上次我把 RabbitMQ 集成到项目中，但是每次使用 RabbitMQ 时都去 New 一个连接，导致并发起不来，所以这次我们就给 RabbitMQ 加一个连接池。

为了偷个懒，我直接用 ChatGPT 教我加。

01 ChatGPT 生成的代码

我输入的关键字如下：

复制代码

rabbitmq 实现一个连接池 java

下面是 GhatGPT 4.0 给我的回复：

java

复制代码

以下是在Java中使用连接池创建RabbitMQ连接的示例：首先，需要添加pika客户端依赖项，例如： <dependency> <groupId>com.rabbitmq</groupId> <artifactId>amqp-client</artifactId> <version>5.9.0</version> </dependency> 创建一个RabbitMQ连接类： import com.rabbitmq.client.Connection; public class RabbitMQConnection { private Connection connection; public RabbitMQConnection(String host, int port, String username, String password) { ConnectionFactory factory = new ConnectionFactory(); factory.setHost(host); factory.setPort(port); factory.setUsername(username); factory.setPassword(password); try { connection = factory.newConnection(); } catch (IOException | TimeoutException e) { e.printStackTrace(); } } public Connection getConnection() { return connection; } public void close() { try { connection.close(); } catch (IOException e) { e.printStackTrace(); } } } 创建一个RabbitMQ连接池类： import java.util.concurrent.BlockingQueue; import java.util.concurrent.LinkedBlockingQueue; public class RabbitMQConnectionPool { private final BlockingQueue<RabbitMQConnection> pool; public RabbitMQConnectionPool(String host, int port, String username, String password, int poolSize) { pool = new LinkedBlockingQueue<>(poolSize); for (int i = 0; i < poolSize; i++) { pool.add(new RabbitMQConnection(host, port, username, password)); } } public RabbitMQConnection getConnection() throws InterruptedException { return pool.take(); } public void returnConnection(RabbitMQConnection connection) { pool.add(connection); } public void close() { pool.forEach(RabbitMQConnection::close); } } 创建一个消费者类： import com.rabbitmq.client.*; public class RabbitMQConsumer implements Runnable { private final RabbitMQConnectionPool connectionPool; private final String queueName; public RabbitMQConsumer(RabbitMQConnectionPool connectionPool, String queueName) { this.connectionPool = connectionPool; this.queueName = queueName; } @Override public void run() { try { Connection connection = connectionPool.getConnection().getConnection(); Channel channel = connection.createChannel(); channel.queueDeclare(queueName, false, false, false, null); channel.basicConsume(queueName, true, (consumerTag, delivery) -> { String message = new String(delivery.getBody(), "UTF-8"); System.out.println("Received message: " + message); }, consumerTag -> {}); } catch (InterruptedException | IOException e) { e.printStackTrace(); } } } 创建一个生产者类： import com.rabbitmq.client.Channel; import com.rabbitmq.client.Connection; import com.rabbitmq.client.MessageProperties; public class RabbitMQProducer { private final RabbitMQConnectionPool connectionPool; private final String queueName; public RabbitMQProducer(RabbitMQConnectionPool connectionPool, String queueName) { this.connectionPool = connectionPool; this.queueName = queueName; } public void send(String message) { try { Connection connection = connectionPool.getConnection().getConnection(); Channel channel = connection.createChannel(); channel.queueDeclare(queueName, false, false, false, null); channel.basicPublish("", queueName, MessageProperties.PERSISTENT_TEXT_PLAIN, message.getBytes()); System.out.println("Sent message: " + message); channel.close(); connection.close(); } catch (InterruptedException|IOException e) { e.printStackTrace(); } } } 创建一个RabbitMQ连接池和多个消费者来消费队列： public class Main { public static void main(String[] args) throws InterruptedException { RabbitMQConnectionPool connectionPool = new RabbitMQConnectionPool("localhost", 5672, "guest", "guest", 10); RabbitMQProducer producer = new RabbitMQProducer(connectionPool, "hello"); producer.send("Hello, world!"); RabbitMQConsumer consumer1 = new RabbitMQConsumer(connectionPool, "hello"); Thread thread1 = new Thread(consumer1); thread1.start(); RabbitMQConsumer consumer2 = new RabbitMQConsumer(connectionPool, "hello"); Thread thread2 = new Thread(consumer2); thread2.start(); thread1.join(); thread2.join(); connectionPool.close(); } } 该示例代码创建了大小为10的连接池，并创建了两个消费者线程来消费队列，并使用一个生产者向队列发送消息。每个消费者从连接池获取连接，并消费队列中的消息，当完成任务后，将连接还回连接池。在实际部署中，您需要根据您的应用程序需求选择合适大小的池，并调整代码以使用正确的参数。

我靠，这代码写的真牛逼，我需要做的事情，就是把里面连接池的部分，集成到我的项目中。

02 给 RabbitMQ 加连接池

2.1 加入连接池

之前我们给 ConnectionFactory 加了个单例工厂，具体可以参考我上一篇文章，由于我们有连接池，所以这个单例工厂就直接废弃了。

typescript

复制代码

/** * @author Louzai * @date 2023/5/10 */ public class RabbitmqConnection { private Connection connection; public RabbitmqConnection(String host, int port, String userName, String password, String virtualhost) { ConnectionFactory connectionFactory = new ConnectionFactory(); connectionFactory.setHost(host); connectionFactory.setPort(port); connectionFactory.setUsername(userName); connectionFactory.setPassword(password); connectionFactory.setVirtualHost(virtualhost); try { connection = connectionFactory.newConnection(); } catch (IOException | TimeoutException e) { e.printStackTrace(); } } /** * 获取链接 * * @return */ public Connection getConnection() { return connection; } /** * 关闭链接 * */ public void close() { try { connection.close(); } catch (IOException e) { e.printStackTrace(); } } }

这个代码就是 ChatGPT 给我写的，我是直接 Copy 过来，然后稍微改动了一下。

arduino

复制代码

public class RabbitmqConnectionPool { private static BlockingQueue<RabbitmqConnection> pool; public static void initRabbitmqConnectionPool(String host, int port, String userName, String password, String virtualhost, Integer poolSize) { pool = new LinkedBlockingQueue<>(poolSize); for (int i = 0; i < poolSize; i++) { pool.add(new RabbitmqConnection(host, port, userName, password, virtualhost)); } } public static RabbitmqConnection getConnection() throws InterruptedException { return pool.take(); } public static void returnConnection(RabbitmqConnection connection) { pool.add(connection); } public static void close() { pool.forEach(RabbitmqConnection::close); } }

2.2 RabbitMQ 发送消费消息

RabbitMQ 发送消息：从连接池拿到连接 -> 创建通道 -> 声明交换机 -> 发送消息 -> 将连接归还连接池。

这里的逻辑基本和之前的一样，只是之前是 New 一个连接，现在是直接从连接池拿到连接，然后最后多了一步归还连接的操作。

typescript

复制代码

@Override public void publishMsg(String exchange, BuiltinExchangeType exchangeType, String toutingKey, String message) { try { //创建连接 RabbitmqConnection rabbitmqConnection = RabbitmqConnectionPool.getConnection(); Connection connection = rabbitmqConnection.getConnection(); //创建消息通道 Channel channel = connection.createChannel(); // 声明exchange中的消息为可持久化，不自动删除 channel.exchangeDeclare(exchange, exchangeType, true, false, null); // 发布消息 channel.basicPublish(exchange, toutingKey, null, message.getBytes()); System.out.println("Publish msg:" + message); channel.close(); RabbitmqConnectionPool.returnConnection(rabbitmqConnection); } catch (InterruptedException | IOException | TimeoutException e) { e.printStackTrace(); } }

RabbitMQ 消费消息：从连接池拿到连接 -> 创建通道 -> 确定消息队列 -> 绑定队列到交换机 -> 接受并消费消息 -> 将连接归还连接池。

同上，这里的逻辑基本和之前的一样，只是多了一个拿连接和归还连接的过程。

java

复制代码

@Override public void consumerMsg(String exchange, String queueName, String routingKey) { try { //创建连接 RabbitmqConnection rabbitmqConnection = RabbitmqConnectionPool.getConnection(); Connection connection = rabbitmqConnection.getConnection(); //创建消息信道 final Channel channel = connection.createChannel(); //消息队列 channel.queueDeclare(queueName, true, false, false, null); //绑定队列到交换机 channel.queueBind(queueName, exchange, routingKey); Consumer consumer = new DefaultConsumer(channel) { @Override public void handleDelivery(String consumerTag, Envelope envelope, AMQP.BasicProperties properties, byte[] body) throws IOException { String message = new String(body, "UTF-8"); System.out.println("Consumer msg:" + message); // 获取Rabbitmq消息，并保存到DB // 说明：这里仅作为示例，如果有多种类型的消息，可以根据消息判定，简单的用 if...else 处理，复杂的用工厂 + 策略模式 notifyService.saveArticleNotify(JsonUtil.toObj(message, UserFootDO.class), NotifyTypeEnum.PRAISE); channel.basicAck(envelope.getDeliveryTag(), false); } }; // 取消自动ack channel.basicConsume(queueName, false, consumer); RabbitmqConnectionPool.returnConnection(rabbitmqConnection); } catch (InterruptedException | IOException e) { e.printStackTrace(); } }

这个代码，其实 ChatGPT 写的有问题，你再回过头去看 ChatGPT 写的代码，发现连接取出，但是没有归还，那会出现什么问题呢？

这里给大家分析一下，由于我们的连接池用的是 BlockingQueue，连接池大小是 5，如果连接全部取出并都不归还，当第 6 个请求过来后，请求就卡住了，导致界面操作会被阻塞，请求完全没有反应。

不要问我怎么知道，因为我是踩坑过来的。