简单聊一聊spring cloud stream和kafka的那点事

Spring Cloud Stream is a framework for building highly scalable event-driven microservices connected with shared messaging systems.

The framework provides a flexible programming model built on already established and familiar Spring idioms and best practices, including support for persistent pub/sub semantics, consumer groups, and stateful partitions.

野生翻译：spring cloud stream是打算统一消息中间件后宫的男人，他身手灵活，身后有靠山spring，会使十八般武器（消息订阅模式啦，消费者组，stateful partitions什么的），目前后宫有东宫娘娘kafka和西宫娘娘rabbitMQ。

八卦党：今天我们扒一扒spring cloud stream和kafka的关系，rabbitMQ就让她在冷宫里面呆着吧。

1、先出场的正宫娘娘：kafka

Apache Kafka® is a distributed streaming platform. What exactly does that mean?

A streaming platform has three key capabilities:

• Publish and subscribe to streams of records, similar to a message queue or enterprise messaging system.
• Store streams of records in a fault-tolerant durable way.
• Process streams of records as they occur.

野生翻译：老娘是个流处理平台，能干的活可多了：

• 能处理发布/订阅消息
• 用很稳的方式保存消息
• 一来就处理，真的很快

总结一句话，就是快、稳、准。

kafka的运行非常简单，从这里下载，然后先运行zookeeper。在最新的kafka的下载包里面也包含了一个zookeeper，可以直接用里面的。zookeeper启动后，需要在kafka的配置文件里面配置好zookeeper的ip和端口，配置文件是config/server.properties。

############################# Zookeeper #############################

# Zookeeper connection string (see zookeeper docs for details).
# This is a comma separated host:port pairs, each corresponding to a zk
# server. e.g. "127.0.0.1:3000,127.0.0.1:3001,127.0.0.1:3002".
# You can also append an optional chroot string to the urls to specify the
# root directory for all kafka znodes.
zookeeper.connect=localhost:2181

# Timeout in ms for connecting to zookeeper
zookeeper.connection.timeout.ms=6000

然后运行bin目录下的命令，启动kafka就可以啦

bin/kafka-server-start.sh -daemon config/server.properties

2、kafka的贴身总管，kafka-manager

kafka虽然启动了，但我们需要了解她的话，还是需要一个总管来汇报情况，我这边用的就是kafka-manager，下载地址在这里。很可惜的是只有源代码的下载，没有可运行版本的，需要自行编译，这个编译速度还挺慢的，我这边提供一个编译好的版本给大家，点这里。

kafka-manager同样需要配置一下和kafka的关系，在conf/application.conf文件里面，不过配置的不是kafka自己，而是kafka挂载的zookeeper。

kafka-manager.zkhosts="localhost:2181"

然后启动bin/kafka-manager就可以了（windows环境下也有kafka-manager.bat可以运行）

这里有个坑，在windows下面运行的话，可能启动失败，提示输入行太长

这个是因为目录太长，把kafak-manager-2.0.0.2目录名缩短就可以正常运行了。

启动后通过Add Cluster把Cluster Zookeeper Host把zookeeper的地址端口填上，Kafka Version的版本一定要和正在使用的kafka版本对上，否则可能看不到kafka的内容。

然后我们就能看到kafka的broker，topic，consumers，partitions等信息了。

3、皇上驾到，spring cloud stream

一切的起点，还在start.spring.io

这黑乎乎的界面是spring为了万圣节搞的事情。和我们相关的是右边这两个依赖，这两个依赖在pom.xml里面对应的是这些

<dependencies>
        <dependency>
			<groupId>org.apache.kafka</groupId>
			<artifactId>kafka-streams</artifactId>
		</dependency>
		<dependency>
			<groupId>org.springframework.cloud</groupId>
			<artifactId>spring-cloud-stream</artifactId>
		</dependency>
		<dependency>
			<groupId>org.springframework.cloud</groupId>
			<artifactId>spring-cloud-stream-binder-kafka-streams</artifactId>
		</dependency>
        <dependency>
			<groupId>org.springframework.cloud</groupId>
			<artifactId>spring-cloud-stream-test-support</artifactId>
			<scope>test</scope>
		</dependency>
</dependencies>
<dependencyManagement>
		<dependencies>
			<dependency>
				<groupId>org.springframework.cloud</groupId>
				<artifactId>spring-cloud-dependencies</artifactId>
				<version>${spring-cloud.version}</version>
				<type>pom</type>
				<scope>import</scope>
			</dependency>
		</dependencies>
	</dependencyManagement>

不过只凭这些还不行，直接运行的话，会提示

Caused by: java.lang.IllegalStateException: Unknown binder configuration: kafka

还需要加上一个依赖包

        <dependency>
			<groupId>org.springframework.cloud</groupId>
			<artifactId>spring-cloud-stream-binder-kafka</artifactId>
		</dependency>

4、发消息，biubiubiu

spring cloud stream项目框架搭好后，我们需要分两个部分，一个是发消息的部分，一个是收消息的地方。我们先看发消息的部分，首先是配置文件,application.yml

spring:
  cloud:
    stream:
      default-binder: kafka #默认的绑定器，
      kafka: #如果用的是rabbitMQ这里填 rabbit
        binder:
          brokers: #Kafka的消息中间件服务器地址
          - localhost:9092
      bindings:
        output: #通道名称
          binder: kafka
          destination: test1 #消息发往的目的地，对应topic
          group: output-group-1 #对应kafka的group
          content-type: text/plain #消息的格式

注意这里的output，表示是发布消息的，和后面订阅消息是对应的。这个output的名字是消息通道名称，是可以自定义的，后面会讲到。

然后我们需要创建一个发布者

import org.springframework.cloud.stream.annotation.EnableBinding;
import org.springframework.cloud.stream.messaging.Source;

@EnableBinding(Source.class)
public class Producer {
	private Source mySource;

	public Producer(Source mySource) {
		super();
		this.mySource = mySource;
	}

	public Source getMysource() {
		return mySource;
	}

	public void setMysource(Source mysource) {
		mySource = mySource;
	}
}

@EnableBinding 按字面理解就知道是绑定通道的，绑定的通道名就是上面的output，Soure.class是spring 提供的，表示这是一个可绑定的发布通道，它的通道名称就是output,和application.yml里面的output对应

源码可以看的很清楚

package org.springframework.cloud.stream.messaging;

import org.springframework.cloud.stream.annotation.Output;
import org.springframework.messaging.MessageChannel;

/**
 * Bindable interface with one output channel.
 *
 * @author Dave Syer
 * @author Marius Bogoevici
 * @see org.springframework.cloud.stream.annotation.EnableBinding
 */
public interface Source {

	/**
	 * Name of the output channel.
	 */
	String OUTPUT = "output";

	/**
	 * @return output channel
	 */
	@Output(Source.OUTPUT)
	MessageChannel output();

}

如果我们需要定义我们自己的通道，可以自己写一个类，比如下面这种，通道名就改成了my-out

import org.springframework.cloud.stream.annotation.Input;
import org.springframework.cloud.stream.annotation.Output;
import org.springframework.messaging.MessageChannel;
import org.springframework.messaging.SubscribableChannel;
public interface MySource {
    String INPUT = "my-in";
    String OUTPUT = "my-out";
    @Input(INPUT)
    SubscribableChannel myInput();
    @Output(OUTPUT)
    MessageChannel myOutput();
}

这样的话，application.yml就要改了

        my-out:
          binder: kafka
          destination: mytest #消息发往的目的地，对应topic
          group: output-group-2 #对应kafka的group
          content-type: text/plain #消息的格式

Product.class的@EnableBinding也需要改，为了做对应，我另外写了一个MyProducer

import org.springframework.cloud.stream.annotation.EnableBinding;

@EnableBinding(MySource.class)
public class MyProducer {
	private MySource mySource;

	public MyProducer(MySource mySource) {
		super();
		this.mySource = mySource;
	}

	public MySource getMysource() {
		return mySource;
	}

	public void setMysource(MySource mysource) {
		mySource = mySource;
	}
}

这样，发布消息的部分就写好了，我们写个controller来发送消息

import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.messaging.support.MessageBuilder;
import org.springframework.web.bind.annotation.RequestBody;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RequestMethod;
import org.springframework.web.bind.annotation.RestController;

import com.wphmoon.kscs.service.ChatMessage;
import com.wphmoon.kscs.service.MyProducer;
import com.wphmoon.kscs.service.Producer;

@RestController
public class MyController {
	@Autowired
	private Producer producer;
	@Autowired
	private MyProducer myProducer;

// get the String message via HTTP, publish it to broker using spring cloud stream
	@RequestMapping(value = "/sendMessage/string", method = RequestMethod.POST)
	public String publishMessageString(@RequestBody String payload) {
// send message to channel output
		producer.getMysource().output().send(MessageBuilder.withPayload(payload).setHeader("type", "string").build());
		return "success";
	}
	@RequestMapping(value = "/sendMyMessage/string", method = RequestMethod.POST)
	public String publishMyMessageString(@RequestBody String payload) {
// send message to channel myoutput
		myProducer.getMysource().myOutput().send(MessageBuilder.withPayload(payload).setHeader("type", "string").build());
		return "success";
	}
}

很简单，直接调用producer发送一个字符串就行了，我使用postman来发起这个动作

消息发送出去了，我们怎么收消息呢？往下看。

5、收消息，来来来

同样的，我们用之前的spring cloud stream项目框架做收消息的部分，首先是application.yml文件

server:
  port: 8081
spring:
  cloud:
    stream:
      default-binder: kafka
      kafka:
        binder:
          brokers:
          - localhost:9092
      bindings:
        input:
         binder: kafka
         destination: test1
         content-type: text/plain
         group: input-group-1
        my-in:
         binder: kafka
         destination: mytest
         content-type: text/plain
         group: input-group-2

重点关注的就是input和my-in ,这个和之前的output和my-out一一对应。

默认和Source类对应的是Sink，这个是官方提供的，代码如下

package org.springframework.cloud.stream.messaging;

import org.springframework.cloud.stream.annotation.Input;
import org.springframework.messaging.SubscribableChannel;

/**
 * Bindable interface with one input channel.
 *
 * @author Dave Syer
 * @author Marius Bogoevici
 * @see org.springframework.cloud.stream.annotation.EnableBinding
 */
public interface Sink {

	/**
	 * Input channel name.
	 */
	String INPUT = "input";

	/**
	 * @return input channel.
	 */
	@Input(Sink.INPUT)
	SubscribableChannel input();

}

调用它的类Consumer用来接收消息，代码如下

import java.time.Instant;
import java.time.ZoneId;
import java.time.format.DateTimeFormatter;
import java.time.format.FormatStyle;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.cloud.stream.annotation.EnableBinding;
import org.springframework.cloud.stream.annotation.StreamListener;
import org.springframework.cloud.stream.messaging.Sink;
import org.springframework.messaging.handler.annotation.Payload;

@EnableBinding(Sink.class)
public class Consumer {
	private static final Logger logger = LoggerFactory.getLogger(Consumer.class);

	@StreamListener(target = Sink.INPUT)
	public void consume(String message) {
		logger.info("recieved a string message : " + message);
	}

	@StreamListener(target = Sink.INPUT, condition = "headers['type']=='chat'")
	public void handle(@Payload ChatMessage message) {
		final DateTimeFormatter df = DateTimeFormatter.ofLocalizedTime(FormatStyle.MEDIUM)
				.withZone(ZoneId.systemDefault());
		final String time = df.format(Instant.ofEpochMilli(message.getTime()));
		logger.info("recieved a complex message : [{}]: {}", time, message.getContents());
	}
}

而我们自定义channel的类MySink和MyConsumer代码如下：

import org.springframework.cloud.stream.annotation.Input;
import org.springframework.messaging.SubscribableChannel;

public interface MySink {
	String INPUT = "my-in";
    @Input(INPUT)
    SubscribableChannel myInput();
}

import java.time.Instant;
import java.time.ZoneId;
import java.time.format.DateTimeFormatter;
import java.time.format.FormatStyle;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.cloud.stream.annotation.EnableBinding;
import org.springframework.cloud.stream.annotation.StreamListener;
import org.springframework.cloud.stream.messaging.Sink;
import org.springframework.messaging.handler.annotation.Payload;

@EnableBinding(MySink.class)
public class MyConsumer {
	private static final Logger logger = LoggerFactory.getLogger(MyConsumer.class);

	@StreamListener(target = MySink.INPUT)
	public void consume(String message) {
		logger.info("recieved a string message : " + message);
	}

	@StreamListener(target = MySink.INPUT, condition = "headers['type']=='chat'")
	public void handle(@Payload ChatMessage message) {
		final DateTimeFormatter df = DateTimeFormatter.ofLocalizedTime(FormatStyle.MEDIUM)
				.withZone(ZoneId.systemDefault());
		final String time = df.format(Instant.ofEpochMilli(message.getTime()));
		logger.info("recieved a complex message : [{}]: {}", time, message.getContents());
	}
}

这样就OK了，当上面我们用postman发了消息后，这边就能直接在日志里面看到

2019-10-29 18:42:39.455  INFO 13556 --- [container-0-C-1] com.wphmoon.kscsclient.MyConsumer        : recieved a string message : 你瞅啥
2019-10-29 18:43:17.017  INFO 13556 --- [container-0-C-1] com.wphmoon.kscsclient.Consumer          : recieved a string message : 你瞅啥

6、到kafka-manager里面再瞅瞅

我们在application.yml里面定义的destination，就是kafka的topic，在kafka-manager的topic list里面可以看到

而接收消息的consumer也可以看到

这就是spring cloud stream和kafka的帝后之恋，不过他们这种政治联姻哪有这么简单，里面复杂的部分我们后面再讲，敬请期待，起驾回宫（野生翻译:The Return of the King）

源代码地址