Ti ringraziamo per la visita. Questa pagina è per il momento disponibile solo in inglese.

Request Processing

Request handlers

Request handlers are responsible for handling one or more types of incoming requests.

You can create request handlers by implementing the RequestHandler interface, or a typed request handler interface, both of which consist of two methods:

  • canHandle, which is called by the SDK to determine if the given handler is capable of processing the incoming request. This method returns true if the handler can handle the request, or false if not. You have the flexibility to choose the conditions on which to base this determination, including the type or parameters of the incoming request, or skill attributes.
  • handle, which is called by the SDK when invoking the request handler. This method contains the handler's request processing logic, and returns an optional Response.

Generic request handlers

The generic RequestHandler interface can be used when you want to handle multiple request types, useful if you have a similar code path across different types of requests.

The following example shows a request handler that is configured to handle an incoming SkillEnabledRequest or SkillDisabledRequest.

public class SkillEventsHandler implements RequestHandler {
    @Override
    public boolean canHandle(HandlerInput input) {
        return input.getRequest() instanceof SkillEnabledRequest
                || input.getRequest() instanceof SkillDisabledRequest;
    }

    @Override
    public Optional<Response> handle(HandlerInput input) {
        logger.info("A skill event was recieved");

        if (input.getRequest() instanceof SkillEnabledRequest) {
            SkillEnabledRequest request = (SkillEnabledRequest) input.getRequest();
            logger.info("User enabled the skill at {}", request.getEventCreationTime());
        }

        if (input.getRequest() instanceof SkillDisabledRequest) {
            SkillDisabledRequest request = (SkillDisabledRequest) input.getRequest();
            logger.info("User disabled the skill at {}", request.getEventCreationTime());
        }
    }
}

Typed request handlers

Request handler interfaces for each SDK supported request type (e.g. IntentRequest, LaunchRequest, SkillEnabledRequest) are available and provide a type safe way of writing handlers for a particular type of request. When using the generic RequestHandler interface you must do a type check, then cast to work against a particular request type. The typed handlers eliminate this requirement and handle the type checking for you. For example, a handler implementing the LaunchRequestHandler interface will only be invoked if the incoming request is a LaunchRequest.

A list of the currently available type specific request handlers is available here.

The following example shows a request handler that is configured to handle the HelloWorldIntent. As you can see, the handler implements the IntentRequestHandler interface since we are going to be handling an intent request.

public class HelloWorldIntentHandler implements IntentRequestHandler {
    @Override
    public boolean canHandle(HandlerInput input, IntentRequest intentRequest) {
        return intentRequest.getIntent().getName().equals("HelloWorldIntent");
    }

    @Override
    public Optional<Response> handle(HandlerInput input, IntentRequest intentRequest) {
        String speechText = "Hello world";

        Intent intent = intentRequest.getIntent();
        String intentName = intent.getName();

        return input.getResponseBuilder()
                .withSpeech(speechText)
                .withSimpleCard(intentName, speechText)
                .build();
    }
}

This type of handler will only be invoked if the incoming request is an IntentRequest. The handler's canHandle method is configured to check the request's intent name, and then return true if the intent name is HelloWorldIntent. A basic "Hello world" response is then generated and returned by the handle method. By modifying the canHandle condition, you can choose to make this handler more granular (for example, by inspecting Intent slot values), or more generic by returning true for any IntentRequest.

CanHandle Predicates

The SDK includes a set of prebuilt canHandle predicates that make it easy to evaluate common conditions, such as intent name, slot values, and attributes. These make it simpler and more concise to write complex canHandle conditional logic. A list of the available Predicates provided with the SDK is available here.

For example, the above sample handler for a HelloWorld intent can have its CanHandle condition simplified by using Predicates:

@Override
public boolean canHandle(HandlerInput input, IntentRequest intentRequest) {
    return input.matches(Predicates.intentName("HelloWorldIntent"));
}

Likewise, the generic RequestHandler example showing a handler for both SkillEnabledRequest or SkillDisabledRequest would look like this:

@Override
public boolean canHandle(HandlerInput input) {
    return input.matches(Predicates.requestType(SkillEnabledRequest.class)
        .or(Predicates.requestType(SkillDisabledRequest.class)));
}

Handler processing order

The SDK calls the canHandle methods on its request handlers in the order in which they were provided to the Skill builder. You can implement as may request handlers as you like for a skill, including multiple handlers for the same type of request (e.g. multiple handlers implementing IntentRequestHandler). In this case, the SDK will access these handlers in the order they were registered and utilize the handler that's canHandle method returns true first.

return Skills.standard()
    .addHandlers(new FooHandler(), new BarHandler(), new BazHandler())
    .build();

In this example, the SDK calls request handlers in the following order:

  1. FooHandler
  2. BarHandler
  3. BazHandler

The SDK always chooses the first handler that is capable of handling a given request. In this example, if both FooHandler and BarHandler are capable of handling a particular request, FooHandler is always be invoked. Keep this in mind when designing and registering request handlers.

Request and response interceptors

The SDK supports request and response interceptors that execute before and after RequestHandler execution, respectively. You can implement request interceptors by implementing the RequestInterceptor interface, or the ResponseInterceptor interface for response interceptors.

Both interceptor interfaces expose a single process method with a void return type. Request interceptors have access to the HandlerInput instance, while response interceptors have access to the HandlerInput as well as the Optional<Response> produced by the RequestHandler.

public class PersistenceSavingResponseInterceptor implements ResponseInterceptor {
    @Override
    public void process(HandlerInput input, Optional<Response> output) {
        input.getAttributesManager().savePersistentAttributes();
    }
}

Request interceptors are invoked immediately prior to execution of the request handler for an incoming request. Request attributes provide a way for request interceptors to pass data and entities on to request handlers.

Response interceptors are invoked immediately after execution of the request handler. Because response interceptors have access to the output generated from execution of the request handler, they are ideal for tasks such as response sanitization and validation.

The following example shows how to register an interceptor with the SDK on the Skill builder:

return Skills.standard()
    .addHandlers(new FooHandler(), new BarHandler(), new BazHandler())
    .addRequestInterceptors(new FooRequestInterceptor())
    .addResponseInterceptors(new BarResponseInterceptor())
    .build();

Exception handlers

Exception handlers are similar to request handlers, but are instead responsible for handling one or more types of exceptions. They are invoked by the SDK when an unhandled exception is thrown during the course of request processing.

All exception handlers must implement the ExceptionHandler interface, consisting of the following two methods:

  • canHandle, which is called by the SDK to determine if the given handler is capable of handling the exception. This method returns true if the handler can handle the exception, or false if not. A catch-all handler can be easily introduced by simply returning true in all cases.
  • handle, which is called by the SDK when invoking the exception handler. This method contains all exception handling logic, and returns an output which optionally may contain a Response.

The following example shows an example exception handler that is configured to handle any exception of type AskSdkException.

public class MyExecptionHandler implements ExceptionHandler {
    @Override
    public boolean canHandle(HandlerInput input, Throwable throwable) {
        return throwable instanceof AskSdkException;
    }

    @Override
    public HandlerOutput handle(HandlerInput input, Throwable throwable) {
        return input.getResponseBuilder()
                    .withSpeech("An error was encountered while handling your request. Try again later.")
                    .build();
    }
}

The handler's canHandle method returns true if the incoming exception is an instance of AskSdkException. The handle method returns a graceful error response to the user.

Exception handlers are executed similarly to request handlers, where the SDK accesses handlers in the order in which they were provided to the Skill.

Handler Input

Request handlers, request and response interceptors, and exception handlers are all passed a HandlerInput instance when invoked. This class exposes various entities useful in request processing, including:

  • RequestEnvelope: Contains the incoming Request and other context.
  • AttributesManager: Provides access to request, session, and persistence attributes.
  • ServiceClientFactory: Constructs service clients capable of calling Alexa APIs.
  • ResponseBuilder: Helps to build responses.
  • Context: Provides an optional, context object passed in by the host container. For example, for skills running on AWS Lambda, this is the context object for the AWS Lambda function.