TextView源码解析-----绘制过程

简介

看段Android官方的简介

Class Overview
Displays text to the user and optionally allows them to edit it. A TextView is a complete text editor, however the basic class is configured to not allow editing; see EditText for a subclass that configures the text view for editing.
To allow users to copy some or all of the TextView’s value and paste it somewhere else, set the XML attribute android:textIsSelectable to “true” or call setTextIsSelectable(true). The textIsSelectable flag allows users to make selection gestures in the TextView, which in turn triggers the system’s built-in copy/paste controls.

TextView主要用于给用户展示文字，并且让用户随意的可以对文字进行编辑。但是普通的TextView是不允许用来编辑的，只有EditText才可以。
如果在XML中设置了android:textIsSelectable 或者在Java代码中调用了setTextIsSelectable(true)方法，就可以允许对TextView的部分或者全部文字进行复制，然后粘贴到其他地方。textIsSelectable 标签是允许用户在TextView上使用选择手势。

顺便提下，大家如果想看API文档的话，可以在
file:///E:/AndroidEnvironment/SDK/docs/reference/android/widget/TextView.html
你安装SDK的目录下/docs/reference/android/widget/TextView.html找到你想要查看控件的API

分析思路

一般自定义view都需要满足2个条件，展示我们期望的UI，正确传递或者接收处理点击或者触摸事件。
所以对于TextView的分析也从这三个地方展开

绘制过程

onMeasure()
onLayout()
onDraw()
事件接收处理

由于TextView继承于View，所以主要分析onTouchEvent()方法就好了
一些和TextView有关的类如何实现，比如Spans，Layout,接收输入的InputConnection

本文基于Android SDK API-19的基础上分析

在分析之前，我们先来看个小彩蛋

不知道这个//TODO是某个哥们自问自答呢，还是别人在对他的代码review的时候给注上的

再分析之前，顺便抛出一个问题供大家思考下，maxEms这个属性到底是用来做什么的？
网上的答案五花八门，在下面的源码中我们可以一窥究竟。

绘制过程

首先来看onMeasure()部分代码

@Override    protected void onMeasure(int widthMeasureSpec, int heightMeasureSpec) {    //首先接收到父容器传递过来的MeasureSpec    //关于MeasureSpec是如果计算的，可以查看之前的博文    //[LinearLayout源码解析](http://blog.csdn.net/wz249863091/article/details/51702980)        int widthMode = MeasureSpec.getMode(widthMeasureSpec);        int heightMode = MeasureSpec.getMode(heightMeasureSpec);        int widthSize = MeasureSpec.getSize(widthMeasureSpec);        int heightSize = MeasureSpec.getSize(heightMeasureSpec);        int width;        int height;        //这里解释下什么叫作boring        //A BoringLayout is a very simple Layout implementation for text that         //fits on a single line and is all left-to-right characters.        //boring就是指布局所用的文本里面不包含任何Span，所有的文本方向都是从左到右的布局，        //并且仅需一行就能显示完全的布局        //这里将TextView和Hint的boring初始化        BoringLayout.Metrics boring = UNKNOWN_BORING;        BoringLayout.Metrics hintBoring = UNKNOWN_BORING;        //获得文字的排序方式。一共有6种        //FIRSTSTRONG_RTL,FIRSTSTRONG_LTR Unicode双向算法        //ANYRTL_LTR        //LTR,RTL 左到右或者右到左排序        //LOCALE        //first strong算法 有兴趣的同学可以自行研究下，一般情况下都是左到右排序        if (mTextDir == null) {            mTextDir = getTextDirectionHeuristic();        }        int des = -1;        boolean fromexisting = false;        //如果宽度是精确模式了，那就那父容器给的宽度当作当前TextView的宽度        if (widthMode == MeasureSpec.EXACTLY) {            // Parent has told us how big to be. So be it.            width = widthSize;        } else {            if (mLayout != null && mEllipsize == null) {            //首先计算下期望值，如果行数大于1就返回-1，否则返回单行宽度            //具体代码贴在下面                des = desired(mLayout);            }            //如果小于0，即行数大于1行，就去判断是否是boring            //isBoring()这个方法也在下面有详细分析，大家可以阅读后            //再回过头来看看            if (des < 0) {                boring = BoringLayout.isBoring(mTransformed, mTextPaint, mTextDir, mBoring);                //阅读过下面的方法，就知道boring是一个Metrics矩阵，                //包含了文本样式 width, ascent, and descen等                if (boring != null) {                    mBoring = boring;                }            } else {                fromexisting = true;            }            //再次判断boring是否为null            //这里有2种情况会为null            //1.des>0,即Textview只显示一行文字，就不会去计算boring的值了            //2.Textview包含的内容不是boring的，多行，有缩进或者包含spann            if (boring == null || boring == UNKNOWN_BORING) {                //如果是多行文字的                if (des < 0) {                    des = (int) FloatMath.ceil(Layout.getDesiredWidth(mTransformed, mTextPaint));                }                width = des;            } else {            //如果是boring模式的就很简单了，把boring刚测量得到的width赋给TextView            //即文字的宽度                width = boring.width;            }            //这里就是加上Drawable的宽度            final Drawables dr = mDrawables;            if (dr != null) {                width = Math.max(width, dr.mDrawableWidthTop);                width = Math.max(width, dr.mDrawableWidthBottom);            }            //这里会再计算一次hint的宽度，流程和上面的一模一样，就不再重复了            if (mHint != null) {                int hintDes = -1;                int hintWidth;                if (mHintLayout != null && mEllipsize == null) {                    hintDes = desired(mHintLayout);                }                if (hintDes < 0) {                    hintBoring = BoringLayout.isBoring(mHint, mTextPaint, mTextDir, mHintBoring);                    if (hintBoring != null) {                        mHintBoring = hintBoring;                    }                }                if (hintBoring == null || hintBoring == UNKNOWN_BORING) {                    if (hintDes < 0) {                        hintDes = (int) FloatMath.ceil(Layout.getDesiredWidth(mHint, mTextPaint));                    }                    hintWidth = hintDes;                } else {                    hintWidth = hintBoring.width;                }                if (hintWidth > width) {                    width = hintWidth;                }            }            //这里再加上padding的值            //顺便说一句，padding的值是在子view里自己算的            //margin的值是在父容器里算的            //在自定义view和viewgroup的时候，千万注意            //width += getCompoundPaddingLeft() + getCompoundPaddingRight();            //在这，就能解答之前的疑问，EMS这个属性到底是干嘛的            //如果我们设置了maxEms这个属性            //public void setMaxEms(int maxems) {           // mMaxWidth = maxems;            //mMaxWidthMode = EMS;            //requestLayout();            //invalidate();                //}            //mMaxWidth的值就是EMS的值            //如果设置了maxLength，那么mMaxWidth的值就是maxWidth的值            //然后再来看如果是EMS模式            //Math.min(width, mMaxWidth * getLineHeight())            //我们的最大宽度就是EMS的值乘以lineHeight的值            //而lineHeight的值 官方是这么解释的            //return the height of one standard line in pixels            //public int getLineHeight() {           // return FastMath.round(mTextPaint.getFontMetricsInt(null) *            //mSpacingMult + mSpacingAdd);           //就是行间距乘以字体大小           //所以在不同行间距和字体大小下，EMS所产生的mMaxWidth也是不同的    }            if (mMaxWidthMode == EMS) {                width = Math.min(width, mMaxWidth * getLineHeight());            } else {                width = Math.min(width, mMaxWidth);            }            if (mMinWidthMode == EMS) {                width = Math.max(width, mMinWidth * getLineHeight());            } else {                width = Math.max(width, mMinWidth);            }            // Check against our minimum width            width = Math.max(width, getSuggestedMinimumWidth());            //如果是Wrap的，会在父容器给的size和实际最大size中取小的            if (widthMode == MeasureSpec.AT_MOST) {                width = Math.min(widthSize, width);            }        }        //最后根据上面计算得到的size-padding的值就是我们单行text实际可以展示的大小        int want = width - getCompoundPaddingLeft() - getCompoundPaddingRight();        int unpaddedWidth = want;        //如果是水平方向可以scroll的，那么宽度就是无限大了，因为可以滑嘛        if (mHorizontallyScrolling) want = VERY_WIDE;        int hintWant = want;        int hintWidth = (mHintLayout == null) ? hintWant : mHintLayout.getWidth();        //这里会牵扯到makeNewLayout（...）这个方法，也会在下面得到详细分析        if (mLayout == null) {            makeNewLayout(want, hintWant, boring, hintBoring,                          width - getCompoundPaddingLeft() - getCompoundPaddingRight(), false);        } else {            final boolean layoutChanged = (mLayout.getWidth() != want) ||                    (hintWidth != hintWant) ||                    (mLayout.getEllipsizedWidth() !=                            width - getCompoundPaddingLeft() - getCompoundPaddingRight());            final boolean widthChanged = (mHint == null) &&                    (mEllipsize == null) &&                    (want > mLayout.getWidth()) &&                    (mLayout instanceof BoringLayout || (fromexisting && des >= 0 && des <= want));            final boolean maximumChanged = (mMaxMode != mOldMaxMode) || (mMaximum != mOldMaximum);            if (layoutChanged || maximumChanged) {                if (!maximumChanged && widthChanged) {                    mLayout.increaseWidthTo(want);                } else {                    makeNewLayout(want, hintWant, boring, hintBoring,                            width - getCompoundPaddingLeft() - getCompoundPaddingRight(), false);                }            } else {                // Nothing has changed            }        }        //然后开始计算高度，这部分代码相对于宽度，就简单的多了        //如果是精确模式，那么高度就等于TextView要求的高度        if (heightMode == MeasureSpec.EXACTLY) {            // Parent has told us how big to be. So be it.            height = heightSize;            mDesiredHeightAtMeasure = -1;        } else {            //计算下想要的高度            //这里逻辑比较简单            //只需要比较下文字高度和hint的高度，取大的那个值就可以了            //至于文字高度和hint高度的计算：            //1.当行高度*行数            //如果设置了Drawable的话，比较2个值得大小，取大的            //如果设置了maxLines或者maxHeight计算下当前高度有没超过最大高度，超过的话取最大高度            //如果设置了minLines或者minHeight的话，比较下当前高度和最小高度，取小的            int desired = getDesiredHeight();            height = desired;            mDesiredHeightAtMeasure = desired;            //如果是warp模式，就取父容器算的和实际需要小的值            if (heightMode == MeasureSpec.AT_MOST) {                height = Math.min(desired, heightSize);            }        }        int unpaddedHeight = height - getCompoundPaddingTop() - getCompoundPaddingBottom();        if (mMaxMode == LINES && mLayout.getLineCount() > mMaximum) {            unpaddedHeight = Math.min(unpaddedHeight, mLayout.getLineTop(mMaximum));        }        /*         * We didn't let makeNewLayout() register to bring the cursor into view,         * so do it here if there is any possibility that it is needed.         */         //这里就是处理下滚动条        if (mMovement != null ||            mLayout.getWidth() > unpaddedWidth ||            mLayout.getHeight() > unpaddedHeight) {            registerForPreDraw();        } else {            scrollTo(0, 0);        }        setMeasuredDimension(width, height);    }

desire()方法

private static int desired(Layout layout) {        //首先获得行数        int n = layout.getLineCount();        CharSequence text = layout.getText();        float max = 0;        // if any line was wrapped, we can't use it.        // but it's ok for the last line not to have a newline        //如果行数大于1，就返回-1        for (int i = 0; i < n - 1; i++) {            if (text.charAt(layout.getLineEnd(i) - 1) != '\n')                return -1;        }        //将宽度和0比较，如果大于0，就取宽度        for (int i = 0; i < n; i++) {            max = Math.max(max, layout.getLineWidth(i));        }        return (int) FloatMath.ceil(max);    }

isBoring()实现

/**     * Returns null if not boring; the width, ascent, and descent in the     * provided Metrics object (or a new one if the provided one was null)     * if boring.     * @hide     */     //如果是boring模式的就返回Metrics object，不是就返回null     //什么是boring模式 开头已经讲过了，根据他的定义也不难猜到这个方法有几个条件判断    public static Metrics isBoring(CharSequence text, TextPaint paint,            TextDirectionHeuristic textDir, Metrics metrics) {        //首先获得一个char型数组，这里值得一提的是TextUtils.obtain(500)这个方法        //在我们自己写代码的时候也可以借鉴，减少内存的交换        //这里把字符串分成了以500个字符为一组        char[] temp = TextUtils.obtain(500);        int length = text.length();        boolean boring = true;        outer:        for (int i = 0; i < length; i += 500) {            int j = i + 500;            //首先判断是否当前组的字符串是否有500个            //没有就取实际长度            if (j > length)                j = length;            //根据长度取出字符串的子串            TextUtils.getChars(text, i, j, temp, 0);            //子串的长度            int n = j - i;            //遍历整个子串            for (int a = 0; a < n; a++) {                char c = temp[a];                //这里有3个条件                //1.如果有换行 \n                //2.如果有缩进 \t                //3.如果不是LTR 左到右模式                //如果有其中1种情况，就视为不是boring模式                if (c == '\n' || c == '\t' || c >= FIRST_RIGHT_TO_LEFT) {                    boring = false;                    break outer;                }            }            if (textDir != null && textDir.isRtl(temp, 0, n)) {               boring = false;               break outer;            }        }        //把temp回收        TextUtils.recycle(temp);        //如果包含了span，那么也视为不是boring模式        if (boring && text instanceof Spanned) {            Spanned sp = (Spanned) text;            Object[] styles = sp.getSpans(0, length, ParagraphStyle.class);            if (styles.length > 0) {                boring = false;            }        }        //如果是boring模式，那就返回Metrics对象        if (boring) {            Metrics fm = metrics;            //首先判断传进来的Metrics是否为空，如果为空，就新建一个对象            if (fm == null) {                fm = new Metrics();            }            //设置TextLine，文本样式            TextLine line = TextLine.obtain();            line.set(paint, text, 0, length, Layout.DIR_LEFT_TO_RIGHT,                    Layout.DIRS_ALL_LEFT_TO_RIGHT, false, null);            fm.width = (int) FloatMath.ceil(line.metrics(fm));            TextLine.recycle(line);            return fm;        } else {            return null;        }    }

makeNewLayout（…）方法

/**     * The width passed in is now the desired layout width,     * not the full view width with padding.     * {@hide}     */    protected void makeNewLayout(int wantWidth, int hintWidth,                                 BoringLayout.Metrics boring,                                 BoringLayout.Metrics hintBoring,                                 int ellipsisWidth, boolean bringIntoView) {        //首先，如果有跑马灯效果，先把跑马灯停了        stopMarquee();        // Update "old" cached values        //把最大宽度和最大行数先保存起来        mOldMaximum = mMaximum;        mOldMaxMode = mMaxMode;        mHighlightPathBogus = true;        if (wantWidth < 0) {            wantWidth = 0;        }        if (hintWidth < 0) {            hintWidth = 0;        }        //获得对其方式        Layout.Alignment alignment = getLayoutAlignment();        final boolean testDirChange = mSingleLine && mLayout != null &&            (alignment == Layout.Alignment.ALIGN_NORMAL ||             alignment == Layout.Alignment.ALIGN_OPPOSITE);        int oldDir = 0;        if (testDirChange) oldDir = mLayout.getParagraphDirection(0);        //是否需要省略号，这个值是根据我们在XML中写的Ellipsize来定的        boolean shouldEllipsize = mEllipsize != null && getKeyListener() == null;        //省略号主要分开始位置，中间位置和结束位置3个常规位置        //还有跑马灯这种非常规位置        //mMarqueeFadeMode分为3种效果        //MARQUEE_FADE_NORMAL        //MARQUEE_FADE_SWITCH_SHOW_ELLIPSIS        //MARQUEE_FADE_SWITCH_SHOW_FADE        //这里判断是否为常规的marquee        final boolean switchEllipsize = mEllipsize == TruncateAt.MARQUEE &&                mMarqueeFadeMode != MARQUEE_FADE_NORMAL;        TruncateAt effectiveEllipsize = mEllipsize;        if (mEllipsize == TruncateAt.MARQUEE &&                mMarqueeFadeMode == MARQUEE_FADE_SWITCH_SHOW_ELLIPSIS) {            effectiveEllipsize = TruncateAt.END_SMALL;        }        //获得排序方向，一般是LTR 左到右        if (mTextDir == null) {            mTextDir = getTextDirectionHeuristic();        }        //获得一个singleLayout,这个方法在下面分析        mLayout = makeSingleLayout(wantWidth, boring, ellipsisWidth, alignment, shouldEllipsize,                effectiveEllipsize, effectiveEllipsize == mEllipsize);        //如果是非常规的跑马灯，需要再创建一个mSavedMarqueeModeLayout，在播放跑马灯的时候        //把这个layout作为mLayout        if (switchEllipsize) {            TruncateAt oppositeEllipsize = effectiveEllipsize == TruncateAt.MARQUEE ?                    TruncateAt.END : TruncateAt.MARQUEE;            mSavedMarqueeModeLayout = makeSingleLayout(wantWidth, boring, ellipsisWidth, alignment,                    shouldEllipsize, oppositeEllipsize, effectiveEllipsize != mEllipsize);        }        shouldEllipsize = mEllipsize != null;        mHintLayout = null;        //如果有默认提示，还要计算hintLayout        if (mHint != null) {            //如果有省略号，那么提示文字的宽度就是实际能分配的宽度            if (shouldEllipsize) hintWidth = wantWidth;            //这段代码看着眼熟不？就是刚在onLayout()里分析的那段            if (hintBoring == UNKNOWN_BORING) {               //如果hint只是单行，无缩进，无spann那么就是把boring的矩阵赋值hintBoring                hintBoring = BoringLayout.isBoring(mHint, mTextPaint, mTextDir,                                                   mHintBoring);                if (hintBoring != null) {                    mHintBoring = hintBoring;                }            }            //这里有3层if嵌套，如果不仔细看，很容易晕            //主要分为3层逻辑            //最外层当前的hint是否是boring的，即单行无span            //第二层主要判断是否需要省略号，当前hint实际需要宽度是否大于计算得到的宽度            //如果仔细看了前面分析可以知道wantWidth,hintWidth和ellipsisWidth            //其实都是一个值 width-paddingleft-paddingRight            //即实际可以提供的宽度            //然后最里面那层就是判断mSavedHintLayout是否为null            //如果为null就make一个新的，不为null就把老的值更新            //先看如果是boring模式            if (hintBoring != null) {                //首先判断hint需要宽度是否小于实际给的宽度                if (hintBoring.width <= hintWidth &&                    (!shouldEllipsize || hintBoring.width <= ellipsisWidth)) {                    if (mSavedHintLayout != null) {                        mHintLayout = mSavedHintLayout.                                replaceOrMake(mHint, mTextPaint,                                hintWidth, alignment, mSpacingMult, mSpacingAdd,                                hintBoring, mIncludePad);                    } else {                        mHintLayout = BoringLayout.make(mHint, mTextPaint,                                hintWidth, alignment, mSpacingMult, mSpacingAdd,                                hintBoring, mIncludePad);                    }                    mSavedHintLayout = (BoringLayout) mHintLayout;                //如果不满足上面的要求                //再判断是否需要省略号，并且需要宽度是否小于实际给的宽度                //如果进入这个if条件的，都是设置了省略号，但是不需要显示的                } else if (shouldEllipsize && hintBoring.width <= hintWidth) {                    if (mSavedHintLayout != null) {                        mHintLayout = mSavedHintLayout.                                replaceOrMake(mHint, mTextPaint,                                hintWidth, alignment, mSpacingMult, mSpacingAdd,                                hintBoring, mIncludePad, mEllipsize,                                ellipsisWidth);                    } else {                        mHintLayout = BoringLayout.make(mHint, mTextPaint,                                hintWidth, alignment, mSpacingMult, mSpacingAdd,                                hintBoring, mIncludePad, mEllipsize,                                ellipsisWidth);                    }                //如果还是不满足                //到了这就应该是需要省略号，但是需要宽度是大于实际给的宽度                //那么就应该显示省略号了                } else if (shouldEllipsize) {                    mHintLayout = new StaticLayout(mHint,                                0, mHint.length(),                                mTextPaint, hintWidth, alignment, mTextDir, mSpacingMult,                                mSpacingAdd, mIncludePad, mEllipsize,                                ellipsisWidth, mMaxMode == LINES ? mMaximum : Integer.MAX_VALUE);                } else {                    mHintLayout = new StaticLayout(mHint, mTextPaint,                            hintWidth, alignment, mTextDir, mSpacingMult, mSpacingAdd,                            mIncludePad);                }            //到了这开始处理不是boring的hint            //如果不是boring的，需要省略号            } else if (shouldEllipsize) {                mHintLayout = new StaticLayout(mHint,                            0, mHint.length(),                            mTextPaint, hintWidth, alignment, mTextDir, mSpacingMult,                            mSpacingAdd, mIncludePad, mEllipsize,                            ellipsisWidth, mMaxMode == LINES ? mMaximum : Integer.MAX_VALUE);            //如果不是boring的，不需要省略号            } else {                mHintLayout = new StaticLayout(mHint, mTextPaint,                        hintWidth, alignment, mTextDir, mSpacingMult, mSpacingAdd,                        mIncludePad);            }        }        //如果文字方向发生变化了，就重新注册OnPreDrawListener        //OnPreDrawListener回调的时机是        //即将绘制视图树时执行的回调函数。这时所有的视图都测量完成并确定了框架。 客户端可以        //使用该方法来调整滚动边框，甚至可以在绘制之前请求新的布局。        if (bringIntoView || (testDirChange && oldDir != mLayout.getParagraphDirection(0))) {            registerForPreDraw();        }        //这里开始处理跑马灯        //如果需要播放跑马灯        if (mEllipsize == TextUtils.TruncateAt.MARQUEE) {            if (!compressText(ellipsisWidth)) {                final int height = mLayoutParams.height;                // If the size of the view does not depend on the size of the text, try to                // start the marquee immediately                //这里值得稍微留意是                //如果当前TextView的宽度不需要依赖内部文字的话                //直接就可以播放跑马灯了                if (height != LayoutParams.WRAP_CONTENT && height != LayoutParams.MATCH_PARENT) {                    startMarquee();                } else {                    // Defer the start of the marquee until we know our width (see setFrame())                    mRestartMarquee = true;                }            }        }        // CursorControllers need a non-null mLayout        if (mEditor != null) mEditor.prepareCursorControllers();    }

分析到这onMeasure()就结束了

让我们来看下onLayout()是如何实现的

 @Override    protected void onLayout(boolean changed, int left, int top, int right, int bottom) {        super.onLayout(changed, left, top, right, bottom);        //这里主要的逻辑就是bringPointIntoView        if (mDeferScroll >= 0) {            int curs = mDeferScroll;            mDeferScroll = -1;            bringPointIntoView(Math.min(curs, mText.length()));        }    }

最后我们看下onDraw()的实现，看看文字是如何被绘制到屏幕上的

@Override    protected void onDraw(Canvas canvas) {        restartMarqueeIfNeeded();        // Draw the background for this view        super.onDraw(canvas);        //首先先计算padding和scorll的值        //还有判断是LTR还是RTS方向        final int compoundPaddingLeft = getCompoundPaddingLeft();        final int compoundPaddingTop = getCompoundPaddingTop();        final int compoundPaddingRight = getCompoundPaddingRight();        final int compoundPaddingBottom = getCompoundPaddingBottom();        final int scrollX = mScrollX;        final int scrollY = mScrollY;        final int right = mRight;        final int left = mLeft;        final int bottom = mBottom;        final int top = mTop;        final boolean isLayoutRtl = isLayoutRtl();        final int offset = getHorizontalOffsetForDrawables();        final int leftOffset = isLayoutRtl ? 0 : offset;        final int rightOffset = isLayoutRtl ? offset : 0 ;        //如果有drawable，那么先绘制draw        final Drawables dr = mDrawables;        if (dr != null) {            /*             * Compound, not extended, because the icon is not clipped             * if the text height is smaller.             */            //计算水平和垂直空间            int vspace = bottom - top - compoundPaddingBottom - compoundPaddingTop;            int hspace = right - left - compoundPaddingRight - compoundPaddingLeft;            // IMPORTANT: The coordinates computed are also used in invalidateDrawable()            // Make sure to update invalidateDrawable() when changing this code.            //开始绘制DrawableLeft            if (dr.mDrawableLeft != null) {            //这里简单介绍下canvas.save()和canvas.restore()            //调用save之后，可以对canvas进行平移和旋转，确定新的原点然后绘制            //等绘制完了之后，可以把原点恢复原状                canvas.save();                canvas.translate(scrollX + mPaddingLeft + leftOffset,                                 scrollY + compoundPaddingTop +                                 (vspace - dr.mDrawableHeightLeft) / 2);                //个人认为TextView整个控件写了十分出彩                //TextView需要绘制背景，文字，Drawable                //谷歌在处理这个控件的时候，把不同的事交给不同的类去完成，充分解耦                //文字部分用Editor和layout处理                //图片部分用Drawable自行绘制                //整个TextView其实只是充当了容器作用                dr.mDrawableLeft.draw(canvas);                canvas.restore();            }            // IMPORTANT: The coordinates computed are also used in invalidateDrawable()            // Make sure to update invalidateDrawable() when changing this code.            if (dr.mDrawableRight != null) {                canvas.save();                canvas.translate(scrollX + right - left - mPaddingRight                        - dr.mDrawableSizeRight - rightOffset,                         scrollY + compoundPaddingTop + (vspace - dr.mDrawableHeightRight) / 2);                dr.mDrawableRight.draw(canvas);                canvas.restore();            }            // IMPORTANT: The coordinates computed are also used in invalidateDrawable()            // Make sure to update invalidateDrawable() when changing this code.            if (dr.mDrawableTop != null) {                canvas.save();                canvas.translate(scrollX + compoundPaddingLeft +                        (hspace - dr.mDrawableWidthTop) / 2, scrollY + mPaddingTop);                dr.mDrawableTop.draw(canvas);                canvas.restore();            }            // IMPORTANT: The coordinates computed are also used in invalidateDrawable()            // Make sure to update invalidateDrawable() when changing this code.            if (dr.mDrawableBottom != null) {                canvas.save();                canvas.translate(scrollX + compoundPaddingLeft +                        (hspace - dr.mDrawableWidthBottom) / 2,                         scrollY + bottom - top - mPaddingBottom - dr.mDrawableSizeBottom);                dr.mDrawableBottom.draw(canvas);                canvas.restore();            }        }        int color = mCurTextColor;        //如果layout为null，通过刚分析的makeNewLayout()方法，再去获得一个Layout        if (mLayout == null) {            assumeLayout();        }        Layout layout = mLayout;        //如果当前没有文字，并且设置了hint，那么就显示hint        if (mHint != null && mText.length() == 0) {            if (mHintTextColor != null) {                color = mCurHintTextColor;            }            layout = mHintLayout;        }        mTextPaint.setColor(color);        mTextPaint.drawableState = getDrawableState();        canvas.save();        //感觉写TextView控件这位工程师对自己写的代码不是很自信，留下了很多疑问        //也许是Review之后忘了删除了，自己看的时候有时候会和有代入感，感觉在给别人review代码        /*  Would be faster if we didn't have to do this. Can we chop the            (displayable) text so that we don't need to do this ever?        */        int extendedPaddingTop = getExtendedPaddingTop();        int extendedPaddingBottom = getExtendedPaddingBottom();        final int vspace = mBottom - mTop - compoundPaddingBottom - compoundPaddingTop;        final int maxScrollY = mLayout.getHeight() - vspace;        //计算矩阵的上下左右4个坐标值        float clipLeft = compoundPaddingLeft + scrollX;        float clipTop = (scrollY == 0) ? 0 : extendedPaddingTop + scrollY;        float clipRight = right - left - compoundPaddingRight + scrollX;        float clipBottom = bottom - top + scrollY -                ((scrollY == maxScrollY) ? 0 : extendedPaddingBottom);        //这里是处理文字阴影        if (mShadowRadius != 0) {            clipLeft += Math.min(0, mShadowDx - mShadowRadius);            clipRight += Math.max(0, mShadowDx + mShadowRadius);            clipTop += Math.min(0, mShadowDy - mShadowRadius);            clipBottom += Math.max(0, mShadowDy + mShadowRadius);        }        //在画布中裁剪出刚计算出来的矩阵大小        canvas.clipRect(clipLeft, clipTop, clipRight, clipBottom);        int voffsetText = 0;        int voffsetCursor = 0;        // translate in by our padding        /* shortcircuit calling getVerticaOffset() */        if ((mGravity & Gravity.VERTICAL_GRAVITY_MASK) != Gravity.TOP) {            voffsetText = getVerticalOffset(false);            voffsetCursor = getVerticalOffset(true);        }        canvas.translate(compoundPaddingLeft, extendedPaddingTop + voffsetText);        final int layoutDirection = getLayoutDirection();        final int absoluteGravity = Gravity.getAbsoluteGravity(mGravity, layoutDirection);        //如果有跑马灯，并且不是MARQUEE_FADE_SWITCH_SHOW_ELLIPSIS模式的话        if (mEllipsize == TextUtils.TruncateAt.MARQUEE &&                mMarqueeFadeMode != MARQUEE_FADE_SWITCH_SHOW_ELLIPSIS) {            //如果当前只有1行显示，并且不是SingleLine的，也不是Gravity.LEFT            if (!mSingleLine && getLineCount() == 1 && canMarquee() &&                    (absoluteGravity & Gravity.HORIZONTAL_GRAVITY_MASK) != Gravity.LEFT) {                final int width = mRight - mLeft;                final int padding = getCompoundPaddingLeft() + getCompoundPaddingRight();                //dx值就是layout的宽度减去实际宽度再减去padding的值，主要是给RTL模式计算偏移量                final float dx = mLayout.getLineRight(0) - (width - padding);                canvas.translate(isLayoutRtl ? -dx : +dx, 0.0f);            }            if (mMarquee != null && mMarquee.isRunning()) {                final float dx = -mMarquee.getScroll();                canvas.translate(isLayoutRtl ? -dx : +dx, 0.0f);            }        }        final int cursorOffsetVertical = voffsetCursor - voffsetText;        //这里终于开始绘制文字了        Path highlight = getUpdatedHighlightPath();        //如果是EditText的就交给mEditor绘制，普通TextView，就交给layout处理        if (mEditor != null) {            mEditor.onDraw(canvas, layout, highlight, mHighlightPaint, cursorOffsetVertical);        } else {            layout.draw(canvas, highlight, mHighlightPaint, cursorOffsetVertical);        }        if (mMarquee != null && mMarquee.shouldDrawGhost()) {            final int dx = (int) mMarquee.getGhostOffset();            canvas.translate(isLayoutRtl ? -dx : dx, 0.0f);            layout.draw(canvas, highlight, mHighlightPaint, cursorOffsetVertical);        }        canvas.restore();    }

总结

上文中主要分析了TextView的整个绘制流程，主要是从过程的角度分析了几个比较重要的阶段。
在下一篇TextView源码分析（二）中会具体分析Layout,Editor和Drawable是如何完成绘制，排版。

简介

分析思路

绘制过程

总结

更多相关文章

随机推荐