Android Handler机制系列文章整体内容如下：
本片文章的主要内容如下：
一、消息的取出
(一)、消息的取出主要是通过Looper的loop方法
代码如下Looper.java 122行
/**
* Run the message queue in this thread. Be sure to call
* {@link #quit()} to end the loop.
*/
public static void loop() {
//第1步
final Looper me = myLooper();
if (me == null) {
throw new RuntimeException("No Looper; Looper.prepare() wasn't called on this thread.");
}
//第2步
final MessageQueue queue = me.mQueue;
// Make sure the identity of this thread is that of the local process,
// and keep track of what that identity token actually is.
Binder.clearCallingIdentity();
final long ident = Binder.clearCallingIdentity();
//第3步
for (;;) {
//第四步
Message msg = queue.next(); // might block
if (msg == null) {
// No message indicates that the message queue is quitting.
return;
}
// This must be in a local variable, in case a UI event sets the logger
final Printer logging = me.mLogging;
if (logging != null) {
logging.println(">>>>> Dispatching to " + msg.target + " " +
msg.callback + ": " + msg.what);
}
final long traceTag = me.mTraceTag;
if (traceTag != 0 && Trace.isTagEnabled(traceTag)) {
Trace.traceBegin(traceTag, msg.target.getTraceName(msg));
}
try {
// 第5步
msg.target.dispatchMessage(msg);
} finally {
if (traceTag != 0) {
Trace.traceEnd(traceTag);
}
}
if (logging != null) {
logging.println("<<<<< Finished to " + msg.target + " " + msg.callback);
}
// Make sure that during the course of dispatching the
// identity of the thread wasn't corrupted.
final long newIdent = Binder.clearCallingIdentity();
if (ident != newIdent) {
Log.wtf(TAG, "Thread identity changed from 0x"
+ Long.toHexString(ident) + " to 0x"
+ Long.toHexString(newIdent) + " while dispatching to "
+ msg.target.getClass().getName() + " "
+ msg.callback + " what=" + msg.what);
}
// 第6步
msg.recycleUnchecked();
}
}
这个方法已经在Android Handler机制4之Looper与Handler简介中说过了，我就重点说下流程，大体上分为6步
由于第1步、第2步和第3步比较简单就不讲解了，而第6步在y已经讲解过，也不讲解了，下面我们来重点说下第4步和第5步
(二)、Message next()方法
PS:在Looper.loop()中获取消息的方式就是调用next()方法。
代码在MessageQueue.java
307行
Message next() {
// Return here if the message loop has already quit and been disposed.
// This can happen if the application tries to restart a looper after quit
// which is not supported.
// 如果消息循环已经退出了。则直接在这里return。因为调用disposed()方法后mPtr=0
final long ptr = mPtr;
if (ptr == 0) {
return null;
}
//记录空闲时处理的IdlerHandler的数量
int pendingIdleHandlerCount = -1; // -1 only during first iteration
// native层用到的变量 ，如果消息尚未到达处理时间，则表示为距离该消息处理事件的总时长，
// 表明Native Looper只需要block到消息需要处理的时间就行了。 所以nextPollTimeoutMillis>0表示还有消息待处理
int nextPollTimeoutMillis = 0;
for (;;) {
if (nextPollTimeoutMillis != 0) {
//刷新下Binder命令，一般在阻塞前调用
Binder.flushPendingCommands();
}
// 调用native层进行消息标示，nextPollTimeoutMillis 为0立即返回，为-1则阻塞等待。
nativePollOnce(ptr, nextPollTimeoutMillis);
//加上同步锁
synchronized (this) {
// Try to retrieve the next message. Return if found.
// 获取开机到现在的时间
final long now = SystemClock.uptimeMillis();
Message prevMsg = null;
// 获取MessageQueue的链表表头的第一个元素
Message msg = mMessages;
// 判断Message是否是障栅，如果是则执行循环，拦截所有同步消息，直到取到第一个异步消息为止
if (msg != null && msg.target == null) {
// 如果能进入这个if，则表面MessageQueue的第一个元素就是障栅(barrier)
// Stalled by a barrier. Find the next asynchronous message in the queue.
// 循环遍历出第一个异步消息，这段代码可以看出障栅会拦截所有同步消息
do {
prevMsg = msg;
msg = msg.next;
//如果msg==null或者msg是异步消息则退出循环，msg==null则意味着已经循环结束
} while (msg != null && !msg.isAsynchronous());
}
// 判断是否有可执行的Message
if (msg != null) {
// 判断该Mesage是否到了被执行的时间。
if (now < msg.when) {
// Next message is not ready. Set a timeout to wake up when it is ready.
// 当Message还没有到被执行时间的时候，记录下一次要执行的Message的时间点
nextPollTimeoutMillis = (int) Math.min(msg.when - now, Integer.MAX_VALUE);
} else {
// Message的被执行时间已到
// Got a message.
// 从队列中取出该Message，并重新构建原来队列的链接
// 刺客说明说有消息，所以不能阻塞
mBlocked = false;
// 如果还有上一个元素
if (prevMsg != null) {
//上一个元素的next(越过自己)直接指向下一个元素
prevMsg.next = msg.next;
} else {
//如果没有上一个元素，则说明是消息队列中的头元素，直接让第二个元素变成头元素
mMessages = msg.next;
}
// 因为要取出msg，所以msg的next不能指向链表的任何元素，所以next要置为null
msg.next = null;
if (DEBUG) Log.v(TAG, "Returning message: " + msg);
// 标记该Message为正处于使用状态，然后返回Message
msg.markInUse();
return msg;
}
} else {
// No more messages.
// 没有任何可执行的Message，重置时间
nextPollTimeoutMillis = -1;
}
// Process the quit message now that all pending messages have been handled.
// 关闭消息队列，返回null，通知Looper停止循环
if (mQuitting) {
dispose();
return null;
}
// If first time idle, then get the number of idlers to run.
// Idle handles only run if the queue is empty or if the first message
// in the queue (possibly a barrier) is due to be handled in the future.
// 当第一次循环的时候才会在空闲的时候去执行IdleHandler，从代码可以看出所谓的空闲状态
// 指的就是当队列中没有任何可执行的Message，这里的可执行有两要求，
// 即该Message不会被障栅拦截，且Message.when到达了执行时间点
if (pendingIdleHandlerCount < 0
&& (mMessages == null || now < mMessages.when)) {
pendingIdleHandlerCount = mIdleHandlers.size();
}
// 这里是消息队列阻塞( 死循环) 的重点，消息队列在阻塞的标示是消息队列中没有任何消息，
// 并且所有的 IdleHandler 都已经执行过一次了
if (pendingIdleHandlerCount <= 0) {
// No idle handlers to run. Loop and wait some more.
mBlocked = true;
continue;
}
// 初始化要被执行的IdleHandler，最少4个
if (mPendingIdleHandlers == null) {
mPendingIdleHandlers = new IdleHandler[Math.max(pendingIdleHandlerCount, 4)];
}
mPendingIdleHandlers = mIdleHandlers.toArray(mPendingIdleHandlers);
}
// Run the idle handlers.
// We only ever reach this code block during the first iteration.
// 开始循环执行所有的IdleHandler，并且根据返回值判断是否保留IdleHandler
for (int i = 0; i < pendingIdleHandlerCount; i++) {
final IdleHandler idler = mPendingIdleHandlers[i];
mPendingIdleHandlers[i] = null; // release the reference to the handler
boolean keep = false;
try {
keep = idler.queueIdle();
} catch (Throwable t) {
Log.wtf(TAG, "IdleHandler threw exception", t);
}
if (!keep) {
synchronized (this) {
mIdleHandlers.remove(idler);
}
}
}
// Reset the idle handler count to 0 so we do not run them again.
// 重点代码，IdleHandler只会在消息队列阻塞之前执行一次，执行之后改标示设置为0，
// 之后就不会再执行，一直到下一次调用MessageQueue.next() 方法。
pendingIdleHandlerCount = 0;
// While calling an idle handler, a new message could have been delivered
// so go back and look again for a pending message without waiting.
// 当执行了IdleHandler 的 处理之后，会消耗一段时间，这时候消息队列里的可能有消息已经到达
// 可执行时间，所以重置该变量回去重新检查消息队列。
nextPollTimeoutMillis = 0;
}
}
总的来说当我们试图产品从MessageQueue中获取一个Message的时候，会分为以下几步
这里有个难点，我简单说下
//**************** 第一部分 ***************
if (msg != null && msg.target == null) {
// Stalled by a barrier. Find the next asynchronous message in the queue.
do {
prevMsg = msg;
msg = msg.next;
} while (msg != null && !msg.isAsynchronous());
}
//============分割线==============
//**************** 第二部分 ***************
if (msg != null) {
if (now < msg.when) {
// Next message is not ready. Set a timeout to wake up when it is ready.
nextPollTimeoutMillis = (int) Math.min(msg.when - now, Integer.MAX_VALUE);
} else {
// Got a message.
mBlocked = false;
if (prevMsg != null) {
prevMsg.next = msg.next;
} else {
mMessages = msg.next;
}
msg.next = null;
if (DEBUG) Log.v(TAG, "Returning message: " + msg);
msg.markInUse();
return msg;
}
} else {
// No more messages.
nextPollTimeoutMillis = -1;
}
PS:
(三)、msg.target.dispatchMessage(msg);方法
我们知道这个方法其实是Handler的dispatchMessage(Message)方法，那我们就来详细看下
代码在Handler.java 93行
/**
* Handle system messages here.
*/
public void dispatchMessage(Message msg) {
if (msg.callback != null) {
//当Message存在回调方法，回调msg.callback.run()方法；
handleCallback(msg);
} else {
if (mCallback != null) {
//当Handler存在Callback成员变量时，回调方法handleMessage()；
if (mCallback.handleMessage(msg)) {
return;
}
}
//Handler自身的回调方法handleMessage()
handleMessage(msg);
}
}
这个方法很简单就是二个条件，三种情况
这里我们可以看到，在分发消息时三个方法的优先级分别如下：
对于很多情况下，消息分发后的处理情况是第3种情况，即Handler.handleMessage()，一般地往往是通过覆写该方法从而实现自己的业务逻辑。
二、消息(Message)的移除
(一) Handler的消息移除
代码如下，因为不复杂，我就合并在一起了
// Handler.java
public final void removeCallbacks(Runnable r) {
mQueue.removeMessages(this, r, null);
}
public final void removeCallbacks(Runnable r, Object token) {
mQueue.removeMessages(this, r, token);
}
public final void removeMessages(int what) {
mQueue.removeMessages(this, what, null);
}
public final void removeMessages(int what, Object object) {
mQueue.removeMessages(this, what, object);
}
public final void removeCallbacksAndMessages(Object token) {
mQueue.removeCallbacksAndMessages(this, token);
}
所以我们知道，Handler里面的删除工作，其实本地都是调用MessageQueue来操作的。
下面我们就来看下MessageQueue是怎么操作的？
(二) MessageQueue的消息移除
MessageQueue的消息移除在其类类的方法如下：
一共有5个方法如下：
那我们就依次讲解下:
移除方法1：void removeMessages(Handler , int , Object )方法
代码在MessageQueue.java 587行
void removeMessages(Handler h, int what, Object object) {
// 第1步
if (h == null) {
return;
}
// 第2步
synchronized (this) {
// 第3步
Message p = mMessages;
// Remove all messages at front.
//第4步
while (p != null && p.target == h && p.what == what
&& (object == null || p.obj == object)) {
Message n = p.next;
mMessages = n;
p.recycleUnchecked();
p = n;
}
// Remove all messages after front.
//第5步
while (p != null) {
Message n = p.next;
if (n != null) {
if (n.target == h && n.what == what
&& (object == null || n.obj == object)) {
Message nn = n.next;
n.recycleUnchecked();
p.next = nn;
continue;
}
}
p = n;
}
}
}
上面的代码大体可以分为5个步骤如下：
总结一下：
移除方法2：void removeMessages(Handler, Runnable,Object)方法
代码在MessageQueue.java 604行
void removeMessages(Handler h, Runnable r, Object object) {
if (h == null || r == null) {
return;
}
synchronized (this) {
Message p = mMessages;
// Remove all messages at front.
while (p != null && p.target == h && p.callback == r
&& (object == null || p.obj == object)) {
Message n = p.next;
mMessages = n;
p.recycleUnchecked();
p = n;
}
// Remove all messages after front.
while (p != null) {
Message n = p.next;
if (n != null) {
if (n.target == h && n.callback == r
&& (object == null || n.obj == object)) {
Message nn = n.next;
n.recycleUnchecked();
p.next = nn;
continue;
}
}
p = n;
}
}
}
里面代码和移除方法1：void removeMessages(Handler , int , Object )基本一致，唯一不同就是筛选条件不同而已。
移除方法3：void removeMessages(Handler, Runnable,Object)方法
代码在MessageQueue.java 689行
void removeCallbacksAndMessages(Handler h, Object object) {
if (h == null) {
return;
}
synchronized (this) {
Message p = mMessages;
// Remove all messages at front.
while (p != null && p.target == h
&& (object == null || p.obj == object)) {
Message n = p.next;
mMessages = n;
p.recycleUnchecked();
p = n;
}
// Remove all messages after front.
while (p != null) {
Message n = p.next;
if (n != null) {
if (n.target == h && (object == null || n.obj == object)) {
Message nn = n.next;
n.recycleUnchecked();
p.next = nn;
continue;
}
}
p = n;
}
}
}
里面代码和移除方法1：void removeMessages(Handler , int , Object )基本一致，唯一不同就是筛选条件不同而已。
移除方法4：void removeAllMessagesLocked()方法
代码在MessageQueue.java 722行
private void removeAllMessagesLocked() {
Message p = mMessages;
while (p != null) {
Message n = p.next;
p.recycleUnchecked();
p = n;
}
mMessages = null;
}
这个方法很简单，就是删除所有的消息
移除方法5：void removeAllFutureMessagesLocked()
代码在MessageQueue.java 732行
private void removeAllFutureMessagesLocked() {
// 第1步
final long now = SystemClock.uptimeMillis();
// 第2步
Message p = mMessages;
if (p != null) {
// 第3步
if (p.when > now) {
removeAllMessagesLocked();
} else {
// 第4步
Message n;
for (;;) {
n = p.next;
if (n == null) {
return;
}
if (n.when > now) {
break;
}
p = n;
}
// 第5步
p.next = null;
do {
p = n;
n = p.next;
p.recycleUnchecked();
} while (n != null);
}
}
}
这个方法大体上分为5个步骤，具体解释如下：
三、关闭消息队列
代码在MessageQueue.java 413行
void quit(boolean safe) {
// 第1步
if (!mQuitAllowed) {
throw new IllegalStateException("Main thread not allowed to quit.");
}
// 第2步
synchronized (this) {
// 第3步
if (mQuitting) {
return;
}
mQuitting = true;
// 第4步
if (safe) {
removeAllFutureMessagesLocked();
} else {
removeAllMessagesLocked();
}
// We can assume mPtr != 0 because mQuitting was previously false.
// 第5步
nativeWake(mPtr);
}
}
这个方法内部大概分为5个步骤
四、查看消息是否存在
Handler机制也存在查找是否存在某条消息的机制，代码如下：
// Handler.java
public final boolean hasMessages(int what) {
return mQueue.hasMessages(this, what, null);
}
public final boolean hasMessages(int what, Object object) {
return mQueue.hasMessages(this, what, object);
}
public final boolean hasCallbacks(Runnable r) {
return mQueue.hasMessages(this, r, null);
}
我们发现其内部都是调用MessageQueue的hasMessages函数，那我们就来看下
(一) boolean hasMessages(Handler h, int what, Object object) 方法
代码在MessageQueue.java 587行
boolean hasMessages(Handler h, int what, Object object) {
//第1步
if (h == null) {
return false;
}
//第2步
synchronized (this) {
//第3步
Message p = mMessages;
//第4步
while (p != null) {
if (p.target == h && p.what == what && (object == null || p.obj == object)) {
return true;
}
p = p.next;
}
return false;
}
}
该方法的主要内容可以分为4个步骤
boolean hasMessages(Handler h, Runnable r, Object object)方法和本方法基本一致，唯一不同就是筛选条件不同而已。我就说讲解了。
五、阻塞非安全执行
这个方法使用场景是Android初始化一个WindowManagerService，应为WindowManagerService不成功，其他组件就不允许继续，所以使用阻塞的方式直到完成。
代码在Handler.java 461行
/**
* Runs the specified task synchronously.
* <p>
* If the current thread is the same as the handler thread, then the runnable
* runs immediately without being enqueued. Otherwise, posts the runnable
* to the handler and waits for it to complete before returning.
* </p><p>
* This method is dangerous! Improper use can result in deadlocks.
* Never call this method while any locks are held or use it in a
* possibly re-entrant manner.
* </p><p>
* This method is occasionally useful in situations where a background thread
* must synchronously await completion of a task that must run on the
* handler's thread. However, this problem is often a symptom of bad design.
* Consider improving the design (if possible) before resorting to this method.
* </p><p>
* One example of where you might want to use this method is when you just
* set up a Handler thread and need to perform some initialization steps on
* it before continuing execution.
* </p><p>
* If timeout occurs then this method returns <code>false</code> but the runnable
* will remain posted on the handler and may already be in progress or
* complete at a later time.
* </p><p>
* When using this method, be sure to use {@link Looper#quitSafely} when
* quitting the looper. Otherwise {@link #runWithScissors} may hang indefinitely.
* (TODO: We should fix this by making MessageQueue aware of blocking runnables.)
* </p>
*
* @param r The Runnable that will be executed synchronously.
* @param timeout The timeout in milliseconds, or 0 to wait indefinitely.
*
* @return Returns true if the Runnable was successfully executed.
* Returns false on failure, usually because the
* looper processing the message queue is exiting.
*
* @hide This method is prone to abuse and should probably not be in the API.
* If we ever do make it part of the API, we might want to rename it to something
* less funny like runUnsafe().
*/
public final boolean runWithScissors(final Runnable r, long timeout) {
// 第1步
if (r == null) {
throw new IllegalArgumentException("runnable must not be null");
}
// 第2步
if (timeout < 0) {
throw new IllegalArgumentException("timeout must be non-negative");
}
// 第3步
// 如果为同一个线程，则直接执行runnable，而不需要加入到消息队列。
if (Looper.myLooper() == mLooper) {
r.run();
return true;
}
// 第4步
BlockingRunnable br = new BlockingRunnable(r);
return br.postAndWait(this, timeout);
}
首先先简单翻译一下注释：
该方法内部的执行流程主要分为4个步骤，如下：
上面涉及到一个咱们之前没有讲解过的类：BlockingRunnable，他是Handler的静态内部类，我们来研究下
(一)、Handler的静态内部类BlockingRunnable
BlockingRunnable是Handler的一个私有内部静态类，利用Object的wait和notifyAll方法实现。
代码在Handler.java
private static final class BlockingRunnable implements Runnable {
private final Runnable mTask;
private boolean mDone;
public BlockingRunnable(Runnable task) {
mTask = task;
}
@Override
public void run() {
try {
mTask.run();
} finally {
synchronized (this) {
mDone = true;
// runnable 执行完之后，会通知wait的线程不再wait
notifyAll();
}
}
}
public boolean postAndWait(Handler handler, long timeout) {
if (!handler.post(this)) {
return false;
}
synchronized (this) {
if (timeout > 0) {
final long expirationTime = SystemClock.uptimeMillis() + timeout;
while (!mDone) {
long delay = expirationTime - SystemClock.uptimeMillis();
if (delay <= 0) {
return false; // timeout
}
// post runnable 之后，将调用线程变为wait状态
try {
wait(delay);
} catch (InterruptedException ex) {
}
}
} else {
while (!mDone) {
// post runnable 之后，将调用线程变为wait状态
try {
wait();
} catch (InterruptedException ex) {
}
}
}
}
return true;
}
}
通过分析源码我们获取的了如下信息：
