Documentation

You are viewing the documentation for the 2.6.0-M4 development release. The latest stable release series is 3.0.x.

§Handling asynchronous results

§Make controllers asynchronous

Internally, Play Framework is asynchronous from the bottom up. Play handles every request in an asynchronous, non-blocking way.

The default configuration is tuned for asynchronous controllers. In other words, the application code should avoid blocking in controllers, i.e., having the controller code wait for an operation. Common examples of such blocking operations are JDBC calls, streaming API, HTTP requests and long computations.

Although it’s possible to increase the number of threads in the default execution context to allow more concurrent requests to be processed by blocking controllers, following the recommended approach of keeping the controllers asynchronous makes it easier to scale and to keep the system responsive under load.

§Creating non-blocking actions

Because of the way Play works, action code must be as fast as possible, i.e., non-blocking. So what should we return from our action if we are not yet able to compute the result? We should return the promise of a result!

Java 8 provides a generic promise API called CompletionStage. A CompletionStage<Result> will eventually be redeemed with a value of type Result. By using a CompletionStage<Result> instead of a normal Result, we are able to return from our action quickly without blocking anything. Play will then serve the result as soon as the promise is redeemed.

§How to create a CompletionStage<Result>

To create a CompletionStage<Result> we need another promise first: the promise that will give us the actual value we need to compute the result:

CompletionStage<Double> promiseOfPIValue = computePIAsynchronously();
// Runs in same thread
CompletionStage<Result> promiseOfResult = promiseOfPIValue.thenApply(pi ->
                ok("PI value computed: " + pi)
);

Play asynchronous API methods give you a CompletionStage. This is the case when you are calling an external web service using the play.libs.WS API, or if you are using Akka to schedule asynchronous tasks or to communicate with Actors using play.libs.Akka.

In this case, using CompletionStage.thenApply will execute the completion stage in the same calling thread as the previous task. This is fine when you have a small amount of CPU bound logic with no blocking.

A simple way to execute a block of code asynchronously and to get a CompletionStage is to use the CompletionStage.supplyAsync() method:

// import static java.util.concurrent.CompletableFuture.supplyAsync;
// creates new task
CompletionStage<Integer> promiseOfInt = supplyAsync(() ->
        intensiveComputation());

Using supplyAsync creates a new task which will be placed on the fork join pool, and may be called from a different thread – although, here it’s using the default executor, and in practice you will specify an executor explicitly.

Only the “*Async” methods from CompletionStage provide asynchronous execution.

§Using HttpExecutionContext

You must supply the HTTP execution context explicitly as an executor when using a Java CompletionStage inside an Action, to ensure that the HTTP.Context remains in scope. If you don’t supply the HTTP execution context, you’ll get “There is no HTTP Context available from here” errors when you call request() or other methods that depend on Http.Context.

You can supply the play.libs.concurrent.HttpExecutionContext instance through dependency injection:

import play.libs.concurrent.HttpExecutionContext;
import play.mvc.*;

import javax.inject.Inject;
import java.util.concurrent.CompletableFuture;
import java.util.concurrent.CompletionStage;

public class MyController extends Controller {

    private HttpExecutionContext httpExecutionContext;

    @Inject
    public MyController(HttpExecutionContext ec) {
        this.httpExecutionContext = ec;
    }

    public CompletionStage<Result> index() {
        // Use a different task with explicit EC
        return calculateResponse().thenApplyAsync(answer -> {
            // uses Http.Context
            ctx().flash().put("info", "Response updated!");
            return ok("answer was " + answer);
        }, httpExecutionContext.current());
    }

    private static CompletionStage<String> calculateResponse() {
        return CompletableFuture.completedFuture("42");
    }
}

Please see Java thread locals for more information on using Java thread locals and HttpExecutionContext.

§Using CustomExecutionContext and HttpExecution

Using a CompletionStage or an HttpExecutionContext is only half of the picture though! At this point you are still on Play’s default ExecutionContext. If you are calling out to a blocking API such as JDBC, then you still will need to have your ExecutionStage run with a different executor, to move it off Play’s rendering thread pool. You can do this by creating a subclass of play.libs.concurrent.CustomExecutionContext with a reference to the custom dispatcher.

Add the following imports:

import play.libs.concurrent.HttpExecution;
import java.util.concurrent.Executor;
import java.util.concurrent.CompletionStage;
import static java.util.concurrent.CompletableFuture.supplyAsync;

Define a custom execution context:

interface MyExecutionContext extends ExecutionContextExecutor {}

// bind MyExecutionContext to MyExecutionContextImpl through DI
public class MyExecutionContextImpl
        extends CustomExecutionContext
        implements MyExecutionContext {

    @javax.inject.Inject
    public MyExecutionContextImpl(ActorSystem actorSystem) {
        // uses a custom thread pool defined in application.conf
        super(actorSystem, "my.dispatcher");
    }
}

You will need to define a custom dispatcher in application.conf, which is done through Akka dispatcher configuration.

Once you have the custom dispatcher, add in the explicit executor and wrap it with HttpException.fromThread:

public class Application extends Controller {

    private MyExecutionContext myExecutionContext;

    @javax.inject.Inject
    public Application(MyExecutionContext myExecutionContext) {
        this.myExecutionContext = myExecutionContext;
    }

    public CompletionStage<Result> index() {
        // Wrap an existing thread pool, using the context from the current thread
        Executor myEc = HttpExecution.fromThread((Executor) myExecutionContext);
        return supplyAsync(() -> intensiveComputation(), myEc)
                .thenApplyAsync(i -> ok("Got result: " + i), myEc);
    }

    public int intensiveComputation() { return 2;}
}

You can’t magically turn synchronous IO into asynchronous by wrapping it in a CompletionStage. If you can’t change the application’s architecture to avoid blocking operations, at some point that operation will have to be executed, and that thread is going to block. So in addition to enclosing the operation in a CompletionStage, it’s necessary to configure it to run in a separate execution context that has been configured with enough threads to deal with the expected concurrency. See Understanding Play thread pools for more information.

§Actions are asynchronous by default

Play actions are asynchronous by default. For instance, in the controller code below, the returned Result is internally enclosed in a promise:

public Result index() {
    return ok("Got request " + request() + "!");
}

Note: Whether the action code returns a Result or a CompletionStage<Result>, both kinds of returned object are handled internally in the same way. There is a single kind of Action, which is asynchronous, and not two kinds (a synchronous one and an asynchronous one). Returning a CompletionStage is a technique for writing non-blocking code.

§Handling time-outs

It is often useful to handle time-outs properly, to avoid having the web browser block and wait if something goes wrong. You can use play.libs.concurrent.Timeout.timeout method to wrap a CompletionStage in a non-blocking timeout.

class MyClass implements play.libs.concurrent.Timeout {
    CompletionStage<Double> callWithOneSecondTimeout() {
        return timeout(computePIAsynchronously(), Duration.ofSeconds(1));
    }
}

Note: Timeout is not the same as cancellation – even in case of timeout, the given future will still complete, even though that completed value is not returned.

Next: Streaming HTTP responses