1. Server-Return-Service-Klasse:
import java.io.IOException; import java.io.InputStream; import java.io.OutputStream; import java.net.Socket; import java.util.logging.Level; import java.util.logging.Logger; public class EchoProtocol implements Runnable { private static final int BUFSIZE = 32; // Size (in bytes) of I/O buffer private Socket clientSocket; // Socket connect to client private Logger logger; // Server logger public EchoProtocol(Socket clientSocket, Logger logger) { this.clientSocket = clientSocket; this.logger = logger; } public static void handleEchoClient(Socket clientSocket, Logger logger) { try { // Get the input and output I/O streams from socket InputStream in = clientSocket.getInputStream(); OutputStream out = clientSocket.getOutputStream(); int recvMsgSize; // Size of received message int totalBytesEchoed = 0; // Bytes received from client byte[] echoBuffer = new byte[BUFSIZE]; // Receive Buffer // Receive until client closes connection, indicated by -1 while ((recvMsgSize = in.read(echoBuffer)) != -1) { out.write(echoBuffer, 0, recvMsgSize); totalBytesEchoed += recvMsgSize; } logger.info("Client " + clientSocket.getRemoteSocketAddress() + ", echoed " + totalBytesEchoed + " bytes."); } catch (IOException ex) { logger.log(Level.WARNING, "Exception in echo protocol", ex); } finally { try { clientSocket.close(); } catch (IOException e) { } } } public void run() { handleEchoClient(this.clientSocket, this.logger); } }
2. Tcp-Server, der für jede Client-Anfrage einen neuen Thread startet:
import java.io.IOException; import java.net.ServerSocket; import java.net.Socket; import java.util.logging.Logger; public class TCPEchoServerThread { public static void main(String[] args) throws IOException { // Create a server socket to accept client connection requests ServerSocket servSock = new ServerSocket(5500); Logger logger = Logger.getLogger("practical"); // Run forever, accepting and spawning a thread for each connection while (true) { Socket clntSock = servSock.accept(); // Block waiting for connection // Spawn thread to handle new connection Thread thread = new Thread(new EchoProtocol(clntSock, logger)); thread.start(); logger.info("Created and started Thread " + thread.getName()); } /* NOT REACHED */ } }
3. Festgelegter Tcp-Server mit der Anzahl von Threads:
import java.io.IOException; import java.net.ServerSocket; import java.net.Socket; import java.util.logging.Level; import java.util.logging.Logger; public class TCPEchoServerPool { public static void main(String[] args) throws IOException { int threadPoolSize = 3; // Fixed ThreadPoolSize final ServerSocket servSock = new ServerSocket(5500); final Logger logger = Logger.getLogger("practical"); // Spawn a fixed number of threads to service clients for (int i = 0; i < threadPoolSize; i++) { Thread thread = new Thread() { public void run() { while (true) { try { Socket clntSock = servSock.accept(); // Wait for a connection EchoProtocol.handleEchoClient(clntSock, logger); // Handle it } catch (IOException ex) { logger.log(Level.WARNING, "Client accept failed", ex); } } } }; thread.start(); logger.info("Created and started Thread = " + thread.getName()); } } }
4. Thread-Pool verwenden (bei der Verwendung von Spring-Threads gelten die Konzepte Warteschlange, maximale Anzahl von Threads, minimale Anzahl von Threads und Timeout)
1. Thread-Pool-Tool Klasse:
import java.util.concurrent.*; /** * 任务执行者 * * @author Watson Xu * @since 1.0.0 <p>2013-6-8 上午10:33:09</p> */ public class ThreadPoolTaskExecutor { private ThreadPoolTaskExecutor() { } private static ExecutorService executor = Executors.newCachedThreadPool(new ThreadFactory() { int count; /* 执行器会在需要自行任务而线程池中没有线程的时候来调用该程序。对于callable类型的调用通过封装以后转化为runnable */ public Thread newThread(Runnable r) { count++; Thread invokeThread = new Thread(r); invokeThread.setName("Courser Thread-" + count); invokeThread.setDaemon(false);// //???????????? return invokeThread; } }); public static void invoke(Runnable task, TimeUnit unit, long timeout) throws TimeoutException, RuntimeException { invoke(task, null, unit, timeout); } public static <T> T invoke(Runnable task, T result, TimeUnit unit, long timeout) throws TimeoutException, RuntimeException { Future<T> future = executor.submit(task, result); T t = null; try { t = future.get(timeout, unit); } catch (TimeoutException e) { throw new TimeoutException("Thread invoke timeout ..."); } catch (Exception e) { throw new RuntimeException(e); } return t; } public static <T> T invoke(Callable<T> task, TimeUnit unit, long timeout) throws TimeoutException, RuntimeException { // 这里将任务提交给执行器,任务已经启动,这里是异步的。 Future<T> future = executor.submit(task); // System.out.println("Task aready in thread"); T t = null; try { /* * 这里的操作是确认任务是否已经完成,有了这个操作以后 * 1)对invoke()的调用线程变成了等待任务完成状态 * 2)主线程可以接收子线程的处理结果 */ t = future.get(timeout, unit); } catch (TimeoutException e) { throw new TimeoutException("Thread invoke timeout ..."); } catch (Exception e) { throw new RuntimeException(e); } return t; } }
2. Skalierbarer TCP-Server:
import java.io.IOException; import java.net.ServerSocket; import java.net.Socket; import java.util.concurrent.TimeUnit; import java.util.logging.Logger; import demo.callable.ThreadPoolTaskExecutor; public class TCPEchoServerExecutor { public static void main(String[] args) throws IOException { // Create a server socket to accept client connection requests ServerSocket servSock = new ServerSocket(5500); Logger logger = Logger.getLogger("practical"); // Run forever, accepting and spawning threads to service each connection while (true) { Socket clntSock = servSock.accept(); // Block waiting for connection //executorService.submit(new EchoProtocol(clntSock, logger)); try { ThreadPoolTaskExecutor.invoke(new EchoProtocol(clntSock, logger), TimeUnit.SECONDS, 3); } catch (Exception e) { } //service.execute(new TimelimitEchoProtocol(clntSock, logger)); } /* NOT REACHED */ } }
Weitere Java-Socket-Programmierbeispiele – Artikel zum TCP-Server-Thread-Pool finden Sie auf der chinesischen PHP-Website!