Testing streams

Dependency

The Akka dependencies are available from Akka’s library repository. To access them there, you need to configure the URL for this repository.

sbt

resolvers += "Akka library repository".at("https://repo.akka.io/maven")

Maven

<project>
  ...
  <repositories>
    <repository>
      <id>akka-repository</id>
      <name>Akka library repository</name>
      <url>https://repo.akka.io/maven</url>
    </repository>
  </repositories>
</project>

Gradle

repositories {
    mavenCentral()
    maven {
        url "https://repo.akka.io/maven"
    }
}

To use Akka Stream TestKit, add the module to your project:

sbt

val AkkaVersion = "2.10.0+31-e778606c-SNAPSHOT"
libraryDependencies += "com.typesafe.akka" %% "akka-stream-testkit" % AkkaVersion % Test

Maven

<properties>
  <scala.binary.version>2.13</scala.binary.version>
</properties>
<dependencyManagement>
  <dependencies>
    <dependency>
      <groupId>com.typesafe.akka</groupId>
      <artifactId>akka-bom_${scala.binary.version}</artifactId>
      <version>2.10.0+31-e778606c-SNAPSHOT</version>
      <type>pom</type>
      <scope>import</scope>
    </dependency>
  </dependencies>
</dependencyManagement>
<dependencies>
  <dependency>
    <groupId>com.typesafe.akka</groupId>
    <artifactId>akka-stream-testkit_${scala.binary.version}</artifactId>
    <scope>test</scope>
  </dependency>
</dependencies>

Gradle

def versions = [
  ScalaBinary: "2.13"
]
dependencies {
  implementation platform("com.typesafe.akka:akka-bom_${versions.ScalaBinary}:2.10.0+31-e778606c-SNAPSHOT")

  testImplementation "com.typesafe.akka:akka-stream-testkit_${versions.ScalaBinary}"
}

Introduction

Verifying behavior of Akka Stream sources, flows and sinks can be done using various code patterns and libraries. Here we will discuss testing these elements using:

• simple sources, sinks and flows;
• sources and sinks in combination with TestProbeTestProbe from the akka-testkit module;
• sources and sinks specifically crafted for writing tests from the akka-stream-testkit module.

It is important to keep your data processing pipeline as separate sources, flows and sinks. This makes them testable by wiring them up to other sources or sinks, or some test harnesses that akka-testkit or akka-stream-testkit provide.

Built-in sources, sinks and operators

Testing a custom sink can be as simple as attaching a source that emits elements from a predefined collection, running a constructed test flow and asserting on the results that sink produced. Here is an example of a test for a sink:

Scala

copysourceval sinkUnderTest =
  Flow[Int].map(_ * 2).toMat(Sink.fold(0)(_ + _))(Keep.right)

val future = Source(1 to 4).runWith(sinkUnderTest)
val result = Await.result(future, 3.seconds)
assert(result == 20)

Java

copysourcefinal Sink<Integer, CompletionStage<Integer>> sinkUnderTest =
    Flow.of(Integer.class)
        .map(i -> i * 2)
        .toMat(Sink.fold(0, (agg, next) -> agg + next), Keep.right());

final CompletionStage<Integer> future =
    Source.from(Arrays.asList(1, 2, 3, 4)).runWith(sinkUnderTest, system);
final Integer result = future.toCompletableFuture().get(3, TimeUnit.SECONDS);
assertEquals(20, result.intValue());

The same strategy can be applied for sources as well. In the next example we have a source that produces an infinite stream of elements. Such source can be tested by asserting that first arbitrary number of elements hold some condition. Here the taketake operator and Sink.seqSink.seq are very useful.

Scala

copysourceval sourceUnderTest = Source.repeat(1).map(_ * 2)

val future = sourceUnderTest.take(10).runWith(Sink.seq)
val result = Await.result(future, 3.seconds)
assert(result == Seq.fill(10)(2))

Java

copysourcefinal Source<Integer, NotUsed> sourceUnderTest = Source.repeat(1).map(i -> i * 2);

final CompletionStage<List<Integer>> future =
    sourceUnderTest.take(10).runWith(Sink.seq(), system);
final List<Integer> result = future.toCompletableFuture().get(3, TimeUnit.SECONDS);
assertEquals(Collections.nCopies(10, 2), result);

When testing a flow we need to attach a source and a sink. As both stream ends are under our control, we can choose sources that tests various edge cases of the flow and sinks that ease assertions.

Scala

copysourceval flowUnderTest = Flow[Int].takeWhile(_ < 5)

val future = Source(1 to 10).via(flowUnderTest).runWith(Sink.fold(Seq.empty[Int])(_ :+ _))
val result = Await.result(future, 3.seconds)
assert(result == (1 to 4))

Java

copysourcefinal Flow<Integer, Integer, NotUsed> flowUnderTest =
    Flow.of(Integer.class).takeWhile(i -> i < 5);

final CompletionStage<Integer> future =
    Source.from(Arrays.asList(1, 2, 3, 4, 5, 6))
        .via(flowUnderTest)
        .runWith(Sink.fold(0, (agg, next) -> agg + next), system);
final Integer result = future.toCompletableFuture().get(3, TimeUnit.SECONDS);
assertEquals(10, result.intValue());

TestKit

Akka Stream offers integration with Actors out of the box. This support can be used for writing stream tests that use familiar TestProbeTestProbe from the akka-testkit API.

One of the more straightforward tests would be to materialize stream to a FutureCompletionStage and then use pipePatterns.pipe pattern to pipe the result of that future to the probe.

Scala

copysourceimport system.dispatcher
import akka.pattern.pipe

val sourceUnderTest = Source(1 to 4).grouped(2)

val probe = TestProbe()
sourceUnderTest.runWith(Sink.seq).pipeTo(probe.ref)
probe.expectMsg(3.seconds, Seq(Seq(1, 2), Seq(3, 4)))

Java

copysourcefinal Source<List<Integer>, NotUsed> sourceUnderTest =
    Source.from(Arrays.asList(1, 2, 3, 4)).grouped(2);

final TestKit probe = new TestKit(system);
final CompletionStage<List<List<Integer>>> future =
    sourceUnderTest.grouped(2).runWith(Sink.head(), system);
akka.pattern.Patterns.pipe(future, system.dispatcher()).to(probe.getRef());
probe.expectMsg(Duration.ofSeconds(3), Arrays.asList(Arrays.asList(1, 2), Arrays.asList(3, 4)));

Instead of materializing to a future, we can use a Sink.actorRefSink.actorRef that sends all incoming elements to the given ActorRefActorRef. Now we can use assertion methods on TestProbeTestProbe and expect elements one by one as they arrive. We can also assert stream completion by expecting for onCompleteMessage which was given to Sink.actorRef.

Scala

copysourcecase object Tick
val sourceUnderTest = Source.tick(0.seconds, 200.millis, Tick)

val probe = TestProbe()
val cancellable = sourceUnderTest
  .to(Sink.actorRef(probe.ref, onCompleteMessage = "completed", onFailureMessage = _ => "failed"))
  .run()

probe.expectMsg(1.second, Tick)
probe.expectNoMessage(100.millis)
probe.expectMsg(3.seconds, Tick)
cancellable.cancel()
probe.expectMsg(3.seconds, "completed")

Java

copysourcefinal Source<Tick, Cancellable> sourceUnderTest =
    Source.tick(Duration.ZERO, Duration.ofMillis(200), Tick.TOCK);

final TestKit probe = new TestKit(system);
final Cancellable cancellable =
    sourceUnderTest.to(Sink.actorRef(probe.getRef(), Tick.COMPLETED)).run(system);
probe.expectMsg(Duration.ofSeconds(3), Tick.TOCK);
probe.expectNoMessage(Duration.ofMillis(100));
probe.expectMsg(Duration.ofSeconds(3), Tick.TOCK);
cancellable.cancel();
probe.expectMsg(Duration.ofSeconds(3), Tick.COMPLETED);

Similarly to Sink.actorRef that provides control over received elements, we can use Source.actorRefSource.actorRef and have full control over elements to be sent.

Scala

copysourceval sinkUnderTest = Flow[Int].map(_.toString).toMat(Sink.fold("")(_ + _))(Keep.right)

val (ref, future) = Source
  .actorRef(
    completionMatcher = {
      case Done =>
        CompletionStrategy.draining
    },
    // Never fail the stream because of a message:
    failureMatcher = PartialFunction.empty,
    bufferSize = 8,
    overflowStrategy = OverflowStrategy.fail)
  .toMat(sinkUnderTest)(Keep.both)
  .run()

ref ! 1
ref ! 2
ref ! 3
ref ! Done

val result = Await.result(future, 3.seconds)
assert(result == "123")

Java

copysourcefinal Sink<Integer, CompletionStage<String>> sinkUnderTest =
    Flow.of(Integer.class)
        .map(i -> i.toString())
        .toMat(Sink.fold("", (agg, next) -> agg + next), Keep.right());

final Pair<ActorRef, CompletionStage<String>> refAndCompletionStage =
    Source.<Integer>actorRef(
            elem -> {
              // complete stream immediately if we send it Done
              if (elem == Done.done()) return Optional.of(CompletionStrategy.immediately());
              else return Optional.empty();
            },
            // never fail the stream because of a message
            elem -> Optional.empty(),
            8,
            OverflowStrategy.fail())
        .toMat(sinkUnderTest, Keep.both())
        .run(system);
final ActorRef ref = refAndCompletionStage.first();
final CompletionStage<String> future = refAndCompletionStage.second();

ref.tell(1, ActorRef.noSender());
ref.tell(2, ActorRef.noSender());
ref.tell(3, ActorRef.noSender());
ref.tell(Done.getInstance(), ActorRef.noSender());

final String result = future.toCompletableFuture().get(1, TimeUnit.SECONDS);
assertEquals("123", result);

Streams TestKit

You may have noticed various code patterns that emerge when testing stream pipelines. Akka Stream has a separate akka-stream-testkit module that provides tools specifically for writing stream tests. This module comes with two main components that are TestSourceTestSource and TestSinkTestSink which provide sources and sinks that materialize to probes that allow fluent API.

Using the TestKit

A sink returned by TestSink.probeTestSink.probe allows manual control over demand and assertions over elements coming downstream.

Scala

copysourceval sourceUnderTest = Source(1 to 4).filter(_ % 2 == 0).map(_ * 2)

sourceUnderTest.runWith(TestSink[Int]()).request(2).expectNext(4, 8).expectComplete()

Java

copysourcefinal Source<Integer, NotUsed> sourceUnderTest =
    Source.from(Arrays.asList(1, 2, 3, 4)).filter(elem -> elem % 2 == 0).map(elem -> elem * 2);

sourceUnderTest
    .runWith(TestSink.probe(system), system)
    .request(2)
    .expectNext(4, 8)
    .expectComplete();

A source returned by TestSource.probeTestSource.probe can be used for asserting demand or controlling when stream is completed or ended with an error.

Scala

copysourceval sinkUnderTest = Sink.cancelled

TestSource[Int]().toMat(sinkUnderTest)(Keep.left).run().expectCancellation()

Java

copysourcefinal Sink<Integer, NotUsed> sinkUnderTest = Sink.cancelled();

TestSource.<Integer>probe(system)
    .toMat(sinkUnderTest, Keep.left())
    .run(system)
    .expectCancellation();

You can also inject exceptions and test sink behavior on error conditions.

Scala

copysourceval sinkUnderTest = Sink.head[Int]

val (probe, future) = TestSource[Int]().toMat(sinkUnderTest)(Keep.both).run()
probe.sendError(new Exception("boom"))

assert(future.failed.futureValue.getMessage == "boom")

Java

copysourcefinal Sink<Integer, CompletionStage<Integer>> sinkUnderTest = Sink.head();

final Pair<TestPublisher.Probe<Integer>, CompletionStage<Integer>> probeAndCompletionStage =
    TestSource.<Integer>probe(system).toMat(sinkUnderTest, Keep.both()).run(system);
final TestPublisher.Probe<Integer> probe = probeAndCompletionStage.first();
final CompletionStage<Integer> future = probeAndCompletionStage.second();
probe.sendError(new Exception("boom"));

ExecutionException exception =
    Assert.assertThrows(
        ExecutionException.class, () -> future.toCompletableFuture().get(3, TimeUnit.SECONDS));
assertEquals("boom", exception.getCause().getMessage());

Test source and sink can be used together in combination when testing flows.

Scala

copysourceval flowUnderTest = Flow[Int].mapAsyncUnordered(2) { sleep =>
  pattern.after(10.millis * sleep, using = system.scheduler)(Future.successful(sleep))
}

val (pub, sub) = TestSource[Int]().via(flowUnderTest).toMat(TestSink[Int]())(Keep.both).run()

sub.request(n = 3)
pub.sendNext(3)
pub.sendNext(2)
pub.sendNext(1)
sub.expectNextUnordered(1, 2, 3)

pub.sendError(new Exception("Power surge in the linear subroutine C-47!"))
val ex = sub.expectError()
assert(ex.getMessage.contains("C-47"))

Java

copysourcefinal Flow<Integer, Integer, NotUsed> flowUnderTest =
    Flow.of(Integer.class)
        .mapAsyncUnordered(
            2,
            sleep ->
                akka.pattern.Patterns.after(
                    Duration.ofMillis(10),
                    system.scheduler(),
                    system.dispatcher(),
                    () -> CompletableFuture.completedFuture(sleep)));

final Pair<TestPublisher.Probe<Integer>, TestSubscriber.Probe<Integer>> pubAndSub =
    TestSource.<Integer>probe(system)
        .via(flowUnderTest)
        .toMat(TestSink.<Integer>probe(system), Keep.both())
        .run(system);
final TestPublisher.Probe<Integer> pub = pubAndSub.first();
final TestSubscriber.Probe<Integer> sub = pubAndSub.second();

sub.request(3);
pub.sendNext(3);
pub.sendNext(2);
pub.sendNext(1);
sub.expectNextUnordered(1, 2, 3);

pub.sendError(new Exception("Power surge in the linear subroutine C-47!"));
final Throwable ex = sub.expectError();
assertTrue(ex.getMessage().contains("C-47"));

Fuzzing Mode

For testing, it is possible to enable a special stream execution mode that exercises concurrent execution paths more aggressively (at the cost of reduced performance) and therefore helps exposing race conditions in tests. To enable this setting add the following line to your configuration:

akka.stream.materializer.debug.fuzzing-mode = on