westwood.c 12.6 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63
/*
 * Westwood Studios Multimedia Formats Demuxer (VQA, AUD)
 * Copyright (c) 2003 The ffmpeg Project
 *
 * This library 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 of the License, or (at your option) any later version.
 *
 * This library 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 this library; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
 */

/**
 * @file westwood.c
 * Westwood Studios VQA & AUD file demuxers
 * by Mike Melanson (melanson@pcisys.net)
 * for more information on the Westwood file formats, visit:
 *   http://www.pcisys.net/~melanson/codecs/
 *   http://www.geocities.com/SiliconValley/8682/aud3.txt
 *
 * Implementation note: There is no definite file signature for AUD files.
 * The demuxer uses a probabilistic strategy for content detection. This
 * entails performing sanity checks on certain header values in order to
 * qualify a file. Refer to wsaud_probe() for the precise parameters.
 */

#include "avformat.h"

#define LE_16(x)  ((((uint8_t*)(x))[1] << 8) | ((uint8_t*)(x))[0])
#define LE_32(x)  ((((uint8_t*)(x))[3] << 24) | \
                   (((uint8_t*)(x))[2] << 16) | \
                   (((uint8_t*)(x))[1] << 8) | \
                    ((uint8_t*)(x))[0])
#define BE_32(x)  ((((uint8_t*)(x))[0] << 24) | \
                   (((uint8_t*)(x))[1] << 16) | \
                   (((uint8_t*)(x))[2] << 8) | \
                    ((uint8_t*)(x))[3])

#define AUD_HEADER_SIZE 12
#define AUD_CHUNK_PREAMBLE_SIZE 8
#define AUD_CHUNK_SIGNATURE 0x0000DEAF

#define FOURCC_TAG( ch0, ch1, ch2, ch3 ) \
        ( (long)(unsigned char)(ch3) | \
        ( (long)(unsigned char)(ch2) << 8 ) | \
        ( (long)(unsigned char)(ch1) << 16 ) | \
        ( (long)(unsigned char)(ch0) << 24 ) )

#define FORM_TAG FOURCC_TAG('F', 'O', 'R', 'M')
#define WVQA_TAG FOURCC_TAG('W', 'V', 'Q', 'A')
#define VQHD_TAG FOURCC_TAG('V', 'Q', 'H', 'D')
#define FINF_TAG FOURCC_TAG('F', 'I', 'N', 'F')
#define SND0_TAG FOURCC_TAG('S', 'N', 'D', '0')
#define SND2_TAG FOURCC_TAG('S', 'N', 'D', '2')
#define VQFR_TAG FOURCC_TAG('V', 'Q', 'F', 'R')

64 65 66 67 68 69 70 71
/* don't know what these tags are for, but acknowledge their existence */
#define CINF_TAG FOURCC_TAG('C', 'I', 'N', 'F')
#define CINH_TAG FOURCC_TAG('C', 'I', 'N', 'H')
#define CIND_TAG FOURCC_TAG('C', 'I', 'N', 'D')
#define PINF_TAG FOURCC_TAG('P', 'I', 'N', 'F')
#define PINH_TAG FOURCC_TAG('P', 'I', 'N', 'H')
#define PIND_TAG FOURCC_TAG('P', 'I', 'N', 'D')

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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237
#define VQA_HEADER_SIZE 0x2A
#define VQA_FRAMERATE 15
#define VQA_VIDEO_PTS_INC (90000 / VQA_FRAMERATE)
#define VQA_PREAMBLE_SIZE 8

typedef struct WsAudDemuxContext {
    int audio_samplerate;
    int audio_channels;
    int audio_bits;
    int audio_type;
    int audio_stream_index;
    int64_t audio_frame_counter;
} WsAudDemuxContext;

typedef struct WsVqaDemuxContext {
    int audio_samplerate;
    int audio_channels;
    int audio_bits;

    int audio_stream_index;
    int video_stream_index;

    int64_t audio_frame_counter;
    int64_t video_pts;
} WsVqaDemuxContext;

static int wsaud_probe(AVProbeData *p)
{
    int field;

    /* Probabilistic content detection strategy: There is no file signature
     * so perform sanity checks on various header parameters:
     *   8000 <= sample rate (16 bits) <= 48000  ==> 40001 acceptable numbers
     *   compression type (8 bits) = 1 or 99     ==> 2 acceptable numbers
     * There is a total of 24 bits. The number space contains 2^24 =
     * 16777216 numbers. There are 40001 * 2 = 80002 acceptable combinations
     * of numbers. There is a 80002/16777216 = 0.48% chance of a false
     * positive.
     */

    if (p->buf_size < AUD_HEADER_SIZE)
        return 0;

    /* check sample rate */
    field = LE_16(&p->buf[0]);
    if ((field < 8000) || (field > 48000))
        return 0;

    /* note: only check for WS IMA (type 99) right now since there is no 
     * support for type 1 */
    if (p->buf[11] != 99)
        return 0;

    /* return 1/2 certainty since this file check is a little sketchy */
    return AVPROBE_SCORE_MAX / 2;
}

static int wsaud_read_header(AVFormatContext *s,
                             AVFormatParameters *ap)
{
    WsAudDemuxContext *wsaud = (WsAudDemuxContext *)s->priv_data;
    ByteIOContext *pb = &s->pb;
    AVStream *st;
    unsigned char header[AUD_HEADER_SIZE];

    if (get_buffer(pb, header, AUD_HEADER_SIZE) != AUD_HEADER_SIZE)
        return -EIO;
    wsaud->audio_samplerate = LE_16(&header[0]);
    if (header[11] == 99)
        wsaud->audio_type = CODEC_ID_ADPCM_IMA_WS;
    else
        return AVERROR_INVALIDDATA;

    /* flag 0 indicates stereo */
    wsaud->audio_channels = (header[10] & 0x1) + 1;
    /* flag 1 indicates 16 bit audio */
    wsaud->audio_bits = (((header[10] & 0x2) >> 1) + 1) * 8;

    /* set the pts reference the same as the sample rate */
    s->pts_num = 1;
    s->pts_den = wsaud->audio_samplerate;

    /* initialize the audio decoder stream */
    st = av_new_stream(s, 0);
    if (!st)
        return AVERROR_NOMEM;
    st->codec.codec_type = CODEC_TYPE_AUDIO;
    st->codec.codec_id = wsaud->audio_type;
    st->codec.codec_tag = 0;  /* no tag */
    st->codec.channels = wsaud->audio_channels;
    st->codec.sample_rate = wsaud->audio_samplerate;
    st->codec.bits_per_sample = wsaud->audio_bits;
    st->codec.bit_rate = st->codec.channels * st->codec.sample_rate *
        st->codec.bits_per_sample / 4;
    st->codec.block_align = st->codec.channels * st->codec.bits_per_sample;

    wsaud->audio_stream_index = st->index;
    wsaud->audio_frame_counter = 0;

    return 0;
}

static int wsaud_read_packet(AVFormatContext *s,
                             AVPacket *pkt)
{
    WsAudDemuxContext *wsaud = (WsAudDemuxContext *)s->priv_data;
    ByteIOContext *pb = &s->pb;
    unsigned char preamble[AUD_CHUNK_PREAMBLE_SIZE];
    unsigned int chunk_size;
    int ret = 0;

    if (get_buffer(pb, preamble, AUD_CHUNK_PREAMBLE_SIZE) !=
        AUD_CHUNK_PREAMBLE_SIZE)
        return -EIO;

    /* validate the chunk */
    if (LE_32(&preamble[4]) != AUD_CHUNK_SIGNATURE)
        return AVERROR_INVALIDDATA;

    chunk_size = LE_16(&preamble[0]);
    if (av_new_packet(pkt, chunk_size))
        return -EIO;
    pkt->stream_index = wsaud->audio_stream_index;
    pkt->pts = wsaud->audio_frame_counter;
    pkt->pts /= wsaud->audio_samplerate;
    if ((ret = get_buffer(pb, pkt->data, chunk_size)) != chunk_size) {
        av_free_packet(pkt);
        ret = -EIO;
    }

    /* 2 samples/byte, 1 or 2 samples per frame depending on stereo */
    wsaud->audio_frame_counter += (chunk_size * 2) / wsaud->audio_channels;

    return ret;
}

static int wsaud_read_close(AVFormatContext *s)
{
//    WsAudDemuxContext *wsaud = (WsAudDemuxContext *)s->priv_data;

    return 0;
}


static int wsvqa_probe(AVProbeData *p)
{
    /* need 12 bytes to qualify */
    if (p->buf_size < 12)
        return 0;

    /* check for the VQA signatures */
    if ((BE_32(&p->buf[0]) != FORM_TAG) ||
        (BE_32(&p->buf[8]) != WVQA_TAG))
        return 0;

    return AVPROBE_SCORE_MAX;
}

static int wsvqa_read_header(AVFormatContext *s,
                             AVFormatParameters *ap)
{
    WsVqaDemuxContext *wsvqa = (WsVqaDemuxContext *)s->priv_data;
    ByteIOContext *pb = &s->pb;
    AVStream *st;
    unsigned char *header;
    unsigned char scratch[VQA_PREAMBLE_SIZE];
238 239
    unsigned int chunk_tag;
    unsigned int chunk_size;
240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268

    /* set the pts reference (1 pts = 1/90000) */
    s->pts_num = 1;
    s->pts_den = 90000;

    /* initialize the video decoder stream */
    st = av_new_stream(s, 0);
    if (!st)
        return AVERROR_NOMEM;
    wsvqa->video_stream_index = st->index;
    st->codec.codec_type = CODEC_TYPE_VIDEO;
    st->codec.codec_id = CODEC_ID_WS_VQA;
    st->codec.codec_tag = 0;  /* no fourcc */

    /* skip to the start of the VQA header */
    url_fseek(pb, 20, SEEK_SET);

    /* the VQA header needs to go to the decoder */
    st->codec.extradata_size = VQA_HEADER_SIZE;
    st->codec.extradata = av_malloc(VQA_HEADER_SIZE);
    header = (unsigned char *)st->codec.extradata;
    if (get_buffer(pb, st->codec.extradata, VQA_HEADER_SIZE) !=
        VQA_HEADER_SIZE) {
        av_free(st->codec.extradata);
        return -EIO;
    }
    st->codec.width = LE_16(&header[6]);
    st->codec.height = LE_16(&header[8]);

269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288
    /* initialize the audio decoder stream is sample rate is non-zero */
    if (LE_16(&header[24])) {
        st = av_new_stream(s, 0);
        if (!st)
            return AVERROR_NOMEM;
        st->codec.codec_type = CODEC_TYPE_AUDIO;
        st->codec.codec_id = CODEC_ID_ADPCM_IMA_WS;
        st->codec.codec_tag = 0;  /* no tag */
        st->codec.sample_rate = LE_16(&header[24]);
        st->codec.channels = header[26];
        st->codec.bits_per_sample = 16;
        st->codec.bit_rate = st->codec.channels * st->codec.sample_rate *
            st->codec.bits_per_sample / 4;
        st->codec.block_align = st->codec.channels * st->codec.bits_per_sample;

        wsvqa->audio_stream_index = st->index;
        wsvqa->audio_samplerate = st->codec.sample_rate;
        wsvqa->audio_channels = st->codec.channels;
        wsvqa->audio_frame_counter = 0;
    }
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
    /* there are 0 or more chunks before the FINF chunk; iterate until
     * FINF has been skipped and the file will be ready to be demuxed */
    do {
        if (get_buffer(pb, scratch, VQA_PREAMBLE_SIZE) != VQA_PREAMBLE_SIZE) {
            av_free(st->codec.extradata);
            return -EIO;
        }
        chunk_tag = BE_32(&scratch[0]);
        chunk_size = BE_32(&scratch[4]);

        /* catch any unknown header tags, for curiousity */
        switch (chunk_tag) {
        case CINF_TAG:
        case CINH_TAG:
        case CIND_TAG:
        case PINF_TAG:
        case PINH_TAG:
        case PIND_TAG:
        case FINF_TAG:
            break;

        default:
            printf (" note: unknown chunk seen (%c%c%c%c)\n",
                scratch[0], scratch[1],
                scratch[2], scratch[3]);
            break;
        }

        url_fseek(pb, chunk_size, SEEK_CUR);
    } while (chunk_tag != FINF_TAG);
320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412

    wsvqa->video_pts = wsvqa->audio_frame_counter = 0;

    return 0;
}

static int wsvqa_read_packet(AVFormatContext *s,
                             AVPacket *pkt)
{
    WsVqaDemuxContext *wsvqa = (WsVqaDemuxContext *)s->priv_data;
    ByteIOContext *pb = &s->pb;
    int ret = 0;
    unsigned char preamble[VQA_PREAMBLE_SIZE];
    unsigned int chunk_type;
    unsigned int chunk_size;
    int skip_byte;

    if (get_buffer(pb, preamble, VQA_PREAMBLE_SIZE) != VQA_PREAMBLE_SIZE)
        return -EIO;

    chunk_type = BE_32(&preamble[0]);
    chunk_size = BE_32(&preamble[4]);
    skip_byte = chunk_size & 0x01;

    if ((chunk_type == SND2_TAG) || (chunk_type == VQFR_TAG)) {

        if (av_new_packet(pkt, chunk_size))
            return -EIO;
        ret = get_buffer(pb, pkt->data, chunk_size);
        if (ret != chunk_size) {
            av_free_packet(pkt);
            ret = -EIO;
        }

        if (chunk_type == SND2_TAG) {
            pkt->stream_index = wsvqa->audio_stream_index;

            pkt->pts = 90000;
            pkt->pts *= wsvqa->audio_frame_counter;
            pkt->pts /= wsvqa->audio_samplerate;

            /* 2 samples/byte, 1 or 2 samples per frame depending on stereo */
            wsvqa->audio_frame_counter += (chunk_size * 2) / 
                wsvqa->audio_channels;
        } else {
            pkt->stream_index = wsvqa->video_stream_index;
            pkt->pts = wsvqa->video_pts;
            wsvqa->video_pts += VQA_VIDEO_PTS_INC;
        }

    } else
        return AVERROR_INVALIDDATA;

    /* stay on 16-bit alignment */
    if (skip_byte)
        url_fseek(pb, 1, SEEK_CUR);

    return ret;
}

static int wsvqa_read_close(AVFormatContext *s)
{
//    WsVqaDemuxContext *wsvqa = (WsVqaDemuxContext *)s->priv_data;

    return 0;
}

static AVInputFormat wsaud_iformat = {
    "wsaud",
    "Westwood Studios audio format",
    sizeof(WsAudDemuxContext),
    wsaud_probe,
    wsaud_read_header,
    wsaud_read_packet,
    wsaud_read_close,
};

static AVInputFormat wsvqa_iformat = {
    "wsvqa",
    "Westwood Studios VQA format",
    sizeof(WsVqaDemuxContext),
    wsvqa_probe,
    wsvqa_read_header,
    wsvqa_read_packet,
    wsvqa_read_close,
};

int westwood_init(void)
{
    av_register_input_format(&wsaud_iformat);
    av_register_input_format(&wsvqa_iformat);
    return 0;
}