LeakCanary的原理,你知道么?

LeakCanary 是由 Square 开源的针对 AndroidJava 的内存泄漏检测工具。

使用

LeakCanary 的集成过程很简单,首先在 build.gradle 文件中添加依赖:

 
   

dependencies {
 debugImplementation 'com.squareup.leakcanary:leakcanary-android:1.5.4'
 releaseImplementation 'com.squareup.leakcanary:leakcanary-android-no-op:1.5.4'
}

debugrelease 版本中使用的是不同的库。LeakCanary 运行时会经常执行 GC 操作,在 release 版本中会影响效率。android-no-op 版本中基本没有逻辑实现,用于 release 版本。

然后实现自己的 Application 类:

 
   

public class ExampleApplication extends Application {

 @Override public void onCreate() {
   super.onCreate();
   if (LeakCanary.isInAnalyzerProcess(this)) {
     // This process is dedicated to LeakCanary for heap analysis.
     // You should not init your app in this process.
     return;
   }
   LeakCanary.install(this);
   // Normal app init code...
 }
}

这样就集成完成了。当 LeakCanary 检测到内存泄露时,会自动弹出 Notification 通知开发者发生内存泄漏的 Activity 和引用链,以便进行修复。

源码分析

从入口函数 LeakCanary.install(this) 开始分析:

LeakCanary.install

LeakCanary.java

 
   

/**
* Creates a {@link RefWatcher} that works out of the box, and starts watching activity
* references (on ICS+).
*/

public static RefWatcher install(Application application) {
 return refWatcher(application).listenerServiceClass(DisplayLeakService.class)
     .excludedRefs(AndroidExcludedRefs.createAppDefaults().build())
     .buildAndInstall();
}

LeakCanary.refWatcher
LeakCanary.java

 
   

/** Builder to create a customized {@link RefWatcher} with appropriate Android defaults. */
public static AndroidRefWatcherBuilder refWatcher(Context context) {
 return new AndroidRefWatcherBuilder(context);
}

refWatcher() 方法新建了一个 AndroidRefWatcherBuilder 对象,该对象继承于 RefWatcherBuilder 类,配置了一些默认参数,利用建造者构建一个 RefWatcher 对象。

AndroidRefWatcherBuilder.listenerServiceClass
AndroidRefWatcherBuilder.java

 
   

public AndroidRefWatcherBuilder listenerServiceClass(
   Class listenerServiceClass)
{
 return heapDumpListener(new ServiceHeapDumpListener(context, listenerServiceClass));
}

RefWatcherBuilder.java

 
   

/** @see HeapDump.Listener */
public final T heapDumpListener(HeapDump.Listener heapDumpListener) {
 this.heapDumpListener = heapDumpListener;
 return self();
}

DisplayLeakService.java

 
   

/**
* Logs leak analysis results, and then shows a notification which will start {@link
* DisplayLeakActivity}.
*
* You can extend this class and override {@link #afterDefaultHandling(HeapDump, AnalysisResult,
* String)} to add custom behavior, e.g. uploading the heap dump.
*/

public class DisplayLeakService extends AbstractAnalysisResultService {}

listenerServiceClass() 方法绑定了一个后台服务 DisplayLeakService,这个服务主要用来分析内存泄漏结果并发送通知。你可以继承并重写这个类来进行一些自定义操作,比如上传分析结果等。

RefWatcherBuilder.excludedRefs
RefWatcherBuilder.java

 
   

public final T excludedRefs(ExcludedRefs excludedRefs) {
 this.excludedRefs = excludedRefs;
 return self();
}

AndroidExcludedRefs.java

 
   

/**
* This returns the references in the leak path that can be ignored for app developers. This
* doesn't mean there is no memory leak, to the contrary. However, some leaks are caused by bugs
* in AOSP or manufacturer forks of AOSP. In such cases, there is very little we can do as app
* developers except by resorting to serious hacks, so we remove the noise caused by those leaks.
*/

public static ExcludedRefs.Builder createAppDefaults() {
 return createBuilder(EnumSet.allOf(AndroidExcludedRefs.class));
}

public static ExcludedRefs.Builder createBuilder(EnumSet refs) {
 ExcludedRefs.Builder excluded = ExcludedRefs.builder();
 for (AndroidExcludedRefs ref : refs) {
   if (ref.applies) {
     ref.add(excluded);
     ((ExcludedRefs.BuilderWithParams) excluded).named(ref.name());
   }
 }
 return excluded;
}

excludedRefs() 方法定义了一些对于开发者可以忽略的路径,意思就是即使这里发生了内存泄漏,LeakCanary 也不会弹出通知。这大多是系统 Bug 导致的,无需用户进行处理。

AndroidRefWatcherBuilder.buildAndInstall

最后调用 buildAndInstall() 方法构建 RefWatcher 实例并开始监听 Activity 的引用:

AndroidRefWatcherBuilder.java

 
   

/**
* Creates a {@link RefWatcher} instance and starts watching activity references (on ICS+).
*/

public RefWatcher buildAndInstall() {
 RefWatcher refWatcher = build();
 if (refWatcher != DISABLED) {
   LeakCanary.enableDisplayLeakActivity(context);
   ActivityRefWatcher.install((Application) context, refWatcher);
 }
 return refWatcher;
}

看一下主要的 build()install() 方法:

RefWatcherBuilder.build

RefWatcherBuilder.java

 
   

/** Creates a {@link RefWatcher}. */
public final RefWatcher build() {
  if (isDisabled()) {
    return RefWatcher.DISABLED;
  }

  ExcludedRefs excludedRefs = this.excludedRefs;
  if (excludedRefs == null) {
    excludedRefs = defaultExcludedRefs();
  }

  HeapDump.Listener heapDumpListener = this.heapDumpListener;
  if (heapDumpListener == null) {
    heapDumpListener = defaultHeapDumpListener();
  }

  DebuggerControl debuggerControl = this.debuggerControl;
  if (debuggerControl == null) {
    debuggerControl = defaultDebuggerControl();
  }

  HeapDumper heapDumper = this.heapDumper;
  if (heapDumper == null) {
    heapDumper = defaultHeapDumper();
  }

  WatchExecutor watchExecutor = this.watchExecutor;
  if (watchExecutor == null) {
    watchExecutor = defaultWatchExecutor();
  }

  GcTrigger gcTrigger = this.gcTrigger;
  if (gcTrigger == null) {
    gcTrigger = defaultGcTrigger();
  }

  return new RefWatcher(watchExecutor, debuggerControl, gcTrigger, heapDumper, heapDumpListener,
          excludedRefs);
}

build() 方法利用建造者模式构建 RefWatcher 实例,看一下其中的主要参数:

  • watchExecutor : 线程控制器,在 onDestroy() 之后并且主线程空闲时执行内存泄漏检测

  • debuggerControl : 判断是否处于调试模式,调试模式中不会进行内存泄漏检测

  • gcTrigger : 用于 GCwatchExecutor 首次检测到可能的内存泄漏,会主动进行 GCGC 之后会再检测一次,仍然泄漏的判定为内存泄漏,进行后续操作

  • heapDumper : dump 内存泄漏处的 heap 信息,写入 hprof 文件

  • heapDumpListener : 解析完 hprof 文件并通知 DisplayLeakService 弹出提醒

  • excludedRefs : 排除可以忽略的泄漏路径

LeakCanary.enableDisplayLeakActivity

接下来就是最核心的 install() 方法,这里就开始观察 Activity 的引用了。在这之前还执行了一步操作,LeakCanary.enableDisplayLeakActivity(context);

 
   

public static void enableDisplayLeakActivity(Context context) {
 setEnabled(context, DisplayLeakActivity.class, true);
}

最后执行到 LeakCanaryInternals#setEnabledBlocking

 
   

public static void setEnabledBlocking(Context appContext, Class componentClass,
   boolean enabled)
{
 ComponentName component = new ComponentName(appContext, componentClass);
 PackageManager packageManager = appContext.getPackageManager();
 int newState = enabled ? COMPONENT_ENABLED_STATE_ENABLED : COMPONENT_ENABLED_STATE_DISABLED;
 // Blocks on IPC.
 packageManager.setComponentEnabledSetting(component, newState, DONT_KILL_APP);
}

这里启用了 DisplayLeakActivity 并且显示应用图标。注意,这是指的不是你自己的应用图标,是一个单独的 LeakCanary 的应用,用于展示内存泄露历史的,入口函数是 DisplayLeakActivity,在 AndroidManifest.xml 中可以看到默认情况下 android:enabled="false" :

 
   

<activity
   android:theme="@style/leak_canary_LeakCanary.Base"
   android:name=".internal.DisplayLeakActivity"
   android:process=":leakcanary"
   android:enabled="false"
   android:label="@string/leak_canary_display_activity_label"
   android:icon="@mipmap/leak_canary_icon"
   android:taskAffinity="com.squareup.leakcanary.${applicationId}"
   >

 <intent-filter>
   <action android:name="android.intent.action.MAIN"/>
   <category android:name="android.intent.category.LAUNCHER"/>
 intent-filter>
activity>

ActivityRefWatcher.install

ActivityRefWatcher.java

 
   

public static void install(Application application, RefWatcher refWatcher) {
 new ActivityRefWatcher(application, refWatcher).watchActivities();
}

public void watchActivities() {
 // Make sure you don't get installed twice.
 stopWatchingActivities();
 application.registerActivityLifecycleCallbacks(lifecycleCallbacks);
}

watchActivities() 方法中先解绑生命周期回调注册 lifecycleCallbacks,再重新绑定,避免重复绑定。lifecycleCallbacks 监听了 Activity 的各个生命周期,在 onDestroy() 中开始检测当前 Activity 的引用。

 
   

private final Application.ActivityLifecycleCallbacks lifecycleCallbacks =
   new Application.ActivityLifecycleCallbacks() {
     @Override public void onActivityCreated(Activity activity, Bundle savedInstanceState) {
     }

     @Override public void onActivityStarted(Activity activity) {
     }

     @Override public void onActivityResumed(Activity activity) {
     }

     @Override public void onActivityPaused(Activity activity) {
     }

     @Override public void onActivityStopped(Activity activity) {
     }

     @Override public void onActivitySaveInstanceState(Activity activity, Bundle outState) {
     }

     @Override public void onActivityDestroyed(Activity activity) {
       ActivityRefWatcher.this.onActivityDestroyed(activity);
     }
   };

   void onActivityDestroyed(Activity activity) {
     refWatcher.watch(activity);
   }

下面着重分析 RefWatcher 是如何检测 Activity 的。

RefWatcher.watch

调用 RefWatcher#watch 检测 Activity
RefWatcher.java

 
   

/**
* Identical to {@link #watch(Object, String)} with an empty string reference name.
*
* @see #watch(Object, String)
*/

public void watch(Object watchedReference) {
 watch(watchedReference, "");
}

/**
* Watches the provided references and checks if it can be GCed. This method is non blocking,
* the check is done on the {@link WatchExecutor} this {@link RefWatcher} has been constructed
* with.
*
* @param referenceName An logical identifier for the watched object.
*/

public void watch(Object watchedReference, String referenceName) {
 if (this == DISABLED) {
   return;
 }
 checkNotNull(watchedReference, "watchedReference");
 checkNotNull(referenceName, "referenceName");
 final long watchStartNanoTime = System.nanoTime();
 String key = UUID.randomUUID().toString();
 retainedKeys.add(key);
 final KeyedWeakReference reference =
     new KeyedWeakReference(watchedReference, key, referenceName, queue);

 ensureGoneAsync(watchStartNanoTime, reference);
}

watch() 方法的参数是 ObjectLeakCanary 并不仅仅是针对 Android 的,它可以检测任何对象的内存泄漏,原理都是一致的。

这里出现了几个新面孔,先来了解一下各自是什么:

  • retainedKeys : 一个 Set 集合,每个检测的对象都对应着一个唯一的 key,存储在 retainedKeys

  • KeyedWeakReference :  自定义的弱引用,持有检测对象和对用的 key

 
   

final class KeyedWeakReference extends WeakReference<Object> {
 public final String key;
 public final String name;

 KeyedWeakReference(Object referent, String key, String name,
     ReferenceQueue referenceQueue) {
   super(checkNotNull(referent, "referent"), checkNotNull(referenceQueue, "referenceQueue"));
   this.key = checkNotNull(key, "key");
   this.name = checkNotNull(name, "name");
 }
}

  • queue : ReferenceQueue 对象,和 KeyedWeakReference 配合使用

这里有个小知识点,弱引用和引用队列 ReferenceQueue 联合使用时,如果弱引用持有的对象被垃圾回收,Java 虚拟机就会把这个弱引用加入到与之关联的引用队列中。即 KeyedWeakReference 持有的 Activity 对象如果被垃圾回收,该对象就会加入到引用队列 queue 中。

接着看看具体的内存泄漏判断过程:

RefWatcher.ensureGoneAsync

 
   

private void ensureGoneAsync(final long watchStartNanoTime, final KeyedWeakReference reference) {
 watchExecutor.execute(new Retryable() {
   @Override public Retryable.Result run() {
     return ensureGone(reference, watchStartNanoTime);
   }
 });
}

通过 watchExecutor 执行检测操作,这里的 watchExecutorAndroidWatchExecutor 对象。

 
   

@Override protected WatchExecutor defaultWatchExecutor() {
 return new AndroidWatchExecutor(DEFAULT_WATCH_DELAY_MILLIS);
}

DEFAULT_WATCH_DELAY_MILLIS 为 5 s。

 
   

public AndroidWatchExecutor(long initialDelayMillis) {
 mainHandler = new Handler(Looper.getMainLooper());
 HandlerThread handlerThread = new HandlerThread(LEAK_CANARY_THREAD_NAME);
 handlerThread.start();
 backgroundHandler = new Handler(handlerThread.getLooper());
 this.initialDelayMillis = initialDelayMillis;
 maxBackoffFactor = Long.MAX_VALUE / initialDelayMillis;
}

看看其中用到的几个对象:

  • mainHandler : 主线程消息队列

  • handlerThread : 后台线程,HandlerThread 对象,线程名为 LeakCanary-Heap-Dump

  • backgroundHandler : 上面的后台线程的消息队列

  • initialDelayMillis : 5 s,即之前的 DEFAULT_WATCH_DELAY_MILLIS

 
   

@Override public void execute(Retryable retryable) {
 if (Looper.getMainLooper().getThread() == Thread.currentThread()) {
   waitForIdle(retryable, 0);
 } else {
   postWaitForIdle(retryable, 0);
 }
}

void postWaitForIdle(final Retryable retryable, final int failedAttempts) {
 mainHandler.post(new Runnable() {
   @Override public void run() {
     waitForIdle(retryable, failedAttempts);
   }
 });
}

void waitForIdle(final Retryable retryable, final int failedAttempts) {
 // This needs to be called from the main thread.
 Looper.myQueue().addIdleHandler(new MessageQueue.IdleHandler() {
   @Override public boolean queueIdle() {
     postToBackgroundWithDelay(retryable, failedAttempts);
     return false;
   }
 });
}

在具体的 execute() 过程中,不管是 waitForIdle 还是 postWaitForIdle,最终还是要切换到主线程中执行。要注意的是,这里的 IdleHandler 到底是什么时候去执行?

我们都知道 Handler 是循环处理 MessageQueue 中的消息的,当消息队列中没有更多消息需要处理的时候,且声明了 IdleHandler 接口,这是就会去处理这里的操作。即指定一些操作,当线程空闲的时候来处理。当主线程空闲时,就会通知后台线程延时 5 秒执行内存泄漏检测工作。

 
   

void postToBackgroundWithDelay(final Retryable retryable, final int failedAttempts) {
 long exponentialBackoffFactor = (long) Math.min(Math.pow(2, failedAttempts), maxBackoffFactor);
 long delayMillis = initialDelayMillis * exponentialBackoffFactor;
 backgroundHandler.postDelayed(new Runnable() {
   @Override public void run() {
     Retryable.Result result = retryable.run();
     if (result == RETRY) {
       postWaitForIdle(retryable, failedAttempts + 1);
     }
   }
 }, delayMillis);
}

下面是真正的检测过程,AndroidWatchExecutor 在执行时调用 ensureGone() 方法:

RefWatcher.ensureGone

 
   

Retryable.Result ensureGone(final KeyedWeakReference reference, final long watchStartNanoTime) {
 long gcStartNanoTime = System.nanoTime();
 long watchDurationMs = NANOSECONDS.toMillis(gcStartNanoTime - watchStartNanoTime);

 removeWeaklyReachableReferences();

 if (debuggerControl.isDebuggerAttached()) {
   // The debugger can create false leaks.
   return RETRY;
 }
 if (gone(reference)) {
   return DONE;
 }
 gcTrigger.runGc();
 removeWeaklyReachableReferences();
 if (!gone(reference)) {
   long startDumpHeap = System.nanoTime();
   long gcDurationMs = NANOSECONDS.toMillis(startDumpHeap - gcStartNanoTime);

   File heapDumpFile = heapDumper.dumpHeap();
   if (heapDumpFile == RETRY_LATER) {
     // Could not dump the heap.
     return RETRY;
   }
   long heapDumpDurationMs = NANOSECONDS.toMillis(System.nanoTime() - startDumpHeap);
   heapdumpListener.analyze(
       new HeapDump(heapDumpFile, reference.key, reference.name, excludedRefs, watchDurationMs,
           gcDurationMs, heapDumpDurationMs));
 }
 return DONE;
}

再重复一次几个变量的含义,retainedKeys 是一个 Set集合,存储检测对象对应的唯一 key 值,queue是一个引用队列,存储被垃圾回收的对象。

主要过程有一下几步:

RefWatcher.emoveWeaklyReachableReferences()

 
   

private void removeWeaklyReachableReferences() {
 // WeakReferences are enqueued as soon as the object to which they point to becomes weakly
 // reachable. This is before finalization or garbage collection has actually happened.
 KeyedWeakReference ref;
 while ((ref = (KeyedWeakReference) queue.poll()) != null) {
   retainedKeys.remove(ref.key);
 }
}

遍历引用队列 queue,判断队列中是否存在当前 Activity 的弱引用,存在则删除 retainedKeys 中对应的引用的 key值。

RefWatcher.gone()

 
   

private boolean gone(KeyedWeakReference reference) {
 return !retainedKeys.contains(reference.key);
}

判断 retainedKeys 中是否包含当前 Activity 引用的 key 值。

如果不包含,说明上一步操作中 retainedKeys 移除了该引用的 key 值,也就说上一步操作之前引用队列 queue 中包含该引用,GC 处理了该引用,未发生内存泄漏,返回 DONE,不再往下执行。

如果包含,并不会立即判定发生内存泄漏,可能存在某个对象已经不可达,但是尚未进入引用队列 queue。这时会主动执行一次 GC 操作之后再次进行判断。

gcTrigger.runGc()

 
   

/**
* Called when a watched reference is expected to be weakly reachable, but hasn't been enqueued
* in the reference queue yet. This gives the application a hook to run the GC before the {@link
* RefWatcher} checks the reference queue again, to avoid taking a heap dump if possible.
*/

public interface GcTrigger {
 GcTrigger DEFAULT = new GcTrigger() {
   @Override public void runGc() {
     // Code taken from AOSP FinalizationTest:
     // https://android.googlesource.com/platform/libcore/+/master/support/src/test/java/libcore/
     // java/lang/ref/FinalizationTester.java
     // System.gc() does not garbage collect every time. Runtime.gc() is
     // more likely to perfom a gc.
     Runtime.getRuntime().gc();
     enqueueReferences();
     System.runFinalization();
   }

   private void enqueueReferences() {
     // Hack. We don't have a programmatic way to wait for the reference queue daemon to move
     // references to the appropriate queues.
     try {
       Thread.sleep(100);
     } catch (InterruptedException e) {
       throw new AssertionError();
     }
   }
 };

 void runGc();
}

注意这里调用 GC 的写法,并不是使用 System.gcSystem.gc 仅仅只是通知系统在合适的时间进行一次垃圾回收操作,实际上并不能保证一定执行。

主动进行 GC 之后会再次进行判定,过程同上。首先调用 removeWeaklyReachableReferences() 清除 retainedKeys 中弱引用的 key 值,再判断是否移除。如果仍然没有移除,判定为内存泄漏。

内存泄露结果处理

AndroidHeapDumper.dumpHeap

判定内存泄漏之后,调用 heapDumper.dumpHeap() 进行处理:

AndroidHeapDumper.java

 
   

@SuppressWarnings("ReferenceEquality") // Explicitly checking for named null.
@Override public File dumpHeap() {
 File heapDumpFile = leakDirectoryProvider.newHeapDumpFile();

 if (heapDumpFile == RETRY_LATER) {
   return RETRY_LATER;
 }

 FutureResult waitingForToast = new FutureResult<>();
 showToast(waitingForToast);

 if (!waitingForToast.wait(5, SECONDS)) {
   CanaryLog.d("Did not dump heap, too much time waiting for Toast.");
   return RETRY_LATER;
 }

 Toast toast = waitingForToast.get();
 try {
   Debug.dumpHprofData(heapDumpFile.getAbsolutePath());
   cancelToast(toast);
   return heapDumpFile;
 } catch (Exception e) {
   CanaryLog.d(e, "Could not dump heap");
   // Abort heap dump
   return RETRY_LATER;
 }
}

leakDirectoryProvider.newHeapDumpFile() 新建了 hprof 文件,然后调用 Debug.dumpHprofData() 方法 dump 当前堆内存并写入刚才创建的文件。

回到 RefWatcher.ensureGone() 方法中,生成 heapDumpFile 文件之后,通过 heapdumpListener 分析。

ServiceHeapDumpListener.analyze

 
   

heapdumpListener.analyze(
         new HeapDump(heapDumpFile, reference.key, reference.name, excludedRefs, watchDurationMs,
             gcDurationMs, heapDumpDurationMs));

这里的 heapdumpListenerServiceHeapDumpListener 对象,接着进入 ServiceHeapDumpListener.runAnalysis() 方法。

 
   

@Override public void analyze(HeapDump heapDump) {
 checkNotNull(heapDump, "heapDump");
 HeapAnalyzerService.runAnalysis(context, heapDump, listenerServiceClass);
}

这里的 listenerServiceClass 指的是 DisplayLeakService.class,文章开头提到的 AndroidRefWatcherBuilder 中进行了配置。

 
   

@Override protected HeapDump.Listener defaultHeapDumpListener() {
 return new ServiceHeapDumpListener(context, DisplayLeakService.class);
}

HeapAnalyzerService.runAnalysis
HeapAnalyzerService.runAnalysis() 方法中启动了它自己,传递了两个参数,DisplayLeakService 类名和要分析的 heapDump。启动自己后,在 onHandleIntent 中进行处理。

 
   

/**
* This service runs in a separate process to avoid slowing down the app process or making it run
* out of memory.
*/

public final class HeapAnalyzerService extends IntentService {

 private static final String LISTENER_CLASS_EXTRA = "listener_class_extra";
 private static final String HEAPDUMP_EXTRA = "heapdump_extra";

 public static void runAnalysis(Context context, HeapDump heapDump,
     Class listenerServiceClass)
{
   Intent intent = new Intent(context, HeapAnalyzerService.class);
   intent.putExtra(LISTENER_CLASS_EXTRA, listenerServiceClass.getName());
   intent.putExtra(HEAPDUMP_EXTRA, heapDump);
   context.startService(intent);
 }

 public HeapAnalyzerService() {
   super(HeapAnalyzerService.class.getSimpleName());
 }

 @Override protected void onHandleIntent(Intent intent) {
   if (intent == null) {
     CanaryLog.d("HeapAnalyzerService received a null intent, ignoring.");
     return;
   }
   String listenerClassName = intent.getStringExtra(LISTENER_CLASS_EXTRA);
   HeapDump heapDump = (HeapDump) intent.getSerializableExtra(HEAPDUMP_EXTRA);

   HeapAnalyzer heapAnalyzer = new HeapAnalyzer(heapDump.excludedRefs);

   AnalysisResult result = heapAnalyzer.checkForLeak(heapDump.heapDumpFile, heapDump.referenceKey);
   AbstractAnalysisResultService.sendResultToListener(this, listenerClassName, heapDump, result);
 }
}

heapAnalyzer.checkForLeak

checkForLeak 方法中主要使用了 Square 公司的另一个库 haha 来分析 Android heap dump,得到结果后回调给 DisplayLeakService

AbstractAnalysisResultService.sendResultToListener

 
   

public static void sendResultToListener(Context context, String listenerServiceClassName,
   HeapDump heapDump, AnalysisResult result)
{
 Class listenerServiceClass;
 try {
   listenerServiceClass = Class.forName(listenerServiceClassName);
 } catch (ClassNotFoundException e) {
   throw new RuntimeException(e);
 }
 Intent intent = new Intent(context, listenerServiceClass);
 intent.putExtra(HEAP_DUMP_EXTRA, heapDump);
 intent.putExtra(RESULT_EXTRA, result);
 context.startService(intent);
}

同样在 onHandleIntent 中进行处理。

DisplayLeakService.onHandleIntent

 
   

@Override protected final void onHandleIntent(Intent intent) {
 HeapDump heapDump = (HeapDump) intent.getSerializableExtra(HEAP_DUMP_EXTRA);
 AnalysisResult result = (AnalysisResult) intent.getSerializableExtra(RESULT_EXTRA);
 try {
   onHeapAnalyzed(heapDump, result);
 } finally {
   //noinspection ResultOfMethodCallIgnored
   heapDump.heapDumpFile.delete();
 }
}

DisplayLeakService.onHeapAnalyzed

调用 onHeapAnalyzed() 之后,会将 hprof 文件删除。

DisplayLeakService.java

 
   

@Override protected final void onHeapAnalyzed(HeapDump heapDump, AnalysisResult result) {
 String leakInfo = leakInfo(this, heapDump, result, true);
 CanaryLog.d("%s", leakInfo);

 boolean resultSaved = false;
 boolean shouldSaveResult = result.leakFound || result.failure != null;
 if (shouldSaveResult) {
   heapDump = renameHeapdump(heapDump);
   resultSaved = saveResult(heapDump, result);
 }

 PendingIntent pendingIntent;
 String contentTitle;
 String contentText;

 if (!shouldSaveResult) {
   contentTitle = getString(R.string.leak_canary_no_leak_title);
   contentText = getString(R.string.leak_canary_no_leak_text);
   pendingIntent = null;
 } else if (resultSaved) {
   pendingIntent = DisplayLeakActivity.createPendingIntent(this, heapDump.referenceKey);

   if (result.failure == null) {
     String size = formatShortFileSize(this, result.retainedHeapSize);
     String className = classSimpleName(result.className);
     if (result.excludedLeak) {
       contentTitle = getString(R.string.leak_canary_leak_excluded, className, size);
     } else {
       contentTitle = getString(R.string.leak_canary_class_has_leaked, className, size);
     }
   } else {
     contentTitle = getString(R.string.leak_canary_analysis_failed);
   }
   contentText = getString(R.string.leak_canary_notification_message);
 } else {
   contentTitle = getString(R.string.leak_canary_could_not_save_title);
   contentText = getString(R.string.leak_canary_could_not_save_text);
   pendingIntent = null;
 }
 // New notification id every second.
 int notificationId = (int) (SystemClock.uptimeMillis() / 1000);
 showNotification(this, contentTitle, contentText, pendingIntent, notificationId);
 afterDefaultHandling(heapDump, result, leakInfo);
}

根据分析结果,调用 showNotification() 方法构建了一个 Notification 向开发者通知内存泄漏。

 
   

public static void showNotification(Context context, CharSequence contentTitle,
   CharSequence contentText, PendingIntent pendingIntent, int notificationId)
{
 NotificationManager notificationManager =
     (NotificationManager) context.getSystemService(Context.NOTIFICATION_SERVICE);

 Notification notification;
 Notification.Builder builder = new Notification.Builder(context) //
     .setSmallIcon(R.drawable.leak_canary_notification)
     .setWhen(System.currentTimeMillis())
     .setContentTitle(contentTitle)
     .setContentText(contentText)
     .setAutoCancel(true)
     .setContentIntent(pendingIntent);
 if (SDK_INT >= O) {
   String channelName = context.getString(R.string.leak_canary_notification_channel);
   setupNotificationChannel(channelName, notificationManager, builder);
 }
 if (SDK_INT < JELLY_BEAN) {
   notification = builder.getNotification();
 } else {
   notification = builder.build();
 }
 notificationManager.notify(notificationId, notification);
}

DisplayLeakService.afterDefaultHandling

最后还会执行一个空实现的方法 afterDefaultHandling

 
   

/**
* You can override this method and do a blocking call to a server to upload the leak trace and
* the heap dump. Don't forget to check {@link AnalysisResult#leakFound} and {@link
* AnalysisResult#excludedLeak} first.
*/

protected void afterDefaultHandling(HeapDump heapDump, AnalysisResult result, String leakInfo) {
}

你可以重写这个方法进行一些自定义的操作,比如向服务器上传泄漏的堆栈信息等。

这样,LeakCanary 就完成了整个内存泄漏检测的过程。可以看到,LeakCanary 的设计思路十分巧妙,同时也很清晰,有很多有意思的知识点,像对于弱引用和 ReferenceQueue 的使用, IdleHandler 的使用,四大组件的开启和关闭等等,都很值的大家去深究。

原创投稿:孙路遥

原文链接:https://www.jianshu.com/p/751f99918711

LeakCanary的原理,你知道么?_第1张图片

你可能感兴趣的:(LeakCanary的原理,你知道么?)