A Fragment is a piece of an application's user interface or behavior that can be placed in an Activity.
Fragment放置在Activity容器中,通常用来作为UI的片段,在日常的开发中也有着广泛的应用,先来看一段常用的代码。
DemoFragment demoFragment = DemoFragment.newInstance("param1", "param2");
Bundle bundle = new Bundle();
demoFragment.setArguments(bundle);
getSupportFragmentManager().beginTransaction()
.add(R.id.fragment_container, demoFragment)
.commit();
这是我们非常常见的代码,借着这段代码,引出我们今天的主题:针对Fragment的全面的源码分析。
Fragment的操作是一种事务操作,什么是事务?🤔简单来说就是一个原子操作,要么被成功执行,否则原来的操作会回滚,各个操作彼此之间互不干扰。我们先整体看下Fragment的操作
序列图。
嗯,看起来有点长😌,不要方,我们先来看看这里面频繁出现的几个类的作用。
Fragment的操作方法一共有七种:
final class BackStackRecord extends FragmentTransaction implements
FragmentManager.BackStackEntry, FragmentManagerImpl.OpGenerator {
@Override
public FragmentTransaction add(Fragment fragment, String tag) {
doAddOp(0, fragment, tag, OP_ADD);
return this;
}
@Override
public FragmentTransaction add(int containerViewId, Fragment fragment) {
doAddOp(containerViewId, fragment, null, OP_ADD);
return this;
}
@Override
public FragmentTransaction add(int containerViewId, Fragment fragment, String tag) {
doAddOp(containerViewId, fragment, tag, OP_ADD);
return this;
}
@Override
public FragmentTransaction replace(int containerViewId, Fragment fragment) {
return replace(containerViewId, fragment, null);
}
@Override
public FragmentTransaction replace(int containerViewId, Fragment fragment, String tag) {
if (containerViewId == 0) {
throw new IllegalArgumentException("Must use non-zero containerViewId");
}
doAddOp(containerViewId, fragment, tag, OP_REPLACE);
return this;
}
@Override
public FragmentTransaction remove(Fragment fragment) {
Op op = new Op();
op.cmd = OP_REMOVE;
op.fragment = fragment;
addOp(op);
return this;
}
@Override
public FragmentTransaction hide(Fragment fragment) {
Op op = new Op();
op.cmd = OP_HIDE;
op.fragment = fragment;
addOp(op);
return this;
}
@Override
public FragmentTransaction show(Fragment fragment) {
Op op = new Op();
op.cmd = OP_SHOW;
op.fragment = fragment;
addOp(op);
return this;
}
@Override
public FragmentTransaction detach(Fragment fragment) {
Op op = new Op();
op.cmd = OP_DETACH;
op.fragment = fragment;
addOp(op);
return this;
}
@Override
public FragmentTransaction attach(Fragment fragment) {
Op op = new Op();
op.cmd = OP_ATTACH;
op.fragment = fragment;
addOp(op);
return this;
}
}
你可以发现,这些方法最终都调用了addOp()方法,Op是什么?🤔Op封装了操作命令、Fragment、动画等内容。上面我们说过BackStackRecord将Fragment与相应应的操作包装起来,传递给FragmentManager调用。
static final class Op {
int cmd;
Fragment fragment;
int enterAnim;
int exitAnim;
int popEnterAnim;
int popExitAnim;
}
cmd对应了响应的操作。
static final int OP_NULL = 0;
static final int OP_ADD = 1;
static final int OP_REPLACE = 2;
static final int OP_REMOVE = 3;
static final int OP_HIDE = 4;
static final int OP_SHOW = 5;
static final int OP_DETACH = 6;
static final int OP_ATTACH = 7;
我们来看看addOp()方法的实现。
final class BackStackRecord extends FragmentTransaction implements
FragmentManager.BackStackEntry, FragmentManagerImpl.OpGenerator {
ArrayList<Op> mOps = new ArrayList<>();
void addOp(Op op) {
mOps.add(op);
op.enterAnim = mEnterAnim;
op.exitAnim = mExitAnim;
op.popEnterAnim = mPopEnterAnim;
op.popExitAnim = mPopExitAnim;
}
}
上面代码的最后一步是commit()方法,该方法提交事务操作,我们来看看它的实现。
final class BackStackRecord extends FragmentTransaction implements
FragmentManager.BackStackEntry, FragmentManagerImpl.OpGenerator {
@Override
public int commit() {
return commitInternal(false);
}
//allowStateLoss是个标志位,表示是否允许状态丢失
int commitInternal(boolean allowStateLoss) {
if (mCommitted) throw new IllegalStateException("commit already called");
if (FragmentManagerImpl.DEBUG) {
Log.v(TAG, "Commit: " + this);
LogWriter logw = new LogWriter(TAG);
PrintWriter pw = new PrintWriter(logw);
dump(" ", null, pw, null);
pw.close();
}
mCommitted = true;
if (mAddToBackStack) {
mIndex = mManager.allocBackStackIndex(this);
} else {
mIndex = -1;
}
mManager.enqueueAction(this, allowStateLoss);
return mIndex;
}
}
可以看到BackStackRecord完成了对Fragment操作的封装,并比较给FragmentManager调用。
从上面的序列图我们可以看出,在commit()方法执行后,会调用FragmentManager.enqueueAction()方法,并通过handler.post()切换到主线程去执行这个Action,执行时间未知。
这个handler正是FragmentActivity里创建的Handler。
final class BackStackRecord extends FragmentTransaction implements
FragmentManager.BackStackEntry, FragmentManagerImpl.OpGenerator {
void executeOps() {
final int numOps = mOps.size();
for (int opNum = 0; opNum < numOps; opNum++) {
final Op op = mOps.get(opNum);
final Fragment f = op.fragment;
f.setNextTransition(mTransition, mTransitionStyle);
//Fragment操作
switch (op.cmd) {
case OP_ADD:
f.setNextAnim(op.enterAnim);
mManager.addFragment(f, false);
break;
case OP_REMOVE:
f.setNextAnim(op.exitAnim);
mManager.removeFragment(f);
break;
case OP_HIDE:
f.setNextAnim(op.exitAnim);
mManager.hideFragment(f);
break;
case OP_SHOW:
f.setNextAnim(op.enterAnim);
mManager.showFragment(f);
break;
case OP_DETACH:
f.setNextAnim(op.exitAnim);
mManager.detachFragment(f);
break;
case OP_ATTACH:
f.setNextAnim(op.enterAnim);
mManager.attachFragment(f);
break;
default:
throw new IllegalArgumentException("Unknown cmd: " + op.cmd);
}
if (!mAllowOptimization && op.cmd != OP_ADD) {
mManager.moveFragmentToExpectedState(f);
}
}
if (!mAllowOptimization) {
// Added fragments are added at the end to comply with prior behavior.
mManager.moveToState(mManager.mCurState, true);
}
}
}
因而,Fragment的操作:
都转换成了FragmentManager的方法:
并调用FragmentManager.moveToState()方法做Fragment的状态迁移。上述的这几种Fragment的操作方法都做了哪些事情呢?🤔
要理解以下方法,我们要先看看Fragment里的几个标志位的含义。
final class FragmentManagerImpl extends FragmentManager implements LayoutInflaterFactory {
//添加Fragment
public void addFragment(Fragment fragment, boolean moveToStateNow) {
if (mAdded == null) {
mAdded = new ArrayList<Fragment>();
}
if (DEBUG) Log.v(TAG, "add: " + fragment);
makeActive(fragment);
if (!fragment.mDetached) {
if (mAdded.contains(fragment)) {
throw new IllegalStateException("Fragment already added: " + fragment);
}
synchronized (mAdded) {
mAdded.add(fragment);
}
fragment.mAdded = true;
fragment.mRemoving = false;
if (fragment.mView == null) {
fragment.mHiddenChanged = false;
}
if (fragment.mHasMenu && fragment.mMenuVisible) {
mNeedMenuInvalidate = true;
}
if (moveToStateNow) {
moveToState(fragment);
}
}
}
//移除Fragment
public void removeFragment(Fragment fragment) {
if (DEBUG) Log.v(TAG, "remove: " + fragment + " nesting=" + fragment.mBackStackNesting);
final boolean inactive = !fragment.isInBackStack();
if (!fragment.mDetached || inactive) {
if (mAdded != null) {
synchronized (mAdded) {
mAdded.remove(fragment);
}
}
if (fragment.mHasMenu && fragment.mMenuVisible) {
mNeedMenuInvalidate = true;
}
fragment.mAdded = false;
fragment.mRemoving = true;
}
}
//隐藏Fragment:将一个Fragment标记成将要隐藏状态,显示工作有completeShowHideFragment(}方法完成
public void hideFragment(Fragment fragment) {
if (DEBUG) Log.v(TAG, "hide: " + fragment);
if (!fragment.mHidden) {
fragment.mHidden = true;
// Toggle hidden changed so that if a fragment goes through show/hide/show
// it doesn't go through the animation.
fragment.mHiddenChanged = !fragment.mHiddenChanged;
}
}
//显示Fragment:将一个Fragment标记成将要显示状态,显示工作有completeShowHideFragment(}方法完成
public void showFragment(Fragment fragment) {
if (DEBUG) Log.v(TAG, "show: " + fragment);
if (fragment.mHidden) {
fragment.mHidden = false;
// Toggle hidden changed so that if a fragment goes through show/hide/show
// it doesn't go through the animation.
fragment.mHiddenChanged = !fragment.mHiddenChanged;
}
}
//将Fragment从宿主Activity分离
public void detachFragment(Fragment fragment) {
if (DEBUG) Log.v(TAG, "detach: " + fragment);
if (!fragment.mDetached) {
fragment.mDetached = true;
if (fragment.mAdded) {
// We are not already in back stack, so need to remove the fragment.
if (mAdded != null) {
if (DEBUG) Log.v(TAG, "remove from detach: " + fragment);
synchronized (mAdded) {
mAdded.remove(fragment);
}
}
if (fragment.mHasMenu && fragment.mMenuVisible) {
mNeedMenuInvalidate = true;
}
fragment.mAdded = false;
}
}
}
//将Fragment关联3到宿主Activity
public void attachFragment(Fragment fragment) {
if (DEBUG) Log.v(TAG, "attach: " + fragment);
if (fragment.mDetached) {
fragment.mDetached = false;
if (!fragment.mAdded) {
if (mAdded == null) {
mAdded = new ArrayList<Fragment>();
}
if (mAdded.contains(fragment)) {
throw new IllegalStateException("Fragment already added: " + fragment);
}
if (DEBUG) Log.v(TAG, "add from attach: " + fragment);
synchronized (mAdded) {
mAdded.add(fragment);
}
fragment.mAdded = true;
if (fragment.mHasMenu && fragment.mMenuVisible) {
mNeedMenuInvalidate = true;
}
}
}
}
}
可以看到这些方法大体类似,差别在于它们处理的标志位不同,这也导致了后续的moveToState()在处理它们的时候回区别对待,具体说来:
detach后的Fragment可以再attach,而remove后的Fragment却不可以,只能重新add。
理解完了Fragment的操作,我们再来看看它的生命周期的变化,这也是我们的重点。
我们先来看一张完整的Fragment生命周期图。
我们都知道Fragment的生命周期依赖于它的宿主Activity,但事实的情况却并不这么简单。
在FragmentManager中,完成Fragment状态变换的主要有四个方法:
moveToState(Fragment f)
moveToState(int newState, boolean always)
moveFragmentToExpectedState(Fragment f)
moveToState(Fragment f, int newState, int transit, int transitionStyle, boolean keepActive)
它们的触发流程也很简单,比方说FragmentActivity触发了onResume()方法。
public class FragmentActivity extends BaseFragmentActivityJB implements
ActivityCompat.OnRequestPermissionsResultCallback,
ActivityCompatApi23.RequestPermissionsRequestCodeValidator {
@Override
protected void onDestroy() {
super.onDestroy();
doReallyStop(false);
mFragments.dispatchDestroy();
mFragments.doLoaderDestroy();
}
}
它会去调用Fragment的dispatchDestory()方法,Fragment又接着会去调用FragmentManager的dispatchDestory()方法。
final class FragmentManagerImpl extends FragmentManager implements LayoutInflaterFactory {
public void dispatchDestroy() {
mDestroyed = true;
execPendingActions();
mExecutingActions = true;
moveToState(Fragment.INITIALIZING, false);
mExecutingActions = false;
mHost = null;
mContainer = null;
mParent = null;
}
}
最终这些处理都会回归到上面这四个方法中来,而这四个方法最终发挥作用当然是最后一个参数最多的方法,其他的方法都只是做了参数的处理和情况的判断。
Fragment定义了六种状态
static final int INITIALIZING = 0; // 未创建
static final int CREATED = 1; // 已创建
static final int ACTIVITY_CREATED = 2; // 宿主Activity已经结束创建
static final int STOPPED = 3; // Fragment的onCreate()方法已完成,onStart()即将开始
static final int STARTED = 4; // Fragment的onCreate()和onStart()方法都已完成,onResume()即将开始
static final int RESUMED = 5; // Fragment的onCreate()、onStart()和onResume()方法都已完成
final class FragmentManagerImpl extends FragmentManager implements LayoutInflaterFactory {
void moveToState(Fragment f, int newState, int transit, int transitionStyle,
boolean keepActive) {
//状态判断
...
//当前状态大于新状态,从上面的状态表可以看出,状态值越小
//就说明处于越早的阶段,一般对应add等操作
if (f.mState < newState) {
...
switch (f.mState) {
//未创建
case Fragment.INITIALIZING:
...
f.onAttach(mHost.getContext());
...
//Fragment被定义在布局文件里的情形,需要先从布局文件里inflate出view
if (f.mFromLayout) {
f.mView = f.performCreateView(f.performGetLayoutInflater(
f.mSavedFragmentState), null, f.mSavedFragmentState);
if (f.mView != null) {
...
f.onViewCreated(f.mView, f.mSavedFragmentState);
} else {
f.mInnerView = null;
}
}
//已创建
case Fragment.CREATED:
if (newState > Fragment.CREATED) {
if (!f.mFromLayout) {
...
if (f.mView != null) {
...
f.onViewCreated(f.mView, f.mSavedFragmentState);
dispatchOnFragmentViewCreated(f, f.mView, f.mSavedFragmentState,
false);
...
} else {
f.mInnerView = null;
}
}
...
f.performActivityCreated(f.mSavedFragmentState);
...
}
case Fragment.ACTIVITY_CREATED:
if (newState > Fragment.ACTIVITY_CREATED) {
f.mState = Fragment.STOPPED;
}
case Fragment.STOPPED:
if (newState > Fragment.STOPPED) {
if (DEBUG) Log.v(TAG, "moveto STARTED: " + f);
f.performStart();
dispatchOnFragmentStarted(f, false);
}
case Fragment.STARTED:
if (newState > Fragment.STARTED) {
if (DEBUG) Log.v(TAG, "moveto RESUMED: " + f);
f.performResume();
dispatchOnFragmentResumed(f, false);
f.mSavedFragmentState = null;
f.mSavedViewState = null;
}
}
}
//当前状态大于新状态,一般对应remove等操作
else if (f.mState > newState) {
switch (f.mState) {
case Fragment.RESUMED:
if (newState < Fragment.RESUMED) {
if (DEBUG) Log.v(TAG, "movefrom RESUMED: " + f);
f.performPause();
dispatchOnFragmentPaused(f, false);
}
case Fragment.STARTED:
if (newState < Fragment.STARTED) {
if (DEBUG) Log.v(TAG, "movefrom STARTED: " + f);
f.performStop();
dispatchOnFragmentStopped(f, false);
}
case Fragment.STOPPED:
if (newState < Fragment.STOPPED) {
if (DEBUG) Log.v(TAG, "movefrom STOPPED: " + f);
f.performReallyStop();
}
case Fragment.ACTIVITY_CREATED:
if (newState < Fragment.ACTIVITY_CREATED) {
...
f.performDestroyView();
dispatchOnFragmentViewDestroyed(f, false);
...
}
case Fragment.CREATED:
if (newState < Fragment.CREATED) {
...
if (f.getAnimatingAway() != null) {
f.setStateAfterAnimating(newState);
newState = Fragment.CREATED;
} else {
....
if (!f.mRetaining) {
f.performDestroy();
dispatchOnFragmentDestroyed(f, false);
} else {
f.mState = Fragment.INITIALIZING;
}
f.performDetach();
dispatchOnFragmentDetached(f, false);
...
}
}
}
}
...
}
}
可以发现进入该方法后会先将Fragment的当前状态与新状态进行比较:
这样便完成了Fragment状态的迁移和生命周期方法的回调。
什么是Fragment回退栈呢?🤔
这个很好理解,和Activity栈相似,放在Activity里的Fragment,如果不做额外处理的话,在点击返回的时候,会直接finish当前Activity,Fragment回退栈就是用来处理Fragment返回的问题。
Fragment的回退栈也是由Fragment来管理的,关于FragmentManger的获取,一是FragmentActivity里的getSupportFragmentManager(),二是Fragment里的getChildFragmentManager(),它们
返回的都是FragmentManagerImpl对象,对Fragment的栈进行管理。
我们先来看看常用的栈操作方法。
入栈
入栈操作通过etSupportFragmentManager.beiginTransaction().addToBackStack()方法完成,它的具体实现在BackRecordStack里。
addToBackStack(String name):入栈,这个方法的实现很简单,就是将BackRecordStack的成员变量mName赋值,mAddToBackStack置true,表示自己要添加进回退栈, 这样在调用commit()方法提交操作时,FragmentManager
会为该Fragment分配栈索引,并将它添加进回退栈列表,供后续出栈的时候调用。
出栈
出栈操作是通过getSupportFragmentManager.popBackStack()等方法完成的,它的具体实现在FragmentManagerImpl里。
我们再来看看这些方法的实现。
final class FragmentManagerImpl extends FragmentManager implements LayoutInflaterFactory {
@Override
public void popBackStack() {
enqueueAction(new PopBackStackState(null, -1, 0), false);
}
@Override
public boolean popBackStackImmediate() {
checkStateLoss();
return popBackStackImmediate(null, -1, 0);
}
}
PopBackStackState实现了OpGenerator接口,封装了将要出栈的Fragment的信息,包括mName、mId与mFlags信息。如果你有细心看,上面我们提到的FragmentTransaction的实现类BackStackRecord
也实现了这个接口。
private class PopBackStackState implements OpGenerator {
final String mName;
final int mId;
final int mFlags;
PopBackStackState(String name, int id, int flags) {
mName = name;
mId = id;
mFlags = flags;
}
@Override
public boolean generateOps(ArrayList<BackStackRecord> records,
ArrayList<Boolean> isRecordPop) {
return popBackStackState(records, isRecordPop, mName, mId, mFlags);
}
}
popBackStack()也调用了enqueueAction()方法,后续的流程和上面的Fragment操作流程是一样的,出栈操作最终对应的是Fragment的remove()操作,因此它对Fragment生命周期的影响和remove()操作相同。
至于popBackStackImmediate()的实现,则就是直接调用执行操作的方法,少了加入队列的等待过程,具体流程也和上面的Fragment操作一样。
final class FragmentManagerImpl extends FragmentManager implements LayoutInflaterFactory {
private boolean popBackStackImmediate(String name, int id, int flags) {
execPendingActions();
ensureExecReady(true);
boolean executePop = popBackStackState(mTmpRecords, mTmpIsPop, name, id, flags);
if (executePop) {
mExecutingActions = true;
try {
optimizeAndExecuteOps(mTmpRecords, mTmpIsPop);
} finally {
cleanupExec();
}
}
doPendingDeferredStart();
burpActive();
return executePop;
}
}
Copyright© 2013-2019