Future is introduced in JDK 1.5 by Doug Lea to represent "the result of an asynchronous computation".
Java 5中引入了Future接口，用于代表异步计算的结果
ExecutorService exe = Executors.newCachedThreadPool();
Runnable runnable = () -> System.out.println("Runnable");
Callable<String> callable = () -> "Callable";
exe.submit(runnable);
Future<String> futureResult = exe.submit(callable);
String result = futureResult.get();
System.out.println(result);
exe.shutdown();
Similar to runnable, callable can also be submitted to executor service. Differently, callable is "A task that returns a result and may throw an exception."
The call to get a future result lead current thread to wait "if necessary for the computation to complete, and then retrieves its result." Below is one implementation of get method.
同时还引入了Callable接口来代表有返回值的任务，把 callable提交给线程池便可以获得一个future。
当我们想要获取callable的结果，或者说想要拿到future的值，可能会需要适时的等待。参照ForkjoinPool中对Future get的实现，基本可以明白，当前线程会一直等到future所代表的异步任务执行完毕。
public final V get() throws InterruptedException, ExecutionException {
int s = (Thread.currentThread() instanceof ForkJoinWorkerThread) ?
doJoin() : externalInterruptibleAwaitDone();
Throwable ex;
if ((s &= DONE_MASK) == CANCELLED)
throw new CancellationException();
if (s == EXCEPTIONAL && (ex = getThrowableException()) != null)
throw new ExecutionException(ex);
return getRawResult();
}
In JDK 1.8, another future, CompletableFuture, is introduced by the same author with epoch-making significance. Completable future is "A {@link Future} that may be explicitly completed (setting its value and status), and may be used as a {@link CompletionStage}, supporting dependent functions and actions that trigger upon its completion."
Java 8中，Doug Lea大神又引入了CompletableFuture 这一大作，既拥有future的特性，同时又可以根据任务的完成状态来触发后续的动作，以一种可同步可异步的调用链方式来执行更为复杂的任务。
Now it comes to the interesting part. Hey, what if we want to convert the legacy future into completable future so as to make full use of the new features?
Well, let's see a snippet from stack overflow http://*.com/questions/23301598/transform-java-future-into-a-completablefuture
现在问题来了，假如我们想要把老代码中的 future 转换成 completable future，怎么做比较好呢？如果第三方库本身拥有完成和失败的回调，整合completable future 还相对容易，就像下面这个从stack overflow搬来的例子一样：
AsynchronousFileChannel open = AsynchronousFileChannel.open(Paths.get("/some/file"));
// ...
CompletableFuture<ByteBuffer> completableFuture = new CompletableFuture<ByteBuffer>();
open.read(buffer, position, null, new CompletionHandler<Integer, Void>() {
@Override
public void completed(Integer result, Void attachment) {
completableFuture.complete(buffer);
}
@Override
public void failed(Throwable exc, Void attachment) {
completableFuture.completeExceptionally(exc);
}
});
completableFuture.thenApply(...)
If the 3rd party library support complete and fail callback, it's convinient to integrate with completable future. But usually, future inteferface won't contain these methods.
Hmmm, feels so close... and unreachable. Wait, how about ListenableFuture introduced by Guava ? It can add callbacks for a future.
然而。。。通常的future接口并没有这些方法╮(￣▽￣")╭ 别急，俗话说，车到山前必有路，让我们回想回想Guava所引入的ListenableFuture, 它可以给future 增加完成和失败的回调。
ListeningExecutorService executor1 = MoreExecutors.listeningDecorator(Executors.newFixedThreadPool(10));
ListenableFuture<Explosion> explosion = executor1.submit(new Callable<Explosion>() {
public Explosion call() {
return pushBigRedButton();
}
});
Futures.addCallback(explosion, new FutureCallback<Explosion>() {
// we want this handler to run immediately after we push the big red button!
public void onSuccess(Explosion explosion) {
walkAwayFrom(explosion);
}
public void onFailure(Throwable thrown) {
battleArchNemesis(); // escaped the explosion!
}
}, executor2);
Pattern matched, problem solved, and everyone is happy now.
Do care about executor1 and executor2, configure the thread pools properly will boost app performance, cheers~
看到熟悉的模型，问题迎刃而解。
注意下上面例子中的executor1 和 executor2，实际运用的时候合理的调节线程池的参数可以更好的提升应用性能。