Android面试题经典之Glide取消加载以及线程池优化

本文首发于公众号“AntDream”，欢迎微信搜索“AntDream”或扫描文章底部二维码关注，和我一起每天进步一点点

Glide通过生命周期取消加载

生命周期回调过程
onStop
--->RequestManager.onStop
-->RequestTracker.pauseRequest
--> SingleRequest.pause
--> SingleRequest.cancel
---> Engine.cancel
---> EngineJob.removeCallback
---> 如果所有的回调监听都移除了
--> DecodeJob.cancel

//EngineJob.class
synchronized void removeCallback(ResourceCallback cb) {
   stateVerifier.throwIfRecycled();
   cbs.remove(cb);
   if (cbs.isEmpty()) {
     cancel();
     boolean isFinishedRunning = hasResource || hasLoadFailed;
     if (isFinishedRunning && pendingCallbacks.get() == 0) {
       release();
     }
   }
}

• 如果任务没有执行，就从队列里移除，取消任务
• 如果任务已经执行，就利用isCancelled标记在停止流程

//DecodeJob.class
public void cancel() {
      isCancelled = true;
      DataFetcherGenerator local = currentGenerator;
      if (local != null) {
          local.cancel();
      }
}

上面最终会调用到具体的获取图片的任务，比如从网络获取图片的HttpUrlFetcher

 //HttpUrlFetcher.class
private volatile boolean isCancelled;
@Override
public void cancel() {
    // TODO: we should consider disconnecting the url connection here, but we can't do so
    // directly because cancel is often called on the main thread.
    isCancelled = true;
}

//内部是通过HttpURLConnection来从网络获取数据
private InputStream loadDataWithRedirects(
      URL url, int redirects, URL lastUrl, Map<String, String> headers) throws HttpException {
    ...

    urlConnection = buildAndConfigureConnection(url, headers);

    try {
      // Connect explicitly to avoid errors in decoders if connection fails.
      urlConnection.connect();
      // Set the stream so that it's closed in cleanup to avoid resource leaks. See #2352.
      stream = urlConnection.getInputStream();
    } catch (IOException e) {
      ...
    }

    if (isCancelled) {
      return null;
    }

    final int statusCode = getHttpStatusCodeOrInvalid(urlConnection);
    if (isHttpOk(statusCode)) {
      return getStreamForSuccessfulRequest(urlConnection);
    } else if (isHttpRedirect(statusCode)) {
      String redirectUrlString = urlConnection.getHeaderField(REDIRECT_HEADER_FIELD);
      ...
      cleanup();
      return loadDataWithRedirects(redirectUrl, redirects + 1, url, headers);
    } else if (statusCode == INVALID_STATUS_CODE) {
      throw new HttpException(statusCode);
    } else {
      ...
    }
}

由于cancel方法一般是在主线程调用的，所以这里采用的是volatile关键字这种方式，在正式获取网络数据时会进行拦截；

• 如果拦截到了，那直接返回null；如果没拦截到，就获取到数据
• 以上最终都会回调到DecodeJob的onDataFetcherReady方法

onDataFetcherReady ---> decodeFromRetrievedData ---> decodeFromData

private void decodeFromRetrievedData() {
     ...
     Resource<R> resource = null;
     try {
       resource = decodeFromData(currentFetcher, currentData, currentDataSource);
     } catch (GlideException e) {
       ...
     }
     if (resource != null) {
       notifyEncodeAndRelease(resource, currentDataSource, isLoadingFromAlternateCacheKey);
     } else {
       //任务被拦截，尝试其他的加载方式
       runGenerators();
     }
}

decodeFromData中会进行判断，如果data为Null就直接返回Null（被拦截时会是null）,这个时候会执行runGenerators方法

runGenerators方法实际上就是加载流程的流转，比如先从文件中加载，文件中没有，就去网络加载，一个个去试这样。当然这里面肯定也有cancel拦截的

private void runGenerators() {
    currentThread = Thread.currentThread();
    startFetchTime = LogTime.getLogTime();
    boolean isStarted = false;
    
    //被cancel拦截就不会尝试其他加载方式，直接任务取消
    
    while (!isCancelled
        && currentGenerator != null
        && !(isStarted = currentGenerator.startNext())) {
      stage = getNextStage(stage);
      currentGenerator = getNextGenerator();

      if (stage == Stage.SOURCE) {
        reschedule();
        return;
      }
    }
    // We've run out of stages and generators, give up.
    if ((stage == Stage.FINISHED || isCancelled) && !isStarted) {
      notifyFailed();
    }
}

所以当加载数据的流程被拦截了，比如网络请求返回null，那到这里这个流程就拦截下来了，直接

notifyFailed任务失败了。

那如果网络加载的时候没有拦截住呢？data就不会为null，就会走notifyEncodeAndRelease方法，最终会一直走到EngineJob的notifyCallbacksOfResult方法，这个方法里面又会有cancel拦截。这样也就是数据加载前加载后都被拦截下来了

Glide中的线程池

自定义ThreadFactory，主要的一个功能是实现限制部分线程网络访问，利用的是严苛模式

@Override
public Thread newThread(@NonNull final Runnable runnable) {
   Thread newThread =
       delegate.newThread(
           new Runnable() {
             @Override
             public void run() {
               if (preventNetworkOperations) {
                 StrictMode.setThreadPolicy(
                     new ThreadPolicy.Builder().detectNetwork().penaltyDeath().build());
               }
               try {
                 runnable.run();
               } catch (Throwable t) {
                 uncaughtThrowableStrategy.handle(t);
               }
             }
           });
   newThread.setName("glide-" + name + "-thread-" + threadNum.getAndIncrement());
   return newThread;
}

线程池主要有这几种：

• AnimationExecutor：加载动画相关，禁止访问网络；如果CPU核心数大于4，就是2个线程，否则是一个线程，核心线程数和最大线程数相同
• diskCacheExecutor：从磁盘加载图片，禁止访问网络；线程数为1，核心线程数和最大线程数相同
• sourceExecutor：可以访问网络，线程数跟CPU核心数有关，不大于4，核心线程数和最大线程数相同
• newUnlimitedSourceExecutor：用于网络请求图片，没有核心线程，只有非核心线程，类似CacheThreadPoll

Glide做了线程池优化，核心线程也会遵循超时策略

public GlideExecutor build() {
  if (TextUtils.isEmpty(name)) {
    throw new IllegalArgumentException(
        "Name must be non-null and non-empty, but given: " + name);
  }
  ThreadPoolExecutor executor =
      new ThreadPoolExecutor(
          corePoolSize,
          maximumPoolSize,
          /*keepAliveTime=*/ threadTimeoutMillis,
          TimeUnit.MILLISECONDS,
          new PriorityBlockingQueue<Runnable>(),
          new DefaultThreadFactory(
              threadFactory, name, uncaughtThrowableStrategy, preventNetworkOperations));

  if (threadTimeoutMillis != NO_THREAD_TIMEOUT) {
    executor.allowCoreThreadTimeOut(true);
  }

  return new GlideExecutor(executor);
}

---

欢迎关注我的公众号AntDream查看更多精彩文章！