feat: add pubsub topic validation

CHANGES.md

@@ -59,6 +59,10 @@
 
 * Added support for using vLLM in the RunInference transform (Python) ([#32528](https://github.com/apache/beam/issues/32528))
 
+## I/Os
+
+* PubsubIO can validate that the Pub/Sub topic exists before running the Read/Write pipeline (Java) ([#32465](https://github.com/apache/beam/pull/32465))
+
 ## New Features / Improvements
 
 * Dataflow worker can install packages from Google Artifact Registry Python repositories (Python) ([#32123](https://github.com/apache/beam/issues/32123)).

sdks/java/io/google-cloud-platform/src/main/java/org/apache/beam/sdk/io/gcp/pubsub/PubsubClient.java

@@ -507,6 +507,9 @@ public abstract void modifyAckDeadline(
   /** Return a list of topics for {@code project}. */
   public abstract List<TopicPath> listTopics(ProjectPath project) throws IOException;
 
+  /** Return {@literal true} if {@code topic} exists. */
+  public abstract boolean isTopicExists(TopicPath topic) throws IOException;
+
   /** Create {@code subscription} to {@code topic}. */
   public abstract void createSubscription(
       TopicPath topic, SubscriptionPath subscription, int ackDeadlineSeconds) throws IOException;

sdks/java/io/google-cloud-platform/src/main/java/org/apache/beam/sdk/io/gcp/pubsub/PubsubGrpcClient.java

@@ -54,6 +54,7 @@
 import io.grpc.Channel;
 import io.grpc.ClientInterceptors;
 import io.grpc.ManagedChannel;
+import io.grpc.StatusRuntimeException;
 import io.grpc.auth.ClientAuthInterceptor;
 import io.grpc.netty.GrpcSslContexts;
 import io.grpc.netty.NegotiationType;
@@ -372,6 +373,21 @@ public List<TopicPath> listTopics(ProjectPath project) throws IOException {
     return topics;
   }
 
+  @Override
+  public boolean isTopicExists(TopicPath topic) throws IOException {
+    GetTopicRequest request = GetTopicRequest.newBuilder().setTopic(topic.getPath()).build();
+    try {
+      publisherStub().getTopic(request);
+      return true;
+    } catch (StatusRuntimeException e) {
+      if (e.getStatus().getCode() == io.grpc.Status.Code.NOT_FOUND) {
+        return false;
+      }
+
+      throw e;
+    }
+  }
+
   @Override
   public void createSubscription(
       TopicPath topic, SubscriptionPath subscription, int ackDeadlineSeconds) throws IOException {

sdks/java/io/google-cloud-platform/src/main/java/org/apache/beam/sdk/io/gcp/pubsub/PubsubIO.java

@@ -50,6 +50,7 @@
 import org.apache.beam.sdk.io.gcp.pubsub.PubsubClient.SubscriptionPath;
 import org.apache.beam.sdk.io.gcp.pubsub.PubsubClient.TopicPath;
 import org.apache.beam.sdk.metrics.Lineage;
+import org.apache.beam.sdk.options.PipelineOptions;
 import org.apache.beam.sdk.options.ValueProvider;
 import org.apache.beam.sdk.options.ValueProvider.NestedValueProvider;
 import org.apache.beam.sdk.options.ValueProvider.StaticValueProvider;
@@ -860,6 +861,8 @@ public abstract static class Read<T> extends PTransform<PBegin, PCollection<T>>
 
     abstract ErrorHandler<BadRecord, ?> getBadRecordErrorHandler();
 
+    abstract boolean getValidate();
+
     abstract Builder<T> toBuilder();
 
     static <T> Builder<T> newBuilder(SerializableFunction<PubsubMessage, T> parseFn) {
@@ -871,6 +874,7 @@ static <T> Builder<T> newBuilder(SerializableFunction<PubsubMessage, T> parseFn)
       builder.setNeedsOrderingKey(false);
       builder.setBadRecordRouter(BadRecordRouter.THROWING_ROUTER);
       builder.setBadRecordErrorHandler(new DefaultErrorHandler<>());
+      builder.setValidate(false);
       return builder;
     }
 
@@ -918,6 +922,8 @@ abstract static class Builder<T> {
       abstract Builder<T> setBadRecordErrorHandler(
           ErrorHandler<BadRecord, ?> badRecordErrorHandler);
 
+      abstract Builder<T> setValidate(boolean validation);
+
       abstract Read<T> build();
     }
 
@@ -1097,6 +1103,11 @@ public Read<T> withErrorHandler(ErrorHandler<BadRecord, ?> badRecordErrorHandler
           .build();
     }
 
+    /** Enable validation of the PubSub Read. */
+    public Read<T> withValidation() {
+      return toBuilder().setValidate(true).build();
+    }
+
     @VisibleForTesting
     /**
      * Set's the internal Clock.
@@ -1262,6 +1273,35 @@ public T apply(PubsubMessage input) {
       return read.setCoder(getCoder());
     }
 
+    @Override
+    public void validate(PipelineOptions options) {
+      if (!getValidate()) {
+        return;
+      }
+
+      PubsubOptions psOptions = options.as(PubsubOptions.class);
+
+      // Validate the existence of the topic.
+      if (getTopicProvider() != null) {
+        PubsubTopic topic = getTopicProvider().get();
+        boolean topicExists = true;
+        try (PubsubClient pubsubClient =
+            getPubsubClientFactory()
+                .newClient(getTimestampAttribute(), getIdAttribute(), psOptions)) {
+          topicExists =
+              pubsubClient.isTopicExists(
+                  PubsubClient.topicPathFromName(topic.project, topic.topic));
+        } catch (Exception e) {
+          throw new RuntimeException(e);
+        }
+
+        if (!topicExists) {
+          throw new IllegalArgumentException(
+              String.format("Pubsub topic '%s' does not exist.", topic));
+        }
+      }
+    }
+
     @Override
     public void populateDisplayData(DisplayData.Builder builder) {
       super.populateDisplayData(builder);
@@ -1341,6 +1381,8 @@ public abstract static class Write<T> extends PTransform<PCollection<T>, PDone>
 
     abstract ErrorHandler<BadRecord, ?> getBadRecordErrorHandler();
 
+    abstract boolean getValidate();
+
     abstract Builder<T> toBuilder();
 
     static <T> Builder<T> newBuilder(
@@ -1350,6 +1392,7 @@ static <T> Builder<T> newBuilder(
       builder.setFormatFn(formatFn);
       builder.setBadRecordRouter(BadRecordRouter.THROWING_ROUTER);
       builder.setBadRecordErrorHandler(new DefaultErrorHandler<>());
+      builder.setValidate(false);
       return builder;
     }
 
@@ -1386,6 +1429,8 @@ abstract Builder<T> setFormatFn(
       abstract Builder<T> setBadRecordErrorHandler(
           ErrorHandler<BadRecord, ?> badRecordErrorHandler);
 
+      abstract Builder<T> setValidate(boolean validation);
+
       abstract Write<T> build();
     }
 
@@ -1396,18 +1441,28 @@ abstract Builder<T> setBadRecordErrorHandler(
      * {@code topic} string.
      */
     public Write<T> to(String topic) {
+      ValueProvider<String> topicProvider = StaticValueProvider.of(topic);
+      validateTopic(topicProvider);
       return to(StaticValueProvider.of(topic));
     }
 
     /** Like {@code topic()} but with a {@link ValueProvider}. */
     public Write<T> to(ValueProvider<String> topic) {
+      validateTopic(topic);
       return toBuilder()
           .setTopicProvider(NestedValueProvider.of(topic, PubsubTopic::fromPath))
           .setTopicFunction(null)
           .setDynamicDestinations(false)
           .build();
     }
 
+    /** Handles validation of {@code topic}. */
+    private static void validateTopic(ValueProvider<String> topic) {
+      if (topic.isAccessible()) {
+        PubsubTopic.fromPath(topic.get());
+      }
+    }
+
     /**
      * Provides a function to dynamically specify the target topic per message. Not compatible with
      * any of the other to methods. If {@link #to} is called again specifying a topic, then this
@@ -1497,6 +1552,11 @@ public Write<T> withErrorHandler(ErrorHandler<BadRecord, ?> badRecordErrorHandle
           .build();
     }
 
+    /** Enable validation of the PubSub Write. */
+    public Write<T> withValidation() {
+      return toBuilder().setValidate(true).build();
+    }
+
     @Override
     public PDone expand(PCollection<T> input) {
       if (getTopicProvider() == null && !getDynamicDestinations()) {
@@ -1566,6 +1626,35 @@ public PDone expand(PCollection<T> input) {
       throw new RuntimeException(); // cases are exhaustive.
     }
 
+    @Override
+    public void validate(PipelineOptions options) {
+      if (!getValidate()) {
+        return;
+      }
+
+      PubsubOptions psOptions = options.as(PubsubOptions.class);
+
+      // Validate the existence of the topic.
+      if (getTopicProvider() != null) {
+        PubsubTopic topic = getTopicProvider().get();
+        boolean topicExists = true;
+        try (PubsubClient pubsubClient =
+            getPubsubClientFactory()
+                .newClient(getTimestampAttribute(), getIdAttribute(), psOptions)) {
+          topicExists =
+              pubsubClient.isTopicExists(
+                  PubsubClient.topicPathFromName(topic.project, topic.topic));
+        } catch (Exception e) {
+          throw new RuntimeException(e);
+        }
+
+        if (!topicExists) {
+          throw new IllegalArgumentException(
+              String.format("Pubsub topic '%s' does not exist.", topic));
+        }
+      }
+    }
+
     @Override
     public void populateDisplayData(DisplayData.Builder builder) {
       super.populateDisplayData(builder);

sdks/java/io/google-cloud-platform/src/main/java/org/apache/beam/sdk/io/gcp/pubsub/PubsubJsonClient.java

@@ -19,6 +19,7 @@
 
 import static org.apache.beam.vendor.guava.v32_1_2_jre.com.google.common.base.Preconditions.checkState;
 
+import com.google.api.client.googleapis.json.GoogleJsonResponseException;
 import com.google.api.client.http.HttpRequestInitializer;
 import com.google.api.services.pubsub.Pubsub;
 import com.google.api.services.pubsub.Pubsub.Projects.Subscriptions;
@@ -310,6 +311,19 @@ public List<TopicPath> listTopics(ProjectPath project) throws IOException {
     return topics;
   }
 
+  @Override
+  public boolean isTopicExists(TopicPath topic) throws IOException {
+    try {
+      pubsub.projects().topics().get(topic.getPath()).execute();
+      return true;
+    } catch (GoogleJsonResponseException e) {
+      if (e.getStatusCode() == 404) {
+        return false;
+      }
+      throw e;
+    }
+  }
+
   @Override
   public void createSubscription(
       TopicPath topic, SubscriptionPath subscription, int ackDeadlineSeconds) throws IOException {

sdks/java/io/google-cloud-platform/src/main/java/org/apache/beam/sdk/io/gcp/pubsub/PubsubTestClient.java

@@ -605,6 +605,12 @@ public List<TopicPath> listTopics(ProjectPath project) throws IOException {
     throw new UnsupportedOperationException();
   }
 
+  @Override
+  public boolean isTopicExists(TopicPath topic) throws IOException {
+    // Always return true for testing purposes.
+    return true;
+  }
+
   @Override
   public void createSubscription(
       TopicPath topic, SubscriptionPath subscription, int ackDeadlineSeconds) throws IOException {

sdks/java/io/google-cloud-platform/src/test/java/org/apache/beam/sdk/io/gcp/pubsub/PubsubGrpcClientTest.java

@@ -40,6 +40,7 @@
 import com.google.pubsub.v1.Topic;
 import io.grpc.ManagedChannel;
 import io.grpc.Server;
+import io.grpc.Status;
 import io.grpc.StatusRuntimeException;
 import io.grpc.inprocess.InProcessChannelBuilder;
 import io.grpc.inprocess.InProcessServerBuilder;
@@ -432,4 +433,43 @@ public void getSchema(GetSchemaRequest request, StreamObserver<Schema> responseO
       server.shutdownNow();
     }
   }
+
+  @Test
+  public void isTopicExists() throws IOException {
+    initializeClient(null, null);
+    TopicPath topicDoesNotExist =
+        PubsubClient.topicPathFromPath("projects/testProject/topics/dontexist");
+    TopicPath topicExists = PubsubClient.topicPathFromPath("projects/testProject/topics/exist");
+
+    PublisherImplBase publisherImplBase =
+        new PublisherImplBase() {
+          @Override
+          public void getTopic(GetTopicRequest request, StreamObserver<Topic> responseObserver) {
+            String topicPath = request.getTopic();
+            if (topicPath.equals(topicDoesNotExist.getPath())) {
+              responseObserver.onError(
+                  new StatusRuntimeException(Status.fromCode(Status.Code.NOT_FOUND)));
+            }
+            if (topicPath.equals(topicExists.getPath())) {
+              responseObserver.onNext(
+                  Topic.newBuilder()
+                      .setName(topicPath)
+                      .setSchemaSettings(
+                          SchemaSettings.newBuilder().setSchema(SCHEMA.getPath()).build())
+                      .build());
+              responseObserver.onCompleted();
+            }
+          }
+        };
+    Server server =
+        InProcessServerBuilder.forName(channelName).addService(publisherImplBase).build().start();
+    try {
+      assertEquals(false, client.isTopicExists(topicDoesNotExist));
+
+      assertEquals(true, client.isTopicExists(topicExists));
+
+    } finally {
+      server.shutdownNow();
+    }
+  }
 }