Cloud-Events Kafka Binding With Spring

Utilize Kafka Protocol Binding for CloudEvents along with Spring-Kafka to easily produce and consume events in a common format.

Will Kennedy

Nov. 13, 20 · Tutorial

Likes (7)

Comment

Save

12.2K Views

This article aims to provide details on how to use the spring-kafka library with the CloudEvents spec in order to quickly get you producing and consuming events with a common structure by way of Kafka. The spring-kafka project provides utilities and templates to interact with Kafka with minimal effort, while the CloudEvents specification describes event data in a common way and provides both a Java-SDK as well as a library for Kafka protocol binding.

The CloudEvents Kafka protocol binding provides two content modes for transferring data.

Structured: event metadata attributes and event data are placed into the Kafka message value section using an event format.
Binary: the value of the event data MUST be placed into the Kafka message's value section as-is, with the content-type header value declaring its media type; all other event attributes MUST be mapped to the Kafka message's header section.

For more details see: https://github.com/cloudevents/spec/blob/master/kafka-protocol-binding.md#13-content-modes

Dependencies

In this tutorial we'll be using spring-kafka 2.5.5.RELEASE and cloudevents-kafka 2.0.0-milestone3. As of this writing, version 2 of the cloud events library hasn't been released, so the latest milestone is used.

    XML
   
 

     
    
xxxxxxxxxx

            
          
        <dependency>
            <groupId>org.springframework.kafka</groupId>
            <artifactId>spring-kafka</artifactId>
            <version>2.5.5.RELEASE</version>
        </dependency>
        <dependency>
            <groupId>io.cloudevents</groupId>
            <artifactId>cloudevents-kafka</artifactId>
            <version>2.0.0-milestone3</version>
        </dependency>
        <dependency>
            <groupId>io.cloudevents</groupId>
            <artifactId>cloudevents-json-jackson</artifactId>
            <version>2.0.0-milestone3</version>
        </dependency>

Binary Producer and Consumer

In order to publish and consumer messages from Kafka, we'll need to create a Producer and Consumer that can support serializing and deserializing Cloud Events.

Producer

The first thing you need is a producer. We can do this by creating a class to help build our Cloud Event supporting producer:

    Java
   
 

     
    
xxxxxxxxxx

            
          
import io.cloudevents.core.format.EventFormat;
import io.cloudevents.core.message.Encoding;
import io.cloudevents.core.provider.EventFormatProvider;
import io.cloudevents.jackson.JsonFormat;
import io.cloudevents.kafka.CloudEventSerializer;
import org.apache.kafka.common.serialization.Serializer;
import org.springframework.kafka.core.DefaultKafkaProducerFactory;
import java.util.HashMap;
import java.util.Map;
import static io.cloudevents.kafka.CloudEventSerializer.ENCODING_CONFIG;
import static io.cloudevents.kafka.CloudEventSerializer.EVENT_FORMAT_CONFIG;
public class CloudEventKafkaProducerFactory {
    public static <K, V> DefaultKafkaProducerFactory<K, V> cloudEventKafkaProducerFactory(
        Map<String, Object> configs,                                                                              Serializer<K> keySerializer,                                                                                         Encoding encoding) {
      
        return cloudEventKafkaProducerFactory(configs, keySerializer,                   encoding, JsonFormat.CONTENT_TYPE);
    }
    @SuppressWarnings("unchecked")
    public static <K, V> DefaultKafkaProducerFactory<K, V> cloudEventKafkaProducerFactory(
      Map<String, Object> configs, 
      Serializer<K> keySerializer, 
      Encoding encoding, 
      String eventFormat) {
      
        //If present, the Kafka message header property content-type MUST be set to the media type of an event format.
        if(Encoding.STRUCTURED.equals(encoding)) {
            EventFormat resolveFormat = EventFormatProvider
              .getInstance()
              .resolveFormat(eventFormat);
            if(resolveFormat == null) {
                eventFormat = JsonFormat.CONTENT_TYPE;
            }
        }
        Map<String, Object> ceSerializerConfigs = new HashMap<>();
        ceSerializerConfigs.put(ENCODING_CONFIG, encoding);
        ceSerializerConfigs.put(EVENT_FORMAT_CONFIG, eventFormat);
        CloudEventSerializer cloudEventSerializer = 
                    new CloudEventSerializer();
        //isKey always false
        cloudEventSerializer.configure(ceSerializerConfigs, false); 
      
        return new DefaultKafkaProducerFactory<>(configs, keySerializer,                (Serializer<V>) cloudEventSerializer);
    }
}

This class will allow us to create a DefaultKafkaProducerFactory that supports either binary or structured encoding. You can also specify the event format depending on your needs.

You can instantiate a ProducerFactory to serialize your data into a structured cloud event using the CloudEventKafkaProducerFactory. You simply need to provide it with your producer configs, the key serializer, and the encoding type, in this case, Encoding.BINARY.

    Java
   
          x
         
DefaultKafkaProducerFactory<String, CloudEvent> binaryProducerFactory = CloudEventKafkaProducerFactory.cloudEventKafkaProducerFactory(producerConfigs, new StringSerializer(), Encoding.BINARY);

Consumer

We'll also need to create a class to help us construct a consumer that supports Cloud Events:

    Java
   
 

     
    
xxxxxxxxxx

            
          
import io.cloudevents.core.format.EventFormat;
import io.cloudevents.core.message.Encoding;
import io.cloudevents.core.provider.EventFormatProvider;
import io.cloudevents.jackson.JsonFormat;
import io.cloudevents.kafka.CloudEventDeserializer;
import org.apache.kafka.common.serialization.Deserializer;
import org.springframework.kafka.core.DefaultKafkaConsumerFactory;
import java.util.HashMap;
import java.util.Map;
import static io.cloudevents.kafka.CloudEventSerializer.ENCODING_CONFIG;
import static io.cloudevents.kafka.CloudEventSerializer.EVENT_FORMAT_CONFIG;
public class CloudEventKafkaConsumerFactory {
    public static <K, V> DefaultKafkaConsumerFactory<K, V>  consumerFactory(
      Map<String, Object> configs,
      Deserializer<K> keySerializer, 
      Encoding encoding) {
        return consumerFactory(configs, 
                               keySerializer, 
                               encoding, 
                               JsonFormat.CONTENT_TYPE);
    }
    @SuppressWarnings("unchecked")
    public static <K, V> DefaultKafkaConsumerFactory<K, V> consumerFactory(
      Map<String, Object> configs, 
      Deserializer<K> keySerializer, 
      Encoding encoding, 
      String eventFormat) {
        //If present, the Kafka message header property content-type MUST be set to the media type of an event format.
        if(Encoding.STRUCTURED.equals(encoding)) {
            EventFormat resolveFormat = EventFormatProvider
              .getInstance()
              .resolveFormat(eventFormat);
            if(resolveFormat == null) {
                eventFormat = JsonFormat.CONTENT_TYPE;
            }
        }
        Map<String, Object> ceDeserializerConfigs = new HashMap<>();
        ceDeserializerConfigs.put(ENCODING_CONFIG, encoding);
        ceDeserializerConfigs.put(EVENT_FORMAT_CONFIG, eventFormat);
        CloudEventDeserializer cloudEventDeserializer = 
          new CloudEventDeserializer();
        //isKey always false
        cloudEventDeserializer.configure(ceDeserializerConfigs, false); 
        return new DefaultKafkaConsumerFactory<>(configs,
                keySerializer,
                (Deserializer<V>) cloudEventDeserializer);
    }
}

This class will allow us to create a DefaultKafkaConsumerFactory that supports either binary or structured encoding which will properly deserialize your Kafka message to a CloudEvent object.

The process of creating a consumer is similar to that of the producer.

    Java
   
xxxxxxxxxx

ConsumerFactory<String, CloudEvent> binaryConsumerFactory =     CloudEventKafkaConsumerFactory.consumerFactory(
  consumerConfigs, 
  new StringDeserializer(), 
  Encoding.BINARY);

Structured Producer and Consumer

Producer

You can create a ProducerFactor to serialize your data into a structured cloud event using the CloudEventKafkaProducerFactory. You simply need to provide it with your producer configs, the key serializer, and the encoding type, in this case, Encoding.STRUCTURED.

    Java
   
xxxxxxxxxx

DefaultKafkaProducerFactory<String, CloudEvent> structuredProducerFactory = CloudEventKafkaProducerFactory.cloudEventKafkaProducerFactory(
  producerConfigs, 
  new StringSerializer(), 
  Encoding.STRUCTURED);

Consumer

The process of creating a consumer is similar to that of the producer.

    Java
   
xxxxxxxxxx

ConsumerFactory<String, CloudEvent> structuredConsumerFactory = CloudEventKafkaConsumerFactory.consumerFactory(
  consumerConfigs, 
  new StringDeserializer(), 
  Encoding.STRUCTURED);

Example Usage

    Java
   
xxxxxxxxxx

DefaultKafkaProducerFactory<String, CloudEvent> binaryProducerFactory = CloudEventKafkaProducerFactory.cloudEventKafkaProducerFactory(
  producerProps, new StringSerializer(), Encoding.BINARY); 
KafkaTemplate<String, CloudEvent> binaryKafkaTemplate = 
  new KafkaTemplate<>(binaryProducerFactory); 
ConsumerFactory<String, CloudEvent> binaryConsumerFactory = CloudEventKafkaConsumerFactory.consumerFactory(
  consumerProps, new StringDeserializer(), Encoding.BINARY);
Consumer<String, CloudEvent> binaryConsumer = binaryConsumerFactory.createConsumer();

With this KafkaTemplate and Consumer, you'll be able to publish and consume CloudEvents with binary encoding. Here is a unit test illustrating this usage:

    Java
   
 

     
    
xxxxxxxxxx

            
          
    @Test
    public void testCloudEventsBinaryMessage() throws Exception {
        binaryKafkaTemplate.send(TOPIC, getCloudEvent(person));
        ConsumerRecord<String, CloudEvent> consumerRecord = KafkaTestUtils
          .getSingleRecord(binaryConsumer, TOPIC);
        CloudEvent consumedEvent = consumerRecord.value();
        Person payload = objectMapper
          .readValue(consumedEvent.getData(),Person.class);
        assertThat(consumedEvent.getId()).isEqualTo(cloudEventID);
        assertThat(consumedEvent.getType()).isEqualTo(cloudEventType);
        assertThat(consumedEvent.getSource()).isEqualTo(cloudEventSource);
        assertThat(payload).isEqualTo(person);
        KafkaUtils.clearConsumerGroupId();
}