Kafka Streams Custom Seder

Just an example of how to use a custom Seder in Kafka Streams Applications. It works with Kafka 2.1.0.

Install

1. Clone the repository.
2. Open the project with your favorite IDE (In my case IntelliJ).
3. The IDE will read the dependencies from pom.xml and will prepare all for you.

Run

The whole code is in src/main/java/myapps. There you can find several files:

• JSONEnricher.java is the main Java class. It contains all the logic of the Kafka Stream App coded.
• Person.java it is the object that we want to represent in our Kafka Topic.
• PersonSerializer.java it is the class which will serialize our Person object.
• PersonDeserializar.java it is the class which will deserialize our Person object.
• PersonSeder.java it is the class which wraps the serializer and the deserializer to be used on JSONEnricher.java.

In order to run this code, you have to:

1. Start Zookeeper and Kafka servers.
2. Create two topics:
   1. streams-plaintext-input as input pipe.
   2. streams-jsonenricher-output as output pipe.
3. Run the Kafka Stream App coded:
   1. Right click on JSONEnricher.java and Run.
4. Start a producer (e.g. kafka-console-producer) and start writing JSON messages to the streams-plaintext-input topic, like:
   • { "name" : "David Corral", "age" : 23 }
   • { "name" : "David Corral", "age" : 23, "location" : "Seville" }
   • { "name" : "David Corral", "age" : 23, "location" : "Seville", "salary" : 23000 }

At this point you can see the transformations performs by the Kafka Stream App in the Run Terminal of your IDE. But also you can start a consumer (e.g. kafka-console-consumer) and retrieve messages from the streams-jsonenricher-output topic.