Spring Integration 1.0 正式发布

来源:百度文库 编辑:神马文学网 时间:2024/04/30 03:59:14
Spring Integration 1.0 正式发布    期待了许久,终于等到Spring Integration 1.0的正式发布。在些先祝贺一了Spring Source公司和Spring integration团队。
   下面是本人争取第一时整理的学习笔记(针对最新版本)

先来看一下实际例子:
The Cafe Sample(小卖部订餐例子)

    小卖部有一个订饮料服务,客户可以通过订单来订购所需要饮料。小卖部提供两种咖啡饮料
        LATTE(拿铁咖啡)和MOCHA(摩卡咖啡)。每种又都分冷饮和热饮
    整个流程如下:
        1.有一个下订单模块,用户可以按要求下一个或多个订单。
        2.有一个订单处理模块,处理订单中那些是关于订购饮料的。
        3.有一个饮料订购处理模块,处理拆分订购的具体是那些种类的饮料,把具体需要生产的饮料要求发给生产模块
        4.有一个生产模块,进行生产。
        5.等生成完成后,有一个订单确认模块(Waiter),把订单的生成的饮料输出。

       
    这个例子利用Spring Integration实现了灵活的,可配置化的模式集成了上述这些服务模块。

Spring Integration提供两种模式的工作方式(Annotation和XML)
先来看一下XML方式,进行示例的开发:
配置文件如下:

    xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
    xmlns:beans="http://www.springframework.org/schema/beans"
    xmlns:stream="http://www.springframework.org/schema/integration/stream"
    xsi:schemaLocation="http://www.springframework.org/schema/beans
            http://www.springframework.org/schema/beans/spring-beans-2.5.xsd
            http://www.springframework.org/schema/integration
            http://www.springframework.org/schema/integration/spring-integration-1.0.xsd
            http://www.springframework.org/schema/integration/stream
            http://www.springframework.org/schema/integration/stream/spring-integration-stream-1.0.xsd">

        
    

        
    
    
    

        
    
    
    
        
         
    
        
    
    
                           method="prepareColdDrink" output-channel="preparedDrinks"/>

         
    
        
    
    
                           method="prepareHotDrink" output-channel="preparedDrinks"/>
        
    
    
                    method="prepareDelivery" output-channel="deliveries"/>

        
    

                    class="org.springframework.integration.samples.cafe.xml.OrderSplitter"/>

                    class="org.springframework.integration.samples.cafe.xml.DrinkRouter"/>

    

    


我们来看一下整体服务是怎么启动的
   
    首先我们来看一下CafeDemo这个类,它触发下定单操作
org.springframework.integration.samples.cafe.xml.CafeDemo
 1 public class CafeDemo {
 2 
 3     public static void main(String[] args) {
 4         ////加载Spring 配置文件 "cafeDemo.xml"
 5         AbstractApplicationContext context = null;
 6         if(args.length > 0) {
 7             context = new FileSystemXmlApplicationContext(args);
 8         }
 9         else {
10             context = new ClassPathXmlApplicationContext("cafeDemo.xml", CafeDemo.class);
11         }
12         //取得 Cafe实列
13         Cafe cafe = (Cafe) context.getBean("cafe");
14         //准备 发送100条消息(订单)
15         for (int i = 1; i <= 100; i++) {
16             Order order = new Order(i);
17             // 一杯热饮  参数说明1.饮料类型 2.数量 3.是否是冷饮(true表示冷饮)
18             order.addItem(DrinkType.LATTE, 2, false);
19             // 一杯冷饮  参数说明1.饮料类型 2.数量 3.是否是冷饮(true表示冷饮)
20             order.addItem(DrinkType.MOCHA, 3, true);
21             //下发订单,把消息发给 orders 队列
22             cafe.placeOrder(order);
23         }
24     }
25 
26 }
下面是Cafe接口的源代码
public interface Cafe {

    //定义GateWay, 把消息发送到 orders 队列, Message的payLoad属性,保存 order参数值
    @Gateway(requestChannel="orders")
    void placeOrder(Order order);

}
OrderSplitter 源代码
 1 public class OrderSplitter {
 2 
 3     //接收 从 orders队列接收的 order 消息后,调用 order.getItems方法
 4     //进行订单的分解, 返回的List可会,被拆分为多个消息后(Message.payLoad),发到指定队列
 5     public List split(Order order) {
 6         return order.getItems();
 7     }
 8 
 9 }
10 
OrderSplitter.split把消息拆分后,变成多个消息,发送到drinks队列.由drinkRouter进行消息的接收。
1 public class DrinkRouter {

3     //从 drinks队列的消息后,根据orderItem的属性,选择路由到不同的队列 coldDrinks或hotDrinks
4     public String resolveOrderItemChannel(OrderItem orderItem) {
5         return (orderItem.isIced()) ? "coldDrinks" : "hotDrinks";
6     }

8 }
下面看一下,如果是一杯冷饮,则消息发送到 coldDrinks队列
接收根据配置,由barista Bean的prepareColdDrink方法接收消息后,进行处理
如果是一杯热饮,则消息发送到 hotDrinks队列
接收根据配置,由barista Bean的prepareHotDrink方法接收消息后,进行处理
 1 public class Barista {
 2 
 3     private long hotDrinkDelay = 5000;
 4 
 5     private long coldDrinkDelay = 1000;
 6 
 7     private AtomicInteger hotDrinkCounter = new AtomicInteger();
 8 
 9     private AtomicInteger coldDrinkCounter = new AtomicInteger();
10 
11 
12     public void setHotDrinkDelay(long hotDrinkDelay) {
13         this.hotDrinkDelay = hotDrinkDelay;
14     }
15 
16     public void setColdDrinkDelay(long coldDrinkDelay) {
17         this.coldDrinkDelay = coldDrinkDelay;
18     }
19 
20     //处理热饮订单,并生成Drink冷料
21     public Drink prepareHotDrink(OrderItem orderItem) {
22         try {
23             Thread.sleep(this.hotDrinkDelay);
24             System.out.println(Thread.currentThread().getName()
25                     + " prepared hot drink #" + hotDrinkCounter.incrementAndGet() + " for order #"
26                     + orderItem.getOrder().getNumber() + ": " + orderItem);
27             return new Drink(orderItem.getOrder().getNumber(), orderItem.getDrinkType(), orderItem.isIced(),
28                     orderItem.getShots());
29         } catch (InterruptedException e) {
30             Thread.currentThread().interrupt();
31             return null;
32         }
33     }
34 
35     //处理冷饮订单,并生成Drink冷料
36     public Drink prepareColdDrink(OrderItem orderItem) {
37         try {
38             Thread.sleep(this.coldDrinkDelay);
39             System.out.println(Thread.currentThread().getName()
40                     + " prepared cold drink #" + coldDrinkCounter.incrementAndGet() + " for order #"
41                     + orderItem.getOrder().getNumber() + ": " + orderItem);
42             return new Drink(orderItem.getOrder().getNumber(), orderItem.getDrinkType(), orderItem.isIced(),
43                     orderItem.getShots());
44         } catch (InterruptedException e) {
45             Thread.currentThread().interrupt();
46             return null;
47         }
48     }
49 
50 }
接下来,已经把订单需要生产的饮料已经完成,现在可以交给服务员(waier)交给客人了。
这里使用的aggregate模式,让服务器等待这个订单的所有饮料生产完后的,交给客户.

下面来介绍该应用

                method="prepareDelivery" output-channel="deliveries"/>
最后,完成订单的消息会发到 waiter队列
 1 public class Waiter {
 2 
 3     public Delivery prepareDelivery(List drinks) {
 4         return new Delivery(drinks);
 5     }
 6 
 7 
 8 }
 9 
10 public class Delivery {
11 
12     private static final String SEPARATOR = "-----------------------";
13 
14 
15     private List deliveredDrinks;
16 
17     private int orderNumber;
18 
19 
20     public Delivery(List deliveredDrinks) {
21         assert(deliveredDrinks.size() > 0);
22         this.deliveredDrinks = deliveredDrinks;
23         this.orderNumber = deliveredDrinks.get(0).getOrderNumber();
24     }
25 
26 
27     public int getOrderNumber() {
28         return orderNumber;
29     }
30 
31     public List getDeliveredDrinks() {
32         return deliveredDrinks;
33     }
34 
35     @Override
36     public String toString() {
37         StringBuffer buffer = new StringBuffer(SEPARATOR + "\n");
38         buffer.append("Order #" + getOrderNumber() + "\n");
39         for (Drink drink : getDeliveredDrinks()) {
40             buffer.append(drink);
41             buffer.append("\n");
42         }
43         buffer.append(SEPARATOR + "\n");
44         return buffer.toString();
45     }
46 
47 }
最后我们使用一个 stream channel adaptor把订单生产完成的饮料输出。
        
    

这样整个流程就执行完了,最终我们的饮料产品就按照订单生产出来了。累了吧,喝咖啡提神着呢!!!

spring-integration官网:http://www.springsource.org/spring-integration

关于 Annotation的介绍,将在下篇介绍。

附:xml配置介绍
Service Activator 配置




6                    ref="somePojo" method="someMethod"/>

 触发指定的方法,接收消息队列配置(触发轮循访问的方式)
 1 
 2     
 3         
 4     

 5 

 6 
 7 
 8     
 9         
10     

11 
 
 触发指定的方法,发送消息
1  
2  
3  
4     
5         
6     


Router
消息路由方式

2     
3         
4             
5             
6         

7     

Aggregator 消息合并
 1 
 2 
 3  4     input-channel="inputChannel" 2
 5     output-channel="outputChannel"  3
 6     discard-channel="discardChannel"  4
 7     ref="aggregatorBean" 5
 8     method="add" 6
 9     completion-strategy="completionStrategyBean"  7
10     completion-strategy-method="checkCompleteness" 8
11     timeout="42" 9
12     send-partial-result-on-timeout="true" 10
13     reaper-interval="135" 11
14     tracked-correlation-id-capacity="99" 12
15     send-timeout="86420000" 13 /> 
16 
17 
18 
19 
20 
21 

The id of the aggregator is optional.

The input channel of the aggregator. Required.

The channel where the aggregator will send the aggregation results. Optional (because incoming messages can specify a reply channel themselves).

The channel where the aggregator will send the messages that timed out (if send-partial-results-on-timeout is false). Optional.

A reference to a bean defined in the application context. The bean must implement the aggregation logic as described above. Required.

A method defined on the bean referenced by ref, that implements the message aggregation algorithm. Optional, with restrictions (see above).

A reference to a bean that implements the decision algorithm as to whether a given message group is complete. The bean can be an implementation of the CompletionStrategy interface or a POJO. In the latter case the completion-strategy-method attribute must be defined as well. Optional (by default, the aggregator .

A method defined on the bean referenced by completion-strategy, that implements the completion decision algorithm. Optional, with restrictions (requires completion-strategy to be present).

The timeout for aggregating messages (counted from the arrival of the first message). Optional.

Whether upon the expiration of the timeout, the aggregator shall try to aggregate the already arrived messages. Optional (false by default).

The interval (in milliseconds) at which a reaper task is executed, checking if there are any timed out groups. Optional.

The capacity of the correlation id tracker. Remembers the already processed correlation ids, preventing the formation of new groups for messages that arrive after their group has been already processed (aggregated or discarded). Optional.

The timeout for sending out messages. Optional.


配置消息合并策略
 1 public class PojoCompletionStrategy {
 2 
 3   public boolean checkCompleteness(List numbers) {
 4     int sum = 0;
 5     for (long number: numbers) {
 6       sum += number;
 7     }
 8     return sum >= maxValue;
 9   }
10 }

Good Luck!
Yours Matthew!