ffmpeg_sdk搭建

网上有很多编译ffmpeg源码的文章，但是通常我们不研究ffmpeg的话，目的就是直接用ffmpeg：

编程环境：win7  + VC6 + FFmpeg-full-SDK-3.2 ；

 

1： 下载 FFmpeg-full-SDK-3.2.rar（目前较新的版本），可以到这里下载；

2：解压后 有 bin、lib、include 三个文件夹，我们可以将解压后的文件夹FFmpeg-full-SDK-3.2放到指定的盘符中；

3：在VC6中，可以这么配置：Tool -> Directoris。。。，然后配置相关设置，（当然也可也将 bin、lib、include 直接放到你的工程目录下）；

4：添加相应的.h文件和.lib文件，如下：

extern "C"
{
#include "libavcodec/avcodec.h"
#include "libavformat/avformat.h"
#include "libswscale/swscale.h"
}
#pragma comment(lib,"avcodec.lib")
#pragma comment(lib,"avdevice.lib")
#pragma comment(lib,"avfilter.lib")
#pragma comment(lib,"avformat.lib")
#pragma comment(lib,"avutil.lib")
#pragma comment(lib,"swscale.lib")

 

这样，你就可以在VC中调用ffmpeg了；注意：ffmpeg 和 cximage 有的定义有冲突的地方，配置VC的时候，只保留ffmpeg就可以了；

 

在 FFmpeg中有一个例子：output_example.c，我们下载就需要在当前的VC环境中调试这个例子，当然，ffmpeg源程序文件夹下还有例子，可以搜索“example”关键字，

当然还有很多可以借鉴的c文件，还有一本书《ffmpeg+tutoria》，里面包含了一些例子，可参考；

为了方便在VC中调用，

1：修改 main函数的名字为任何名字，如miantest；当然参数目前先不要变；

2：在VC6中编译是提示有一些错误，不过这些提示都是强制转换的提示，所以添加强制转换就可以了；

3：将bin文件中的dll放到当前工程中，或自己配置一些工程；

4：为了方便应用，我这里定义一个 myffmpeg.h 这个文件：头文件包含 + lib 包含 + output_example.c 的修正内容，配置完成ffmpeg后，添加myffmpeg.h文件到工程，

myffmpeg.h  如下：

 

#pragma once
 
extern "C"
{
#include "libavcodec/avcodec.h"
#include "libavformat/avformat.h"
#include "libswscale/swscale.h"
}
#pragma comment(lib,"avcodec.lib")
#pragma comment(lib,"avdevice.lib")
#pragma comment(lib,"avfilter.lib")
#pragma comment(lib,"avformat.lib")
#pragma comment(lib,"avutil.lib")
#pragma comment(lib,"swscale.lib")
 

/*
 * Libavformat API example: Output a media file in any supported
 * libavformat format. The default codecs are used.
 *
 * Copyright (c) 2003 Fabrice Bellard
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this software and associated documentation files (the "Software"), to deal
 * in the Software without restriction, including without limitation the rights
 * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
 * copies of the Software, and to permit persons to whom the Software is
 * furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be included in
 * all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 * THE SOFTWARE.
 */
#include
#include
#include
#include

#ifndef M_PI
#define M_PI 3.14159265358979323846
#endif

#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;//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, int 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 = (CodecID)codec_id;
    c->codec_type = CODEC_TYPE_AUDIO;

    /* put sample parameters */
    c->bit_rate = 64000;
    c->sample_rate = 44100;
    c->channels = 2;
    return st;
}
//实际上就是初始化
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;
    //上面三行 如何初始化的？
    audio_outbuf_size = 10000;
    audio_outbuf = (uint8_t*)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 = (int16_t*)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;
        v = (int)(sin(t) * 10000);
        for(i = 0; i < nb_channels; i++)
    for(j=0;j            *q++ = v;
        t += tincr;
        tincr += tincr2;
    }
}
//重点
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->pts != AV_NOPTS_VALUE)
        pkt.pts= av_rescale_q(c->coded_frame->pts, c->time_base, st->time_base);
    pkt.flags |= PKT_FLAG_KEY;
    pkt.stream_index= st->index;
    pkt.data= audio_outbuf;

    /* write the compressed frame in the media file */
    if (av_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, int 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 = (CodecID)codec_id;
    c->codec_type = CODEC_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(!strcmp(oc->oformat->name, "mp4") || !strcmp(oc->oformat->name, "mov") || !strcmp(oc->oformat->name, "3gp"))
        c->flags |= CODEC_FLAG_GLOBAL_HEADER;

    return st;
}

static AVFrame *alloc_picture(int 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 = (uint8_t *)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 = (uint8_t*)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        for(x=0;x            pict->data[0][y * pict->linesize[0] + x] = x + y + i * 3;
        }
    }

    /* Cb and Cr */
    for(y=0;y        for(x=0;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 |= PKT_FLAG_KEY;
        pkt.stream_index= st->index;
        pkt.data= (uint8_t *)picture;
        pkt.size= sizeof(AVPicture);

        ret = av_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 |= 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_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 maintest(int argc, char **argv)//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();
    //下面一小段
    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];

    /* auto detect the output format from the name. default is
       mpeg. */
    fmt = guess_format(NULL, filename, NULL);
    if (!fmt) {
        printf("Could not deduce output format from file extension: using MPEG.\n");
        fmt = guess_format("mpeg", NULL, NULL);
    }
    if (!fmt) {
        fprintf(stderr, "Could not find suitable output format\n");
        exit(1);
    }

    /* allocate the output media context */
    oc = av_alloc_format_context();
    if (!oc) {
        fprintf(stderr, "Memory error\n");
        exit(1);
    }
    oc->oformat = fmt;
    snprintf(oc->filename, sizeof(oc->filename), "%s", filename);//注意下

    /* 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);
    }

    /* set the output parameters (must be done even if no
       parameters). */
    if (av_set_parameters(oc, NULL) < 0) {
        fprintf(stderr, "Invalid output format parameters\n");
        exit(1);
    }

    dump_format(oc, 0, filename, 1);
    //从上面到这儿的部分非常的重要
    /* 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 (url_fopen(&oc->pb, filename, URL_WRONLY) < 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);
        }
    }

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

    /* write the trailer, if any */
    av_write_trailer(oc);

    /* 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 */
        url_fclose(oc->pb);
    }

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

    return 0;
}

 

 

 

 

 

//=================================================================================================================================

 

 

 

这里还有一篇文章不错：http://hi.baidu.com/ljmaierjia/blog/item/4d9462fa50d66950d7887d29.html

在VS2008下调试output_example.c（详细步骤）

注意：这个是编译Debug版本的。
必要资源：FFMPEG SDK 3.2（已经编译好的，可以去下载）
1.建立一个空的VC++工程
  建立好之后，将FFMPEG SDK 3.2下的output_example.c文件拷贝至路径“%Project Dir%\%Project Name%\”下
  其“%Project Dir%”为你所建立的工程的路径，“%Project Name%”为工程名。我的工程路径为“D:\ARMDEV\FFMPEG\VS FFMPEG\example”
  工程名为example，因此我就要将output_example.c拷贝至路径“D:\ARM DEV\FFMPEG\VS FFMPEG\example\example”下。
2.拷贝头文件
  将FFMPEG SDK 3.2下的include文件夹下的所有文件及文件夹拷贝至output_example.c同一个目录下。
3.为工程导入output_example.c以及头文件
  A:在解决方案视图里右键单击“Source Files”->Add->Exsiting Item 然后选择工程路径下的output_example.c文件。
  B:在解决方案视图里右键单击“Header Files”->Add->Exsiting Item 然后选择工程路径下的前面拷贝过去的所有的头文件。
4.修改头文件
  修改你所引用到的头文件。因为output_example.c里只引用了2个头文件avformat.h,swscale.h
  所以分别在这两个头文件中加入如下代码：
#ifdef __cplusplus
extern "C" {
#endif
/*.你要引用的头文件的内容..*/
#ifdef __cplusplus
}
#endif
前半部分放在头文件的最开头，后半部分放在头文件的最末尾。这样就可以引用正常使用ffmpeg提供的函数了。

5.拷贝lib文件
  将FFMPEG SDK 3.2下的lib文件夹拷贝至output_example.c同一个目录下。
6.为项目添加所需要的lib文件
  A:在Project->Properties->Linker->General->Additional Library Directories下输入“$(SolutionDir)\$(ProjectName)\lib”,然后点OK。
  B:在Project->Properties->Linker->Input->Additional Dependencies下输入“avcodec.lib avdevice.lib avfilter.lib avformat.lib avutil.lib swscale.lib”，然后点OK。
7.修改output_example.c文件
  将output_example.c里的“#include "avformat.h"”和“#include "swscale.h"”
  修改为“#include "libavformat/avformat.h"”和“#include "libswscale/swscale.h"”。
8.修改编译输出路径
  在Project->Properties->Linker->General->Output Directory下输入“$(SolutionDir)\$(ProjectName)\$(ConfigurationName)”，然后点OK。
好了，开始编译。会发现找不到stdint.h这个文件，于是将出现问题的头文件中的“include ”改为“include "stdint.h"”，继续编译，直到所有头文件中的“include ”都改为“include "stdint.h"”为止。
好了，到这一步编译已经没有问题了，可是运行的时候会出现找不到avformat.dll等信息。要想能运行必须为工程添加相应的dll文件。
9.添加dll文件
  将FFMPEG SDK 3.2下的bin文件夹下的所有文件拷贝至工程文件夹下的Debug文件夹下，这个时候就可以运行output_example.exe了。
可能还有别的办法，比如可能很多头文件是不用的，但是我这里全部都添加进来了，原因是我是新手...只要能调通就万事大吉了，呵呵
希望这个帖子能帮助到一些跟我一样的菜鸟朋友么，大家加油，哈哈。