ffmpeg 打开视频流太慢（下）

前面的博文中已经交代过，ffmpeg打开视频慢主要是因为av_find_stream_info 耗时久。下面给出重写查找音视频stream info的一段代码，用来替代av_find_stream_info 。

static int try_decode_frame(AVFormatContext *s, AVStream *st, AVPacket *avpkt,
            AVDictionary **options)
        {
            const AVCodec *codec;
            int got_picture = , ret = ;
            AVFrame *frame = av_frame_alloc();
            AVSubtitle subtitle;
            AVPacket pkt = *avpkt;

            if (!frame)
                return AVERROR(ENOMEM);

            if (!avcodec_is_open(st->codec) &&
                st->info->found_decoder <=  &&
                (st->codec->codec_id != -st->info->found_decoder || !st->codec->codec_id)) {
                    AVDictionary *thread_opt = NULL;
                    codec =  avcodec_find_decoder( st->codec->codec_id);
                    if (!codec) {
                        st->info->found_decoder = -st->codec->codec_id;
                        ret                     = -;
                        goto fail;
                    }

                    /* Force thread count to 1 since the H.264 decoder will not extract
                    * SPS and PPS to extradata during multi-threaded decoding. */
                    av_dict_set(options ? options : &thread_opt, "threads", "", );
                    ret = avcodec_open2(st->codec, codec, options ? options : &thread_opt);
                    if (!options)
                        av_dict_free(&thread_opt);
                    if (ret < ) {
                        st->info->found_decoder = -st->codec->codec_id;
                        goto fail;
                    }
                    st->info->found_decoder = ;
            } else if (!st->info->found_decoder)
                st->info->found_decoder = ;

            if (st->info->found_decoder < ) {
                ret = -;
                goto fail;
            }

            got_picture = ;
            switch (st->codec->codec_type) {
            case AVMEDIA_TYPE_VIDEO:
                if(st->codec->width >  && st->codec->height >  && st->codec->pix_fmt != AV_PIX_FMT_NONE)
                {
                    ret = ;
                    break;
                }
                ret = avcodec_decode_video2(st->codec, frame,
                    &got_picture, &pkt);
                break;
            case AVMEDIA_TYPE_AUDIO:
                ret = avcodec_decode_audio4(st->codec, frame, &got_picture, &pkt);
                break;
            case AVMEDIA_TYPE_SUBTITLE:
                ret = avcodec_decode_subtitle2(st->codec, &subtitle,
                    &got_picture, &pkt);
                ret = pkt.size;
                break;
            default:
                break;
            }
            if (ret >= ) {
                if (got_picture)
                    st->nb_decoded_frames++;
                pkt.data += ret;
                pkt.size -= ret;
                ret       = got_picture;
            }

            if (!pkt.data && !got_picture)
                ret = -;

fail:
            av_frame_free(&frame);
            return ret;
        }
        static int ff_alloc_extradata(AVCodecContext *avctx, int size)
        {
            int ret;

            if (size <  || size >= INT32_MAX - FF_INPUT_BUFFER_PADDING_SIZE) {
                avctx->extradata_size = ;
                return AVERROR(EINVAL);
            }
            avctx->extradata = (uint8_t *)av_malloc(size + FF_INPUT_BUFFER_PADDING_SIZE);
            if (avctx->extradata) {
                memset(avctx->extradata + size, , FF_INPUT_BUFFER_PADDING_SIZE);
                avctx->extradata_size = size;
                ret = ;
            } else {
                avctx->extradata_size = ;
                ret = AVERROR(ENOMEM);
            }
            return ret;
        }
        static AVPacket *add_to_pktbuf(AVPacketList **packet_buffer, AVPacket *pkt,
            AVPacketList **plast_pktl)
        {
            AVPacketList *pktl = (AVPacketList *)av_mallocz(sizeof(AVPacketList));
            if (!pktl)
                return NULL;

            if (*packet_buffer)
                (*plast_pktl)->next = pktl;
            else
                *packet_buffer = pktl;

            /* Add the packet in the buffered packet list. */
            *plast_pktl = pktl;
            pktl->pkt   = *pkt;
            return &pktl->pkt;
        }
        static void free_packet_buffer(AVPacketList **pkt_buf, AVPacketList **pkt_buf_end)
        {
            while (*pkt_buf) {
                AVPacketList *pktl = *pkt_buf;
                *pkt_buf = pktl->next;
                av_free_packet(&pktl->pkt);
                av_freep(&pktl);
            }
            *pkt_buf_end = NULL;
        }
        static int ff_check_interrupt(AVIOInterruptCB *cb)
        {
            int ret;
            if (cb && cb->callback && (ret = cb->callback(cb->opaque)))
                return ret;
            return ;
        }

        static int has_codec_parameters(AVStream *st, const char **errmsg_ptr,bool keypacket = false)
        {
            AVCodecContext *avctx = st->codec;

#define FAIL(errmsg) do {                                         \
    if (errmsg_ptr)                                           \
    *errmsg_ptr = errmsg;                                 \
    return ;                                                 \
            } while ()

        /*    if(keypacket)
            {
                static int packetCount = 1;
                packetCount++;
                if(packetCount > 1)
                    cout << "packetCount too much" <<endl;
            }*/
            switch (avctx->codec_type) {
            case AVMEDIA_TYPE_AUDIO:
                if(avctx->codec_id == AV_CODEC_ID_NONE)
                    break;
                if (st->info->found_decoder >=  &&
                    avctx->sample_fmt == AV_SAMPLE_FMT_NONE)
                    FAIL("unspecified sample format");
                if (!avctx->sample_rate)
                    FAIL("unspecified sample rate");
                if (!avctx->channels)
                    FAIL("unspecified number of channels");
                if (st->info->found_decoder >=  && !st->nb_decoded_frames && avctx->codec_id == AV_CODEC_ID_DTS)
                    FAIL("no decodable DTS frames");
                break;
            case AVMEDIA_TYPE_VIDEO:
                if (!avctx->width)
                    FAIL("unspecified size");
                if (st->info->found_decoder >=  && avctx->pix_fmt == AV_PIX_FMT_NONE)
                    FAIL("unspecified pixel format");
                break;
            case AVMEDIA_TYPE_DATA:
                return ;
            case AVMEDIA_TYPE_UNKNOWN:
                FAIL("unspecified codec type");
                break;
            default:
                return ;
            }

            return ;
        }
        static    const AVCodec *find_decoder(AVFormatContext *s, AVStream *st, enum AVCodecID codec_id)
        {
            if (st->codec->codec)
            return st->codec->codec;

            switch (st->codec->codec_type)
            {
            case AVMEDIA_TYPE_VIDEO:
                if (s->video_codec)    return s->video_codec;
                break;
            case AVMEDIA_TYPE_AUDIO:
                if (s->audio_codec)    return s->audio_codec;
                break;
            case AVMEDIA_TYPE_SUBTITLE:
                if (s->subtitle_codec) return s->subtitle_codec;
                break;
            }
            return avcodec_find_decoder(codec_id);
        }

        static int alloc_and_copy_exterdata(AVCodecContext *dest, const AVCodecContext *src,size_t size ,size_t pad)
        {
            if(src->extradata && size > )
            {
                dest->extradata = (uint8_t *)av_malloc(size + pad);
                if(dest->extradata == nullptr)
                {
                    goto fail;
                }
                memcpy(dest->extradata, src->extradata, size);
                if(pad)
                {
                    memset((uint8_t *)dest->extradata + size, , pad);
                }
            }
            return ;
            fail:
             av_freep(&dest->extradata);
             return AVERROR(ENOMEM);
        }

        static int alloc_and_copy_intra_matrix(AVCodecContext *dest, const AVCodecContext *src,size_t size ,size_t pad)
        {
            if(src->intra_matrix && size > )
            {
                dest->intra_matrix = (uint16_t *)av_malloc(size + pad);
                if(dest->intra_matrix == nullptr)
                {
                    goto fail;
                }
                memcpy(dest->intra_matrix, src->intra_matrix, size);
                if(pad)
                {
                    memset((uint8_t *)dest->intra_matrix + size, , pad);
                }
            }

            fail:
             av_freep(&dest->intra_matrix);
             return AVERROR(ENOMEM);
        }

        static int alloc_and_copy_inter_matrix(AVCodecContext *dest, const AVCodecContext *src,size_t size ,size_t pad)
        {
            if(src->inter_matrix && size > )
            {
                dest->inter_matrix = (uint16_t *)av_malloc(size + pad);
                if(dest->inter_matrix == nullptr)
                {
                    goto fail;
                }
                memcpy(dest->inter_matrix, src->inter_matrix, size);
                if(pad)
                {
                    memset((uint8_t *)dest->inter_matrix + size, , pad);
                }
            }

            fail:
             av_freep(&dest->inter_matrix);
             return AVERROR(ENOMEM);
        }

        static int alloc_and_copy_rc_override(AVCodecContext *dest, const AVCodecContext *src,size_t size ,size_t pad)
        {
            if(src->rc_override && size > )
            {
                dest->rc_override = (RcOverride *)av_malloc(size + pad);
                if(dest->rc_override == nullptr)
                {
                    goto fail;
                }
                memcpy(dest->rc_override, src->rc_override, size);
                if(pad)
                {
                    memset((uint8_t *)dest->rc_override + size, , pad);
                }
            }

            fail:
             av_freep(&dest->rc_override);
             return AVERROR(ENOMEM);
        }
        static int avcodec_copy_context_private(AVCodecContext *dest, const AVCodecContext *src)
        {
            if(!src)
            {
                return -;
            }
            const AVCodec *orig_codec = dest->codec;
            uint8_t *orig_priv_data = (uint8_t *)dest->priv_data;

            if (avcodec_is_open(dest)) { // check that the dest context is uninitialized
                av_log(dest, AV_LOG_ERROR,
                       "Tried to copy AVCodecContext %p into already-initialized %p\n",
                       src, dest);
                return AVERROR(EINVAL);
            }

            av_opt_free(dest);

            memcpy(dest, src, sizeof(*dest));

            dest->priv_data       = orig_priv_data;

            if (orig_priv_data)
                av_opt_copy(orig_priv_data, src->priv_data);

            dest->codec           = orig_codec;

            /* set values specific to opened codecs back to their default state */
            dest->slice_offset    = NULL;
            dest->hwaccel         = NULL;
            dest->internal        = NULL;

            /* reallocate values that should be allocated separately */
            //dest->rc_eq           = NULL;
            dest->extradata       = NULL;
            dest->intra_matrix    = NULL;
            dest->inter_matrix    = NULL;
            dest->rc_override     = NULL;
            dest->subtitle_header = NULL;
        /*    if (src->rc_eq) {
                dest->rc_eq = av_strdup(src->rc_eq);
                if (!dest->rc_eq)
                    return AVERROR(ENOMEM);
            }*/
            int ret = ;
            ret = alloc_and_copy_exterdata(dest, src, src->extradata_size, FF_INPUT_BUFFER_PADDING_SIZE);
            if(ret < )
            {
                return ret ;
            }
            ret = alloc_and_copy_intra_matrix(dest, src,  * sizeof(int16_t), );
            if(ret < )
            {
                return ret ;
            }
            ret = alloc_and_copy_inter_matrix(dest, src,  * sizeof(int16_t), );
            if(ret < )
            {
                return ret ;
            }
            ret = alloc_and_copy_rc_override(dest, src,src->rc_override_count * sizeof(*src->rc_override), );
            return ret ;
        }

        static int avformatFindStreamInfo(AVFormatContext *ic, AVDictionary **options,std::vector<shared_ptr<AVPacket>> &packets,int timeout = )
        {
            int i, count, ret = , j;
            int64_t read_size;
            AVStream *st;
            // AVPacket  *pkt;
            AVPacket packet;
            int64_t old_offset  = ;
            int64_t last_offset  = ;
            bool firstKeyPacket = false;
            int tryCount = ;
            int orig_nb_streams = ic->nb_streams;
            int flush_codecs    = ic->probesize > ;
            int64_t startTime = av_gettime();
            int64_t max_analyze_duration = ic->max_analyze_duration2;
            for (i = ; i < ic->nb_streams; i++) {
                const AVCodec *codec;
                AVDictionary *thread_opt = NULL;
                st = ic->streams[i];

                if (st->codec->codec_type == AVMEDIA_TYPE_VIDEO ||
                    st->codec->codec_type == AVMEDIA_TYPE_SUBTITLE) {
                        if (!st->codec->time_base.num)
                            st->codec->time_base = st->time_base;
                }

                if (!st->parser && !(ic->flags & AVFMT_FLAG_NOPARSE)) {
                    st->parser = av_parser_init(st->codec->codec_id);
                    if (st->parser) {
                        if (st->need_parsing == AVSTREAM_PARSE_HEADERS) {
                            st->parser->flags |= PARSER_FLAG_COMPLETE_FRAMES;
                        } else if (st->need_parsing == AVSTREAM_PARSE_FULL_RAW) {
                            st->parser->flags |= PARSER_FLAG_USE_CODEC_TS;
                        }
                    } else if (st->need_parsing) {

                    }
                }
                codec = avcodec_find_decoder(st->codec->codec_id);

                av_dict_set(options ? &options[i] : &thread_opt, "threads", "", );
                if (!has_codec_parameters(st, NULL) && st->request_probe <= ) {
                    if (codec && !st->codec->codec)
                        if (avcodec_open2(st->codec, codec, options ? &options[i] : &thread_opt) < )
                            av_log(ic, AV_LOG_WARNING,
                            "Failed to open codec in av_find_stream_info\n");
                }
                if (!options)
                    av_dict_free(&thread_opt);
            }

            for (i = ; i < ic->nb_streams; i++) {
                ic->streams[i]->info->fps_first_dts = AV_NOPTS_VALUE;
                ic->streams[i]->info->fps_last_dts  = AV_NOPTS_VALUE;
            }

            count = ;
            read_size = ;
            bool saveAudioFlag = false;
            shared_ptr<AVPacket> lastAudioPacket[AudioPacketsNum] = {nullptr};

            for (;;) {
                if (ff_check_interrupt(&ic->interrupt_callback)) {
                    ret = AVERROR_EXIT;
                    av_log(ic, AV_LOG_DEBUG, "interrupted\n");
                    break;
                }
                if(av_gettime() - startTime > timeout *  * )
                {
                    ret = -;
                    break;
                }
                /* check if one codec still needs to be handled */
                for (i = ; i < ic->nb_streams; i++) {
                    st = ic->streams[i];
                    if (!has_codec_parameters(st, NULL,firstKeyPacket))
                        break;

                    if (st->parser && st->parser->parser->split &&
                        !st->codec->extradata)
                        break;
                }
                if (i == ic->nb_streams && ic->nb_streams > )
                {
                    break;
                }

                shared_ptr<AVPacket> pkt((AVPacket*)av_malloc(sizeof(AVPacket)), [&](AVPacket *p) { av_free_packet(p); av_freep(&p); });
                av_init_packet(pkt.get());
                old_offset  = avio_tell(ic->pb);
                ret = av_read_frame(ic, pkt.get());
                if (ret == AVERROR(EAGAIN))
                {
                    std::this_thread::sleep_for(std::chrono::milliseconds());
                    continue;
                }
                if (ret < ) {
                    /* EOF or error*/
                    break;
                }

                if(ic->nb_streams == ) return -;
                auto codeType = ic->streams[pkt->stream_index]->codec->codec_type;
                if(codeType == AVMEDIA_TYPE_VIDEO && pkt->flags&AV_PKT_FLAG_KEY)
                {
                    firstKeyPacket = true;
                }
                if(!firstKeyPacket && codeType == AVMEDIA_TYPE_VIDEO)
                {
                    continue;
                }

                shared_ptr<AVPacket> pkt1((AVPacket*)av_malloc(sizeof(AVPacket)), [&](AVPacket *p) { av_free_packet(p); av_freep(&p); });
                av_init_packet(pkt1.get());
                av_copy_packet(pkt1.get(),pkt.get());
                packets.push_back(pkt1);

                st = ic->streams[pkt->stream_index];
                if (!(st->disposition & AV_DISPOSITION_ATTACHED_PIC))
                    read_size += pkt->size;

                if (pkt->dts != AV_NOPTS_VALUE && st->codec_info_nb_frames > ) {
                    /* check for non-increasing dts */
                    if (st->info->fps_last_dts != AV_NOPTS_VALUE &&
                        st->info->fps_last_dts >= pkt->dts) {

                            st->info->fps_first_dts =
                                st->info->fps_last_dts  = AV_NOPTS_VALUE;
                    }
                    /* Check for a discontinuity in dts. If the difference in dts
                    * is more than 1000 times the average packet duration in the
                    * sequence, we treat it as a discontinuity. */
                    if (st->info->fps_last_dts != AV_NOPTS_VALUE &&
                        st->info->fps_last_dts_idx > st->info->fps_first_dts_idx &&
                        (pkt->dts - st->info->fps_last_dts) /  >
                        (st->info->fps_last_dts     - st->info->fps_first_dts) /
                        (st->info->fps_last_dts_idx - st->info->fps_first_dts_idx)) {

                            st->info->fps_first_dts =
                                st->info->fps_last_dts  = AV_NOPTS_VALUE;
                    }

                    /* update stored dts values */
                    if (st->info->fps_first_dts == AV_NOPTS_VALUE) {
                        st->info->fps_first_dts     = pkt->dts;
                        st->info->fps_first_dts_idx = st->codec_info_nb_frames;
                    }
                    st->info->fps_last_dts     = pkt->dts;
                    st->info->fps_last_dts_idx = st->codec_info_nb_frames;
                }

                if (st->parser && st->parser->parser->split && !st->codec->extradata) {
                    int i = st->parser->parser->split(st->codec, pkt->data, pkt->size);
                    if (i >  && i < FF_MAX_EXTRADATA_SIZE) {
                        if (ff_alloc_extradata(st->codec, i))
                            return AVERROR(ENOMEM);
                        memcpy(st->codec->extradata, pkt->data,
                            st->codec->extradata_size);
                    }
                }
                auto decodeRet = try_decode_frame(ic, st, pkt.get(),
                    (options && i < orig_nb_streams) ? &options[i] : NULL);
                st->codec_info_nb_frames++;

            }

            for (i = ; i < ic->nb_streams; i++) {
                st = ic->streams[i];
                avcodec_close(st->codec);
            }
            for (i = ; i < ic->nb_streams; i++)
            {
                st = ic->streams[i];
                if (st->codec->codec_type == AVMEDIA_TYPE_VIDEO)
                {
                    if (st->codec->codec_id == AV_CODEC_ID_RAWVIDEO && !st->codec->codec_tag && !st->codec->bits_per_coded_sample) {
                        uint32_t tag= avcodec_pix_fmt_to_codec_tag(st->codec->pix_fmt);
                        /*  if (avpriv_find_pix_fmt(ff_raw_pix_fmt_tags, tag) == st->codec->pix_fmt)
                        st->codec->codec_tag= tag;*/
                    }

                    if (!st->r_frame_rate.num) {
                        if (    st->codec->time_base.den * (int64_t) st->time_base.num
                            <= st->codec->time_base.num * st->codec->ticks_per_frame * (int64_t) st->time_base.den) {
                                st->r_frame_rate.num = st->codec->time_base.den;
                                st->r_frame_rate.den = st->codec->time_base.num * st->codec->ticks_per_frame;
                        } else {
                            st->r_frame_rate.num = st->time_base.den;
                            st->r_frame_rate.den = st->time_base.num;
                        }
                    }
                }
                else if (st->codec->codec_type == AVMEDIA_TYPE_AUDIO)
                {
                    if (!st->codec->bits_per_coded_sample)
                        st->codec->bits_per_coded_sample =
                        av_get_bits_per_sample(st->codec->codec_id);
                    // set stream disposition based on audio service type
                    switch (st->codec->audio_service_type) {
                    case AV_AUDIO_SERVICE_TYPE_EFFECTS:
                        st->disposition = AV_DISPOSITION_CLEAN_EFFECTS;
                        break;
                    case AV_AUDIO_SERVICE_TYPE_VISUALLY_IMPAIRED:
                        st->disposition = AV_DISPOSITION_VISUAL_IMPAIRED;
                        break;
                    case AV_AUDIO_SERVICE_TYPE_HEARING_IMPAIRED:
                        st->disposition = AV_DISPOSITION_HEARING_IMPAIRED;
                        break;
                    case AV_AUDIO_SERVICE_TYPE_COMMENTARY:
                        st->disposition = AV_DISPOSITION_COMMENT;
                        break;
                    case AV_AUDIO_SERVICE_TYPE_KARAOKE:
                        st->disposition = AV_DISPOSITION_KARAOKE;
                        break;
                    }
                }
            }

find_stream_info_err:

            for (i = ; i < ic->nb_streams; i++) {
                st = ic->streams[i];
                if (ic->streams[i]->codec->codec_type != AVMEDIA_TYPE_AUDIO)
                    ic->streams[i]->codec->thread_count = ;
                if (st->info)
                    av_freep(&st->info->duration_error);
                av_freep(&ic->streams[i]->info);
            }
            if(ic->nb_streams == )
                ret = -;
            return ret;
        }

这段代码的宗旨是找到音视频必要的信息后，立即返回。方法参数里传入vector> packets 是为了保存在查找的过程中已经从context中读取的视频包。如果不保存，这些视频包将会丢掉，同样会增加视频打开的时间。经现场测试，主流的摄像机例如海康，英飞拓，中威等都可以用它来打开。注意：文件不可以用这种方式打开。如果使用这段代码打不开视频或有疑问，请联系我：350197870。

视频下载地址：http://www.chungen90.com/?news_3/

Demo下载地址： http://www.chungen90.com/?news_2