Integrate Coyote with a unit testing framework

Common unit testing frameworks like MSTest, xUnit.net and nunit cannot easily call the coyote command line tool for testing. In this case you can use the Coyote TestingEngine directly.

Note that before running a Coyote test, you must first use coyote rewrite from the coyote command line tool (typically as a post-build task) to rewrite your binaries so that Coyote can inject logic that takes control of the concurrent execution during testing. Learn more about this step in rewriting binaries.

The Coyote TestingEngine is included in the Microsoft.Coyote.Test package. The following shows a complete example using xUnit. The project simply includes xUnit and the Coyote packages:

<Project Sdk="Microsoft.NET.Sdk">
  <PropertyGroup>
    <TargetFramework>net8.0</TargetFramework>
  </PropertyGroup>
  <ItemGroup>
    <PackageReference Include="Microsoft.Coyote" Version="1.4.1" />
    <PackageReference Include="Microsoft.Coyote.Test" Version="1.4.1" />
    <PackageReference Include="xunit" Version="2.4.2" />
    <PackageReference Include="xunit.runner.visualstudio" Version="2.4.5">
      <PrivateAssets>all</PrivateAssets>
      <IncludeAssets>runtime; build; native; contentfiles; analyzers; 
                     buildtransitive</IncludeAssets>
    </PackageReference>
  </ItemGroup>
</Project>

And then your [Fact] method which runs as an xUnit unit test can create a TestingEngine to run a Coyote test method. This test code can run in the Visual Studio Test Explorer or from a dotnet test command line:

public class Test
{
    ITestOutputHelper Output;

    public Test(ITestOutputHelper output)
    {
        this.Output = output;
    }


    [Fact(Timeout = 5000)]
    public void RunCoyoteTest()
    {
        var config = Configuration.Create();
        TestingEngine engine = TestingEngine.Create(config, CoyoteTestMethod);
        engine.Run();
        var report = engine.TestReport;
        Output.WriteLine("Coyote found {0} bug.", report.NumOfFoundBugs);
        Assert.True(report.NumOfFoundBugs == 0, $"Coyote found {report.NumOfFoundBugs} bug(s).");
    }

    private async Task CoyoteTestMethod()
    {
        // This is running as a Coyote test.
        await Task.Delay(10);
        Specification.Assert(false, "This test failed!");
    }
}

This will produce the following test output because the Specification.Assert is hard wired to fail.

Coyote found 1 bug.

Most of the command line options you see on coyote test are available in the Configuration class. Use the With* helper methods to set the various configurations, for example, to specify --strategy prioritization together with --strategy-value 10 use the following:

var config = Configuration.Create().WithPrioritizationStrategy(false, 10);

For --iterations use WithTestingIterations and so on.

If you want the rich Coyote log files, you can use the TryEmitTraces method on the TestingEngine to produce those log files in the folder of your choice like this:

List<string> filenames = new List<string>(engine.TryEmitTraces("d:\\temp\\test", "mytest"));
foreach (var item in filenames)
{
    Output.WriteLine("See log file: {0}", item);
}

Note: TryEmitTraces is an iterator method, which means you must iterate the result in order to produce the log files. You will see the following output:

Coyote found 1 bugs
See log file: d:\temp\test\mytest_0.txt
See log file: d:\temp\test\mytest_0.trace

And the log file contains the familiar output of coyote test as follows:

<TestLog> Running test.
<ErrorLog> This test failed!
<StackTrace>
at Microsoft.Coyote.SystematicTesting.OperationScheduler.NotifyAssertionFailure(
    String text, Boolean killTasks, Boolean cancelExecution)
   at Microsoft.Coyote.SystematicTesting.ControlledRuntime.Assert(Boolean predicate, String s)
   at ConsoleApp14.Test.CoyoteTestMethod()
   at System.Runtime.CompilerServices.AsyncMethodBuilderCore.Start[TStateMachine]()
   at ConsoleApp14.Test.CoyoteTestMethod()
   at Microsoft.Coyote.SystematicTesting.ControlledRuntime.RunTestb__0d.MoveNext()
   at System.Runtime.CompilerServices.AsyncMethodBuilderCore.Start[TStateMachine]()
   at Microsoft.Coyote.SystematicTesting.ControlledRuntime.RunTestb__0()
   at System.Threading.Tasks.Task.InnerInvoke()
   at System.Threading.Tasks.Task.c.cctorb__274_0(Object obj)
   at System.Threading.ExecutionContext.RunFromThreadPoolDispatchLoop(
       Thread threadPoolThread, ExecutionContext executionContext, ContextCallback callback)
   at System.Threading.Tasks.Task.ExecuteWithThreadLocal(Task currentTaskSlot)
   at System.Threading.Tasks.Task.ExecuteEntryUnsafe(Thread threadPoolThread)
   at System.Threading.Tasks.Task.ExecuteFromThreadPool(Thread threadPoolThread)
   at System.Threading.ThreadPoolWorkQueue.Dispatch()
   at System.Threading._ThreadPoolWaitCallback.PerformWaitCallback()

<TestLog> Exploration finished [found a bug using the 'random' strategy].
<StrategyLog> Testing statistics:
<StrategyLog> Found 1 bug.
<StrategyLog> Exploration statistics:
<StrategyLog> Explored 1 schedule: 1 fair and 0 unfair.
<StrategyLog> Found 100.00% buggy schedules.
<StrategyLog> Number of scheduling decisions in fair terminating schedules: 
              3 (min), 3 (avg), 3 (max).

The TestEngine.Create method has overloads for supporting Coyote test methods with the following signatures:

Action
Action<ICoyoteRuntime>
Action<IActorRuntime>
Func<Task>
Func<ICoyoteRuntime, Task>
Func<IActorRuntime, Task>

Notice that you never create an ICoyoteRuntime or IActorRuntime yourself, the TestingEngine will do that for you so it can provide the non-production systematic test version of those runtimes.

Replaying a trace

You can also easily replay and debug a trace, similar to using coyote replay from the command line tool. To do this you need to configure the TestingEngine to run in replay mode:

var trace = ...
var config = Configuration.Create().WithReproducibleTrace(trace);

The input to the WithReproducibleTrace method should either be the contents of a .trace file or the string value of TestingEngine.ReproducibleTrace (from a previous run).

Then you add breakpoints to debug and replay as follows:

var trace = ...
var config = Configuration.Create().WithReproducibleTrace(trace);
TestingEngine engine = TestingEngine.Create(config, CoyoteTestMethod);
engine.Run();

Testing actors

Actors run asynchronously, so you will need to design your actors in a way such that you know when they have finished doing what they are supposed to do. One way to do that is to use the Coyote TaskCompletionSource<bool> as follows:

class TestConfigEvent : Event
{
    public TaskCompletionSource<bool> Completed = TaskCompletionSource.Create<bool>();
}

private async Task CoyoteTestActors(IActorRuntime runtime)
{
    // this method can be run by the Coyote TestingEngine.
    TestConfigEvent config = new TestConfigEvent();
    runtime.CreateActor(typeof(MyTestActor), config);
    await config.Completed.Task;
    Output.WriteLine("Coyote actor test passed");
}

Where MyTestActor sets the result on the TaskCompletionSource as follows:

[OnEventDoAction(typeof(MyEvent), nameof(HandleEvent))]
class MyTestActor : Actor
{
    TestConfigEvent config;

    protected override System.Threading.Tasks.Task OnInitializeAsync(Event initialEvent)
    {
        config = (TestConfigEvent)initialEvent;
        var actor = this.CreateActor(typeof(MyActor));
        this.SendEvent(actor, new MyEvent() { Caller = this.Id });

        return base.OnInitializeAsync(initialEvent);
    }

    private void HandleEvent(Event e)
    {
        config.Completed.SetResult(true);
    }
}

This test actor creates MyActor, sends an event to it, waits for a response, then sets the TaskCompletionSource result. MyActor is a simple ping-pong style actor:

class MyEvent : Event
{
    public ActorId Caller;
}

[OnEventDoAction(typeof(MyEvent), nameof(HandleEvent))]
class MyActor : Actor
{
    private void HandleEvent(Event e)
    {
        ActorId caller = ((MyEvent)e).Caller;
        this.SendEvent(caller, new MyEvent() { Caller = this.Id });
    }
}

If you run this test setting WithXmlLogEnabled(true) on the Configuration you will get the following DGML diagram showing you what happened during this test:

unittest

See the following API documentation for more information: