前言

在上一篇文章中,我们知道了RabbitMQ的消息流程如下:

但在具体的使用中,我们还需知道exchange的类型,因为不同的类型对应不同的队列和路由规则。

在rabbitmq中,exchange有4个类型:direct,topic,fanout,header。

direct exchange

此类型的exchange路由规则很简单:

exchange在和queue进行binding时会设置routingkey

channel.QueueBind(queue: "create_pdf_queue",
exchange: "pdf_events",
routingKey: "pdf_create",
arguments: null);

然后我们在将消息发送到exchange时会设置对应的routingkey

channel.BasicPublish(exchange: "pdf_events",
routingKey: "pdf_create",
basicProperties: properties,
body: body);

在direct类型的exchange中,只有这两个routingkey完全相同,exchange才会选择对应的binging进行消息路由。

具体的流程如下:

通过代码可以会理解好一点:

var factory = new ConnectionFactory() { HostName = "localhost" };
using (var connection = factory.CreateConnection())
using (var channel = connection.CreateModel())
{
// Direct类型的exchange, 名称 pdf_events
channel.ExchangeDeclare(exchange: "pdf_events",
type: ExchangeType.Direct,
durable: true,
autoDelete: false,
arguments: null); // 创建create_pdf_queue队列
channel.QueueDeclare(queue: "create_pdf_queue",
durable: true,
exclusive: false,
autoDelete: false,
arguments: null); //创建 pdf_log_queue队列
channel.QueueDeclare(queue: "pdf_log_queue",
durable: true,
exclusive: false,
autoDelete: false,
arguments: null); //绑定 pdf_events --> create_pdf_queue 使用routingkey:pdf_create
channel.QueueBind(queue: "create_pdf_queue",
exchange: "pdf_events",
routingKey: "pdf_create",
arguments: null); //绑定 pdf_events --> pdf_log_queue 使用routingkey:pdf_log
channel.QueueBind(queue: "pdf_log_queue",
exchange: "pdf_events",
routingKey: "pdf_log",
arguments: null); var message = "Demo some pdf creating...";
var body = Encoding.UTF8.GetBytes(message);
var properties = channel.CreateBasicProperties();
properties.Persistent = true; //发送消息到exchange :pdf_events ,使用routingkey: pdf_create
//通过binding routinekey的比较,次消息会路由到队列 create_pdf_queue
channel.BasicPublish(exchange: "pdf_events",
routingKey: "pdf_create",
basicProperties: properties,
body: body); message = "pdf loging ...";
body = Encoding.UTF8.GetBytes(message);
properties = channel.CreateBasicProperties();
properties.Persistent = true; //发送消息到exchange :pdf_events ,使用routingkey: pdf_log
//通过binding routinekey的比较,次消息会路由到队列 pdf_log_queue
channel.BasicPublish(exchange: "pdf_events",
routingKey: "pdf_log",
basicProperties: properties,
body: body); }

topic exchange

此类型exchange和上面的direct类型差不多,但direct类型要求routingkey完全相等,这里的routingkey可以有通配符:'*','#'.

其中'*'表示匹配一个单词, '#'则表示匹配没有或者多个单词

如上图第一个binding:

  • exchange: agreements
  • queue A:  berlin_agreements
  • binding routingkey: agreements.eu.berlin.#

第二个binding:

  • exchange: agreements
  • queue B: all_agreements
  • binding routingkey: agreements.#

第三个binding:

  • exchange: agreements
  • queue c: headstore_agreements
  • binding routingkey: agreements.eu.*.headstore

所以如果我们消息的routingkey为agreements.eu.berlin那么符合第一和第二个binding,但最后一个不符合,具体的代码如下:

var factory = new ConnectionFactory() { HostName = "localhost" };
using (var connection = factory.CreateConnection())
using (var channel = connection.CreateModel())
{
// Topic类型的exchange, 名称 agreements
channel.ExchangeDeclare(exchange: "agreements",
type: ExchangeType.Topic,
durable: true,
autoDelete: false,
arguments: null); // 创建berlin_agreements队列
channel.QueueDeclare(queue: "berlin_agreements",
durable: true,
exclusive: false,
autoDelete: false,
arguments: null); //创建 all_agreements 队列
channel.QueueDeclare(queue: "all_agreements",
durable: true,
exclusive: false,
autoDelete: false,
arguments: null); //创建 headstore_agreements 队列
channel.QueueDeclare(queue: "headstore_agreements",
durable: true,
exclusive: false,
autoDelete: false,
arguments: null); //绑定 agreements --> berlin_agreements 使用routingkey:agreements.eu.berlin.#
channel.QueueBind(queue: "berlin_agreements",
exchange: "agreements",
routingKey: "agreements.eu.berlin.#",
arguments: null); //绑定 agreements --> all_agreements 使用routingkey:agreements.#
channel.QueueBind(queue: "all_agreements",
exchange: "agreements",
routingKey: "agreements.#",
arguments: null); //绑定 agreements --> headstore_agreements 使用routingkey:agreements.eu.*.headstore
channel.QueueBind(queue: "headstore_agreements",
exchange: "agreements",
routingKey: "agreements.eu.*.headstore",
arguments: null); var message = "hello world";
var body = Encoding.UTF8.GetBytes(message);
var properties = channel.CreateBasicProperties();
properties.Persistent = true; //发送消息到exchange :agreements ,使用routingkey: agreements.eu.berlin
//agreements.eu.berlin 匹配 agreements.eu.berlin.# 和agreements.#
//agreements.eu.berlin 不匹配 agreements.eu.*.headstore
//最终次消息会路由到队里:berlin_agreements(agreements.eu.berlin.#) 和 all_agreements(agreements.#)
channel.BasicPublish(exchange: "agreements",
routingKey: "agreements.eu.berlin",
basicProperties: properties,
body: body); }

fanout exchange

此exchange的路由规则很简单直接将消息路由到所有绑定的队列中,无须对消息的routingkey进行匹配操作。

header exchange

此类型的exchange和以上三个都不一样,其路由的规则是根据header来判断,其中的header就是以下方法的arguments参数:

Dictionary<string, object> aHeader = new Dictionary<string, object>();
aHeader.Add("format", "pdf");
aHeader.Add("type", "report");
aHeader.Add("x-match", "all");
channel.QueueBind(queue: "queue.A",
exchange: "agreements",
routingKey: string.Empty,
arguments: aHeader);

其中的x-match为特殊的header,可以为all则表示要匹配所有的header,如果为any则表示只要匹配其中的一个header即可。

在发布消息的时候就需要传入header值:

var properties = channel.CreateBasicProperties();
properties.Persistent = true;
Dictionary<string, object> mHeader1 = new Dictionary<string, object>();
mHeader1.Add("format", "pdf");
mHeader1.Add("type", "report");
properties.Headers = mHeader1;

具体的规则可以看以下代码:

var factory = new ConnectionFactory() { HostName = "localhost" };
using (var connection = factory.CreateConnection())
using (var channel = connection.CreateModel())
{
// Headers类型的exchange, 名称 agreements
channel.ExchangeDeclare(exchange: "agreements",
type: ExchangeType.Headers,
durable: true,
autoDelete: false,
arguments: null); // 创建queue.A队列
channel.QueueDeclare(queue: "queue.A", durable: true, exclusive: false, autoDelete: false, arguments: null); //创建 queue.B 队列
channel.QueueDeclare(queue: "queue.B", durable: true, exclusive: false, autoDelete: false, arguments: null); //创建 queue.C 队列
channel.QueueDeclare(queue: "queue.C", durable: true, exclusive: false, autoDelete: false, arguments: null); //绑定 agreements --> queue.A 使用arguments (format=pdf, type=report, x-match=all)
Dictionary<string, object> aHeader = new Dictionary<string, object>();
aHeader.Add("format", "pdf");
aHeader.Add("type", "report");
aHeader.Add("x-match", "all");
channel.QueueBind(queue: "queue.A",
exchange: "agreements",
routingKey: string.Empty,
arguments: aHeader); //绑定 agreements --> queue.B 使用arguments (format=pdf, type=log, x-match=any)
Dictionary<string, object> bHeader = new Dictionary<string, object>();
bHeader.Add("format", "pdf");
bHeader.Add("type", "log");
bHeader.Add("x-match", "any");
channel.QueueBind(queue: "queue.B",
exchange: "agreements",
routingKey: string.Empty,
arguments: bHeader); //绑定 agreements --> queue.C 使用arguments (format=zip, type=report, x-match=all)
Dictionary<string, object> cHeader = new Dictionary<string, object>();
cHeader.Add("format", "zip");
cHeader.Add("type", "report");
cHeader.Add("x-match", "all");
channel.QueueBind(queue: "queue.C",
exchange: "agreements",
routingKey: string.Empty,
arguments: cHeader); string message1 = "hello world";
var body = Encoding.UTF8.GetBytes(message1);
var properties = channel.CreateBasicProperties();
properties.Persistent = true;
Dictionary<string, object> mHeader1 = new Dictionary<string, object>();
mHeader1.Add("format", "pdf");
mHeader1.Add("type", "report");
properties.Headers = mHeader1;
//此消息路由到 queue.A 和 queue.B
//queue.A 的binding (format=pdf, type=report, x-match=all)
//queue.B 的binding (format = pdf, type = log, x - match = any)
channel.BasicPublish(exchange: "agreements",
routingKey: string.Empty,
basicProperties: properties,
body: body); string message2 = "hello world";
body = Encoding.UTF8.GetBytes(message2);
properties = channel.CreateBasicProperties();
properties.Persistent = true;
Dictionary<string, object> mHeader2 = new Dictionary<string, object>();
mHeader2.Add("type", "log");
properties.Headers = mHeader2;
//x-match 配置queue.B
//queue.B 的binding (format = pdf, type = log, x-match = any)
channel.BasicPublish(exchange: "agreements",
routingKey: string.Empty,
basicProperties: properties,
body: body); string message3= "hello world";
body = Encoding.UTF8.GetBytes(message3);
properties = channel.CreateBasicProperties();
properties.Persistent = true;
Dictionary<string, object> mHeader3 = new Dictionary<string, object>();
mHeader3.Add("format", "zip");
properties.Headers = mHeader3;
//配置失败,不会被路由
channel.BasicPublish(exchange: "agreements",
routingKey: string.Empty,
basicProperties: properties,
body: body); }

总计

以上就是exchange 类型的总结,一般来说direct和topic用来具体的路由消息,如果要用广播的消息一般用fanout的exchange。

header类型用的比较少,但还是知道一点好。

RabbitMQ Exchange类型详解的更多相关文章

  1. RabbitMQ基础知识详解

    什么是MQ? MQ全称为Message Queue, 消息队列(MQ)是一种应用程序对应用程序的通信方法.MQ是消费-生产者模型的一个典型的代表,一端往消息队列中不断写入消息,而另一端则可以读取队列中 ...

  2. RabbitMQ,Apache的ActiveMQ,阿里RocketMQ,Kafka,ZeroMQ,MetaMQ,Redis也可实现消息队列,RabbitMQ的应用场景以及基本原理介绍,RabbitMQ基础知识详解,RabbitMQ布曙

    消息队列及常见消息队列介绍 2017-10-10 09:35操作系统/客户端/人脸识别 一.消息队列(MQ)概述 消息队列(Message Queue),是分布式系统中重要的组件,其通用的使用场景可以 ...

  3. C++11 并发指南六(atomic 类型详解四 C 风格原子操作介绍)

    前面三篇文章<C++11 并发指南六(atomic 类型详解一 atomic_flag 介绍)>.<C++11 并发指南六( <atomic> 类型详解二 std::at ...

  4. C++11 并发指南六( <atomic> 类型详解二 std::atomic )

    C++11 并发指南六(atomic 类型详解一 atomic_flag 介绍)  一文介绍了 C++11 中最简单的原子类型 std::atomic_flag,但是 std::atomic_flag ...

  5. C#进阶系列——WebApi 接口返回值不困惑:返回值类型详解

    前言:已经有一个月没写点什么了,感觉心里空落落的.今天再来篇干货,想要学习Webapi的园友们速速动起来,跟着博主一起来学习吧.之前分享过一篇 C#进阶系列——WebApi接口传参不再困惑:传参详解  ...

  6. C++11 并发指南六(atomic 类型详解三 std::atomic (续))

    C++11 并发指南六( <atomic> 类型详解二 std::atomic ) 介绍了基本的原子类型 std::atomic 的用法,本节我会给大家介绍C++11 标准库中的 std: ...

  7. 服务启动项 Start类型详解

    注册表的服务启动项 Start类型详解 HKLM\SYSTEM\CurrentControlSet\services\ 下的服务项.不论有没有在services.msc服务管理控制台中显示,在注册表中 ...

  8. c# WebApi之接口返回类型详解

    c# WebApi之接口返回类型详解 https://blog.csdn.net/lwpoor123/article/details/78644998

  9. C++之string类型详解

    C++之string类型详解 之所以抛弃char*的字符串而选用C++标准程序库中的string类,是因为他和前者比较起来,不必担心内存是否足够.字符串长度等等,而且作为一个泛型类出现,他集成的操作函 ...

随机推荐

  1. 打印Ibatis最后,SQL声明

    做项目时,满足这一需求.我们希望最终打印出在数据库运行SQL声明,这些都普遍遇到了一些一般性问题.我会去Appfuse,结果这次没有成功.它是有相关的配置,可是好像没实用.我也就没有深查下去.我想这种 ...

  2. c#开发之多国语言解决方案gnu.gettext + poedit

    1.工具简介 1.1.关于i18n i18n其来源是英文单词 internationalization的首末字符i和n,18为中间的字符数是“国际化”的简称. i10n为资源本地化,全称为Locali ...

  3. android:由URL载入中ImageView

    两种方法: 方法一: Bitmap bimage= getBitmapFromURL(bannerpath); image.setImageBitmap(bimage); public static ...

  4. hudson任务配置说明

    hudson任务配置说明 Discard Old Builds:hudson默认保留过去的构建,勾选此选项,则可以设置构建记录的有效期: (帮助:这里控制着您想要在hudson所在的磁盘把构建记录存储 ...

  5. sql简单实用的统计汇总案例参考

    USE [PM]GO/****** 对象:  StoredProcedure [dbo].[LfangSatstics]    脚本日期: 08/24/2013 10:57:48 ******/SET ...

  6. HBase 手动 flush 机制梳理

    对应 HBase 版本0.94.1,对照了开源的版本和工作使用的某发行版 问题:在 HBase shell 里面输入 flush 'table_or_region_name'之后,发生了什么?具体的实 ...

  7. 开发框架(OrchardNoCMS)介绍(一)

    基于ASP.NET MVC的热插拔模块式开发框架(OrchardNoCMS)介绍(一) Orchard CMS是针对CMS开发的,对于很多开发需求来说,内容管理这块儿可能并不需要,而需要它的模块式开发 ...

  8. IIS Express添加MIME映射

    最近在使用fontawesome字体时,在浏览器控制台看到 fontawesome-webfont.woff2?v=4.3.0 无法访问的错误,检查了一下文件确实存在并且路径也对,这就奇怪了! 在控制 ...

  9. CloudFormation

    亚马逊云服务之CloudFormation   亚马逊的Web Service其实包含了一套云服务.云服务主要分为三种: IaaS: Infrastructure as a service,基础设施即 ...

  10. Stream使用教程

    现在,计算机CPU运行速度的快速发展已经远远超过了内存的访问速度.越来越多的程序性能被内存带宽所限制,而不是计算机的CPU运行速率. Stream benchmark是一个简单的合成基准测试程序,衡量 ...