采用线程池进行异步任务处理

时间 2019-11-18

标签采用线程进行异步任务处理繁體版

原文原文链接

建立线程池

阿里JAVA编码规约，建议采用ThreadPoolExecutor建立线程池。json

private static ExecutorService simpleExecutorService = new ThreadPoolExecutor(
            200,
            300,
            0L,
            TimeUnit.MICROSECONDS,
            new LinkedBlockingDeque<Runnable>(10000),
            new ThreadPoolExecutor.DiscardPolicy());

在同步操做过程当中经过线程池完成异步操做

public void doSomething(final String message) {
        simpleExecutorService.execute(new Runnable() {
            @Override
            public void run() {
                try {
                    Thread.sleep(3000);
                    System.out.println("step 2");
                    System.out.println("message=>" + message);
                } catch (InterruptedException e) {
                    e.printStackTrace();
                }
            }
        });

        System.out.println("step 1");
    }

进行测试

ThreadUtil threadUtil = new ThreadUtil();
threadUtil.doSomething("a thread pool demo");

输出结果

step 1
step 2
message=>a thread pool demo

@Async

在Spring3.x以后框架已经支持采用@Async注解进行异步执行了。多线程

被@Async修饰的方法叫作异步方法，这些异步方法会在新的线程中进行处理，不影响主线程的顺序执行。并发

无返回值执行

@Component
@Slf4j
public class AsyncTask {

    @Async
    public void dealNoReturnTask(){
        log.info("Thread {} deal No Return Task start", Thread.currentThread().getName());
        try {
            Thread.sleep(3000);
        } catch (InterruptedException e) {
            e.printStackTrace();
        }
        log.info("Thread {} deal No Return Task end at {}", Thread.currentThread().getName(), System.currentTimeMillis());
    }
}

进行调用：框架

@SpringBootTest(classes = SpringbootApplication.class)
@RunWith(SpringJUnit4ClassRunner.class)
@Slf4j
public class AsyncTest {

    @Autowired
    private AsyncTask asyncTask;

    @Test
    public void testDealNoReturnTask(){
        asyncTask.dealNoReturnTask();
        try {
            log.info("begin to deal other Task!");
            Thread.sleep(10000);
        } catch (InterruptedException e) {
            e.printStackTrace();
        }
    }

有返回值执行

@Async
    public Future<String> dealHaveReturnTask() {
        try {
            Thread.sleep(3000);
        } catch (InterruptedException e) {
            e.printStackTrace();
        }
        JSONObject jsonObject = new JSONObject();
        jsonObject.put("thread", Thread.currentThread().getName());
        jsonObject.put("time", System.currentTimeMillis());
        return new AsyncResult<String>(jsonObject.toJSONString());
    }

判断任务是否取消：异步

@Test
    public void testDealHaveReturnTask() throws Exception {

        Future<String> future = asyncTask.dealHaveReturnTask();
        log.info("begin to deal other Task!");
        while (true) {
            if(future.isCancelled()){
                log.info("deal async task is Cancelled");
                break;
            }
            if (future.isDone() ) {
                log.info("deal async task is Done");
                log.info("return result is " + future.get());
                break;
            }
            log.info("wait async task to end ...");
            Thread.sleep(1000);
        }
    }

异步执行结果异常处理

咱们能够实现AsyncConfigurer接口，也能够继承AsyncConfigurerSupport类来实现在方法getAsyncExecutor()中建立线程池的时候，必须使用 executor.initialize()，否则在调用时会报线程池未初始化的异常。若是使用threadPoolTaskExecutor()来定义bean，则不须要初始化async

@Configuration
@EnableAsync
@Slf4j
public class AsyncConfig implements AsyncConfigurer {

//    @Bean
//    public ThreadPoolTaskExecutor threadPoolTaskExecutor(){
//        ThreadPoolTaskExecutor executor = new ThreadPoolTaskExecutor();
//        executor.setCorePoolSize(10);
//        executor.setMaxPoolSize(100);
//        executor.setQueueCapacity(100);
//        return executor;
//    }

    @Override
    public Executor getAsyncExecutor() {
        ThreadPoolTaskExecutor executor = new ThreadPoolTaskExecutor();
        executor.setCorePoolSize(10);
        executor.setMaxPoolSize(100);
        executor.setQueueCapacity(100);
        executor.setThreadNamePrefix("AsyncExecutorThread-");
        executor.initialize(); //若是不初始化，致使找到不到执行器
        return executor;
    }
    @Override
    public AsyncUncaughtExceptionHandler getAsyncUncaughtExceptionHandler() {
        return new AsyncExceptionHandler();
    }
}

异步异常处理类：ide

@Slf4j
public class AsyncExceptionHandler implements AsyncUncaughtExceptionHandler {
    @Override
    public void handleUncaughtException(Throwable ex, Method method, Object... params) {
        log.info("Async method: {} has uncaught exception,params:{}", method.getName(), JSON.toJSONString(params));

        if (ex instanceof AsyncException) {
            AsyncException asyncException = (AsyncException) ex;
            log.info("asyncException:{}",asyncException.getErrorMessage());
        }

        log.info("Exception :");
        ex.printStackTrace();
    }
}

@Data
@AllArgsConstructor
public class AsyncException extends Exception {
    private int code;
    private String errorMessage;
}

在无返回值的异步调用中，异步处理抛出异常，AsyncExceptionHandler的handleUncaughtException()会捕获指定异常，原有任务还会继续运行，直到结束。
在有返回值的异步调用中，异步处理抛出异常，会直接抛出异常，异步任务结束，原有处理结束执行。

Future或FutureTask

须要结合Callable测试

public class CallableDemo implements Callable<Integer> {

    private int sum;

    @Override
    public Integer call() throws Exception {
        System.out.println("Callable子线程开始计算啦！");
        Thread.sleep(2000);

        for(int i=0 ;i<5000;i++){
            sum=sum+i;
        }
        System.out.println("Callable子线程计算结束！");
        return sum;
    }
}

Future模式

//建立线程池
        ExecutorService es = Executors.newSingleThreadExecutor();
        //建立Callable对象任务
        CallableDemo calTask = new CallableDemo();

        //提交任务并获取执行结果
        Future<Integer> future = es.submit(calTask);

        //关闭线程池
        es.shutdown();

        try {
            System.out.println("主线程在执行其余任务");

            if (future.get() != null) {
                //输出获取到的结果
                System.out.println("future.get()-->" + future.get());
            } else {
                //输出获取到的结果
                System.out.println("future.get()未获取到结果");
            }

        } catch (Exception e) {
            e.printStackTrace();
        }
        System.out.println("主线程在执行完成");

FutureTask模式

//建立线程池
        ExecutorService es = Executors.newSingleThreadExecutor();
        //建立Callable对象任务
        CallableDemo calTask = new CallableDemo();

        //建立FutureTask
        FutureTask<Integer> future = new FutureTask<>(calTask);
        // future.run();  // 因为FutureTask继承于Runable，因此也能够直接调用run方法执行
        //执行任务
        es.submit(future); // 效果同上面直接调用run方法

        //关闭线程池
        es.shutdown();

        try {
            System.out.println("主线程在执行其余任务");

            if (future.get() != null) {
                //输出获取到的结果
                System.out.println("future.get()-->" + future.get());
            } else {
                //输出获取到的结果
                System.out.println("future.get()未获取到结果");
            }

        } catch (Exception e) {
            e.printStackTrace();
        }
        System.out.println("主线程在执行完成");

归并异步执行结果

public class FutureDemo{
 
     public static void main(String[] args)  {
         Long start = System.currentTimeMillis();
         //开启多线程
         ExecutorService exs = Executors.newFixedThreadPool(10);
         try {
             //结果集
             List<Integer> list = new ArrayList<Integer>();
             List<Future<Integer>> futureList = new ArrayList<Future<Integer>>();
             //1.高速提交10个任务，每一个任务返回一个Future入list
             for(int i=0;i<10;i++){
                 futureList.add(exs.submit(new CallableTask(i+1)));
             }

             Long getResultStart = System.currentTimeMillis();
             System.out.println("结果归集开始时间="+new Date());
             //2.结果归集，遍历futureList,高速轮询（模拟实现了并发）获取future状态成功完成后获取结果，退出当前循环
             for (Future<Integer> future : futureList) {
                  //CPU高速轮询：每一个future都并发轮循，判断完成状态而后获取结果，这一行，是本实现方案的精髓所在。即有10个future在高速轮询，完成一个future的获取结果，就关闭一个轮询
                 while (true) {
                    //获取future成功完成状态，若是想要限制每一个任务的超时时间，取消本行的状态判断+future.get(1000*1, TimeUnit.MILLISECONDS)+catch超时异常使用便可。 
                    if (future.isDone()&& !future.isCancelled()) {
                         Integer i = future.get();//获取结果
                         System.out.println("任务i="+i+"获取完成!"+new Date());
                         list.add(i);
                         break;//当前future获取结果完毕，跳出while
                     } else {
                          //每次轮询休息1毫秒（CPU纳秒级），避免CPU高速轮循耗空CPU---》新手别忘记这个
                         Thread.sleep(1);
                     }
                 }
             }

             System.out.println("list="+list);

             System.out.println("总耗时="+(System.currentTimeMillis()-start)+",取结果归集耗时="+(System.currentTimeMillis()-getResultStart));
            
         } catch (Exception e) {
             e.printStackTrace();
         } finally {
             exs.shutdown();
         }
     }

static class CallableTask implements Callable<Integer>{
         Integer i;
         
         public CallableTask(Integer i) {
             super();
             this.i=i;
         }
 
         @Override
         public Integer call() throws Exception {
             if(i==1){
                 Thread.sleep(3000);//任务1耗时3秒
             }else if(i==5){
                 Thread.sleep(5000);//任务5耗时5秒
             }else{
                 Thread.sleep(1000);//其它任务耗时1秒
             }
             System.out.println("task线程："+Thread.currentThread().getName()+"任务i="+i+",完成！");  
             return i;
         }
     }
 }