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
|
/* $Id: decmpeg2.c,v 1.12 2005/03/03 16:30:42 titer Exp $
This file is part of the HandBrake source code.
Homepage: <http://handbrake.fr/>.
It may be used under the terms of the GNU General Public License. */
#include "hb.h"
#include "hbffmpeg.h"
#include "mpeg2dec/mpeg2.h"
/* Cadence tracking */
#ifndef PIC_FLAG_REPEAT_FIRST_FIELD
#define PIC_FLAG_REPEAT_FIRST_FIELD 256
#endif
#define TOP_FIRST PIC_FLAG_TOP_FIELD_FIRST
#define PROGRESSIVE PIC_FLAG_PROGRESSIVE_FRAME
#define COMPOSITE PIC_FLAG_COMPOSITE_DISPLAY
#define SKIP PIC_FLAG_SKIP
#define TAGS PIC_FLAG_TAGS
#define REPEAT_FIRST PIC_FLAG_REPEAT_FIRST_FIELD
#define COMPOSITE_MASK PIC_MASK_COMPOSITE_DISPLAY
#define TB 8
#define BT 16
#define BT_PROG 32
#define BTB_PROG 64
#define TB_PROG 128
#define TBT_PROG 256
#define NTAGS 8
/**********************************************************************
* hb_libmpeg2_t
*********************************************************************/
typedef struct hb_libmpeg2_s
{
mpeg2dec_t * libmpeg2;
const mpeg2_info_t * info;
hb_job_t * job;
hb_title_t * title;
int width;
int height;
int rate;
double aspect_ratio;
enum PixelFormat pixfmt;
int got_iframe; /* set when we get our first iframe */
int look_for_iframe; /* need an iframe to add chap break */
int look_for_break; /* need gop start to add chap break */
int cur_tag; /* index of current tag */
uint32_t nframes; /* number of frames we've decoded */
int64_t last_pts;
int64_t first_pts;
int cadence[12];
int flag;
hb_list_t * list_subtitle;
hb_buffer_t * last_cc1_buf;
struct {
int64_t start; // start time of this frame
hb_buffer_t * cc_buf; // captions for this frame
} tags[NTAGS];
} hb_libmpeg2_t;
/**********************************************************************
* hb_libmpeg2_init
**********************************************************************
*
*********************************************************************/
static hb_libmpeg2_t * hb_libmpeg2_init()
{
hb_libmpeg2_t * m = calloc( sizeof( hb_libmpeg2_t ), 1 );
m->libmpeg2 = mpeg2_init();
m->info = mpeg2_info( m->libmpeg2 );
m->last_pts = -1;
m->first_pts = -1;
int i;
for ( i = 0; i < NTAGS; ++i )
{
m->tags[i].start = -1;
}
return m;
}
// send cc_buf to the CC decoder(s)
static void cc_send_to_decoder( hb_libmpeg2_t *m, hb_buffer_t *cc_buf )
{
hb_subtitle_t *subtitle;
// if there's more than one decoder for the captions send a copy
// of the buffer to all but the last. Then send the buffer to
// the last one (usually there's just one decoder so the 'while' is skipped).
int i = 0, n = hb_list_count( m->list_subtitle );
while ( --n > 0 )
{
// make a copy of the buf then forward it to the decoder
hb_buffer_t *cpy = hb_buffer_init( cc_buf->size );
hb_buffer_copy_settings( cpy, cc_buf );
memcpy( cpy->data, cc_buf->data, cc_buf->size );
subtitle = hb_list_item( m->list_subtitle, i++ );
hb_fifo_push( subtitle->fifo_in, cpy );
}
subtitle = hb_list_item( m->list_subtitle, i );
hb_fifo_push( subtitle->fifo_in, cc_buf );
}
static void hb_mpeg2_cc( hb_libmpeg2_t *m, const uint8_t *cc_block )
{
uint8_t cc_hdr = *cc_block;
if ( ( cc_hdr & 0x4 ) == 0 )
// not valid - ignore
return;
switch (cc_hdr & 3)
{
case 0:
// CC1 stream
if ( ( cc_block[1] & 0x7f ) == 0 && ( cc_block[2] & 0x7f ) == 0 )
// just padding - ignore
return;
if ( m->last_cc1_buf )
{
// new data from the same time as last call - add to buffer
int len = m->last_cc1_buf->size;
hb_buffer_realloc( m->last_cc1_buf, len + 2 );
memcpy( m->last_cc1_buf->data + len, cc_block+1, 2 );
m->last_cc1_buf->size = len + 2;
return;
}
// allocate a new buffer and copy the caption data into it.
// (we don't send it yet because we don't know what timestamp to use).
hb_buffer_t *cc_buf = hb_buffer_init( 2 );
if( !cc_buf )
return;
memcpy( cc_buf->data, cc_block+1, 2 );
m->last_cc1_buf = cc_buf;
break;
#ifdef notyet
case 1:
// CC2 stream
//process608( cc_block+1, 2, &m->cc608 );
break;
case 2: //EIA-708
// DTVCC packet data
// Fall through
case 3: //EIA-708
{
uint8_t temp[4];
temp[0]=cc_valid;
temp[1]=cc_type;
temp[2]=cc_block[1];
temp[3]=cc_block[2];
do_708 ((const unsigned char *) temp, 4);
}
break;
#endif
default:
break;
}
}
static inline int have_captions( const uint8_t *user_data, uint32_t len )
{
return len >= 6 &&
( ( user_data[0] == 0x43 && user_data[1] == 0x43 ) ||
( user_data[0] == 0x47 && user_data[1] == 0x41 &&
user_data[2] == 0x39 && user_data[3] == 0x34 &&
user_data[4] == 3 && (user_data[5] & 0x40) ) );
}
static void do_one_dvd_cc( hb_libmpeg2_t *m, const uint8_t *header, int field1 )
{
uint8_t data[3];
data[0] = ( header[0] == 0xff && 0 == field1 )? 0x04 : 0x05;
data[1] = header[1];
data[2] = header[2];
hb_mpeg2_cc( m, data );
data[0] = ( header[3] == 0xff && 1 == field1 )? 0x04 : 0x05;
data[1] = header[4];
data[2] = header[5];
hb_mpeg2_cc( m, data );
}
// extract all the captions in the current frame and send them downstream
// to the decoder.
//
// (this routine should only be called if there are captions in the current
// frame. I.e., only if a call to 'have_captions' returns true.)
static void extract_mpeg2_captions( hb_libmpeg2_t *m )
{
const uint8_t *user_data = m->info->user_data;
int dvd_captions = user_data[0] == 0x43;
int capcount, field1packet = 0;
const uint8_t *header = &user_data[4];
if ( !dvd_captions )
{
// ATSC encapsulated captions - header starts one byte later
// and has an extra unused byte following it.
capcount = header[1] & 0x1f;
header += 3;
}
else
{
// DVD captions
if ( ( header[0] & 0x80 ) == 0x00 )
field1packet=1; /* expect Field 1 second */
capcount=(header[0] & 0x1e) / 2;
header++;
}
int i;
for( i=0; i<capcount; i++ )
{
if ( !dvd_captions )
{
hb_mpeg2_cc( m, header );
header += 3;
}
else
{
do_one_dvd_cc( m, header, field1packet );
header += 6;
}
}
if ( dvd_captions )
{
// Deal with extra closed captions some DVDs have.
while( header[0]==0xfe || header[0]==0xff )
{
do_one_dvd_cc( m, header, field1packet );
header += 6;
}
}
}
static void next_tag( hb_libmpeg2_t *m, hb_buffer_t *buf_es )
{
m->cur_tag = ( m->cur_tag + 1 ) & (NTAGS-1);
if ( m->tags[m->cur_tag].start >= 0 || m->tags[m->cur_tag].cc_buf )
{
if ( m->tags[m->cur_tag].start < 0 ||
( m->got_iframe && m->tags[m->cur_tag].start >= m->first_pts ) )
hb_log("mpeg2 tag botch: pts %"PRId64", tag pts %"PRId64" buf 0x%p",
buf_es->start, m->tags[m->cur_tag].start, m->tags[m->cur_tag].cc_buf);
if ( m->tags[m->cur_tag].cc_buf )
hb_buffer_close( &m->tags[m->cur_tag].cc_buf );
}
m->tags[m->cur_tag].start = buf_es->start;
mpeg2_tag_picture( m->libmpeg2, m->cur_tag, 0 );
}
static hb_buffer_t *hb_copy_frame( hb_job_t *job, int width, int height,
enum PixelFormat pixfmt,
uint8_t* y, uint8_t *u, uint8_t *v )
{
int dst_w = width, dst_h = height;
if ( job )
{
dst_w = job->title->width;
dst_h = job->title->height;
}
int dst_wh = dst_w * dst_h;
hb_buffer_t *buf = hb_video_buffer_init( dst_w, dst_h );
buf->start = -1;
if ( dst_w != width || dst_h != height || pixfmt == PIX_FMT_YUV422P )
{
// we're encoding and the frame dimensions don't match the title dimensions -
// rescale & matte Y, U, V into our output buf.
AVPicture in, out;
avpicture_alloc(&in, pixfmt, width, height );
avpicture_alloc(&out, PIX_FMT_YUV420P, dst_w, dst_h );
int src_wh = width * height;
if ( pixfmt == PIX_FMT_YUV422P )
{
memcpy( in.data[0], y, src_wh );
memcpy( in.data[1], u, src_wh >> 1 );
memcpy( in.data[2], v, src_wh >> 1 );
}
else
{
memcpy( in.data[0], y, src_wh );
memcpy( in.data[1], u, src_wh >> 2 );
memcpy( in.data[2], v, src_wh >> 2 );
}
struct SwsContext *context = sws_getContext( width, height, pixfmt,
dst_w, dst_h, PIX_FMT_YUV420P,
SWS_LANCZOS|SWS_ACCURATE_RND,
NULL, NULL, NULL );
sws_scale( context, in.data, in.linesize, 0, height, out.data, out.linesize );
sws_freeContext( context );
uint8_t *data = buf->data;
memcpy( data, out.data[0], dst_wh );
data += dst_wh;
// U & V planes are 1/4 the size of Y plane.
dst_wh >>= 2;
memcpy( data, out.data[1], dst_wh );
data += dst_wh;
memcpy( data, out.data[2], dst_wh );
avpicture_free( &out );
avpicture_free( &in );
}
else
{
// we're scanning or the frame dimensions match the title's dimensions
// so we can do a straight copy.
uint8_t *data = buf->data;
memcpy( data, y, dst_wh );
data += dst_wh;
// U & V planes are 1/4 the size of Y plane.
dst_wh >>= 2;
memcpy( data, u, dst_wh );
data += dst_wh;
memcpy( data, v, dst_wh );
}
return buf;
}
/**********************************************************************
* hb_libmpeg2_decode
**********************************************************************
*
*********************************************************************/
static int hb_libmpeg2_decode( hb_libmpeg2_t * m, hb_buffer_t * buf_es,
hb_list_t * list_raw )
{
mpeg2_state_t state;
hb_buffer_t * buf;
if ( buf_es->size )
{
/* Feed libmpeg2 */
if( buf_es->start >= 0 )
{
next_tag( m, buf_es );
}
mpeg2_buffer( m->libmpeg2, buf_es->data, buf_es->data + buf_es->size );
}
for( ;; )
{
state = mpeg2_parse( m->libmpeg2 );
if( state == STATE_BUFFER )
{
/* Require some more data */
break;
}
// if the user requested captions, process
// any captions found in the current frame.
if ( m->list_subtitle && m->last_pts >= 0 &&
have_captions( m->info->user_data, m->info->user_data_len ) )
{
extract_mpeg2_captions( m );
// if we don't have a tag for the captions, make one
if ( m->last_cc1_buf && m->tags[m->cur_tag].cc_buf != m->last_cc1_buf )
{
if (m->tags[m->cur_tag].cc_buf)
{
hb_log("mpeg2 tag botch2: pts %"PRId64", tag pts %"PRId64" buf 0x%p",
buf_es->start, m->tags[m->cur_tag].start, m->tags[m->cur_tag].cc_buf);
hb_buffer_close( &m->tags[m->cur_tag].cc_buf );
}
// see if we already made a tag for the timestamp. If so we
// can just use it, otherwise make a new tag.
if (m->tags[m->cur_tag].start < 0)
{
next_tag( m, buf_es );
}
m->tags[m->cur_tag].cc_buf = m->last_cc1_buf;
}
}
if( state == STATE_SEQUENCE )
{
if( !( m->width && m->height && m->rate ) )
{
m->width = m->info->sequence->width;
m->height = m->info->sequence->height;
m->rate = m->info->sequence->frame_period;
if ( m->aspect_ratio <= 0 && m->height &&
m->info->sequence->pixel_height )
{
/* mpeg2_parse doesn't store the aspect ratio. Instead
* it keeps the pixel width & height that would cause
* the storage width & height to come out in the correct
* aspect ratio. Convert these back to aspect ratio.
*/
double ar_numer = m->width * m->info->sequence->pixel_width;
double ar_denom = m->height * m->info->sequence->pixel_height;
m->aspect_ratio = ar_numer / ar_denom;
}
}
if ( m->info->sequence->width >> 1 == m->info->sequence->chroma_width &&
m->info->sequence->height >> 1 == m->info->sequence->chroma_height )
{
m->pixfmt = PIX_FMT_YUV420P;
}
else
{
m->pixfmt = PIX_FMT_YUV422P;
}
}
else if( state == STATE_GOP && m->look_for_break)
{
// we were looking for a gop to add a chapter break - we found it
// so now start looking for an iframe.
m->look_for_iframe = m->look_for_break;
m->look_for_break = 0;
}
else if( ( state == STATE_SLICE || state == STATE_END ) &&
m->info->display_fbuf )
{
m->last_cc1_buf = NULL;
if( !m->got_iframe && ( m->info->display_picture->flags &
PIC_MASK_CODING_TYPE ) == PIC_FLAG_CODING_TYPE_I )
{
// we got an iframe so we can start decoding video now
m->got_iframe = 1;
}
if( m->got_iframe )
{
buf = hb_copy_frame( m->job, m->info->sequence->width,
m->info->sequence->height,
m->pixfmt,
m->info->display_fbuf->buf[0],
m->info->display_fbuf->buf[1],
m->info->display_fbuf->buf[2] );
buf->sequence = buf_es->sequence;
hb_buffer_t *cc_buf = NULL;
if( m->info->display_picture->flags & PIC_FLAG_TAGS )
{
int t = m->info->display_picture->tag;
buf->start = m->tags[t].start;
cc_buf = m->tags[t].cc_buf;
m->tags[t].start = -1;
m->tags[t].cc_buf = NULL;
}
if( buf->start < 0 && m->last_pts >= 0 )
{
/* For some reason nb_fields is sometimes 1 while it
should be 2 */
buf->start = m->last_pts +
MAX( 2, m->info->display_picture->nb_fields ) *
m->info->sequence->frame_period / 600;
}
if ( buf->start >= 0 )
{
m->last_pts = buf->start;
}
// if we were accumulating captions we now know the timestamp
// so ship them to the decoder.
if ( cc_buf )
{
cc_buf->start = m->last_pts;
cc_send_to_decoder( m, cc_buf );
}
if( m->look_for_iframe && ( m->info->display_picture->flags &
PIC_MASK_CODING_TYPE ) == PIC_FLAG_CODING_TYPE_I )
{
// we were waiting for an iframe to insert a chapter mark
// and we have one.
buf->new_chap = m->look_for_iframe;
m->look_for_iframe = 0;
const char *chap_name = "";
if ( m->job && buf->new_chap > 0 &&
hb_list_item( m->job->title->list_chapter,
buf->new_chap - 1 ) )
{
hb_chapter_t * c = hb_list_item( m->job->title->list_chapter,
buf->new_chap - 1 );
chap_name = c->title;
}
hb_log( "mpeg2: \"%s\" (%d) at frame %u time %"PRId64,
chap_name, buf->new_chap, m->nframes, buf->start );
}
else if ( m->nframes == 0 )
{
// this is the first frame returned by the decoder
m->first_pts = buf->start;
if ( m->job && hb_list_item( m->job->title->list_chapter,
m->job->chapter_start - 1 ) )
{
hb_chapter_t * c = hb_list_item( m->job->title->list_chapter,
m->job->chapter_start - 1 );
hb_log( "mpeg2: \"%s\" (%d) at frame %u time %"PRId64,
c->title, m->job->chapter_start, m->nframes, buf->start );
}
}
++m->nframes;
m->flag = m->info->display_picture->flags;
/* Uncomment this block to see frame-by-frame picture flags, as the video encodes.
hb_log("***** MPEG 2 Picture Info for PTS %lld *****", buf->start);
if( m->flag & TOP_FIRST )
hb_log("MPEG2 Flag: Top field first");
if( m->flag & PROGRESSIVE )
hb_log("MPEG2 Flag: Progressive");
if( m->flag & COMPOSITE )
hb_log("MPEG2 Flag: Composite");
if( m->flag & SKIP )
hb_log("MPEG2 Flag: Skip!");
if( m->flag & TAGS )
hb_log("MPEG2 Flag: TAGS");
if(fm->lag & REPEAT_FIRST )
hb_log("MPEG2 Flag: Repeat first field");
if( m->flag & COMPOSITE_MASK )
hb_log("MPEG2 Flag: Composite mask");
hb_log("fields: %d", m->info->display_picture->nb_fields);
*/
/* Rotate the cadence tracking. */
int i = 0;
for(i=11; i > 0; i--)
{
m->cadence[i] = m->cadence[i-1];
}
if ( !(m->flag & PROGRESSIVE) && !(m->flag & TOP_FIRST) )
{
/* Not progressive, not top first...
That means it's probably bottom
first, 2 fields displayed.
*/
//hb_log("MPEG2 Flag: Bottom field first, 2 fields displayed.");
m->cadence[0] = BT;
}
else if ( !(m->flag & PROGRESSIVE) && (m->flag & TOP_FIRST) )
{
/* Not progressive, top is first,
Two fields displayed.
*/
//hb_log("MPEG2 Flag: Top field first, 2 fields displayed.");
m->cadence[0] = TB;
}
else if ( (m->flag & PROGRESSIVE) && !(m->flag & TOP_FIRST) && !( m->flag & REPEAT_FIRST ) )
{
/* Progressive, but noting else.
That means Bottom first,
2 fields displayed.
*/
//hb_log("MPEG2 Flag: Progressive. Bottom field first, 2 fields displayed.");
m->cadence[0] = BT_PROG;
}
else if ( (m->flag & PROGRESSIVE) && !(m->flag & TOP_FIRST) && ( m->flag & REPEAT_FIRST ) )
{
/* Progressive, and repeat. .
That means Bottom first,
3 fields displayed.
*/
//hb_log("MPEG2 Flag: Progressive repeat. Bottom field first, 3 fields displayed.");
m->cadence[0] = BTB_PROG;
}
else if ( (m->flag & PROGRESSIVE) && (m->flag & TOP_FIRST) && !( m->flag & REPEAT_FIRST ) )
{
/* Progressive, top first.
That means top first,
2 fields displayed.
*/
//hb_log("MPEG2 Flag: Progressive. Top field first, 2 fields displayed.");
m->cadence[0] = TB_PROG;
}
else if ( (m->flag & PROGRESSIVE) && (m->flag & TOP_FIRST) && ( m->flag & REPEAT_FIRST ) )
{
/* Progressive, top, repeat.
That means top first,
3 fields displayed.
*/
//hb_log("MPEG2 Flag: Progressive repeat. Top field first, 3 fields displayed.");
m->cadence[0] = TBT_PROG;
}
if ( (m->cadence[2] <= TB) && (m->cadence[1] <= TB) && (m->cadence[0] > TB) && (m->cadence[11]) )
hb_log("%fs: Video -> Film", (float)buf->start / 90000);
if ( (m->cadence[2] > TB) && (m->cadence[1] <= TB) && (m->cadence[0] <= TB) && (m->cadence[11]) )
hb_log("%fs: Film -> Video", (float)buf->start / 90000);
/* Store picture flags for later use by filters */
buf->flags = m->info->display_picture->flags;
hb_list_add( list_raw, buf );
}
}
else if( state == STATE_INVALID )
{
mpeg2_reset( m->libmpeg2, 0 );
}
/*
* Add closed captions to the title if we are scanning (no job).
*
* Just because we don't add this doesn't mean that there aren't any when
* we scan, just that noone said anything. So you should be able to add
* closed captions some other way (See decmpeg2Init() for alternative
* approach of assuming that there are always CC, which is probably
* safer - however I want to leave the autodetect in here for now to
* see how it goes).
*/
if( !m->job && m->title &&
have_captions( m->info->user_data, m->info->user_data_len ) )
{
hb_subtitle_t *subtitle;
int i = 0;
while ( ( subtitle = hb_list_item( m->title->list_subtitle, i++ ) ) )
{
/*
* Let's call them 608 subs for now even if they aren't,
* since they are the only types we grok.
*/
if( subtitle->source == CC608SUB )
{
break;
}
}
if( ! subtitle )
{
subtitle = calloc( sizeof( hb_subtitle_t ), 1 );
subtitle->track = 0;
subtitle->id = 0;
subtitle->format = TEXTSUB;
subtitle->source = CC608SUB;
subtitle->config.dest = PASSTHRUSUB;
subtitle->type = 5;
snprintf( subtitle->lang, sizeof( subtitle->lang ), "Closed Captions");
/*
* The language of the subtitles will be the same as the first audio
* track, i.e. the same as the video.
*/
hb_audio_t *audio = hb_list_item( m->title->list_audio, 0 );
if( audio )
{
snprintf( subtitle->iso639_2, sizeof( subtitle->iso639_2 ),
audio->config.lang.iso639_2);
} else {
snprintf( subtitle->iso639_2, sizeof( subtitle->iso639_2 ), "und");
}
hb_list_add( m->title->list_subtitle, subtitle );
}
}
}
return 1;
}
/**********************************************************************
* hb_libmpeg2_close
**********************************************************************
*
*********************************************************************/
static void hb_libmpeg2_close( hb_libmpeg2_t ** _m )
{
hb_libmpeg2_t * m = *_m;
mpeg2_close( m->libmpeg2 );
int i;
for ( i = 0; i < NTAGS; ++i )
{
if ( m->tags[i].cc_buf )
hb_buffer_close( &m->tags[i].cc_buf );
}
free( m );
*_m = NULL;
}
/**********************************************************************
* The decmpeg2 work object
**********************************************************************
*
*********************************************************************/
struct hb_work_private_s
{
hb_libmpeg2_t * libmpeg2;
hb_list_t * list;
};
/**********************************************************************
* hb_work_decmpeg2_init
**********************************************************************
*
*********************************************************************/
static int decmpeg2Init( hb_work_object_t * w, hb_job_t * job )
{
hb_work_private_t * pv;
pv = calloc( 1, sizeof( hb_work_private_t ) );
w->private_data = pv;
pv->libmpeg2 = hb_libmpeg2_init();
pv->list = hb_list_init();
pv->libmpeg2->job = job;
if( job && job->title ) {
pv->libmpeg2->title = job->title;
}
/*
* If not scanning, then are we supposed to extract Closed Captions
* and send them to the decoder?
*/
if( job && hb_list_count( job->list_subtitle ) > 0 )
{
hb_subtitle_t *subtitle;
int i = 0;
for( i = 0; i < hb_list_count( job->list_subtitle ); i++ )
while ( ( subtitle = hb_list_item( job->list_subtitle, i++ ) ) != NULL )
{
if( subtitle->source == CC608SUB )
{
if ( ! pv->libmpeg2->list_subtitle )
{
pv->libmpeg2->list_subtitle = hb_list_init();
}
hb_list_add(pv->libmpeg2->list_subtitle, subtitle);
}
}
}
return 0;
}
/**********************************************************************
* Work
**********************************************************************
*
*********************************************************************/
static int decmpeg2Work( hb_work_object_t * w, hb_buffer_t ** buf_in,
hb_buffer_t ** buf_out )
{
hb_work_private_t * pv = w->private_data;
hb_buffer_t * buf, * last = NULL;
int status = HB_WORK_OK;
if( w->title && pv && pv->libmpeg2 && !pv->libmpeg2->title ) {
pv->libmpeg2->title = w->title;
}
// The reader found a chapter break. Remove it from the input
// stream. If we're reading (as opposed to scanning) start looking
// for the next GOP start since that's where the chapter begins.
if( (*buf_in)->new_chap )
{
if ( pv->libmpeg2->job )
{
pv->libmpeg2->look_for_break = (*buf_in)->new_chap;
}
(*buf_in)->new_chap = 0;
}
hb_libmpeg2_decode( pv->libmpeg2, *buf_in, pv->list );
/* if we got an empty buffer signaling end-of-stream send it downstream */
if ( (*buf_in)->size == 0 )
{
hb_list_add( pv->list, *buf_in );
*buf_in = NULL;
status = HB_WORK_DONE;
/*
* Purge any pending caption buffer then let the Closed Captions decoder
* know that it is the end of the data.
*/
if ( pv->libmpeg2->list_subtitle )
{
if ( pv->libmpeg2->last_cc1_buf )
{
cc_send_to_decoder( pv->libmpeg2, pv->libmpeg2->last_cc1_buf );
pv->libmpeg2->last_cc1_buf = NULL;
}
cc_send_to_decoder( pv->libmpeg2, hb_buffer_init( 0 ) );
}
}
*buf_out = NULL;
while( ( buf = hb_list_item( pv->list, 0 ) ) )
{
hb_list_rem( pv->list, buf );
if( last )
{
last->next = buf;
last = buf;
}
else
{
*buf_out = buf;
last = buf;
}
}
return status;
}
/**********************************************************************
* Close
**********************************************************************
*
*********************************************************************/
static void decmpeg2Close( hb_work_object_t * w )
{
hb_work_private_t * pv = w->private_data;
// don't log during scan
if ( pv->libmpeg2->job )
{
hb_log( "mpeg2 done: %d frames", pv->libmpeg2->nframes );
}
if ( pv->libmpeg2->last_cc1_buf )
{
hb_buffer_close( &pv->libmpeg2->last_cc1_buf );
}
hb_list_close( &pv->list );
if ( pv->libmpeg2->list_subtitle )
{
hb_list_close( &pv->libmpeg2->list_subtitle );
}
hb_libmpeg2_close( &pv->libmpeg2 );
free( pv );
}
static int decmpeg2Info( hb_work_object_t *w, hb_work_info_t *info )
{
hb_work_private_t *pv = w->private_data;
memset( info, 0, sizeof(*info) );
if ( pv && pv->libmpeg2 && pv->libmpeg2->info && pv->libmpeg2->info->sequence )
{
hb_libmpeg2_t *m = pv->libmpeg2;
info->width = m->width;
info->height = m->height;
info->pixel_aspect_width = m->info->sequence->pixel_width;
info->pixel_aspect_height = m->info->sequence->pixel_height;
info->aspect = m->aspect_ratio;
// if the frame is progressive & NTSC DVD height report it as 23.976 FPS
// so that scan can autodetect NTSC film
info->rate = 27000000;
info->rate_base = ( m->info->display_fbuf && m->info->display_picture &&
(m->info->display_picture->flags & PROGRESSIVE) &&
(m->height == 480 ) ) ? 1126125 : m->rate;
info->bitrate = m->info->sequence->byte_rate * 8;
info->profile = m->info->sequence->profile_level_id >> 4;
info->level = m->info->sequence->profile_level_id & 0xf;
info->name = "mpeg2";
return 1;
}
return 0;
}
hb_work_object_t hb_decmpeg2 =
{
WORK_DECMPEG2,
"MPEG-2 decoder (libmpeg2)",
decmpeg2Init,
decmpeg2Work,
decmpeg2Close,
decmpeg2Info
};
|