Android源碼分析(Handler機(jī)制)

源碼基于安卓8.0分析結(jié)果

關(guān)鍵類ActivityThread、Handler、Looper、Message、MessageQueue

  • ActivityThread中的流程:應(yīng)用程序入口是在ActivityThread的main方法中,程序啟動(dòng),底層去調(diào)用C/C++去調(diào)用main方法
  • ActivityThread中的main的方法
 /*
將當(dāng)前線程初始化為一個(gè)活套,將其標(biāo)記為
*應(yīng)用程序的主要活套。應(yīng)用程序的主要套接字
*是由Android環(huán)境創(chuàng)建的,所以您永遠(yuǎn)不需要
*自己調(diào)用這個(gè)函數(shù)
 */
 public static void main(String[] args) {
        Looper.prepareMainLooper();
        ActivityThread thread = new ActivityThread();
        thread.attach(false);
        if (sMainThreadHandler == null) { sMainThreadHandler = thread.getHandler(); }
        //if(false){}之類的語句,這種寫法是方便調(diào)試的,通過一個(gè)標(biāo)志就可以控制某些代碼是否執(zhí)行,比如說是否輸出一些系統(tǒng)的Log
        if (false) { myLooper().setMessageLogging(new LogPrinter(Log.DEBUG, "ActivityThread")); }
        Looper.loop();
}
  • Looper.prepareMainLooper();方法
    ////Looper的prepare方法,并且關(guān)聯(lián)到主線程
    public static void prepareMainLooper() {
        //Only one Looper may be created per thread"
        // false意思不允許我們程序員退出(面向我們開發(fā)者),因?yàn)檫@是在主線程里面
        // TODO: 2018/5/17   
        prepare(false);
        synchronized (Looper.class) {
            if (sMainLooper != null) {
                throw new IllegalStateException("The main Looper has already been prepared.");
            }
            //把Looper設(shè)置為主線程的Looper
            sMainLooper = myLooper();
        }
    }
  • 關(guān)于prepare(false)方法: Only one Looper may be created per thread 也就是說,一個(gè)線程只有一個(gè)Looper對象,要不然會(huì)拋出異常
   private static void prepare(boolean quitAllowed) {
        if (sThreadLocal.get() != null) {
            throw new RuntimeException("Only one Looper may be created per thread");
        }
        sThreadLocal.set(new Looper(quitAllowed));
    }
  • 關(guān)于ThreadLocal的set方法,可以找到ThreadLocal的構(gòu)造函數(shù),底層的實(shí)現(xiàn)是一個(gè)Entry的數(shù)組.
 ThreadLocal<Looper> sThreadLocal = new ThreadLocal<Looper>();
 //1、ThreadLocal的set方法
  public void set(T value) {
        Thread t = Thread.currentThread();
        ThreadLocalMap map = getMap(t);
        if (map != null)
            map.set(this, value);
        else
            createMap(t, value);
    }
 //2、ThreadLocal的createMap方法
  void createMap(Thread t, T firstValue) {
        t.threadLocals = new ThreadLocalMap(this, firstValue);
    }

ThreadLocalMap的構(gòu)造函數(shù)里面有一個(gè)長度為16的Entry的數(shù)組,當(dāng)然這個(gè)機(jī)制和HashMap差不多,也有擴(kuò)容機(jī)制,就是當(dāng)容器裝不下了,在此的基礎(chǔ)上增加一倍的長度,同時(shí)把原來的數(shù)據(jù)copy到新的Entry數(shù)組中

 //3、ThreadLocal的構(gòu)造函數(shù)
     ThreadLocalMap(ThreadLocal<?> firstKey, Object firstValue) {
            table = new Entry[INITIAL_CAPACITY];
            int i = firstKey.threadLocalHashCode & (INITIAL_CAPACITY - 1);
            table[i] = new Entry(firstKey, firstValue);
            size = 1;
            setThreshold(INITIAL_CAPACITY);
        }

ThreadLocalMap的(擴(kuò)容機(jī)制)Double the capacity of the table.

      /**
         * Double the capacity of the table.
         */
        private void resize() {
            Entry[] oldTab = table;
            int oldLen = oldTab.length;
            int newLen = oldLen * 2;
            Entry[] newTab = new Entry[newLen];
            int count = 0;

            for (int j = 0; j < oldLen; ++j) {
                Entry e = oldTab[j];
                if (e != null) {
                    ThreadLocal<?> k = e.get();
                    if (k == null) {
                        e.value = null; // Help the GC
                    } else {
                        int h = k.threadLocalHashCode & (newLen - 1);
                        while (newTab[h] != null)
                            h = nextIndex(h, newLen);
                        newTab[h] = e;
                        count++;
                    }
                }
            }

            setThreshold(newLen);
            size = count;
            table = newTab;
        }
  • 同時(shí)關(guān)注Entry的類,可以發(fā)現(xiàn)這是WeakReference的子類,關(guān)系到了弱引用:弱引用是比軟引用更弱的一種的引用的類型,只有弱引用指向的對象的生命周期更短,當(dāng)垃圾回收器掃描到只有具有弱引用的對象的時(shí)候,不敢當(dāng)前空間是否不足,都會(huì)對弱引用對象進(jìn)行回收,不太明白的可以看我另外一篇文章 安卓代碼、圖片、布局、網(wǎng)絡(luò)和電量優(yōu)化
  static class Entry extends WeakReference<ThreadLocal<?>> {
            /** The value associated with this ThreadLocal. */
            Object value;
            Entry(ThreadLocal<?> k, Object v) {
                super(k);
                value = v;
            }
        }

為什么我要提起它???可能我理解的不太準(zhǔn)確,肯定不太準(zhǔn)確,在我現(xiàn)在工作中,維護(hù)和開發(fā)一個(gè)硬件的應(yīng)用早餐機(jī)(用戶通過App預(yù)定早餐,第二天早上去機(jī)器上取早餐),就像蜂巢的快遞柜一樣,取早餐的機(jī)器,在深圳工作的大佬,可能也許看見過我們的機(jī)器,PLC、安卓、物聯(lián)網(wǎng)這篇文章有詳細(xì)的介紹。在測試過程中,由于App常駐在前臺(tái),有幾率導(dǎo)致App直接掛掉,通過日志發(fā)現(xiàn)是內(nèi)存不足,直接kill了這個(gè)App,我想這里可能就是這個(gè)原因,這個(gè)一個(gè)弱應(yīng)用,只要虛擬機(jī)掃描導(dǎo)致這里了,我不管你了,我直接把你回收掉。僅僅是個(gè)人的理解,同時(shí)我們安卓的開發(fā)板也不太穩(wěn)定,如果在這一點(diǎn)有見解的大佬,歡迎討論,謝謝了

  • Looper.loop();根據(jù)我們的常識(shí)知道,如果程序沒有死循環(huán)的話,執(zhí)行完main函數(shù)(比如構(gòu)建視圖等等代碼)以后就會(huì)立馬退出了。之所以我們的APP能夠一直運(yùn)行著,就是因?yàn)長ooper.loop()里面是一個(gè)死循環(huán)
    • 1、 首先拿到Looper對象(me),如果當(dāng)前的線程沒有Looper,那么就會(huì)拋出異常, // TODO: 2018/5/17 在子線程中創(chuàng)建handler的話,需要looper也要準(zhǔn)備好 ,要不然會(huì)報(bào)錯(cuò)。這就是為什么在子線程里面創(chuàng)建Handler如果不手動(dòng)創(chuàng)建和啟動(dòng)Looper會(huì)報(bào)錯(cuò)的原因
      • 這個(gè)Looper對象就是通過sThreadLocal.get();細(xì)心的話可以發(fā)現(xiàn)前面已經(jīng)sThreadLocal.set()
   public static @Nullable Looper myLooper() {
        return sThreadLocal.get();
    }
  • 2、然后拿到Looper的成員變量MessageQueue,在MessageQueue里面不斷地去取消息,關(guān)于MessageQueue的next方法如下:如果這個(gè)msg為null的,這個(gè)結(jié)束掉這個(gè)
  • 3、msg.target.dispatchMessage(msg)就是處理消息,緊接著在loop方法的最后調(diào)用了msg.recycleUnchecked()這就是回收了Message。
  • 4、我們平時(shí)寫Handler的時(shí)候不需要我們手動(dòng)回收,因?yàn)楣雀璧墓こ處熞呀?jīng)有考慮到這方面的問題了。消息是在Handler分發(fā)處理之后就會(huì)被自動(dòng)回收的:
    public static void loop() {
        final Looper me = myLooper();
        if (me == null) {
            // TODO: 2018/5/17   在子線程中創(chuàng)建handler的話,需要looper也要準(zhǔn)備好 ,要不然會(huì)報(bào)錯(cuò)
            // 1、 首先拿到Looper對象(me),如果當(dāng)前的線程沒有Looper,那么就會(huì)拋出異常,
            // 這就是為什么在子線程里面創(chuàng)建Handler如果不手動(dòng)創(chuàng)建和啟動(dòng)Looper會(huì)報(bào)錯(cuò)的原因
            throw new RuntimeException("No Looper; Looper.prepare() wasn't called on this thread.");
        }
        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();

        for (; ; ) {
            // TODO: 2018/5/17   Message
           // 2、然后拿到Looper的成員變量MessageQueue,在MessageQueue里面不斷地去取消息,關(guān)于MessageQueue的next方法如下:
            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 slowDispatchThresholdMs = me.mSlowDispatchThresholdMs;

            final long traceTag = me.mTraceTag;
            if (traceTag != 0 && Trace.isTagEnabled(traceTag)) {
                Trace.traceBegin(traceTag, msg.target.getTraceName(msg));
            }
            final long start = (slowDispatchThresholdMs == 0) ? 0 : SystemClock.uptimeMillis();
            final long end;
          //  msg.target.dispatchMessage(msg)就是處理消息,緊接著在loop方法的最后調(diào)用了msg.recycleUnchecked()這就是回收了Message。
            // TODO: 2018/5/17
            處理消息
            try {
                處理消息
                // TODO: 2018/5/17  msg中的target 就是handler的本體的對象  ,直接去handler中發(fā)送這個(gè)對象
                msg.target.dispatchMessage(msg);
                end = (slowDispatchThresholdMs == 0) ? 0 : SystemClock.uptimeMillis();
            } finally {
                if (traceTag != 0) {
                    Trace.traceEnd(traceTag);
                }
            }
            if (slowDispatchThresholdMs > 0) {
                final long time = end - start;
                if (time > slowDispatchThresholdMs) {
                    Slog.w(TAG, "Dispatch took " + time + "ms on "
                            + Thread.currentThread().getName() + ", h=" +
                            msg.target + " cb=" + msg.callback + " msg=" + msg.what);
                }
            }

            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);
            }
            // TODO: 2018/5/17
            我們平時(shí)寫Handler的時(shí)候不需要我們手動(dòng)回收,因?yàn)楣雀璧墓こ處熞呀?jīng)有考慮到這方面的問題了。消息是在Handler分發(fā)處理之后就會(huì)被自動(dòng)回收的:
            msg.recycleUnchecked();
        }
    }

  • 關(guān)于Message類
    1、對象是實(shí)現(xiàn)了Parcelable接口的,因?yàn)镸essage消息可能需要跨進(jìn)程通信,這時(shí)候就需要進(jìn)程序列化以及反序列化操作了。
public final class Message implements Parcelable Message   

2、obtain()得到Message對象,其中的設(shè)計(jì)模式享元模式:我見過最好的Demo,理解:采用一個(gè)共享類避免大量擁有相同的內(nèi)容的“小類的開銷”
享元模式德優(yōu)缺點(diǎn):優(yōu)點(diǎn)在于大幅度的降低內(nèi)存中對象的數(shù)量,但是,它做到這一點(diǎn)代價(jià)優(yōu)點(diǎn)高,享元模式使得系統(tǒng)更加復(fù)雜為了使對象可以共享,需要將一些狀態(tài)外部化,這使得一些程序邏輯更加的復(fù)雜享元模式將享元對象的狀態(tài)外部化,而讀取外部狀態(tài)使得運(yùn)行的時(shí)間稍微變長,更多的Demo可以看這篇文章二十三種設(shè)計(jì)模式

 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();
    }

3、消息的回收機(jī)制方法一:這個(gè)方法調(diào)用的時(shí)機(jī)是在MessageQueueen中。MessageQueueen.queueMessage(Message msg, long when)

    public void recycle() {
        if (isInUse()) {
            if (gCheckRecycle) {
                throw new IllegalStateException("This message cannot be recycled because it "
                        + "is still in use.");
            }
            return;
        }
        recycleUnchecked();
    }

4、消息的回收機(jī)制方法二:調(diào)用的地方是在 Looper.loop();谷歌的工程師幫我們調(diào)用,所以我們在開發(fā)過程中,沒有去調(diào)用這個(gè)消息回收,哈哈,向谷歌致敬,同時(shí)這個(gè)方法也會(huì)在recycle中調(diào)用。

 void recycleUnchecked() {
        // Mark the message as in use while it remains in the recycled object pool.
        // Clear out all other details.
        flags = FLAG_IN_USE;
        what = 0;
        arg1 = 0;
        arg2 = 0;
        obj = null;
        replyTo = null;
        sendingUid = -1;
        when = 0;
        target = null;
        callback = null;
        data = null;

        synchronized (sPoolSync) {
            if (sPoolSize < MAX_POOL_SIZE) {
                next = sPool;
                sPool = this;
                sPoolSize++;
            }
        }
    }
  • 關(guān)于MessageQueue類
    1、 Message next() 方法:看到消息的取出用到了一些native方法,這樣做是為了獲得更高的效率,消息的去取出并不是直接就從隊(duì)列的頭部取出的,而是根據(jù)了消息的when時(shí)間參數(shù)有關(guān)的,因?yàn)槲覀兛梢园l(fā)送延時(shí)消息、也可以發(fā)送一個(gè)指定時(shí)間點(diǎn)的消息
    • 1、for循環(huán)的使用native 方法
    • 2、根據(jù)時(shí)間戳獲取消息
  Message next() {
        final long ptr = mPtr;
        if (ptr == 0) {
            return null;
        }
        int pendingIdleHandlerCount = -1; // -1 only during first iteration
        int nextPollTimeoutMillis = 0;
        //防止被反射修改了這個(gè)標(biāo)記,直接寫出for循環(huán)
        for (;;) {
            if (nextPollTimeoutMillis != 0) {
                Binder.flushPendingCommands();
            }
            //native 方法
            nativePollOnce(ptr, nextPollTimeoutMillis);

            synchronized (this) {
                // Try to retrieve the next message.  Return if found.
                //拿到當(dāng)前的時(shí)間戳
                final long now = SystemClock.uptimeMillis();
                Message prevMsg = null;
                Message msg = mMessages;
                //判斷頭指針的Target(Handler是否為空(因?yàn)轭^指針只是一個(gè)指針的作用))
                if (msg != null && msg.target == null) {
                    // Stalled by a barrier.  Find the next asynchronous message in the queue.
                    do {
                        //遍歷下一條Message
                        prevMsg = msg;
                        msg = msg.next;
                    } while (msg != null && !msg.isAsynchronous());
                }
                if (msg != null) {
                    if (now < msg.when) {
                        //還沒有到執(zhí)行的時(shí)間
                        nextPollTimeoutMillis = (int) Math.min(msg.when - now, Integer.MAX_VALUE);
                    } else {
                        //到了執(zhí)行時(shí)間,直接返回
                        mBlocked = false;
                        if (prevMsg != null) {
                            //拿出消息,斷開鏈表
                            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;
        }
    }

2、 boolean enqueueMessage(Message msg, long when)方法中調(diào)用了 msg.recycle();

  boolean enqueueMessage(Message msg, long when) {
        if (msg.target == null) {
            throw new IllegalArgumentException("Message must have a target.");
        }
        if (msg.isInUse()) {
            throw new IllegalStateException(msg + " This message is already in use.");
        }

        synchronized (this) {
            if (mQuitting) {
                IllegalStateException e = new IllegalStateException(
                        msg.target + " sending message to a Handler on a dead thread");
                Log.w(TAG, e.getMessage(), e);
                // TODO: 2018/5/21  在這里調(diào)用的 釋放消息
                msg.recycle();
                return false;
            }
}

2、quit的方法: Looper.myLooper().quit();調(diào)用的就是下面的方法 只不過safe==false

 void quit(boolean safe) {
        if (!mQuitAllowed) {
            throw new IllegalStateException("Main thread not allowed to quit.");
        }

        synchronized (this) {
            if (mQuitting) {
                return;
            }
            //置位正在退出的標(biāo)志
            mQuitting = true;
           //清空所有消息
            if (safe) {
                //安全的(系統(tǒng)的),未來未處理的消息都移除
                removeAllFutureMessagesLocked();
            } else {
                //如果是不安全的,例如我們自己定義的消息,就一次性全部移除掉
                removeAllMessagesLocked();
            }

            // We can assume mPtr != 0 because mQuitting was previously false.
            nativeWake(mPtr);
        }
    }
  • 安全的(系統(tǒng)的),未來未處理的消息都移除
 private void removeAllFutureMessagesLocked() {
        final long now = SystemClock.uptimeMillis();
        Message p = mMessages;
        if (p != null) {
            if (p.when > now) {
                //如果所有消息都處理完了,就一次性把全部消息移除掉
                removeAllMessagesLocked();
            } else {
                //否則就通過for循環(huán)拿到還沒有把還沒有執(zhí)行的Message,利用do循環(huán)
                //把這些未處理的消息通過recycleUnchecked方法回收,放回到消息池里面
                Message n;
                for (;;) {
                    n = p.next;
                    if (n == null) {
                        return;
                    }
                    if (n.when > now) {
                        break;
                    }
                    p = n;
                }
                p.next = null;
                do {
                    p = n;
                    n = p.next;
                    p.recycleUnchecked();
                } while (n != null);
            }
        }
    }
  • 如果是不安全的,例如我們自己定義的消息,就一次性全部移除掉
    private void removeAllMessagesLocked() {
        Message p = mMessages;
        while (p != null) {
            Message n = p.next;
            p.recycleUnchecked();
            p = n;
        }
        mMessages = null;
    }
  • 關(guān)于Handler發(fā)送消息流程
  • 通過一個(gè)Handler發(fā)送一個(gè)延遲5s的消息,
  innerHandler.postDelayed(new Runnable() {
            @Override
            public void run() {
                //todo
            }
        },5000);
  • 調(diào)用到Handler中的postDelayed方法
    public final boolean postDelayed(Runnable r, long delayMillis) {
        return sendMessageDelayed(getPostMessage(r), delayMillis);
    }
  • getPostMessage(r)方法:通過obtain得到一個(gè)Message的對象
   private static Message getPostMessage(Runnable r) {
        Message m = Message.obtain();
        m.callback = r;
        return m;
    }
  • sendMessageDelayed(getPostMessage(r), delayMillis):這里的delayMillis的時(shí)間小于0的話,也會(huì)為0
  public final boolean sendMessageDelayed(Message msg, long delayMillis)
    {
        if (delayMillis < 0) {
            delayMillis = 0;
        }
        return sendMessageAtTime(msg, SystemClock.uptimeMillis() + delayMillis);
    }
  • 關(guān)于sendMessageAtTime(msg, SystemClock.uptimeMillis() + delayMillis);SystemClock.uptimeMillis() 系統(tǒng)的時(shí)間返回為milliseconds==毫秒,在這個(gè)方法就可以看出,MessageQueue不可獲取
  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);
    }
  • 那么關(guān)于mQueue何時(shí)初始化的呢,請看代碼分析
    1、我們平時(shí)都是new Handler(),開始使用的
    public Handler() {
        this(null, false);
    }

直接調(diào)用了這里的方法,同時(shí)這個(gè)構(gòu)造方法是Hide了的,在外界調(diào)用不掉,為啥把它Handler,我還不太懂,反正可以注意到 mQueue = mLooper.mQueue; 原MessageQueue是在Looper中初始化的,ok,往下走

   public Handler(Callback callback, boolean async) {
        if (FIND_POTENTIAL_LEAKS) {
            final Class<? extends Handler> klass = getClass();
            if ((klass.isAnonymousClass() || klass.isMemberClass() || klass.isLocalClass()) &&
                    (klass.getModifiers() & Modifier.STATIC) == 0) {
                Log.w(TAG, "The following Handler class should be static or leaks might occur: " +
                    klass.getCanonicalName());
            }
        }

        mLooper = Looper.myLooper();
        if (mLooper == null) {
            throw new RuntimeException(
                "Can't create handler inside thread that has not called Looper.prepare()");
        }
        mQueue = mLooper.mQueue;
        mCallback = callback;
        mAsynchronous = async;
    }

Looper的構(gòu)造方法,可以看到MessageQueue初始化

   private Looper(boolean quitAllowed) {
        mQueue = new MessageQueue(quitAllowed);
        mThread = Thread.currentThread();
    }

那么Looper又在哪里初始化的呢:通過代碼可以發(fā)現(xiàn)prepare()方法中初始化,通過前面的代碼的分析又在ActivityThread中的main方法,通過調(diào)用Looper.prepareMainLooper()方法

   private static void prepare(boolean quitAllowed) {
        if (sThreadLocal.get() != null) {
            throw new RuntimeException("Only one Looper may be created per thread");
        }
        sThreadLocal.set(new Looper(quitAllowed));
    
  • 關(guān)于enqueueMessage(queue, msg, uptimeMillis);其實(shí)也就是調(diào)用到MessageQueue.enqueueMessage方法
   private boolean enqueueMessage(MessageQueue queue, Message msg, long uptimeMillis) {
        msg.target = this;
        if (mAsynchronous) {
            msg.setAsynchronous(true);
        }
        return queue.enqueueMessage(msg, uptimeMillis);
    }
  • 關(guān)于MessageQueue.enqueueMessage():對msg一些的賦值,同時(shí)呢,也調(diào)用了,本地方法,這樣性能很高,如果真的需要看懂源碼的流程,一定打個(gè)斷點(diǎn),一步步的走下去,就可以看到很良好的結(jié)果。
 boolean enqueueMessage(Message msg, long when) {
        //1、目標(biāo)為空,那么拋出異常
        if (msg.target == null) {
            throw new IllegalArgumentException("Message must have a target.");
        }
        //2、如果這個(gè)消息已經(jīng)被使用了的話,也拋出異常
        // /*package*/ boolean isInUse() {
        //        return ((flags & FLAG_IN_USE) == FLAG_IN_USE);
        //    }
        if (msg.isInUse()) {
            throw new IllegalStateException(msg + " This message is already in use.");
        }

        synchronized (this) {
            //3、如果是退出了,就是App退出了,退出了的標(biāo)記
            if (mQuitting) {
                IllegalStateException e = new IllegalStateException(
                        msg.target + " sending message to a Handler on a dead thread");
                Log.w(TAG, e.getMessage(), e);
                // TODO: 2018/5/21  在這里調(diào)用的 釋放消息
                msg.recycle();
                return false;
            }
            //4、標(biāo)記它正在使用中,
            msg.markInUse();
            //5、當(dāng)前的時(shí)間
            msg.when = when;
            Message p = mMessages;
            boolean needWake;
            if (p == null || when == 0 || when < p.when) {
                // 6、新的頭,如果阻塞,喚醒事件隊(duì)列。
                msg.next = p;
                mMessages = msg;
                needWake = mBlocked;
            } else {
                // 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.
                //插入隊(duì)列中間。通常我們不必醒來
                //增加事件隊(duì)列,除非隊(duì)列頭上有障礙物。
                //消息是隊(duì)列中最早的異步消息。
                needWake = mBlocked && p.target == null && msg.isAsynchronous();
                Message prev;
                //7、for保持消息的不斷的移動(dòng)
                for (;;) {
                    //前一個(gè)消息,如果走到這里,那么這個(gè)p不會(huì)為null
                    prev = p;
                    //把這個(gè)消息下一個(gè)賦值給P,如果下個(gè)值為null的話,就直接break
                    p = p.next;
                    if (p == null || when < p.when) {
                        break;
                    }
                    //8、需要醒來,同時(shí)消息是異步的
                    if (needWake && p.isAsynchronous()) {
                        needWake = false;
                    }
                }
                msg.next = p; // invariant: p == prev.next
                prev.next = msg;
            }

            // We can assume mPtr != 0 because mQuitting is false.
            if (needWake) {
                nativeWake(mPtr);
            }
        }
        return true;
    }
  • 這樣MessageQueue.next() 方法,就不斷的取出Message,做相應(yīng)的動(dòng)作,
  • 如何分發(fā)消息呢?還得看Looper.loop()方法
 public static void loop() {
   //前面省略了方法
    for (;;) {
            //這樣MessageQueue.next() 方法,就不斷的取出Message
            Message msg = queue.next(); // might block
            msg.target.dispatchMessage(msg);
   //省略了方法
   }

Handle system messages here. 這樣就把消息分發(fā)下去了!

    /**
     * Handle system messages here.
     */
    public void dispatchMessage(Message msg) {
        // TODO: 2018/5/17
        //這個(gè)callback呢,即使他媽的一個(gè)線程
        if (msg.callback != null) {
            handleCallback(msg);
        } else {
            //兩個(gè)都沒有的話,就去把這個(gè)消息發(fā)送到handleMessage中去
            if (mCallback != null) {
                if (mCallback.handleMessage(msg)) {
                    return;
                }
            }
            handleMessage(msg);
        }
    }
Handler機(jī)制.png
  • 說明幾點(diǎn)
    • 1、 安卓的程序的入口的函數(shù)是ActivityThread.main(),反正每個(gè)App啟動(dòng)都會(huì)經(jīng)過它,具體為啥,我也不清楚
    • 2、首先初始化的是Looper,Looper的構(gòu)造方法初始化MessageQueue,然后ThreadLocal.set()方法保存,原來是ThreadLocal里面有個(gè)ThreadLocalMap容器,底層的原理和HashMap差不多,有個(gè)初始長度為16的Entry數(shù)組,也有擴(kuò)容機(jī)制
private Looper(boolean quitAllowed) {
        mQueue = new MessageQueue(quitAllowed);
        mThread = Thread.currentThread();
    }
  • 3、由于Entry繼承的是一個(gè)WeakReference類,那么是弱應(yīng)用的子類,當(dāng)內(nèi)存不足,掃描到這里,就不被回收,導(dǎo)致App被kill,系統(tǒng)回到Launch,(當(dāng)然這僅僅是我的假設(shè),不正確),安卓系統(tǒng)不會(huì)把內(nèi)置的軟件給kill,不如說時(shí)間,主題,launch,如果要?dú)ⅲ椭荒軞㈤_發(fā)者的應(yīng)用了
  • 4、在ActivityThread.main()后,有個(gè)Looper.loop(),可以得出在:子線程中創(chuàng)建handler的話,需要looper也要準(zhǔn)備好 ,要不然會(huì)報(bào)錯(cuò)。這就是為什么在子線程里面創(chuàng)建Handler如果不手動(dòng)創(chuàng)建和啟動(dòng)Looper會(huì)報(bào)錯(cuò)的原因
   final Looper me = myLooper();
        if (me == null) {
            // TODO: 2018/5/17   在子線程中創(chuàng)建handler的話,需要looper也要準(zhǔn)備好 ,要不然會(huì)報(bào)錯(cuò)
            // 1、 首先拿到Looper對象(me),如果當(dāng)前的線程沒有Looper,那么就會(huì)拋出異常,
            // 這就是為什么在子線程里面創(chuàng)建Handler如果不手動(dòng)創(chuàng)建和啟動(dòng)Looper會(huì)報(bào)錯(cuò)的原因
            throw new RuntimeException("No Looper; Looper.prepare() wasn't called on this thread.");
        }
  • 5、程序能夠不斷的運(yùn)行著的原因,是Looper.loop中是一個(gè)死循環(huán),當(dāng)消息隊(duì)列沒有消息了,程序就會(huì)退出
  • 6、消息的分發(fā)msg.target.dispatchMessage(msg); msg.target其實(shí)就是Handler對象,可以看到分發(fā)消息的最終結(jié)果,也可以從這里表明Message的使用有好多種可能。
    public void dispatchMessage(Message msg) {
        // TODO: 2018/5/17
        //這個(gè)callback呢,即使他媽的一個(gè)線程
        if (msg.callback != null) {
            handleCallback(msg);
        } else {
            //兩個(gè)都沒有的話,就去把這個(gè)消息發(fā)送到handleMessage中去
            if (mCallback != null) {
                if (mCallback.handleMessage(msg)) {
                    return;
                }
            }
            handleMessage(msg);
        }
    }
  • 7、Message中,目前是我見過享元模式最好的實(shí)現(xiàn)的方式。更多的Demo可以看這篇文章二十三種設(shè)計(jì)模式
  • 8、為啥沒有消息回收,因?yàn)楣雀韫こ處煟呀?jīng)幫我們做了。致敬谷歌
  • 9、MessageQueue.next(),使用的是本地方法,因?yàn)樾实膯栴},需要更高的效率,所以需要本地,原理說實(shí)話,我想扯一下,看了好多文檔,發(fā)現(xiàn)我自己也不明白,反正不是我們常規(guī)的隊(duì)列中取出,而是根據(jù)when時(shí)間參數(shù)有關(guān)。
  • 10、Handler,發(fā)送消息的姿勢很多(注意是姿勢),需要不斷的嘗試,在集合源碼,就可以發(fā)現(xiàn)新大陸
  • 11、后續(xù)會(huì)講到View的繪制啊 RESUME_ACTIVITYactivity獲取焦點(diǎn),底層通過的也是Handler,在ActivityThread 內(nèi)部類H 繼承的是Handler,這里也是View繪制的開始,后續(xù)會(huì)寫一篇文章分析。
 private class H extends Handler{
        public void handleMessage(Message msg) {
            case RESUME_ACTIVITY:
                    Trace.traceBegin(Trace.TRACE_TAG_ACTIVITY_MANAGER, "activityResume");
                    SomeArgs args = (SomeArgs) msg.obj;
                    handleResumeActivity((IBinder) args.arg1, true, args.argi1 != 0, true,
          }
}
  • 12、退出程序其實(shí)就是 mLooper.myLooper().quit();
                case EXIT_APPLICATION:
                    if (mInitialApplication != null) {
                        mInitialApplication.onTerminate();
                    }
                    //退出Looper的循環(huán) 這里實(shí)際上是調(diào)用了MessageQueue的quit,清空所有Message。
                    mLooper.myLooper().quit();
                    break;
最后編輯于
?著作權(quán)歸作者所有,轉(zhuǎn)載或內(nèi)容合作請聯(lián)系作者
平臺(tái)聲明:文章內(nèi)容(如有圖片或視頻亦包括在內(nèi))由作者上傳并發(fā)布,文章內(nèi)容僅代表作者本人觀點(diǎn),簡書系信息發(fā)布平臺(tái),僅提供信息存儲(chǔ)服務(wù)。

推薦閱讀更多精彩內(nèi)容