stize / infrastructure Goto Github PK

The current unit testing approach doesn't support evaluating Pulumi values.

This was due to a Pulumi limitation (or lack of documentation) on how Output<T> values could be evaluated.

As per this Pulumi documentation, looks like now Pulumi allows the evaluation of Outputs and is now possible to create more comprehensive Unit Tests:

It requires of a Extension method:

public static class TestingExtensions
{
    public static Task<T> GetValueAsync<T>(this Output<T> output)
    {
        var tcs = new TaskCompletionSource<T>();
        output.Apply(v => { tcs.SetResult(v); return v; });
        return tcs.Task;
    }
}

And now we can evaluate Output values:

[Fact]
public async Task ResourceGroupHasEnvironmentTag()
{
    var resources = await TestAsync();
    var resourceGroup = resources.OfType<ResourceGroup>().First();

    var tags = await resourceGroup.Tags.GetValueAsync();
    tags.Should().NotBeNull("Tags must be defined");
    tags.Should().ContainKey("Environment");
}

We need to provide basic support to Azure Storage, including

• Tiers
• Kinds
• Public/Private
• VNET Service Endpoints

Following this announce, we need to perform the migration from pulumi azure-nextgen to pulumi azure-native.

Add Azure Networking support, including:

• Network Security Groups
• Security Rules
• Network Interfaces
• Application Security Groups

All these networking extensions should support the standard naming and tagging strategy for Stize

Add most commonly used additional features to SQL database and server modules.

• Database
• Server

We currently have some quality metrics using codacy, but we are still missing code coverage, which definitely help us check how thoroughly tested our code is.

We need to complete the setup to generate and push code coverage information to codacy.

The codacy documentation recommends using open cover or dotcover for this, but unfortunately it doesn't work with Unix systems, which breaks our multi-platform support.

Instead, we should use Coverlet and generate the code coverage reports in opencover format.

• Add coverlet to the project and make sure it is able to generate code coverage information in opencover format
• Update the cd pipeline to generate the code coverage information and push result to Codacy
• Add the Codacy API key to the Github Actions environment (Request support to one of the project owners)

Pulumi migration issues:
Mock methods NewResourceAsync() and CallAsync have been updated to take new parameters.

This means the RunAsync() method in the Testing.cs class needs updating and ALL of the mocks need updating.

I've assigned myself this task, but I wanted to let people know that Mocks will be formatted slightly differently from now on.

See link for more information on the update.

Dependabot couldn't find a .(cs|vb|fs)proj for this project.

Dependabot requires a .(cs|vb|fs)proj to evaluate your project's current .NET dependencies. It had expected to find one at the path: /<anything>.(cs|vb|fs)proj.

If this isn't a .NET project, or if it is a library, you may wish to disable updates for it from within Dependabot.

View the update logs.

We have to add active geo-replication support.

Active geo-replication happens at database level. I would start focusing on the database-side of things (meaning users should create two SQL Servers manually). Once the geo-replication is working we can think if we add some extension methods that automatically creates a second SQL Server and adds the database into it.