_name> and activate it using the appropriate script for your operating system (e.g., .\envName\Scripts\activate.ps1).

Exploring Directory Structure

• VNV: Python virtual environment, isolating project dependencies.
• VS Code: Visual Studio Code folder with function-specific settings.
• .gitignore: Excludes files and folders from Git commits.
• functionapp: Contains function code and runtime dependencies for local execution.
• host.json: Configuration file for function behavior (application insights, versioning, extensions).
• local.settings.json: Local settings (connection strings, key-value pairs, environment variables).
• requirements.txt: Lists Python package dependencies.

Key Points

• local.settings.json allows use of the AzureWebJobsStorage key with developmentStorage=true for local Azure Storage Emulator connection.
• Encrypt secrets in local.settings.json using settings add --encrypted for secure public environments.

Creating Functions

• Use app.func to define functions within the functionapp folder.
• Decorators (@) define function attributes (e.g., @app.func, @app.route).
• @app.func registers a function for deployment.
• @app.route defines an HTTP function's endpoint path.
• anon = True provides anonymous access to HTTP functions.
• The req parameter accesses incoming HTTP request data.

Example Function Code

from azure.functions import HttpRequest, HttpResponse

def main(req: HttpRequest) -> HttpResponse:
    name = req.params.get('name')
    if name:
        return HttpResponse(f"Hello {name}!")
    else:
        return HttpResponse(f"Hello World!")

Project Set Up

• The example project demonstrates an HTTP trigger function.
• The function app contains one or more functions.
• requirements.txt installs Python packages using pip install -r requirements.txt.
• Always activate the virtual environment before running code.
• Visual Studio Code's Azure Functions extension helps manage the project (optional).
• Command-line alternatives exist for Azure Functions management.

Azure Functions - HTTP Requests

• Functions use the @app.route decorator.
• A response returns a status code and message content.
• funk start starts the local function app.
• funk stop stops the local function app.
• Functions accept request parameters (e.g., ?name=<value>).

Azure Functions - Storage Triggers

• The Azure Storage Emulator is for local development.
• Azure Storage Explorer allows interacting with the emulator.
• You can create a local storage account with Azure Storage Explorer.
• Use the @func.blob_trigger decorator for Blob Trigger functions.
• The connection parameter sets the storage connection string (Azure Web Jobs Storage, Azure Emulator).
• Blob trigger functions respond to file uploads to a specified container and path.

Azure Functions - Blob Storage Triggers

• Blob trigger functions read uploaded files to a container path.
• func.input_stream reads data from the blob trigger function.
• Modules like csv and codecs process Blob Storage data.

Azure Functions - Organizing Functions

• Functions group into blueprints using func.blueprint for better organization and maintainability.
• Blueprints register into the Azure Function container using app.register_blueprint.

Learn about Azure Functions for running Python scripts in Azure. Explore triggers, input bindings, and output bindings for efficient data processing. Simplify deployment with infrastructure management.