Question

我想构建一个线程，每当我将一个项目添加到队列时，它就会侦听一个队列并执行一个作业。

但我对如何开发它并不太了解。我已经尝试过一些来自RxJava2的Flowable示例，但是没有认识到如何做到这一点。

我对Android和Java中的所有示例都是开放的，也许消息处理程序或执行程序将是一个简单的解决方案。可悲的是没有诀窍。特别是RxJava2会很棒。

更新

换句话说，我想在它上面构建一个队列机制，因为长日志显示为分离的时间，并且只要其中两个在近时间调用它们就会混合。

public final class Logcat {

   private static final String TAG = "HOWDY";

   public static void v(String message) {
       Log.v(TAG, message);
   }

   public static void d(String message) {
       Log.d(TAG, message); 
       //TODO I will add a for-loop later for long messages to make sure to show all of them for each method.
   }

   public static void e(Throwable throwable) {
       Log.e(TAG, throwable.getMessage());
   }

   public static void e(String message) {
       Log.e(TAG, message);
   }

   public static void e(ApiError error) {
       Log.e(TAG, error.message);
   }
}

Answer 1

好的，我会这样做..

import java.util.Queue;
import java.util.concurrent.CompletableFuture;
import java.util.concurrent.ConcurrentLinkedQueue;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.TimeUnit;
import java.util.concurrent.atomic.AtomicBoolean;

import io.reactivex.Observable;
import io.reactivex.schedulers.Schedulers;
import io.reactivex.subjects.PublishSubject;

public class DemoRxJava2 {

    public static void testWithQueue() {

        CompletableFuture<String> allDone = new CompletableFuture<>();
        AtomicBoolean submitDone = new AtomicBoolean(Boolean.FALSE);
        final Queue<Long> queue = new ConcurrentLinkedQueue<>();

        Observable.interval(2, TimeUnit.SECONDS)
        .takeWhile(tick -> !queue.isEmpty() || !submitDone.get())
        .flatMap(tick -> {
            return Observable.create(sub -> {
                while (!queue.isEmpty()) {
                    sub.onNext(queue.poll());
                }
                sub.onComplete();
            });
        })
        .subscribeOn(Schedulers.single())
        .doOnSubscribe(dis -> System.out.println("Queue processing active"))
        .doOnComplete(() -> {
            System.out.println("Queue processing done");
            allDone.complete("DONE");
        })
        .subscribe(nextTs -> System.out.printf("[%s] : Processing tx : %d\n", Thread.currentThread().getName(), nextTs));

        Observable.interval(1,TimeUnit.SECONDS)
        .take(10)
        .doOnSubscribe(dis -> System.out.println("Job submitter start"))
        .doOnNext(tick -> {
            long ms = System.currentTimeMillis() / 1000;
            System.out.printf("[%s] : Submitting tx : %d\n", Thread.currentThread().getName(), ms);
            queue.add(ms);
        })
        .doOnComplete(() -> submitDone.set(Boolean.TRUE))
        .blockingSubscribe();

        try {
            allDone.get();
        } catch (InterruptedException | ExecutionException e) {
            e.printStackTrace();
        }
    }

    public static void testWithSubject() {

        CompletableFuture<String> allDone = new CompletableFuture<>();

        PublishSubject<Long> queue = PublishSubject.create();

        queue.observeOn(Schedulers.single())
        .flatMap(tx -> Observable.just(tx).delay(2, TimeUnit.SECONDS))
        .doOnSubscribe(dis -> System.out.println("Queue processing active"))
        .doOnComplete(() -> allDone.complete("DONE"))
        .subscribe(nextTs -> System.out.printf("[%s] : Processing tx : %d\n", Thread.currentThread().getName(), nextTs));

        Observable.interval(1, TimeUnit.SECONDS)
        .take(10)
        .doOnSubscribe(dis -> System.out.println("Job submitter start"))
        .doOnNext(tick -> {
            long ms = System.currentTimeMillis() / 1000;
            System.out.printf("[%s] : Submitting tx : %d\n", Thread.currentThread().getName(), ms);
            queue.onNext(ms);
        })
        .doOnComplete(() -> queue.onComplete())
        .blockingSubscribe();

        //wait until all done
        try {
            allDone.get();
        } catch (InterruptedException | ExecutionException e) {
            e.printStackTrace();
        }
    }

    public static void main(String[] args) {
        testWithQueue();
        testWithSubject();
    }
}

这只是演示了如何使用RxJava处理单独线程中的对象队列，使其适应您的需要

Answer 2

这就是我的方式。

interface ILog {

   String TAG = "HOWDY";

   void display();
}

我的抽象课;

abstract class AbstractLog implements ILog {

   String mLog;

   AbstractLog(@NonNull String log) {
       mLog = log;
   }
}

这是我的具体课程;还有另一个类，如Verbose等。

public class ErrorLog extends AbstractLog {

   ErrorLog(@NonNull String log) {
       super(log);
   }

   ErrorLog(@NonNull Throwable throwable) {
       super(throwable.getMessage());
   }

   ErrorLog(@NonNull ApiError error) {
       super(error.message);
   }

   @Override
   public void display() {
      Log.e(TAG, mLog);
   }
}

这是与./进行互动的班级开发人员。

public final class Logcat {

   private static LogQueue sQueue = new LogQueue();

   public static void v(String log) {
       Message message = new Message();
       message.obj = new VerboseLog(log);

       sQueue.sendMessage(message);
   }

   public static void d(String log) {
       Message message = new Message();
       message.obj = new DebugLog(log);

       sQueue.sendMessage(message);
   }

   public static void e(Throwable throwable) {
       Message message = new Message();
       message.obj = new ErrorLog(throwable);

       sQueue.sendMessage(message);
   }

   public static void e(String log) {
       Message message = new Message();
       message.obj = new ErrorLog(log);

       sQueue.sendMessage(message);
   }

   public static void e(ApiError error) {
       Message message = new Message();
       message.obj = new ErrorLog(error);

       sQueue.sendMessage(message);
   }

   private static class LogQueue extends Handler {

       @Override
       public void handleMessage(Message msg) {
           super.handleMessage(msg);

           ILog log = (ILog) msg.obj;
           log.display();
       }
   }
}

希望对别人有所帮助。

创建一个操作作业的队列

2 个答案: