ffmpeg 音视频流分离合并实例

#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <math.h>

#include "libavformat/avformat.h"
#include "libswscale/swscale.h"

#undef exit

/* 5 seconds stream duration */
#define STREAM_DURATION   5.0
#define STREAM_FRAME_RATE 25 /* 25 images/s */
#define STREAM_NB_FRAMES  ((int)(STREAM_DURATION * STREAM_FRAME_RATE))
#define STREAM_PIX_FMT PIX_FMT_YUV420P /* default pix_fmt */

static int sws_flags = SWS_BICUBIC;

/**************************************************************/
/* audio output */

float t, tincr, tincr2;
int16_t *samples;
uint8_t *audio_outbuf;
int audio_outbuf_size;
int audio_input_frame_size;

/*
 * add an audio output stream
 */
static AVStream *add_audio_stream(AVFormatContext *oc, enum CodecID codec_id)
{
    AVCodecContext *c;
    AVStream *st;

    st = av_new_stream(oc, 1);
    if (!st) {
        fprintf(stderr, "Could not alloc stream\n");
        exit(1);
    }

    c = st->codec;
    c->codec_id = codec_id;
    c->codec_type = AVMEDIA_TYPE_AUDIO;

    /* put sample parameters */
    c->sample_fmt = AV_SAMPLE_FMT_S16;
    c->bit_rate = 64000;
    c->sample_rate = 44100;
    c->channels = 2;

    // some formats want stream headers to be separate
    if(oc->oformat->flags & AVFMT_GLOBALHEADER)
        c->flags |= CODEC_FLAG_GLOBAL_HEADER;

    return st;
}

static void open_audio(AVFormatContext *oc, AVStream *st)
{
    AVCodecContext *c;
    AVCodec *codec;

    c = st->codec;

    /* find the audio encoder */
    codec = avcodec_find_encoder(c->codec_id);
    if (!codec) {
        fprintf(stderr, "codec not found\n");
        exit(1);
    }

    /* open it */
    if (avcodec_open(c, codec) < 0) {
        fprintf(stderr, "could not open codec\n");
        exit(1);
    }

    /* init signal generator */
    t = 0;
    tincr = 2 * M_PI * 110.0 / c->sample_rate;
    /* increment frequency by 110 Hz per second */
    tincr2 = 2 * M_PI * 110.0 / c->sample_rate / c->sample_rate;

    audio_outbuf_size = 10000;
    audio_outbuf = av_malloc(audio_outbuf_size);

    /* ugly hack for PCM codecs (will be removed ASAP with new PCM
       support to compute the input frame size in samples */
    if (c->frame_size <= 1) {
        audio_input_frame_size = audio_outbuf_size / c->channels;
        switch(st->codec->codec_id) {
        case CODEC_ID_PCM_S16LE:
        case CODEC_ID_PCM_S16BE:
        case CODEC_ID_PCM_U16LE:
        case CODEC_ID_PCM_U16BE:
            audio_input_frame_size >>= 1;
            break;
        default:
            break;
        }
    } else {
        audio_input_frame_size = c->frame_size;
    }
    samples = av_malloc(audio_input_frame_size * 2 * c->channels);
}

/* prepare a 16 bit dummy audio frame of 'frame_size' samples and
   'nb_channels' channels */
static void get_audio_frame(int16_t *samples, int frame_size, int nb_channels)
{
    int j, i, v;
    int16_t *q;

    q = samples;
    for(j=0;j<frame_size;j++) {
        v = (int)(sin(t) * 10000);
        for(i = 0; i < nb_channels; i++)
            *q++ = v;
        t += tincr;
        tincr += tincr2;
    }
}

static void write_audio_frame(AVFormatContext *oc, AVStream *st)
{
    AVCodecContext *c;
    AVPacket pkt;
    av_init_packet(&pkt);

    c = st->codec;

    get_audio_frame(samples, audio_input_frame_size, c->channels);

    pkt.size= avcodec_encode_audio(c, audio_outbuf, audio_outbuf_size, samples);

    if (c->coded_frame && c->coded_frame->pts != AV_NOPTS_VALUE)
        pkt.pts= av_rescale_q(c->coded_frame->pts, c->time_base, st->time_base);
    pkt.flags |= AV_PKT_FLAG_KEY;
    pkt.stream_index= st->index;
    pkt.data= audio_outbuf;

    /* write the compressed frame in the media file */
    if (av_interleaved_write_frame(oc, &pkt) != 0) {
        fprintf(stderr, "Error while writing audio frame\n");
        exit(1);
    }
}

static void close_audio(AVFormatContext *oc, AVStream *st)
{
    avcodec_close(st->codec);

    av_free(samples);
    av_free(audio_outbuf);
}

/**************************************************************/
/* video output */

AVFrame *picture, *tmp_picture;
uint8_t *video_outbuf;
int frame_count, video_outbuf_size;

/* add a video output stream */
static AVStream *add_video_stream(AVFormatContext *oc, enum CodecID codec_id)
{
    AVCodecContext *c;
    AVStream *st;

    st = av_new_stream(oc, 0);
    if (!st) {
        fprintf(stderr, "Could not alloc stream\n");
        exit(1);
    }

    c = st->codec;
    c->codec_id = codec_id;
    c->codec_type = AVMEDIA_TYPE_VIDEO;

    /* put sample parameters */
    c->bit_rate = 400000;
    /* resolution must be a multiple of two */
    c->width = 352;
    c->height = 288;
    /* time base: this is the fundamental unit of time (in seconds) in terms
       of which frame timestamps are represented. for fixed-fps content,
       timebase should be 1/framerate and timestamp increments should be
       identically 1. */
    c->time_base.den = STREAM_FRAME_RATE;
    c->time_base.num = 1;
    c->gop_size = 12; /* emit one intra frame every twelve frames at most */
    c->pix_fmt = STREAM_PIX_FMT;
    if (c->codec_id == CODEC_ID_MPEG2VIDEO) {
        /* just for testing, we also add B frames */
        c->max_b_frames = 2;
    }
    if (c->codec_id == CODEC_ID_MPEG1VIDEO){
        /* Needed to avoid using macroblocks in which some coeffs overflow.
           This does not happen with normal video, it just happens here as
           the motion of the chroma plane does not match the luma plane. */
        c->mb_decision=2;
    }
    // some formats want stream headers to be separate
    if(oc->oformat->flags & AVFMT_GLOBALHEADER)
        c->flags |= CODEC_FLAG_GLOBAL_HEADER;

    return st;
}

static AVFrame *alloc_picture(enum PixelFormat pix_fmt, int width, int height)
{
    AVFrame *picture;
    uint8_t *picture_buf;
    int size;

    picture = avcodec_alloc_frame();
    if (!picture)
        return NULL;
    size = avpicture_get_size(pix_fmt, width, height);
    picture_buf = av_malloc(size);
    if (!picture_buf) {
        av_free(picture);
        return NULL;
    }
    avpicture_fill((AVPicture *)picture, picture_buf,
                   pix_fmt, width, height);
    return picture;
}

static void open_video(AVFormatContext *oc, AVStream *st)
{
    AVCodec *codec;
    AVCodecContext *c;

    c = st->codec;

    /* find the video encoder */
    codec = avcodec_find_encoder(c->codec_id);
    if (!codec) {
        fprintf(stderr, "codec not found\n");
        exit(1);
    }

    /* open the codec */
    if (avcodec_open(c, codec) < 0) {
        fprintf(stderr, "could not open codec\n");
        exit(1);
    }

    video_outbuf = NULL;
    if (!(oc->oformat->flags & AVFMT_RAWPICTURE)) {
        /* allocate output buffer */
        /* XXX: API change will be done */
        /* buffers passed into lav* can be allocated any way you prefer,
           as long as they're aligned enough for the architecture, and
           they're freed appropriately (such as using av_free for buffers
           allocated with av_malloc) */
        video_outbuf_size = 200000;
        video_outbuf = av_malloc(video_outbuf_size);
    }

    /* allocate the encoded raw picture */
    picture = alloc_picture(c->pix_fmt, c->width, c->height);
    if (!picture) {
        fprintf(stderr, "Could not allocate picture\n");
        exit(1);
    }

    /* if the output format is not YUV420P, then a temporary YUV420P
       picture is needed too. It is then converted to the required
       output format */
    tmp_picture = NULL;
    if (c->pix_fmt != PIX_FMT_YUV420P) {
        tmp_picture = alloc_picture(PIX_FMT_YUV420P, c->width, c->height);
        if (!tmp_picture) {
            fprintf(stderr, "Could not allocate temporary picture\n");
            exit(1);
        }
    }
}

/* prepare a dummy image */
static void fill_yuv_image(AVFrame *pict, int frame_index, int width, int height)
{
    int x, y, i;

    i = frame_index;

    /* Y */
    for(y=0;y<height;y++) {
        for(x=0;x<width;x++) {
            pict->data[0][y * pict->linesize[0] + x] = x + y + i * 3;
        }
    }

    /* Cb and Cr */
    for(y=0;y<height/2;y++) {
        for(x=0;x<width/2;x++) {
            pict->data[1][y * pict->linesize[1] + x] = 128 + y + i * 2;
            pict->data[2][y * pict->linesize[2] + x] = 64 + x + i * 5;
        }
    }
}

static void write_video_frame(AVFormatContext *oc, AVStream *st)
{
    int out_size, ret;
    AVCodecContext *c;
    static struct SwsContext *img_convert_ctx;

    c = st->codec;

    if (frame_count >= STREAM_NB_FRAMES) {
        /* no more frame to compress. The codec has a latency of a few
           frames if using B frames, so we get the last frames by
           passing the same picture again */
    } else {
        if (c->pix_fmt != PIX_FMT_YUV420P) {
            /* as we only generate a YUV420P picture, we must convert it
               to the codec pixel format if needed */
            if (img_convert_ctx == NULL) {
                img_convert_ctx = sws_getContext(c->width, c->height,
                                                 PIX_FMT_YUV420P,
                                                 c->width, c->height,
                                                 c->pix_fmt,
                                                 sws_flags, NULL, NULL, NULL);
                if (img_convert_ctx == NULL) {
                    fprintf(stderr, "Cannot initialize the conversion context\n");
                    exit(1);
                }
            }
            fill_yuv_image(tmp_picture, frame_count, c->width, c->height);
            sws_scale(img_convert_ctx, tmp_picture->data, tmp_picture->linesize,
                      0, c->height, picture->data, picture->linesize);
        } else {
            fill_yuv_image(picture, frame_count, c->width, c->height);
        }
    }

    if (oc->oformat->flags & AVFMT_RAWPICTURE) {
        /* raw video case. The API will change slightly in the near
           futur for that */
        AVPacket pkt;
        av_init_packet(&pkt);

        pkt.flags |= AV_PKT_FLAG_KEY;
        pkt.stream_index= st->index;
        pkt.data= (uint8_t *)picture;
        pkt.size= sizeof(AVPicture);

        ret = av_interleaved_write_frame(oc, &pkt);
    } else {
        /* encode the image */
        out_size = avcodec_encode_video(c, video_outbuf, video_outbuf_size, picture);
        /* if zero size, it means the image was buffered */
        if (out_size > 0) {
            AVPacket pkt;
            av_init_packet(&pkt);

            if (c->coded_frame->pts != AV_NOPTS_VALUE)
                pkt.pts= av_rescale_q(c->coded_frame->pts, c->time_base, st->time_base);
            if(c->coded_frame->key_frame)
                pkt.flags |= AV_PKT_FLAG_KEY;
            pkt.stream_index= st->index;
            pkt.data= video_outbuf;
            pkt.size= out_size;

            /* write the compressed frame in the media file */
            ret = av_interleaved_write_frame(oc, &pkt);
        } else {
            ret = 0;
        }
    }
    if (ret != 0) {
        fprintf(stderr, "Error while writing video frame\n");
        exit(1);
    }
    frame_count++;
}

static void close_video(AVFormatContext *oc, AVStream *st)
{
    avcodec_close(st->codec);
    av_free(picture->data[0]);
    av_free(picture);
    if (tmp_picture) {
        av_free(tmp_picture->data[0]);
        av_free(tmp_picture);
    }
    av_free(video_outbuf);
}

/**************************************************************/
/* media file output */

int main(int argc, char **argv)
{
    const char *filename;
    AVOutputFormat *fmt;
    AVFormatContext *oc;
    AVStream *audio_st, *video_st;
    double audio_pts, video_pts;
    int i;

    /* initialize libavcodec, and register all codecs and formats */
    av_register_all();

    if (argc != 2) {
        printf("usage: %s output_file\n"
               "API example program to output a media file with libavformat.\n"
               "The output format is automatically guessed according to the file extension.\n"
               "Raw images can also be output by using '%%d' in the filename\n"
               "\n", argv[0]);
        exit(1);
    }

    filename = argv[1];

    /* allocate the output media context */
    avformat_alloc_output_context2(&oc, NULL, NULL, filename);
    if (!oc) {
        printf("Could not deduce output format from file extension: using MPEG.\n");
        avformat_alloc_output_context2(&oc, NULL, "mpeg", filename);
    }
    if (!oc) {
        exit(1);
    }
    fmt= oc->oformat;

    /* add the audio and video streams using the default format codecs
       and initialize the codecs */
    video_st = NULL;
    audio_st = NULL;
    if (fmt->video_codec != CODEC_ID_NONE) {
        video_st = add_video_stream(oc, fmt->video_codec);
    }
    if (fmt->audio_codec != CODEC_ID_NONE) {
        audio_st = add_audio_stream(oc, fmt->audio_codec);
    }

    av_dump_format(oc, 0, filename, 1);

    /* now that all the parameters are set, we can open the audio and
       video codecs and allocate the necessary encode buffers */
    if (video_st)
        open_video(oc, video_st);
    if (audio_st)
        open_audio(oc, audio_st);

    /* open the output file, if needed */
    if (!(fmt->flags & AVFMT_NOFILE)) {
        if (avio_open(&oc->pb, filename, AVIO_FLAG_WRITE) < 0) {
            fprintf(stderr, "Could not open '%s'\n", filename);
            exit(1);
        }
    }

    /* write the stream header, if any */
    av_write_header(oc);

    for(;;) {
        /* compute current audio and video time */
        if (audio_st)
            audio_pts = (double)audio_st->pts.val * audio_st->time_base.num / audio_st->time_base.den;
        else
            audio_pts = 0.0;

        if (video_st)
            video_pts = (double)video_st->pts.val * video_st->time_base.num / video_st->time_base.den;
        else
            video_pts = 0.0;

        if ((!audio_st || audio_pts >= STREAM_DURATION) &&
            (!video_st || video_pts >= STREAM_DURATION))
            break;

        /* write interleaved audio and video frames */
        if (!video_st || (video_st && audio_st && audio_pts < video_pts)) {
            write_audio_frame(oc, audio_st);
        } else {
            write_video_frame(oc, video_st);
        }
    }

    /* write the trailer, if any.  the trailer must be written
     * before you close the CodecContexts open when you wrote the
     * header; otherwise write_trailer may try to use memory that
     * was freed on av_codec_close() */
    av_write_trailer(oc);

    /* close each codec */
    if (video_st)
        close_video(oc, video_st);
    if (audio_st)
        close_audio(oc, audio_st);

    /* free the streams */
    for(i = 0; i < oc->nb_streams; i++) {
        av_freep(&oc->streams[i]->codec);
        av_freep(&oc->streams[i]);
    }

    if (!(fmt->flags & AVFMT_NOFILE)) {
        /* close the output file */
        avio_close(oc->pb);
    }

    /* free the stream */
    av_free(oc);

    return 0;
}