C# 使用.NET的SocketAsyncEventArgs实现高效能多并发TCPSocket通信
简介：
SocketAsyncEventArgs是一个套接字操作得类，主要作用是实现socket消息的异步接收和发送，跟Socket的BeginSend和BeginReceive方法异步处理没有多大区别，它的优势在于完成端口的实现来处理大数据的并发情况。
- BufferManager类, 管理传输流的大小
- SocketEventPool类: 管理SocketAsyncEventArgs的一个应用池. 有效地重复使用.
- AsyncUserToken类: 这个可以根据自己的实际情况来定义.主要作用就是存储客户端的信息.
- SocketManager类: 核心,实现Socket监听,收发信息等操作.
- 额外功能 1.自动检测无效连接并断开 2.自动释放资源
BufferManager类
using System;
using System.Collections.Generic;
using System.Linq;
using System.Net.Sockets;
using System.Text;
namespace Plates.Service
{
class BufferManager
{
int m_numBytes; // the total number of bytes controlled by the buffer pool
byte[] m_buffer; // the underlying byte array maintained by the Buffer Manager
Stack m_freeIndexPool; //
int m_currentIndex;
int m_bufferSize;
public BufferManager(int totalBytes, int bufferSize)
{
m_numBytes = totalBytes;
m_currentIndex = 0;
m_bufferSize = bufferSize;
m_freeIndexPool = new Stack();
}
// Allocates buffer space used by the buffer pool
public void InitBuffer()
{
// create one big large buffer and divide that
// out to each SocketAsyncEventArg object
m_buffer = new byte[m_numBytes];
}
// Assigns a buffer from the buffer pool to the
// specified SocketAsyncEventArgs object
//
// true if the buffer was successfully set, else false
public bool SetBuffer(SocketAsyncEventArgs args)
{
if (m_freeIndexPool.Count > 0)
{
args.SetBuffer(m_buffer, m_freeIndexPool.Pop(), m_bufferSize);
}
else
{
if ((m_numBytes - m_bufferSize) < m_currentIndex)
{
return false;
}
args.SetBuffer(m_buffer, m_currentIndex, m_bufferSize);
m_currentIndex += m_bufferSize;
}
return true;
}
// Removes the buffer from a SocketAsyncEventArg object.
// This frees the buffer back to the buffer pool
public void FreeBuffer(SocketAsyncEventArgs args)
{
m_freeIndexPool.Push(args.Offset);
args.SetBuffer(null, 0, 0);
}
}
}
SocketEventPool类:
using System;
using System.Collections.Generic;
using System.Linq;
using System.Net.Sockets;
using System.Text;
namespace Plates.Service
{
class SocketEventPool
{
Stack m_pool;
public SocketEventPool(int capacity)
{
m_pool = new Stack(capacity);
}
public void Push(SocketAsyncEventArgs item)
{
if (item == null) { throw new ArgumentNullException("Items added to a SocketAsyncEventArgsPool cannot be null"); }
lock (m_pool)
{
m_pool.Push(item);
}
}
// Removes a SocketAsyncEventArgs instance from the pool
// and returns the object removed from the pool
public SocketAsyncEventArgs Pop()
{
lock (m_pool)
{
return m_pool.Pop();
}
}
// The number of SocketAsyncEventArgs instances in the pool
public int Count
{
get { return m_pool.Count; }
}
public void Clear()
{
m_pool.Clear();
}
}
} AsyncUserToken类
using System;
using System.Collections;
using System.Collections.Generic;
using System.Linq;
using System.Net;
using System.Net.Sockets;
using System.Text;
namespace Plates.Service
{
class AsyncUserToken
{
///
/// 客户端IP地址
///
public IPAddress IPAddress { get; set; }
///
/// 远程地址
///
public EndPoint Remote { get; set; }
///
/// 通信SOKET
///
public Socket Socket { get; set; }
///
/// 连接时间
///
public DateTime ConnectTime { get; set; }
///
/// 所属用户信息
///
public UserInfoModel UserInfo { get; set; }
///
/// 数据缓存区
///
public List Buffer { get; set; }
public AsyncUserToken()
{
this.Buffer = new List();
}
}
} SocketManager类
using Plates.Common;
using System;
using System.Collections;
using System.Collections.Generic;
using System.Linq;
using System.Net;
using System.Net.Sockets;
using System.Text;
using System.Threading;
namespace Plates.Service
{
class SocketManager
{
private int m_maxConnectNum; //最大连接数
private int m_revBufferSize; //最大接收字节数
BufferManager m_bufferManager;
const int opsToAlloc = 2;
Socket listenSocket; //监听Socket
SocketEventPool m_pool;
int m_clientCount; //连接的客户端数量
Semaphore m_maxNumberAcceptedClients;
List m_clients; //客户端列表
#region 定义委托
///
/// 客户端连接数量变化时触发
///
/// 当前增加客户的个数(用户退出时为负数,增加时为正数,一般为1)
/// 增加用户的信息
public delegate void OnClientNumberChange(int num, AsyncUserToken token);
///
/// 接收到客户端的数据
///
/// 客户端
/// 客户端数据
public delegate void OnReceiveData(AsyncUserToken token, byte[] buff);
#endregion
#region 定义事件
///
/// 客户端连接数量变化事件
///
public event OnClientNumberChange ClientNumberChange;
///
/// 接收到客户端的数据事件
///
public event OnReceiveData ReceiveClientData;
#endregion
#region 定义属性
///
/// 获取客户端列表
///
public List ClientList { get { return m_clients; } }
#endregion
///
/// 构造函数
///
/// 最大连接数
/// 缓存区大小
public SocketManager(int numConnections, int receiveBufferSize)
{
m_clientCount = 0;
m_maxConnectNum = numConnections;
m_revBufferSize = receiveBufferSize;
// allocate buffers such that the maximum number of sockets can have one outstanding read and
//write posted to the socket simultaneously
m_bufferManager = new BufferManager(receiveBufferSize * numConnections * opsToAlloc, receiveBufferSize);
m_pool = new SocketEventPool(numConnections);
m_maxNumberAcceptedClients = new Semaphore(numConnections, numConnections);
}
///
/// 初始化
///
public void Init()
{
// Allocates one large byte buffer which all I/O operations use a piece of. This gaurds
// against memory fragmentation
m_bufferManager.InitBuffer();
m_clients = new List();
// preallocate pool of SocketAsyncEventArgs objects
SocketAsyncEventArgs readWriteEventArg;
for (int i = 0; i < m_maxConnectNum; i++)
{
readWriteEventArg = new SocketAsyncEventArgs();
readWriteEventArg.Completed += new EventHandler(IO_Completed);
readWriteEventArg.UserToken = new AsyncUserToken();
// assign a byte buffer from the buffer pool to the SocketAsyncEventArg object
m_bufferManager.SetBuffer(readWriteEventArg);
// add SocketAsyncEventArg to the pool
m_pool.Push(readWriteEventArg);
}
}
///
/// 启动服务
///
///
public bool Start(IPEndPoint localEndPoint)
{
try
{
m_clients.Clear();
listenSocket = new Socket(localEndPoint.AddressFamily, SocketType.Stream, ProtocolType.Tcp);
listenSocket.Bind(localEndPoint);
// start the server with a listen backlog of 100 connections
listenSocket.Listen(m_maxConnectNum);
// post accepts on the listening socket
StartAccept(null);
return true;
}
catch (Exception)
{
return false;
}
}
///
/// 停止服务
///
public void Stop()
{
foreach (AsyncUserToken token in m_clients)
{
try
{
token.Socket.Shutdown(SocketShutdown.Both);
}
catch (Exception) { }
}
try
{
listenSocket.Shutdown(SocketShutdown.Both);
}
catch (Exception) { }
listenSocket.Close();
int c_count = m_clients.Count;
lock (m_clients) { m_clients.Clear(); }
if (ClientNumberChange != null)
ClientNumberChange(-c_count, null);
}
public void CloseClient(AsyncUserToken token)
{
try
{
token.Socket.Shutdown(SocketShutdown.Both);
}
catch (Exception) { }
}
// Begins an operation to accept a connection request from the client
//
// The context object to use when issuing
// the accept operation on the server's listening socket
public void StartAccept(SocketAsyncEventArgs acceptEventArg)
{
if (acceptEventArg == null)
{
acceptEventArg = new SocketAsyncEventArgs();
acceptEventArg.Completed += new EventHandler(AcceptEventArg_Completed);
}
else
{
// socket must be cleared since the context object is being reused
acceptEventArg.AcceptSocket = null;
}
m_maxNumberAcceptedClients.WaitOne();
if (!listenSocket.AcceptAsync(acceptEventArg))
{
ProcessAccept(acceptEventArg);
}
}
// This method is the callback method associated with Socket.AcceptAsync
// operations and is invoked when an accept operation is complete
//
void AcceptEventArg_Completed(object sender, SocketAsyncEventArgs e)
{
ProcessAccept(e);
}
private void ProcessAccept(SocketAsyncEventArgs e)
{
try
{
Interlocked.Increment(ref m_clientCount);
// Get the socket for the accepted client connection and put it into the
//ReadEventArg object user token
SocketAsyncEventArgs readEventArgs = m_pool.Pop();
AsyncUserToken userToken = (AsyncUserToken)readEventArgs.UserToken;
userToken.Socket = e.AcceptSocket;
userToken.ConnectTime = DateTime.Now;
userToken.Remote = e.AcceptSocket.RemoteEndPoint;
userToken.IPAddress = ((IPEndPoint)(e.AcceptSocket.RemoteEndPoint)).Address;
lock (m_clients) { m_clients.Add(userToken); }
if (ClientNumberChange != null)
ClientNumberChange(1, userToken);
if (!e.AcceptSocket.ReceiveAsync(readEventArgs))
{
ProcessReceive(readEventArgs);
}
}
catch (Exception me)
{
RuncomLib.Log.LogUtils.Info(me.Message + "\r\n" + me.StackTrace);
}
// Accept the next connection request
if (e.SocketError == SocketError.OperationAborted) return;
StartAccept(e);
}
void IO_Completed(object sender, SocketAsyncEventArgs e)
{
// determine which type of operation just completed and call the associated handler
switch (e.LastOperation)
{
case SocketAsyncOperation.Receive:
ProcessReceive(e);
break;
case SocketAsyncOperation.Send:
ProcessSend(e);
break;
default:
throw new ArgumentException("The last operation completed on the socket was not a receive or send");
}
}
// This method is invoked when an asynchronous receive operation completes.
// If the remote host closed the connection, then the socket is closed.
// If data was received then the data is echoed back to the client.
//
private void ProcessReceive(SocketAsyncEventArgs e)
{
try
{
// check if the remote host closed the connection
AsyncUserToken token = (AsyncUserToken)e.UserToken;
if (e.BytesTransferred > 0 && e.SocketError == SocketError.Success)
{
//读取数据
byte[] data = new byte[e.BytesTransferred];
Array.Copy(e.Buffer, e.Offset, data, 0, e.BytesTransferred);
lock (token.Buffer)
{
token.Buffer.AddRange(data);
}
//注意:你一定会问,这里为什么要用do-while循环?
//如果当客户发送大数据流的时候,e.BytesTransferred的大小就会比客户端发送过来的要小,
//需要分多次接收.所以收到包的时候,先判断包头的大小.够一个完整的包再处理.
//如果客户短时间内发送多个小数据包时, 服务器可能会一次性把他们全收了.
//这样如果没有一个循环来控制,那么只会处理第一个包,
//剩下的包全部留在token.Buffer中了,只有等下一个数据包过来后,才会放出一个来.
do
{
//判断包的长度
byte[] lenBytes = token.Buffer.GetRange(0, 4).ToArray();
int packageLen = BitConverter.ToInt32(lenBytes, 0);
if (packageLen > token.Buffer.Count - 4)
{ //长度不够时,退出循环,让程序继续接收
break;
}
//包够长时,则提取出来,交给后面的程序去处理
byte[] rev = token.Buffer.GetRange(4, packageLen).ToArray();
//从数据池中移除这组数据
lock (token.Buffer)
{
token.Buffer.RemoveRange(0, packageLen + 4);
}
//将数据包交给后台处理,这里你也可以新开个线程来处理.加快速度.
if(ReceiveClientData != null)
ReceiveClientData(token, rev);
//这里API处理完后,并没有返回结果,当然结果是要返回的,却不是在这里, 这里的代码只管接收.
//若要返回结果,可在API处理中调用此类对象的SendMessage方法,统一打包发送.不要被微软的示例给迷惑了.
} while (token.Buffer.Count > 4);
//继续接收. 为什么要这么写,请看Socket.ReceiveAsync方法的说明
if (!token.Socket.ReceiveAsync(e))
this.ProcessReceive(e);
}
else
{
CloseClientSocket(e);
}
}
catch (Exception xe)
{
RuncomLib.Log.LogUtils.Info(xe.Message + "\r\n" + xe.StackTrace);
}
}
// This method is invoked when an asynchronous send operation completes.
// The method issues another receive on the socket to read any additional
// data sent from the client
//
//
private void ProcessSend(SocketAsyncEventArgs e)
{
if (e.SocketError == SocketError.Success)
{
// done echoing data back to the client
AsyncUserToken token = (AsyncUserToken)e.UserToken;
// read the next block of data send from the client
bool willRaiseEvent = token.Socket.ReceiveAsync(e);
if (!willRaiseEvent)
{
ProcessReceive(e);
}
}
else
{
CloseClientSocket(e);
}
}
//关闭客户端
private void CloseClientSocket(SocketAsyncEventArgs e)
{
AsyncUserToken token = e.UserToken as AsyncUserToken;
lock (m_clients) { m_clients.Remove(token); }
//如果有事件,则调用事件,发送客户端数量变化通知
if (ClientNumberChange != null)
ClientNumberChange(-1, token);
// close the socket associated with the client
try
{
token.Socket.Shutdown(SocketShutdown.Send);
}
catch (Exception) { }
token.Socket.Close();
// decrement the counter keeping track of the total number of clients connected to the server
Interlocked.Decrement(ref m_clientCount);
m_maxNumberAcceptedClients.Release();
// Free the SocketAsyncEventArg so they can be reused by another client
e.UserToken = new AsyncUserToken();
m_pool.Push(e);
}
///
/// 对数据进行打包,然后再发送
///
///
///
/// 使用方法：
SocketManager m_socket = new SocketManager(200, 1024);
m_socket.Init();
m_socket.Start(new IPEndPoint(IPAddress.Any, 13909));
//m_socket.Stop();