The Apache Pekko Kafka Connector project is an open source initiative to implement stream-aware and reactive integration pipelines for Java and Scala. It is built on top of @extrefPekko Streams, and has been designed from the ground up to understand streaming natively and provide a DSL for reactive and stream-oriented programming, with built-in support for backpressure. Apache Pekko Streams is a Reactive Streams and JDK 9+ @extrefjava.util.concurrent.Flow-compliant implementation and therefore @extreffully interoperable with other implementations.
This Apache Pekko Connectors Kafka connector lets you connect Apache Kafka to Apache Pekko Streams. It was formerly known as Apache Pekko Streams Kafka and even Reactive Kafka.
@@project-info{ projectId=“core” }
| Kafka client | Scala Versions | Apache Pekko version | Apache Pekko Connectors Kafka Connector |
|---|---|---|---|
| 3.7.1+ | 2.12, 2.13, 3.3 | 1.1.0+ (but 1.0.0+ should be ok) | 1.1.0 |
@@@ note
As Kafka‘s client protocol negotiates the version to use with the Kafka broker, you may use a Kafka client version that is different than the Kafka broker’s version.
These clients can communicate with brokers that are version 2.1.0 or newer. Older or newer brokers may not support certain features. You will receive an UnsupportedVersionException when invoking an API that is not available on the running broker version.
Check even Confluent's Versions and Interoperability page for more details.
-- @extref:Javadoc for KafkaConsumer
@@@
@@dependency [Maven,sbt,Gradle] { group=org.apache.pekko artifact=pekko-connectors-kafka_$scala.binary.version$ version=$project.version$ symbol2=PekkoVersion value2=“$pekko.version$” group2=org.apache.pekko artifact2=pekko-stream_$scala.binary.version$ version2=PekkoVersion }
This connector depends on Apache Pekko 1.0.x and note that it is important that all pekko-* dependencies are in the same version, so it is recommended to depend on them explicitly to avoid problems with transient dependencies causing an unlucky mix of versions.
Apache Pekko Connectors Kafka APIs accept a typed @apidoc[org.apache.pekko.actor.typed.ActorSystem] or a classic @apidoc[org.apache.pekko.actor.ActorSystem] because both implement the @apidoc[org.apache.pekko.actor.ClassicActorSystemProvider] @scala[trait]@java[interface]. There are some Apache Pekko Connectors Kafka APIs that only accept classic a @apidoc[org.apache.pekko.actor.ActorRef], such as the @refrebalance listener API, but otherwise there is no difference between running Apache Pekko Connectors Kafka and any other Apache Pekko Streams implementation with a typed @apidoc[org.apache.pekko.actor.typed.ActorSystem]. For more information on Apache Pekko classic and typed interoperability read the @extrefCoexistence page of the Apache Pekko Documentation.
The table below shows Apache Pekko Connectors Kafka's direct dependencies and the second tab shows all libraries it depends on transitively.
@@dependencies { projectId=“core” }
Following Apache Pekko's conventions, there are two separate packages named org.apache.pekko.kafka.scaladsl and org.apache.pekko.kafka.javadsl with the API for Scala and Java. These packages contain Producer and Consumer classes with factory methods for the various Apache Pekko Streams Flow, Sink and Source that are producing or consuming messages to/from Kafka.
A few self-contained examples using Apache Pekko Connectors are available as Apache Pekko Connectors Samples.
To read and see how others use Apache Pekko Connectors, see the Apache Pekko Connectors documentation's Webinars, Presentations and Articles listing.
Please feel free to contribute to Apache Pekko Connectors and the Apache Pekko Connectors Kafka connector by reporting issues you identify, or by suggesting changes to the code. Please refer to our contributing instructions to learn how it can be done.
We want Apache Pekko and Apache Pekko Connectors to strive in a welcoming and open atmosphere and expect all contributors to respect our code of conduct.
@@@ index
@@@