当前位置:网站首页>再探Handler(下)(Handler核心原理最全解析)
再探Handler(下)(Handler核心原理最全解析)
2022-06-27 03:13:00 【AD钙奶-lalala】
再探Handler(上)(Handler核心原理最全解析)_AD钙奶-lalala的博客-CSDN博客
我们都知道可以在主线程直接创建Handler,那么问题来了:我们可以在子线程创建一个Handler吗?如何去做呢?
方式1:
public class MainActivity extends AppCompatActivity {
@Override
protected void onCreate(Bundle savedInstanceState) {
super.onCreate(savedInstanceState);
setContentView(R.layout.activity_main);
test();
try {
Thread.sleep(3000);
}catch (Exception e){
e.printStackTrace();
}
Message msg = Message.obtain();
msg.what = 1;
msg.obj = "lzy";
mHandler.sendMessageDelayed(msg, 2000);
}
Handler mHandler;
@SuppressLint("HandlerLeak")
public void test() {
new Thread(new Runnable() {
@Override
public void run() {
Looper.prepare();
mHandler = new Handler() {
@Override
public void handleMessage(@NonNull Message msg) {
Toast.makeText(MainActivity.this,
msg.obj.toString(),
Toast.LENGTH_LONG).show();
super.handleMessage(msg);
}
};
Looper.loop();
}
}).start();
}
}这种方式肯定是可以实现主线程到子线程的通信的,但是这种方式很不友好。首先我们不确定handler对象何时创建好,第二个这个handler只能用在一个地方。那我们就应该思考如何去改进。
系统其实给我们提供了一个类HandlerThread,我们可以参考这个类来优化我们的代码。
方式2:
public class LzyHandlerThread extends Thread {
Looper mLooper;
public LzyHandlerThread(String name) {
super(name);
}
@Override
public void run() {
Looper.prepare();
synchronized (this){
mLooper = Looper.myLooper();
notifyAll();
}
Looper.loop();
}
public Looper getLooper() {
if (!isAlive()) {
return null;
}
synchronized (this) {
while (isAlive() && mLooper == null) {
try {
wait();
} catch (InterruptedException e) {
e.printStackTrace();
}
}
}
return mLooper;
}
}我们来思考一下,这个类为什么要这样设计?首先我们要明白notifyAll()和wait()必须要都必须使用在synchronized包住的同步代码块或者同步方法之中。
- wait():一旦执行此方法,当前线程就会进入阻塞状态,并且释放同步锁;
- notifyAll():一旦执行此方法,就会唤醒所有被wait()阻塞的线程。
看下使用:
public class MainActivity extends AppCompatActivity {
@Override
protected void onCreate(Bundle savedInstanceState) {
super.onCreate(savedInstanceState);
setContentView(R.layout.activity_main);
LzyHandlerThread lzyHandlerThread = new LzyHandlerThread("子线程1");
lzyHandlerThread.start();
Handler handler = new Handler(lzyHandlerThread.getLooper()) {
@Override
public void handleMessage(@NonNull Message msg) {
Log.e("lzy",Thread.currentThread().getName());
super.handleMessage(msg);
}
};
Message msg = Message.obtain();
msg.obj = "hello_k";
handler.sendMessageDelayed(msg, 5000);
}
}打印:
2022-06-23 21:03:24.044 6312-6333/com.example.lzyhandler E/lzy: 子线程1我们回过头来思考一下设计的时候为什么需要用到notifyAll和wait?
其实主要是由于并发问题,调用线程start方法后就回去执行run方法,随后初始化Handler需要传入Looper对象,而run方法的执行和Handler的初始化是在两个线程里面执行的,也就是说getLooper方法是执行在主线程的,run是执行在我们创建的名为子线程1的线程的。这样我们就无法保证mLooper在getLooper里面一定不为空。如果为空,而且主线程抢到了锁,就让子线程执行到赋值步骤阻塞。然后主线程释放锁,子线程获取锁,继续赋值操作,完成后唤醒主线程继续执行同步代码块里面的代码,注意wait不会阻塞主线程。
我们继续来思考下一个问题:子线程维护的Looper,消息队列无消息时处理方案是什么?有什么用?主线程呢?
我们来看一看我们前面设计的代码有什么隐患,当我们消息处理完毕后,Looper.loop我们知道里面是一个死循环,这样的话,MessageQueue <- Looper <- Thread <- Handler <- MainActivity这一条引用链就不会断,造成内存泄漏。那我们该如何去处理这个问题呢?
我们可以参考HandlerThread的源码,我们注意到这样一个方法:
public boolean quit() {
Looper looper = getLooper();
if (looper != null) {
looper.quit();
return true;
}
return false;
}我们再去Looper源码里面去看quit方法:
public void quit() {
mQueue.quit(false);
}再深入MessageQueue里面看quit方法:
void quit(boolean safe) {
if (!mQuitAllowed) {
throw new IllegalStateException("Main thread not allowed to quit.");
}
synchronized (this) {
if (mQuitting) {
return;
}
mQuitting = true;
if (safe) {
removeAllFutureMessagesLocked();
} else {
removeAllMessagesLocked();
}
// We can assume mPtr != 0 because mQuitting was previously false.
nativeWake(mPtr);
}
}
再来看Looper的loop方法:
public static void loop() {
···
for (;;) {
Message msg = queue.next();
if (msg == null) {
// No message indicates that the message queue is quitting.
return;
}
...
}
}我们再看看MessageQueue的next方法,我这里只挑重点:
Message next() {
...
if (mQuitting) {
dispose();
return null;
}
...
}调用MessageQueue方法后会将mQuiting设置为true,这样next就会返回null,loop死循环就会跳出。主线程的Looper里面的死循环是不能退出的,退出了程序也就没了。
我们都知道Handler是可以发延迟消息的,那么问题又来了:Handler是如何处理发送延时消息的呢?
我们还是来看源码:
public final boolean sendMessageDelayed(Message msg, long delayMillis)
{
if (delayMillis < 0) {
delayMillis = 0;
}
return sendMessageAtTime(msg, SystemClock.uptimeMillis() + delayMillis);
}最终延时的时间会换算成一个具体的时间。最终会走到MessageQueue的enqueueMesasge方法(不理解调用流程的去看上一篇文章):
boolean enqueueMessage(Message msg, long when) {
···
synchronized (this) {
···
msg.markInUse();
msg.when = when;
Message p = mMessages;
boolean needWake;
if (p == null || when == 0 || when < p.when) {
msg.next = p;
mMessages = msg;
needWake = mBlocked;
} else {
needWake = mBlocked && p.target == null && msg.isAsynchronous();
Message prev;
for (;;) {
prev = p;
p = p.next;
if (p == null || when < p.when) {
break;
}
if (needWake && p.isAsynchronous()) {
needWake = false;
}
}
msg.next = p; // invariant: p == prev.next
prev.next = msg;
}
if (needWake) {
nativeWake(mPtr);
}
}
return true;
}
注意:如果when < p.when,说明什么?很明显了,就是新插入的消息执行时间小于链表第一个消息的时间,这个时候将新消息插入链表头。if里面的代码就是这个意思,能看懂吧,看不懂的赶紧回去补下链表的数据结构。简而言之:插入消息的时候已经按时间顺序排列好了!
那么我们又有疑惑,如果链表头的消息执行时间仍然在后面不在当前该如何处理呢?这个时候就需要看一下去消息的函数,上MessageQueue的next方法:
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.
final long ptr = mPtr;
if (ptr == 0) {
return null;
}
int pendingIdleHandlerCount = -1; // -1 only during first iteration
int nextPollTimeoutMillis = 0;
for (;;) {
if (nextPollTimeoutMillis != 0) {
Binder.flushPendingCommands();
}
nativePollOnce(ptr, nextPollTimeoutMillis);
synchronized (this) {
// Try to retrieve the next message. Return if found.
final long now = SystemClock.uptimeMillis();
Message prevMsg = null;
Message msg = mMessages;
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());
}
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;
}
// Process the quit message now that all pending messages have been handled.
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.
if (pendingIdleHandlerCount < 0
&& (mMessages == null || now < mMessages.when)) {
pendingIdleHandlerCount = mIdleHandlers.size();
}
if (pendingIdleHandlerCount <= 0) {
// No idle handlers to run. Loop and wait some more.
mBlocked = true;
continue;
}
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.
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.
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.
nextPollTimeoutMillis = 0;
}
}这个方法很长,我们只需要关注我们需要关注的核心点,我们看这几行代码:
if (now < msg.when) {
nextPollTimeoutMillis = (int) Math.min(msg.when - now, Integer.MAX_VALUE);
}老哥是不是见笑了,原来谷歌的代码也是如此朴实无华,当前时间小于消息执行时间的话,计算时间差。这一次循环并不会返回msg,也不会退出循环,进入下一次循环。注意:
nativePollOnce(ptr, nextPollTimeoutMillis);
//注:
private native void nativePollOnce(long ptr, int timeoutMillis); 这次一个native方法,什么意思呢,就是等一段时间再执行呗,底层运用了epoll机制。这里就不细说了,有时间单独讲。
还记得我们的demo里面Message是如何创建的吗?Message.obtain,我们为什么不直接new 一个Message?
我们知道,整个主线程的运行是基于Looper的loop方法的,届时会有大量的消息被创建,如果没有复用机制,将会频繁的触发GC!而GC触发的时候,是会暂停进程中所有线程的。频繁GC必然会导致卡顿!
我们再来看看obtain到底做了怎样的优化:
public static Message obtain() {
synchronized (sPoolSync) {
if (sPool != null) {
Message m = sPool;
sPool = m.next;
m.next = null;
m.flags = 0; // clear in-use flag
sPoolSize--;
return m;
}
}
return new Message();
}很明显:设计了一个缓存消息链表,如果缓存链表有对象,直接取出来一个。
最后一个问题了:Handler在没有消息处理时时阻塞的还是非阻塞的?为什么不会出现ANR?
我们还是来看源码:
if (pendingIdleHandlerCount <= 0) {
// No idle handlers to run. Loop and wait some more.
mBlocked = true;
continue;
}没有消息的时候,Looper的loop方法里面这一行代码:
Message msg = queue.next(); // might block会阻塞。ANR出现的原因是定时器机制,跟Handler没什么关系。比如说在主线程执行耗时操作,主线程一直在执行耗时操作而没有办法干别的,比如说刷新UI啥的,很多系统消息无法处理,定时器就会及时报出ANR的错误提醒。handler如果是空闲状态,说明没有任何消息需要处理,如果有消息了,阻塞就会消失,又怎么会ANR呢?
边栏推荐
- 清华&华为等 综述 | 语义通信:原则与挑战
- 流沙画模拟器源码
- Anaconda3安装过程及安装后缺失大量文件,没有scripts等目录
- servlet与JSP期末复习考点梳理 42问42答
- 正则表达式:语法
- 2021:Zero-shot Visual Question Answering using Knowledge Graphs使用知识图的零次视觉问答
- 超级详细,2 万字详解,吃透 ES!
- 人群模拟
- 2021:Passage Retrieval for Outside-KnowledgeVisual Question Answering通道检索的外部知识视觉问答
- Super détaillé, 20 000 caractères détaillés, mangez à travers es!
猜你喜欢

Yuantou firm offer weekly record 20220627

Mmdetection uses yolox to train its own coco data set

Pat class a 1024 palindromic number

Anaconda3 is missing a large number of files during and after installation, and there are no scripts and other directories

pytorch 22 8种Dropout方法的简介 及 基于Dropout用4行代码快速实现DropBlock

平均风向风速计算(单位矢量法)

Flink学习2:应用场景

Quicksand painting simulator source code

STM32入门介绍

2021:AdaVQA: Overcoming Language Priors with Adapted Margin Cosine Loss∗自适应的边缘余弦损失解决语言先验
随机推荐
2021:Passage Retrieval for Outside-KnowledgeVisual Question Answering通道检索的外部知识视觉问答
Mmdetection valueerror: need at least one array to concatenate solution
【一起上水硕系列】Day 6
JWT certification process and use cases
Career outlook, money outlook and happiness outlook
How to solve the problem of low applet utilization
I found a JSON visualization tool artifact. I love it!
IDEA中好用的插件
DAMA、DCMM等数据管理框架各个能力域的划分是否合理?有内在逻辑吗?
2019LXMERT:Learning Cross-Modality Encoder Representations from Transformers
paddlepaddle 21 基于dropout实现用4行代码dropblock
[micro service sentinel] degradation rules slow call proportion abnormal proportion abnormal constant
mmdetection ValueError: need at least one array to concatenate解决方案
Regular expressions: Syntax
PAT甲级 1026 Table Tennis
Cvpr2022 | pointdistiller: structured knowledge distillation for efficient and compact 3D detection
pytorch 22 8种Dropout方法的简介 及 基于Dropout用4行代码快速实现DropBlock
PAT甲级 1024 Palindromic Number
SQLite reader plug-in tests SQLite syntax
超级详细,2 万字详解,吃透 ES!