标签:
林炳文Evankaka原创作品。转载请注明出处http://blog.csdn.net/evankaka
在上一文章中,笔者介绍了线程池及其内部的原理。今天主要讲的也是和线程相关的内容。一般情况下,使用Runnable接口、Thread实现的线程我们都是无法返回结果的。但是如果对一些场合需要线程返回的结果。就要使用用Callable、Future、FutureTask、CompletionService这几个类。Callable只能在ExecutorService的线程池中跑,但有返回结果,也可以通过返回的Future对象查询执行状态。Future 本身也是一种设计模式,它是用来取得异步任务的结果,
所以来看看它们的源码信息
1、Callable
看看其源码:
public interface Callable<V> { V call() throws Exception; }
它只有一个call方法,并且有一个返回V,是泛型。可以认为这里返回V就是线程返回的结果。
ExecutorService接口:线程池执行调度框架
<T> Future<T> submit(Callable<T> task); <T> Future<T> submit(Runnable task, T result); Future<?> submit(Runnable task);
2、Future
Future是我们最常见的
public interface Future<V> { //试图取消对此任务的执行。如果任务已完成、或已取消,或者由于某些其他原因而无法取消,则此尝试将失败。当调用 cancel 时,如果调用成功,而此任务尚未启动 //,则此任务将永不运行。如果任务已经启动,则 //mayInterruptIfRunning 参数确定是否应该以试图停止任务的方式来中断执行此任务的线程。此方法返回后,对 isDone() 的后续调用将始终返回 true。如果此方法返 //回 true,则对 isCancelled() //的后续调用将始终返回 true。 boolean cancel(boolean mayInterruptIfRunning); //如果在任务正常完成前将其取消,则返回 true。 boolean isCancelled(); //如果任务已完成,则返回 true。 可能由于正常终止、异常或取消而完成,在所有这些情况中,此方法都将返回 true。 boolean isDone(); //等待线程结果返回,会阻塞 V get() throws InterruptedException, ExecutionException; //设置超时时间 V get(long timeout, TimeUnit unit) throws InterruptedException, ExecutionException, TimeoutException; }
从源码看其继承关系如下:
其源码如下:
public class FutureTask<V> implements RunnableFuture<V> { //真正用来执行线程的类 private final Sync sync; //构造方法1,从Callable来创建FutureTask public FutureTask(Callable<V> callable) { if (callable == null) throw new NullPointerException(); sync = new Sync(callable); } //构造方法2,从Runnable来创建FutureTask,V就是线程执行返回结果 public FutureTask(Runnable runnable, V result) { sync = new Sync(Executors.callable(runnable, result)); } //和Futrue一样 public boolean isCancelled() { return sync.innerIsCancelled(); } //和Futrue一样 public boolean isDone() { return sync.innerIsDone(); } //和Futrue一样 public boolean cancel(boolean mayInterruptIfRunning) { return sync.innerCancel(mayInterruptIfRunning); } //和Futrue一样 public V get() throws InterruptedException, ExecutionException { return sync.innerGet(); } //和Futrue一样 public V get(long timeout, TimeUnit unit) throws InterruptedException, ExecutionException, TimeoutException { return sync.innerGet(unit.toNanos(timeout)); } //线程结束后的操作 protected void done() { } //设置结果 protected void set(V v) { sync.innerSet(v); } //设置异常 protected void setException(Throwable t) { sync.innerSetException(t); } //线程执行入口 public void run() { sync.innerRun(); } //重置 protected boolean runAndReset() { return sync.innerRunAndReset(); } //这个类才是真正执行、关闭线程的类 private final class Sync extends AbstractQueuedSynchronizer { private static final long serialVersionUID = -7828117401763700385L; //线程运行状态 private static final int RUNNING = 1; private static final int RAN = 2; private static final int CANCELLED = 4; private final Callable<V> callable; private V result; private Throwable exception; //线程实例 private volatile Thread runner; //构造函数 Sync(Callable<V> callable) { this.callable = callable; } 。。。。 } }FutureTask类是Future 的一个实现,并实现了Runnable,所以可通过Excutor(线程池) 来执行,也可传递给Thread对象执行。如果在主线程中需要执行比较耗时的操作时,但又不想阻塞主线程时,可以把这些作业交给Future对象在后台完成,当主线程将来需要时,就可以通过Future对象获得后台作业的计算结果或者执行状态。 Executor框架利用FutureTask来完成异步任务,并可以用来进行任何潜在的耗时的计算。一般FutureTask多用于耗时的计算,主线程可以在完成自己的任务后,再去获取结果。FutureTask类既可以使用new Thread(Runnable r)放到一个新线程中跑,也可以使用ExecutorService.submit(Runnable r)放到线程池中跑,而且两种方式都可以获取返回结果,但实质是一样的,即如果要有返回结果那么构造函数一定要注入一个Callable对象。
1、Future实例
package com.func.axc.futuretask; import java.util.Random; import java.util.concurrent.Callable; import java.util.concurrent.ExecutionException; import java.util.concurrent.ExecutorService; import java.util.concurrent.Executors; import java.util.concurrent.Future; /** * 功能概要: * * @author linbingwen * @since 2016年6月8日 */ public class FutureTest { /** * @author linbingwen * @since 2016年6月8日 * @param args */ public static void main(String[] args) { System.out.println("main Thread begin at:"+ System.nanoTime()); ExecutorService executor = Executors.newCachedThreadPool(); HandleCallable task1 = new HandleCallable("1"); HandleCallable task2 = new HandleCallable("2"); HandleCallable task3 = new HandleCallable("3"); Future<Integer> result1 = executor.submit(task1); Future<Integer> result2 = executor.submit(task2); Future<Integer> result3 = executor.submit(task3); executor.shutdown(); try { Thread.sleep(1000); } catch (InterruptedException e1) { e1.printStackTrace(); } try { System.out.println("task1运行结果:"+result1.get()); System.out.println("task2运行结果:"+result2.get()); System.out.println("task3运行结果:"+result3.get()); } catch (InterruptedException e) { e.printStackTrace(); } catch (ExecutionException e) { e.printStackTrace(); } System.out.println("main Thread finish at:"+ System.nanoTime()); } } class HandleCallable implements Callable<Integer>{ private String name; public HandleCallable(String name) { this.name = name; } @Override public Integer call() throws Exception { System.out.println("task"+ name + "开始进行计算"); Thread.sleep(3000); int sum = new Random().nextInt(300); int result = 0; for (int i = 0; i < sum; i++) result += i; return result; } }执行结果:
2、FutureTask
方法一、直接通过New Thread来启动线程
package com.func.axc.futuretask; import java.util.Random; import java.util.concurrent.Callable; import java.util.concurrent.ExecutionException; import java.util.concurrent.FutureTask; import org.springframework.scheduling.config.Task; /** * 功能概要: * * @author linbingwen * @since 2016年5月31日 */ public class FutrueTaskTest { public static void main(String[] args) { //采用直接启动线程的方法 System.out.println("main Thread begin at:"+ System.nanoTime()); MyTask task1 = new MyTask("1"); FutureTask<Integer> result1 = new FutureTask<Integer>(task1); Thread thread1 = new Thread(result1); thread1.start(); MyTask task2 = new MyTask("2"); FutureTask<Integer> result2 = new FutureTask<Integer>(task2); Thread thread2 = new Thread(result2); thread2.start(); try { Thread.sleep(1000); } catch (InterruptedException e1) { e1.printStackTrace(); } try { System.out.println("task1返回结果:" + result1.get()); System.out.println("task2返回结果:" + result2.get()); } catch (InterruptedException e) { e.printStackTrace(); } catch (ExecutionException e) { e.printStackTrace(); } System.out.println("main Thread finish at:"+ System.nanoTime()); } } class MyTask implements Callable<Integer> { private String name; public MyTask(String name) { this.name = name; } @Override public Integer call() throws Exception { System.out.println("task"+ name + "开始进行计算"); Thread.sleep(3000); int sum = new Random().nextInt(300); int result = 0; for (int i = 0; i < sum; i++) result += i; return result; } }
方法二、通过线程池来启动线程
package com.func.axc.futuretask; import java.util.Random; import java.util.concurrent.Callable; import java.util.concurrent.ExecutionException; import java.util.concurrent.ExecutorService; import java.util.concurrent.Executors; import java.util.concurrent.Future; /** * 功能概要: * * @author linbingwen * @since 2016年5月31日 */ public class FutrueTaskTest2 { public static void main(String[] args) { System.out.println("main Thread begin at:"+ System.nanoTime()); ExecutorService executor = Executors.newCachedThreadPool(); MyTask2 task1 = new MyTask2("1"); MyTask2 task2 = new MyTask2("2"); Future<Integer> result1 = executor.submit(task1); Future<Integer> result2 = executor.submit(task2); executor.shutdown(); try { Thread.sleep(1000); } catch (InterruptedException e1) { e1.printStackTrace(); } try { System.out.println("task1返回结果:" + result1.get()); System.out.println("task2返回结果:" + result2.get()); } catch (InterruptedException e) { e.printStackTrace(); } catch (ExecutionException e) { e.printStackTrace(); } System.out.println("main Thread finish at:"+ System.nanoTime()); } } class MyTask2 implements Callable<Integer> { private String name; public MyTask2(String name) { this.name = name; } @Override public Integer call() throws Exception { System.out.println("task"+ name + "开始进行计算"); Thread.sleep(3000); int sum = new Random().nextInt(300); int result = 0; for (int i = 0; i < sum; i++) result += i; return result; } }执行结果:
这个光看其单词,就可以猜到它应该是一个线程执行完成后相关的服务,没错。它就是一个将线程池执行结果放入到一个Blockqueueing的类。那么它和Future或FutureTask有什么不同呢?其实在上面的例子中,笔者用的实例可能不太好。如果在线程池中我们使用Future或FutureTask来取得返回结果,比如。我们开了100条线程。但是这些线程的执行时间是未知的。但是我们又需要返回结果。每执行一条线程就根据结果做一次相应的操作。如果是Future或FutureTask。我们只能通过一个循环,不断的遍历线程池里的线程。取得其执行状态。然后再来取结果。这样效率就太低了,有可能发生一条线程执行完毕了,但我们不能立刻知道它处理完成了。还得通过一个循环来判断。基本上面的这种问题,所以出了CompletionService。
CompletionService原理不是很难,它就是将一组线程的执行结果放入一个BlockQueueing当中。这里线程的执行结果放入到Blockqueue的顺序只和这个线程的执行时间有关。和它们的启动顺序无关。并且你无需自己在去写很多判断哪个线程是否执行完成,它里面会去帮你处理。
首先看看其源码:
package java.util.concurrent; public interface CompletionService<V> { //提交线程任务 Future<V> submit(Callable<V> task); //提交线程任务 Future<V> submit(Runnable task, V result); //阻塞等待 Future<V> take() throws InterruptedException; //非阻塞等待 Future<V> poll(); //带时间的非阻塞等待 Future<V> poll(long timeout, TimeUnit unit) throws InterruptedException; }上面只是一个接口类,其实现类如下:
package java.util.concurrent; public class ExecutorCompletionService<V> implements CompletionService<V> { private final Executor executor;//线程池类 private final AbstractExecutorService aes; private final BlockingQueue<Future<V>> completionQueue;//存放线程执行结果的阻塞队列 //内部封装的一个用来执线程的FutureTask private class QueueingFuture extends FutureTask<Void> { QueueingFuture(RunnableFuture<V> task) { super(task, null); this.task = task; } protected void done() { completionQueue.add(task); }//线程执行完成后调用此函数将结果放入阻塞队列 private final Future<V> task; } private RunnableFuture<V> newTaskFor(Callable<V> task) { if (aes == null) return new FutureTask<V>(task); else return aes.newTaskFor(task); } private RunnableFuture<V> newTaskFor(Runnable task, V result) { if (aes == null) return new FutureTask<V>(task, result); else return aes.newTaskFor(task, result); } //构造函数,这里一般传入一个线程池对象executor的实现类 public ExecutorCompletionService(Executor executor) { if (executor == null) throw new NullPointerException(); this.executor = executor; this.aes = (executor instanceof AbstractExecutorService) ? (AbstractExecutorService) executor : null; this.completionQueue = new LinkedBlockingQueue<Future<V>>();//默认的是链表阻塞队列 } //构造函数,可以自己设定阻塞队列 public ExecutorCompletionService(Executor executor, BlockingQueue<Future<V>> completionQueue) { if (executor == null || completionQueue == null) throw new NullPointerException(); this.executor = executor; this.aes = (executor instanceof AbstractExecutorService) ? (AbstractExecutorService) executor : null; this.completionQueue = completionQueue; } //提交线程任务,其实最终还是executor去提交 public Future<V> submit(Callable<V> task) { if (task == null) throw new NullPointerException(); RunnableFuture<V> f = newTaskFor(task); executor.execute(new QueueingFuture(f)); return f; } //提交线程任务,其实最终还是executor去提交 public Future<V> submit(Runnable task, V result) { if (task == null) throw new NullPointerException(); RunnableFuture<V> f = newTaskFor(task, result); executor.execute(new QueueingFuture(f)); return f; } public Future<V> take() throws InterruptedException { return completionQueue.take(); } public Future<V> poll() { return completionQueue.poll(); } public Future<V> poll(long timeout, TimeUnit unit) throws InterruptedException { return completionQueue.poll(timeout, unit); } }从源码中可以知道。最终还是线程还是提交到Executor当中去运行,所以构造函数中需要Executor参数来实例化。而每次有线程执行完成后往阻塞队列添加一个Future。
这是上面的RunnableFuture,这是每次往线程池是放入的线程。
public interface RunnableFuture<V> extends Runnable, Future<V> { void run(); }
1、与Future的区别使用:
自定义一个Callable
class HandleFuture<Integer> implements Callable<Integer> { private Integer num; public HandleFuture(Integer num) { this.num = num; } @Override public Integer call() throws Exception { Thread.sleep(3*100); System.out.println(Thread.currentThread().getName()); return num; } }首先是Futuer
public static void FutureTest() throws InterruptedException, ExecutionException { System.out.println("main Thread begin:"); ExecutorService executor = Executors.newCachedThreadPool(); List<Future<Integer>> result = new ArrayList<Future<Integer>>(); for (int i = 0;i<10;i++) { Future<Integer> submit = executor.submit(new HandleFuture(i)); result.add(submit); } executor.shutdown(); for (int i = 0;i<10;i++) {//一个一个等待返回结果 System.out.println("返回结果:"+result.get(i).get()); } System.out.println("main Thread end:"); }执行结果:
接下来换成CompletionService来做:
public static void CompleTest() throws InterruptedException, ExecutionException { System.out.println("main Thread begin:"); ExecutorService executor = Executors.newCachedThreadPool(); // 构建完成服务 CompletionService<Integer> completionService = new ExecutorCompletionService<Integer>(executor); for (int i = 0;i<10;i++) { completionService.submit(new HandleFuture(i)); } for (int i = 0;i<10;i++) {//一个一个等待返回结果 System.out.println("返回结果:"+completionService.take().get()); } System.out.println("main Thread end:"); }
Java并发编程与技术内幕:Callable、Future、FutureTask、CompletionService
标签:
原文地址:http://blog.csdn.net/evankaka/article/details/51610635