spring整合消息隊列rabbitmq

時間 2019-12-05

原文原文鏈接

spring你們太熟，就很少說了java

rabbitmq一個amqp的隊列服務實現,具體介紹請參考本文http://lynnkong.iteye.com/blog/1699684 spring

本文側重介紹如何將rabbitmq整合到項目中apache

ps:本文只是簡單一個整合介紹，屬於拋磚引玉，具體實現還需你們深刻研究哈..json

1.首先是生產者配置緩存

<?xml version="1.0" encoding="UTF-8"?>
<beans xmlns="http://www.springframework.org/schema/beans"
       xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
       xmlns:context="http://www.springframework.org/schema/context"
       xmlns:rabbit="http://www.springframework.org/schema/rabbit"
       xsi:schemaLocation="
            http://www.springframework.org/schema/beans
                http://www.springframework.org/schema/beans/spring-beans.xsd
            http://www.springframework.org/schema/context
                http://www.springframework.org/schema/context/spring-context.xsd
            http://www.springframework.org/schema/rabbit
                http://www.springframework.org/schema/rabbit/spring-rabbit-1.0.xsd">

    
   <!-- 鏈接服務配置  -->
   <rabbit:connection-factory id="connectionFactory" host="localhost" username="guest"
        password="guest" port="5672"  />
        
   <rabbit:admin connection-factory="connectionFactory"/>
   
   <!-- queue 隊列聲明-->
   <rabbit:queue id="queue_one" durable="true" auto-delete="false" exclusive="false" name="queue_one"/>
   
   
   <!-- exchange queue binging key 綁定 -->
    <rabbit:direct-exchange name="my-mq-exchange" durable="true" auto-delete="false" id="my-mq-exchange">
        <rabbit:bindings>
            <rabbit:binding queue="queue_one" key="queue_one_key"/>
        </rabbit:bindings>
    </rabbit:direct-exchange>
    
    <-- spring amqp默認的是jackson 的一個插件,目的將生產者生產的數據轉換爲json存入消息隊列，因爲fastjson的速度快於jackson,這裏替換爲fastjson的一個實現 -->
    <bean id="jsonMessageConverter"  class="mq.convert.FastJsonMessageConverter"></bean>
    
    <-- spring template聲明-->
    <rabbit:template exchange="my-mq-exchange" id="amqpTemplate"  connection-factory="connectionFactory"  message-converter="jsonMessageConverter"/>
</beans>

2.fastjson messageconver插件實現ide

import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.springframework.amqp.core.Message;
import org.springframework.amqp.core.MessageProperties;
import org.springframework.amqp.support.converter.AbstractMessageConverter;
import org.springframework.amqp.support.converter.MessageConversionException;

import fe.json.FastJson;

public class FastJsonMessageConverter  extends AbstractMessageConverter {
	private static Log log = LogFactory.getLog(FastJsonMessageConverter.class);

	public static final String DEFAULT_CHARSET = "UTF-8";

	private volatile String defaultCharset = DEFAULT_CHARSET;
	
	public FastJsonMessageConverter() {
		super();
		//init();
	}
	
	public void setDefaultCharset(String defaultCharset) {
		this.defaultCharset = (defaultCharset != null) ? defaultCharset
				: DEFAULT_CHARSET;
	}
	
	public Object fromMessage(Message message)
			throws MessageConversionException {
		return null;
	}
	
	public <T> T fromMessage(Message message,T t) {
		String json = "";
		try {
			json = new String(message.getBody(),defaultCharset);
		} catch (UnsupportedEncodingException e) {
			e.printStackTrace();
		}
		return (T) FastJson.fromJson(json, t.getClass());
	}	
	

	protected Message createMessage(Object objectToConvert,
			MessageProperties messageProperties)
			throws MessageConversionException {
		byte[] bytes = null;
		try {
			String jsonString = FastJson.toJson(objectToConvert);
			bytes = jsonString.getBytes(this.defaultCharset);
		} catch (UnsupportedEncodingException e) {
			throw new MessageConversionException(
					"Failed to convert Message content", e);
		} 
		messageProperties.setContentType(MessageProperties.CONTENT_TYPE_JSON);
		messageProperties.setContentEncoding(this.defaultCharset);
		if (bytes != null) {
			messageProperties.setContentLength(bytes.length);
		}
		return new Message(bytes, messageProperties);

	}
}

3.生產者端調用this

import java.util.List;

import org.springframework.amqp.core.AmqpTemplate;


public class MyMqGatway {
	
	@Autowired
	private AmqpTemplate amqpTemplate;
	
	public void sendDataToCrQueue(Object obj) {
		amqpTemplate.convertAndSend("queue_one_key", obj);
	}	
}

4.消費者端配置(與生產者端大同小異)插件

<?xml version="1.0" encoding="UTF-8"?>
<beans xmlns="http://www.springframework.org/schema/beans"
       xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
       xmlns:context="http://www.springframework.org/schema/context"
       xmlns:rabbit="http://www.springframework.org/schema/rabbit"
       xsi:schemaLocation="
            http://www.springframework.org/schema/beans
                http://www.springframework.org/schema/beans/spring-beans.xsd
            http://www.springframework.org/schema/context
                http://www.springframework.org/schema/context/spring-context.xsd
            http://www.springframework.org/schema/rabbit
                http://www.springframework.org/schema/rabbit/spring-rabbit-1.0.xsd">

    
   <!-- 鏈接服務配置  -->
   <rabbit:connection-factory id="connectionFactory" host="localhost" username="guest"
        password="guest" port="5672"  />
        
   <rabbit:admin connection-factory="connectionFactory"/>
   
   <!-- queue 隊列聲明-->
   <rabbit:queue id="queue_one" durable="true" auto-delete="false" exclusive="false" name="queue_one"/>
   
   
   <!-- exchange queue binging key 綁定 -->
    <rabbit:direct-exchange name="my-mq-exchange" durable="true" auto-delete="false" id="my-mq-exchange">
        <rabbit:bindings>
            <rabbit:binding queue="queue_one" key="queue_one_key"/>
        </rabbit:bindings>
    </rabbit:direct-exchange>

    
     
    <!-- queue litener  觀察 監聽模式 當有消息到達時會通知監聽在對應的隊列上的監聽對象-->
    <rabbit:listener-container connection-factory="connectionFactory" acknowledge="auto" task-executor="taskExecutor">
        <rabbit:listener queues="queue_one" ref="queueOneLitener"/>
    </rabbit:listener-container>
</beans>

5.消費者端調用線程

import org.springframework.amqp.core.Message;
import org.springframework.amqp.core.MessageListener;

public class QueueOneLitener implements  MessageListener{
	@Override
	public void onMessage(Message message) {
		System.out.println(" data :" + message.getBody());
	}
}

6.因爲消費端當隊列有數據到達時，對應監聽的對象就會被通知到，沒法作到批量獲取，批量入庫，所以能夠在消費端緩存一個臨時隊列，將mq取出來的數據存入本地隊列，後臺線程定時批量處理便可code

打個廣告：創業項目www.1zheke.com，有培訓訴求的人能夠支持一下，或者周圍有培訓需求的人，幫忙擴散一下，限量1折課，獎學拿到手軟，很是感謝