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
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
|
/*
* General DV muxer/demuxer
* Copyright (c) 2003 Roman Shaposhnick
*
* Many thanks to Dan Dennedy <dan@dennedy.org> for providing wealth
* of DV technical info.
*
* Raw DV format
* Copyright (c) 2002 Fabrice Bellard.
*
* 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
*/
#include <time.h>
#include "avformat.h"
#include "dvdata.h"
#include "dv.h"
struct DVDemuxContext {
AVFormatContext* fctx;
AVStream* vst;
AVStream* ast[2];
AVPacket audio_pkt[2];
uint8_t audio_buf[2][8192];
int ach;
int frames;
uint64_t abytes;
};
struct DVMuxContext {
const DVprofile* sys; /* Current DV profile. E.g.: 525/60, 625/50 */
uint8_t frame_buf[144000]; /* frame under contruction */
FifoBuffer audio_data; /* Fifo for storing excessive amounts of PCM */
int frames; /* Number of a current frame */
time_t start_time; /* Start time of recording */
uint8_t aspect; /* Aspect ID 0 - 4:3, 7 - 16:9 */
int has_audio; /* frame under contruction has audio */
int has_video; /* frame under contruction has video */
};
enum dv_section_type {
dv_sect_header = 0x1f,
dv_sect_subcode = 0x3f,
dv_sect_vaux = 0x56,
dv_sect_audio = 0x76,
dv_sect_video = 0x96,
};
enum dv_pack_type {
dv_header525 = 0x3f, /* see dv_write_pack for important details on */
dv_header625 = 0xbf, /* these two packs */
dv_timecode = 0x13,
dv_audio_source = 0x50,
dv_audio_control = 0x51,
dv_audio_recdate = 0x52,
dv_audio_rectime = 0x53,
dv_video_source = 0x60,
dv_video_control = 0x61,
dv_viedo_recdate = 0x62,
dv_video_rectime = 0x63,
dv_unknown_pack = 0xff,
};
/*
* The reason why the following three big ugly looking tables are
* here is my lack of DV spec IEC 61834. The tables were basically
* constructed to make code that places packs in SSYB, VAUX and
* AAUX blocks very simple and table-driven. They conform to the
* SMPTE 314M and the output of my personal DV camcorder, neither
* of which is sufficient for a reliable DV stream producing. Thus
* while code is still in development I'll be gathering input from
* people with different DV equipment and modifying the tables to
* accommodate all the quirks. Later on, if possible, some of them
* will be folded into smaller tables and/or switch-if logic. For
* now, my only excuse is -- they don't eat up that much of a space.
*/
static const int dv_ssyb_packs_dist[12][6] = {
{ 0x13, 0x13, 0x13, 0x13, 0x13, 0x13 },
{ 0x13, 0x13, 0x13, 0x13, 0x13, 0x13 },
{ 0x13, 0x13, 0x13, 0x13, 0x13, 0x13 },
{ 0x13, 0x13, 0x13, 0x13, 0x13, 0x13 },
{ 0x13, 0x13, 0x13, 0x13, 0x13, 0x13 },
{ 0x13, 0x13, 0x13, 0x13, 0x13, 0x13 },
{ 0x13, 0x62, 0x63, 0x13, 0x62, 0x63 },
{ 0x13, 0x62, 0x63, 0x13, 0x62, 0x63 },
{ 0x13, 0x62, 0x63, 0x13, 0x62, 0x63 },
{ 0x13, 0x62, 0x63, 0x13, 0x62, 0x63 },
{ 0x13, 0x62, 0x63, 0x13, 0x62, 0x63 },
{ 0x13, 0x62, 0x63, 0x13, 0x62, 0x63 },
};
static const int dv_vaux_packs_dist[12][15] = {
{ 0x60, 0x61, 0x62, 0x63, 0xff, 0xff, 0xff, 0xff, 0xff,
0x60, 0x61, 0x62, 0x63, 0xff, 0xff },
{ 0x60, 0x61, 0x62, 0x63, 0xff, 0xff, 0xff, 0xff, 0xff,
0x60, 0x61, 0x62, 0x63, 0xff, 0xff },
{ 0x60, 0x61, 0x62, 0x63, 0xff, 0xff, 0xff, 0xff, 0xff,
0x60, 0x61, 0x62, 0x63, 0xff, 0xff },
{ 0x60, 0x61, 0x62, 0x63, 0xff, 0xff, 0xff, 0xff, 0xff,
0x60, 0x61, 0x62, 0x63, 0xff, 0xff },
{ 0x60, 0x61, 0x62, 0x63, 0xff, 0xff, 0xff, 0xff, 0xff,
0x60, 0x61, 0x62, 0x63, 0xff, 0xff },
{ 0x60, 0x61, 0x62, 0x63, 0xff, 0xff, 0xff, 0xff, 0xff,
0x60, 0x61, 0x62, 0x63, 0xff, 0xff },
{ 0x60, 0x61, 0x62, 0x63, 0xff, 0xff, 0xff, 0xff, 0xff,
0x60, 0x61, 0x62, 0x63, 0xff, 0xff },
{ 0x60, 0x61, 0x62, 0x63, 0xff, 0xff, 0xff, 0xff, 0xff,
0x60, 0x61, 0x62, 0x63, 0xff, 0xff },
{ 0x60, 0x61, 0x62, 0x63, 0xff, 0xff, 0xff, 0xff, 0xff,
0x60, 0x61, 0x62, 0x63, 0xff, 0xff },
{ 0x60, 0x61, 0x62, 0x63, 0xff, 0xff, 0xff, 0xff, 0xff,
0x60, 0x61, 0x62, 0x63, 0xff, 0xff },
{ 0x60, 0x61, 0x62, 0x63, 0xff, 0xff, 0xff, 0xff, 0xff,
0x60, 0x61, 0x62, 0x63, 0xff, 0xff },
{ 0x60, 0x61, 0x62, 0x63, 0xff, 0xff, 0xff, 0xff, 0xff,
0x60, 0x61, 0x62, 0x63, 0xff, 0xff },
};
static const int dv_aaux_packs_dist[12][9] = {
{ 0xff, 0xff, 0xff, 0x50, 0x51, 0x52, 0x53, 0xff, 0xff },
{ 0x50, 0x51, 0x52, 0x53, 0xff, 0xff, 0xff, 0xff, 0xff },
{ 0xff, 0xff, 0xff, 0x50, 0x51, 0x52, 0x53, 0xff, 0xff },
{ 0x50, 0x51, 0x52, 0x53, 0xff, 0xff, 0xff, 0xff, 0xff },
{ 0xff, 0xff, 0xff, 0x50, 0x51, 0x52, 0x53, 0xff, 0xff },
{ 0x50, 0x51, 0x52, 0x53, 0xff, 0xff, 0xff, 0xff, 0xff },
{ 0xff, 0xff, 0xff, 0x50, 0x51, 0x52, 0x53, 0xff, 0xff },
{ 0x50, 0x51, 0x52, 0x53, 0xff, 0xff, 0xff, 0xff, 0xff },
{ 0xff, 0xff, 0xff, 0x50, 0x51, 0x52, 0x53, 0xff, 0xff },
{ 0x50, 0x51, 0x52, 0x53, 0xff, 0xff, 0xff, 0xff, 0xff },
{ 0xff, 0xff, 0xff, 0x50, 0x51, 0x52, 0x53, 0xff, 0xff },
{ 0x50, 0x51, 0x52, 0x53, 0xff, 0xff, 0xff, 0xff, 0xff },
};
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;
}
static int dv_audio_frame_size(const DVprofile* sys, int frame)
{
return sys->audio_samples_dist[frame % (sizeof(sys->audio_samples_dist)/
sizeof(sys->audio_samples_dist[0]))];
}
static int dv_write_pack(enum dv_pack_type pack_id, DVMuxContext *c, uint8_t* buf)
{
struct tm tc;
time_t ct;
int ltc_frame;
buf[0] = (uint8_t)pack_id;
switch (pack_id) {
case dv_header525: /* I can't imagine why these two weren't defined as real */
case dv_header625: /* packs in SMPTE314M -- they definitely look like ones */
buf[1] = 0xf8 | /* reserved -- always 1 */
(0 & 0x07); /* APT: Track application ID */
buf[2] = (0 << 7) | /* TF1: audio data is 0 - valid; 1 - invalid */
(0x0f << 3) | /* reserved -- always 1 */
(0 & 0x07); /* AP1: Audio application ID */
buf[3] = (0 << 7) | /* TF2: video data is 0 - valid; 1 - invalid */
(0x0f << 3) | /* reserved -- always 1 */
(0 & 0x07); /* AP2: Video application ID */
buf[4] = (0 << 7) | /* TF3: subcode(SSYB) is 0 - valid; 1 - invalid */
(0x0f << 3) | /* reserved -- always 1 */
(0 & 0x07); /* AP3: Subcode application ID */
break;
case dv_timecode:
ct = (time_t)(c->frames / ((float)c->sys->frame_rate /
(float)c->sys->frame_rate_base));
brktimegm(ct, &tc);
/*
* LTC drop-frame frame counter drops two frames (0 and 1) every
* minute, unless it is exactly divisible by 10
*/
ltc_frame = (c->frames + 2*ct/60 - 2*ct/600) % c->sys->ltc_divisor;
buf[1] = (0 << 7) | /* Color fame: 0 - unsync; 1 - sync mode */
(1 << 6) | /* Drop frame timecode: 0 - nondrop; 1 - drop */
((ltc_frame / 10) << 4) | /* Tens of frames */
(ltc_frame % 10); /* Units of frames */
buf[2] = (1 << 7) | /* Biphase mark polarity correction: 0 - even; 1 - odd */
((tc.tm_sec / 10) << 4) | /* Tens of seconds */
(tc.tm_sec % 10); /* Units of seconds */
buf[3] = (1 << 7) | /* Binary group flag BGF0 */
((tc.tm_min / 10) << 4) | /* Tens of minutes */
(tc.tm_min % 10); /* Units of minutes */
buf[4] = (1 << 7) | /* Binary group flag BGF2 */
(1 << 6) | /* Binary group flag BGF1 */
((tc.tm_hour / 10) << 4) | /* Tens of hours */
(tc.tm_hour % 10); /* Units of hours */
break;
case dv_audio_source: /* AAUX source pack */
buf[1] = (0 << 7) | /* locked mode */
(1 << 6) | /* reserved -- always 1 */
(dv_audio_frame_size(c->sys, c->frames) -
c->sys->audio_min_samples[0]);
/* # of samples */
buf[2] = (0 << 7) | /* multi-stereo */
(0 << 5) | /* #of audio channels per block: 0 -- 1 channel */
(0 << 4) | /* pair bit: 0 -- one pair of channels */
0; /* audio mode */
buf[3] = (1 << 7) | /* res */
(1 << 6) | /* multi-language flag */
(c->sys->dsf << 5) | /* system: 60fields/50fields */
0; /* definition: 0 -- SD (525/625) */
buf[4] = (1 << 7) | /* emphasis: 1 -- off */
(0 << 6) | /* emphasis time constant: 0 -- reserved */
(0 << 3) | /* frequency: 0 -- 48Khz, 1 -- 44,1Khz, 2 -- 32Khz */
0; /* quantization: 0 -- 16bit linear, 1 -- 12bit nonlinear */
break;
case dv_audio_control:
buf[1] = (0 << 6) | /* copy protection: 0 -- unrestricted */
(1 << 4) | /* input source: 1 -- digital input */
(3 << 2) | /* compression: 3 -- no information */
0; /* misc. info/SMPTE emphasis off */
buf[2] = (1 << 7) | /* recording start point: 1 -- no */
(1 << 6) | /* recording end point: 1 -- no */
(1 << 3) | /* recording mode: 1 -- original */
7;
buf[3] = (1 << 7) | /* direction: 1 -- forward */
0x20; /* speed */
buf[4] = (1 << 7) | /* reserved -- always 1 */
0x7f; /* genre category */
break;
case dv_audio_recdate:
case dv_viedo_recdate: /* VAUX recording date */
ct = c->start_time + (time_t)(c->frames /
((float)c->sys->frame_rate / (float)c->sys->frame_rate_base));
brktimegm(ct, &tc);
buf[1] = 0xff; /* ds, tm, tens of time zone, units of time zone */
/* 0xff is very likely to be "unknown" */
buf[2] = (3 << 6) | /* reserved -- always 1 */
((tc.tm_mday / 10) << 4) | /* Tens of day */
(tc.tm_mday % 10); /* Units of day */
buf[3] = /* we set high 4 bits to 0, shouldn't we set them to week? */
((tc.tm_mon / 10) << 4) | /* Tens of month */
(tc.tm_mon % 10); /* Units of month */
buf[4] = (((tc.tm_year % 100) / 10) << 4) | /* Tens of year */
(tc.tm_year % 10); /* Units of year */
break;
case dv_audio_rectime: /* AAUX recording time */
case dv_video_rectime: /* VAUX recording time */
ct = c->start_time + (time_t)(c->frames /
((float)c->sys->frame_rate / (float)c->sys->frame_rate_base));
brktimegm(ct, &tc);
buf[1] = (3 << 6) | /* reserved -- always 1 */
0x3f; /* tens of frame, units of frame: 0x3f - "unknown" ? */
buf[2] = (1 << 7) | /* reserved -- always 1 */
((tc.tm_sec / 10) << 4) | /* Tens of seconds */
(tc.tm_sec % 10); /* Units of seconds */
buf[3] = (1 << 7) | /* reserved -- always 1 */
((tc.tm_min / 10) << 4) | /* Tens of minutes */
(tc.tm_min % 10); /* Units of minutes */
buf[4] = (3 << 6) | /* reserved -- always 1 */
((tc.tm_hour / 10) << 4) | /* Tens of hours */
(tc.tm_hour % 10); /* Units of hours */
break;
case dv_video_source:
buf[1] = 0xff; /* reserved -- always 1 */
buf[2] = (1 << 7) | /* B/W: 0 - b/w, 1 - color */
(1 << 6) | /* following CLF is valid - 0, invalid - 1 */
(3 << 4) | /* CLF: color frames id (see ITU-R BT.470-4) */
0xf; /* reserved -- always 1 */
buf[3] = (3 << 6) | /* reserved -- always 1 */
(c->sys->dsf << 5) | /* system: 60fields/50fields */
0; /* signal type video compression */
buf[4] = 0xff; /* VISC: 0xff -- no information */
break;
case dv_video_control:
buf[1] = (0 << 6) | /* Copy generation management (CGMS) 0 -- free */
0x3f; /* reserved -- always 1 */
buf[2] = 0xc8 | /* reserved -- always b11001xxx */
c->aspect;
buf[3] = (1 << 7) | /* Frame/field flag 1 -- frame, 0 -- field */
(1 << 6) | /* First/second field flag 0 -- field 2, 1 -- field 1 */
(1 << 5) | /* Frame change flag 0 -- same picture as before, 1 -- different */
(1 << 4) | /* 1 - interlaced, 0 - noninterlaced */
0xc; /* reserved -- always b1100 */
buf[4] = 0xff; /* reserved -- always 1 */
break;
default:
buf[1] = buf[2] = buf[3] = buf[4] = 0xff;
}
return 5;
}
static inline int dv_write_dif_id(enum dv_section_type t, uint8_t seq_num,
uint8_t dif_num, uint8_t* buf)
{
buf[0] = (uint8_t)t; /* Section type */
buf[1] = (seq_num<<4) | /* DIF seq number 0-9 for 525/60; 0-11 for 625/50 */
(0 << 3) | /* FSC: for 50Mb/s 0 - first channel; 1 - second */
7; /* reserved -- always 1 */
buf[2] = dif_num; /* DIF block number Video: 0-134, Audio: 0-8 */
return 3;
}
static inline int dv_write_ssyb_id(uint8_t syb_num, uint8_t fr, uint8_t* buf)
{
if (syb_num == 0 || syb_num == 6) {
buf[0] = (fr<<7) | /* FR ID 1 - first half of each channel; 0 - second */
(0<<4) | /* AP3 (Subcode application ID) */
0x0f; /* reserved -- always 1 */
}
else if (syb_num == 11) {
buf[0] = (fr<<7) | /* FR ID 1 - first half of each channel; 0 - second */
0x7f; /* reserved -- always 1 */
}
else {
buf[0] = (fr<<7) | /* FR ID 1 - first half of each channel; 0 - second */
(0<<4) | /* APT (Track application ID) */
0x0f; /* reserved -- always 1 */
}
buf[1] = 0xf0 | /* reserved -- always 1 */
(syb_num & 0x0f); /* SSYB number 0 - 11 */
buf[2] = 0xff; /* reserved -- always 1 */
return 3;
}
static void dv_format_frame(DVMuxContext *c, uint8_t* buf)
{
int i, j, k;
for (i = 0; i < c->sys->difseg_size; i++) {
memset(buf, 0xff, 80 * 6); /* First 6 DIF blocks are for control data */
/* DV header: 1DIF */
buf += dv_write_dif_id(dv_sect_header, i, 0, buf);
buf += dv_write_pack((c->sys->dsf ? dv_header625 : dv_header525), c, buf);
buf += 72; /* unused bytes */
/* DV subcode: 2DIFs */
for (j = 0; j < 2; j++) {
buf += dv_write_dif_id( dv_sect_subcode, i, j, buf);
for (k = 0; k < 6; k++) {
buf += dv_write_ssyb_id(k, (i < c->sys->difseg_size/2), buf);
buf += dv_write_pack(dv_ssyb_packs_dist[i][k], c, buf);
}
buf += 29; /* unused bytes */
}
/* DV VAUX: 3DIFs */
for (j = 0; j < 3; j++) {
buf += dv_write_dif_id(dv_sect_vaux, i, j, buf);
for (k = 0; k < 15 ; k++)
buf += dv_write_pack(dv_vaux_packs_dist[i][k], c, buf);
buf += 2; /* unused bytes */
}
/* DV Audio/Video: 135 Video DIFs + 9 Audio DIFs */
for (j = 0; j < 135; j++) {
if (j%15 == 0) {
memset(buf, 0xff, 80);
buf += dv_write_dif_id(dv_sect_audio, i, j/15, buf);
buf += 77; /* audio control & shuffled PCM audio */
}
buf += dv_write_dif_id(dv_sect_video, i, j, buf);
buf += 77; /* 1 video macro block: 1 bytes control
4 * 14 bytes Y 8x8 data
10 bytes Cr 8x8 data
10 bytes Cb 8x8 data */
}
}
}
static void dv_inject_audio(DVMuxContext *c, const uint8_t* pcm, uint8_t* frame_ptr)
{
int i, j, d, of, size;
size = 4 * dv_audio_frame_size(c->sys, c->frames);
for (i = 0; i < c->sys->difseg_size; i++) {
frame_ptr += 6 * 80; /* skip DIF segment header */
for (j = 0; j < 9; j++) {
dv_write_pack(dv_aaux_packs_dist[i][j], c, &frame_ptr[3]);
for (d = 8; d < 80; d+=2) {
of = c->sys->audio_shuffle[i][j] + (d - 8)/2 * c->sys->audio_stride;
if (of*2 >= size)
continue;
frame_ptr[d] = pcm[of*2+1]; // FIXME: may be we have to admit
frame_ptr[d+1] = pcm[of*2]; // that DV is a big endian PCM
}
frame_ptr += 16 * 80; /* 15 Video DIFs + 1 Audio DIF */
}
}
}
static void dv_inject_video(DVMuxContext *c, const uint8_t* video_data, uint8_t* frame_ptr)
{
int i, j;
int ptr = 0;
for (i = 0; i < c->sys->difseg_size; i++) {
ptr += 6 * 80; /* skip DIF segment header */
for (j = 0; j < 135; j++) {
if (j%15 == 0)
ptr += 80; /* skip Audio DIF */
ptr += 3;
memcpy(frame_ptr + ptr, video_data + ptr, 77);
ptr += 77;
}
}
}
/*
* 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);
}
/*
* 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 ffmpeg 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* pcm, uint8_t* pcm2)
{
int size, i, j, d, of, smpls, freq, quant, half_ch;
uint16_t lc, rc;
const DVprofile* sys;
const uint8_t* as_pack;
as_pack = dv_extract_pack(frame, dv_audio_source);
if (!as_pack) /* No audio ? */
return 0;
sys = dv_frame_profile(frame);
smpls = as_pack[1] & 0x3f; /* samples in this frame - min. samples */
freq = (as_pack[4] >> 3) & 0x07; /* 0 - 48KHz, 1 - 44,1kHz, 2 - 32 kHz */
quant = as_pack[4] & 0x07; /* 0 - 16bit linear, 1 - 12bit nonlinear */
if (quant > 1)
return -1; /* Unsupported quantization */
size = (sys->audio_min_samples[freq] + smpls) * 4; /* 2ch, 2bytes */
half_ch = sys->difseg_size/2;
/* 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) {
if (!pcm2)
break;
else
pcm = pcm2;
}
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: may be 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: may be 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: may be 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;
const DVprofile* sys;
int freq, smpls, quant, i, ach;
sys = dv_frame_profile(frame);
as_pack = dv_extract_pack(frame, dv_audio_source);
if (!as_pack || !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 - 32 kHz */
quant = as_pack[4] & 0x07; /* 0 - 16bit linear, 1 - 12bit nonlinear */
ach = (quant && freq == 2) ? 2 : 1;
/* Dynamic handling of the audio streams in DV */
for (i=0; i<ach; i++) {
if (!c->ast[i]) {
c->ast[i] = av_new_stream(c->fctx, 0);
if (!c->ast[i])
break;
av_set_pts_info(c->ast[i], 64, 1, 30000);
c->ast[i]->codec.codec_type = CODEC_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 |= 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 (sys->audio_min_samples[freq] + smpls) * 4; /* 2ch, 2bytes */;
}
static int dv_extract_video_info(DVDemuxContext *c, uint8_t* frame)
{
const DVprofile* sys;
const uint8_t* vsc_pack;
AVCodecContext* avctx;
int apt, is16_9;
int size = 0;
sys = dv_frame_profile(frame);
if (sys) {
avctx = &c->vst->codec;
avctx->frame_rate = sys->frame_rate;
avctx->frame_rate_base = sys->frame_rate_base;
av_set_pts_info(c->vst, 64, sys->frame_rate_base, sys->frame_rate);
avctx->width = sys->width;
avctx->height = sys->height;
avctx->pix_fmt = 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)));
avctx->sample_aspect_ratio = sys->sar[is16_9];
size = sys->frame_size;
}
return size;
}
/*
* The following 6 functions constitute our interface to the world
*/
int dv_assemble_frame(DVMuxContext *c, AVStream* st,
const uint8_t* data, int data_size, uint8_t** frame)
{
uint8_t pcm[8192];
int fsize, reqasize;
*frame = &c->frame_buf[0];
if (c->has_audio && c->has_video) { /* must be a stale frame */
dv_format_frame(c, *frame);
c->frames++;
if (c->has_audio > 0)
c->has_audio = 0;
c->has_video = 0;
}
if (st->codec.codec_type == CODEC_TYPE_VIDEO) {
/* FIXME: we have to have more sensible approach than this one */
if (c->has_video)
av_log(&st->codec, AV_LOG_ERROR, "Can't process DV frame #%d. Insufficient audio data or severe sync problem.\n", c->frames);
dv_inject_video(c, data, *frame);
c->has_video = 1;
data_size = 0;
if (c->has_audio < 0)
goto out;
}
reqasize = 4 * dv_audio_frame_size(c->sys, c->frames);
fsize = fifo_size(&c->audio_data, c->audio_data.rptr);
if (st->codec.codec_type == CODEC_TYPE_AUDIO || (c->has_video && fsize >= reqasize)) {
if (fsize + data_size >= reqasize && !c->has_audio) {
if (fsize >= reqasize) {
fifo_read(&c->audio_data, &pcm[0], reqasize, &c->audio_data.rptr);
} else {
fifo_read(&c->audio_data, &pcm[0], fsize, &c->audio_data.rptr);
memcpy(&pcm[fsize], &data[0], reqasize - fsize);
data += reqasize - fsize;
data_size -= reqasize - fsize;
}
dv_inject_audio(c, &pcm[0], *frame);
c->has_audio = 1;
}
/* FIXME: we have to have more sensible approach than this one */
if (fifo_size(&c->audio_data, c->audio_data.rptr) + data_size >= 100*AVCODEC_MAX_AUDIO_FRAME_SIZE)
av_log(&st->codec, AV_LOG_ERROR, "Can't process DV frame #%d. Insufficient video data or severe sync problem.\n", c->frames);
fifo_write(&c->audio_data, (uint8_t *)data, data_size, &c->audio_data.wptr);
}
out:
return (c->has_audio && c->has_video) ? c->sys->frame_size : 0;
}
DVMuxContext* dv_init_mux(AVFormatContext* s)
{
DVMuxContext *c;
AVStream *vst = NULL;
AVStream *ast = NULL;
int i;
if (s->nb_streams > 2)
return NULL;
c = av_mallocz(sizeof(DVMuxContext));
if (!c)
return NULL;
/* We have to sort out where audio and where video stream is */
for (i=0; i<s->nb_streams; i++) {
switch (s->streams[i]->codec.codec_type) {
case CODEC_TYPE_VIDEO:
vst = s->streams[i];
break;
case CODEC_TYPE_AUDIO:
ast = s->streams[i];
break;
default:
goto bail_out;
}
}
/* Some checks -- DV format is very picky about its incoming streams */
if (!vst || vst->codec.codec_id != CODEC_ID_DVVIDEO)
goto bail_out;
if (ast && (ast->codec.codec_id != CODEC_ID_PCM_S16LE ||
ast->codec.sample_rate != 48000 ||
ast->codec.channels != 2))
goto bail_out;
c->sys = dv_codec_profile(&vst->codec);
if (!c->sys)
goto bail_out;
/* Ok, everything seems to be in working order */
c->frames = 0;
c->has_audio = ast ? 0 : -1;
c->has_video = 0;
c->start_time = (time_t)s->timestamp;
c->aspect = 0; /* 4:3 is the default */
if ((int)(av_q2d(vst->codec.sample_aspect_ratio) * vst->codec.width / vst->codec.height * 10) == 17) /* 16:9 */
c->aspect = 0x07;
if (ast && fifo_init(&c->audio_data, 100*AVCODEC_MAX_AUDIO_FRAME_SIZE) < 0)
goto bail_out;
dv_format_frame(c, &c->frame_buf[0]);
return c;
bail_out:
av_free(c);
return NULL;
}
void dv_delete_mux(DVMuxContext *c)
{
fifo_free(&c->audio_data);
}
DVDemuxContext* dv_init_demux(AVFormatContext *s)
{
DVDemuxContext *c;
c = av_mallocz(sizeof(DVDemuxContext));
if (!c)
return NULL;
c->vst = av_new_stream(s, 0);
if (!c->vst) {
av_free(c);
return NULL;
}
c->fctx = s;
c->ast[0] = c->ast[1] = NULL;
c->ach = 0;
c->frames = 0;
c->abytes = 0;
c->vst->codec.codec_type = CODEC_TYPE_VIDEO;
c->vst->codec.codec_id = CODEC_ID_DVVIDEO;
c->vst->codec.bit_rate = 25000000;
c->vst->start_time = 0;
return c;
}
int 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 dv_produce_packet(DVDemuxContext *c, AVPacket *pkt,
uint8_t* buf, int buf_size)
{
int size, i;
const DVprofile* sys = dv_frame_profile(buf);
if (buf_size < 4 || buf_size < 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;
}
dv_extract_audio(buf, c->audio_buf[0], c->audio_buf[1]);
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 |= 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 = dv_codec_profile(&c->vst->codec);
int64_t offset;
int64_t size = url_filesize(url_fileno(&s->pb));
int64_t max_offset = ((size-1) / sys->frame_size) * sys->frame_size;
offset = sys->frame_size * timestamp;
if (offset > max_offset) offset = max_offset;
else if (offset < 0) offset = 0;
return offset;
}
void dv_flush_audio_packets(DVDemuxContext *c)
{
c->audio_pkt[0].size = c->audio_pkt[1].size = 0;
}
/************************************************************
* Implementation of the easiest DV storage of all -- raw DV.
************************************************************/
typedef struct RawDVContext {
uint8_t buf[144000];
DVDemuxContext* dv_demux;
} RawDVContext;
static int dv_read_header(AVFormatContext *s,
AVFormatParameters *ap)
{
RawDVContext *c = s->priv_data;
const DVprofile* sys;
c->dv_demux = dv_init_demux(s);
if (!c->dv_demux)
return -1;
if (get_buffer(&s->pb, c->buf, 4) <= 0 || url_fseek(&s->pb, -4, SEEK_CUR) < 0)
return AVERROR_IO;
sys = dv_frame_profile(c->buf);
s->bit_rate = av_rescale(sys->frame_size * 8, sys->frame_rate, sys->frame_rate_base);
return 0;
}
static int dv_read_packet(AVFormatContext *s, AVPacket *pkt)
{
int size;
RawDVContext *c = s->priv_data;
size = dv_get_packet(c->dv_demux, pkt);
if (size < 0) {
if (get_buffer(&s->pb, c->buf, 4) <= 0)
return AVERROR_IO;
size = dv_frame_profile(c->buf)->frame_size;
if (get_buffer(&s->pb, c->buf + 4, size - 4) <= 0)
return AVERROR_IO;
size = 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);
const DVprofile* sys = dv_codec_profile(&c->vst->codec);
c->frames= offset / sys->frame_size;
if (c->ach)
c->abytes= av_rescale(c->frames,
c->ast[0]->codec.bit_rate * (int64_t)sys->frame_rate_base,
8*sys->frame_rate);
dv_flush_audio_packets(c);
return url_fseek(&s->pb, offset, SEEK_SET);
}
static int dv_read_close(AVFormatContext *s)
{
RawDVContext *c = s->priv_data;
av_free(c->dv_demux);
return 0;
}
static int dv_write_header(AVFormatContext *s)
{
s->priv_data = dv_init_mux(s);
if (!s->priv_data) {
av_log(s, AV_LOG_ERROR, "Can't initialize DV format!\n"
"Make sure that you supply exactly two streams:\n"
" video: 25fps or 29.97fps, audio: 2ch/48Khz/PCM\n");
return -1;
}
return 0;
}
static int dv_write_packet(struct AVFormatContext *s, AVPacket *pkt)
{
uint8_t* frame;
int fsize;
fsize = dv_assemble_frame((DVMuxContext *)s->priv_data, s->streams[pkt->stream_index],
pkt->data, pkt->size, &frame);
if (fsize > 0) {
put_buffer(&s->pb, frame, fsize);
put_flush_packet(&s->pb);
}
return 0;
}
/*
* We might end up with some extra A/V data without matching counterpart.
* E.g. video data without enough audio to write the complete frame.
* Currently we simply drop the last frame. I don't know whether this
* is the best strategy of all
*/
static int dv_write_trailer(struct AVFormatContext *s)
{
dv_delete_mux((DVMuxContext *)s->priv_data);
return 0;
}
static AVInputFormat dv_iformat = {
"dv",
"DV video format",
sizeof(RawDVContext),
NULL,
dv_read_header,
dv_read_packet,
dv_read_close,
dv_read_seek,
.extensions = "dv,dif",
};
static AVOutputFormat dv_oformat = {
"dv",
"DV video format",
NULL,
"dv",
sizeof(DVMuxContext),
CODEC_ID_PCM_S16LE,
CODEC_ID_DVVIDEO,
dv_write_header,
dv_write_packet,
dv_write_trailer,
};
int ff_dv_init(void)
{
av_register_input_format(&dv_iformat);
av_register_output_format(&dv_oformat);
return 0;
}
|