Azure Functions with SSE

• This sample application demonstrates how to implement SSE (Server-Sent Events) in Azure Functions.
• The UI code has been created using Chainlit.
• The backend is developed using the Azure Function App Python SDK.
• The SDK and implementation use Azure Function Programming model v2.

Endpoint: HTTP Trigger

• azure: https://<APP_NAME>.azurewebsites.net/api/<FUNCTION_NAME>
• local: http://localhost:<PORT>/api/<FUNCTION_NAME>
  • [GET] http://localhost:7071/api/stream-openai

Code and test Azure Functions locally

• Code and test Azure Functions locally

• Develop Azure Functions locally using Core Tools

• Install the Azure Functions Core Tools & Visual Code : Azure Extension

• Rename local.settings.template.json to local.settings.json - Fill your values in local.settings.json

• The command must be run in a virtual environment.

  • Functions: func start
  • UI application: chainlit run ui_app.py -w

• [Optional] Change venv path > .vscode > settings.json > "azureFunctions.pythonVenv": "venv"

Publish your local code to a function app in Azure

• cmd: func azure functionapp publish <FunctionAppName>

SSE（Server-Sent Events）

• Azure Functions: Support for HTTP Streams in Python is now in Preview!
  • When deploying, add the following application settings: "PYTHON_ENABLE_INIT_INDEXING": "1". If you are deploying to Linux Consumption, also add "PYTHON_ISOLATE_WORKER_DEPENDENCIES": "1". When running locally, you also need to add these same settings to the local.settings.json project file.