上一篇文章最后結束在ViewRootImpl.setView()函數，這個函數之后發生了什么事情，我們接着分析。
1. ViewRootImpl簡介
ViewRootImpl作為視圖層次中的頂層，實現了View和WindowManager之間需要的協議，與SystemServer進程的WindowManagerService有交互，具體實現了WindowManagerGlobal內部的大部分功能。
1.1 ViewRootImpl的定義：
public final class ViewRootImpl implements ViewParent, View.AttachInfo.Callbacks, HardwareRenderer.HardwareDrawCallbacks { ...... final IWindowSession mWindowSession; final W mWindow; //W繼承於IWindow.Stub類。 View mView; // 由setView()函數賦值。 private final Surface mSurface = new Surface(); //mSurface可能存在多線程訪問，需要加鎖保護。 ...... public ViewRootImpl(Context context, Display display) { ...... mWindowSession = WindowManagerGlobal.getWindowSession(); mWindow = new W(this); ...... } }
mWindowSession將被用於建立Activity的ViewRootImpl和WindowManagerService的關系。創建代碼如下：
getWindowSession()@WindowManagerGlobal.java
public static IWindowSession getWindowSession() { synchronized (WindowManagerGlobal.class) { if (sWindowSession == null) { try { InputMethodManager imm = InputMethodManager.getInstance(); IWindowManager windowManager = getWindowManagerService(); //得到WindowMangerService的Binder代理. sWindowSession = windowManager.openSession(imm.getClient(), imm.getInputContext()); ///openSession操作是一個使用Binder通信的跨進程調用。
...... } catch (RemoteException e) { ...... } } return sWindowSession; } }
getWindowManagerService()@WindowManagerGlobal.java
public static IWindowManager getWindowManagerService() { synchronized (WindowManagerGlobal.class) { if (sWindowManagerService == null) { sWindowManagerService = IWindowManager.Stub.asInterface(ServiceManager.getService("window")); } return sWindowManagerService; } }
openSession()@WindowManagerService.java
public IWindowSession openSession(IInputMethodClient client, IInputContext inputContext) { ...... Session session = new Session(this, client, inputContext); return session; }
可以看到，通過WindowManagerService.openSession()獲得一個Session對象。類Session繼承於IWindowSession.Stub，並實現了IBinder接口。它支持Binder通信，並屬於Bn端，即響應請求的服務端。
1.2 ViewRootImpl與WindowManagerService的關系
ViewRootImpl通過mWindowSession（IWindowSession）與WMS進程進行跨進程通信。
ViewRootImpl的mWindow是基於Binder通信的類，而且W是IWindow的Bn端，用於響應請求。
它們之間屬於跨進程通信，關系如下圖：
每個APP進程都會和WMS建立一個IWindowSession會話，APP進程利用這個會話與WMS通信。WMS利用IWindow進行事件通知，每當發生一些事件（按鍵觸摸事件）時，WMS就會告訴某個IWindow.
事件分發大致流程：
(1) WMS所在的SystemServer進程接收到按鍵事件。
(2) WMS找到UI位於屏幕頂端的進程所對應的IWindow對象，這個是Bp端對象。
(3) 調用這個IWindow對象的dispatchKey。 IWindow對象的Bn端位於ViewRootImpl中，ViewRootImpl根據內部View的位置信息找到真正處理這個事件的View，最后調用dispatchKey函數完成事件處理。
每個按鍵事件都會被轉化成一個消息，由系統將這個消息加入到對應進程的消息隊列中。該進程在處理消息派發時，根據消息的句柄找到對應的Window，進而由Window處理掉這個消息。
2. ViewRootImpl.setView()函數
setView()@ViewRootImpl.java
public void setView(View view, WindowManager.LayoutParams attrs, View panelParentView) { synchronized (this) { if (mView == null) { mView = view; //傳進來的view是PhoneWindow的DecorView，保存起來。 mViewLayoutDirectionInitial = mView.getRawLayoutDirection(); mFallbackEventHandler.setView(view); mWindowAttributes.copyFrom(attrs); ...... mSoftInputMode = attrs.softInputMode; // InputMode可以影響輸入法彈出時視圖應該怎么顯示。 ...... mAdded = true; int res; /* = WindowManagerImpl.ADD_OKAY; */ //在被加到WindowManager之前調度第一次layout,確保收到系統事件之前重新進行了布局。 requestLayout(); if ((mWindowAttributes.inputFeatures & WindowManager.LayoutParams.INPUT_FEATURE_NO_INPUT_CHANNEL) == 0) { mInputChannel = new InputChannel(); //創建InputChannel對象 } try { mOrigWindowType = mWindowAttributes.type; mAttachInfo.mRecomputeGlobalAttributes = true; collectViewAttributes(); res = mWindowSession.addToDisplay(mWindow, mSeq, mWindowAttributes, //注意第一個參數是mWindow. 繼承IWindow.Stub類。 getHostVisibility(), mDisplay.getDisplayId(), mAttachInfo.mContentInsets, mInputChannel); } catch (RemoteException e) { ...... }
// 創建InputQueue和WindowInputEventReceiver與窗口輸入事件相關的對象。 if (mInputChannel != null) { if (mInputQueueCallback != null) { mInputQueue = new InputQueue(); mInputQueueCallback.onInputQueueCreated(mInputQueue); } mInputEventReceiver = new WindowInputEventReceiver(mInputChannel, Looper.myLooper()); } mAddedTouchMode = (res & WindowManagerGlobal.ADD_FLAG_IN_TOUCH_MODE) != 0; // 根據res設置touch mode。 mAppVisible = (res & WindowManagerGlobal.ADD_FLAG_APP_VISIBLE) != 0; // 根據res設置visible屬性。 ...... } } }
3. performTraversals()函數
requestLayout()函數的功能就是觸發UI繪制。結合時序圖。調用3至調用15，是消息發送傳遞及Callback設置過程。最后會由TraversalRunnable的run函數中調用ViewRootImpl.doTraversal()函數去調用performTraversals()開始進行UI繪制。
performTraversals()函數很復雜，我們先專注於時序圖展示的內容。
performTraversals()@ViewRootImpl.java
private void performTraversals() { ...... final View host = mView; if (mFirst || windowShouldResize || insetsChanged || viewVisibilityChanged || params != null) {
......
try {
......
relayoutResult = relayoutWindow(params, viewVisibility, insetsPending);
}
......
if (!mStopped) { boolean focusChangedDueToTouchMode = ensureTouchModeLocally( (relayoutResult&WindowManagerGlobal.RELAYOUT_RES_IN_TOUCH_MODE) != 0); if (focusChangedDueToTouchMode || mWidth != host.getMeasuredWidth() || mHeight != host.getMeasuredHeight() || contentInsetsChanged) { int childWidthMeasureSpec = getRootMeasureSpec(mWidth, lp.width); //獲取測量模式。lp.width和lp.height在創建ViewGroup實例時等於MATCH_PARENTint childHeightMeasureSpec = getRootMeasureSpec(mHeight, lp.height); // Ask host how big it wants to be performMeasure(childWidthMeasureSpec, childHeightMeasureSpec); // 根據測量模式進行測量。 ......
layoutRequested = true; } } } final boolean didLayout = layoutRequested && !mStopped; boolean triggerGlobalLayoutListener = didLayout || attachInfo.mRecomputeGlobalAttributes; if (didLayout) { performLayout(lp, desiredWindowWidth, desiredWindowHeight); // By this point all views have been sized and positioned ...... } ...... boolean cancelDraw = attachInfo.mTreeObserver.dispatchOnPreDraw() || viewVisibility != View.VISIBLE; if (!cancelDraw && !newSurface) { if (!skipDraw || mReportNextDraw) { ......
performDraw(); } }
...... }
3.1 relayoutWindow()函數
relayoutWindow()@ViewRootImpl.java
private int relayoutWindow(WindowManager.LayoutParams params, int viewVisibility, boolean insetsPending) throws RemoteException { ...... // 調用WMS的relayout函數。通過該函數修改一個window的參數，提供新的參數，它將返回屏幕上window的新的一幀(忽略位置)，和該window的surface。如果window當前是隱藏的，那么surface是無效的，否則就可以使用它來畫window的內容。 int relayoutResult = mWindowSession.relayout( mWindow, // 需要修改參數的window mSeq, // 排序的序列號 params, // 要應用到window的新參數 (int) (mView.getMeasuredWidth() * appScale + 0.5f), //window想要的寬 (int) (mView.getMeasuredHeight() * appScale + 0.5f), //window想要的高 viewVisibility, // window的root view的可見性 insetsPending ? WindowManagerGlobal.RELAYOUT_INSETS_PENDING : 0, mWinFrame, // outFrame, 放着屏幕上新position/size的Rect mPendingOverscanInsets, mPendingContentInsets, mPendingVisibleInsets, mPendingConfiguration, mSurface // 存放新顯示surface的對象。 ); ...... return relayoutResult; }
3.2 performMeasure()
在執行performMesure()時，需要傳入測量模式，這里簡單介紹下MeasureSpec。MeasureSpec封裝了從父view傳 給子view的布局要求，每個MeasureSpec代表對寬度或高度的要求。每個MeasureSpec值由大小和模式組成，高16位表示mode，低 16位表示size，即父view的size。
MeasureSpec有三種可能的模式：
UNSPECIFIED： 父view沒有強加任何限制給子view，子view可以有它想要的size。
EXACTLY：父view決定子view的確切size，子view將被指定這些邊界，而忽視它想要的size.
AT_MOST: 父view對子view沒有限制，子view可以獲得它自己指定的size。
對於DecorView來說，mode一般為EXACTLY，size為屏幕的寬高。
看下父view中是如何得到MeasureSpec的：
getRootMeasureSpec()@ViewRootImpl.java
// 根據layout params計算window中root view的MeasureSpec. // windowSize: window可用的寬或高 // window的寬或高的layout params. private static int getRootMeasureSpec(int windowSize, int rootDimension) { int measureSpec; switch (rootDimension) { case ViewGroup.LayoutParams.MATCH_PARENT: // Window can't resize. Force root view to be windowSize. // MeasureSpec.EXACTLY: 父控件決定子控件的確切大小，子控件與父控件一樣大，子控件要求的大小被忽視。 measureSpec = MeasureSpec.makeMeasureSpec(windowSize, MeasureSpec.EXACTLY); break; case ViewGroup.LayoutParams.WRAP_CONTENT: // Window can resize. Set max size for root view. // MeasureSpec.AT_MOST：子控件的大小可以為它指定的大小。 measureSpec = MeasureSpec.makeMeasureSpec(windowSize, MeasureSpec.AT_MOST); break; default: // Window wants to be an exact size. Force root view to be that size. measureSpec = MeasureSpec.makeMeasureSpec(rootDimension, MeasureSpec.EXACTLY); break; } return measureSpec; }
performMeasure()@ViewRootImpl.java
private void performMeasure(int childWidthMeasureSpec, int childHeightMeasureSpec) { Trace.traceBegin(Trace.TRACE_TAG_VIEW, "measure"); try { mView.measure(childWidthMeasureSpec, childHeightMeasureSpec); // 調用mView的measure。 } finally { Trace.traceEnd(Trace.TRACE_TAG_VIEW); } }
measure()@View.java
// 這個函數計算出view的大小，然后設置實際的寬和高。每個view的實際寬高都是由父view和自身共同決定的。 // 實際測量工作在onMeasure()函數中執行，子類也能重載onMesure()函數實現對自身的測量。 public final void measure(int widthMeasureSpec, int heightMeasureSpec) { // final方法，子類不可以重寫 ......if ((mPrivateFlags & PFLAG_FORCE_LAYOUT) == PFLAG_FORCE_LAYOUT || widthMeasureSpec != mOldWidthMeasureSpec || heightMeasureSpec != mOldHeightMeasureSpec) { // first clears the measured dimension flag mPrivateFlags &= ~PFLAG_MEASURED_DIMENSION_SET; resolveRtlPropertiesIfNeeded(); int cacheIndex = (mPrivateFlags & PFLAG_FORCE_LAYOUT) == PFLAG_FORCE_LAYOUT ? -1 : mMeasureCache.indexOfKey(key); if (cacheIndex < 0 || sIgnoreMeasureCache) { // 調用onMeasure方法, 測量自己，這個調用需要重新設置measured dimension flag。 onMeasure(widthMeasureSpec, heightMeasureSpec); mPrivateFlags3 &= ~PFLAG3_MEASURE_NEEDED_BEFORE_LAYOUT; } ...... // flag not set, setMeasuredDimension() was not invoked, we raise // an exception to warn the developer if ((mPrivateFlags & PFLAG_MEASURED_DIMENSION_SET) != PFLAG_MEASURED_DIMENSION_SET) { // 子類重寫onMeausre方法的時，必須調用setMeasuredDimension或者super.onMeasure方法，不然就會走到這里報錯。 // setMeasuredDimension中回去改變mPrivateFlags的值 throw new IllegalStateException("onMeasure() did not set the" + " measured dimension by calling" + " setMeasuredDimension()"); } mPrivateFlags |= PFLAG_LAYOUT_REQUIRED; } mOldWidthMeasureSpec = widthMeasureSpec; mOldHeightMeasureSpec = heightMeasureSpec; mMeasureCache.put(key, ((long) mMeasuredWidth) << 32 | (long) mMeasuredHeight & 0xffffffffL); // suppress sign extension }
onMeasure()@View.java
// 提供view和它的內容的測量寬高，應該由子類重寫這個函數來提供准確有效的測量值。這個函數的兩個參數都是由父view傳過來的，代表父view的規格。 // 子類重寫這個函數的時候必須調用setMeasuredDimension()保存這個view的測量寬和高，確保測量的寬和高至少是view的最小寬高。 protected void onMeasure(int widthMeasureSpec, int heightMeasureSpec) { setMeasuredDimension(getDefaultSize(getSuggestedMinimumWidth(), widthMeasureSpec), getDefaultSize(getSuggestedMinimumHeight(), heightMeasureSpec)); }
setMeasuredDimension()@View.java
// 設置測量結果及flags.
protected final void setMeasuredDimension(int measuredWidth, int measuredHeight) { boolean optical = isLayoutModeOptical(this); if (optical != isLayoutModeOptical(mParent)) { Insets insets = getOpticalInsets(); int opticalWidth = insets.left + insets.right; int opticalHeight = insets.top + insets.bottom; measuredWidth += optical ? opticalWidth : -opticalWidth; measuredHeight += optical ? opticalHeight : -opticalHeight; } // 賦值給mMeasuredWidth，getMeasuredWidth()函數返回的就是這個值。
// measure的目的就是對view樹中的每個view的mMeasuredWidth和mMesuredHeight賦值。這兩個值被賦值，該view的測量工作才算結束。 mMeasuredWidth = measuredWidth; mMeasuredHeight = measuredHeight; // 設置flag, 這就是重寫onMeasure方法時如果不調用setMeasuredDimension方法時, measure會拋出異常的原因。 mPrivateFlags |= PFLAG_MEASURED_DIMENSION_SET; }
View，ViewGroup，FrameLayout，DecorView的關系如下：
DecorView.onMesure()@PhoneWindow.java
protected void onMeasure(int widthMeasureSpec, int heightMeasureSpec) { final DisplayMetrics metrics = getContext().getResources().getDisplayMetrics(); final boolean isPortrait = metrics.widthPixels < metrics.heightPixels; final int widthMode = getMode(widthMeasureSpec); final int heightMode = getMode(heightMeasureSpec); boolean fixedWidth = false; if (widthMode == AT_MOST) { final TypedValue tvw = isPortrait ? mFixedWidthMinor : mFixedWidthMajor; if (tvw != null && tvw.type != TypedValue.TYPE_NULL) { final int w; if (tvw.type == TypedValue.TYPE_DIMENSION) { w = (int) tvw.getDimension(metrics); } else if (tvw.type == TypedValue.TYPE_FRACTION) { w = (int) tvw.getFraction(metrics.widthPixels, metrics.widthPixels); } else { w = 0; } if (w > 0) { final int widthSize = MeasureSpec.getSize(widthMeasureSpec); widthMeasureSpec = MeasureSpec.makeMeasureSpec(Math.min(w, widthSize), EXACTLY); fixedWidth = true; } } } if (heightMode == AT_MOST) { final TypedValue tvh = isPortrait ? mFixedHeightMajor : mFixedHeightMinor; if (tvh != null && tvh.type != TypedValue.TYPE_NULL) { final int h; if (tvh.type == TypedValue.TYPE_DIMENSION) { h = (int) tvh.getDimension(metrics); } else if (tvh.type == TypedValue.TYPE_FRACTION) { h = (int) tvh.getFraction(metrics.heightPixels, metrics.heightPixels); } else { h = 0; } if (h > 0) { final int heightSize = MeasureSpec.getSize(heightMeasureSpec); heightMeasureSpec = MeasureSpec.makeMeasureSpec(Math.min(h, heightSize), EXACTLY); } } } super.onMeasure(widthMeasureSpec, heightMeasureSpec); // 調用FrameLayout.onMeasure().
int width = getMeasuredWidth(); boolean measure = false; widthMeasureSpec = MeasureSpec.makeMeasureSpec(width, EXACTLY); if (!fixedWidth && widthMode == AT_MOST) { final TypedValue tv = isPortrait ? mMinWidthMinor : mMinWidthMajor; if (tv.type != TypedValue.TYPE_NULL) { final int min; if (tv.type == TypedValue.TYPE_DIMENSION) { min = (int)tv.getDimension(metrics); } else if (tv.type == TypedValue.TYPE_FRACTION) { min = (int)tv.getFraction(metrics.widthPixels, metrics.widthPixels); } else { min = 0; } if (width < min) { widthMeasureSpec = MeasureSpec.makeMeasureSpec(min, EXACTLY); measure = true; } } } if (measure) { super.onMeasure(widthMeasureSpec, heightMeasureSpec); } }
onMesure()@FrameLayout.java
protected void onMeasure(int widthMeasureSpec, int heightMeasureSpec) { int count = getChildCount(); // 獲得子view的數目 final boolean measureMatchParentChildren = MeasureSpec.getMode(widthMeasureSpec) != MeasureSpec.EXACTLY || MeasureSpec.getMode(heightMeasureSpec) != MeasureSpec.EXACTLY; mMatchParentChildren.clear(); int maxHeight = 0; int maxWidth = 0; int childState = 0; for (int i = 0; i < count; i++) { // 測量每個子View final View child = getChildAt(i); if (mMeasureAllChildren || child.getVisibility() != GONE) { measureChildWithMargins(child, widthMeasureSpec, 0, heightMeasureSpec, 0); final LayoutParams lp = (LayoutParams) child.getLayoutParams(); maxWidth = Math.max(maxWidth, child.getMeasuredWidth() + lp.leftMargin + lp.rightMargin); maxHeight = Math.max(maxHeight, child.getMeasuredHeight() + lp.topMargin + lp.bottomMargin); childState = combineMeasuredStates(childState, child.getMeasuredState()); if (measureMatchParentChildren) { if (lp.width == LayoutParams.MATCH_PARENT || lp.height == LayoutParams.MATCH_PARENT) { mMatchParentChildren.add(child); } } } } // Account for padding too maxWidth += getPaddingLeftWithForeground() + getPaddingRightWithForeground(); maxHeight += getPaddingTopWithForeground() + getPaddingBottomWithForeground(); // Check against our minimum height and width maxHeight = Math.max(maxHeight, getSuggestedMinimumHeight()); maxWidth = Math.max(maxWidth, getSuggestedMinimumWidth()); // Check against our foreground's minimum height and width final Drawable drawable = getForeground(); if (drawable != null) { maxHeight = Math.max(maxHeight, drawable.getMinimumHeight()); maxWidth = Math.max(maxWidth, drawable.getMinimumWidth()); }
// 設置自己的size, 即ViewGroup的size. 可以看到，它的size由父view及子view共同決定。 setMeasuredDimension(resolveSizeAndState(maxWidth, widthMeasureSpec, childState), resolveSizeAndState(maxHeight, heightMeasureSpec, childState << MEASURED_HEIGHT_STATE_SHIFT)); count = mMatchParentChildren.size(); if (count > 1) { // 對於match parent的child進行測量。 for (int i = 0; i < count; i++) { final View child = mMatchParentChildren.get(i); final MarginLayoutParams lp = (MarginLayoutParams) child.getLayoutParams(); int childWidthMeasureSpec; int childHeightMeasureSpec; if (lp.width == LayoutParams.MATCH_PARENT) { childWidthMeasureSpec = MeasureSpec.makeMeasureSpec(getMeasuredWidth() - getPaddingLeftWithForeground() - getPaddingRightWithForeground() - lp.leftMargin - lp.rightMargin, MeasureSpec.EXACTLY); } else { childWidthMeasureSpec = getChildMeasureSpec(widthMeasureSpec, getPaddingLeftWithForeground() + getPaddingRightWithForeground() + lp.leftMargin + lp.rightMargin, lp.width); } if (lp.height == LayoutParams.MATCH_PARENT) { childHeightMeasureSpec = MeasureSpec.makeMeasureSpec(getMeasuredHeight() - getPaddingTopWithForeground() - getPaddingBottomWithForeground() - lp.topMargin - lp.bottomMargin, MeasureSpec.EXACTLY); } else { childHeightMeasureSpec = getChildMeasureSpec(heightMeasureSpec, getPaddingTopWithForeground() + getPaddingBottomWithForeground() + lp.topMargin + lp.bottomMargin, lp.height); } child.measure(childWidthMeasureSpec, childHeightMeasureSpec); } } }
measureChildWithMargins()@ViewGroup.java
// 要求view的一個child測量自己，同時考慮MeasureSpec要求和padding和margins. protected void measureChildWithMargins(View child, int parentWidthMeasureSpec, int widthUsed, int parentHeightMeasureSpec, int heightUsed) {
// 獲取子view的LayoutParams final MarginLayoutParams lp = (MarginLayoutParams) child.getLayoutParams(); // 調整MeasureSpec，通過傳進來的父view的MeasureSpec及子view自身的LayoutParams，共同決定子view的MeasureSpec. final int childWidthMeasureSpec = getChildMeasureSpec(parentWidthMeasureSpec, mPaddingLeft + mPaddingRight + lp.leftMargin + lp.rightMargin + widthUsed, lp.width); final int childHeightMeasureSpec = getChildMeasureSpec(parentHeightMeasureSpec, mPaddingTop + mPaddingBottom + lp.topMargin + lp.bottomMargin + heightUsed, lp.height); // 調用子view的measure()，又重新開始上面measure的流程。子view的measure()中會回調子view的onMeasure()函數。 child.measure(childWidthMeasureSpec, childHeightMeasureSpec); }
getChildMeasureSpec()@ViewGroup.java
public static int getChildMeasureSpec(int spec, int padding, int childDimension) { //獲取parent view的mode和size
int specMode = MeasureSpec.getMode(spec); int specSize = MeasureSpec.getSize(spec); //parent size與padding的差值(即parent中除去padding后剩余的大小)，若小於0則直接返回0. int size = Math.max(0, specSize - padding); int resultSize = 0; int resultMode = 0; switch (specMode) { //根據parent mode進行處理 // Parent has imposed an exact size on us case MeasureSpec.EXACTLY: if (childDimension >= 0) { // 若child的layout_width或者layout_height在xml或者代碼中指定了大於等於0的值，
// 則設置child的size值為指定的值，mode為EXACTLY.
resultSize = childDimension; resultMode = MeasureSpec.EXACTLY; } else if (childDimension == LayoutParams.MATCH_PARENT) { // Child wants to be our size. So be it. // 若child的layout_width或者layout_height在xml或者代碼中指定為MATCH_PARENT，
// 則設置child的size值為size，mode為EXACTLY.
resultSize = size; resultMode = MeasureSpec.EXACTLY; } else if (childDimension == LayoutParams.WRAP_CONTENT) { // Child wants to determine its own size. It can't be bigger than us. // 若child的layout_width或者layout_height在xml或者代碼中指定為WRAP_CONTENT，
// 則設置child的size值為指定的值，mode為AT_MOST.
resultSize = size; resultMode = MeasureSpec.AT_MOST; } break; // Parent has imposed a maximum size on us // 下面兩個分支類似
case MeasureSpec.AT_MOST: ......break; // Parent asked to see how big we want to be case MeasureSpec.UNSPECIFIED: ......break; }
//使用mode和size生成MeasureSpec返回。 return MeasureSpec.makeMeasureSpec(resultSize, resultMode); }
可以看到，getChildMeasureSpec()是通過其父view提供的MeasureSpec參數得到specMode和specSize，然后根據計算出來的specMode以及子view的childDimension（layout_width或layout_height）來計算自身的measureSpec，如果其本身包含子視圖，則計算出來的measureSpec將作為調用其子視圖measure函數的參數，同時也作為自身調用setMeasuredDimension()的參數，如果其不包含子視圖則默認情況下最終會調用onMeasure的默認實現，並最終調用到setMeasuredDimension()。
performMeasure()小結：
(1) mView.measure() , 即調用View.measure()方法。
(2) View.measure()內部調用onMeasure()方法，因為這里的mView是DecorView，調用的是DecorView.onMeasure()， 這個函數內部調用super.onMeasure()。因為DecorView繼承於FrameLayout，所以super.onMeasure()即調用FrameLayout的onMeasure()方法。
(3) FrameLayout.onMeasure()對每個子view進行遍歷測量，對每個子view調用measureChildWithMargins()方法測量子view。
(4) measureChildWithMargins()內部調用getChildMeasureSpec()計算出傳給子view的MeasureSpec，並調用子view的measure()方法。
(5) FrameLayout.onMeasure()中根據父view及(3)測量出來的子view的MeasureSpec信息調用setMeasuredDimension()設置自身的size。
(6) 使用View的getMeasuredWidth()和getMeasuredHeight()方法來獲取view測量的寬高，必須保證這兩個方法在onMeasure流程之后被調用才能返回有效值。
(7) ViewGroup的子類必須要求LayoutParams繼承於MarginLayoutParams，否則無法使用layout_margin參數。
3.3 performLayout()
performLayout()@ViewRootImpl.java
private void performLayout(WindowManager.LayoutParams lp, int desiredWindowWidth, int desiredWindowHeight) { mLayoutRequested = false; mScrollMayChange = true; mInLayout = true; final View host = mView; Trace.traceBegin(Trace.TRACE_TAG_VIEW, "layout"); try { // 把performMeasure測量的結果傳入layout()函數 host.layout(0, 0, host.getMeasuredWidth(), host.getMeasuredHeight()); mInLayout = false; ...... } finally { Trace.traceEnd(Trace.TRACE_TAG_VIEW); } mInLayout = false; }
這個函數內調用mView.layout()方法，而ViewGroup中重寫了layout()方法，並調用了super.layout(), 即View.layout()。
layout()@View.java
// 給view及它的子孫的size及position賦值。 // 這是布局機制的第二階段(第一階段是measuring)。在這個階段，每個父view都會調用其所有子view的layout()來放置它們。 // 子類中不應該重載這個方法，若子類中有children, 則該子類應該重載onLayout()函數，在這個函數中調用每個child的layout. public void layout(int l, int t, int r, int b) { ......int oldL = mLeft; int oldT = mTop; int oldB = mBottom; int oldR = mRight; // 這部分是判斷這個View的大小是否已經發生了變化，來判斷是否需要重新layout。判斷的方式就是通過將l,t,r,b參數傳給setFrame()賦值給
// mLeft，mTop, mRight，mBottom。 // isLayoutModeOptical(): 如果mParent是一個使用可見范圍進行編排的ViewGroup, 返回true. boolean changed = isLayoutModeOptical(mParent) ? setOpticalFrame(l, t, r, b) : setFrame(l, t, r, b); if (changed || (mPrivateFlags & PFLAG_LAYOUT_REQUIRED) == PFLAG_LAYOUT_REQUIRED) { // 調用onLayout方法 onLayout(changed, l, t, r, b); mPrivateFlags &= ~PFLAG_LAYOUT_REQUIRED; ...... } mPrivateFlags &= ~PFLAG_FORCE_LAYOUT; mPrivateFlags3 |= PFLAG3_IS_LAID_OUT; }
我們先來看View的onLayout()方法，
onLayout()@View.java
protected void onLayout(boolean changed, int left, int top, int right, int bottom) { }
這是一個空方法，這是因為Layout需要ViewGroup來控制進行。
再來看ViewGroup的onLayout()方法：
protected abstract void onLayout(boolean changed, int l, int t, int r, int b);
onLayout在ViewGroup中是個抽象方法，所以ViewGroup的子類都需要實現該方法。FrameLayout繼承於ViewGroup，我們來看下FrameLayout.onLayout()的實現。
onLayout()@FrameLayout.java
protected void onLayout(boolean changed, int left, int top, int right, int bottom) { layoutChildren(left, top, right, bottom, false /* no force left gravity */); } void layoutChildren(int left, int top, int right, int bottom, boolean forceLeftGravity) { final int count = getChildCount(); final int parentLeft = getPaddingLeftWithForeground(); final int parentRight = right - left - getPaddingRightWithForeground(); final int parentTop = getPaddingTopWithForeground(); final int parentBottom = bottom - top - getPaddingBottomWithForeground(); mForegroundBoundsChanged = true; for (int i = 0; i < count; i++) { final View child = getChildAt(i); if (child.getVisibility() != GONE) {
// 獲取子view的LayoutParams final LayoutParams lp = (LayoutParams) child.getLayoutParams(); // measure過程計算處理的子view的寬高 final int width = child.getMeasuredWidth(); final int height = child.getMeasuredHeight(); int childLeft; int childTop; int gravity = lp.gravity; if (gravity == -1) { gravity = DEFAULT_CHILD_GRAVITY; } final int layoutDirection = getLayoutDirection(); final int absoluteGravity = Gravity.getAbsoluteGravity(gravity, layoutDirection); final int verticalGravity = gravity & Gravity.VERTICAL_GRAVITY_MASK; // 根據absoluteGravity計算childLeft switch (absoluteGravity & Gravity.HORIZONTAL_GRAVITY_MASK) { case Gravity.CENTER_HORIZONTAL: childLeft = parentLeft + (parentRight - parentLeft - width) / 2 + lp.leftMargin - lp.rightMargin; break; case Gravity.RIGHT: if (!forceLeftGravity) { childLeft = parentRight - width - lp.rightMargin; break; } case Gravity.LEFT: default: childLeft = parentLeft + lp.leftMargin; } // 根據verticalGravity計算childTop switch (verticalGravity) { case Gravity.TOP: childTop = parentTop + lp.topMargin; break; case Gravity.CENTER_VERTICAL: childTop = parentTop + (parentBottom - parentTop - height) / 2 + lp.topMargin - lp.bottomMargin; break; case Gravity.BOTTOM: childTop = parentBottom - height - lp.bottomMargin; break; default: childTop = parentTop + lp.topMargin; } //調用子child的layout方法設置child的位置。 child.layout(childLeft, childTop, childLeft + width, childTop + height); } } }
FrameLayout.onLayout()函數也展示了一般的自定義ViewGroup重載onLayout函數的實現方式。onLayout()函數的目的就是確定子view們在父view中的具體位置，通常做法是用一個for循環調用每個子view的layout(l, t, r, b)函數，傳入不同的l，t，r，b值確定子view的位置。
performLayout()小結：
(1) 調用host.layout(), 即View.layout()。
(2) View.layout()中調用setFrame()或者setOpticalFrame()賦值給View的mLeft，mTop，mRight，mBottom變量並判斷位置是否變化，若變化則調用onLayout()。
(3) 因為host是mView就是DecorView，也就是FrameLayout的子類，所以實際上是FrameLayout.onLayout()被調用。
(4) 在FrameLayout.onLayout()函數中遍歷每個子view, 並調用每個子view的layout()設置子view的位置。
(5) View.layout()函數可以被重載，ViewGroup.layout不能被重載，ViewGroup的onLayout為abstract的，子類必須重載實現自己的位置邏輯。
(6) measure()過程的結果是得到每個view的measuredWidth和measuredHeight，layout()過程的結果是得到每個view的mLeft，mTop，mRight，mBottom.這些值都是相對於父view而言的。
(7) layout_XXX的布局屬性都是針對包含子view的ViewGroup的，當對一個沒有父容器的view設置相關layout_XXX屬性是沒有意義的。
(8) 使用View的getWidth()和getHeight()方法來獲取view測量的寬高時，必須保證這兩個方法在onLayout流程之后被調用才能返回有效值。
3.4 performDraw()
performDraw()@ViewRootImpl.java
private void performDraw() { if (mAttachInfo.mDisplayState == Display.STATE_OFF && !mReportNextDraw) { return; } final boolean fullRedrawNeeded = mFullRedrawNeeded; mFullRedrawNeeded = false; mIsDrawing = true; Trace.traceBegin(Trace.TRACE_TAG_VIEW, "draw"); try { // 開始draw draw(fullRedrawNeeded); } finally { mIsDrawing = false; Trace.traceEnd(Trace.TRACE_TAG_VIEW); } ...... }
draw()@ViewRootImpl.java
private void draw(boolean fullRedrawNeeded) { Surface surface = mSurface; if (!surface.isValid()) { return; } ......if (!dirty.isEmpty() || mIsAnimating) { if (attachInfo.mHardwareRenderer != null && attachInfo.mHardwareRenderer.isEnabled()) { ...... attachInfo.mHardwareRenderer.draw(mView, attachInfo, this, animating ? null : mCurrentDirty); // 使用硬件繪制 } else { ......// 使用軟件進行繪制。 if (!drawSoftware(surface, attachInfo, yoff, scalingRequired, dirty)) { return; } } } if (animating) { mFullRedrawNeeded = true; scheduleTraversals(); // 如果在動畫中，重新發送繪制消息 } }
drawSoftware()@ViewRootImpl.java
private boolean drawSoftware(Surface surface, AttachInfo attachInfo, int yoff, boolean scalingRequired, Rect dirty) { // 使用軟件渲染繪制。 Canvas canvas; try { int left = dirty.left; int top = dirty.top; int right = dirty.right; int bottom = dirty.bottom; canvas = mSurface.lockCanvas(dirty); //從mSurface中lock一塊Canvas. The dirty rectangle can be modified by Surface.lockCanvas() //no inspection Constant Conditions if (left != dirty.left || top != dirty.top || right != dirty.right || bottom != dirty.bottom) { attachInfo.mIgnoreDirtyState = true; } // TODO: Do this in native canvas.setDensity(mDensity); } catch (Surface.OutOfResourcesException e) { handleOutOfResourcesException(e); return false; } catch (IllegalArgumentException e) { // Don't assume this is due to out of memory, it could be // something else, and if it is something else then we could // kill stuff (or ourself) for no reason. mLayoutRequested = true; // ask wm for a new surface next time. return false; } try { // 如果該位圖格式包含alpha通道，需要在繪制之前把它清除。這樣child就可以在透明背景下重新合成繪制。 // 如果使用了偏移，需要清理偏移不會出現的區域，以免垃圾遺留在空白區域上。 if (!canvas.isOpaque() || yoff != 0) { canvas.drawColor(0, PorterDuff.Mode.CLEAR); } dirty.setEmpty(); mIsAnimating = false; attachInfo.mDrawingTime = SystemClock.uptimeMillis(); mView.mPrivateFlags |= View.PFLAG_DRAWN; try { canvas.translate(0, -yoff); if (mTranslator != null) { mTranslator.translateCanvas(canvas); } canvas.setScreenDensity(scalingRequired ? mNoncompatDensity : 0); attachInfo.mSetIgnoreDirtyState = false; mView.draw(canvas); // 調用View.draw()由view自己在canvas上進行繪制。 drawAccessibilityFocusedDrawableIfNeeded(canvas); } finally { if (!attachInfo.mSetIgnoreDirtyState) { // Only clear the flag if it was not set during the mView.draw() call attachInfo.mIgnoreDirtyState = false; } } } finally { try { surface.unlockCanvasAndPost(canvas); // unlock canvas } catch (IllegalArgumentException e) { mLayoutRequested = true; // ask wm for a new surface next time. //noinspection ReturnInsideFinallyBlock return false; } } return true; }
draw()@View.java
// 在給定的canvas上手動渲染view及所有子view.這個函數被調用之前，該view必須已經完整走了一遍layout. // 當實現一個view時，應該實現onDraw()而不是重載這個函數。如果要重載這個函數，調用super.draw()。 public void draw(Canvas canvas) { if (mClipBounds != null) { canvas.clipRect(mClipBounds); } final int privateFlags = mPrivateFlags; final boolean dirtyOpaque = (privateFlags & PFLAG_DIRTY_MASK) == PFLAG_DIRTY_OPAQUE && (mAttachInfo == null || !mAttachInfo.mIgnoreDirtyState); mPrivateFlags = (privateFlags & ~PFLAG_DIRTY_MASK) | PFLAG_DRAWN; /* * Draw traversal performs several drawing steps which must be executed * in the appropriate order: * 1. Draw the background * 2. If necessary, save the canvas' layers to prepare for fading * 3. Draw view's content * 4. Draw children * 5. If necessary, draw the fading edges and restore layers * 6. Draw decorations (scrollbars for instance) */ // Step 1, draw the background, if needed int saveCount; if (!dirtyOpaque) { final Drawable background = mBackground; // 獲取xml中通過android:background屬性或者代碼中setBackgroundColor()，setBackgroundResource()等進行賦值的背景Drawable. if (background != null) { final int scrollX = mScrollX; final int scrollY = mScrollY; // 根據layout過程確定的view的位置設置背景繪制區域。 if (mBackgroundSizeChanged) { background.setBounds(0, 0, mRight - mLeft, mBottom - mTop); mBackgroundSizeChanged = false; } if ((scrollX | scrollY) == 0) { background.draw(canvas); // 調用Drawble.draw()完成繪制工作。 } else { canvas.translate(scrollX, scrollY); background.draw(canvas); canvas.translate(-scrollX, -scrollY); } } } // skip step 2 & 5 if possible (common case) final int viewFlags = mViewFlags; boolean horizontalEdges = (viewFlags & FADING_EDGE_HORIZONTAL) != 0; boolean verticalEdges = (viewFlags & FADING_EDGE_VERTICAL) != 0; if (!verticalEdges && !horizontalEdges) { // 沒有邊緣效果 // Step 3, draw the content if (!dirtyOpaque) onDraw(canvas); // 對view的內容進行繪制 // Step 4, draw the children dispatchDraw(canvas); // 對當前所有子view進行繪制，如果當前的view沒有子view，就不需要繪制。// Step 6, draw decorations (scrollbars). scroolbars第一次喚醒時進行繪制。
// 注意，每一個view都是有水平和垂直滾動條的，只是大多數情況不顯示而已。 onDrawScrollBars(canvas); if (mOverlay != null && !mOverlay.isEmpty()) { mOverlay.getOverlayView().dispatchDraw(canvas); } // we're done... return; } ......
}
View.onDraw()方法是一個空方法，而ViewGroup也沒有重寫該方法，需要子類去實現具體的onDraw()邏輯，這是因為每個view的內容是各不相同的。
View.dispatchDraw()是一個空函數，有子view時才進行繪制，看下在ViewGroup中的實現。
dispatchDraw()@ViewGroup.java
protected void dispatchDraw(Canvas canvas) { final int count = mChildrenCount; final View[] children = mChildren; int flags = mGroupFlags; ......
if ((flags & FLAG_USE_CHILD_DRAWING_ORDER) == 0) { for (int i = 0; i < count; i++) { final View child = children[i]; if ((child.mViewFlags & VISIBILITY_MASK) == VISIBLE || child.getAnimation() != null) { more |= drawChild(canvas, child, drawingTime); } } } else { for (int i = 0; i < count; i++) { final View child = children[getChildDrawingOrder(count, i)]; if ((child.mViewFlags & VISIBILITY_MASK) == VISIBLE || child.getAnimation() != null) { more |= drawChild(canvas, child, drawingTime); } } } // Draw any disappearing views that have animations if (mDisappearingChildren != null) { final ArrayList<View> disappearingChildren = mDisappearingChildren; final int disappearingCount = disappearingChildren.size() - 1; // Go backwards -- we may delete as animations finish for (int i = disappearingCount; i >= 0; i--) { final View child = disappearingChildren.get(i); more |= drawChild(canvas, child, drawingTime); } } ......// mGroupFlags might have been updated by drawChild() flags = mGroupFlags; if ((flags & FLAG_INVALIDATE_REQUIRED) == FLAG_INVALIDATE_REQUIRED) { invalidate(true); } ......
}
drawChild()@ViewGroup.java
protected boolean drawChild(Canvas canvas, View child, long drawingTime) { return child.draw(canvas, this, drawingTime); //調用View.draw()，child的繪制重新走上面流程。 }
performDraw()小結：
(1) View默認不會繪制內容的，需要子類自己去實現實際的繪制。
(2) View的繪制是利用onDraw()函數傳入的Canvas進行的。
(3) 看源碼中有很多關於動畫的邏輯，需要區分View動畫和ViewGroup布局動畫。View動畫是指View自身的動畫，可以通過setAnimation()進行設置。ViewGroup布局動畫是針對ViewGroup顯示內部子view時設置的動畫，可以在xml文件中對ViewGroup設置layoutAnimation屬性。
(4) 子view在獲取canvas的剪切區時不用關注padding，padding會被自動處理掉。
(5) 默認情況下子ViewGroup.drawChild()的順序與子view被添加的順序一致，可以通過ViewGroupDrawingOrder()函數提供不同的順序。
3.5 View.invalidate()和View.postInvalidate()介紹
上一篇文章中，ViewGroup.addView()調用了invalidate()函數，現在來介紹下這個函數的內容。
3.5.1 View.invalidate()的作用
invalidate()的作用是廢止整個view，如果這個view是可見的，那么在將來某個時間點onDraw()會被調用。這個函數必須在UI線程中調用，在非UI線程可以調用postInvalidate()。
這個函數有三種具體實現：
(1) void invalidate(boolean invalidateCache) : 完整的廢止會將drawing cache一起廢止, 當view的內容或者dimensions不變的時候，這個函數可以設置不廢止cache。
(2) void invalidate(Rect dirty) : 標記由dirty定義的區域為需要繪制的區域。
(3) void invalidate(int l, int t, int r, int b) : 標記由rect(l,t,r,b)定義的區域，dirty rect的坐標是相對於view的。
3.5.2 常見會引起invalidate()調用的原因
(1) 直接調用invalidate()方法請求重新draw，只會繪制調用者本身。
(2) 調用setSelection()方法請求重新draw，只會繪制調用者本身。
(3) 調用setVisibility()方法, 當view的狀態從INVISIBLE變成VISIBLE時會間接調用invalidate方法，從而繪制該view。當view的狀態從 INVISIBLEVISIBLE變為GONE狀態時會間接調用requestLayout和invalidate()方法，同時由於view樹大小發生 了變化，所以會請求measure過程以及draw過程，同樣只繪制需要重新繪制的view。
(4) 調用setEnabled()方法請求重新draw，不會重新繪制任何view包括該調用者本身。
(5) 調用requestFocus()方法請求view樹的draw過程，只繪制需要重繪的view。
3.5.3 invalidate()源碼分析
View.invalidate()會調用父view的invalidateChild()函數，實際上調用的是ViewGroup.invalidateChild()函數。
invalidateChild()@ViewGroup.java
public final void invalidateChild(View child, final Rect dirty) { ViewParent parent = this; ...... do { ...... parent = parent.invalidateChildInParent(location, dirty); //層層循環向上調用，知道ViewRootImpl返回null. ...... } while (parent != null); ...... }
invalidateChild()實際上是一個從當前view向上父view回溯的過程。每一層的父view都將自己的顯示區域與傳入的dirty Rect做交集。
invalidateChild()函數調用最后會傳到ViewRootImpl.invalidateChindInParent()。
invalidateChindInParent()@ViewRootImpl
public ViewParent invalidateChildInParent(int[] location, Rect dirty) { checkThread(); ...... final Rect localDirty = mDirty; ...... localDirty.union(dirty.left, dirty.top, dirty.right, dirty.bottom); // 融合dirty rect與當前的rect. ...... scheduleTraversals(); // 回到上文第3點， 會異步觸發執行performTraversals()。 ...... return null; }
ViewRootImpl.invalidateChindInParent()直接返回null，也就是ViewGroup.invalidateChild()中do-while循環結束的條件。
調用invalidate()的實質就是從當前view層層上傳到父級，直到ViewRootImpl觸發scheduleTraversals()函數，整個view樹重新走一遍上面的繪制流程。
3.5.4 postInvalidate()函數
View.invalidate()是在UI線程里運行的，在其它線程中只能調用View.postInvalidate()，該函數最終會通過ViewRootImpl類的handler發送MSG_INVALIDATE消息，最終又在UI線程中執行View.invalidate()。
3.6 requestLayout()函數
requestLayout()@View.java
public void requestLayout() { ...... if (mParent != null && !mParent.isLayoutRequested()) { mParent.requestLayout(); } ...... }
當view的requestLayout()函數被調用時，實際上是層層向父級傳遞，知道調用ViewRootImpl.requestLayout()為止。
requestLayout()@ViewRootImpl.java
public void requestLayout() { if (!mHandlingLayoutInLayoutRequest) { checkThread(); mLayoutRequested = true; scheduleTraversals(); } }
scheduleTraversals()即時序圖中的第3個調用，后面的流程與上面描述的類似了。需要注意的是requestLayout()函數會觸發measure和layout，但不會draw，不會重新繪制包括調用者本身的任何view。
4. addToDisplay()@Session.java
通過addToDisplay()將window對象添加到WMS中，WMS將管理這個window。
public int addToDisplay(IWindow window, int seq, WindowManager.LayoutParams attrs, int viewVisibility, int displayId, Rect outContentInsets, InputChannel outInputChannel) { // 調用WindowManagerService的addWindow()函數 return mService.addWindow(this, window, seq, attrs, viewVisibility, displayId, outContentInsets, outInputChannel); }
addWindow()@WindowManagerService.java
public int addWindow(Session session, IWindow client, int seq, WindowManager.LayoutParams attrs, int viewVisibility, int displayId, Rect outContentInsets, InputChannel outInputChannel) { ...... WindowState win = null; ...... // 創建一個WindowState， WMS中維護的是WindowState對象. win = new WindowState(this, session, client, token, attachedWindow, appOp[0], seq, attrs, viewVisibility, displayContent); ...... win.attach(); // 調用WindowState.attach()函數。 mWindowMap.put(client.asBinder(), win); ...... return res; }
attach()@WindowState.java
void attach() { mSession.windowAddedLocked(); }
windowAddedLocked()@Session.java
void windowAddedLocked() { if (mSurfaceSession == null) { mSurfaceSession = new SurfaceSession(); //創建SurfaceSeesion對象 mService.mSessions.add(this); } mNumWindow++; }
（時序圖中調用26以后就是在WindowManagerService中與底層交互，控制底層顯示實際內容了，在本文中暫不考慮)