温故知新:探究Android UI 绘制刷新流程

一、说明:

  1. 基于之前的了解知道ui的绘制最终会走到Android的ViewRootImplscheduleTraversals进行发送接收vsync信号绘制,在ViewRootImpl中还会进行主线程检测,也就是我们所谓子线程更新ui会抛出异常。可看activity启动流程参考:juejin.cn/post/708344...

  2. 像我们常用的刷新ui,invalidate,requestLayout方法,(按我之前的理解在ViewRootImpl初始化添加后,在子线程中刷新ui一定会崩溃:如下图)

二、问题:invalidate一定会导致异常崩溃?

2.1、例子:子线程更新TextView文本(注意这里是TextView,为什么是它而不是ImageView,因为我的背景就是使用的TextView,使用它的时候发现了invalidate,requestLayout方法的区别 )

某天我在onResume中利用子线程更新了TextView的一段代码,发现并没有抛出异常崩溃,代码如下:

kotlin 复制代码
 override fun onResume() {
        super.onResume()
        mBind.btTest.setOnClickListener{
            lifecycleScope.launch(Dispatchers.IO) {
                mBind.btTest.text = "子线程点击改变:${Thread.currentThread().name}"
            }
        }
    }

我在想为什么呢? ,看代码: 一步步debug:TextView控件中:

typescript 复制代码
1.、
public final void setText(CharSequence text) {
    setText(text, mBufferType);
}
2.、
public void setText(CharSequence text, BufferType type) {
    setText(text, type, true, 0);

    if (mCharWrapper != null) {
        mCharWrapper.mChars = null;
    }
}
3.、
private void setText(CharSequence text, BufferType type,
                     boolean notifyBefore, int oldlen) {
     ...省略
    if (mLayout != null) {
        checkForRelayout();
    }
    ...省略
    
}

以上主要看第三步中的 checkForRelayout()检测是否需要重绘,方法如下

scss 复制代码
@UnsupportedAppUsage
private void checkForRelayout() {
    // If we have a fixed width, we can just swap in a new text layout
    // if the text height stays the same or if the view height is fixed.

    if ((mLayoutParams.width != LayoutParams.WRAP_CONTENT
            || (mMaxWidthMode == mMinWidthMode && mMaxWidth == mMinWidth))
            && (mHint == null || mHintLayout != null)
            && (mRight - mLeft - getCompoundPaddingLeft() - getCompoundPaddingRight() > 0)) {
        // Static width, so try making a new text layout.

        int oldht = mLayout.getHeight();
        int want = mLayout.getWidth();
        int hintWant = mHintLayout == null ? 0 : mHintLayout.getWidth();

        /*
         * No need to bring the text into view, since the size is not
         * changing (unless we do the requestLayout(), in which case it
         * will happen at measure).
         */
        makeNewLayout(want, hintWant, UNKNOWN_BORING, UNKNOWN_BORING,
                      mRight - mLeft - getCompoundPaddingLeft() - getCompoundPaddingRight(),
                      false);
        
        //1.检测文本的显示类型,就是我们的过长省略号这种
        if (mEllipsize != TextUtils.TruncateAt.MARQUEE) {
            // In a fixed-height view, so use our new text layout.
            if (mLayoutParams.height != LayoutParams.WRAP_CONTENT
                    && mLayoutParams.height != LayoutParams.MATCH_PARENT) {
                autoSizeText();
                invalidate();
                return;
            }

            // Dynamic height, but height has stayed the same,
            // so use our new text layout.
            if (mLayout.getHeight() == oldht
                    && (mHintLayout == null || mHintLayout.getHeight() == oldht)) {
                autoSizeText();
                invalidate();
                return;
            }
        }

        // We lose: the height has changed and we have a dynamic height.
        // Request a new view layout using our new text layout.
        requestLayout();
        invalidate();
    } else {
        // Dynamic width, so we have no choice but to request a new
        // view layout with a new text layout.
        nullLayouts();
        requestLayout();
        invalidate();
    }
}

从上面的checkForRelayout()方法中的if (mEllipsize != TextUtils.TruncateAt.MARQUEE)条件知道成立,因为我们没有设置过mEllipsize = 跑马灯效果,所以走了invalidate()方法然后直接return截断,不会走后面的requestLayout()方法,至于requestLayout() 与 invalidate()的区别我就不讲了,不是本文的重点,感兴趣可以参考:juejin.cn/post/701745...

2.2、分析requestLayout方法

基于之前的知识我知道调用requestLayout()方法会崩溃,至于为什么调用requestLayout()方法会崩溃?

我们先看requestLayout()方法,暂停一会invalidate()的跟进

requestLayout()方法代码如下:

kotlin 复制代码
public void requestLayout() {
    if (mMeasureCache != null) mMeasureCache.clear();

    if (mAttachInfo != null && mAttachInfo.mViewRequestingLayout == null) {
        // Only trigger request-during-layout logic if this is the view requesting it,
        // not the views in its parent hierarchy
        ViewRootImpl viewRoot = getViewRootImpl();
        if (viewRoot != null && viewRoot.isInLayout()) {
            if (!viewRoot.requestLayoutDuringLayout(this)) {
                return;
            }
        }
        mAttachInfo.mViewRequestingLayout = this;
    }

    mPrivateFlags |= PFLAG_FORCE_LAYOUT;
    mPrivateFlags |= PFLAG_INVALIDATED;

    if (mParent != null && !mParent.isLayoutRequested()) {
        mParent.requestLayout();
    }
    if (mAttachInfo != null && mAttachInfo.mViewRequestingLayout == this) {
        mAttachInfo.mViewRequestingLayout = null;
    }
}

requestLayout()方法中会循环递归调用 mParent.requestLayout()方法,直到找到ViewRootImpl中的requestLayout()方法,而它的方法做了线程检测如下图:这就是requestLayout()方法会崩溃的原因。

验证猜想:TextView设置跑马灯属性,使上面的if (mEllipsize != TextUtils.TruncateAt.MARQUEE)不成立,走下面的requestLayout()方法,代码如下:

kotlin 复制代码
 override fun onResume() {
        super.onResume()
        mBind.btTest.ellipsize = TextUtils.TruncateAt.valueOf("MARQUEE")
        mBind.btTest.setOnClickListener{
            lifecycleScope.launch(Dispatchers.IO) {
                mBind.btTest.text = "子线程点击改变:${Thread.currentThread().name}"
            }
        }
    }

果然点击后崩溃:

2.3、继续分析invalidate()方法,为什么不会导致textview的更新崩溃

看代码在View.java文件中

csharp 复制代码
public void invalidate() {
    invalidate(true);
}
typescript 复制代码
public void invalidate(boolean invalidateCache) {
    invalidateInternal(0, 0, mRight - mLeft, mBottom - mTop, invalidateCache, true);
}
java 复制代码
void invalidateInternal(int l, int t, int r, int b, boolean invalidateCache,
        boolean fullInvalidate) {
    if (mGhostView != null) {
        mGhostView.invalidate(true);
        return;
    }

    if (skipInvalidate()) {
        return;
    }

    // Reset content capture caches
    mPrivateFlags4 &= ~PFLAG4_CONTENT_CAPTURE_IMPORTANCE_MASK;
    mContentCaptureSessionCached = false;

    if ((mPrivateFlags & (PFLAG_DRAWN | PFLAG_HAS_BOUNDS)) == (PFLAG_DRAWN | PFLAG_HAS_BOUNDS)
            || (invalidateCache && (mPrivateFlags & PFLAG_DRAWING_CACHE_VALID) == PFLAG_DRAWING_CACHE_VALID)
            || (mPrivateFlags & PFLAG_INVALIDATED) != PFLAG_INVALIDATED
            || (fullInvalidate && isOpaque() != mLastIsOpaque)) {
        if (fullInvalidate) {
            mLastIsOpaque = isOpaque();
            mPrivateFlags &= ~PFLAG_DRAWN;
        }

        mPrivateFlags |= PFLAG_DIRTY;

        if (invalidateCache) {
            mPrivateFlags |= PFLAG_INVALIDATED;
            mPrivateFlags &= ~PFLAG_DRAWING_CACHE_VALID;
        }

        // Propagate the damage rectangle to the parent view.
        final AttachInfo ai = mAttachInfo;
        final ViewParent p = mParent;
        if (p != null && ai != null && l < r && t < b) {
            final Rect damage = ai.mTmpInvalRect;
            damage.set(l, t, r, b);
            p.invalidateChild(this, damage);
        }

        // Damage the entire projection receiver, if necessary.
        if (mBackground != null && mBackground.isProjected()) {
            final View receiver = getProjectionReceiver();
            if (receiver != null) {
                receiver.damageInParent();
            }
        }
    }
}

核心 代码是上面第三段invalidateInternal方法中的invalidateChild方法

它回调到ViewGroup中的invalidateChild方法

看:invalidateChild如下图:我们知道

if (attachInfo != null && attachInfo.mHardwareAccelerated) 条件成立attachInfo不为空 ,且硬件加速被开启(从API 14 (3.0)起。硬件加速默认开启)。 attachInfo 是一个view在attach至其父window被赋值的一系列信息。参考:juejin.cn/post/710110...

所以条件成立后走的onDescendantInvalidated方法 如下:

less 复制代码
@CallSuper
public void onDescendantInvalidated(@NonNull View child, @NonNull View target) {
    /*
     * HW-only, Rect-ignoring damage codepath
     *
     * We don't deal with rectangles here, since RenderThread native code computes damage for
     * everything drawn by HWUI (and SW layer / drawing cache doesn't keep track of damage area)
     */

    // if set, combine the animation flag into the parent
    mPrivateFlags |= (target.mPrivateFlags & PFLAG_DRAW_ANIMATION);

    if ((target.mPrivateFlags & ~PFLAG_DIRTY_MASK) != 0) {
        // We lazily use PFLAG_DIRTY, since computing opaque isn't worth the potential
        // optimization in provides in a DisplayList world.
        mPrivateFlags = (mPrivateFlags & ~PFLAG_DIRTY_MASK) | PFLAG_DIRTY;

        // simplified invalidateChildInParent behavior: clear cache validity to be safe...
        mPrivateFlags &= ~PFLAG_DRAWING_CACHE_VALID;
    }

    // ... and mark inval if in software layer that needs to repaint (hw handled in native)
    if (mLayerType == LAYER_TYPE_SOFTWARE) {
        // Layered parents should be invalidated. Escalate to a full invalidate (and note that
        // we do this after consuming any relevant flags from the originating descendant)
        mPrivateFlags |= PFLAG_INVALIDATED | PFLAG_DIRTY;
        target = this;
    }

    if (mParent != null) {
        mParent.onDescendantInvalidated(this, target);
    }
}

上面一段代码核心是 mParent.onDescendantInvalidated(this, target); 类似于requestLayout()方法 onDescendantInvalidated中会循环递归调用 mParent.onDescendantInvalidated(this, target);方法,直到找到ViewRootImpl中的onDescendantInvalidated(this, target)方法,而它的方法没做线程检测如下图:这就是开了硬件加速后invalidate方法不会崩溃的原因。如下图:直接走scheduleTraversals绘制刷新有兴趣可看:juejin.cn/post/686375...

而关闭硬件加速后会怎样呢? 继续看invalidateChild方法

java 复制代码
@Deprecated
@Override
public final void invalidateChild(View child, final Rect dirty) {
    final AttachInfo attachInfo = mAttachInfo;
    if (attachInfo != null && attachInfo.mHardwareAccelerated) {
        // HW accelerated fast path
        onDescendantInvalidated(child, child);
        return;
    }

    ViewParent parent = this;
    if (attachInfo != null) {
     
       ...

        do {
             ....
            parent = parent.invalidateChildInParent(location, dirty);
               
            ....
        } while (parent != null);
    }
}

上面一段核心是 parent = parent.invalidateChildInParent(location, dirty);方法 同理while循环不停调用 invalidateChildInParent方法直到找到ViewRootImpl中的invalidateChildInParent(int[] location, Rect dirty)方法,如下图内部进行了线程检测

**验证猜想关闭硬件加速:android:hardwareAccelerated="false"**果然崩溃了。

三、总结

这就是我遇到的问题:单纯的根据TextView在子线程可以更新得出的结论,总的来说要想不崩溃还得绕过ViewRootImpl中的checkThread的检测。至于研究它有什么用,只有知道理解源码的流程,才能写出更好的东西。

参考:

juejin.cn/post/710110...

juejin.cn/post/701745...

juejin.cn/post/686375...

相关推荐
码农爱java6 小时前
设计模式--抽象工厂模式【创建型模式】
java·设计模式·面试·抽象工厂模式·原理·23种设计模式·java 设计模式
Jiude6 小时前
算法题题解记录——双变量问题的 “枚举右,维护左”
python·算法·面试
长安051112 小时前
面试经典题目:LeetCode134_加油站
c++·算法·面试
正在绘制中13 小时前
Java重要面试名词整理(一):MySQL&JVM&Tomcat
java·开发语言·面试
沉默王二14 小时前
虾皮开的很高,还有签字费。
后端·面试·github
Do14 小时前
时间系列三:实现毫秒级倒计时
前端·javascript·面试
小蜗牛慢慢爬行14 小时前
有关异步场景的 10 大 Spring Boot 面试问题
java·开发语言·网络·spring boot·后端·spring·面试
ThisIsClark15 小时前
【后端面试总结】MySQL主从复制逻辑的技术介绍
mysql·面试·职场和发展
程序猿进阶16 小时前
深入解析 Spring WebFlux:原理与应用
java·开发语言·后端·spring·面试·架构·springboot
LCG元1 天前
【面试问题】JIT 是什么?和 JVM 什么关系?
面试·职场和发展