移步Android Handler机制详解
- 1、Handler发送消息
- 2、Handler的send方案
- 3、Handler的post方案
1 Handler发送消息
send方案.png
post方案.png
- send方案发送消息(需要回调才能接收消息)
- sendMessage(Message) 立即发送Message到消息队列
- sendMessageAtFrontOfQueue(Message) 立即发送Message到队列，而且是放在队列的最前面
- sendMessageAtTime(Message,long) 设置时间，发送Message到队列
- sendMessageDelayed(Message,long) 延时若干毫秒后，发送Message到队列
- post方案 立即发送Message到消息队列
- post(Runnable) 立即发送Message到消息队列
- postAtFrontOfQueue(Runnable) 立即发送Message到队列，而且是放在队列的最前面
- postAtTime(Runnable,long) 设置时间，发送Message到队列
- postDelayed(Runnable,long) 在延时若干毫秒后，发送Message到队列
2 Handler的send方案
以Handler的sendMessage(Message msg)为例子
2.1 sendMessage(Message msg)方法
代码在Handler.java 505行
/**
* Pushes a message onto the end of the message queue after all pending messages
* before the current time. It will be received in {@link #handleMessage},
* in the thread attached to this handler.
*
* @return Returns true if the message was successfully placed in to the
* message queue. Returns false on failure, usually because the
* looper processing the message queue is exiting.
*/
public final boolean sendMessage(Message msg)
{
return sendMessageDelayed(msg, 0);
}
- 在当前时间，在所有待处理消息之后，将消息推送到消息队列的末尾。在和当前线程关联的的Handler里面的handleMessage将收到这条消息
2.1.1 boolean sendMessageDelayed(Message msg, long delayMillis)
代码在Handler.java 565行
/**
* Enqueue a message into the message queue after all pending messages
* before (current time + delayMillis). You will receive it in
* {@link #handleMessage}, in the thread attached to this handler.
*
* @return Returns true if the message was successfully placed in to the
* message queue. Returns false on failure, usually because the
* looper processing the message queue is exiting. Note that a
* result of true does not mean the message will be processed -- if
* the looper is quit before the delivery time of the message
* occurs then the message will be dropped.
*/
public final boolean sendMessageDelayed(Message msg, long delayMillis)
{
if (delayMillis < 0) {
delayMillis = 0;
}
return sendMessageAtTime(msg, SystemClock.uptimeMillis() + delayMillis);
}
该方法内部就做了两件事
- 1、判断delayMillis是否小于0
- 2、调用了public boolean sendMessageAtTime(Message msg, long uptimeMillis)方法
2.1.2 boolean sendMessageAtTime(Message msg, long uptimeMillis)
代码在Handler.java 592行
/**
* Enqueue a message into the message queue after all pending messages
* before the absolute time (in milliseconds) uptimeMillis.
* The time-base is {@link android.os.SystemClock#uptimeMillis}.
* Time spent in deep sleep will add an additional delay to execution.
* You will receive it in {@link #handleMessage}, in the thread attached
* to this handler.
*
* @param uptimeMillis The absolute time at which the message should be
* delivered, using the
* {@link android.os.SystemClock#uptimeMillis} time-base.
*
* @return Returns true if the message was successfully placed in to the
* message queue. Returns false on failure, usually because the
* looper processing the message queue is exiting. Note that a
* result of true does not mean the message will be processed -- if
* the looper is quit before the delivery time of the message
* occurs then the message will be dropped.
*/
public boolean sendMessageAtTime(Message msg, long uptimeMillis) {
MessageQueue queue = mQueue;
if (queue == null) {
RuntimeException e = new RuntimeException(
this + " sendMessageAtTime() called with no mQueue");
Log.w("Looper", e.getMessage(), e);
return false;
}
return enqueueMessage(queue, msg, uptimeMillis);
}
- 以android系统的SystemClock的uptimeMillis()为基准，以毫秒为基本单位的绝对时间下，在所有待处理消息后，将消息放到消息队列中。
- 深度睡眠中的时间将会延迟执行的时间，你将在和当前线程办的规定的Handler中的handleMessage中收到该消息。
- 因为通常我们理解的异步是指新开一个线程，但是这里不是，因为异步的也是发送到looper所绑定的消息队列中，这里的异步主要是针对Message中的障栅(Barrier)而言的，当出现障栅(Barrier)的时候，同步的会被阻塞，而异步的则不会。所以这个异步仅仅是一个标记而已。
该方法内部就做了两件事
- 1、获取消息队列，并对该消息队列做非空判断，如果为null，直接返回false，表示消息发送失败
- 2、调用了boolean enqueueMessage(MessageQueue queue, Message msg, long uptimeMillis)方法
2.1.3 boolean enqueueMessage(MessageQueue queue, Message msg, long uptimeMillis)
代码在Handler.java 626行
private boolean enqueueMessage(MessageQueue queue, Message msg, long uptimeMillis) {
msg.target = this;
if (mAsynchronous) {
msg.setAsynchronous(true);
}
return queue.enqueueMessage(msg, uptimeMillis);
}
本方法内部做了三件事
- 设置msg的target变量，并将target指向自己
- 如果Handler的mAsynchronous值为true(默认为false，即不设置)，则设置msg的flags值，让是否异步在Handler和Message达成统一。
- 调用MessageQueue的enqueueMessage()方法
2.1.4 boolean enqueueMessage(Message msg, long when)方法
代码在MessageQueue.java 533行
boolean enqueueMessage(Message msg, long when) {
//第一步
if (msg.target == null) {
throw new IllegalArgumentException("Message must have a target.");
}
// 第二步
if (msg.isInUse()) {
throw new IllegalStateException(msg + " This message is already in use.");
}
// 第三步
synchronized (this) {
// 第四步
//判断消息队列是否正在关闭
if (mQuitting) {
IllegalStateException e = new IllegalStateException(
msg.target + " sending message to a Handler on a dead thread");
Log.w(TAG, e.getMessage(), e);
msg.recycle();
return false;
}
// 第五步
msg.markInUse();
msg.when = when;
Message p = mMessages;
boolean needWake;
// 第六步
//根据when的比较来判断要添加的Message是否应该放在队列头部，当第一个添加消息的时候，
// 测试队列为空，所以该Message也应该位于头部。
if (p == null || when == 0 || when < p.when) {
// New head, wake up the event queue if blocked.
// 把msg的下一个元素设置为p
msg.next = p;
// 把msg设置为链表的头部元素
mMessages = msg;
// 如果有阻塞，则需要唤醒
needWake = mBlocked;
} else {
// 第七步
// Inserted within the middle of the queue. Usually we don't have to wake
// up the event queue unless there is a barrier at the head of the queue
// and the message is the earliest asynchronous message in the queue.
//除非消息队列的头部是障栅(barrier)，或者消息队列的第一个消息是异步消息，
//否则如果是插入到中间位置，我们通常不唤醒消息队列，
// 第八步
needWake = mBlocked && p.target == null && msg.isAsynchronous();
Message prev;
// 第九步
// 不断遍历消息队列，根据when的比较找到合适的插入Message的位置。
for (;;) {
prev = p;
p = p.next;
// 第十步
if (p == null || when < p.when) {
break;
}
// 第十一 步
if (needWake && p.isAsynchronous()) {
needWake = false;
}
}
// 第十二步
msg.next = p; // invariant: p == prev.next
prev.next = msg;
}
// 第十三步
// We can assume mPtr != 0 because mQuitting is false.
if (needWake) {
nativeWake(mPtr);
}
}
// 第十四步
return true;
}
- 第1步骤、 判断msg的target变量是否为null，如果为null，则为障栅(barrier)，而障栅(barrier)入队则是通过postSyncBarrier()方法入队，所以msg的target一定有值
- 第2步骤、 判断msg的标志位，因为此时的msg应该是要入队，意味着msg的标志位应该显示还未被使用。如果显示已使用，明显有问题，直接抛异常。
- 第3步骤、 加入同步锁。
- 第4步骤、 判断消息队列是否正在被关闭，如果是正在被关闭，则return false告诉消息入队是失败，并且回收消息
- 第5步骤、 设置msg的when并且修改msg的标志位，msg标志位显示为已使用
- 第6步骤、 如果p==null则说明消息队列中的链表的头部元素为null；when == 0 表示立即执行；when< p.when 表示 msg的执行时间早与链表中的头部元素的时间，所以上面三个条件，那个条件成立，都要把msg设置成消息队列中链表的头部是元素
- 第7步骤、 如果上面三个条件都不满足则说明要把msg插入到中间的位置，不需要插入到头部
- 第8步骤、 如果头部元素不是障栅(barrier)或者异步消息，而且还是插入中间的位置，我们是不唤醒消息队列的。
- 第9步骤、 进入一个死循环，将p的值赋值给prev，前面的带我们知道，p指向的是mMessage，所以这里是将prev指向了mMessage，在下一次循环的时候，prev则指向了第一个message，一次类推。接着讲p指向了p.next也就是mMessage.next，也就是消息队列链表中的第二个元素。这一步骤实现了消息指针的移动，此时p表示的消息队列中第二个元素。
- 第10步骤、 p==null，则说明没有下一个元素，即消息队列到头了，跳出循环；p!=null&&when < p.when 则说明当前需要入队的这个message的执行时间是小于队列中这个任务的执行时间的，也就是说这个需要入队的message需要比队列中这个message先执行，则说明这个位置刚刚是适合这个message的，所以跳出循环。 如果上面的两个条件都不满足，则说明这个位置还不是放置这个需要入队的message，则继续跟链表中后面的元素，也就是继续跟消息队列中的下一个消息进行对比，直到满足条件或者到达队列的末尾。
- 第11步骤、 因为没有满足条件，说明队列中还有消息，不需要唤醒。
- 第12步骤、 跳出循环后主要做了两件事：事件A，将入队的这个消息的next指向循环中获取到的应该排在这个消息之后message。事件B，将msg前面的message.next指向了msg。这样就将一个message完成了入队。
- 第13步骤、 如果需要唤醒，则唤醒，具体请看后面的Handler中的Native详解。
- 第14步骤、 返回true，告知入队成功。
插入消息后，有条件地执行nativeWake去唤醒epoll。needWake的值依赖mBlocked——当进入next()方法时没有待处理的消息，mBlock为true，有消息并返回到looper时，mBlock为false。
结合next()和enqueueMessage()方法，得知nativeWake被调用的条件为：
- next()方法在等待新消息，且新插入消息的为链表头时。needWake为true
- 设置了Sync Barrier，且插入的消息是“异步”的。needWake为true
2.2 boolean sendMessageAtFrontOfQueue(Message msg)
代码在Handler.java 615行
/**
* Enqueue a message at the front of the message queue, to be processed on
* the next iteration of the message loop. You will receive it in
* {@link #handleMessage}, in the thread attached to this handler.
* This method is only for use in very special circumstances -- it
* can easily starve the message queue, cause ordering problems, or have
* other unexpected side-effects.
*
* @return Returns true if the message was successfully placed in to the
* message queue. Returns false on failure, usually because the
* looper processing the message queue is exiting.
*/
public final boolean sendMessageAtFrontOfQueue(Message msg) {
MessageQueue queue = mQueue;
if (queue == null) {
RuntimeException e = new RuntimeException(
this + " sendMessageAtTime() called with no mQueue");
Log.w("Looper", e.getMessage(), e);
return false;
}
return enqueueMessage(queue, msg, 0);
}
- 在消息队列的最前面插入一个消息，在消息循环的下一次迭代中进行处理。
- 由于它可以轻松的解决消息队列的排序问题和其他的意外副作用。
方法内部的实现和boolean sendMessageAtTime(Message msg, long uptimeMillis)大体上一致，唯一的区别就是该方法在调用enqueueMessage(MessageQueue, Message, long)方法的时候，最后一个参数是0而已。
2.3 boolean sendEmptyMessage(int what)
代码在Handler.java 517行
/**
* Sends a Message containing only the what value.
*
* @return Returns true if the message was successfully placed in to the
* message queue. Returns false on failure, usually because the
* looper processing the message queue is exiting.
*/
public final boolean sendEmptyMessage(int what)
{
return sendEmptyMessageDelayed(what, 0);
}
发送一个仅有what的Message
boolean sendEmptyMessageDelayed(int what, long delayMillis)
代码在Handler.java 531行
/**
* Sends a Message containing only the what value, to be delivered
* after the specified amount of time elapses.
* @see #sendMessageDelayed(android.os.Message, long)
*
* @return Returns true if the message was successfully placed in to the
* message queue. Returns false on failure, usually because the
* looper processing the message queue is exiting.
*/
public final boolean sendEmptyMessageDelayed(int what, long delayMillis) {
Message msg = Message.obtain();
msg.what = what;
return sendMessageDelayed(msg, delayMillis);
}
- 发送一个仅有what的Message，并且延迟特定的时间发送
- 这个方法内部主要就是做了3件事
1、调用Message.obtain();从消息对象池中获取一个空的Message。
2、设置这个Message的what值
3、调用sendMessageDelayed(Message,long) 将这个消息方法
2.4 boolean sendEmptyMessageAtTime(int what, long uptimeMillis)
代码在Handler.java 547行
/**
* Sends a Message containing only the what value, to be delivered
* at a specific time.
* @see #sendMessageAtTime(android.os.Message, long)
*
* @return Returns true if the message was successfully placed in to the
* message queue. Returns false on failure, usually because the
* looper processing the message queue is exiting.
*/
public final boolean sendEmptyMessageAtTime(int what, long uptimeMillis) {
Message msg = Message.obtain();
msg.what = what;
return sendMessageAtTime(msg, uptimeMillis);
}
- 发送一个仅有what的Message，并且在特定的时间发送
- 这个方法内部主要就是做了3件事
- 1、调用Message.obtain();从消息对象池中获取一个空的Message。
- 2、设置这个Message的what值
- 3、调用sendMessageAtTime(Message,long) 将这个消息方法
2.5 小结
这些send方案都会从这里或者那里最终走到boolean enqueueMessage(MessageQueue queue, Message msg, long uptimeMillis)
5713484-e56c82c4e36768f5.png
3 Handler的post方案
3.1 boolean post(Runnable r)方法
代码在Handler.java 324行
/**
* Causes the Runnable r to be added to the message queue.
* The runnable will be run on the thread to which this handler is
* attached.
*
* @param r The Runnable that will be executed.
*
* @return Returns true if the Runnable was successfully placed in to the
* message queue. Returns false on failure, usually because the
* looper processing the message queue is exiting.
*/
public final boolean post(Runnable r)
{
return sendMessageDelayed(getPostMessage(r), 0);
}
将一个Runnable添加到消息队列中，这个runnable将会在和当前Handler关联的线程中被执行。
就是调用了sendMessageDelayed(Message, long);这个方法，所以可见boolean post(Runnable r)这个方法最终还是走到上面说到的send的流程中
最终调用boolean enqueueMessage(MessageQueue queue, Message msg, long uptimeMillis)。
3.1.1 Message getPostMessage(Runnable r)方法
代码在Handler.java 725行
private static Message getPostMessage(Runnable r) {
Message m = Message.obtain();
m.callback = r;
return m;
}
代码很简单，主要是做了两件事
- 通过Message.obtain()从消息对象池中获取一个空的Message
- 将这空的Message的callback变量指向Runnable
最后返回这个Message m。
3.1.2 小结
- boolean post(Runnable r)方法的内置也是通过Message.obtain()来获取一个Message对象m，然后仅仅把m的callback指向参数r而已。
- 最后最终通过调用send方案的某个流程最终调用到boolean enqueueMessage(MessageQueue queue, Message msg, long uptimeMillis)
3.2 boolean postAtTime(Runnable r, long uptimeMillis)方法
代码在Handler.java 347行
/**
* Causes the Runnable r to be added to the message queue, to be run
* at a specific time given by uptimeMillis.
* The time-base is {@link android.os.SystemClock#uptimeMillis}.
* Time spent in deep sleep will add an additional delay to execution.
* The runnable will be run on the thread to which this handler is attached.
*
* @param r The Runnable that will be executed.
* @param uptimeMillis The absolute time at which the callback should run,
* using the {@link android.os.SystemClock#uptimeMillis} time-base.
*
* @return Returns true if the Runnable was successfully placed in to the
* message queue. Returns false on failure, usually because the
* looper processing the message queue is exiting. Note that a
* result of true does not mean the Runnable will be processed -- if
* the looper is quit before the delivery time of the message
* occurs then the message will be dropped.
*/
public final boolean postAtTime(Runnable r, long uptimeMillis)
{
return sendMessageAtTime(getPostMessage(r), uptimeMillis);
}
- 将一个Runnable添加到消息队列中，这个runnable将会一个特定的时间被执行，这个时间是以android.os.SystemClock.uptimeMillis()为基准。
- 如果在深度睡眠下，会推迟执行的时间，这个Runnable将会在和当前Hander关联的线程中被执行。
- 方法内部也是先是调用getPostMessage(Runnable)来获取一个Message，这个Message的callback字段指向了这个Runnable，然后调用sendMessageAtTime(Message,long)。
3.3 boolean postAtTime(Runnable r, Object token, long uptimeMillis)方法
代码在Handler.java 372行
/**
* Causes the Runnable r to be added to the message queue, to be run
* at a specific time given by uptimeMillis.
* The time-base is {@link android.os.SystemClock#uptimeMillis}.
* Time spent in deep sleep will add an additional delay to execution.
* The runnable will be run on the thread to which this handler is attached.
*
* @param r The Runnable that will be executed.
* @param uptimeMillis The absolute time at which the callback should run,
* using the {@link android.os.SystemClock#uptimeMillis} time-base.
*
* @return Returns true if the Runnable was successfully placed in to the
* message queue. Returns false on failure, usually because the
* looper processing the message queue is exiting. Note that a
* result of true does not mean the Runnable will be processed -- if
* the looper is quit before the delivery time of the message
* occurs then the message will be dropped.
*
* @see android.os.SystemClock#uptimeMillis
*/
public final boolean postAtTime(Runnable r, Object token, long uptimeMillis)
{
return sendMessageAtTime(getPostMessage(r, token), uptimeMillis);
}
- 这个方法和上个方法唯一不同就是多了Object参数，而这个参数仅仅是把Message.obtain();获取的Message的obj字段的指向第二个入参token而已。最后也是调用sendMessageAtTime(Message,long)。
2.4 boolean postDelayed(Runnable r, long delayMillis)方法
代码在Handler.java 396行
/**
* Causes the Runnable r to be added to the message queue, to be run
* after the specified amount of time elapses.
* The runnable will be run on the thread to which this handler
* is attached.
* The time-base is {@link android.os.SystemClock#uptimeMillis}.
* Time spent in deep sleep will add an additional delay to execution.
*
* @param r The Runnable that will be executed.
* @param delayMillis The delay (in milliseconds) until the Runnable
* will be executed.
*
* @return Returns true if the Runnable was successfully placed in to the
* message queue. Returns false on failure, usually because the
* looper processing the message queue is exiting. Note that a
* result of true does not mean the Runnable will be processed --
* if the looper is quit before the delivery time of the message
* occurs then the message will be dropped.
*/
public final boolean postDelayed(Runnable r, long delayMillis)
{
return sendMessageDelayed(getPostMessage(r), delayMillis);
}
这个方法也很简单，就是依旧是通过getPostMessage(Runnable)来获取一个Message，最后调用sendMessageDelayed(Message,long)而已。
3.5 boolean postAtFrontOfQueue(Runnable r)方法
代码在Handler.java 416行
/**
* Posts a message to an object that implements Runnable.
* Causes the Runnable r to executed on the next iteration through the
* message queue. The runnable will be run on the thread to which this
* handler is attached.
* This method is only for use in very special circumstances -- it
* can easily starve the message queue, cause ordering problems, or have
* other unexpected side-effects.
*
* @param r The Runnable that will be executed.
*
* @return Returns true if the message was successfully placed in to the
* message queue. Returns false on failure, usually because the
* looper processing the message queue is exiting.
*/
public final boolean postAtFrontOfQueue(Runnable r)
{
return sendMessageAtFrontOfQueue(getPostMessage(r));
}
就是依旧是通过getPostMessage(Runnable)来获取一个Message，最后调用sendMessageAtFrontOfQueue(Message)而已。
3.6 小结
- Handler的post方案:都是通过Message getPostMessage(Runnable )中调用Message m = Message.obtain();来获取一个空的Message，然后把这个Message的callback变量指向了Runnable，最终调用相应的send方案而已。
所以我们可以这样说:
Handler的发送消息(障栅除外)，无论是通过send方案还是pos方案最终都会做走到 boolean enqueueMessage(MessageQueue queue, Message msg, long uptimeMillis) 中
5713484-5d5c06c67b925a3d.png
参考
Android Handler机制7之消息发送