上一篇文章,只是简单地介绍了用guava的ListenableFuture实现异步,那如果需要多重回调呢?
方法 | 描述 |
---|---|
transform | 加一个回调函数 |
allAsList | 返回一个ListenableFuture ,该ListenableFuture 返回的result是一个List,List中的值是每个ListenableFuture的返回值,假如传入的其中之一fails或者cancel,这个Future fails 或者canceled |
successAsList | 返回一个ListenableFuture ,该Future的结果包含所有成功的Future,按照原来的顺序,当其中之一Failed或者cancel,则用null替代 |
链式回调:
package cn.dubby.listeningfuture;
import com.google.common.util.concurrent.*;
import java.util.concurrent.Callable;
import java.util.concurrent.CountDownLatch;
import java.util.concurrent.Executors;
/**
* Created by dubby on 16/4/29.
*/
public class ListeningFutureDemo {
public static void main(String[] args) {
final CountDownLatch latch = new CountDownLatch(1);
final ListeningExecutorService service = MoreExecutors.listeningDecorator(Executors.newFixedThreadPool(10));
ListenableFuture<String> explosion = service.submit(new Callable<String>() {
public String call() throws Exception {
System.out.println("任务线程正在执行...");
try {
Thread.sleep(1000);
} catch (Exception e) {
e.printStackTrace();
}
return "任务线程的结果";
}
});
ListenableFuture<String> first = Futures.transform(explosion, new AsyncFunction<String, String>() {
public ListenableFuture<String> apply(final String input) throws Exception {
ListenableFuture<String> temp = service.submit(new Callable<String>() {
public String call() throws Exception {
System.out.println("第1个回调线程正在执行...");
try {
Thread.sleep(1000);
} catch (Exception e) {
e.printStackTrace();
}
return input + " & 第1个回调线程的结果 ";
}
});
return temp;
}
}, service);
ListenableFuture<String> second = Futures.transform(first, new AsyncFunction<String, String>() {
public ListenableFuture<String> apply(final String input) throws Exception {
ListenableFuture<String> temp = service.submit(new Callable<String>() {
public String call() throws Exception {
System.out.println("第2个回调线程正在执行...");
try {
Thread.sleep(1000);
} catch (Exception e) {
e.printStackTrace();
}
return input + " & 第2个回调线程的结果 ";
}
});
return temp;
}
}, service);
ListenableFuture<String> third = Futures.transform(second, new AsyncFunction<String, String>() {
public ListenableFuture<String> apply(final String input) throws Exception {
ListenableFuture<String> temp = service.submit(new Callable<String>() {
public String call() throws Exception {
System.out.println("第3个回调线程正在执行...");
try {
Thread.sleep(1000);
} catch (Exception e) {
e.printStackTrace();
}
return input + " & 第3个回调线程的结果 ";
}
});
return temp;
}
}, service);
ListenableFuture<String> forth = Futures.transform(third, new AsyncFunction<String, String>() {
public ListenableFuture<String> apply(final String input) throws Exception {
ListenableFuture<String> temp = service.submit(new Callable<String>() {
public String call() throws Exception {
System.out.println("第4个回调线程正在执行...");
try {
Thread.sleep(1000);
} catch (Exception e) {
e.printStackTrace();
}
return input + " & 第4个回调线程的结果 ";
}
});
return temp;
}
}, service);
Futures.addCallback(forth, new FutureCallback<String>() {
public void onSuccess(String result) {
latch.countDown();
System.out.println("结果: " + result);
}
public void onFailure(Throwable t) {
System.out.println(t.getMessage());
}
});
try {
latch.await();
} catch (InterruptedException e) {
e.printStackTrace();
}
service.shutdown();
}
}
一个Future,一共进行了两次处理,同样的,可以添加更多的回调函数。简化了操作。