Once you have implemented your Request Handler, you will want to dispatch Commands or Events to that Handler.
Registering a Handler¶
In order for a Command Dispatcher to find a Handler for your Command or Event you need to register the association between that Command or Event and your Handler.
The Subscriber Registry is where you register your Handlers.
var subscriberRegistry = new SubscriberRegistry(); subscriberRegistry.Register<GreetingCommand, GreetingCommandHandler>();
Once you have registered your Handlers, you can dispatch requests to them. To do that you simply use the CommandProcessor.Send() method passing in an instance of your command.
Building a Command Dispatcher¶
We associate a Subscriber Registry with a Command Processor by passing it into the constructor of the Command Processor. For convenience, we provide a Commmand Processor Builder that helps you configure new instances of Command Processor.
var logger = LogProvider.For<Program>(); var registry = new SubscriberRegistry(); registry.Register<GreetingCommand, GreetingCommandHandler>(); var builder = CommandProcessorBuilder.With() .Handlers(new HandlerConfiguration( subscriberRegistry: registry, handlerFactory: new SimpleHandlerFactory(logger) )) .DefaultPolicy() .NoTaskQueues() .RequestContextFactory(new InMemoryRequestContextFactory()); var commandProcessor = builder.Build();
We cover configuration of a **Command Processor** in more detail later.
Returning results to the caller.¶
We use Command-Query separation so a Command does not have return value and CommandDispatcher.Send() does not return anything.
This in turn leads to a set of questions that we need to answer about common scenarios:
- How do I handle failure? With no return value, what do I do if my handler fails
- How do I pass information back to the caller? Creation scenarios particularly seem to require the caller knows about identies for created entities.
We discuss these issues below.
If we don’t allow return values, what do you do on failure?
- The basic failure strategy is to throw an exception. This will terminate the request handling pipeline.
- If you want to support Retry, and Circuit Breaker you can use our support for Polly Policies
- You can also build your own exception handling into your Pipeline.
- Finally you can use our support for a Fallback handler to provide backstop exception handling.
Passing Information to the Caller¶
Sometimes you need to provide information to the caller about the success of the operation. The most common requirement is the Identity of a new created Entity so that you can query for it. For example you are implementing a REST API and in response to a POST request you create a new entity and want to return the entity body in the HTTP response body.
The best approach is to generate the Identity to use for the new Entity and pass that as a parameter on the Command, such as Guid or Hi-Lo identity.
But what if you are not be able to do this or want to support it for performance reasons?
In that case add a property to the Command that you can initialize from the Handler, for example create a NewEntityIdentity property in your command that you write the new entity’s identity to in the Handler, and then inspect the property in your Command in the calling code after the call to CommandDispatcher.Send() completes.
Note that you cannot use this strategy with CommandDispatcher.Send() as you have no way to update the Command in process.
Using the base class when dispatching a message¶
All Command or Event messages derive from IRequest and ICommand and IEvent respectively. So it may seem natural to create a collection of them, for example List<IRequest>, and then process a set of messages by enumerating over them.
When you try this, you will encounter the issue that we dispatch based on the concrete type of the Command or Event. In other words the type you register via the SubscriberRegistry. Because CommandProcessor.Send() is actually CommandProcessor.Send<T>() you need to provide the concrete type in the call for the compiler to determine the type to use with the cool as the concrete type.
If you try this:
ICommand command = new GreetingCommand("Ian"); commandProcessor.Send(command);
Then you will get this error: “ArgumentException “No command handler was found for the typeof command Brighter.commandprocessor.ICommand - a command should have exactly one handler.”“
Now, you don’t see this issue if you pass the concrete type in, so the compiler can correctly resolve the run-time type.
So what can you do if you must pass the base class to the Command Processor i.e. because you are using a list.
The workaround is to use the dynamic keyword. Using the dynamic keyword means that the type will be evaluated using RTTI, which will successfully pick up the type that you need.
ICommand command = new GreetingCommand("Ian"); commandProcessor.Send((dynamic)command);
See this discussion for more.