summaryrefslogtreecommitdiffstats
path: root/libavformat/dv.c
blob: e02fb53f98a0b0293e21c3a242a37ab3abcbba93 (plain)
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
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
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
238
239
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
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
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
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
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
/*
 * General DV muxer/demuxer
 * Copyright (c) 2003 Roman Shaposhnik
 *
 * Many thanks to Dan Dennedy <dan@dennedy.org> for providing wealth
 * of DV technical info.
 *
 * Raw DV format
 * Copyright (c) 2002 Fabrice Bellard
 *
 * 50 Mbps (DVCPRO50) and 100 Mbps (DVCPRO HD) support
 * Copyright (c) 2006 Daniel Maas <dmaas@maasdigital.com>
 * Funded by BBC Research & Development
 *
 * 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
 */
#include <time.h>
#include "avformat.h"
#include "internal.h"
#include "libavcodec/dv_profile.h"
#include "libavcodec/dvdata.h"
#include "libavutil/intreadwrite.h"
#include "libavutil/mathematics.h"
#include "dv.h"

struct DVDemuxContext {
    const DVprofile*  sys;    /* Current DV profile. E.g.: 525/60, 625/50 */
    AVFormatContext*  fctx;
    AVStream*         vst;
    AVStream*         ast[4];
    AVPacket          audio_pkt[4];
    uint8_t           audio_buf[4][8192];
    int               ach;
    int               frames;
    uint64_t          abytes;
};

static inline uint16_t dv_audio_12to16(uint16_t sample)
{
    uint16_t shift, result;

    sample = (sample < 0x800) ? sample : sample | 0xf000;
    shift  = (sample & 0xf00) >> 8;

    if (shift < 0x2 || shift > 0xd) {
        result = sample;
    } else if (shift < 0x8) {
        shift--;
        result = (sample - (256 * shift)) << shift;
    } else {
        shift = 0xe - shift;
        result = ((sample + ((256 * shift) + 1)) << shift) - 1;
    }

    return result;
}

/*
 * This is the dumbest implementation of all -- it simply looks at
 * a fixed offset and if pack isn't there -- fails. We might want
 * to have a fallback mechanism for complete search of missing packs.
 */
static const uint8_t* dv_extract_pack(uint8_t* frame, enum dv_pack_type t)
{
    int offs;

    switch (t) {
    case dv_audio_source:
        offs = (80*6 + 80*16*3 + 3);
        break;
    case dv_audio_control:
        offs = (80*6 + 80*16*4 + 3);
        break;
    case dv_video_control:
        offs = (80*5 + 48 + 5);
        break;
    default:
        return NULL;
    }

    return frame[offs] == t ? &frame[offs] : NULL;
}

static const int dv_audio_frequency[3] = {
    48000, 44100, 32000,
};

/*
 * There's a couple of assumptions being made here:
 * 1. By default we silence erroneous (0x8000/16bit 0x800/12bit) audio samples.
 *    We can pass them upwards when libavcodec will be ready to deal with them.
 * 2. We don't do software emphasis.
 * 3. Audio is always returned as 16bit linear samples: 12bit nonlinear samples
 *    are converted into 16bit linear ones.
 */
static int dv_extract_audio(uint8_t* frame, uint8_t* ppcm[4],
                            const DVprofile *sys)
{
    int size, chan, i, j, d, of, smpls, freq, quant, half_ch;
    uint16_t lc, rc;
    const uint8_t* as_pack;
    uint8_t *pcm, ipcm;

    as_pack = dv_extract_pack(frame, dv_audio_source);
    if (!as_pack)    /* No audio ? */
        return 0;

    smpls =  as_pack[1] & 0x3f;       /* samples in this frame - min. samples */
    freq  = (as_pack[4] >> 3) & 0x07; /* 0 - 48kHz, 1 - 44,1kHz, 2 - 32kHz */
    quant =  as_pack[4] & 0x07;       /* 0 - 16bit linear, 1 - 12bit nonlinear */

    if (quant > 1)
        return -1; /* unsupported quantization */

    if (freq >= FF_ARRAY_ELEMS(dv_audio_frequency))
        return AVERROR_INVALIDDATA;

    size = (sys->audio_min_samples[freq] + smpls) * 4; /* 2ch, 2bytes */
    half_ch = sys->difseg_size / 2;

    /* We work with 720p frames split in half, thus even frames have
     * channels 0,1 and odd 2,3. */
    ipcm = (sys->height == 720 && !(frame[1] & 0x0C)) ? 2 : 0;

    /* for each DIF channel */
    for (chan = 0; chan < sys->n_difchan; chan++) {
        /* next stereo channel (50Mbps and 100Mbps only) */
        pcm = ppcm[ipcm++];
        if (!pcm)
            break;

        /* for each DIF segment */
        for (i = 0; i < sys->difseg_size; i++) {
            frame += 6 * 80; /* skip DIF segment header */
            if (quant == 1 && i == half_ch) {
                /* next stereo channel (12bit mode only) */
                pcm = ppcm[ipcm++];
                if (!pcm)
                    break;
            }

            /* for each AV sequence */
            for (j = 0; j < 9; j++) {
                for (d = 8; d < 80; d += 2) {
                    if (quant == 0) {  /* 16bit quantization */
                        of = sys->audio_shuffle[i][j] + (d - 8) / 2 * sys->audio_stride;
                        if (of*2 >= size)
                            continue;

                        pcm[of*2]   = frame[d+1]; // FIXME: maybe we have to admit
                        pcm[of*2+1] = frame[d];   //        that DV is a big-endian PCM
                        if (pcm[of*2+1] == 0x80 && pcm[of*2] == 0x00)
                            pcm[of*2+1] = 0;
                    } else {           /* 12bit quantization */
                        lc = ((uint16_t)frame[d]   << 4) |
                             ((uint16_t)frame[d+2] >> 4);
                        rc = ((uint16_t)frame[d+1] << 4) |
                             ((uint16_t)frame[d+2] & 0x0f);
                        lc = (lc == 0x800 ? 0 : dv_audio_12to16(lc));
                        rc = (rc == 0x800 ? 0 : dv_audio_12to16(rc));

                        of = sys->audio_shuffle[i%half_ch][j] + (d - 8) / 3 * sys->audio_stride;
                        if (of*2 >= size)
                            continue;

                        pcm[of*2]   = lc & 0xff; // FIXME: maybe we have to admit
                        pcm[of*2+1] = lc >> 8;   //        that DV is a big-endian PCM
                        of = sys->audio_shuffle[i%half_ch+half_ch][j] +
                            (d - 8) / 3 * sys->audio_stride;
                        pcm[of*2]   = rc & 0xff; // FIXME: maybe we have to admit
                        pcm[of*2+1] = rc >> 8;   //        that DV is a big-endian PCM
                        ++d;
                    }
                }

                frame += 16 * 80; /* 15 Video DIFs + 1 Audio DIF */
            }
        }
    }

    return size;
}

static int dv_extract_audio_info(DVDemuxContext* c, uint8_t* frame)
{
    const uint8_t* as_pack;
    int freq, stype, smpls, quant, i, ach;

    as_pack = dv_extract_pack(frame, dv_audio_source);
    if (!as_pack || !c->sys) {    /* No audio ? */
        c->ach = 0;
        return 0;
    }

    smpls =  as_pack[1] & 0x3f;       /* samples in this frame - min. samples */
    freq  = (as_pack[4] >> 3) & 0x07; /* 0 - 48kHz, 1 - 44,1kHz, 2 - 32kHz */
    stype = (as_pack[3] & 0x1f);      /* 0 - 2CH, 2 - 4CH, 3 - 8CH */
    quant =  as_pack[4] & 0x07;       /* 0 - 16bit linear, 1 - 12bit nonlinear */

    if (freq >= FF_ARRAY_ELEMS(dv_audio_frequency)) {
        av_log(c->fctx, AV_LOG_ERROR,
               "Unrecognized audio sample rate index (%d)\n", freq);
        return 0;
    }

    if (stype > 3) {
        av_log(c->fctx, AV_LOG_ERROR, "stype %d is invalid\n", stype);
        c->ach = 0;
        return 0;
    }

    /* note: ach counts PAIRS of channels (i.e. stereo channels) */
    ach = ((int[4]){  1,  0,  2,  4})[stype];
    if (ach == 1 && quant && freq == 2)
        ach = 2;

    /* Dynamic handling of the audio streams in DV */
    for (i = 0; i < ach; i++) {
       if (!c->ast[i]) {
           c->ast[i] = avformat_new_stream(c->fctx, NULL);
           if (!c->ast[i])
               break;
           avpriv_set_pts_info(c->ast[i], 64, 1, 30000);
           c->ast[i]->codec->codec_type = AVMEDIA_TYPE_AUDIO;
           c->ast[i]->codec->codec_id   = CODEC_ID_PCM_S16LE;

           av_init_packet(&c->audio_pkt[i]);
           c->audio_pkt[i].size         = 0;
           c->audio_pkt[i].data         = c->audio_buf[i];
           c->audio_pkt[i].stream_index = c->ast[i]->index;
           c->audio_pkt[i].flags       |= AV_PKT_FLAG_KEY;
       }
       c->ast[i]->codec->sample_rate = dv_audio_frequency[freq];
       c->ast[i]->codec->channels    = 2;
       c->ast[i]->codec->bit_rate    = 2 * dv_audio_frequency[freq] * 16;
       c->ast[i]->start_time         = 0;
    }
    c->ach = i;

    return (c->sys->audio_min_samples[freq] + smpls) * 4; /* 2ch, 2bytes */;
}

static int dv_extract_video_info(DVDemuxContext *c, uint8_t* frame)
{
    const uint8_t* vsc_pack;
    AVCodecContext* avctx;
    int apt, is16_9;
    int size = 0;

    if (c->sys) {
        avctx = c->vst->codec;

        avpriv_set_pts_info(c->vst, 64, c->sys->time_base.num,
                        c->sys->time_base.den);
        avctx->time_base= c->sys->time_base;
        if (!avctx->width){
            avctx->width = c->sys->width;
            avctx->height = c->sys->height;
        }
        avctx->pix_fmt = c->sys->pix_fmt;

        /* finding out SAR is a little bit messy */
        vsc_pack = dv_extract_pack(frame, dv_video_control);
        apt      = frame[4] & 0x07;
        is16_9   = (vsc_pack && ((vsc_pack[2] & 0x07) == 0x02 ||
                                (!apt && (vsc_pack[2] & 0x07) == 0x07)));
        c->vst->sample_aspect_ratio = c->sys->sar[is16_9];
        avctx->bit_rate = av_rescale_q(c->sys->frame_size, (AVRational){8,1},
                                       c->sys->time_base);
        size = c->sys->frame_size;
    }
    return size;
}

/*
 * The following 3 functions constitute our interface to the world
 */

DVDemuxContext* avpriv_dv_init_demux(AVFormatContext *s)
{
    DVDemuxContext *c;

    c = av_mallocz(sizeof(DVDemuxContext));
    if (!c)
        return NULL;

    c->vst = avformat_new_stream(s, NULL);
    if (!c->vst) {
        av_free(c);
        return NULL;
    }

    c->fctx                   = s;
    c->vst->codec->codec_type = AVMEDIA_TYPE_VIDEO;
    c->vst->codec->codec_id   = CODEC_ID_DVVIDEO;
    c->vst->codec->bit_rate   = 25000000;
    c->vst->start_time        = 0;

    return c;
}

int avpriv_dv_get_packet(DVDemuxContext *c, AVPacket *pkt)
{
    int size = -1;
    int i;

    for (i = 0; i < c->ach; i++) {
       if (c->ast[i] && c->audio_pkt[i].size) {
           *pkt = c->audio_pkt[i];
           c->audio_pkt[i].size = 0;
           size = pkt->size;
           break;
       }
    }

    return size;
}

int avpriv_dv_produce_packet(DVDemuxContext *c, AVPacket *pkt,
                      uint8_t* buf, int buf_size)
{
    int size, i;
    uint8_t *ppcm[4] = {0};

    if (buf_size < DV_PROFILE_BYTES ||
        !(c->sys = avpriv_dv_frame_profile(c->sys, buf, buf_size)) ||
        buf_size < c->sys->frame_size) {
          return -1;   /* Broken frame, or not enough data */
    }

    /* Queueing audio packet */
    /* FIXME: in case of no audio/bad audio we have to do something */
    size = dv_extract_audio_info(c, buf);
    for (i = 0; i < c->ach; i++) {
       c->audio_pkt[i].size = size;
       c->audio_pkt[i].pts  = c->abytes * 30000*8 / c->ast[i]->codec->bit_rate;
       ppcm[i] = c->audio_buf[i];
    }
    if (c->ach)
        dv_extract_audio(buf, ppcm, c->sys);

    /* We work with 720p frames split in half, thus even frames have
     * channels 0,1 and odd 2,3. */
    if (c->sys->height == 720) {
        if (buf[1] & 0x0C) {
            c->audio_pkt[2].size = c->audio_pkt[3].size = 0;
        } else {
            c->audio_pkt[0].size = c->audio_pkt[1].size = 0;
            c->abytes += size;
        }
    } else {
        c->abytes += size;
    }

    /* Now it's time to return video packet */
    size = dv_extract_video_info(c, buf);
    av_init_packet(pkt);
    pkt->data         = buf;
    pkt->size         = size;
    pkt->flags       |= AV_PKT_FLAG_KEY;
    pkt->stream_index = c->vst->id;
    pkt->pts          = c->frames;

    c->frames++;

    return size;
}

static int64_t dv_frame_offset(AVFormatContext *s, DVDemuxContext *c,
                              int64_t timestamp, int flags)
{
    // FIXME: sys may be wrong if last dv_read_packet() failed (buffer is junk)
    const DVprofile* sys = avpriv_dv_codec_profile(c->vst->codec);
    int64_t offset;
    int64_t size = avio_size(s->pb) - s->data_offset;
    int64_t max_offset = ((size-1) / sys->frame_size) * sys->frame_size;

    offset = sys->frame_size * timestamp;

    if (size >= 0 && offset > max_offset) offset = max_offset;
    else if (offset < 0) offset = 0;

    return offset + s->data_offset;
}

void ff_dv_offset_reset(DVDemuxContext *c, int64_t frame_offset)
{
    c->frames= frame_offset;
    if (c->ach)
        c->abytes= av_rescale_q(c->frames, c->sys->time_base,
                                (AVRational){8, c->ast[0]->codec->bit_rate});
    c->audio_pkt[0].size = c->audio_pkt[1].size = 0;
    c->audio_pkt[2].size = c->audio_pkt[3].size = 0;
}

/************************************************************
 * Implementation of the easiest DV storage of all -- raw DV.
 ************************************************************/

typedef struct RawDVContext {
    DVDemuxContext* dv_demux;
    uint8_t         buf[DV_MAX_FRAME_SIZE];
} RawDVContext;

static int dv_read_header(AVFormatContext *s)
{
    unsigned state, marker_pos = 0;
    RawDVContext *c = s->priv_data;

    c->dv_demux = avpriv_dv_init_demux(s);
    if (!c->dv_demux)
        return -1;

    state = avio_rb32(s->pb);
    while ((state & 0xffffff7f) != 0x1f07003f) {
        if (s->pb->eof_reached) {
            av_log(s, AV_LOG_ERROR, "Cannot find DV header.\n");
            return -1;
        }
        if (state == 0x003f0700 || state == 0xff3f0700)
            marker_pos = avio_tell(s->pb);
        if (state == 0xff3f0701 && avio_tell(s->pb) - marker_pos == 80) {
            avio_seek(s->pb, -163, SEEK_CUR);
            state = avio_rb32(s->pb);
            break;
        }
        state = (state << 8) | avio_r8(s->pb);
    }
    AV_WB32(c->buf, state);

    if (avio_read(s->pb, c->buf + 4, DV_PROFILE_BYTES - 4) <= 0 ||
        avio_seek(s->pb, -DV_PROFILE_BYTES, SEEK_CUR) < 0)
        return AVERROR(EIO);

    c->dv_demux->sys = avpriv_dv_frame_profile(c->dv_demux->sys, c->buf, DV_PROFILE_BYTES);
    if (!c->dv_demux->sys) {
        av_log(s, AV_LOG_ERROR, "Can't determine profile of DV input stream.\n");
        return -1;
    }

    s->bit_rate = av_rescale_q(c->dv_demux->sys->frame_size, (AVRational){8,1},
                               c->dv_demux->sys->time_base);

    return 0;
}


static int dv_read_packet(AVFormatContext *s, AVPacket *pkt)
{
    int size;
    RawDVContext *c = s->priv_data;

    size = avpriv_dv_get_packet(c->dv_demux, pkt);

    if (size < 0) {
        if (!c->dv_demux->sys)
            return AVERROR(EIO);
        size = c->dv_demux->sys->frame_size;
        if (avio_read(s->pb, c->buf, size) <= 0)
            return AVERROR(EIO);

        size = avpriv_dv_produce_packet(c->dv_demux, pkt, c->buf, size);
    }

    return size;
}

static int dv_read_seek(AVFormatContext *s, int stream_index,
                       int64_t timestamp, int flags)
{
    RawDVContext *r   = s->priv_data;
    DVDemuxContext *c = r->dv_demux;
    int64_t offset    = dv_frame_offset(s, c, timestamp, flags);

    if (avio_seek(s->pb, offset, SEEK_SET) < 0)
        return -1;

    ff_dv_offset_reset(c, offset / c->sys->frame_size);
    return 0;
}

static int dv_read_close(AVFormatContext *s)
{
    RawDVContext *c = s->priv_data;
    av_free(c->dv_demux);
    return 0;
}

static int dv_probe(AVProbeData *p)
{
    unsigned state, marker_pos = 0;
    int i;
    int matches = 0;
    int secondary_matches = 0;

    if (p->buf_size < 5)
        return 0;

    state = AV_RB32(p->buf);
    for (i = 4; i < p->buf_size; i++) {
        if ((state & 0xffffff7f) == 0x1f07003f)
            matches++;
        // any section header, also with seq/chan num != 0,
        // should appear around every 12000 bytes, at least 10 per frame
        if ((state & 0xff07ff7f) == 0x1f07003f)
            secondary_matches++;
        if (state == 0x003f0700 || state == 0xff3f0700)
            marker_pos = i;
        if (state == 0xff3f0701 && i - marker_pos == 80)
            matches++;
        state = (state << 8) | p->buf[i];
    }

    if (matches && p->buf_size / matches < 1024*1024) {
        if (matches > 4 || (secondary_matches >= 10 && p->buf_size / secondary_matches < 24000))
            return AVPROBE_SCORE_MAX*3/4; // not max to avoid dv in mov to match
        return AVPROBE_SCORE_MAX/4;
    }
    return 0;
}

#if CONFIG_DV_DEMUXER
AVInputFormat ff_dv_demuxer = {
    .name           = "dv",
    .long_name      = NULL_IF_CONFIG_SMALL("DV video format"),
    .priv_data_size = sizeof(RawDVContext),
    .read_probe     = dv_probe,
    .read_header    = dv_read_header,
    .read_packet    = dv_read_packet,
    .read_close     = dv_read_close,
    .read_seek      = dv_read_seek,
    .extensions     = "dv,dif",
};
#endif
OpenPOWER on IntegriCloud