Skip to content

Latest commit

 

History

History
699 lines (564 loc) · 23.9 KB

File metadata and controls

699 lines (564 loc) · 23.9 KB

Reactive Messaging

Overview

Asynchronous messaging is a commonly used form of communication in the world of microservices. While its possible to start building your reactive streams directly by combining operators and connecting them to reactive APIs, with Helidon SE Reactive Messaging, you can now use prepared tools for repetitive use case scenarios .

<dependency>
    <groupId>io.helidon.messaging</groupId>
    <artifactId>helidon-messaging</artifactId>
</dependency>

Usage

For example connecting your streams to external services usually requires a lot of boiler-plate code for configuration handling, backpressure propagation, acknowledgement and more.

For such tasks there is a system of connectors, emitters and means to orchestrate them in Helidon, called Reactive Messaging. It’s basically an API for connecting and configuring Connectors and Emitters with your reactive streams through so called Channels.

You may wonder how Reactive Messaging relates to MicroProfile Reactive Messaging. As the making of connectors or even configuring them can be repetitive task leading to the same results, Helidon SE Reactive Messaging supports very same configuration format for connectors as its MicroProfile counterpart does. Also, MP Connectors are reusable in Helidon SE Messaging with some limitation(there is no CDI in Helidon SE). All Messaging connectors in Helidon are made to be universally usable by Helidon MP and SE.

Channel

Channel is a named pair of Publisher and Subscriber. Channels can be connected together by processors. Registering of Publisher or Subscriber for a channel can be done by Messaging API, or configured implicitly for using registered connectors for generating such Publisher or Subscriber.

Example of simple channel:
Channel<String> channel1 = Channel.create("channel1");

Messaging.builder()
        .publisher(channel1, Multi.just("message 1", "message 2")
                                  .map(Message::of))
        .listener(channel1, s -> System.out.println("Intecepted message " + s))
        .build()
        .start();

Processor

Processor is a typical reactive processor acting as a Subscriber to upstream and as a Publisher to downstream. In terms of reactive messaging it is able to connect two channels to one reactive stream.

Example of processor usage:
Channel<String> firstChannel = Channel.create("first-channel");
Channel<String> secondChannel = Channel.create("second-channel");

Messaging.builder()
        .publisher(secondChannel, ReactiveStreams.of("test1", "test2", "test3")
                .map(Message::of))
        .processor(secondChannel, firstChannel, ReactiveStreams.<Message<String>>builder()
                .map(Message::getPayload)
                .map(String::toUpperCase)
                .map(Message::of)
        )
        .subscriber(firstChannel, ReactiveStreams.<Message<String>>builder()
                .peek(Message::ack)
                .map(Message::getPayload)
                .forEach(s -> System.out.println("Consuming message " + s)))
        .build()
        .start();

>Consuming message TEST1
>Consuming message TEST2
>Consuming message TEST3

Message

Reactive Messaging in Helidon SE uses the same concept of message wrapping as MicroProfile messaging. The only notable difference is that SE Messaging does almost no implicit or automatic acknowledgement due to no magic philosophy of Helidon SE.

Only exception to this are variants of methods Messaging.Builder#listener and Messaging.Builder#processor with consumer or function params, conveniently unwrapping payload for you. After such implicit unwrapping is not possible to do a manual acknowledgement, therefore implicit ack before callback is executed is necessary.

Connectors

Connector concept is a way for connecting channels to external sources. To make creation and usage of connectors as easy and versatile as possible, Helidon SE Messaging uses same API for connectors like MicroProfile Reactive Messaging does. This allows connectors to be usable in both flavors of Helidon with one limitation which is that connector has to be able to work without CDI.

Example of such a versatile connectors in Helidon:

Messaging Connector

Connector for Reactive Messaging is a factory producing Publishers and Subscribers for Channels in Reactive Messaging. Messaging connector is just an implementation of IncomingConnectorFactory, OutgoingConnectorFactory or both.

Example connector example-connector:
@Connector("example-connector")
public class ExampleConnector implements IncomingConnectorFactory, OutgoingConnectorFactory {

   @Override
   public PublisherBuilder<? extends Message<?>> getPublisherBuilder(Config config) {
       return ReactiveStreams.of("foo", "bar")
               .map(Message::of);
   }

   @Override
   public SubscriberBuilder<? extends Message<?>, Void> getSubscriberBuilder(Config config) {
       return ReactiveStreams.<Message<?>>builder()
               .map(Message::getPayload)
               .forEach(o -> System.out.println("Connector says: " + o));
   }
}
Example of channel to connector mapping config:
mp.messaging.outgoing.to-connector-channel.connector: example-connector
mp.messaging.incoming.from-connector-channel.connector: example-connector
Example producing to connector:
Config config = Config.create();

Messaging.builder()
         .config(config)
         .connector(new ExampleConnector())
         .publisher(Channel.create("to-connector-channel"),
                ReactiveStreams.of("fee", "fie")
                    .map(Message::of)
         )
         .build()
         .start();

> Connector says: fee
> Connector says: fie
Example consuming from connector:
Messaging.builder()
        .connector(new ExampleConnector())
        .subscriber(Channel.create("from-connector-channel"),
                ReactiveStreams.<Message<String>>builder()
                    .peek(Message::ack)
                    .map(Message::getPayload)
                    .forEach(s -> System.out.println("Consuming: " + s))
        )
        .build()
        .start();

> Consuming: foo
> Consuming: bar
Configuration for Messaging connector

Messaging connector in Helidon SE can be configured explicitly by API or implicitly by config following notation of MicroProfile Reactive Messaging.

Configuration is being supplied to connector by Messaging implementation, two mandatory attributes are always present:

  • channel-name name of the channel which has this connector configured as Publisher or Subscriber, Channel.create('name-of-channel') in case of explicit configuration or mp.messaging.incoming.name-of-channel.connector: connector-name in case of implicit config

  • connector name of the connector @Connector("connector-name")

Example connector accessing configuration:
@Connector("example-connector")
public class ExampleConnector implements IncomingConnectorFactory {

    @Override
    public PublisherBuilder<? extends Message<?>> getPublisherBuilder(final Config config) {

        String firstPropValue = config.getValue("first-test-prop", String.class);(1)
        String secondPropValue = config.getValue("second-test-prop", String.class);

        return ReactiveStreams.of(firstPropValue, secondPropValue)
                .map(Message::of);
    }
}
  1. Config context is merged from channel and connector contexts

Explicit Config for Messaging connector

An explicit config for channel’s publisher is possible with Channel.Builder#publisherConfig(Config config) and for subscriber with Channel.Builder#subscriberConfig(Config config). Supplied Helidon Config is merged with mandatory attributes and any implicit config found. Resulting config is served to Connector.

Example consuming from Kafka connector with explicit config:
String kafkaServer = config.get("app.kafka.bootstrap.servers").asString().get();
String topic = config.get("app.kafka.topic").asString().get();

Channel<String> fromKafka = Channel.<String>builder()(1)(2)
        .name("from-kafka")
        .publisherConfig(KafkaConnector.configBuilder()
                .bootstrapServers(kafkaServer)
                .groupId("example-group-" + session.getId())
                .topic(topic)
                .autoOffsetReset(KafkaConfigBuilder.AutoOffsetReset.LATEST)
                .enableAutoCommit(true)
                .keyDeserializer(StringDeserializer.class)
                .valueDeserializer(StringDeserializer.class)
                .build()
        )
        .build();

KafkaConnector kafkaConnector = KafkaConnector.create();(3)

Messaging messaging = Messaging.builder()
        .connector(kafkaConnector)
        .listener(fromKafka, payload -> {
            System.out.println("Kafka says: " + payload);
        })
        .build()
        .start();
  1. Prepare channel for connecting kafka connector with specific publisher configuration → listener,

  2. Channel → connector mapping is automatic when using KafkaConnector.configBuilder()

  3. Prepare Kafka connector, can be used by any channel

Implicit Config for Messaging connector

Implicit config without any hard-coding is possible with Helidon Config following notation of MicroProfile Reactive Messaging.

Example of channel to connector mapping config with custom properties:
mp.messaging.incoming.from-connector-channel.connector: example-connector(1)
mp.messaging.incoming.from-connector-channel.first-test-prop: foo(2)
mp.messaging.connector.example-connector.second-test-prop: bar(3)
  1. Channel → Connector mapping

  2. Channel configuration properties

  3. Connector configuration properties

Example consuming from connector:
Config config = Config.create();

Messaging.builder()
        .config(config)
        .connector(new ExampleConnector())
        .listener(Channel.create("from-connector-channel"),
                    s -> System.out.println("Consuming: " + s))
        .build()
        .start();

> Consuming: foo
> Consuming: bar

Reusability in MP Messaging

As the API is the same for MicroProfile Reactive Messaging connectors, all that is needed to make connector work in both ways is annotating it with @ApplicationScoped. Such connector is treated as a bean in Helidon MP.

For specific information about creating messaging connectors for Helidon MP visit MicroProfile Reactive Messaging.

Kafka Connector

Maven dependency
<dependency>
    <groupId>io.helidon.messaging.kafka</groupId>
    <artifactId>helidon-messaging-kafka</artifactId>
</dependency>
Reactive Kafka Connector

Connecting streams to Kafka with Reactive Messaging couldn’t be easier.

Explicit config with config builder for Kafka Connector
Example of consuming from Kafka:
String kafkaServer = config.get("app.kafka.bootstrap.servers").asString().get();
String topic = config.get("app.kafka.topic").asString().get();

Channel<String> fromKafka = Channel.<String>builder()(1)(2)
        .name("from-kafka")
        .publisherConfig(KafkaConnector.configBuilder()
                .bootstrapServers(kafkaServer)
                .groupId("example-group-" + session.getId())
                .topic(topic)
                .autoOffsetReset(KafkaConfigBuilder.AutoOffsetReset.LATEST)
                .enableAutoCommit(true)
                .keyDeserializer(StringDeserializer.class)
                .valueDeserializer(StringDeserializer.class)
                .build()
        )
        .build();

KafkaConnector kafkaConnector = KafkaConnector.create();(3)

Messaging messaging = Messaging.builder()
        .connector(kafkaConnector)
        .listener(fromKafka, payload -> {
            System.out.println("Kafka says: " + payload);
        })
        .build()
        .start();
  1. Prepare a channel for connecting kafka connector with specific publisher configuration → listener

  2. Channel → connector mapping is automatic when using KafkaConnector.configBuilder()

  3. Prepare Kafka connector, can be used by any channel

Example of producing to Kafka:
String kafkaServer = config.get("app.kafka.bootstrap.servers").asString().get();
String topic = config.get("app.kafka.topic").asString().get();

Channel<String> toKafka = Channel.<String>builder()(1)(2)
        .subscriberConfig(KafkaConnector.configBuilder()
                .bootstrapServers(kafkaServer)
                .topic(topic)
                .keySerializer(StringSerializer.class)
                .valueSerializer(StringSerializer.class)
                .build()
        ).build();

KafkaConnector kafkaConnector = KafkaConnector.create();(3)

messaging = Messaging.builder()
        .publisher(toKafka, Multi.just("test1", "test2").map(Message::of))
        .connector(kafkaConnector)
        .build()
        .start();
  1. Prepare a channel for connecting kafka connector with specific publisher configuration → listener

  2. Channel → connector mapping is automatic when using KafkaConnector.configBuilder()

  3. Prepare Kafka connector, can be used by any channel

Implicit Helidon Config for Kafka Connector
Example of connector config:
mp.messaging:

  incoming.from-kafka:
    connector: helidon-kafka
    topic: messaging-test-topic-1
    auto.offset.reset: latest # (1)
    enable.auto.commit: true
    group.id: example-group-id

  outgoing.to-kafka:
    connector: helidon-kafka
    topic: messaging-test-topic-1

  connector:
    helidon-kafka:
      bootstrap.servers: localhost:9092 # (2)
      key.serializer: org.apache.kafka.common.serialization.StringSerializer
      value.serializer: org.apache.kafka.common.serialization.StringSerializer
      key.deserializer: org.apache.kafka.common.serialization.StringDeserializer
      value.deserializer: org.apache.kafka.common.serialization.StringDeserializer
  1. Kafka client consumer’s property auto.offset.reset configuration for from-kafka channel only

  2. Kafka client’s property bootstrap.servers configuration for all channels using the connector

Example of consuming from Kafka:
Config config = Config.create();

Channel<String> fromKafka = Channel.create("from-kafka");

KafkaConnector kafkaConnector = KafkaConnector.create();(1)

Messaging messaging = Messaging.builder()
        .config(config)
        .connector(kafkaConnector)
        .listener(fromKafka, payload -> {
            System.out.println("Kafka says: " + payload);
        })
        .build()
        .start();
  1. Prepare Kafka connector, can be used by any channel

Example of producing to Kafka:
Config config = Config.create();

Channel<String> toKafka = Channel.create("to-kafka");

KafkaConnector kafkaConnector = KafkaConnector.create();(1)

messaging = Messaging.builder()
        .config(config)
        .publisher(toKafka, Multi.just("test1", "test2").map(Message::of))
        .connector(kafkaConnector)
        .build()
        .start();
  1. Prepare Kafka connector, can be used by any channel

Don’t forget to check out the examples with pre-configured Kafka docker image, for easy testing:

JMS Connector

Maven dependency
<dependency>
    <groupId>io.helidon.messaging.jms</groupId>
    <artifactId>helidon-messaging-jms</artifactId>
</dependency>
Reactive JMS Connector

Connecting streams to JMS with Reactive Messaging couldn’t be easier.

Explicit config with config builder for JMS Connector
Example of consuming from JMS:
Channel<String> fromJms = Channel.<String>builder()(1)(2)
        .name("from-jms")
        .publisherConfig(JmsConnector.configBuilder()
                .jndiInitialFactory(ActiveMQInitialContextFactory.class)
                .jndiProviderUrl("tcp://127.0.0.1:61616")
                .type(JmsConfigBuilder.Type.QUEUE)
                .destination("se-example-queue-1")
                .build()
        )
        .build();

JmsConnector jmsConnector = JmsConnector.create();(3)

Messaging messaging = Messaging.builder()
        .connector(jmsConnector)
        .listener(fromJms, payload -> {
            System.out.println("Jms says: " + payload);
        })
        .build()
        .start();
  1. Prepare a channel for connecting jms connector with specific publisher configuration → listener

  2. Channel → connector mapping is automatic when using JmsConnector.configBuilder()

  3. Prepare JMS connector, can be used by any channel

Example of producing to JMS:
Channel<String> toJms = Channel.<String>builder()(1)(2)
        .subscriberConfig(JmsConnector.configBuilder()
                .jndiInitialFactory(ActiveMQInitialContextFactory.class)
                .jndiProviderUrl("tcp://127.0.0.1:61616")
                .type(JmsConfigBuilder.Type.QUEUE)
                .destination("se-example-queue-1")
                .build()
        ).build();

JmsConnector jmsConnector = JmsConnector.create();(3)

messaging = Messaging.builder()
        .publisher(toJms, Multi.just("test1", "test2").map(Message::of))
        .connector(jmsConnector)
        .build()
        .start();
  1. Prepare a channel for connecting jms connector with specific publisher configuration → listener

  2. Channel → connector mapping is automatic when using JmsConnector.configBuilder()

  3. Prepare JMS connector, can be used by any channel

Implicit Helidon Config for JMS Connector
Example of connector config:
mp.messaging:

  incoming.from-jms:
    connector: helidon-jms
    destination: se-example-queue-1
    session-group-id: session-group-1
    type: queue

  outgoing.to-jms:
    connector: helidon-jms
      destination: se-example-queue-1
      type: queue

  connector:
    helidon-jms:
      jndi:
        jms-factory: ConnectionFactory
        env-properties:
          java.naming.factory.initial: org.apache.activemq.jndi.ActiveMQInitialContextFactory
          java.naming.provider.url: tcp://127.0.0.1:61616
Example of consuming from JMS:
Config config = Config.create();

Channel<String> fromJms = Channel.create("from-jms");

JmsConnector jmsConnector = JmsConnector.create();(1)

Messaging messaging = Messaging.builder()
        .config(config)
        .connector(jmsConnector)
        .listener(fromJms, payload -> {
            System.out.println("Jms says: " + payload);
        })
        .build()
        .start();
  1. Prepare JMS connector, can be used by any channel

Example of producing to JMS:
Config config = Config.create();

Channel<String> toJms = Channel.create("to-jms");

JmsConnector jmsConnector = JmsConnector.create();(1)

messaging = Messaging.builder()
        .config(config)
        .publisher(toJms, Multi.just("test1", "test2").map(Message::of))
        .connector(jmsConnector)
        .build()
        .start();
  1. Prepare JMS connector, can be used by any channel

Don’t forget to check out the examples with pre-configured ActiveMQ docker image, for easy testing:

AQ Connector

Maven dependency
<dependency>
    <groupId>io.helidon.messaging.aq</groupId>
    <artifactId>helidon-messaging-aq</artifactId>
</dependency>
Reactive Oracle AQ Connector
Sending and receiving
Example of producing to and consuming from Oracle AQ:
PoolDataSource pds = PoolDataSourceFactory.getPoolDataSource();(1)
pds.setConnectionFactoryClassName("oracle.jdbc.pool.OracleDataSource");
pds.setURL("jdbc:oracle:thin:@(DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(Host=192.168.0.123)(Port=1521))(CONNECT_DATA=(SID=XE)))");
pds.setUser("frank");
pds.setPassword("frank");

AqConnector seConn = AqConnector.builder()(2)
    .dataSource("test-ds", pds)
    .build();

Channel<String> toAq = Channel.<String>builder()(3)
    .name("toAq")
    .subscriberConfig(AqConnector.configBuilder()
        .queue("example_queue_1")
        .dataSource("test-ds")
        .build())
    .build();

Channel<String> fromAq = Channel.<String>builder()(4)
    .name("fromAq")
    .publisherConfig(AqConnector.configBuilder()
        .queue("example_queue_1")
        .dataSource("test-ds")
        .build())
    .build();

Messaging.builder()(5)
    .connector(seConn)
    .publisher(toAq, Multi.just("Hello", "world", "from", "Oracle", "DB!").map(Message::of))(6)
    .listener(fromAq, s -> System.out.pritln("Message received: "+s))(7)
    .build()
    .start();
  1. Prepare Oracle UCP

  2. Setup AQ connector and provide datasource with an identifier test-ds

  3. Setup channel for sending messages to queue example_queue_1 with datasource test-ds

  4. Setup channel for receiving messages from queue example_queue_1 with datasource test-ds

  5. Register connector and channels

  6. Add a publisher for several test messages to publish them to example_queue_1 immediately

  7. Subscribe callback for any message coming from example_queue_1