utils.c

/*
 * various utility functions for use within Libav
 * Copyright (c) 2000, 2001, 2002 Fabrice Bellard
 *
 * This file is part of Libav.
 *
 * Libav is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * Libav is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with Libav; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

#undef NDEBUG
#include <assert.h>
#include <stdarg.h>
#include <stdint.h>

#include "config.h"

#include "libavutil/avassert.h"
#include "libavutil/avstring.h"
#include "libavutil/dict.h"
#include "libavutil/internal.h"
#include "libavutil/mathematics.h"
#include "libavutil/opt.h"
#include "libavutil/parseutils.h"
#include "libavutil/pixdesc.h"
#include "libavutil/time.h"

#include "libavcodec/bytestream.h"
#include "libavcodec/internal.h"

#include "audiointerleave.h"
#include "avformat.h"
#include "id3v2.h"
#include "internal.h"
#include "metadata.h"
#if CONFIG_NETWORK
#include "network.h"
#endif
#include "riff.h"
#include "url.h"

/**
 * @file
 * various utility functions for use within Libav
 */

unsigned avformat_version(void)
{
    return LIBAVFORMAT_VERSION_INT;
}

const char *avformat_configuration(void)
{
    return LIBAV_CONFIGURATION;
}

const char *avformat_license(void)
{
#define LICENSE_PREFIX "libavformat license: "
    return LICENSE_PREFIX LIBAV_LICENSE + sizeof(LICENSE_PREFIX) - 1;
}

/* an arbitrarily chosen "sane" max packet size -- 50M */
#define SANE_CHUNK_SIZE (50000000)

/* Read the data in sane-sized chunks and append to pkt.
 * Return the number of bytes read or an error. */
static int append_packet_chunked(AVIOContext *s, AVPacket *pkt, int size)
{
    int64_t chunk_size = size;
    int64_t orig_pos   = pkt->pos; // av_grow_packet might reset pos
    int orig_size      = pkt->size;
    int ret = 0;

    do {
        int prev_size = pkt->size;
        int read_size;

        /* When the caller requests a lot of data, limit it to the amount
         * left in file or SANE_CHUNK_SIZE when it is not known. */
        if (size > SANE_CHUNK_SIZE) {
            int64_t filesize = avio_size(s) - avio_tell(s);
            chunk_size = FFMAX(filesize, SANE_CHUNK_SIZE);
        }
        read_size = FFMIN(size, chunk_size);

        ret = av_grow_packet(pkt, read_size);
        if (ret < 0)
            break;

        ret = avio_read(s, pkt->data + prev_size, read_size);
        if (ret != read_size) {
            av_shrink_packet(pkt, prev_size + FFMAX(ret, 0));
            break;
        }

        size -= read_size;
    } while (size > 0);

    pkt->pos = orig_pos;
    if (!pkt->size)
        av_packet_unref(pkt);
    return pkt->size > orig_size ? pkt->size - orig_size : ret;
}

int av_get_packet(AVIOContext *s, AVPacket *pkt, int size)
{
    av_init_packet(pkt);
    pkt->data = NULL;
    pkt->size = 0;
    pkt->pos  = avio_tell(s);

    return append_packet_chunked(s, pkt, size);
}

int av_append_packet(AVIOContext *s, AVPacket *pkt, int size)
{
    if (!pkt->size)
        return av_get_packet(s, pkt, size);
    return append_packet_chunked(s, pkt, size);
}

int av_filename_number_test(const char *filename)
{
    char buf[1024];
    return filename &&
           (av_get_frame_filename(buf, sizeof(buf), filename, 1) >= 0);
}

static int set_codec_from_probe_data(AVFormatContext *s, AVStream *st,
                                     AVProbeData *pd, int score)
{
    static const struct {
        const char *name;
        enum AVCodecID id;
        enum AVMediaType type;
    } fmt_id_type[] = {
        { "aac",       AV_CODEC_ID_AAC,        AVMEDIA_TYPE_AUDIO },
        { "ac3",       AV_CODEC_ID_AC3,        AVMEDIA_TYPE_AUDIO },
        { "dts",       AV_CODEC_ID_DTS,        AVMEDIA_TYPE_AUDIO },
        { "eac3",      AV_CODEC_ID_EAC3,       AVMEDIA_TYPE_AUDIO },
        { "h264",      AV_CODEC_ID_H264,       AVMEDIA_TYPE_VIDEO },
        { "latm",      AV_CODEC_ID_AAC_LATM,   AVMEDIA_TYPE_AUDIO },
        { "m4v",       AV_CODEC_ID_MPEG4,      AVMEDIA_TYPE_VIDEO },
        { "mp3",       AV_CODEC_ID_MP3,        AVMEDIA_TYPE_AUDIO },
        { "mpegvideo", AV_CODEC_ID_MPEG2VIDEO, AVMEDIA_TYPE_VIDEO },
        { 0 }
    };
    AVInputFormat *fmt = av_probe_input_format2(pd, 1, &score);

    if (fmt) {
        int i;
        av_log(s, AV_LOG_DEBUG,
               "Probe with size=%d, packets=%d detected %s with score=%d\n",
               pd->buf_size, MAX_PROBE_PACKETS - st->probe_packets,
               fmt->name, score);
        for (i = 0; fmt_id_type[i].name; i++) {
            if (!strcmp(fmt->name, fmt_id_type[i].name)) {
                st->codecpar->codec_id   = fmt_id_type[i].id;
                st->codecpar->codec_type = fmt_id_type[i].type;
#if FF_API_LAVF_AVCTX
FF_DISABLE_DEPRECATION_WARNINGS
                st->codec->codec_type = st->codecpar->codec_type;
                st->codec->codec_id   = st->codecpar->codec_id;
FF_ENABLE_DEPRECATION_WARNINGS
#endif
                break;
            }
        }
    }
    return !!fmt;
}

/************************************************************/
/* input media file */

/* Open input file and probe the format if necessary. */
static int init_input(AVFormatContext *s, const char *filename,
                      AVDictionary **options)
{
    int ret;
    AVProbeData pd = { filename, NULL, 0 };

    if (s->pb) {
        s->flags |= AVFMT_FLAG_CUSTOM_IO;
        if (!s->iformat)
            return av_probe_input_buffer(s->pb, &s->iformat, filename,
                                         s, 0, s->probesize);
        else if (s->iformat->flags & AVFMT_NOFILE)
            return AVERROR(EINVAL);
        return 0;
    }

    if ((s->iformat && s->iformat->flags & AVFMT_NOFILE) ||
        (!s->iformat && (s->iformat = av_probe_input_format(&pd, 0))))
        return 0;

    ret = s->io_open(s, &s->pb, filename, AVIO_FLAG_READ, options);
    if (ret < 0)
        return ret;
    if (s->iformat)
        return 0;
    return av_probe_input_buffer(s->pb, &s->iformat, filename,
                                 s, 0, s->probesize);
}

static int add_to_pktbuf(AVPacketList **packet_buffer, AVPacket *pkt,
                         AVPacketList **plast_pktl, int ref)
{
    AVPacketList *pktl = av_mallocz(sizeof(AVPacketList));
    int ret;

    if (!pktl)
        return AVERROR(ENOMEM);

    if (ref) {
        if ((ret = av_packet_ref(&pktl->pkt, pkt)) < 0) {
            av_free(pktl);
            return ret;
        }
    } else {
        pktl->pkt = *pkt;
    }

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

    /* Add the packet in the buffered packet list. */
    *plast_pktl = pktl;
    return 0;
}

static int queue_attached_pictures(AVFormatContext *s)
{
    int i, ret;
    for (i = 0; i < s->nb_streams; i++)
        if (s->streams[i]->disposition & AV_DISPOSITION_ATTACHED_PIC &&
            s->streams[i]->discard < AVDISCARD_ALL) {

            ret = add_to_pktbuf(&s->internal->raw_packet_buffer,
                                &s->streams[i]->attached_pic,
                                &s->internal->raw_packet_buffer_end, 1);
            if (ret < 0)
                return ret;
        }
    return 0;
}

#if FF_API_LAVF_AVCTX
FF_DISABLE_DEPRECATION_WARNINGS
static int update_stream_avctx(AVFormatContext *s)
{
    int i, ret;
    for (i = 0; i < s->nb_streams; i++) {
        AVStream *st = s->streams[i];

        if (!st->internal->need_codec_update)
            continue;

        ret = avcodec_parameters_to_context(st->codec, st->codecpar);
        if (ret < 0)
            return ret;

        st->internal->need_codec_update = 0;
    }
    return 0;
}
FF_ENABLE_DEPRECATION_WARNINGS
#endif

int avformat_open_input(AVFormatContext **ps, const char *filename,
                        AVInputFormat *fmt, AVDictionary **options)
{
    AVFormatContext *s = *ps;
    int i, ret = 0;
    AVDictionary *tmp = NULL;
    ID3v2ExtraMeta *id3v2_extra_meta = NULL;

    if (!s && !(s = avformat_alloc_context()))
        return AVERROR(ENOMEM);
    if (fmt)
        s->iformat = fmt;

    if (options)
        av_dict_copy(&tmp, *options, 0);

    if ((ret = av_opt_set_dict(s, &tmp)) < 0)
        goto fail;

    if ((ret = init_input(s, filename, &tmp)) < 0)
        goto fail;

    /* Check filename in case an image number is expected. */
    if (s->iformat->flags & AVFMT_NEEDNUMBER) {
        if (!av_filename_number_test(filename)) {
            ret = AVERROR(EINVAL);
            goto fail;
        }
    }

    s->duration = s->start_time = AV_NOPTS_VALUE;
    av_strlcpy(s->filename, filename ? filename : "", sizeof(s->filename));

    /* Allocate private data. */
    if (s->iformat->priv_data_size > 0) {
        if (!(s->priv_data = av_mallocz(s->iformat->priv_data_size))) {
            ret = AVERROR(ENOMEM);
            goto fail;
        }
        if (s->iformat->priv_class) {
            *(const AVClass **) s->priv_data = s->iformat->priv_class;
            av_opt_set_defaults(s->priv_data);
            if ((ret = av_opt_set_dict(s->priv_data, &tmp)) < 0)
                goto fail;
        }
    }

    /* e.g. AVFMT_NOFILE formats will not have a AVIOContext */
    if (s->pb)
        ff_id3v2_read(s, ID3v2_DEFAULT_MAGIC, &id3v2_extra_meta);

    if (s->iformat->read_header)
        if ((ret = s->iformat->read_header(s)) < 0)
            goto fail;

    if (id3v2_extra_meta &&
        (ret = ff_id3v2_parse_apic(s, &id3v2_extra_meta)) < 0)
        goto fail;
    ff_id3v2_free_extra_meta(&id3v2_extra_meta);

    if ((ret = queue_attached_pictures(s)) < 0)
        goto fail;

    if (s->pb && !s->internal->data_offset)
        s->internal->data_offset = avio_tell(s->pb);

    s->internal->raw_packet_buffer_remaining_size = RAW_PACKET_BUFFER_SIZE;

#if FF_API_LAVF_AVCTX
    update_stream_avctx(s);
#endif

    for (i = 0; i < s->nb_streams; i++)
        s->streams[i]->internal->orig_codec_id = s->streams[i]->codecpar->codec_id;

    if (options) {
        av_dict_free(options);
        *options = tmp;
    }
    *ps = s;
    return 0;

fail:
    ff_id3v2_free_extra_meta(&id3v2_extra_meta);
    av_dict_free(&tmp);
    if (s->pb && !(s->flags & AVFMT_FLAG_CUSTOM_IO))
        avio_close(s->pb);
    avformat_free_context(s);
    *ps = NULL;
    return ret;
}

/*******************************************************/

static int probe_codec(AVFormatContext *s, AVStream *st, const AVPacket *pkt)
{
    if (st->codecpar->codec_id == AV_CODEC_ID_PROBE) {
        AVProbeData *pd = &st->probe_data;
        av_log(s, AV_LOG_DEBUG, "probing stream %d\n", st->index);
        --st->probe_packets;

        if (pkt) {
            int err;
            if ((err = av_reallocp(&pd->buf, pd->buf_size + pkt->size +
                                   AVPROBE_PADDING_SIZE)) < 0)
                return err;
            memcpy(pd->buf + pd->buf_size, pkt->data, pkt->size);
            pd->buf_size += pkt->size;
            memset(pd->buf + pd->buf_size, 0, AVPROBE_PADDING_SIZE);
        } else {
            st->probe_packets = 0;
            if (!pd->buf_size) {
                av_log(s, AV_LOG_ERROR,
                       "nothing to probe for stream %d\n", st->index);
                return 0;
            }
        }

        if (!st->probe_packets ||
            av_log2(pd->buf_size) != av_log2(pd->buf_size - pkt->size)) {
            set_codec_from_probe_data(s, st, pd, st->probe_packets > 0
                                                 ? AVPROBE_SCORE_MAX / 4 : 0);
            if (st->codecpar->codec_id != AV_CODEC_ID_PROBE) {
                pd->buf_size = 0;
                av_freep(&pd->buf);
                av_log(s, AV_LOG_DEBUG, "probed stream %d\n", st->index);
            }
        }
    }
    return 0;
}

int ff_read_packet(AVFormatContext *s, AVPacket *pkt)
{
    int ret, i, err;
    AVStream *st;

    for (;;) {
        AVPacketList *pktl = s->internal->raw_packet_buffer;

        if (pktl) {
            *pkt = pktl->pkt;
            st   = s->streams[pkt->stream_index];
            if (st->codecpar->codec_id != AV_CODEC_ID_PROBE ||
                !st->probe_packets ||
                s->internal->raw_packet_buffer_remaining_size < pkt->size) {
                AVProbeData *pd;
                if (st->probe_packets)
                    if ((err = probe_codec(s, st, NULL)) < 0)
                        return err;
                pd = &st->probe_data;
                av_freep(&pd->buf);
                pd->buf_size = 0;
                s->internal->raw_packet_buffer                 = pktl->next;
                s->internal->raw_packet_buffer_remaining_size += pkt->size;
                av_free(pktl);
                return 0;
            }
        }

        pkt->data = NULL;
        pkt->size = 0;
        av_init_packet(pkt);
        ret = s->iformat->read_packet(s, pkt);
        if (ret < 0) {
            if (!pktl || ret == AVERROR(EAGAIN))
                return ret;
            for (i = 0; i < s->nb_streams; i++) {
                st = s->streams[i];
                if (st->probe_packets)
                    if ((err = probe_codec(s, st, NULL)) < 0)
                        return err;
            }
            continue;
        }

        if (!pkt->buf) {
            AVPacket tmp = { 0 };
            ret = av_packet_ref(&tmp, pkt);
            if (ret < 0)
                return ret;
            *pkt = tmp;
        }

        if ((s->flags & AVFMT_FLAG_DISCARD_CORRUPT) &&
            (pkt->flags & AV_PKT_FLAG_CORRUPT)) {
            av_log(s, AV_LOG_WARNING,
                   "Dropped corrupted packet (stream = %d)\n",
                   pkt->stream_index);
            av_packet_unref(pkt);
            continue;
        }

        st = s->streams[pkt->stream_index];

        switch (st->codecpar->codec_type) {
        case AVMEDIA_TYPE_VIDEO:
            if (s->video_codec_id)
                st->codecpar->codec_id = s->video_codec_id;
            break;
        case AVMEDIA_TYPE_AUDIO:
            if (s->audio_codec_id)
                st->codecpar->codec_id = s->audio_codec_id;
            break;
        case AVMEDIA_TYPE_SUBTITLE:
            if (s->subtitle_codec_id)
                st->codecpar->codec_id = s->subtitle_codec_id;
            break;
        }

        if (!pktl && (st->codecpar->codec_id != AV_CODEC_ID_PROBE ||
                      !st->probe_packets))
            return ret;

        err = add_to_pktbuf(&s->internal->raw_packet_buffer, pkt,
                            &s->internal->raw_packet_buffer_end, 0);
        if (err)
            return err;
        s->internal->raw_packet_buffer_remaining_size -= pkt->size;

        if ((err = probe_codec(s, st, pkt)) < 0)
            return err;
    }
}

/**********************************************************/

/**
 * Return the frame duration in seconds. Return 0 if not available.
 */
void ff_compute_frame_duration(AVFormatContext *s, int *pnum, int *pden, AVStream *st,
                               AVCodecParserContext *pc, AVPacket *pkt)
{
    AVRational codec_framerate = s->iformat ? st->internal->avctx->framerate :
                                              (AVRational){ 0, 1 };
    int frame_size;

    *pnum = 0;
    *pden = 0;
    switch (st->codecpar->codec_type) {
    case AVMEDIA_TYPE_VIDEO:
        if (st->avg_frame_rate.num) {
            *pnum = st->avg_frame_rate.den;
            *pden = st->avg_frame_rate.num;
        } else if (st->time_base.num * 1000LL > st->time_base.den) {
            *pnum = st->time_base.num;
            *pden = st->time_base.den;
        } else if (codec_framerate.den * 1000LL > codec_framerate.num) {
            *pnum = codec_framerate.den;
            *pden = codec_framerate.num;
            if (pc && pc->repeat_pict) {
                if (*pnum > INT_MAX / (1 + pc->repeat_pict))
                    *pden /= 1 + pc->repeat_pict;
                else
                    *pnum *= 1 + pc->repeat_pict;
            }
            /* If this codec can be interlaced or progressive then we need
             * a parser to compute duration of a packet. Thus if we have
             * no parser in such case leave duration undefined. */
            if (st->internal->avctx->ticks_per_frame > 1 && !pc)
                *pnum = *pden = 0;
        }
        break;
    case AVMEDIA_TYPE_AUDIO:
        frame_size = av_get_audio_frame_duration2(st->codecpar, pkt->size);
        if (frame_size <= 0 || st->codecpar->sample_rate <= 0)
            break;
        *pnum = frame_size;
        *pden = st->codecpar->sample_rate;
        break;
    default:
        break;
    }
}

static int is_intra_only(enum AVCodecID id)
{
    const AVCodecDescriptor *d = avcodec_descriptor_get(id);
    if (!d)
        return 0;
    if (d->type == AVMEDIA_TYPE_VIDEO && !(d->props & AV_CODEC_PROP_INTRA_ONLY))
        return 0;
    return 1;
}

static void update_initial_timestamps(AVFormatContext *s, int stream_index,
                                      int64_t dts, int64_t pts)
{
    AVStream *st       = s->streams[stream_index];
    AVPacketList *pktl = s->internal->packet_buffer;

    if (st->first_dts != AV_NOPTS_VALUE ||
        dts           == AV_NOPTS_VALUE ||
        st->cur_dts   == AV_NOPTS_VALUE)
        return;

    st->first_dts = dts - st->cur_dts;
    st->cur_dts   = dts;

    for (; pktl; pktl = pktl->next) {
        if (pktl->pkt.stream_index != stream_index)
            continue;
        // FIXME: think more about this check
        if (pktl->pkt.pts != AV_NOPTS_VALUE && pktl->pkt.pts == pktl->pkt.dts)
            pktl->pkt.pts += st->first_dts;

        if (pktl->pkt.dts != AV_NOPTS_VALUE)
            pktl->pkt.dts += st->first_dts;

        if (st->start_time == AV_NOPTS_VALUE && pktl->pkt.pts != AV_NOPTS_VALUE)
            st->start_time = pktl->pkt.pts;
    }
    if (st->start_time == AV_NOPTS_VALUE)
        st->start_time = pts;
}

static void update_initial_durations(AVFormatContext *s, AVStream *st,
                                     int stream_index, int duration)
{
    AVPacketList *pktl = s->internal->packet_buffer;
    int64_t cur_dts    = 0;

    if (st->first_dts != AV_NOPTS_VALUE) {
        cur_dts = st->first_dts;
        for (; pktl; pktl = pktl->next) {
            if (pktl->pkt.stream_index == stream_index) {
                if (pktl->pkt.pts != pktl->pkt.dts  ||
                    pktl->pkt.dts != AV_NOPTS_VALUE ||
                    pktl->pkt.duration)
                    break;
                cur_dts -= duration;
            }
        }
        pktl          = s->internal->packet_buffer;
        st->first_dts = cur_dts;
    } else if (st->cur_dts)
        return;

    for (; pktl; pktl = pktl->next) {
        if (pktl->pkt.stream_index != stream_index)
            continue;
        if (pktl->pkt.pts == pktl->pkt.dts  &&
            pktl->pkt.dts == AV_NOPTS_VALUE &&
            !pktl->pkt.duration) {
            pktl->pkt.dts = cur_dts;
            if (!st->internal->avctx->has_b_frames)
                pktl->pkt.pts = cur_dts;
            cur_dts += duration;
            if (st->codecpar->codec_type != AVMEDIA_TYPE_AUDIO)
                pktl->pkt.duration = duration;
        } else
            break;
    }
    if (st->first_dts == AV_NOPTS_VALUE)
        st->cur_dts = cur_dts;
}

static void compute_pkt_fields(AVFormatContext *s, AVStream *st,
                               AVCodecParserContext *pc, AVPacket *pkt)
{
    int num, den, presentation_delayed, delay, i;
    int64_t offset;

    if (s->flags & AVFMT_FLAG_NOFILLIN)
        return;

    if ((s->flags & AVFMT_FLAG_IGNDTS) && pkt->pts != AV_NOPTS_VALUE)
        pkt->dts = AV_NOPTS_VALUE;

    /* do we have a video B-frame ? */
    delay = st->internal->avctx->has_b_frames;
    presentation_delayed = 0;

    /* XXX: need has_b_frame, but cannot get it if the codec is
     *  not initialized */
    if (delay &&
        pc && pc->pict_type != AV_PICTURE_TYPE_B)
        presentation_delayed = 1;

    if (pkt->pts != AV_NOPTS_VALUE && pkt->dts != AV_NOPTS_VALUE &&
        st->pts_wrap_bits < 63 &&
        pkt->dts - (1LL << (st->pts_wrap_bits - 1)) > pkt->pts) {
        pkt->dts -= 1LL << st->pts_wrap_bits;
    }

    /* Some MPEG-2 in MPEG-PS lack dts (issue #171 / input_file.mpg).
     * We take the conservative approach and discard both.
     * Note: If this is misbehaving for an H.264 file, then possibly
     * presentation_delayed is not set correctly. */
    if (delay == 1 && pkt->dts == pkt->pts &&
        pkt->dts != AV_NOPTS_VALUE && presentation_delayed) {
        av_log(s, AV_LOG_DEBUG, "invalid dts/pts combination\n");
        pkt->dts = AV_NOPTS_VALUE;
    }

    if (pkt->duration == 0 && st->codecpar->codec_type != AVMEDIA_TYPE_AUDIO) {
        ff_compute_frame_duration(s, &num, &den, st, pc, pkt);
        if (den && num) {
            pkt->duration = av_rescale_rnd(1, num * (int64_t) st->time_base.den,
                                           den * (int64_t) st->time_base.num,
                                           AV_ROUND_DOWN);

            if (pkt->duration != 0 && s->internal->packet_buffer)
                update_initial_durations(s, st, pkt->stream_index,
                                         pkt->duration);
        }
    }

    /* Correct timestamps with byte offset if demuxers only have timestamps
     * on packet boundaries */
    if (pc && st->need_parsing == AVSTREAM_PARSE_TIMESTAMPS && pkt->size) {
        /* this will estimate bitrate based on this frame's duration and size */
        offset = av_rescale(pc->offset, pkt->duration, pkt->size);
        if (pkt->pts != AV_NOPTS_VALUE)
            pkt->pts += offset;
        if (pkt->dts != AV_NOPTS_VALUE)
            pkt->dts += offset;
    }

    /* This may be redundant, but it should not hurt. */
    if (pkt->dts != AV_NOPTS_VALUE &&
        pkt->pts != AV_NOPTS_VALUE &&
        pkt->pts > pkt->dts)
        presentation_delayed = 1;

    av_log(NULL, AV_LOG_TRACE,
            "IN delayed:%d pts:%"PRId64", dts:%"PRId64" "
            "cur_dts:%"PRId64" st:%d pc:%p\n",
            presentation_delayed, pkt->pts, pkt->dts, st->cur_dts,
            pkt->stream_index, pc);
    /* Interpolate PTS and DTS if they are not present. We skip H.264
     * currently because delay and has_b_frames are not reliably set. */
    if ((delay == 0 || (delay == 1 && pc)) &&
        st->codecpar->codec_id != AV_CODEC_ID_H264) {
        if (presentation_delayed) {
            /* DTS = decompression timestamp */
            /* PTS = presentation timestamp */
            if (pkt->dts == AV_NOPTS_VALUE)
                pkt->dts = st->last_IP_pts;
            update_initial_timestamps(s, pkt->stream_index, pkt->dts, pkt->pts);
            if (pkt->dts == AV_NOPTS_VALUE)
                pkt->dts = st->cur_dts;

            /* This is tricky: the dts must be incremented by the duration
             * of the frame we are displaying, i.e. the last I- or P-frame. */
            if (st->last_IP_duration == 0)
                st->last_IP_duration = pkt->duration;
            if (pkt->dts != AV_NOPTS_VALUE)
                st->cur_dts = pkt->dts + st->last_IP_duration;
            st->last_IP_duration = pkt->duration;
            st->last_IP_pts      = pkt->pts;
            /* Cannot compute PTS if not present (we can compute it only
             * by knowing the future. */
        } else if (pkt->pts != AV_NOPTS_VALUE ||
                   pkt->dts != AV_NOPTS_VALUE ||
                   pkt->duration              ||
                   st->codecpar->codec_type == AVMEDIA_TYPE_AUDIO) {
            int duration = pkt->duration;
            if (!duration && st->codecpar->codec_type == AVMEDIA_TYPE_AUDIO) {
                ff_compute_frame_duration(s, &num, &den, st, pc, pkt);
                if (den && num) {
                    duration = av_rescale_rnd(1,
                                              num * (int64_t) st->time_base.den,
                                              den * (int64_t) st->time_base.num,
                                              AV_ROUND_DOWN);
                    if (duration != 0 && s->internal->packet_buffer)
                        update_initial_durations(s, st, pkt->stream_index,
                                                 duration);
                }
            }

            if (pkt->pts != AV_NOPTS_VALUE || pkt->dts != AV_NOPTS_VALUE ||
                duration) {
                /* presentation is not delayed : PTS and DTS are the same */
                if (pkt->pts == AV_NOPTS_VALUE)
                    pkt->pts = pkt->dts;
                update_initial_timestamps(s, pkt->stream_index, pkt->pts,
                                          pkt->pts);
                if (pkt->pts == AV_NOPTS_VALUE)
                    pkt->pts = st->cur_dts;
                pkt->dts = pkt->pts;
                if (pkt->pts != AV_NOPTS_VALUE)
                    st->cur_dts = pkt->pts + duration;
            }
        }
    }

    if (pkt->pts != AV_NOPTS_VALUE && delay <= MAX_REORDER_DELAY) {
        st->pts_buffer[0] = pkt->pts;
        for (i = 0; i<delay && st->pts_buffer[i] > st->pts_buffer[i + 1]; i++)
            FFSWAP(int64_t, st->pts_buffer[i], st->pts_buffer[i + 1]);
        if (pkt->dts == AV_NOPTS_VALUE)
            pkt->dts = st->pts_buffer[0];
        // We skipped it above so we try here.
        if (st->codecpar->codec_id == AV_CODEC_ID_H264)
            // This should happen on the first packet
            update_initial_timestamps(s, pkt->stream_index, pkt->dts, pkt->pts);
        if (pkt->dts > st->cur_dts)
            st->cur_dts = pkt->dts;
    }

    av_log(NULL, AV_LOG_TRACE,
            "OUTdelayed:%d/%d pts:%"PRId64", dts:%"PRId64" cur_dts:%"PRId64"\n",
            presentation_delayed, delay, pkt->pts, pkt->dts, st->cur_dts);

    /* update flags */
    if (is_intra_only(st->codecpar->codec_id))
        pkt->flags |= AV_PKT_FLAG_KEY;
#if FF_API_CONVERGENCE_DURATION
FF_DISABLE_DEPRECATION_WARNINGS
    if (pc)
        pkt->convergence_duration = pc->convergence_duration;
FF_ENABLE_DEPRECATION_WARNINGS
#endif
}

static void free_packet_buffer(AVPacketList **pkt_buf, AVPacketList **pkt_buf_end)
{
    while (*pkt_buf) {
        AVPacketList *pktl = *pkt_buf;
        *pkt_buf = pktl->next;
        av_packet_unref(&pktl->pkt);
        av_freep(&pktl);
    }
    *pkt_buf_end = NULL;
}

/**
 * Parse a packet, add all split parts to parse_queue.
 *
 * @param pkt Packet to parse, NULL when flushing the parser at end of stream.
 */
static int parse_packet(AVFormatContext *s, AVPacket *pkt, int stream_index)
{
    AVPacket out_pkt = { 0 }, flush_pkt = { 0 };
    AVStream *st = s->streams[stream_index];
    uint8_t *data = pkt ? pkt->data : NULL;
    int size      = pkt ? pkt->size : 0;
    int ret = 0, got_output = 0;

    if (!pkt) {
        av_init_packet(&flush_pkt);
        pkt        = &flush_pkt;
        got_output = 1;
    }

    while (size > 0 || (pkt == &flush_pkt && got_output)) {
        int len;

        av_init_packet(&out_pkt);
        len = av_parser_parse2(st->parser, st->internal->avctx,
                               &out_pkt.data, &out_pkt.size, data, size,
                               pkt->pts, pkt->dts, pkt->pos);

        pkt->pts = pkt->dts = AV_NOPTS_VALUE;
        /* increment read pointer */
        data += len;
        size -= len;

        got_output = !!out_pkt.size;

        if (!out_pkt.size)
            continue;

        if (pkt->side_data) {
            out_pkt.side_data       = pkt->side_data;
            out_pkt.side_data_elems = pkt->side_data_elems;
            pkt->side_data          = NULL;
            pkt->side_data_elems    = 0;
        }

        /* set the duration */
        out_pkt.duration = 0;
        if (st->codecpar->codec_type == AVMEDIA_TYPE_AUDIO) {
            if (st->internal->avctx->sample_rate > 0) {
                out_pkt.duration =
                    av_rescale_q_rnd(st->parser->duration,
                                     (AVRational) { 1, st->internal->avctx->sample_rate },
                                     st->time_base,
                                     AV_ROUND_DOWN);
            }
        }

        out_pkt.stream_index = st->index;
        out_pkt.pts          = st->parser->pts;
        out_pkt.dts          = st->parser->dts;
        out_pkt.pos          = st->parser->pos;

        if (st->parser->key_frame == 1 ||
            (st->parser->key_frame == -1 &&
             st->parser->pict_type == AV_PICTURE_TYPE_I))
            out_pkt.flags |= AV_PKT_FLAG_KEY;

        compute_pkt_fields(s, st, st->parser, &out_pkt);

        if ((s->iformat->flags & AVFMT_GENERIC_INDEX) &&
            out_pkt.flags & AV_PKT_FLAG_KEY) {
            ff_reduce_index(s, st->index);
            av_add_index_entry(st, st->parser->frame_offset, out_pkt.dts,
                               0, 0, AVINDEX_KEYFRAME);
        }

        if ((ret = add_to_pktbuf(&s->internal->parse_queue, &out_pkt,
                                 &s->internal->parse_queue_end,
                                 1))) {
            av_packet_unref(&out_pkt);
            goto fail;
        }
    }

    /* end of the stream => close and free the parser */
    if (pkt == &flush_pkt) {
        av_parser_close(st->parser);
        st->parser = NULL;
    }

fail:
    av_packet_unref(pkt);
    return ret;
}

static int read_from_packet_buffer(AVPacketList **pkt_buffer,
                                   AVPacketList **pkt_buffer_end,
                                   AVPacket      *pkt)
{
    AVPacketList *pktl;
    av_assert0(*pkt_buffer);
    pktl        = *pkt_buffer;
    *pkt        = pktl->pkt;
    *pkt_buffer = pktl->next;
    if (!pktl->next)
        *pkt_buffer_end = NULL;
    av_freep(&pktl);
    return 0;
}

static int read_frame_internal(AVFormatContext *s, AVPacket *pkt)
{
    int ret = 0, i, got_packet = 0;
    AVDictionary *metadata = NULL;

    av_init_packet(pkt);

    while (!got_packet && !s->internal->parse_queue) {
        AVStream *st;
        AVPacket cur_pkt;

        /* read next packet */
        ret = ff_read_packet(s, &cur_pkt);
        if (ret < 0) {
            if (ret == AVERROR(EAGAIN))
                return ret;
            /* flush the parsers */
            for (i = 0; i < s->nb_streams; i++) {
                st = s->streams[i];
                if (st->parser && st->need_parsing)
                    parse_packet(s, NULL, st->index);
            }
            /* all remaining packets are now in parse_queue =>
             * really terminate parsing */
            break;
        }
        ret = 0;
        st  = s->streams[cur_pkt.stream_index];

        if (cur_pkt.pts != AV_NOPTS_VALUE &&
            cur_pkt.dts != AV_NOPTS_VALUE &&
            cur_pkt.pts < cur_pkt.dts) {
            av_log(s, AV_LOG_WARNING,
                   "Invalid timestamps stream=%d, pts=%"PRId64", "
                   "dts=%"PRId64", size=%d\n",
                   cur_pkt.stream_index, cur_pkt.pts,
                   cur_pkt.dts, cur_pkt.size);
        }
        if (s->debug & FF_FDEBUG_TS)
            av_log(s, AV_LOG_DEBUG,
                   "ff_read_packet stream=%d, pts=%"PRId64", dts=%"PRId64", "
                   "size=%d, duration=%"PRId64", flags=%d\n",
                   cur_pkt.stream_index, cur_pkt.pts, cur_pkt.dts,
                   cur_pkt.size, cur_pkt.duration, cur_pkt.flags);

        if (st->need_parsing && !st->parser && !(s->flags & AVFMT_FLAG_NOPARSE)) {
            st->parser = av_parser_init(st->codecpar->codec_id);
            if (!st->parser)
                /* no parser available: just output the raw packets */
                st->need_parsing = AVSTREAM_PARSE_NONE;
            else if (st->need_parsing == AVSTREAM_PARSE_HEADERS)
                st->parser->flags |= PARSER_FLAG_COMPLETE_FRAMES;
            else if (st->need_parsing == AVSTREAM_PARSE_FULL_ONCE)
                st->parser->flags |= PARSER_FLAG_ONCE;
        }

        if (!st->need_parsing || !st->parser) {
            /* no parsing needed: we just output the packet as is */
            *pkt = cur_pkt;
            compute_pkt_fields(s, st, NULL, pkt);
            if ((s->iformat->flags & AVFMT_GENERIC_INDEX) &&
                (pkt->flags & AV_PKT_FLAG_KEY) && pkt->dts != AV_NOPTS_VALUE) {
                ff_reduce_index(s, st->index);
                av_add_index_entry(st, pkt->pos, pkt->dts,
                                   0, 0, AVINDEX_KEYFRAME);
            }
            got_packet = 1;
        } else if (st->discard < AVDISCARD_ALL) {
            if ((ret = parse_packet(s, &cur_pkt, cur_pkt.stream_index)) < 0)
                return ret;
        } else {
            /* free packet */
            av_packet_unref(&cur_pkt);
        }
    }

    if (!got_packet && s->internal->parse_queue)
        ret = read_from_packet_buffer(&s->internal->parse_queue, &s->internal->parse_queue_end, pkt);

    av_opt_get_dict_val(s, "metadata", AV_OPT_SEARCH_CHILDREN, &metadata);
    if (metadata) {