Spring-data-redis: 分布式队列

Redis中list数据结构，具有“双端队列”的特性，同时redis具有持久数据的能力，因此redis实现分布式队列是非常安全可靠的。它类似于JMS中的“Queue”，只不过功能和可靠性(事务性)并没有JMS严格。

Redis中的队列阻塞时，整个connection都无法继续进行其他操作，因此在基于连接池设计是需要注意。

我们通过spring-data-redis，来实现“同步队列”，设计风格类似与JMS。

一.配置文件：

<beans xmlns="http://www.springframework.org/schema/beans"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://www.springframework.org/schema/beans http://www.springframework.org/schema/beans/spring-beans.xsd" default-autowire="byName">
    <bean id="jedisPoolConfig" class="redis.clients.jedis.JedisPoolConfig">
        <property name="maxActive" value="32"></property>
        <property name="maxIdle" value="6"></property>
        <property name="maxWait" value="15000"></property>
        <property name="minEvictableIdleTimeMillis" value="300000"></property>
        <property name="numTestsPerEvictionRun" value="3"></property>
        <property name="timeBetweenEvictionRunsMillis" value="60000"></property>
        <property name="whenExhaustedAction" value="1"></property>
    </bean>
    <bean id="jedisConnectionFactory" class="org.springframework.data.redis.connection.jedis.JedisConnectionFactory" destroy-method="destroy">
        <property name="poolConfig" ref="jedisPoolConfig"></property>
        <property name="hostName" value="127.0.0.1"></property>
        <property name="port" value="6379"></property>
        <property name="password" value="0123456"></property>
        <property name="timeout" value="15000"></property>
        <property name="usePool" value="true"></property>
    </bean>
    <bean id="jedisTemplate" class="org.springframework.data.redis.core.RedisTemplate">
        <property name="connectionFactory" ref="jedisConnectionFactory"></property>
        <property name="defaultSerializer">
            <bean class="org.springframework.data.redis.serializer.StringRedisSerializer"/>
        </property>
    </bean>
    <bean id="jedisQueueListener" class="com.sample.redis.sdr.QueueListener"/>
    <bean id="jedisQueue" class="com.sample.redis.sdr.RedisQueue" destroy-method="destroy">
        <property name="redisTemplate" ref="jedisTemplate"></property>
        <property name="key" value="user:queue"></property>
        <property name="listener" ref="jedisQueueListener"></property>
    </bean>
</beans>

二.程序实例：

1) QueueListener：当队列中有数据时，可以执行类似于JMS的回调操作。

public interface RedisQueueListener<T> {

    public void onMessage(T value);
}

public class QueueListener<String> implements RedisQueueListener<String> {

    @Override
    public void onMessage(String value) {
        System.out.println(value);

    }

}

2) RedisQueue:队列操作，内部封装redisTemplate实例；如果配置了“listener”，那么queue将采用“消息回调”的方式执行，listenerThread是一个后台线程，用来自动处理“队列信息”。如果不配置“listener”，那么你可以将redisQueue注入到其他spring bean中，手动去“take”数据即可。

public class RedisQueue<T> implements InitializingBean,DisposableBean{
    private RedisTemplate redisTemplate;
    private String key;
    private int cap = Short.MAX_VALUE;//最大阻塞的容量，超过容量将会导致清空旧数据
    private byte[] rawKey;
    private RedisConnectionFactory factory;
    private RedisConnection connection;//for blocking
    private BoundListOperations<String, T> listOperations;//noblocking

    private Lock lock = new ReentrantLock();//基于底层IO阻塞考虑

    private RedisQueueListener listener;//异步回调
    private Thread listenerThread;

    private boolean isClosed;

    public void setRedisTemplate(RedisTemplate redisTemplate) {
        this.redisTemplate = redisTemplate;
    }

    public void setListener(RedisQueueListener listener) {
        this.listener = listener;
    }

    public void setKey(String key) {
        this.key = key;
    }


    @Override
    public void afterPropertiesSet() throws Exception {
        factory = redisTemplate.getConnectionFactory();
        connection = RedisConnectionUtils.getConnection(factory);
        rawKey = redisTemplate.getKeySerializer().serialize(key);
        listOperations = redisTemplate.boundListOps(key);
        if(listener != null){
            listenerThread = new ListenerThread();
            listenerThread.setDaemon(true);
            listenerThread.start();
        }
    }


    /**
     * blocking
     * remove and get last item from queue:BRPOP
     * @return
     */
    public T takeFromTail(int timeout) throws InterruptedException{
        lock.lockInterruptibly();
        try{
            List<byte[]> results = connection.bRPop(timeout, rawKey);
            if(CollectionUtils.isEmpty(results)){
                return null;
            }
            return (T)redisTemplate.getValueSerializer().deserialize(results.get(1));
        }finally{
            lock.unlock();
        }
    }

    public T takeFromTail() throws InterruptedException{
        return takeFromHead(0);
    }

    /**
     * 从队列的头，插入
     */
    public void pushFromHead(T value){
        listOperations.leftPush(value);
    }

    public void pushFromTail(T value){
        listOperations.rightPush(value);
    }

    /**
     * noblocking
     * @return null if no item in queue
     */
    public T removeFromHead(){
        return listOperations.leftPop();
    }

    public T removeFromTail(){
        return listOperations.rightPop();
    }

    /**
     * blocking
     * remove and get first item from queue:BLPOP
     * @return
     */
    public T takeFromHead(int timeout) throws InterruptedException{
        lock.lockInterruptibly();
        try{
            List<byte[]> results = connection.bLPop(timeout, rawKey);
            if(CollectionUtils.isEmpty(results)){
                return null;
            }
            return (T)redisTemplate.getValueSerializer().deserialize(results.get(1));
        }finally{
            lock.unlock();
        }
    }

    public T takeFromHead() throws InterruptedException{
        return takeFromHead(0);
    }

    @Override
    public void destroy() throws Exception {
        if(isClosed){
            return;
        }
        shutdown();
        RedisConnectionUtils.releaseConnection(connection, factory);
    }

    private void shutdown(){
        try{
            listenerThread.interrupt();
        }catch(Exception e){
            //
        }
    }

    class ListenerThread extends Thread {

        @Override
        public void run(){
            try{
                while(true){
                    T value = takeFromHead();//cast exceptionyou should check.
                    //逐个执行
                    if(value != null){
                        try{
                            listener.onMessage(value);
                        }catch(Exception e){
                            //
                        }
                    }
                }
            }catch(InterruptedException e){
                //
            }
        }
    }

}

3) 使用与测试：

public static void main(String[] args) throws Exception{
    ClassPathXmlApplicationContext context = new ClassPathXmlApplicationContext("classpath:spring-redis-beans.xml");
    RedisQueue<String> redisQueue = (RedisQueue)context.getBean("jedisQueue");
    redisQueue.pushFromHead("test:app");
    Thread.sleep(15000);
    redisQueue.pushFromHead("test:app");
    Thread.sleep(15000);
    redisQueue.destroy();
}

在程序运行期间，你可以通过redis-cli(客户端窗口)执行“lpush”，你会发现程序的控制台仍然能够正常打印队列信息。