How to: 使用 数据流 实现生产者-消费者模式

 

producer把消息发送到消息块，consumer从块读取消息。

安装：

1. Install-Package Microsoft.Tpl.Dataflow
2.  
3. using System.Threading.Tasks.Dataflow;

 

解释：

Produce方法随机生成字节，并Post到ITargetBlock对象；

Consumer方法从ISourceBlock对象读取字节；

可以使用BufferBlock来同时扮演源和目标对象。

Post（）：同步发送消息。

Complete（）：表明当前块（source block）已经没有数据更多的数据了。

Consumer方法使用await和async操作符异步地计算总的字节数。

OutputAvailableAsync（）：从source block收到一个通知（接收到Complete的通知），表明没有更多的数据可用。

 

1. public
   static
   class DataflowProducerConsumer
2. {
3.     // Demonstrates the production end of the producer and consumer pattern.
4.     static
   void Produce(ITargetBlock<byte[]> target)
5.     {
6.         Random rand = new Random();
7.  
8.         // In a loop, fill a buffer with random data and
9.         // post the buffer to the target block.
10.         for (int i = 0; i < 100; i++)
11.         {
12.             // Create an array to hold random byte data.
13.             byte[] buffer = new
    byte[1024];
14.  
15.             // Fill the buffer with random bytes.
16.             rand.NextBytes(buffer);
17.  
18.             // Post the result to the message block.
19.             target.Post(buffer);
20.         }
21.  
22.         // Set the target to the completed state to signal to the consumer
23.         // that no more data will be available.
24.         target.Complete();
25.     }
26.  
27.     // Demonstrates the consumption end of the producer and consumer pattern.
28.     static async Task<int> ConsumeAsync(ISourceBlock<byte[]> source)
29.     {
30.         // Initialize a counter to track the number of bytes that are processed.
31.         int bytesProcessed = 0;
32.  
33.         // Read from the source buffer until the source buffer has no
34.         // available output data.
35.         while (await source.OutputAvailableAsync())
36.         {
37.             byte[] data = source.Receive();
38.  
39.             // Increment the count of bytes received.
40.             bytesProcessed += data.Length;
41.         }
42.  
43.         return bytesProcessed;
44.     }
45.  
46.     static
    void Run(string[] args)
47.     {
48.         // Create a BufferBlock<byte[]> object. This object serves as the
49.         // target block for the producer and the source block for the consumer.
50.         var buffer = new BufferBlock<byte[]>();
51.  
52.         // Start the consumer. The Consume method runs asynchronously.
53.         var consumer = ConsumeAsync(buffer);
54.  
55.         // Post source data to the dataflow block.
56.         Produce(buffer);
57.  
58.         // Wait for the consumer to process all data.
59.         consumer.Wait();
60.  
61.         // Print the count of bytes processed to the console.
62.         Console.WriteLine("Processed {0} bytes.", consumer.Result);
63.     }
64. }

 

参考：https://msdn.microsoft.com/en-us/library/hh228601(v=vs.110).aspx