打算写两篇文章简单分析FFmpeg的写文件用到的3个函数：avformat_write_header()，av_write_frame()以及av_write_trailer()。其中av_write_frame()用于写视频数据，，avformat_write_header()用于写视频文件头，而av_write_trailer()用于写视频文件尾。本文首先分析avformat_write_header()。PS：需要注意的是，尽管这3个函数功能是配套的，但是它们的前缀却不一样，写文件头Header的函数前缀是“avformat_”，其他两个函数前缀是“av_”（不太明白其中的原因）。avformat_write_header()的声明位于libavformat\avformat.h，如下所示。/** * Allocate the stream private data and write the stream header to * an output media file. * * @param s Media file handle, must be allocated with avformat_alloc_context(). *Its oformat field must be set to the desired output format; *Its pb field must be set to an already opened AVIOContext. * @param options An AVDictionary filled with AVFormatContext and muxer-private options. *On return this parameter will be destroyed and replaced with a dict containing *options that were not found. May be NULL. * * @return 0 on success, negative AVERROR on failure. * * @see av_opt_find, av_dict_set, avio_open, av_oformat_next. */int avformat_write_header(AVFormatContext *s, AVDictionary **options);简单解释一下它的参数的含义：s：用于输出的AVFormatContext。options：额外的选项，目前没有深入研究过，一般为NULL。函数正常执行后返回值等于0。该函数最典型的例子可以参考：最简单的基于FFMPEG的视频编码器（YUV编码为H.264）函数调用关系图
avformat_write_header()的调用关系如下图所示。
avformat_write_header()avformat_write_header()的定义位于libavformat\mux.c，如下所示。int avformat_write_header(AVFormatContext *s, AVDictionary **options){int ret = 0;if (ret = init_muxer(s, options))return ret;if (s->oformat->write_header) {ret = s->oformat->write_header(s);if (ret >= 0 && s->pb && s->pb->error < 0)ret = s->pb->error;if (ret < 0)return ret;if (s->flush_packets && s->pb && s->pb->error >= 0 && s->flags & AVFMT_FLAG_FLUSH_PACKETS)avio_flush(s->pb);}if ((ret = init_pts(s)) < 0)return ret;if (s->avoid_negative_ts < 0) {av_assert2(s->avoid_negative_ts == AVFMT_AVOID_NEG_TS_AUTO);if (s->oformat->flags & (AVFMT_TS_NEGATIVE | AVFMT_NOTIMESTAMPS)) {s->avoid_negative_ts = 0;} elses->avoid_negative_ts = AVFMT_AVOID_NEG_TS_MAKE_NON_NEGATIVE;}return 0;}从源代码可以看出，avformat_write_header()完成了以下工作：（1）调用init_muxer()初始化复用器（2）调用AVOutputFormat的write_header()下面看一下这两个函数。init_muxer()init_muxer()用于初始化复用器，它的定义如下所示。static int init_muxer(AVFormatContext *s, AVDictionary **options){int ret = 0, i;AVStream *st;AVDictionary *tmp = NULL;AVCodecContext *codec = NULL;AVOutputFormat *of = s->oformat;AVDictionaryEntry *e;if (options)av_dict_copy(&tmp, *options, 0);if ((ret = av_opt_set_dict(s, &tmp)) < 0)goto fail;if (s->priv_data && s->oformat->priv_class && *(const AVClass**)s->priv_data==s->oformat->priv_class &&(ret = av_opt_set_dict2(s->priv_data, &tmp, AV_OPT_SEARCH_CHILDREN)) < 0)goto fail;#if FF_API_LAVF_BITEXACTif (s->nb_streams && s->streams[0]->codec->flags & CODEC_FLAG_BITEXACT)s->flags |= AVFMT_FLAG_BITEXACT;#endif// some sanity checksif (s->nb_streams == 0 && !(of->flags & AVFMT_NOSTREAMS)) {av_log(s, AV_LOG_ERROR, "No streams to mux were specified\n");ret = AVERROR(EINVAL);goto fail;}for (i = 0; i < s->nb_streams; i++) {st = s->streams[i];codec = st->codec;#if FF_API_LAVF_CODEC_TBFF_DISABLE_DEPRECATION_WARNINGSif (!st->time_base.num && codec->time_base.num) {av_log(s, AV_LOG_WARNING, "Using AVStream.codec.time_base as a ""timebase hint to the muxer is deprecated. Set ""AVStream.time_base instead.\n");avpriv_set_pts_info(st, 64, codec->time_base.num, codec->time_base.den);}FF_ENABLE_DEPRECATION_WARNINGS#endifif (!st->time_base.num) {/* fall back on the default timebase values */if (codec->codec_type == AVMEDIA_TYPE_AUDIO && codec->sample_rate)avpriv_set_pts_info(st, 64, 1, codec->sample_rate);elseavpriv_set_pts_info(st, 33, 1, 90000);}switch (codec->codec_type) {case AVMEDIA_TYPE_AUDIO:if (codec->sample_rate <= 0) {av_log(s, AV_LOG_ERROR, "sample rate not set\n");ret = AVERROR(EINVAL);goto fail;}if (!codec->block_align)codec->block_align = codec->channels *av_get_bits_per_sample(codec->codec_id) >> 3;break;case AVMEDIA_TYPE_VIDEO:if ((codec->width <= 0 || codec->height <= 0) &&!(of->flags & AVFMT_NODIMENSIONS)) {av_log(s, AV_LOG_ERROR, "dimensions not set\n");ret = AVERROR(EINVAL);goto fail;}if (av_cmp_q(st->sample_aspect_ratio, codec->sample_aspect_ratio)&& FFABS(av_q2d(st->sample_aspect_ratio) – av_q2d(codec->sample_aspect_ratio)) > 0.004*av_q2d(st->sample_aspect_ratio)) {if (st->sample_aspect_ratio.num != 0 &&st->sample_aspect_ratio.den != 0 &&codec->sample_aspect_ratio.num != 0 &&codec->sample_aspect_ratio.den != 0) {av_log(s, AV_LOG_ERROR, "Aspect ratio mismatch between muxer ""(%d/%d) and encoder layer (%d/%d)\n",st->sample_aspect_ratio.num, st->sample_aspect_ratio.den,codec->sample_aspect_ratio.num,codec->sample_aspect_ratio.den);ret = AVERROR(EINVAL);goto fail;}}break;}if (of->codec_tag) {if ( codec->codec_tag&& codec->codec_id == AV_CODEC_ID_RAWVIDEO&& ( av_codec_get_tag(of->codec_tag, codec->codec_id) == 0|| av_codec_get_tag(of->codec_tag, codec->codec_id) == MKTAG(‘r’, ‘a’, ‘w’, ‘ ‘))&& !validate_codec_tag(s, st)) {// the current rawvideo encoding system ends up setting// the wrong codec_tag for avi/mov, we override it herecodec->codec_tag = 0;}if (codec->codec_tag) {if (!validate_codec_tag(s, st)) {char tagbuf[32], tagbuf2[32];av_get_codec_tag_string(tagbuf, sizeof(tagbuf), codec->codec_tag);av_get_codec_tag_string(tagbuf2, sizeof(tagbuf2), av_codec_get_tag(s->oformat->codec_tag, codec->codec_id));av_log(s, AV_LOG_ERROR,"Tag %s/0x%08x incompatible with output codec id ‘%d’ (%s)\n",tagbuf, codec->codec_tag, codec->codec_id, tagbuf2);ret = AVERROR_INVALIDDATA;goto fail;}} elsecodec->codec_tag = av_codec_get_tag(of->codec_tag, codec->codec_id);}if (of->flags & AVFMT_GLOBALHEADER &&!(codec->flags & CODEC_FLAG_GLOBAL_HEADER))av_log(s, AV_LOG_WARNING,"Codec for stream %d does not use global headers ""but container format requires global headers\n", i);if (codec->codec_type != AVMEDIA_TYPE_ATTACHMENT)s->internal->nb_interleaved_streams++;}if (!s->priv_data && of->priv_data_size > 0) {s->priv_data = av_mallocz(of->priv_data_size);if (!s->priv_data) {ret = AVERROR(ENOMEM);goto fail;}if (of->priv_class) {*(const AVClass **)s->priv_data = of->priv_class;av_opt_set_defaults(s->priv_data);if ((ret = av_opt_set_dict2(s->priv_data, &tmp, AV_OPT_SEARCH_CHILDREN)) < 0)goto fail;}}/* set muxer identification string */if (!(s->flags & AVFMT_FLAG_BITEXACT)) {av_dict_set(&s->metadata, "encoder", LIBAVFORMAT_IDENT, 0);} else {av_dict_set(&s->metadata, "encoder", NULL, 0);}for (e = NULL; e = av_dict_get(s->metadata, "encoder-", e, AV_DICT_IGNORE_SUFFIX); ) {av_dict_set(&s->metadata, e->key, NULL, 0);}if (options) {av_dict_free(options);*options = tmp;}return 0;fail:av_dict_free(&tmp);return ret;}init_muxer()代码很长，但是它所做的工作比较简单，可以概括成两个字：检查。函数的流程可以概括成以下几步：（1）将传入的AVDictionary形式的选项设置到AVFormatContext（2）遍历AVFormatContext中的每个AVStream，并作如下检查：a)AVStream的time_base是否正确设置。如果发现AVStream的time_base没有设置，则会调用avpriv_set_pts_info()进行设置。b)对于音频，检查采样率设置是否正确；对于视频，检查宽、高、宽高比。c)其他一些检查，不再详述。AVOutputFormat->write_header()avformat_write_header()中最关键的地方就是调用了AVOutputFormat的write_header()。write_header()是AVOutputFormat中的一个函数指针，指向写文件头的函数。不同的AVOutputFormat有不同的write_header()的实现方法。在这里我们举例子看一下FLV封装格式对应的AVOutputFormat，它的定义位于libavformat\flvenc.c，如下所示。AVOutputFormat ff_flv_muxer = {.name= "flv",.long_name= NULL_IF_CONFIG_SMALL("FLV (Flash Video)"),.mime_type= "video/x-flv",.extensions= "flv",.priv_data_size = sizeof(FLVContext),.audio_codec = CONFIG_LIBMP3LAME ? AV_CODEC_ID_MP3 : AV_CODEC_ID_ADPCM_SWF,.video_codec = AV_CODEC_ID_FLV1,.write_header = flv_write_header,.write_packet = flv_write_packet,.write_trailer = flv_write_trailer,.codec_tag= (const AVCodecTag* const []) {flv_video_codec_ids, flv_audio_codec_ids, 0},.flags= AVFMT_GLOBALHEADER | AVFMT_VARIABLE_FPS |AVFMT_TS_NONSTRICT,};从ff_flv_muxer的定义中可以看出，write_header()指向的函数为flv_write_header()。我们继续看一下flv_write_header()函数。flv_write_header()的定义同样位于libavformat\flvenc.c，如下所示。static int flv_write_header(AVFormatContext *s){int i;AVIOContext *pb = s->pb;FLVContext *flv = s->priv_data;int64_t data_size;//设置参数for (i = 0; i < s->nb_streams; i++) {AVCodecContext *enc = s->streams[i]->codec;FLVStreamContext *sc;switch (enc->codec_type) {case AVMEDIA_TYPE_VIDEO:if (s->streams[i]->avg_frame_rate.den &&s->streams[i]->avg_frame_rate.num) {//设置帧率，从AVStream拷贝过来flv->framerate = av_q2d(s->streams[i]->avg_frame_rate);}if (flv->video_enc) {av_log(s, AV_LOG_ERROR,"at most one video stream is supported in flv\n");return AVERROR(EINVAL);}//视频编码的AVCodecContextflv->video_enc = enc;if (enc->codec_tag == 0) {av_log(s, AV_LOG_ERROR, "Video codec ‘%s’ for stream %d is not compatible with FLV\n",avcodec_get_name(enc->codec_id), i);return AVERROR(EINVAL);}if (enc->codec_id == AV_CODEC_ID_MPEG4 ||enc->codec_id == AV_CODEC_ID_H263) {int error = s->strict_std_compliance > FF_COMPLIANCE_UNOFFICIAL;av_log(s, error ? AV_LOG_ERROR : AV_LOG_WARNING,"Codec %s is not supported in the official FLV specification,\n", avcodec_get_name(enc->codec_id));if (error) {av_log(s, AV_LOG_ERROR,"use vstrict=-1 / -strict -1 to use it anyway.\n");return AVERROR(EINVAL);}} else if (enc->codec_id == AV_CODEC_ID_VP6) {av_log(s, AV_LOG_WARNING,"Muxing VP6 in flv will produce flipped video on playback.\n");}break;case AVMEDIA_TYPE_AUDIO:if (flv->audio_enc) {av_log(s, AV_LOG_ERROR,"at most one audio stream is supported in flv\n");return AVERROR(EINVAL);}//音频编码的AVCodecContextflv->audio_enc = enc;if (get_audio_flags(s, enc) < 0)return AVERROR_INVALIDDATA;if (enc->codec_id == AV_CODEC_ID_PCM_S16BE)av_log(s, AV_LOG_WARNING,"16-bit big-endian audio in flv is valid but most likely unplayable (hardware dependent); use s16le\n");break;case AVMEDIA_TYPE_DATA:if (enc->codec_id != AV_CODEC_ID_TEXT && enc->codec_id != AV_CODEC_ID_NONE) {av_log(s, AV_LOG_ERROR, "Data codec ‘%s’ for stream %d is not compatible with FLV\n",avcodec_get_name(enc->codec_id), i);return AVERROR_INVALIDDATA;}flv->data_enc = enc;break;default:av_log(s, AV_LOG_ERROR, "Codec type ‘%s’ for stream %d is not compatible with FLV\n",av_get_media_type_string(enc->codec_type), i);return AVERROR(EINVAL);}avpriv_set_pts_info(s->streams[i], 32, 1, 1000); /* 32 bit pts in ms */sc = av_mallocz(sizeof(FLVStreamContext));if (!sc)return AVERROR(ENOMEM);s->streams[i]->priv_data = sc;sc->last_ts = -1;}flv->delay = AV_NOPTS_VALUE;//开始写入//Signatureavio_write(pb, "FLV", 3);//Versionavio_w8(pb, 1);//“！！”意思是把非0转换成1//Flagsavio_w8(pb, FLV_HEADER_FLAG_HASAUDIO * !!flv->audio_enc +FLV_HEADER_FLAG_HASVIDEO * !!flv->video_enc);//Header sizeavio_wb32(pb, 9);//Header结束//Previous Tag Sizeavio_wb32(pb, 0);for (i = 0; i < s->nb_streams; i++)if (s->streams[i]->codec->codec_tag == 5) {avio_w8(pb, 8);// message typeavio_wb24(pb, 0); // include flagsavio_wb24(pb, 0); // time stampavio_wb32(pb, 0); // reservedavio_wb32(pb, 11); // sizeflv->reserved = 5;}write_metadata(s, 0);for (i = 0; i < s->nb_streams; i++) {AVCodecContext *enc = s->streams[i]->codec;if (enc->codec_id == AV_CODEC_ID_AAC || enc->codec_id == AV_CODEC_ID_H264 || enc->codec_id == AV_CODEC_ID_MPEG4) {int64_t pos;avio_w8(pb, enc->codec_type == AVMEDIA_TYPE_VIDEO ?FLV_TAG_TYPE_VIDEO : FLV_TAG_TYPE_AUDIO);avio_wb24(pb, 0); // size patched lateravio_wb24(pb, 0); // tsavio_w8(pb, 0); // ts extavio_wb24(pb, 0); // streamidpos = avio_tell(pb);if (enc->codec_id == AV_CODEC_ID_AAC) {avio_w8(pb, get_audio_flags(s, enc));avio_w8(pb, 0); // AAC sequence headeravio_write(pb, enc->extradata, enc->extradata_size);} else {avio_w8(pb, enc->codec_tag | FLV_FRAME_KEY); // flagsavio_w8(pb, 0); // AVC sequence headeravio_wb24(pb, 0); // composition timeff_isom_write_avcc(pb, enc->extradata, enc->extradata_size);}data_size = avio_tell(pb) – pos;avio_seek(pb, -data_size – 10, SEEK_CUR);avio_wb24(pb, data_size);avio_skip(pb, data_size + 10 – 3);avio_wb32(pb, data_size + 11); // previous tag size}}return 0;}从源代码可以看出，flv_write_header()完成了FLV文件头的写入工作。该函数的工作可以大体分为以下两部分：（1）给FLVContext设置参数（2）写文件头，以及相关的Tag写文件头的代码很短，如下所示。avio_write(pb, "FLV", 3);avio_w8(pb, 1);avio_w8(pb, FLV_HEADER_FLAG_HASAUDIO * !!flv->audio_enc +FLV_HEADER_FLAG_HASVIDEO * !!flv->video_enc);avio_wb32(pb, 9);可以参考下图中FLV文件头的定义比对一下上面的代码。
雷霄骅leixiaohua1020@126.com
人生谁无少年时，甜苦酸辛各自知。
