aboutsummaryrefslogtreecommitdiffstats
path: root/libavcodec/proresdec_lgpl.c
blob: 4bdf39246129d92c00bf1348ef9e7e6db5a25aa6 (plain) (blame)
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
/*
 * Apple ProRes compatible decoder
 *
 * Copyright (c) 2010-2011 Maxim Poliakovski
 *
 * This file is part of FFmpeg.
 *
 * FFmpeg 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.
 *
 * FFmpeg 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 FFmpeg; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

/**
 * @file
 * This is a decoder for Apple ProRes 422 SD/HQ/LT/Proxy and ProRes 4444.
 * It is used for storing and editing high definition video data in Apple's Final Cut Pro.
 *
 * @see http://wiki.multimedia.cx/index.php?title=Apple_ProRes
 */

#define LONG_BITSTREAM_READER // some ProRes vlc codes require up to 28 bits to be read at once

#include <stdint.h>

#include "libavutil/intmath.h"
#include "avcodec.h"
#include "idctdsp.h"
#include "internal.h"
#include "proresdata.h"
#include "proresdsp.h"
#include "get_bits.h"

typedef struct ProresThreadData {
    const uint8_t *index;            ///< pointers to the data of this slice
    int slice_num;
    int x_pos, y_pos;
    int slice_width;
    int prev_slice_sf;               ///< scalefactor of the previous decoded slice
    DECLARE_ALIGNED(16, int16_t, blocks)[8 * 4 * 64];
    DECLARE_ALIGNED(16, int16_t, qmat_luma_scaled)[64];
    DECLARE_ALIGNED(16, int16_t, qmat_chroma_scaled)[64];
} ProresThreadData;

typedef struct ProresContext {
    ProresDSPContext dsp;
    AVFrame    *frame;
    ScanTable  scantable;
    int        scantable_type;           ///< -1 = uninitialized, 0 = progressive, 1/2 = interlaced

    int        frame_type;               ///< 0 = progressive, 1 = top-field first, 2 = bottom-field first
    int        pic_format;               ///< 2 = 422, 3 = 444
    uint8_t    qmat_luma[64];            ///< dequantization matrix for luma
    uint8_t    qmat_chroma[64];          ///< dequantization matrix for chroma
    int        qmat_changed;             ///< 1 - global quantization matrices changed
    int        total_slices;            ///< total number of slices in a picture
    ProresThreadData *slice_data;
    int        pic_num;
    int        chroma_factor;
    int        mb_chroma_factor;
    int        num_chroma_blocks;       ///< number of chrominance blocks in a macroblock
    int        num_x_slices;
    int        num_y_slices;
    int        slice_width_factor;
    int        slice_height_factor;
    int        num_x_mbs;
    int        num_y_mbs;
    int        alpha_info;
} ProresContext;


static av_cold int decode_init(AVCodecContext *avctx)
{
    ProresContext *ctx = avctx->priv_data;

    ctx->total_slices     = 0;
    ctx->slice_data       = NULL;

    avctx->bits_per_raw_sample = PRORES_BITS_PER_SAMPLE;
    ff_proresdsp_init(&ctx->dsp, avctx);

    ctx->scantable_type = -1;   // set scantable type to uninitialized
    memset(ctx->qmat_luma, 4, 64);
    memset(ctx->qmat_chroma, 4, 64);

    return 0;
}


static int decode_frame_header(ProresContext *ctx, const uint8_t *buf,
                               const int data_size, AVCodecContext *avctx)
{
    int hdr_size, version, width, height, flags;
    const uint8_t *ptr;

    hdr_size = AV_RB16(buf);
    if (hdr_size > data_size) {
        av_log(avctx, AV_LOG_ERROR, "frame data too small\n");
        return AVERROR_INVALIDDATA;
    }

    version = AV_RB16(buf + 2);
    if (version >= 2) {
        av_log(avctx, AV_LOG_ERROR,
               "unsupported header version: %d\n", version);
        return AVERROR_INVALIDDATA;
    }

    width  = AV_RB16(buf + 8);
    height = AV_RB16(buf + 10);
    if (width != avctx->width || height != avctx->height) {
        av_log(avctx, AV_LOG_ERROR,
               "picture dimension changed: old: %d x %d, new: %d x %d\n",
               avctx->width, avctx->height, width, height);
        return AVERROR_INVALIDDATA;
    }

    ctx->frame_type = (buf[12] >> 2) & 3;
    if (ctx->frame_type > 2) {
        av_log(avctx, AV_LOG_ERROR,
               "unsupported frame type: %d\n", ctx->frame_type);
        return AVERROR_INVALIDDATA;
    }

    ctx->chroma_factor     = (buf[12] >> 6) & 3;
    ctx->mb_chroma_factor  = ctx->chroma_factor + 2;
    ctx->num_chroma_blocks = (1 << ctx->chroma_factor) >> 1;
    ctx->alpha_info        = buf[17] & 0xf;

    if (ctx->alpha_info > 2) {
        av_log(avctx, AV_LOG_ERROR, "Invalid alpha mode %d\n", ctx->alpha_info);
        return AVERROR_INVALIDDATA;
    }
    if (avctx->skip_alpha) ctx->alpha_info = 0;

    switch (ctx->chroma_factor) {
    case 2:
        avctx->pix_fmt = ctx->alpha_info ? AV_PIX_FMT_YUVA422P10
                                         : AV_PIX_FMT_YUV422P10;
        break;
    case 3:
        avctx->pix_fmt = ctx->alpha_info ? AV_PIX_FMT_YUVA444P10
                                         : AV_PIX_FMT_YUV444P10;
        break;
    default:
        av_log(avctx, AV_LOG_ERROR,
               "unsupported picture format: %d\n", ctx->pic_format);
        return AVERROR_INVALIDDATA;
    }

    if (ctx->scantable_type != ctx->frame_type) {
        if (!ctx->frame_type)
            ff_init_scantable(ctx->dsp.idct_permutation, &ctx->scantable,
                              ff_prores_progressive_scan);
        else
            ff_init_scantable(ctx->dsp.idct_permutation, &ctx->scantable,
                              ff_prores_interlaced_scan);
        ctx->scantable_type = ctx->frame_type;
    }

    if (ctx->frame_type) {      /* if interlaced */
        ctx->frame->interlaced_frame = 1;
        ctx->frame->top_field_first  = ctx->frame_type & 1;
    } else {
        ctx->frame->interlaced_frame = 0;
    }

    avctx->color_primaries = buf[14];
    avctx->color_trc       = buf[15];
    avctx->colorspace      = buf[16];

    ctx->qmat_changed = 0;
    ptr   = buf + 20;
    flags = buf[19];
    if (flags & 2) {
        if (ptr - buf > hdr_size - 64) {
            av_log(avctx, AV_LOG_ERROR, "header data too small\n");
            return AVERROR_INVALIDDATA;
        }
        if (memcmp(ctx->qmat_luma, ptr, 64)) {
            memcpy(ctx->qmat_luma, ptr, 64);
            ctx->qmat_changed = 1;
        }
        ptr += 64;
    } else {
        memset(ctx->qmat_luma, 4, 64);
        ctx->qmat_changed = 1;
    }

    if (flags & 1) {
        if (ptr - buf > hdr_size - 64) {
            av_log(avctx, AV_LOG_ERROR, "header data too small\n");
            return -1;
        }
        if (memcmp(ctx->qmat_chroma, ptr, 64)) {
            memcpy(ctx->qmat_chroma, ptr, 64);
            ctx->qmat_changed = 1;
        }
    } else {
        memset(ctx->qmat_chroma, 4, 64);
        ctx->qmat_changed = 1;
    }

    return hdr_size;
}


static int decode_picture_header(ProresContext *ctx, const uint8_t *buf,
                                 const int data_size, AVCodecContext *avctx)
{
    int   i, hdr_size, pic_data_size, num_slices;
    int   slice_width_factor, slice_height_factor;
    int   remainder, num_x_slices;
    const uint8_t *data_ptr, *index_ptr;

    hdr_size = data_size > 0 ? buf[0] >> 3 : 0;
    if (hdr_size < 8 || hdr_size > data_size) {
        av_log(avctx, AV_LOG_ERROR, "picture header too small\n");
        return AVERROR_INVALIDDATA;
    }

    pic_data_size = AV_RB32(buf + 1);
    if (pic_data_size > data_size) {
        av_log(avctx, AV_LOG_ERROR, "picture data too small\n");
        return AVERROR_INVALIDDATA;
    }

    slice_width_factor  = buf[7] >> 4;
    slice_height_factor = buf[7] & 0xF;
    if (slice_width_factor > 3 || slice_height_factor) {
        av_log(avctx, AV_LOG_ERROR,
               "unsupported slice dimension: %d x %d\n",
               1 << slice_width_factor, 1 << slice_height_factor);
        return AVERROR_INVALIDDATA;
    }

    ctx->slice_width_factor  = slice_width_factor;
    ctx->slice_height_factor = slice_height_factor;

    ctx->num_x_mbs = (avctx->width + 15) >> 4;
    ctx->num_y_mbs = (avctx->height +
                      (1 << (4 + ctx->frame->interlaced_frame)) - 1) >>
                     (4 + ctx->frame->interlaced_frame);

    remainder    = ctx->num_x_mbs & ((1 << slice_width_factor) - 1);
    num_x_slices = (ctx->num_x_mbs >> slice_width_factor) + (remainder & 1) +
                   ((remainder >> 1) & 1) + ((remainder >> 2) & 1);

    num_slices = num_x_slices * ctx->num_y_mbs;
    if (num_slices != AV_RB16(buf + 5)) {
        av_log(avctx, AV_LOG_ERROR, "invalid number of slices\n");
        return AVERROR_INVALIDDATA;
    }

    if (ctx->total_slices != num_slices) {
        av_freep(&ctx->slice_data);
        ctx->slice_data = av_malloc_array(num_slices + 1, sizeof(ctx->slice_data[0]));
        if (!ctx->slice_data)
            return AVERROR(ENOMEM);
        ctx->total_slices = num_slices;
    }

    if (hdr_size + num_slices * 2 > data_size) {
        av_log(avctx, AV_LOG_ERROR, "slice table too small\n");
        return AVERROR_INVALIDDATA;
    }

    /* parse slice table allowing quick access to the slice data */
    index_ptr = buf + hdr_size;
    data_ptr = index_ptr + num_slices * 2;

    for (i = 0; i < num_slices; i++) {
        ctx->slice_data[i].index = data_ptr;
        ctx->slice_data[i].prev_slice_sf = 0;
        data_ptr += AV_RB16(index_ptr + i * 2);
    }
    ctx->slice_data[i].index = data_ptr;
    ctx->slice_data[i].prev_slice_sf = 0;

    if (data_ptr > buf + data_size) {
        av_log(avctx, AV_LOG_ERROR, "out of slice data\n");
        return -1;
    }

    return pic_data_size;
}


/**
 * Read an unsigned rice/exp golomb codeword.
 */
static inline int decode_vlc_codeword(GetBitContext *gb, unsigned codebook)
{
    unsigned int rice_order, exp_order, switch_bits;
    unsigned int buf, code;
    int log, prefix_len, len;

    OPEN_READER(re, gb);
    UPDATE_CACHE(re, gb);
    buf = GET_CACHE(re, gb);

    /* number of prefix bits to switch between Rice and expGolomb */
    switch_bits = (codebook & 3) + 1;
    rice_order  = codebook >> 5;        /* rice code order */
    exp_order   = (codebook >> 2) & 7;  /* exp golomb code order */

    log = 31 - av_log2(buf); /* count prefix bits (zeroes) */

    if (log < switch_bits) { /* ok, we got a rice code */
        if (!rice_order) {
            /* shortcut for faster decoding of rice codes without remainder */
            code = log;
            LAST_SKIP_BITS(re, gb, log + 1);
        } else {
            prefix_len = log + 1;
            code = (log << rice_order) + NEG_USR32(buf << prefix_len, rice_order);
            LAST_SKIP_BITS(re, gb, prefix_len + rice_order);
        }
    } else { /* otherwise we got a exp golomb code */
        len  = (log << 1) - switch_bits + exp_order + 1;
        code = NEG_USR32(buf, len) - (1 << exp_order) + (switch_bits << rice_order);
        LAST_SKIP_BITS(re, gb, len);
    }

    CLOSE_READER(re, gb);

    return code;
}

#define LSB2SIGN(x) (-((x) & 1))
#define TOSIGNED(x) (((x) >> 1) ^ LSB2SIGN(x))

/**
 * Decode DC coefficients for all blocks in a slice.
 */
static inline void decode_dc_coeffs(GetBitContext *gb, int16_t *out,
                                    int nblocks)
{
    int16_t prev_dc;
    int     i, sign;
    int16_t delta;
    unsigned int code;

    code   = decode_vlc_codeword(gb, FIRST_DC_CB);
    out[0] = prev_dc = TOSIGNED(code);

    out   += 64; /* move to the DC coeff of the next block */
    delta  = 3;

    for (i = 1; i < nblocks; i++, out += 64) {
        code = decode_vlc_codeword(gb, ff_prores_dc_codebook[FFMIN(FFABS(delta), 3)]);

        sign     = -(((delta >> 15) & 1) ^ (code & 1));
        delta    = (((code + 1) >> 1) ^ sign) - sign;
        prev_dc += delta;
        out[0]   = prev_dc;
    }
}

#define MAX_PADDING 16

/**
 * Decode AC coefficients for all blocks in a slice.
 */
static inline int decode_ac_coeffs(GetBitContext *gb, int16_t *out,
                                   int blocks_per_slice,
                                   int plane_size_factor,
                                   const uint8_t *scan)
{
    int pos, block_mask, run, level, sign, run_cb_index, lev_cb_index;
    int max_coeffs, bits_left;

    /* set initial prediction values */
    run   = 4;
    level = 2;

    max_coeffs = blocks_per_slice << 6;
    block_mask = blocks_per_slice - 1;

    for (pos = blocks_per_slice - 1; pos < max_coeffs;) {
        run_cb_index = ff_prores_run_to_cb_index[FFMIN(run, 15)];
        lev_cb_index = ff_prores_lev_to_cb_index[FFMIN(level, 9)];

        bits_left = get_bits_left(gb);
        if (bits_left <= 0 || (bits_left <= MAX_PADDING && !show_bits(gb, bits_left)))
            return 0;

        run = decode_vlc_codeword(gb, ff_prores_ac_codebook[run_cb_index]);
        if (run < 0)
            return AVERROR_INVALIDDATA;

        bits_left = get_bits_left(gb);
        if (bits_left <= 0 || (bits_left <= MAX_PADDING && !show_bits(gb, bits_left)))
            return AVERROR_INVALIDDATA;

        level = decode_vlc_codeword(gb, ff_prores_ac_codebook[lev_cb_index]) + 1;
        if (level < 0)
            return AVERROR_INVALIDDATA;

        pos += run + 1;
        if (pos >= max_coeffs)
            break;

        sign = get_sbits(gb, 1);
        out[((pos & block_mask) << 6) + scan[pos >> plane_size_factor]] =
            (level ^ sign) - sign;
    }

    return 0;
}


/**
 * Decode a slice plane (luma or chroma).
 */
static int decode_slice_plane(ProresContext *ctx, ProresThreadData *td,
                              const uint8_t *buf,
                              int data_size, uint16_t *out_ptr,
                              int linesize, int mbs_per_slice,
                              int blocks_per_mb, int plane_size_factor,
                              const int16_t *qmat, int is_chroma)
{
    GetBitContext gb;
    int16_t *block_ptr;
    int mb_num, blocks_per_slice, ret;

    blocks_per_slice = mbs_per_slice * blocks_per_mb;

    memset(td->blocks, 0, 8 * 4 * 64 * sizeof(*td->blocks));

    init_get_bits(&gb, buf, data_size << 3);

    decode_dc_coeffs(&gb, td->blocks, blocks_per_slice);

    ret = decode_ac_coeffs(&gb, td->blocks, blocks_per_slice,
                           plane_size_factor, ctx->scantable.permutated);
    if (ret < 0)
        return ret;

    /* inverse quantization, inverse transform and output */
    block_ptr = td->blocks;

    if (!is_chroma) {
        for (mb_num = 0; mb_num < mbs_per_slice; mb_num++, out_ptr += blocks_per_mb * 4) {
            ctx->dsp.idct_put(out_ptr,                    linesize, block_ptr, qmat);
            block_ptr += 64;
            if (blocks_per_mb > 2) {
                ctx->dsp.idct_put(out_ptr + 8,            linesize, block_ptr, qmat);
                block_ptr += 64;
            }
            ctx->dsp.idct_put(out_ptr + linesize * 4,     linesize, block_ptr, qmat);
            block_ptr += 64;
            if (blocks_per_mb > 2) {
                ctx->dsp.idct_put(out_ptr + linesize * 4 + 8, linesize, block_ptr, qmat);
                block_ptr += 64;
            }
        }
    } else {
        for (mb_num = 0; mb_num < mbs_per_slice; mb_num++, out_ptr += blocks_per_mb * 4) {
            ctx->dsp.idct_put(out_ptr,                    linesize, block_ptr, qmat);
            block_ptr += 64;
            ctx->dsp.idct_put(out_ptr + linesize * 4,     linesize, block_ptr, qmat);
            block_ptr += 64;
            if (blocks_per_mb > 2) {
                ctx->dsp.idct_put(out_ptr + 8,            linesize, block_ptr, qmat);
                block_ptr += 64;
                ctx->dsp.idct_put(out_ptr + linesize * 4 + 8, linesize, block_ptr, qmat);
                block_ptr += 64;
            }
        }
    }
    return 0;
}


static void unpack_alpha(GetBitContext *gb, uint16_t *dst, int num_coeffs,
                         const int num_bits)
{
    const int mask = (1 << num_bits) - 1;
    int i, idx, val, alpha_val;

    idx       = 0;
    alpha_val = mask;
    do {
        do {
            if (get_bits1(gb))
                val = get_bits(gb, num_bits);
            else {
                int sign;
                val  = get_bits(gb, num_bits == 16 ? 7 : 4);
                sign = val & 1;
                val  = (val + 2) >> 1;
                if (sign)
                    val = -val;
            }
            alpha_val = (alpha_val + val) & mask;
            if (num_bits == 16)
                dst[idx++] = alpha_val >> 6;
            else
                dst[idx++] = (alpha_val << 2) | (alpha_val >> 6);
            if (idx >= num_coeffs) {
                break;
            }
        } while (get_bits1(gb));
        val = get_bits(gb, 4);
        if (!val)
            val = get_bits(gb, 11);
        if (idx + val > num_coeffs)
            val = num_coeffs - idx;
        if (num_bits == 16)
            for (i = 0; i < val; i++)
                dst[idx++] = alpha_val >> 6;
        else
            for (i = 0; i < val; i++)
                dst[idx++] = (alpha_val << 2) | (alpha_val >> 6);
    } while (idx < num_coeffs);
}

/**
 * Decode alpha slice plane.
 */
static void decode_alpha_plane(ProresContext *ctx, ProresThreadData *td,
                               const uint8_t *buf, int data_size,
                               uint16_t *out_ptr, int linesize,
                               int mbs_per_slice)
{
    GetBitContext gb;
    int i;
    uint16_t *block_ptr;

    memset(td->blocks, 0, 8 * 4 * 64 * sizeof(*td->blocks));

    init_get_bits(&gb, buf, data_size << 3);

    if (ctx->alpha_info == 2)
        unpack_alpha(&gb, td->blocks, mbs_per_slice * 4 * 64, 16);
    else
        unpack_alpha(&gb, td->blocks, mbs_per_slice * 4 * 64, 8);

    block_ptr = td->blocks;

    for (i = 0; i < 16; i++) {
        memcpy(out_ptr, block_ptr, 16 * mbs_per_slice * sizeof(*out_ptr));
        out_ptr   += linesize >> 1;
        block_ptr += 16 * mbs_per_slice;
    }
}

static int decode_slice(AVCodecContext *avctx, void *tdata)
{
    ProresThreadData *td = tdata;
    ProresContext *ctx = avctx->priv_data;
    int mb_x_pos  = td->x_pos;
    int mb_y_pos  = td->y_pos;
    int pic_num   = ctx->pic_num;
    int slice_num = td->slice_num;
    int mbs_per_slice = td->slice_width;
    const uint8_t *buf;
    uint8_t *y_data, *u_data, *v_data, *a_data;
    AVFrame *pic = ctx->frame;
    int i, sf, slice_width_factor;
    int slice_data_size, hdr_size;
    int y_data_size, u_data_size, v_data_size, a_data_size;
    int y_linesize, u_linesize, v_linesize, a_linesize;
    int coff[4];
    int ret;

    buf             = ctx->slice_data[slice_num].index;
    slice_data_size = ctx->slice_data[slice_num + 1].index - buf;

    slice_width_factor = av_log2(mbs_per_slice);

    y_data     = pic->data[0];
    u_data     = pic->data[1];
    v_data     = pic->data[2];
    a_data     = pic->data[3];
    y_linesize = pic->linesize[0];
    u_linesize = pic->linesize[1];
    v_linesize = pic->linesize[2];
    a_linesize = pic->linesize[3];

    if (pic->interlaced_frame) {
        if (!(pic_num ^ pic->top_field_first)) {
            y_data += y_linesize;
            u_data += u_linesize;
            v_data += v_linesize;
            if (a_data)
                a_data += a_linesize;
        }
        y_linesize <<= 1;
        u_linesize <<= 1;
        v_linesize <<= 1;
        a_linesize <<= 1;
    }
    y_data += (mb_y_pos << 4) * y_linesize + (mb_x_pos << 5);
    u_data += (mb_y_pos << 4) * u_linesize + (mb_x_pos << ctx->mb_chroma_factor);
    v_data += (mb_y_pos << 4) * v_linesize + (mb_x_pos << ctx->mb_chroma_factor);
    if (a_data)
        a_data += (mb_y_pos << 4) * a_linesize + (mb_x_pos << 5);

    if (slice_data_size < 6) {
        av_log(avctx, AV_LOG_ERROR, "slice data too small\n");
        return AVERROR_INVALIDDATA;
    }

    /* parse slice header */
    hdr_size    = buf[0] >> 3;
    coff[0]     = hdr_size;
    y_data_size = AV_RB16(buf + 2);
    coff[1]     = coff[0] + y_data_size;
    u_data_size = AV_RB16(buf + 4);
    coff[2]     = coff[1] + u_data_size;
    v_data_size = hdr_size > 7 ? AV_RB16(buf + 6) : slice_data_size - coff[2];
    coff[3]     = coff[2] + v_data_size;
    a_data_size = ctx->alpha_info ? slice_data_size - coff[3] : 0;

    /* if V or alpha component size is negative that means that previous
       component sizes are too large */
    if (v_data_size < 0 || a_data_size < 0 || hdr_size < 6) {
        av_log(avctx, AV_LOG_ERROR, "invalid data size\n");
        return AVERROR_INVALIDDATA;
    }

    sf = av_clip(buf[1], 1, 224);
    sf = sf > 128 ? (sf - 96) << 2 : sf;

    /* scale quantization matrixes according with slice's scale factor */
    /* TODO: this can be SIMD-optimized a lot */
    if (ctx->qmat_changed || sf != td->prev_slice_sf) {
        td->prev_slice_sf = sf;
        for (i = 0; i < 64; i++) {
            td->qmat_luma_scaled[ctx->dsp.idct_permutation[i]]   = ctx->qmat_luma[i]   * sf;
            td->qmat_chroma_scaled[ctx->dsp.idct_permutation[i]] = ctx->qmat_chroma[i] * sf;
        }
    }

    /* decode luma plane */
    ret = decode_slice_plane(ctx, td, buf + coff[0], y_data_size,
                             (uint16_t*) y_data, y_linesize,
                             mbs_per_slice, 4, slice_width_factor + 2,
                             td->qmat_luma_scaled, 0);

    if (ret < 0)
        return ret;

    /* decode U chroma plane */
    ret = decode_slice_plane(ctx, td, buf + coff[1], u_data_size,
                             (uint16_t*) u_data, u_linesize,
                             mbs_per_slice, ctx->num_chroma_blocks,
                             slice_width_factor + ctx->chroma_factor - 1,
                             td->qmat_chroma_scaled, 1);
    if (ret < 0)
        return ret;

    /* decode V chroma plane */
    ret = decode_slice_plane(ctx, td, buf + coff[2], v_data_size,
                             (uint16_t*) v_data, v_linesize,
                             mbs_per_slice, ctx->num_chroma_blocks,
                             slice_width_factor + ctx->chroma_factor - 1,
                             td->qmat_chroma_scaled, 1);
    if (ret < 0)
        return ret;

    /* decode alpha plane if available */
    if (a_data && a_data_size)
        decode_alpha_plane(ctx, td, buf + coff[3], a_data_size,
                           (uint16_t*) a_data, a_linesize,
                           mbs_per_slice);

    return 0;
}


static int decode_picture(ProresContext *ctx, int pic_num,
                          AVCodecContext *avctx)
{
    int slice_num, slice_width, x_pos, y_pos;

    slice_num = 0;

    ctx->pic_num = pic_num;
    for (y_pos = 0; y_pos < ctx->num_y_mbs; y_pos++) {
        slice_width = 1 << ctx->slice_width_factor;

        for (x_pos = 0; x_pos < ctx->num_x_mbs && slice_width;
             x_pos += slice_width) {
            while (ctx->num_x_mbs - x_pos < slice_width)
                slice_width >>= 1;

            ctx->slice_data[slice_num].slice_num   = slice_num;
            ctx->slice_data[slice_num].x_pos       = x_pos;
            ctx->slice_data[slice_num].y_pos       = y_pos;
            ctx->slice_data[slice_num].slice_width = slice_width;

            slice_num++;
        }
    }

    return avctx->execute(avctx, decode_slice,
                          ctx->slice_data, NULL, slice_num,
                          sizeof(ctx->slice_data[0]));
}


#define MOVE_DATA_PTR(nbytes) buf += (nbytes); buf_size -= (nbytes)

static int decode_frame(AVCodecContext *avctx, void *data, int *got_frame,
                        AVPacket *avpkt)
{
    ProresContext *ctx = avctx->priv_data;
    const uint8_t *buf = avpkt->data;
    int buf_size       = avpkt->size;
    int frame_hdr_size, pic_num, pic_data_size;

    ctx->frame            = data;
    ctx->frame->pict_type = AV_PICTURE_TYPE_I;
    ctx->frame->key_frame = 1;

    /* check frame atom container */
    if (buf_size < 28 || buf_size < AV_RB32(buf) ||
        AV_RB32(buf + 4) != FRAME_ID) {
        av_log(avctx, AV_LOG_ERROR, "invalid frame\n");
        return AVERROR_INVALIDDATA;
    }

    MOVE_DATA_PTR(8);

    frame_hdr_size = decode_frame_header(ctx, buf, buf_size, avctx);
    if (frame_hdr_size < 0)
        return AVERROR_INVALIDDATA;

    MOVE_DATA_PTR(frame_hdr_size);

    if (ff_get_buffer(avctx, ctx->frame, 0) < 0)
        return -1;

    for (pic_num = 0; ctx->frame->interlaced_frame - pic_num + 1; pic_num++) {
        pic_data_size = decode_picture_header(ctx, buf, buf_size, avctx);
        if (pic_data_size < 0)
            return AVERROR_INVALIDDATA;

        if (decode_picture(ctx, pic_num, avctx))
            return -1;

        MOVE_DATA_PTR(pic_data_size);
    }

    ctx->frame = NULL;
    *got_frame = 1;

    return avpkt->size;
}


static av_cold int decode_close(AVCodecContext *avctx)
{
    ProresContext *ctx = avctx->priv_data;

    av_freep(&ctx->slice_data);

    return 0;
}


AVCodec ff_prores_lgpl_decoder = {
    .name           = "prores_lgpl",
    .long_name      = NULL_IF_CONFIG_SMALL("Apple ProRes (iCodec Pro)"),
    .type           = AVMEDIA_TYPE_VIDEO,
    .id             = AV_CODEC_ID_PRORES,
    .priv_data_size = sizeof(ProresContext),
    .init           = decode_init,
    .close          = decode_close,
    .decode         = decode_frame,
    .capabilities   = CODEC_CAP_DR1 | CODEC_CAP_SLICE_THREADS,
};