wavenc.c 12.1 KB
Newer Older
1 2 3 4
/*
 * WAV muxer
 * Copyright (c) 2001, 2002 Fabrice Bellard
 *
Paul B Mahol's avatar
Paul B Mahol committed
5 6 7
 * Sony Wave64 muxer
 * Copyright (c) 2012 Paul B Mahol
 *
8 9 10
 * WAV muxer RF64 support
 * Copyright (c) 2013 Daniel Verkamp <daniel@drv.nu>
 *
11
 * This file is part of FFmpeg.
12
 *
13
 * FFmpeg is free software; you can redistribute it and/or
14 15 16 17
 * 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.
 *
18
 * FFmpeg is distributed in the hope that it will be useful,
19 20 21 22 23
 * 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
24
 * License along with FFmpeg; if not, write to the Free Software
25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

#include <stdint.h>
#include <string.h>

#include "libavutil/dict.h"
#include "libavutil/common.h"
#include "libavutil/mathematics.h"
#include "libavutil/opt.h"

#include "avformat.h"
#include "avio.h"
#include "avio_internal.h"
#include "internal.h"
#include "riff.h"

42 43 44 45
#define RF64_AUTO   (-1)
#define RF64_NEVER  0
#define RF64_ALWAYS 1

46 47 48
typedef struct WAVMuxContext {
    const AVClass *class;
    int64_t data;
49
    int64_t fact_pos;
50
    int64_t ds64;
51 52 53 54
    int64_t minpts;
    int64_t maxpts;
    int last_duration;
    int write_bext;
55
    int rf64;
56 57
} WAVMuxContext;

Paul B Mahol's avatar
Paul B Mahol committed
58
#if CONFIG_WAV_MUXER
59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116
static inline void bwf_write_bext_string(AVFormatContext *s, const char *key, int maxlen)
{
    AVDictionaryEntry *tag;
    int len = 0;

    if (tag = av_dict_get(s->metadata, key, NULL, 0)) {
        len = strlen(tag->value);
        len = FFMIN(len, maxlen);
        avio_write(s->pb, tag->value, len);
    }

    ffio_fill(s->pb, 0, maxlen - len);
}

static void bwf_write_bext_chunk(AVFormatContext *s)
{
    AVDictionaryEntry *tmp_tag;
    uint64_t time_reference = 0;
    int64_t bext = ff_start_tag(s->pb, "bext");

    bwf_write_bext_string(s, "description", 256);
    bwf_write_bext_string(s, "originator", 32);
    bwf_write_bext_string(s, "originator_reference", 32);
    bwf_write_bext_string(s, "origination_date", 10);
    bwf_write_bext_string(s, "origination_time", 8);

    if (tmp_tag = av_dict_get(s->metadata, "time_reference", NULL, 0))
        time_reference = strtoll(tmp_tag->value, NULL, 10);
    avio_wl64(s->pb, time_reference);
    avio_wl16(s->pb, 1);  // set version to 1

    if (tmp_tag = av_dict_get(s->metadata, "umid", NULL, 0)) {
        unsigned char umidpart_str[17] = {0};
        int i;
        uint64_t umidpart;
        int len = strlen(tmp_tag->value+2);

        for (i = 0; i < len/16; i++) {
            memcpy(umidpart_str, tmp_tag->value + 2 + (i*16), 16);
            umidpart = strtoll(umidpart_str, NULL, 16);
            avio_wb64(s->pb, umidpart);
        }
        ffio_fill(s->pb, 0, 64 - i*8);
    } else
        ffio_fill(s->pb, 0, 64); // zero UMID

    ffio_fill(s->pb, 0, 190); // Reserved

    if (tmp_tag = av_dict_get(s->metadata, "coding_history", NULL, 0))
        avio_put_str(s->pb, tmp_tag->value);

    ff_end_tag(s->pb, bext);
}

static int wav_write_header(AVFormatContext *s)
{
    WAVMuxContext *wav = s->priv_data;
    AVIOContext *pb = s->pb;
117
    int64_t fmt;
118

119 120 121 122
    if (wav->rf64 == RF64_ALWAYS) {
        ffio_wfourcc(pb, "RF64");
        avio_wl32(pb, -1); /* RF64 chunk size: use size in ds64 */
    } else {
Daniel Verkamp's avatar
Daniel Verkamp committed
123 124
        ffio_wfourcc(pb, "RIFF");
        avio_wl32(pb, 0); /* file length */
125 126
    }

127 128
    ffio_wfourcc(pb, "WAVE");

129 130 131 132 133 134 135 136
    if (wav->rf64 != RF64_NEVER) {
        /* write empty ds64 chunk or JUNK chunk to reserve space for ds64 */
        ffio_wfourcc(pb, wav->rf64 == RF64_ALWAYS ? "ds64" : "JUNK");
        avio_wl32(pb, 28); /* chunk size */
        wav->ds64 = avio_tell(pb);
        ffio_fill(pb, 0, 28);
    }

137 138 139 140 141 142 143 144 145 146 147
    /* format header */
    fmt = ff_start_tag(pb, "fmt ");
    if (ff_put_wav_header(pb, s->streams[0]->codec) < 0) {
        av_log(s, AV_LOG_ERROR, "%s codec not supported in WAVE format\n",
               s->streams[0]->codec->codec ? s->streams[0]->codec->codec->name : "NONE");
        return -1;
    }
    ff_end_tag(pb, fmt);

    if (s->streams[0]->codec->codec_tag != 0x01 /* hence for all other than PCM */
        && s->pb->seekable) {
148
        wav->fact_pos = ff_start_tag(pb, "fact");
149
        avio_wl32(pb, 0);
150
        ff_end_tag(pb, wav->fact_pos);
151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188
    }

    if (wav->write_bext)
        bwf_write_bext_chunk(s);

    avpriv_set_pts_info(s->streams[0], 64, 1, s->streams[0]->codec->sample_rate);
    wav->maxpts = wav->last_duration = 0;
    wav->minpts = INT64_MAX;

    /* info header */
    ff_riff_write_info(s);

    /* data header */
    wav->data = ff_start_tag(pb, "data");

    avio_flush(pb);

    return 0;
}

static int wav_write_packet(AVFormatContext *s, AVPacket *pkt)
{
    AVIOContext *pb  = s->pb;
    WAVMuxContext    *wav = s->priv_data;
    avio_write(pb, pkt->data, pkt->size);
    if(pkt->pts != AV_NOPTS_VALUE) {
        wav->minpts        = FFMIN(wav->minpts, pkt->pts);
        wav->maxpts        = FFMAX(wav->maxpts, pkt->pts);
        wav->last_duration = pkt->duration;
    } else
        av_log(s, AV_LOG_ERROR, "wav_write_packet: NOPTS\n");
    return 0;
}

static int wav_write_trailer(AVFormatContext *s)
{
    AVIOContext *pb  = s->pb;
    WAVMuxContext    *wav = s->priv_data;
189 190 191
    int64_t file_size, data_size;
    int64_t number_of_samples = 0;
    int rf64 = 0;
192 193 194 195 196 197

    avio_flush(pb);

    if (s->pb->seekable) {
        /* update file size */
        file_size = avio_tell(pb);
198 199 200 201
        data_size = file_size - wav->data;
        if (wav->rf64 == RF64_ALWAYS || (wav->rf64 == RF64_AUTO && file_size - 8 > UINT32_MAX)) {
            rf64 = 1;
        } else {
Daniel Verkamp's avatar
Daniel Verkamp committed
202 203 204
            avio_seek(pb, 4, SEEK_SET);
            avio_wl32(pb, (uint32_t)(file_size - 8));
            avio_seek(pb, file_size, SEEK_SET);
205

Daniel Verkamp's avatar
Daniel Verkamp committed
206 207
            ff_end_tag(pb, wav->data);
            avio_flush(pb);
208
        }
209

Daniel Verkamp's avatar
Daniel Verkamp committed
210 211 212
        number_of_samples = av_rescale(wav->maxpts - wav->minpts + wav->last_duration,
                                       s->streams[0]->codec->sample_rate * (int64_t)s->streams[0]->time_base.num,
                                       s->streams[0]->time_base.den);
213 214 215

        if(s->streams[0]->codec->codec_tag != 0x01) {
            /* Update num_samps in fact chunk */
216
            avio_seek(pb, wav->fact_pos, SEEK_SET);
217 218 219 220
            if (rf64 || (wav->rf64 == RF64_AUTO && number_of_samples > UINT32_MAX)) {
                rf64 = 1;
                avio_wl32(pb, -1);
            } else {
Daniel Verkamp's avatar
Daniel Verkamp committed
221 222 223
                avio_wl32(pb, number_of_samples);
                avio_seek(pb, file_size, SEEK_SET);
                avio_flush(pb);
224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247
            }
        }

        if (rf64) {
            /* overwrite RIFF with RF64 */
            avio_seek(pb, 0, SEEK_SET);
            ffio_wfourcc(pb, "RF64");
            avio_wl32(pb, -1);

            /* write ds64 chunk (overwrite JUNK if rf64 == RF64_AUTO) */
            avio_seek(pb, wav->ds64 - 8, SEEK_SET);
            ffio_wfourcc(pb, "ds64");
            avio_wl32(pb, 28);                  /* ds64 chunk size */
            avio_wl64(pb, file_size - 8);       /* RF64 chunk size */
            avio_wl64(pb, data_size);           /* data chunk size */
            avio_wl64(pb, number_of_samples);   /* fact chunk number of samples */
            avio_wl32(pb, 0);                   /* number of table entries for non-'data' chunks */

            /* write -1 in data chunk size */
            avio_seek(pb, wav->data - 4, SEEK_SET);
            avio_wl32(pb, -1);

            avio_seek(pb, file_size, SEEK_SET);
            avio_flush(pb);
248 249 250 251 252 253 254 255 256
        }
    }
    return 0;
}

#define OFFSET(x) offsetof(WAVMuxContext, x)
#define ENC AV_OPT_FLAG_ENCODING_PARAM
static const AVOption options[] = {
    { "write_bext", "Write BEXT chunk.", OFFSET(write_bext), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, ENC },
257 258 259 260
    { "rf64",       "Use RF64 header rather than RIFF for large files.",    OFFSET(rf64), AV_OPT_TYPE_INT,   { .i64 = RF64_NEVER  },-1, 1, ENC, "rf64" },
    { "auto",       "Write RF64 header if file grows large enough.",        0,            AV_OPT_TYPE_CONST, { .i64 = RF64_AUTO   }, 0, 0, ENC, "rf64" },
    { "always",     "Always write RF64 header regardless of file size.",    0,            AV_OPT_TYPE_CONST, { .i64 = RF64_ALWAYS }, 0, 0, ENC, "rf64" },
    { "never",      "Never write RF64 header regardless of file size.",     0,            AV_OPT_TYPE_CONST, { .i64 = RF64_NEVER  }, 0, 0, ENC, "rf64" },
261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285
    { NULL },
};

static const AVClass wav_muxer_class = {
    .class_name = "WAV muxer",
    .item_name  = av_default_item_name,
    .option     = options,
    .version    = LIBAVUTIL_VERSION_INT,
};

AVOutputFormat ff_wav_muxer = {
    .name              = "wav",
    .long_name         = NULL_IF_CONFIG_SMALL("WAV / WAVE (Waveform Audio)"),
    .mime_type         = "audio/x-wav",
    .extensions        = "wav",
    .priv_data_size    = sizeof(WAVMuxContext),
    .audio_codec       = AV_CODEC_ID_PCM_S16LE,
    .video_codec       = AV_CODEC_ID_NONE,
    .write_header      = wav_write_header,
    .write_packet      = wav_write_packet,
    .write_trailer     = wav_write_trailer,
    .flags             = AVFMT_TS_NONSTRICT,
    .codec_tag         = (const AVCodecTag* const []){ ff_codec_wav_tags, 0 },
    .priv_class        = &wav_muxer_class,
};
Paul B Mahol's avatar
Paul B Mahol committed
286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326
#endif /* CONFIG_WAV_MUXER */

#if CONFIG_W64_MUXER
#include "w64.h"

static void start_guid(AVIOContext *pb, const uint8_t *guid, int64_t *pos)
{
    *pos = avio_tell(pb);

    avio_write(pb, guid, 16);
    avio_wl64(pb, INT64_MAX);
}

static void end_guid(AVIOContext *pb, int64_t start)
{
    int64_t end, pos = avio_tell(pb);

    end = FFALIGN(pos, 8);
    ffio_fill(pb, 0, end - pos);
    avio_seek(pb, start + 16, SEEK_SET);
    avio_wl64(pb, end - start);
    avio_seek(pb, end, SEEK_SET);
}

static int w64_write_header(AVFormatContext *s)
{
    WAVMuxContext *wav = s->priv_data;
    AVIOContext *pb = s->pb;
    int64_t start;
    int ret;

    avio_write(pb, ff_w64_guid_riff, sizeof(ff_w64_guid_riff));
    avio_wl64(pb, -1);
    avio_write(pb, ff_w64_guid_wave, sizeof(ff_w64_guid_wave));
    start_guid(pb, ff_w64_guid_fmt, &start);
    if ((ret = ff_put_wav_header(pb, s->streams[0]->codec)) < 0) {
        av_log(s, AV_LOG_ERROR, "%s codec not supported\n",
               s->streams[0]->codec->codec ? s->streams[0]->codec->codec->name : "NONE");
        return ret;
    }
    end_guid(pb, start);
Paul B Mahol's avatar
Paul B Mahol committed
327 328 329 330 331 332 333 334

    if (s->streams[0]->codec->codec_tag != 0x01 /* hence for all other than PCM */
        && s->pb->seekable) {
        start_guid(pb, ff_w64_guid_fact, &wav->fact_pos);
        avio_wl64(pb, 0);
        end_guid(pb, wav->fact_pos);
    }

Paul B Mahol's avatar
Paul B Mahol committed
335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352
    start_guid(pb, ff_w64_guid_data, &wav->data);

    return 0;
}

static int w64_write_trailer(AVFormatContext *s)
{
    AVIOContext    *pb = s->pb;
    WAVMuxContext *wav = s->priv_data;
    int64_t file_size;

    if (pb->seekable) {
        end_guid(pb, wav->data);

        file_size = avio_tell(pb);
        avio_seek(pb, 16, SEEK_SET);
        avio_wl64(pb, file_size);

Paul B Mahol's avatar
Paul B Mahol committed
353 354 355 356 357 358 359 360 361 362 363
        if (s->streams[0]->codec->codec_tag != 0x01) {
            int64_t number_of_samples;

            number_of_samples = av_rescale(wav->maxpts - wav->minpts + wav->last_duration,
                                           s->streams[0]->codec->sample_rate * (int64_t)s->streams[0]->time_base.num,
                                           s->streams[0]->time_base.den);
            avio_seek(pb, wav->fact_pos + 24, SEEK_SET);
            avio_wl64(pb, number_of_samples);
        }

        avio_seek(pb, file_size, SEEK_SET);
Paul B Mahol's avatar
Paul B Mahol committed
364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383
        avio_flush(pb);
    }

    return 0;
}

AVOutputFormat ff_w64_muxer = {
    .name              = "w64",
    .long_name         = NULL_IF_CONFIG_SMALL("Sony Wave64"),
    .extensions        = "w64",
    .priv_data_size    = sizeof(WAVMuxContext),
    .audio_codec       = AV_CODEC_ID_PCM_S16LE,
    .video_codec       = AV_CODEC_ID_NONE,
    .write_header      = w64_write_header,
    .write_packet      = wav_write_packet,
    .write_trailer     = w64_write_trailer,
    .flags             = AVFMT_TS_NONSTRICT,
    .codec_tag         = (const AVCodecTag* const []){ ff_codec_wav_tags, 0 },
};
#endif /* CONFIG_W64_MUXER */