扣丁书屋

Android源码进阶之Glide缓存机制原理详解

前言

Android缓存机制:如果没有缓存,在大量的网络请求从远程获取图片时会造成网络流量的浪费,加载速度较慢,用户体验不好;

今天我们就来聊聊Glide的缓存机制

一、Glide中缓存概念简述

Glide将它分成了两个模块,一个是内存缓存,一个是硬盘缓存;

1、内存缓存

  • 内存缓存又分为两级,一级是LruCache缓存,一级是弱引用缓存
  • 内存缓存的作用:防止应用重复将图片数据读取到内存当中。
  • LruCache缓存:不在使用中的图片使用LruCache来进行缓存。
  • 弱引用缓存:把正在使用中的图片使用弱引用来进行缓存,这样的目的保护正在使用的资源不会被LruCache算法回收。

2、硬盘缓存

硬盘缓存的作用:防止应用重复从网络或其他地方重复下载和读取数据;

3、图片请求步骤

开始一个新的图片请求之前检查以下多级的缓存:

  • 内存缓存:该图片是否最近被加载过并仍存在于内存中?即LruCache缓存;
  • 活动资源:现在是否有另一个 View 正在展示这张图片?也就是弱引用缓存;
  • 资源类型:该图片是否之前曾被解码、转换并写入过磁盘缓存?
  • 数据来源:构建这个图片的资源是否之前曾被写入过文件缓存?
  • 前两步检查图片是否在内存中,如果是则直接返回图片。后两步则检查图片是否在磁盘上,以便快速但异步地返回图片;
  • 如果四个步骤都未能找到图片,则Glide会返回到原始资源以取回数据(原始文件,Uri, Url等);
  • 图片存的顺序是:弱引用、内存、磁盘;
  • 图片取的顺序是:内存、弱引用、磁盘;

4、Glide中Bitmap复用机制

  • Bitmap复用机制:将已经不需要使用的数据空间重新拿来使用,减少内存抖动(指在短时间内有大量的对象被创建或者被回收的现象);
  • BitmapFactory.Options.inMutable是Glide能够复用Bitmap的基石,是BitmapFactory提供的一个参数,表示该Bitmap是可变的,支持复用的。BitmapFactory.Options中提供了两个属性:inMutable、inBitmap。当进行Bitmap复用时,需要设置inMutable为true,inBitmap设置被复用的已经存在的Bitmap。Bitmap复用池使用LRU算法实现;

二、缓存源码流程

memory cache和disk cache在Glide创建的时候也被创建了,Glide创建的代码在GlideBuilder.build(Context)方法

@NonNull
Glide build(@NonNull Context context) {
  if (memoryCache == null) {
    memoryCache = new LruResourceCache(memorySizeCalculator.getMemoryCacheSize());
  }
  if (diskCacheFactory == null) {
    diskCacheFactory = new InternalCacheDiskCacheFactory(context);
  }
  if (engine == null) {
    engine =
        new Engine(
            memoryCache,
            diskCacheFactory,
            ...);
  }
  return new Glide(
      ...
      memoryCache,
      ...);
}

1、内存缓存-memoryCache

通过代码可以看到 memoryCache 被放入 Engine 和 Glide 实例中。在Engine中利用memoryCache进行存取操作,Glide 实例中的memoryCache是用来在内存紧张的时候,通知memoryCache释放内存。Glide实现了ComponentCallbacks2接口,在Glide创建完成后,通过applicationContext.registerComponentCallbacks(glide)似的 Glide 实例可以监听内存紧张的信号。

// Glide
@Override
public void onTrimMemory(int level) {
  trimMemory(level);
}
public void trimMemory(int level) {
  // Engine asserts this anyway when removing resources, fail faster and consistently
  Util.assertMainThread();
  // memory cache needs to be trimmed before bitmap pool to trim re-pooled Bitmaps too. See #687.
  memoryCache.trimMemory(level);
  bitmapPool.trimMemory(level);
  arrayPool.trimMemory(level);
}

memoryCache是一个使用LRU(least recently used)算法实现的内存缓存类LruResourceCache,继承至LruCache类,并实现了MemoryCache接口。LruCache定义了LRU算法实现相关的操作,而MemoryCache定义的是内存缓存相关的操作。

LruCache 的实现是利用了 LinkedHashMap 的这种数据结构的一个特性( accessOrder=true 基于访问顺序 )再加上对 LinkedHashMap 的数据操作上锁实现的缓存策略。

当调用 put()方法时,就会在集合中添加元素,并调用

trimToSize()判断缓存是否已满,如果满了就用 LinkedHashMap 的迭代器删除队尾元素,即近期最少访问的元素。

当调用 get()方法访问缓存对象时,就会调用 LinkedHashMap 的 get()方法获得对应集合元素,同时会更新该元素到队头

2、磁盘缓存

diskCacheFactory是创建DiskCache的Factory,DiskCache接口定义

public interface DiskCache {
  interface Factory {
    /** 250 MB of cache. */
    int DEFAULT_DISK_CACHE_SIZE = 250 * 1024 * 1024;
    String DEFAULT_DISK_CACHE_DIR = "image_manager_disk_cache";
    @Nullable
    DiskCache build();
  }
  interface Writer {
    boolean write(@NonNull File file);
  }
  @Nullable
  File get(Key key);
  void put(Key key, Writer writer);
  @SuppressWarnings("unused")
  void delete(Key key);
  void clear();
}

接着再来看下DiskCache.Factory的默认实现:InternalCacheDiskCacheFactory

public final class InternalCacheDiskCacheFactory extends DiskLruCacheFactory {
  public InternalCacheDiskCacheFactory(Context context) {
    this(context, DiskCache.Factory.DEFAULT_DISK_CACHE_DIR,
        DiskCache.Factory.DEFAULT_DISK_CACHE_SIZE);
  }
  public InternalCacheDiskCacheFactory(Context context, long diskCacheSize) {
    this(context, DiskCache.Factory.DEFAULT_DISK_CACHE_DIR, diskCacheSize);
  }
  public InternalCacheDiskCacheFactory(final Context context, final String diskCacheName,
                                       long diskCacheSize) {
    super(new CacheDirectoryGetter() {
      @Override
      public File getCacheDirectory() {
        File cacheDirectory = context.getCacheDir();
        if (cacheDirectory == null) {
          return null;
        }
        if (diskCacheName != null) {
          return new File(cacheDirectory, diskCacheName);
        }
        return cacheDirectory;
      }
    }, diskCacheSize);
  }
}

由以上代码可以看出:默认会创建一个250M的缓存目录,其路径为/data/data/{package}/cache/image_manager_disk_cache/

继续看其父类DiskLruCacheFactory的代码

public class DiskLruCacheFactory implements DiskCache.Factory {
  private final long diskCacheSize;
  private final CacheDirectoryGetter cacheDirectoryGetter;
  public interface CacheDirectoryGetter {
    File getCacheDirectory();
  }
  ...
  public DiskLruCacheFactory(CacheDirectoryGetter cacheDirectoryGetter, long diskCacheSize) {
    this.diskCacheSize = diskCacheSize;
    this.cacheDirectoryGetter = cacheDirectoryGetter;
  }
  @Override
  public DiskCache build() {
    File cacheDir = cacheDirectoryGetter.getCacheDirectory();
    if (cacheDir == null) {
      return null;
    }
    if (!cacheDir.mkdirs() && (!cacheDir.exists() || !cacheDir.isDirectory())) {
      return null;
    }
    return DiskLruCacheWrapper.create(cacheDir, diskCacheSize);
  }
}

DiskLruCacheFactory.build()方法会返回一个DiskLruCacheWrapper类的实例,看下DiskLruCacheWrapper的实现

public class DiskLruCacheWrapper implements DiskCache {
  private static final String TAG = "DiskLruCacheWrapper";
  private static final int APP_VERSION = 1;
  private static final int VALUE_COUNT = 1;
  private static DiskLruCacheWrapper wrapper;
  private final SafeKeyGenerator safeKeyGenerator;
  private final File directory;
  private final long maxSize;
  private final DiskCacheWriteLocker writeLocker = new DiskCacheWriteLocker();
  private DiskLruCache diskLruCache;
  @SuppressWarnings("deprecation")
  public static DiskCache create(File directory, long maxSize) {
    return new DiskLruCacheWrapper(directory, maxSize);
  }
  @Deprecated
  @SuppressWarnings({"WeakerAccess", "DeprecatedIsStillUsed"})
  protected DiskLruCacheWrapper(File directory, long maxSize) {
    this.directory = directory;
    this.maxSize = maxSize;
    this.safeKeyGenerator = new SafeKeyGenerator();
  }
  private synchronized DiskLruCache getDiskCache() throws IOException {
    if (diskLruCache == null) {
      diskLruCache = DiskLruCache.open(directory, APP_VERSION, VALUE_COUNT, maxSize);
    }
    return diskLruCache;
  }
  @Override
  public File get(Key key) {
    String safeKey = safeKeyGenerator.getSafeKey(key);
    File result = null;
    try {
      final DiskLruCache.Value value = getDiskCache().get(safeKey);
      if (value != null) {
        result = value.getFile(0);
      }
    } catch (IOException e) {
      ...
    }
    return result;
  }
  @Override
  public void put(Key key, Writer writer) {
    String safeKey = safeKeyGenerator.getSafeKey(key);
    writeLocker.acquire(safeKey);
    try {
      try {
        DiskLruCache diskCache = getDiskCache();
        Value current = diskCache.get(safeKey);
        ...
        DiskLruCache.Editor editor = diskCache.edit(safeKey);
        ...
        try {
          File file = editor.getFile(0);
          if (writer.write(file)) {
            editor.commit();
          }
        } finally {
          editor.abortUnlessCommitted();
        }
      } catch (IOException e) {
        ...
      }
    } finally {
      writeLocker.release(safeKey);
    }
  }
  ...
}

里面包装了一个DiskLruCache,该类主要是为DiskLruCache提供了一个根据Key生成safeKey的SafeKeyGenerator以及写锁DiskCacheWriteLocker。

回到GlideBuilder.build(Context)中,diskCacheFactory会被传进Engine中,在Engine的构造方法中会被包装成为一个LazyDiskCacheProvider,在被需要的时候调用getDiskCache()方法,这样就会调用factory的build()方法返回一个DiskCache。代码如下:

private static class LazyDiskCacheProvider implements DecodeJob.DiskCacheProvider {
    private final DiskCache.Factory factory;
    private volatile DiskCache diskCache;
    LazyDiskCacheProvider(DiskCache.Factory factory) {
      this.factory = factory;
    }
    ...
    @Override
    public DiskCache getDiskCache() {
      if (diskCache == null) {
        synchronized (this) {
          if (diskCache == null) {
            diskCache = factory.build();
          }
          if (diskCache == null) {
            diskCache = new DiskCacheAdapter();
          }
        }
      }
      return diskCache;
    }
  }

LazyDiskCacheProvider会在Engine后面的初始化流程中作为入参传到DecodeJobFactory的构造器。在DecodeJobFactory创建DecodeJob时也会作为入参会传进去,DecodeJob中会以全局变量保存此LazyDiskCacheProvider,在资源加载完毕并展示后,会进行缓存的存储。同时,DecodeJob也会在DecodeHelper初始化时,将此DiskCacheProvider设置进去,供ResourceCacheGenerator、DataCacheGenerator读取缓存,供SourceGenerator写入缓存

3、 ActiveResources

ActiveResources在Engine的构造器中被创建,在ActiveResources的构造器中会启动一个后台优先级级别(THREAD_PRIORITY_BACKGROUND)的线程,在该线程中会调用cleanReferenceQueue()方法一直循环清除ReferenceQueue中的将要被GC的Resource。

final class ActiveResources {
  private final boolean isActiveResourceRetentionAllowed;
  private final Executor monitorClearedResourcesExecutor;
  @VisibleForTesting
  final Map<Key, ResourceWeakReference> activeEngineResources = new HashMap<>();
  private final ReferenceQueue<EngineResource<?>> resourceReferenceQueue = new ReferenceQueue<>();
  private volatile boolean isShutdown;
  ActiveResources(boolean isActiveResourceRetentionAllowed) {
    this(
        isActiveResourceRetentionAllowed,
        java.util.concurrent.Executors.newSingleThreadExecutor(
            new ThreadFactory() {
              @Override
              public Thread newThread(@NonNull final Runnable r) {
                return new Thread(
                    new Runnable() {
                      @Override
                      public void run() {
                        Process.setThreadPriority(Process.THREAD_PRIORITY_BACKGROUND);
                        r.run();
                      }
                    },
                    "glide-active-resources");
              }
            }));
  }
  @VisibleForTesting
  ActiveResources(
      boolean isActiveResourceRetentionAllowed, Executor monitorClearedResourcesExecutor) {
    this.isActiveResourceRetentionAllowed = isActiveResourceRetentionAllowed;
    this.monitorClearedResourcesExecutor = monitorClearedResourcesExecutor;
    monitorClearedResourcesExecutor.execute(
        new Runnable() {
          @Override
          public void run() {
            cleanReferenceQueue();
          }
        });
  }
  @SuppressWarnings("WeakerAccess")
  @Synthetic void cleanReferenceQueue() {
    while (!isShutdown) {
      try {
        ResourceWeakReference ref = (ResourceWeakReference) resourceReferenceQueue.remove();
        cleanupActiveReference(ref);
        // This section for testing only.
        DequeuedResourceCallback current = cb;
        if (current != null) {
          current.onResourceDequeued();
        }
        // End for testing only.
      } catch (InterruptedException e) {
        Thread.currentThread().interrupt();
      }
    }
  }
}

先来看看ActiveResources的activate方法(保存)、deactivate方法(删除)的方法

synchronized void activate(Key key, EngineResource<?> resource) {
    ResourceWeakReference toPut =
        new ResourceWeakReference(
            key, resource, resourceReferenceQueue, isActiveResourceRetentionAllowed);
    ResourceWeakReference removed = activeEngineResources.put(key, toPut);
    if (removed != null) {
      removed.reset();
    }
  }
  synchronized void deactivate(Key key) {
    ResourceWeakReference removed = activeEngineResources.remove(key);
    if (removed != null) {
      removed.reset();
    }
  }

activate方法会将参数封装成为一个ResourceWeakReference,然后放入map中,如果对应的key之前有值,那么调用之前值的reset方法进行清除。deactivate方法先在map中移除,然后调用resource的reset方法进行清除。ResourceWeakReference继承WeakReference,内部只是保存了Resource的一些属性。

static final class ResourceWeakReference extends WeakReference<EngineResource<?>> {
  @SuppressWarnings("WeakerAccess") @Synthetic final Key key;
  @SuppressWarnings("WeakerAccess") @Synthetic final boolean isCacheable;
  @Nullable @SuppressWarnings("WeakerAccess") @Synthetic Resource<?> resource;
  @Synthetic
  @SuppressWarnings("WeakerAccess")
  ResourceWeakReference(
      @NonNull Key key,
      @NonNull EngineResource<?> referent,
      @NonNull ReferenceQueue<? super EngineResource<?>> queue,
      boolean isActiveResourceRetentionAllowed) {
    super(referent, queue);
    this.key = Preconditions.checkNotNull(key);
    this.resource =
        referent.isCacheable() && isActiveResourceRetentionAllowed
            ? Preconditions.checkNotNull(referent.getResource()) : null;
    isCacheable = referent.isCacheable();
  }
}

构造方法中调用了super(referent, queue),这样做可以让将要被GC的对象放入到ReferenceQueue中。而ActiveResources.cleanReferenceQueue()方法会一直尝试从queue中获取将要被GC的resource,然后调用cleanupActiveReference方法将resource从activeEngineResources中移除。cleanupActiveReference源码如下:

void cleanupActiveReference(@NonNull ResourceWeakReference ref) {
    synchronized (listener) {
      synchronized (this) {
        // 移除active资源
        activeEngineResources.remove(ref.key);
        if (!ref.isCacheable || ref.resource == null) {
          return;
        }
        // 构造新的 Resource
        EngineResource<?> newResource =
            new EngineResource<>(ref.resource, /*isCacheable=*/ true, /*isRecyclable=*/ false);
        newResource.setResourceListener(ref.key, listener);
        // 回调Engine的onResourceReleased方法
        // 这会导致此资源从active变成memory cache状态
        listener.onResourceReleased(ref.key, newResource);
      }
    }
  }

Engine实现了EngineResource.ResourceListener,此处的listener就是Engine,最终会回调Engine.onResourceReleased

@Override
  public synchronized void onResourceReleased(Key cacheKey, EngineResource<?> resource) {
    activeResources.deactivate(cacheKey);
    if (resource.isCacheable()) {
      cache.put(cacheKey, resource);
    } else {
      resourceRecycler.recycle(resource);
    }
  }

如果资源可以被缓存,则缓存到 memory cache,否则对资源进行回收

4、磁盘缓存读取

我们分析下缓存的存取代码。我们看下

public synchronized <R> LoadStatus load(...) {
  EngineKey key = keyFactory.buildKey(model, signature, width, height, transformations,
      resourceClass, transcodeClass, options);
  EngineResource<?> active = loadFromActiveResources(key, isMemoryCacheable);
  if (active != null) {
    cb.onResourceReady(active, DataSource.MEMORY_CACHE);
    return null;
  }
  EngineResource<?> cached = loadFromCache(key, isMemoryCacheable);
  if (cached != null) {
    cb.onResourceReady(cached, DataSource.MEMORY_CACHE);
    return null;
  }
  EngineJob<?> current = jobs.get(key, onlyRetrieveFromCache);
  if (current != null) {
    current.addCallback(cb, callbackExecutor);
    return new LoadStatus(cb, current);
  }
  EngineJob<R> engineJob =
      engineJobFactory.build(...);
  DecodeJob<R> decodeJob =
      decodeJobFactory.build(...);
  jobs.put(key, engineJob);
  engineJob.addCallback(cb, callbackExecutor);
  engineJob.start(decodeJob);
  return new LoadStatus(cb, engineJob);
}

缓存需要根据EngineKey去存取,先看下EngineKey的构造方法


EngineKey(
      Object model,
      Key signature,
      int width
      int height,
      Map<Class<?>, Transformation<?>> transformations,
      Class<?> resourceClass,
      Class<?> transcodeClass,
      Options options)
  • model:load方法传的参数;
  • signature:BaseRequestOptions的成员变量,默认会是EmptySignature.obtain()
  • 在加载本地resource资源时会变成ApplicationVersionSignature.obtain(context);
  • width、height:如果没有指定override(int size),那么将得到view的size;
  • transformations:默认会基于ImageView的scaleType设置对应的四个Transformation;
  • 如果指定了transform,那么就基于该值进行设置;
  • resourceClass:解码后的资源,如果没有asBitmap、asGif,一般会是Object;
  • transcodeClass:最终要转换成的数据类型,根据as方法确定,加载本地res或者网络URL,都会调用asDrawable,所以为Drawable
  • options:如果没有设置过transform,此处会根据ImageView的scaleType默认指定一个option;
  • 所以,在多次加载同一个model的过程中,只要上述任何一个参数有改变,都不会认为是同一个key;

回到Engine.load方法,从缓存加载成功后的回调cb.onResourceReady(cached, DataSource.MEMORY_CACHE);可以看到:active状态的资源和memory cache状态的资源都是DataSource.MEMORY_CACHE,并且加载的资源都是 EngineResource 对象,该对象内部采用了引用计数去判断资源是否被释放,如果引用计数为0,那么会调用listener.onResourceReleased(key, this)方法通知外界此资源已经释放了。这里的listener是ResourceListener类型的接口,只有一个onResourceReleased(Key key, EngineResource<?> resource)方法,Engine实现了该接口,此处的listener就是Engine。在Engine.onResourceReleased方法中会判断资源是否可缓存,可缓存则将此资源放入memory cache中,否则回收掉该资源,代码如下:

public synchronized void onResourceReleased(Key cacheKey, EngineResource<?> resource) {
    // 从activeResources中移除
    activeResources.deactivate(cacheKey);
    if (resource.isCacheable()) {
      // 存入 MemoryCache
      cache.put(cacheKey, resource);
    } else {
      resourceRecycler.recycle(resource);
    }
  }

继续回到Engine.load方法,先来看下active资源获取的方法

@Nullable
  private EngineResource<?> loadFromActiveResources(Key key, boolean isMemoryCacheable) {
    // 设置skipMemoryCache(true),则isMemoryCacheable为false,跳过ActiveResources
    if (!isMemoryCacheable) {
      return null;
    }
    EngineResource<?> active = activeResources.get(key);
    if (active != null) {
      // 命中缓存,引用计数+1
      active.acquire();
    }
    return active;
  }

继续分析cached资源获取的方法,如果从active资源中没有获取到缓存,则继续从内存缓存中查找

 private EngineResource<?> loadFromCache(Key key, boolean isMemoryCacheable) {
    // 设置skipMemoryCache(true),则isMemoryCacheable为false,跳过ActiveResources
    if (!isMemoryCacheable) {
      return null;
    }
    EngineResource<?> cached = getEngineResourceFromCache(key);
    if (cached != null) {
      // 命中缓存,引用计数+1
      cached.acquire();
      // 将此资源从memoryCache中移到activeResources中
      activeResources.activate(key, cached);
    }
    return cached;
  }

如果从memoryCache中获取到资源则将此资源从memoryCache中移到activeResources中。第一次加载的时候activeResources和memoryCache中都没有缓存的,后面继续通过DecodeJob和EngineJob去加载资源。DecoceJob实现了Runnable接口,然后会被EngineJob.start方法提交到对应的线程池中去执行。在DecoceJob的run方法中,会依次从ResourceCacheGenerator和DataCacheGenerator中去取缓存数据,当这两者都取不到的情况下,会交给SourceGenerator加载网络图片或者本地资源。resource资源和data资源都是磁盘缓存中的资源。

先看下 ResourceCacheGenerator.startNext

@Override
  public boolean startNext() {
    // list里面只有一个GlideUrl对象
    List<Key> sourceIds = helper.getCacheKeys();
    if (sourceIds.isEmpty()) {
      return false;
    }
    // 获得了三个可以到达的registeredResourceClasses
    // GifDrawable、Bitmap、BitmapDrawable
    List<Class<?>> resourceClasses = helper.getRegisteredResourceClasses();
    if (resourceClasses.isEmpty()) {
      if (File.class.equals(helper.getTranscodeClass())) {
        return false;
      }
      throw new IllegalStateException(
         "Failed to find any load path from " + helper.getModelClass() + " to "
             + helper.getTranscodeClass());
    }
    // 遍历sourceIds中的每一个key、resourceClasses中每一个class,以及其他的一些值组成key
    // 尝试在磁盘缓存中以key找到缓存文件
    while (modelLoaders == null || !hasNextModelLoader()) {
      resourceClassIndex++;
      if (resourceClassIndex >= resourceClasses.size()) {
        sourceIdIndex++;
        if (sourceIdIndex >= sourceIds.size()) {
          return false;
        }
        resourceClassIndex = 0;
      }
      Key sourceId = sourceIds.get(sourceIdIndex);
      Class<?> resourceClass = resourceClasses.get(resourceClassIndex);
      Transformation<?> transformation = helper.getTransformation(resourceClass);
      // PMD.AvoidInstantiatingObjectsInLoops Each iteration is comparatively expensive anyway,
      // we only run until the first one succeeds, the loop runs for only a limited
      // number of iterations on the order of 10-20 in the worst case.
      // 构造key
      currentKey =
          new ResourceCacheKey(// NOPMD AvoidInstantiatingObjectsInLoops
              helper.getArrayPool(),
              sourceId,
              helper.getSignature(),
              helper.getWidth(),
              helper.getHeight(),
              transformation,
              resourceClass,
              helper.getOptions());
      // 查找缓存文件
      cacheFile = helper.getDiskCache().get(currentKey);
      // 如果找到了缓存文件,循环条件则会为false,退出循环
      if (cacheFile != null) {
        sourceKey = sourceId;
        // 1. 找出注入时以File.class为modelClass的注入代码
        // 2. 调用所有注入的factory.build方法得到ModelLoader
        // 3 .过滤掉不可能处理model的ModelLoader
        // 此时的modelLoaders值为:
        // [ByteBufferFileLoader, FileLoader, FileLoader, UnitModelLoader]
        modelLoaders = helper.getModelLoaders(cacheFile);
        modelLoaderIndex = 0;
      }
    }
    // 如果找到了缓存文件,hasNextModelLoader()方法则会为true,可以执行循环
    // 没有找到缓存文件,则不会进入循环,会直接返回false
    loadData = null;
    boolean started = false;
    while (!started && hasNextModelLoader()) {
      ModelLoader<File, ?> modelLoader = modelLoaders.get(modelLoaderIndex++);
      // 在循环中会依次判断某个ModelLoader能不能加载此文件
      loadData = modelLoader.buildLoadData(cacheFile,
          helper.getWidth(), helper.getHeight(), helper.getOptions());
      if (loadData != null && helper.hasLoadPath(loadData.fetcher.getDataClass())) {
        started = true;
        // 如果某个ModelLoader可以,那么就调用其fetcher进行加载数据       
// 加载成功或失败会通知自身
        loadData.fetcher.loadData(helper.getPriority(), this);
      }
    }
    return started;
  }

该方法的相关注释代码里都有标明。找缓存时key的类型为ResourceCacheKey,我们先来看下ResourceCacheKey的构成


currentKey =
          new ResourceCacheKey(// NOPMD AvoidInstantiatingObjectsInLoops
              helper.getArrayPool(),
              sourceId,
              helper.getSignature(),
              helper.getWidth(),
              helper.getHeight(),
              transformation,
              resourceClass,
              helper.getOptions());
ResourceCacheKey(
      ArrayPool arrayPool,
      Key sourceKey,
      Key signature,
      int width,
      int height,
      Transformation<?> appliedTransformation,
      Class<?> decodedResourceClass,
      Options options)
  • arrayPool:默认值是LruArrayPool,不参与key的equals方法;
  • sourceKey:如果请求的是URL,此处就是GlideUrl(GlideUrl implements Key);
  • signature:BaseRequestOptions的成员变量,默认会是EmptySignature.obtain(),
  • 在加载本地resource资源时会变成ApplicationVersionSignature.obtain(context);
  • width、height:如果没有指定override(int size),那么将得到view的size;
  • appliedTransformation:默认会根据ImageView的scaleType设置对应的BitmapTransformation;
  • 如果指定了transform,那么就会是指定的值;
  • decodedResourceClass:可以被编码成的资源类型,如BitmapDrawable等;
  • options:如果没有设置过transform,此处会根据ImageView的scaleType默认指定一个option;

在ResourceCacheKey中,arrayPool并没有参与equals方法;

生成ResourceCacheKey之后会根据key去磁盘缓存中查找cacheFile = helper.getDiskCache().get(currentKey);

helper.getDiskCache()返回DiskCache接口,它的实现类是DiskLruCacheWrapper,看下DiskLruCacheWrapper.get方法

@Override
  public File get(Key key) {
    String safeKey = safeKeyGenerator.getSafeKey(key);
    ...
    File result = null;
    try {
      final DiskLruCache.Value value = getDiskCache().get(safeKey);
      if (value != null) {
        result = value.getFile(0);
      }
    } catch (IOException e) {
      ...
    }
    return result;
  }

这里调用SafeKeyGenerator生成了一个String类型的SafeKey,实际上就是对原始key中每个字段都使用SHA-256加密,然后将得到的字节数组转换为16进制的字符串。生成SafeKey后,接着根据SafeKey去DiskCache里面找对应的缓存文件,然后返回文件。

回到ResourceCacheGenerator.startNext方法中,如果找到了缓存会调用loadData.fetcher.loadData(helper.getPriority(), this);这里的 fetcher 是 ByteBufferFetcher,ByteBufferFetcher的loadData方法中最终会执行callback.onDataReady(result)这里callback是ResourceCacheGenerator

  public void onDataReady(Object data) {
    cb.onDataFetcherReady(sourceKey, data, loadData.fetcher, DataSource.RESOURCE_DISK_CACHE,
        currentKey);
  }

ResourceCacheGenerator的onDataReady方法又会回调DecodeJob的onDataFetcherReady方法进行后续的解码操作。

如果ResourceCacheGenerator没有找到缓存,就会交给DataCacheGenerator继续查找缓存。该类大体流程和ResourceCacheGenerator一样,有点不同的是,DataCacheGenerator的构造器有两个构造器,其中的DataCacheGenerator(List, DecodeHelper<?>, FetcherReadyCallback)构造器是给SourceGenerator准备的。因为如果没有磁盘缓存,那么从源头加载后,肯定需要进行磁盘缓存操作的。所以,SourceGenerator会将加载后的资源保存到磁盘中,然后转交给DataCacheGenerator从磁盘中取出交给ImageView展示。

看下DataCacheGenerator.startNext

public boolean startNext() {
    while (modelLoaders == null || !hasNextModelLoader()) {
      sourceIdIndex++;
      if (sourceIdIndex >= cacheKeys.size()) {
        return false;
      }
      Key sourceId = cacheKeys.get(sourceIdIndex);
      ...
      Key originalKey = new DataCacheKey(sourceId, helper.getSignature());
      cacheFile = helper.getDiskCache().get(originalKey);
      ...
    while (!started && hasNextModelLoader()) {
      ModelLoader<File, ?> modelLoader = modelLoaders.get(modelLoaderIndex++);
      loadData =
          modelLoader.buildLoadData(cacheFile, helper.getWidth(), helper.getHeight(),
              helper.getOptions());
      if (loadData != null && helper.hasLoadPath(loadData.fetcher.getDataClass())) {
        started = true;
        loadData.fetcher.loadData(helper.getPriority(), this);
      }
    }
    return started;
  }

这里的originalKey是DataCacheKey类型的,DataCacheKey构造方法如下

DataCacheKey(Key sourceKey, Key signature)

这里的sourceKey和signature与ResourceCacheKey中的两个变量一致,从这里就可以看出:DataCache缓存的是原始的数据,ResourceCache缓存的是是被解码、转换后的数据。

如果DataCacheGenerator没有取到缓存,那么会交给SourceGenerator从源头加载。看下SourceGenerator的startNext方法

@Override
  public boolean startNext() {
    // 首次运行dataToCache为null
    if (dataToCache != null) {
      Object data = dataToCache;
      dataToCache = null;
      cacheData(data);
    }
    // 首次运行sourceCacheGenerator为null
    if (sourceCacheGenerator != null && sourceCacheGenerator.startNext()) {
      return true;
    }
    sourceCacheGenerator = null;
    loadData = null;
    boolean started = false;
    while (!started && hasNextModelLoader()) {
      loadData = helper.getLoadData().get(loadDataListIndex++);
      if (loadData != null
          && (helper.getDiskCacheStrategy().isDataCacheable(loadData.fetcher.getDataSource())
          || helper.hasLoadPath(loadData.fetcher.getDataClass()))) {
        started = true;
        loadData.fetcher.loadData(helper.getPriority(), this);
      }
    }
    return started;
  }

加载成功后依然会回调SourceGenerator的onDataReady方法

@Override
  public void onDataReady(Object data) {
    DiskCacheStrategy diskCacheStrategy = helper.getDiskCacheStrategy();
    if (data != null && diskCacheStrategy.isDataCacheable(loadData.fetcher.getDataSource())) {
      dataToCache = data;
      // cb 为 DecodeJob
      cb.reschedule();
    } else {
      // cb 为 DecodeJob
      cb.onDataFetcherReady(loadData.sourceKey, data, loadData.fetcher,
          loadData.fetcher.getDataSource(), originalKey);
    }
  }

先判断获取到的数据是否需要进行磁盘缓存,如果需要磁盘缓存,则经过DecodeJob、EngineJob的调度,重新调用SourceGenerator.startNext方法,此时dataToCache已经被赋值,则会调用cacheData(data);进行磁盘缓存的写入,并转交给DataCacheGenerator完成后续的处理;否则就通知DecodeJob已经加载成功。

先看下SourceGenerator的startNext方法中调用的SourceGenerator.cacheData(data)

private void cacheData(Object dataToCache) {
    long startTime = LogTime.getLogTime();
    try {
      Encoder<Object> encoder = helper.getSourceEncoder(dataToCache);
      DataCacheWriter<Object> writer =
          new DataCacheWriter<>(encoder, dataToCache, helper.getOptions());
      originalKey = new DataCacheKey(loadData.sourceKey, helper.getSignature());
      helper.getDiskCache().put(originalKey, writer);
      ...
    } finally {
      loadData.fetcher.cleanup();
    }
    sourceCacheGenerator =
        new DataCacheGenerator(Collections.singletonList(loadData.sourceKey), helper, this);
  }

cacheData方法先构建了一个DataCacheKey将data写入了磁盘,然后new了一个DataCacheGenerator赋值给sourceCacheGenerator。回到startNext继续向下执行,此时sourceCacheGenerator不为空,就调用其startNext()方法从磁盘中加载刚写入磁盘的数据,并返回true让DecodeJob停止尝试获取数据。此时,从磁盘缓存中读取数据的逻辑已经完成,接下来是写磁盘缓存。

假如SourceGenerator的onDataReady方法中的磁盘缓存策略不可用,则会回调DecodeJob.onDataFetcherReady方法

// DecodeJob
  @Override
  public void onDataFetcherReady(Key sourceKey, Object data, DataFetcher<?> fetcher,
      DataSource dataSource, Key attemptedKey) {
    this.currentSourceKey = sourceKey;
    this.currentData = data;
    this.currentFetcher = fetcher;
    this.currentDataSource = dataSource;
    this.currentAttemptingKey = attemptedKey;
    if (Thread.currentThread() != currentThread) {
      runReason = RunReason.DECODE_DATA;
      callback.reschedule(this);
    } else {
      GlideTrace.beginSection("DecodeJob.decodeFromRetrievedData");
      try {
        decodeFromRetrievedData();
      } finally {
        GlideTrace.endSection();
      }
    }
  }
  private void decodeFromRetrievedData() {
    ...
    Resource<R> resource = null;
    try {
      resource = decodeFromData(currentFetcher, currentData, currentDataSource);
    } catch (GlideException e) {
      e.setLoggingDetails(currentAttemptingKey, currentDataSource);
      throwables.add(e);
    }
    if (resource != null) {
      notifyEncodeAndRelease(resource, currentDataSource);
    } else {
      runGenerators();
    }
  }

decodeFromRetrievedData();后续的方法调用链在之前的文章中分析过,主要做的事情就是:将原始的data数据转变为可以供ImageView显示的resource数据并将其显示在ImageView上。

将原始的data数据转变为resource数据后,会调用DecodeJob.onResourceDecoded(dataSource, decoded)

@Synthetic
  @NonNull
  <Z> Resource<Z> onResourceDecoded(DataSource dataSource,
      @NonNull Resource<Z> decoded) {
    @SuppressWarnings("unchecked")
    Class<Z> resourceSubClass = (Class<Z>) decoded.get().getClass();
    Transformation<Z> appliedTransformation = null;
    Resource<Z> transformed = decoded;
    // 不是 resource cache时要transform
    if (dataSource != DataSource.RESOURCE_DISK_CACHE) {
      appliedTransformation = decodeHelper.getTransformation(resourceSubClass);
      transformed = appliedTransformation.transform(glideContext, decoded, width, height);
    }
    // TODO: Make this the responsibility of the Transformation.
    if (!decoded.equals(transformed)) {
      decoded.recycle();
    }
    final EncodeStrategy encodeStrategy;
    final ResourceEncoder<Z> encoder;
    if (decodeHelper.isResourceEncoderAvailable(transformed)) {
      encoder = decodeHelper.getResultEncoder(transformed);
      encodeStrategy = encoder.getEncodeStrategy(options);
    } else {
      encoder = null;
      encodeStrategy = EncodeStrategy.NONE;
    }
    Resource<Z> result = transformed;
    boolean isFromAlternateCacheKey = !decodeHelper.isSourceKey(currentSourceKey);
    if (diskCacheStrategy.isResourceCacheable(isFromAlternateCacheKey, dataSource,
        encodeStrategy)) {
      if (encoder == null) {
        throw new Registry.NoResultEncoderAvailableException(transformed.get().getClass());
      }
      final Key key;
      switch (encodeStrategy) {
        case SOURCE:
          key = new DataCacheKey(currentSourceKey, signature);
          break;
        case TRANSFORMED:
          key =
              new ResourceCacheKey(
                  decodeHelper.getArrayPool(),
                  currentSourceKey,
                  signature,
                  width,
                  height,
                  appliedTransformation,
                  resourceSubClass,
                  options);
          break;
        default:
          throw new IllegalArgumentException("Unknown strategy: " + encodeStrategy);
      }
      LockedResource<Z> lockedResult = LockedResource.obtain(transformed);
      deferredEncodeManager.init(key, encoder, lockedResult);
      result = lockedResult;
    }
    return result;
  } 

然后是此过程中的磁盘缓存过程,影响的因素有encodeStrategy、DiskCacheStrategy.isResourceCacheable。encodeStrategy根据resource数据的类型来判断,如果是Bitmap或BitmapDrawable,那么就是TRANSFORMED;如果是GifDrawable,那么就是SOURCE。磁盘缓存策略默认是DiskCacheStrategy.AUTOMATIC。源码如下:

 public static final DiskCacheStrategy AUTOMATIC = new DiskCacheStrategy() {
        public boolean isDataCacheable(DataSource dataSource) {
            return dataSource == DataSource.REMOTE;
        }
        public boolean isResourceCacheable(boolean isFromAlternateCacheKey, DataSource dataSource, EncodeStrategy encodeStrategy) {
            return (isFromAlternateCacheKey && dataSource == DataSource.DATA_DISK_CACHE || dataSource == DataSource.LOCAL) && encodeStrategy == EncodeStrategy.TRANSFORMED;
        }
        public boolean decodeCachedResource() {
            return true;
        }
        public boolean decodeCachedData() {
            return true;
        }
    };
  • 只有dataSource为DataSource.LOCAL且encodeStrategy为EncodeStrategy.TRANSFORMED时,才允许缓存。也就是只有本地的resource数据为Bitmap或BitmapDrawable的资源才可以缓存。
  • 在DecodeJob.onResourceDecoded中会调用deferredEncodeManager.init(key, encoder, lockedResult);去初始化deferredEncodeManager。

在DecodeJob的decodeFromRetrievedData();中拿到resource数据后会调用notifyEncodeAndRelease(resource, currentDataSource)利用deferredEncodeManager对象进行磁盘缓存的写入;

otifyEncodeAndRelease(Resource<R> resource, DataSource dataSource) {
    ...
    // 通知回调,资源已经就绪
    notifyComplete(result, dataSource);
    stage = Stage.ENCODE;
    try {
      if (deferredEncodeManager.hasResourceToEncode()) {
        deferredEncodeManager.encode(diskCacheProvider, options);
      }
    } finally {
      if (lockedResource != null) {
        lockedResource.unlock();
      }
    }
    onEncodeComplete();
  }

deferredEncodeManager.encode行磁盘缓存的写入

// DecodeJob
private static class DeferredEncodeManager<Z> {
  private Key key;
  private ResourceEncoder<Z> encoder;
  private LockedResource<Z> toEncode;
  @Synthetic
  DeferredEncodeManager() { }
  // We just need the encoder and resource type to match, which this will enforce.
  @SuppressWarnings("unchecked")
  <X> void init(Key key, ResourceEncoder<X> encoder, LockedResource<X> toEncode) {
    this.key = key;
    this.encoder = (ResourceEncoder<Z>) encoder;
    this.toEncode = (LockedResource<Z>) toEncode;
  }
  void encode(DiskCacheProvider diskCacheProvider, Options options) {
    GlideTrace.beginSection("DecodeJob.encode");
    try {
      // 存入磁盘缓存
      diskCacheProvider.getDiskCache().put(key,
          new DataCacheWriter<>(encoder, toEncode, options));
    } finally {
      toEncode.unlock();
      GlideTrace.endSection();
    }
  }
  boolean hasResourceToEncode() {
    return toEncode != null;
  }
  void clear() {
    key = null;
    encoder = null;
    toEncode = null;
  }
}

diskCacheProvider.getDiskCache()获取到DiskLruCacheWrapper,并调用DiskLruCacheWrapper的put写入。DiskLruCacheWrapper在写入的时候会使用到写锁DiskCacheWriteLocker,锁对象由对象池WriteLockPool创建,写锁WriteLock实现是一个不公平锁ReentrantLock。

在缓存写入前,会判断key对应的value存不存在,若存在则不写入。缓存的真正写入会由DataCacheWriter交给ByteBufferEncoder和StreamEncoder两个具体类来写入,前者负责将ByteBuffer写入到文件,后者负责将InputStream写入到文件。

目前为止,磁盘缓存的读写流程都已分析完成;

5、内存缓存:ActiveResource与MemoryCache读取

回到DecodeJob.notifyEncodeAndRelease方法中,经过notifyComplete、EngineJob.onResourceReady、notifyCallbacksOfResult方法中。

在该方法中一方面会将原始的resource包装成一个EngineResource,然后通过回调传给Engine.onEngineJobComplete

@Override
  public synchronized void onEngineJobComplete(
      EngineJob<?> engineJob, Key key, EngineResource<?> resource) {
    // 设置资源的回调为自己,这样在资源释放时会通知自己的回调方法
    if (resource != null) {
      resource.setResourceListener(key, this);
      // 将资源放入activeResources中,资源变为active状态
      if (resource.isCacheable()) {
        activeResources.activate(key, resource);
      }
    }
    // 将engineJob从Jobs中移除
    jobs.removeIfCurrent(key, engineJob);
  }

在这里会将资源放入activeResources中,资源变为active状态。后面会使用Executors.mainThreadExecutor()调用SingleRequest.onResourceReady回调进行资源的显示。在触发回调前后各有一个地方会对engineResource进行acquire()和release()操作,这两个操作分别发生在notifyCallbacksOfResult()方法的incrementPendingCallbacks、decrementPendingCallbacks()调用中

@Synthetic
void notifyCallbacksOfResult() {
  ResourceCallbacksAndExecutors copy;
  Key localKey;
  EngineResource<?> localResource;
  synchronized (this) {
    ...
    engineResource = engineResourceFactory.build(resource, isCacheable);
    ...
    hasResource = true;
    copy = cbs.copy();
    incrementPendingCallbacks(copy.size() + 1);
    localKey = key;
    localResource = engineResource;
  }
  listener.onEngineJobComplete(this, localKey, localResource);
  for (final ResourceCallbackAndExecutor entry : copy) {
    entry.executor.execute(new CallResourceReady(entry.cb));
  }
  decrementPendingCallbacks();
}
synchronized void incrementPendingCallbacks(int count) {
  ...
  if (pendingCallbacks.getAndAdd(count) == 0 && engineResource != null) {
    engineResource.acquire();
  }
}
synchronized void decrementPendingCallbacks() {
  ...
  int decremented = pendingCallbacks.decrementAndGet();
  if (decremented == 0) {
    if (engineResource != null) {
      engineResource.release();
    }
    release();
  }
}
private class CallResourceReady implements Runnable {
  private final ResourceCallback cb;
  CallResourceReady(ResourceCallback cb) {
    this.cb = cb;
  }
  @Override
  public void run() {
    synchronized (EngineJob.this) {
      if (cbs.contains(cb)) {
        // Acquire for this particular callback.
        engineResource.acquire();
        callCallbackOnResourceReady(cb);
        removeCallback(cb);
      }
      decrementPendingCallbacks();
    }
  }
}

CallResourceReady的run方法中也会调用engineResource.acquire(),上面的代码调用结束后,engineResource的引用计数为1。engineResource的引用计数会在RequestManager.onDestory方法中最终调用SingleRequest.clear()方法,SingleRequest.clear()内部调用releaseResource()、Engine.release 进行释放,这样引用计数就变为0。引用计数就变为0后会通知Engine将此资源从active状态变成memory cache状态。如果我们再次加载资源时可以从memory cache中加载,那么资源又会从memory cache状态变成active状态。也就是说,在资源第一次显示后,我们关闭页面,资源会由active变成memory cache;然后我们再次进入页面,加载时会命中memory cache,从而又变成active状态

总结

  • 读取内存缓存时,先从LruCache算法机制的内存缓存读取,再从弱引用机制的内存缓存读取;
  • 写入内存缓存时,先写入 弱引用机制 的内存缓存,等到图片不再被使用时,再写入到 LruCache算法机制的内存缓存;
  • 读取磁盘缓存时,先读取转换后图片的缓存,再读取原始图片的缓存;

https://mp.weixin.qq.com/s/nUggwwzxl6McEyqzuVhAnQ

最多阅读

简化Android的UI开发 2年以前  |  514531次阅读
Android 深色模式适配原理分析 1年以前  |  25458次阅读
Android 样式系统 | 主题背景覆盖 1年以前  |  7287次阅读
Android Studio 生成so文件 及调用 1年以前  |  4335次阅读
30分钟搭建一个android的私有Maven仓库 2年以前  |  4047次阅读
Android设计与开发工作流 2年以前  |  3846次阅读
Google Enjarify:可代替dex2jar的dex反编译 2年以前  |  3746次阅读
Android多渠道打包工具:apptools 2年以前  |  3395次阅读
Google Java编程风格规范(中文版) 2年以前  |  3288次阅读
Android-模块化-面向接口编程 1年以前  |  3247次阅读
Android UI基本技术点 2年以前  |  3225次阅读
Android死锁初探 1年以前  |  3194次阅读
Android内存异常机制(用户空间)_NE 1年以前  |  3162次阅读
移动端常见崩溃指标 1年以前  |  3115次阅读

手机扫码阅读