libavformat/iamfenc.c

   1 /*
   2  * IAMF muxer
   3  * Copyright (c) 2023 James Almer
   4  *
   5  * This file is part of FFmpeg.
   6  *
   7  * FFmpeg is free software; you can redistribute it and/or
   8  * modify it under the terms of the GNU Lesser General Public
   9  * License as published by the Free Software Foundation; either
  10  * version 2.1 of the License, or (at your option) any later version.
  11  *
  12  * FFmpeg is distributed in the hope that it will be useful,
  13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  15  * Lesser General Public License for more details.
  16  *
  17  * You should have received a copy of the GNU Lesser General Public
  18  * License along with FFmpeg; if not, write to the Free Software
  19  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  20  */
  21
  22 #include <stdint.h>
  23
  24 #include "avformat.h"
  25 #include "iamf.h"
  26 #include "iamf_writer.h"
  27 #include "internal.h"
  28 #include "mux.h"
  29
  30 typedef struct IAMFMuxContext {
  31     IAMFContext iamf;
  32
  33     int64_t descriptors_offset;
  34     int update_extradata;
  35
  36     int first_stream_id;
  37 } IAMFMuxContext;
  38
  39 static int iamf_init(AVFormatContext *s)
  40 {
  41     IAMFMuxContext *const c = s->priv_data;
  42     IAMFContext *const iamf = &c->iamf;
  43     int nb_audio_elements = 0, nb_mix_presentations = 0;
  44     int ret;
  45
  46     for (int i = 0; i < s->nb_streams; i++) {
  47         if (s->streams[i]->codecpar->codec_type != AVMEDIA_TYPE_AUDIO ||
  48             (s->streams[i]->codecpar->codec_tag != MKTAG('m','p','4','a') &&
  49              s->streams[i]->codecpar->codec_tag != MKTAG('O','p','u','s') &&
  50              s->streams[i]->codecpar->codec_tag != MKTAG('f','L','a','C') &&
  51              s->streams[i]->codecpar->codec_tag != MKTAG('i','p','c','m'))) {
  52             av_log(s, AV_LOG_ERROR, "Unsupported codec id %s\n",
  53                    avcodec_get_name(s->streams[i]->codecpar->codec_id));
  54             return AVERROR(EINVAL);
  55         }
  56
  57         if (s->streams[i]->codecpar->ch_layout.nb_channels > 2) {
  58             av_log(s, AV_LOG_ERROR, "Unsupported channel layout on stream #%d\n", i);
  59             return AVERROR(EINVAL);
  60         }
  61
  62         for (int j = 0; j < i; j++) {
  63             if (s->streams[i]->id == s->streams[j]->id) {
  64                 av_log(s, AV_LOG_ERROR, "Duplicated stream id %d\n", s->streams[j]->id);
  65                 return AVERROR(EINVAL);
  66             }
  67         }
  68     }
  69
  70     if (s->nb_stream_groups <= 1) {
  71         av_log(s, AV_LOG_ERROR, "There must be at least two stream groups\n");
  72         return AVERROR(EINVAL);
  73     }
  74
  75     for (int i = 0; i < s->nb_stream_groups; i++) {
  76         const AVStreamGroup *stg = s->stream_groups[i];
  77
  78         if (stg->type == AV_STREAM_GROUP_PARAMS_IAMF_AUDIO_ELEMENT)
  79             nb_audio_elements++;
  80         if (stg->type == AV_STREAM_GROUP_PARAMS_IAMF_MIX_PRESENTATION)
  81             nb_mix_presentations++;
  82     }
  83     if ((nb_audio_elements < 1 || nb_audio_elements > 2) || nb_mix_presentations < 1) {
  84         av_log(s, AV_LOG_ERROR, "There must be >= 1 and <= 2 IAMF_AUDIO_ELEMENT and at least "
  85                                 "one IAMF_MIX_PRESENTATION stream groups\n");
  86         return AVERROR(EINVAL);
  87     }
  88
  89     for (int i = 0; i < s->nb_stream_groups; i++) {
  90         const AVStreamGroup *stg = s->stream_groups[i];
  91         if (stg->type != AV_STREAM_GROUP_PARAMS_IAMF_AUDIO_ELEMENT)
  92             continue;
  93
  94         ret = ff_iamf_add_audio_element(iamf, stg, s);
  95         if (ret < 0)
  96             return ret;
  97     }
  98
  99     for (int i = 0; i < s->nb_stream_groups; i++) {
 100         const AVStreamGroup *stg = s->stream_groups[i];
 101         if (stg->type != AV_STREAM_GROUP_PARAMS_IAMF_MIX_PRESENTATION)
 102             continue;
 103
 104         ret = ff_iamf_add_mix_presentation(iamf, stg, s);
 105         if (ret < 0)
 106             return ret;
 107     }
 108
 109     c->first_stream_id = s->streams[0]->id;
 110
 111     return 0;
 112 }
 113
 114 static int iamf_write_header(AVFormatContext *s)
 115 {
 116     IAMFMuxContext *const c = s->priv_data;
 117     IAMFContext *const iamf = &c->iamf;
 118     int ret;
 119
 120     c->descriptors_offset = avio_tell(s->pb);
 121     ret = ff_iamf_write_descriptors(iamf, s->pb, s);
 122     if (ret < 0)
 123         return ret;
 124
 125     c->first_stream_id = s->streams[0]->id;
 126
 127     return 0;
 128 }
 129
 130 static int iamf_write_packet(AVFormatContext *s, AVPacket *pkt)
 131 {
 132     IAMFMuxContext *const c = s->priv_data;
 133     AVStream *st = s->streams[pkt->stream_index];
 134     int ret = 0;
 135
 136     if (st->id == c->first_stream_id)
 137         ret = ff_iamf_write_parameter_blocks(&c->iamf, s->pb, pkt, s);
 138     if (!ret)
 139         ret = ff_iamf_write_audio_frame(&c->iamf, s->pb, st->id, pkt);
 140     if (!ret && !pkt->size)
 141         c->update_extradata = 1;
 142
 143     return ret;
 144 }
 145
 146 static int iamf_write_trailer(AVFormatContext *s)
 147 {
 148     const IAMFMuxContext *const c = s->priv_data;
 149     const IAMFContext *const iamf = &c->iamf;
 150     int64_t pos;
 151     int ret;
 152
 153     if (!c->update_extradata || !(s->pb->seekable & AVIO_SEEKABLE_NORMAL))
 154         return 0;
 155
 156     pos = avio_tell(s->pb);
 157     avio_seek(s->pb, c->descriptors_offset, SEEK_SET);
 158     ret = ff_iamf_write_descriptors(iamf, s->pb, s);
 159     if (ret < 0)
 160         return ret;
 161
 162     avio_seek(s->pb, pos, SEEK_SET);
 163
 164     return 0;
 165 }
 166
 167 static void iamf_deinit(AVFormatContext *s)
 168 {
 169     IAMFMuxContext *const c = s->priv_data;
 170     IAMFContext *const iamf = &c->iamf;
 171
 172     ff_iamf_uninit_context(iamf);
 173 }
 174
 175 static const AVCodecTag iamf_codec_tags[] = {
 176     { AV_CODEC_ID_AAC,       MKTAG('m','p','4','a') },
 177     { AV_CODEC_ID_FLAC,      MKTAG('f','L','a','C') },
 178     { AV_CODEC_ID_OPUS,      MKTAG('O','p','u','s') },
 179     { AV_CODEC_ID_PCM_S16LE, MKTAG('i','p','c','m') },
 180     { AV_CODEC_ID_PCM_S16BE, MKTAG('i','p','c','m') },
 181     { AV_CODEC_ID_PCM_S24LE, MKTAG('i','p','c','m') },
 182     { AV_CODEC_ID_PCM_S24BE, MKTAG('i','p','c','m') },
 183     { AV_CODEC_ID_PCM_S32LE, MKTAG('i','p','c','m') },
 184     { AV_CODEC_ID_PCM_S32BE, MKTAG('i','p','c','m') },
 185     { AV_CODEC_ID_NONE,      MKTAG('i','p','c','m') }
 186 };
 187
 188 const FFOutputFormat ff_iamf_muxer = {
 189     .p.name            = "iamf",
 190     .p.long_name       = NULL_IF_CONFIG_SMALL("Raw Immersive Audio Model and Formats"),
 191     .p.extensions      = "iamf",
 192     .priv_data_size    = sizeof(IAMFMuxContext),
 193     .p.audio_codec     = AV_CODEC_ID_OPUS,
 194     .init              = iamf_init,
 195     .deinit            = iamf_deinit,
 196     .write_header      = iamf_write_header,
 197     .write_packet      = iamf_write_packet,
 198     .write_trailer     = iamf_write_trailer,
 199     .p.codec_tag       = (const AVCodecTag* const []){ iamf_codec_tags, NULL },
 200     .p.flags           = AVFMT_GLOBALHEADER | AVFMT_NOTIMESTAMPS,
 201 };