1,AwesomeEvent 这个是同步相应的事件而做的一个类,跟framework层的looper和handler作用相似,player有一些异步操作比如解析文件,这些操作比较耗时,做异步操作然后做回调会有更好的用户体验
struct AwesomeEvent : public TimedEventQueue::Event
继承自TimedEventQueue::Event ,关于c++的内部类可以参考以前的blog,TimedEventQueue会起一个thread,来做event的post操作
2, AwesomeRemoteRenderer
AwesomeLocalRenderer
这两个类是往surface上post数据,而remote据说是因为omx node的使用,还没有细研究,而localrenderer会根据有无硬件加速调用不同的方法
3 构造函数会初始化event,这些后面再说
mediaplayer的一般方法流程是
setdatasource
prepareAsync
start
现在只针对filesource做流程分析
setdatasource 一般会设置file的uri
mUri = uri;现在只是设置一下uri,下面有段注释
// The actual work will be done during preparation in the call to
// ::finishSetDataSource_l to avoid blocking the calling thread in
// setDataSource for any significant time.
现在不会做其他的操作,因为会消耗太多的时间有可能导致上层的ANR,这个操作是同步操作
重头戏就在prepareAsync里面了,看这个函数
if (!mQueueStarted) {
mQueue.start();
mQueueStarted = true;
}//会把eventqueue打开,为以后的event post做准备,这个时候会建立一个线程
mAsyncPrepareEvent = new AwesomeEvent(
this, &AwesomePlayer::onPrepareAsyncEvent);
mQueue.postEvent(mAsyncPrepareEvent);
//为了能异步,postevent
到onPrepareAsyncEvent()
if (mUri.size() > 0) {
status_t err = finishSetDataSource_l();
if (err != OK) {
abortPrepare(err);
return;
}
}
finishSetDataSource_l();//这个就是上面注释的那个函数 这个函数会根据source类型建立不同的extractor
针对filesource,实际就是这个dataSource = DataSource::CreateFromURI(mUri.string(), &mUriHeaders);//uriheaders为null,不用管它
接着跳转 到source = new FileSource(uri+7);建立filesource,它会注册不同的extractor,sniffer出不同的mimetype,操作filesource
接着回到 finishSetDataSource_l(),
sp<MediaExtractor> extractor = MediaExtractor::Create(dataSource,mime);
这个时候datasource会sniffer出不同的extractor,从而根据mime type建立不同的extractor
最后就是setdatasource这个本来应该第一步做的操作
这个时候会解析出audio track和viedo track,为下面的codec做准备
if (mVideoTrack != NULL && mVideoSource == NULL) {
// Use NPT timestamps if playing
// RTSP streaming with video only content
// (no audio to drive the clock for media time)
uint32_t flags = 0;
if (mRTSPController != NULL && mAudioTrack == NULL) {
flags |= OMXCodec::kUseNptTimestamp;
}
status_t err = initVideoDecoder(flags);
if (err != OK) {
abortPrepare(err);
return;
}
}
if (mAudioTrack != NULL && mAudioSource == NULL) {
status_t err = initAudioDecoder();
if (err != OK) {
abortPrepare(err);
return;
}
}
这个就是涉及最关键的部分了 codec部分
从上面下来 audiotrack 和videotrack都已经有了 这个时候就要初始化codec了
initVideoDecoder(flags);
initAudioDecoder();
{
mVideoSource = OMXCodec::Create(
mClient.interface(), mVideoTrack->getFormat(),
false, // createEncoder
mVideoTrack,
NULL, flags);
OMXCodec是个很关键的类,这个就到了前几天写的stagefright的code里面了,我会继续写那个,不在这里说了,总之会返回一个Videosource,这个是根据videotrack解码过的数据
if (mVideoSource != NULL) {
int64_t durationUs;
if (mVideoTrack->getFormat()->findInt64(kKeyDuration, &durationUs)) {
Mutex::Autolock autoLock(mMiscStateLock);
if (mDurationUs < 0 || durationUs > mDurationUs) {
mDurationUs = durationUs;
}
}
CHECK(mVideoTrack->getFormat()->findInt32(kKeyWidth, &mVideoWidth));
CHECK(mVideoTrack->getFormat()->findInt32(kKeyHeight, &mVideoHeight));
status_t err = mVideoSource->start(); //这个时候videosource和viedotrack都会准备好,申请buffer
if (err != OK) {
mVideoSource.clear();
return err;
}
}
return mVideoSource != NULL ? OK : UNKNOWN_ERROR;
}
弄好以后就会更改状态,做上层的prepareAsync的回调
数据已经准备了,
play_l()
if ((mVideoSource != NULL) && (!mVideoBuffer))
{
// Changes to fix Audio starts playing before video.
// First video frame is returned late as it is referenced to decode subsequent P and B frames.
// For higher resolutions (e.g. 1080p) this returning time is significant.
// We need to trigger Video decoder earlier than audio so that Video catch up with audio in time.
MediaSource::ReadOptions options;
if (mSeeking) {
LOGV("seeking to %lld us (%.2f secs)", mSeekTimeUs, mSeekTimeUs / 1E6);
options.setSeekTo(
mSeekTimeUs, MediaSource::ReadOptions::SEEK_CLOSEST_SYNC);
}
for (;;) {
status_t err = mVideoSource->read(&mVideoBuffer, &options);
options.clearSeekTo();
if (err != OK) {
CHECK_EQ(mVideoBuffer, NULL);
if (err == INFO_FORMAT_CHANGED) {
LOGV("VideoSource signalled format change.");
if (mVideoRenderer != NULL) {
mVideoRendererIsPreview = false;
initRenderer_l();
}
continue;
}
break;
}
break;
}
}
if (mVideoSource != NULL) {
// Kick off video playback
postVideoEvent_l();
}
就会把得到的mVideoBuffer render到surface上面
而audio会调用audioplayer
mAudioPlayer = new AudioPlayer(mAudioSink, this);
mAudioPlayer->setSource(mAudioSource);
// We've already started the MediaSource in order to enable
// the prefetcher to read its data.
status_t err = mAudioPlayer->start(
true /* sourceAlreadyStarted */);
还有音视频同步,seek的问题,下次再写
ps-----这篇是针对2.3的代码,4.0以后又发生了一些变化,后面的问题就没有写完,会写一些4.0相关的东西