书接上文，我们接着看如何利用FFmpeg解码音视频
代码示例
static jint
MediaPlayer_play(JNIEnv *env, jobject thiz, jstring url, jobject surface) {
int ret = -1;
const char *input = env->GetStringUTFChars(url, NULL);
if (input == NULL) {
LOGE("input is invalid");
return ret;
}
av_register_all();
AVFormatContext *formatContext = avformat_alloc_context();
int openState = avformat_open_input(&formatContext, input, NULL, NULL);
if (openState < 0) {
char errbuf[128];
if (av_strerror(openState, errbuf, sizeof(errbuf)) == 0) {
LOGE("open file error, the rease is %s", errbuf);
}
return ret;
}
av_dump_format(formatContext, 0, input, 0);
if (avformat_find_stream_info(formatContext, NULL) < 0) {
LOGE("read stream error");
return ret;
}
int video_stream_index = -1;
LOGD("int this video file, the streams num：%d", formatContext->nb_streams);
for (int i = 0; i < formatContext->nb_streams; i++) {
if (formatContext->streams[i]->codecpar->codec_type == AVMEDIA_TYPE_VIDEO) {
video_stream_index = i;
break;
}
}
//audio decode
int audio_stream_index = -1;
for (int i = 0; i < formatContext->nb_streams; i++) {
if (formatContext->streams[i]->codecpar->codec_type == AVMEDIA_TYPE_AUDIO) {
audio_stream_index = i;
break;
}
}
LOGE("audio_stream_index=%d", audio_stream_index);
if (audio_stream_index == -1) {
LOGE("no audio stream");
return ret;
}
if (video_stream_index == -1) {
LOGE("vide stream is error");
return ret;
}
AVCodecParameters *codecParameters = formatContext->streams[video_stream_index]->codecpar;
AVCodec *videoDecoder = avcodec_find_decoder(codecParameters->codec_id);
if (videoDecoder == NULL) {
LOGD("can not find codec for this video");
return ret;
}
AVCodecParameters *audioCodecParameters = formatContext->streams[audio_stream_index]->codecpar;
AVCodec *audioDecoder = avcodec_find_decoder(audioCodecParameters->codec_id);
AVCodecContext *codecContext = avcodec_alloc_context3(videoDecoder);
if (codecContext == NULL) {
LOGE("alloc codec context failed!");
return ret;
}
AVCodecContext *audioContext = avcodec_alloc_context3(audioDecoder);
if (audioContext == NULL) {
LOGE("alloc audio context failed!");
return ret;
}
if (avcodec_parameters_to_context(codecContext, codecParameters) < 0) {
LOGE("inflate codec context failed!");
return ret;
}
if (avcodec_parameters_to_context(audioContext, audioCodecParameters) < 0) {
LOGE("inflate audio codec context failed!");
return ret;
}
if (avcodec_open2(codecContext, videoDecoder, NULL) < 0) {
LOGE("init context error");
return ret;
}
if (avcodec_open2(audioContext, audioDecoder, NULL) < 0) {
LOGE("init audio context error");
return ret;
}
AVPixelFormat dstFormat = AV_PIX_FMT_RGBA;
AVPacket *packet = av_packet_alloc();
AVFrame *frame = av_frame_alloc();
//分配解码后的每一数据信息的结构体（指针）
AVFrame *outFrame = av_frame_alloc();
//分配最终显示出来的目标帧信息的结构体（指针）
uint8_t *out_buffer = (uint8_t *)av_malloc(av_image_get_buffer_size(dstFormat, codecContext->width, codecContext -> height, 1));
av_image_fill_arrays(outFrame->data, outFrame->linesize,
out_buffer, dstFormat, codecContext->width, codecContext->height, 1);
SwsContext *swsContext = sws_getContext(codecContext->width,
codecContext->height,
codecContext->pix_fmt,
codecContext->width,
codecContext->height,
dstFormat,
SWS_BICUBIC,
NULL,
NULL,
NULL);
if (swsContext == NULL) {
LOGE("swsContext is NULL");
return ret;
}
ANativeWindow *nativeWindow = ANativeWindow_fromSurface(env, surface);
//定义绘图缓冲区
ANativeWindow_Buffer outBuffer;
SwrContext *swrCtx = swr_alloc();
enum AVSampleFormat in_sample_fmt = audioContext->sample_fmt;
enum AVSampleFormat out_sample_fmt = AV_SAMPLE_FMT_S16;
int in_sample_rate = audioContext->sample_rate;
int out_sample_rate = 44100;
LOGE("run_audio_decodec start");
uint64_t in_ch_layout = audioContext->channel_layout;
uint64_t out_ch_layout = AV_CH_LAYOUT_STEREO;
swr_alloc_set_opts(swrCtx, out_ch_layout, out_sample_fmt, out_sample_rate,
in_ch_layout, in_sample_fmt, in_sample_rate, 0, NULL);
swr_init(swrCtx);
int out_channel_nb = av_get_channel_layout_nb_channels(out_ch_layout);
uint8_t *out_buffer_audio = static_cast<uint8_t *>(av_malloc(MAX_AUDIO_FRME_SIZE));
audioTrack = static_cast<AudioTrack *>(malloc(sizeof(AudioTrack)));
jobject at = env->CallObjectMethod(javaCallback->thiz, javaCallback->createAudioTrackId, out_sample_rate, out_channel_nb);
audioTrack->audiotrack = env->NewGlobalRef(at);
audioTrack->audio_track_class = env->GetObjectClass(audioTrack->audiotrack);
audioTrack->play_id = env->GetMethodID(audioTrack->audio_track_class, "play", "()V");
audioTrack->stop_id = env->GetMethodID(audioTrack->audio_track_class, "stop", "()V");
audioTrack->write_id = env->GetMethodID(audioTrack->audio_track_class, "write", "([BII)I");
env->CallVoidMethod(audioTrack->audiotrack, audioTrack->play_id);
//通过设置宽高限制缓冲区中的像素数量，而非屏幕的物流显示尺寸。
//如果缓冲区与物理屏幕的显示尺寸不相符，则实际显示可能会是拉伸，或者被压缩的图像
ANativeWindow_setBuffersGeometry(nativeWindow, codecContext->width, codecContext->height,
WINDOW_FORMAT_RGBA_8888);
while (av_read_frame(formatContext, packet) == 0) {
if (packet->stream_index == video_stream_index) {
int sendPacketState = avcodec_send_packet(codecContext, packet);
if (sendPacketState == 0) {
int receiveFrameState = avcodec_receive_frame(codecContext, frame);
if (receiveFrameState == 0) {
//锁定窗口绘图界面
ANativeWindow_lock(nativeWindow, &outBuffer, NULL);
sws_scale(swsContext,
((const uint8_t *const*) frame->data),
frame->linesize,
0,
frame->height,
outFrame->data,
outFrame->linesize);
uint8_t *dst = static_cast<uint8_t *>(outBuffer.bits);
//解码后的像素数据首地址
//这里由于使用的是RGBA格式，所以解码图像数据只保存在data[0]中。但如果是YUV就会有data[0]
//data[1],data[2]
uint8_t *src = outFrame->data[0];
int oneLineByte = outBuffer.stride * 4;
int srcStride = outFrame->linesize[0];
for (int i = 0; i < codecContext->height; i++) {
memcpy(dst + i * oneLineByte, src + i * srcStride, srcStride);
}
ANativeWindow_unlockAndPost(nativeWindow);
//进行短暂休眠。如果休眠时间太长会导致播放的每帧画面有延迟感，如果短会有加速播放的感觉。
//一般一每秒60帧——16毫秒一帧的时间进行休眠
usleep(1000 * 20);
} else if (receiveFrameState == AVERROR(EAGAIN)) {
LOGE("decode error AVERROR(EAGAIN)");
} else if (receiveFrameState == AVERROR_EOF) {
LOGE("decode error AVERROR_EOF");
} else if (receiveFrameState == AVERROR(EINVAL)) {
LOGE("decode error AVERROR(EINVAL)");
} else if (receiveFrameState == AVERROR(ENOMEM)) {
LOGE("decode error AVERROR(ENOMEM)");
} else {
LOGE("decode error unknown");
}
} else if(sendPacketState == AVERROR(EAGAIN)) {
LOGD("send decode failed：AVERROR(EAGAIN)");
} else if (sendPacketState == AVERROR_EOF) {
LOGD("send decode failed：AVERROR_EOF");
} else if (sendPacketState == AVERROR(EINVAL)) {
LOGD("send decode failed：AVERROR(EINVAL)");
} else if (sendPacketState == AVERROR(ENOMEM)) {
LOGD("send decode failed：AVERROR(ENOMEM)");
} else {
LOGD("send decode failed：unknow");
}
} else if (packet->stream_index == audio_stream_index) {
int sendPacketState = avcodec_send_packet(audioContext, packet);
if (sendPacketState == 0) {
while (avcodec_receive_frame(audioContext, frame) == 0) {
AVRational tb = (AVRational){1, frame->sample_rate};
if (frame->pts != AV_NOPTS_VALUE) {
frame->pts = av_rescale_q(frame->pts, audioContext->time_base, tb);
} else if (frame->pts != AV_NOPTS_VALUE) {
frame->pts = av_rescale_q(frame->pkt_pts, av_codec_get_pkt_timebase(audioContext), tb);
} else {
frame->pts = packet->dts;
}
int len2 = swr_convert(swrCtx, &out_buffer_audio, MAX_AUDIO_FRME_SIZE,
(const uint8_t **)(frame->extended_data), frame->nb_samples);
if (len2 < 0) {
LOGE("swr_convert FAILED!!!");
}
int out_buffer_size = av_samples_get_buffer_size(NULL, out_channel_nb, frame->nb_samples, out_sample_fmt, 1);
jbyteArray audio_sample_array = env->NewByteArray(out_buffer_size);
jbyte *sample_bytep = env->GetByteArrayElements(audio_sample_array, NULL);
memcpy(sample_bytep, out_buffer_audio, out_buffer_size);
env->ReleaseByteArrayElements(audio_sample_array, sample_bytep, 0);
env->CallIntMethod(audioTrack->audiotrack, audioTrack->write_id, audio_sample_array, 0, out_buffer_size);
env->DeleteLocalRef(audio_sample_array);
}
} else {
LOGE("send package error");
}
}
av_packet_unref(packet);
}
LOGE("play complete");
ANativeWindow_release(nativeWindow);
av_frame_free(&outFrame);
av_frame_free(&frame);
av_packet_free(&packet);
avcodec_free_context(&codecContext);
avformat_close_input(&formatContext);
avformat_free_context(formatContext);
env->ReleaseStringUTFChars(url, input);
return ret;
}
网友评论