Using Azure App Configuration in .NET 5 Functions

Azure .NET 5 Functions

A lot has been said about .NET 5 support for Azure Functions. The most important news: it is now officially supported! 🥳 To enable running
Azure Functions with .NET 5, the new Isolated Model enables Functions to run as an out-of-process language worker separate from the
Azure Functions runtime. This way you'll have full control over the Function's dependencies on one hand, and new features like a middleware
pipeline on the other. For an awesome write-up about everything around this, check out Oscar van Tol's post
Azure Functions in .NET 5 and beyond.

Azure App Configuration

Azure App Configuration provides a service to centrally manage application settings and feature flags. Modern programs, especially programs
running in a cloud, generally have many components that are distributed in nature. Spreading configuration settings across these components
can lead to hard-to-troubleshoot errors during an application deployment. Use App Configuration to store all the settings for your
application and secure their accesses in one place.

What is Azure App Configuration?

For an introduction to Azure App Configuration you can check my previous post
Azure App Configuration: an introduction.

Bringing the two together

To enable using App Configuration in .NET 5 Functions, you can use the AddAzureAppConfiguration extension method to add App Configuration
as a source to the IConfigurationBuilder. This extension method can be found in NuGet package
Microsoft.Extensions.Configuration.AzureAppConfiguration.
By doing this, you can simply call GetValue on IConfiguration to get a value. Based on the order you add configuration sources to the
builder, you will be able to fallback through different source.

For example: if you first add Environment Variables (which includes settings.json files, too!) and add App Configuration second, asking for
a setting 'Setting1' will first try to get that value from App Configuration. If it can't be found there, the application will search for
the setting 'Setting1' in Environment variables.

Wiring up your Functions App

Program.cs:

public static void Main()
{
    var host = new HostBuilder()
        .ConfigureAppConfiguration(c =>
        {
            // Add Environment Variables since we need to get the App Configuration connection string from settings.
            c.AddEnvironmentVariables();
            // Call Build() so we can get values from the IConfiguration.
            var config = c.Build();
            var connectionString = config.GetValue<string>("AppConfigurationConnectionString");
            // Add Azure App Configuration with the connectionstring.
            c.AddAzureAppConfiguration(connectionString);
        })
        .ConfigureFunctionsWorkerDefaults()
    v.Build();

    host.Run();
}

Getting a value from App Configuration in a Function

Triggered.cs (an HttpTriggered Function):

public class Triggered
{
    // Private field to hold IConfiguration
    private IConfiguration _configuration;

    // Constructor that gets IConfiguration injected
    public Triggered(IConfiguration configuration)
    {
        _configuration = configuration;    
    }

    [Function("Triggered")]
    public HttpResponseData Run([HttpTrigger(AuthorizationLevel.Function, "get")] HttpRequestData req,
        FunctionContext executionContext)
    {
        var response = req.CreateResponse(HttpStatusCode.OK);
        response.Headers.Add("Content-Type", "text/plain; charset=utf-8");
        // Call GetVallue<string> on IConfiguration to get the setting, this calls out to App Configuration
        var settingOne = _configuration.GetValue<string>("Setting1");
        response.WriteString($"The value of the setting is {settingOne}!");

        return response;
    }
}

I tried this with a complete connection string, so including the access key. Of course this can also work without an access key using a
Managed Identity. This eliminates the need for having access keys in code or configuration.

Hope this helps. Happy coding!