playframework – 在play framework 2.4.x中使用依赖注入测试actor

前端之家收集整理的这篇文章主要介绍了playframework – 在play framework 2.4.x中使用依赖注入测试actor前端之家小编觉得挺不错的,现在分享给大家,也给大家做个参考。
如何测试依赖注入创建的actor?在我的应用程序中,我可以通过命名注入获得ActorRef:
public MyClass {
    @Inject
    @Named("ping")
    ActorRef mPingRef;
}

如何在我的测试中获得此参考?

这是我的演员:

public class PingActor extends UntypedActor {
    @Inject
    public PingActor(Configuration configuration) {
         ... // Use config
    }


    @Override
    public void onReceive(Object message) throws Exception {
        if (message instanceof Ping) {
            getSender().tell(new Pong(),getSelf());
        }
    }

    public static class Ping {}
    public static class Pong {}
}

我已经使用自己的模块配置了我的应用程序:

public class MyModule extends AbstractModule implements AkkaGuiceSupport {
    private final Configuration mConfig;

    public MyModule(Environment environment,Configuration configuration){
        this.mConfig = configuration;
    }

    @Override
    protected void configure() {
        bindActor(PingActor.class,"ping");
    }
}

该模块在application.conf中启用:

play.modules.enabled += "com.my.package.MyModule"
这个解决方案适用于PlayScala,但它应该与PlayJava的机制相同:

所以我得到了我的GuiceModule:

class CommonModule extends AbstractModule with AkkaGuiceSupport {
  override def configure(): Unit = {
    bindActor[SomeActor]("actor-name")
  }
}

然后测试(我从我的测试中删除了一些东西,所以它可能无法直接编译):

import akka.actor.{ActorRef,ActorSystem}
import akka.testkit.{TestKit,TestProbe}
import module.CommonModule
import org.specs2.mutable.Specification
import org.specs2.specification.Scope
import play.api.inject._
import play.api.inject.guice.GuiceApplicationBuilder
import play.api.test.Helpers._

class SwitchUpdateActorSpec extends Specification {

  "MyActor" should {

    val actorSystem = ActorSystem("test")
    class Actors extends TestKit(actorSystem) with Scope

    val app = new GuiceApplicationBuilder(modules = Seq(new CommonModule))
      .overrides(bind[ActorSystem].toInstance(actorSystem))
      .build()


    "respond with 'ok' upon receiving a message" in new Actors {
      running(app) {
        private val injector: Injector = app.injector
        private val actor: ActorRef = injector.instanceOf(BindingKey(classOf[ActorRef]).qualifiedWith("actor-name"))

        val probe = TestProbe()
        actor.tell("hi there!",probe.ref)

        probe.expectMsg("ok")
      }
    }
  }    
}

所以我做的是:

>创建一个全新的ActorSystem
>将该actorSystem包装在Akka的TestKit中(libraryDependencies =“com.typesafe.akka”%%“akka-testkit”%“2.4.1”)
>使用GuiceApplicationBuilder应用覆盖
>然后直接使用app.injector访问我的guice配置的actor

当您查看在MyModule.configure()方法中使用的bindActor的实现时,会发生很明显的情况:

def bindActor[T <: Actor: ClassTag](name: String,props: Props => Props = identity): Unit = {
    accessBinder.bind(classOf[ActorRef])
      .annotatedWith(Names.named(name))
      .toProvider(Providers.guicify(Akka.providerOf[T](name,props)))
      .asEagerSingleton()
  }

猜你在找的设计模式相关文章