RabbitMQ~消费者实时与消息服务器保持通话

这个文章主要介绍简单的消费者的实现,rabbitMQ实现的消费者可以对消息服务器进行实时监听,当有消息(生产者把消息推到服务器上之后),消费者可以自动去消费它,这通常是开启一个进程去维护这个对话,它与消息服务器保持一个TCP的长连接,整个这个过程于rabbitMQ为我们提供,程序开发人员只需要实现自己的回调方法即可.

简单的rabbitMQ消费者

/// <summary>
    /// 消息消费者
    /// </summary>
    public class RabbitMqSubscriber : Lind.DDD.Commons.DisposableBase
    {
        private readonly string exchangeName;
        private readonly string queueName;
        private readonly IConnection connection;
        private readonly IModel channel;
        private bool disposed;

        /// <summary>
        /// 从消息服务器拉到消息后触发
        /// </summary>
        public event EventHandler<MessageReceivedEventArgs> MessageReceived;

        /// <summary>
        /// Initializes a new instance of <c>RabbitMqMessageSubscriber</c> class.
        /// </summary>
        /// <param name="uri"></param>
        /// <param name="exchangeName"></param>
        /// <param name="queueName"></param>
        public RabbitMqSubscriber(string uri, string queueName, string userName = "", string password = "")
        {
            this.exchangeName = exchangeName;
            this.queueName = queueName;
            var factory = new ConnectionFactory() { Uri = uri };
            if (!string.IsNullOrWhiteSpace(userName))
                factory.UserName = userName;
            if (!string.IsNullOrWhiteSpace(password))
                factory.Password = password;
            this.connection = factory.CreateConnection();
            this.channel = connection.CreateModel();
        }

        public void Subscribe()
        {
            channel.QueueDeclare(
                queue: this.queueName, 
                durable: false,//持久化
                exclusive: false, //独占,只能被一个consumer使用
                autoDelete: false,//自己删除,在最后一个consumer完成后删除它
                arguments: null);
            var consumer = new EventingBasicConsumer(channel);
            consumer.Received += (sender, e) =>
            {
                var body = e.Body;
                var json = Encoding.UTF8.GetString(body);
                var message = JsonConvert.DeserializeObject(json, new JsonSerializerSettings { TypeNameHandling = TypeNameHandling.All });
                this.OnMessageReceived(new MessageReceivedEventArgs(message));
                channel.BasicAck(e.DeliveryTag, multiple: false);
            };
            channel.BasicConsume(queue: queueName,
                                 noAck: false,
                                 consumer: consumer);
        }

        private void OnMessageReceived(MessageReceivedEventArgs e)
        {
            this.MessageReceived?.Invoke(this, e);
        }

        protected override void Finalize(bool disposing)
        {
            if (disposing)
            {
                if (!disposed)
                {
                    this.channel.Dispose();
                    this.connection.Dispose();
                    disposed = true;
                }
            }
        }
    }

简单调用

class Program
    {
        static void Main(string[] args)
        {
            var subscriber = new Lind.DDD.RabbitMq.RabbitMqSubscriber("amqp://localhost:5672", "zzl");
            subscriber.MessageReceived += Subscriber_MessageReceived;
            subscriber.Subscribe();
            Console.ReadKey();
        }

        private static void Subscriber_MessageReceived(object sender, RabbitMq.MessageReceivedEventArgs e)
        {
            Console.WriteLine("消费者2->消费了一个消息{0}", e.Message);
            Lind.DDD.Logger.LoggerFactory.Instance.Logger_Debug("消费者2->消费了一个消息{0}" + e.Message);
            Thread.Sleep(2000);
        }

    }

实时拉消息

RabbitMQ消息模型

通过上面图我们可以更容易和清晰的去理解rabbitmq的工作流程.

作者：仓储大叔，张占岭，
荣誉：微软MVP