aboutsummaryrefslogtreecommitdiffstats
path: root/libavcodec/tta.c
blob: 8d49bc80d2cf4278adceea9b84e804503fe52e53 (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
/*
 * TTA (The Lossless True Audio) decoder
 * Copyright (c) 2006 Alex Beregszaszi
 *
 * 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
 * TTA (The Lossless True Audio) decoder
 * @see http://www.true-audio.com/
 * @see http://tta.corecodec.org/
 * @author Alex Beregszaszi
 */

#define BITSTREAM_READER_LE
//#define DEBUG
#include <limits.h>
#include "avcodec.h"
#include "get_bits.h"

#define FORMAT_SIMPLE    1
#define FORMAT_ENCRYPTED 2

#define MAX_ORDER 16
typedef struct TTAFilter {
    int32_t shift, round, error, mode;
    int32_t qm[MAX_ORDER];
    int32_t dx[MAX_ORDER];
    int32_t dl[MAX_ORDER];
} TTAFilter;

typedef struct TTARice {
    uint32_t k0, k1, sum0, sum1;
} TTARice;

typedef struct TTAChannel {
    int32_t predictor;
    TTAFilter filter;
    TTARice rice;
} TTAChannel;

typedef struct TTAContext {
    AVCodecContext *avctx;
    AVFrame frame;
    GetBitContext gb;

    int format, channels, bps, data_length;
    int frame_length, last_frame_length, total_frames;

    int32_t *decode_buffer;

    TTAChannel *ch_ctx;
} TTAContext;

static const uint32_t shift_1[] = {
    0x00000001, 0x00000002, 0x00000004, 0x00000008,
    0x00000010, 0x00000020, 0x00000040, 0x00000080,
    0x00000100, 0x00000200, 0x00000400, 0x00000800,
    0x00001000, 0x00002000, 0x00004000, 0x00008000,
    0x00010000, 0x00020000, 0x00040000, 0x00080000,
    0x00100000, 0x00200000, 0x00400000, 0x00800000,
    0x01000000, 0x02000000, 0x04000000, 0x08000000,
    0x10000000, 0x20000000, 0x40000000, 0x80000000,
    0x80000000, 0x80000000, 0x80000000, 0x80000000,
    0x80000000, 0x80000000, 0x80000000, 0x80000000
};

static const uint32_t * const shift_16 = shift_1 + 4;

static const int32_t ttafilter_configs[4][2] = {
    {10, 1},
    {9, 1},
    {10, 1},
    {12, 0}
};

static void ttafilter_init(TTAFilter *c, int32_t shift, int32_t mode) {
    memset(c, 0, sizeof(TTAFilter));
    c->shift = shift;
   c->round = shift_1[shift-1];
//    c->round = 1 << (shift - 1);
    c->mode = mode;
}

// FIXME: copy paste from original
static inline void memshl(register int32_t *a, register int32_t *b) {
    *a++ = *b++;
    *a++ = *b++;
    *a++ = *b++;
    *a++ = *b++;
    *a++ = *b++;
    *a++ = *b++;
    *a++ = *b++;
    *a = *b;
}

// FIXME: copy paste from original
// mode=1 encoder, mode=0 decoder
static inline void ttafilter_process(TTAFilter *c, int32_t *in, int32_t mode) {
    register int32_t *dl = c->dl, *qm = c->qm, *dx = c->dx, sum = c->round;

    if (!c->error) {
        sum += *dl++ * *qm, qm++;
        sum += *dl++ * *qm, qm++;
        sum += *dl++ * *qm, qm++;
        sum += *dl++ * *qm, qm++;
        sum += *dl++ * *qm, qm++;
        sum += *dl++ * *qm, qm++;
        sum += *dl++ * *qm, qm++;
        sum += *dl++ * *qm, qm++;
        dx += 8;
    } else if(c->error < 0) {
        sum += *dl++ * (*qm -= *dx++), qm++;
        sum += *dl++ * (*qm -= *dx++), qm++;
        sum += *dl++ * (*qm -= *dx++), qm++;
        sum += *dl++ * (*qm -= *dx++), qm++;
        sum += *dl++ * (*qm -= *dx++), qm++;
        sum += *dl++ * (*qm -= *dx++), qm++;
        sum += *dl++ * (*qm -= *dx++), qm++;
        sum += *dl++ * (*qm -= *dx++), qm++;
    } else {
        sum += *dl++ * (*qm += *dx++), qm++;
        sum += *dl++ * (*qm += *dx++), qm++;
        sum += *dl++ * (*qm += *dx++), qm++;
        sum += *dl++ * (*qm += *dx++), qm++;
        sum += *dl++ * (*qm += *dx++), qm++;
        sum += *dl++ * (*qm += *dx++), qm++;
        sum += *dl++ * (*qm += *dx++), qm++;
        sum += *dl++ * (*qm += *dx++), qm++;
    }

    *(dx-0) = ((*(dl-1) >> 30) | 1) << 2;
    *(dx-1) = ((*(dl-2) >> 30) | 1) << 1;
    *(dx-2) = ((*(dl-3) >> 30) | 1) << 1;
    *(dx-3) = ((*(dl-4) >> 30) | 1);

    // compress
    if (mode) {
        *dl = *in;
        *in -= (sum >> c->shift);
        c->error = *in;
    } else {
        c->error = *in;
        *in += (sum >> c->shift);
        *dl = *in;
    }

    if (c->mode) {
        *(dl-1) = *dl - *(dl-1);
        *(dl-2) = *(dl-1) - *(dl-2);
        *(dl-3) = *(dl-2) - *(dl-3);
    }

    memshl(c->dl, c->dl + 1);
    memshl(c->dx, c->dx + 1);
}

static void rice_init(TTARice *c, uint32_t k0, uint32_t k1)
{
    c->k0 = k0;
    c->k1 = k1;
    c->sum0 = shift_16[k0];
    c->sum1 = shift_16[k1];
}

static int tta_get_unary(GetBitContext *gb)
{
    int ret = 0;

    // count ones
    while (get_bits_left(gb) > 0 && get_bits1(gb))
        ret++;
    return ret;
}

static const int64_t tta_channel_layouts[7] = {
    AV_CH_LAYOUT_STEREO,
    AV_CH_LAYOUT_STEREO|AV_CH_LOW_FREQUENCY,
    AV_CH_LAYOUT_QUAD,
    0,
    AV_CH_LAYOUT_5POINT1_BACK,
    AV_CH_LAYOUT_5POINT1_BACK|AV_CH_BACK_CENTER,
    AV_CH_LAYOUT_7POINT1_WIDE
};

static av_cold int tta_decode_init(AVCodecContext * avctx)
{
    TTAContext *s = avctx->priv_data;
    int i;

    s->avctx = avctx;

    // 30bytes includes a seektable with one frame
    if (avctx->extradata_size < 30)
        return -1;

    init_get_bits(&s->gb, avctx->extradata, avctx->extradata_size * 8);
    if (show_bits_long(&s->gb, 32) == AV_RL32("TTA1"))
    {
        /* signature */
        skip_bits(&s->gb, 32);

        s->format = get_bits(&s->gb, 16);
        if (s->format > 2) {
            av_log(s->avctx, AV_LOG_ERROR, "Invalid format\n");
            return -1;
        }
        if (s->format == FORMAT_ENCRYPTED) {
            av_log_missing_feature(s->avctx, "Encrypted TTA", 0);
            return AVERROR(EINVAL);
        }
        avctx->channels = s->channels = get_bits(&s->gb, 16);
        if (s->channels > 1 && s->channels < 9)
            avctx->channel_layout = tta_channel_layouts[s->channels-2];
        avctx->bits_per_coded_sample = get_bits(&s->gb, 16);
        s->bps = (avctx->bits_per_coded_sample + 7) / 8;
        avctx->sample_rate = get_bits_long(&s->gb, 32);
        s->data_length = get_bits_long(&s->gb, 32);
        skip_bits(&s->gb, 32); // CRC32 of header

        if (s->channels == 0) {
            av_log(s->avctx, AV_LOG_ERROR, "Invalid number of channels\n");
            return AVERROR_INVALIDDATA;
        }

        switch(s->bps) {
        case 1: avctx->sample_fmt = AV_SAMPLE_FMT_U8; break;
        case 2:
            avctx->sample_fmt = AV_SAMPLE_FMT_S16;
            avctx->bits_per_raw_sample = 16;
            break;
        case 3:
            avctx->sample_fmt = AV_SAMPLE_FMT_S32;
            avctx->bits_per_raw_sample = 24;
            break;
        //case 4: avctx->sample_fmt = AV_SAMPLE_FMT_S32; break;
        default:
            av_log(avctx, AV_LOG_ERROR, "Invalid/unsupported sample format.\n");
            return AVERROR_INVALIDDATA;
        }

        // prevent overflow
        if (avctx->sample_rate > 0x7FFFFF) {
            av_log(avctx, AV_LOG_ERROR, "sample_rate too large\n");
            return AVERROR(EINVAL);
        }
        s->frame_length = 256 * avctx->sample_rate / 245;

        s->last_frame_length = s->data_length % s->frame_length;
        s->total_frames = s->data_length / s->frame_length +
                        (s->last_frame_length ? 1 : 0);

        av_log(s->avctx, AV_LOG_DEBUG, "format: %d chans: %d bps: %d rate: %d block: %d\n",
            s->format, avctx->channels, avctx->bits_per_coded_sample, avctx->sample_rate,
            avctx->block_align);
        av_log(s->avctx, AV_LOG_DEBUG, "data_length: %d frame_length: %d last: %d total: %d\n",
            s->data_length, s->frame_length, s->last_frame_length, s->total_frames);

        // FIXME: seek table
        for (i = 0; i < s->total_frames; i++)
            skip_bits(&s->gb, 32);
        skip_bits(&s->gb, 32); // CRC32 of seektable

        if(s->frame_length >= UINT_MAX / (s->channels * sizeof(int32_t))){
            av_log(avctx, AV_LOG_ERROR, "frame_length too large\n");
            return -1;
        }

        s->decode_buffer = av_mallocz(sizeof(int32_t)*s->frame_length*s->channels);
        if (!s->decode_buffer)
            return AVERROR(ENOMEM);
        s->ch_ctx = av_malloc(avctx->channels * sizeof(*s->ch_ctx));
        if (!s->ch_ctx) {
            av_freep(&s->decode_buffer);
            return AVERROR(ENOMEM);
        }
    } else {
        av_log(avctx, AV_LOG_ERROR, "Wrong extradata present\n");
        return -1;
    }

    avcodec_get_frame_defaults(&s->frame);
    avctx->coded_frame = &s->frame;

    return 0;
}

static int tta_decode_frame(AVCodecContext *avctx, void *data,
                            int *got_frame_ptr, AVPacket *avpkt)
{
    const uint8_t *buf = avpkt->data;
    int buf_size = avpkt->size;
    TTAContext *s = avctx->priv_data;
    int i, ret;
    int cur_chan = 0, framelen = s->frame_length;
    int32_t *p;

    init_get_bits(&s->gb, buf, buf_size*8);

    // FIXME: seeking
    s->total_frames--;
    if (!s->total_frames && s->last_frame_length)
        framelen = s->last_frame_length;

    /* get output buffer */
    s->frame.nb_samples = framelen;
    if ((ret = avctx->get_buffer(avctx, &s->frame)) < 0) {
        av_log(avctx, AV_LOG_ERROR, "get_buffer() failed\n");
        return ret;
    }

    // decode directly to output buffer for 24-bit sample format
    if (s->bps == 3)
        s->decode_buffer = (int32_t *)s->frame.data[0];

    // init per channel states
    for (i = 0; i < s->channels; i++) {
        s->ch_ctx[i].predictor = 0;
        ttafilter_init(&s->ch_ctx[i].filter, ttafilter_configs[s->bps-1][0], ttafilter_configs[s->bps-1][1]);
        rice_init(&s->ch_ctx[i].rice, 10, 10);
    }

    for (p = s->decode_buffer; p < s->decode_buffer + (framelen * s->channels); p++) {
        int32_t *predictor = &s->ch_ctx[cur_chan].predictor;
        TTAFilter *filter = &s->ch_ctx[cur_chan].filter;
        TTARice *rice = &s->ch_ctx[cur_chan].rice;
        uint32_t unary, depth, k;
        int32_t value;

        unary = tta_get_unary(&s->gb);

        if (unary == 0) {
            depth = 0;
            k = rice->k0;
        } else {
            depth = 1;
            k = rice->k1;
            unary--;
        }

        if (get_bits_left(&s->gb) < k)
            return -1;

        if (k) {
            if (k > MIN_CACHE_BITS)
                return -1;
            value = (unary << k) + get_bits(&s->gb, k);
        } else
            value = unary;

        // FIXME: copy paste from original
        switch (depth) {
        case 1:
            rice->sum1 += value - (rice->sum1 >> 4);
            if (rice->k1 > 0 && rice->sum1 < shift_16[rice->k1])
                rice->k1--;
            else if(rice->sum1 > shift_16[rice->k1 + 1])
                rice->k1++;
            value += shift_1[rice->k0];
        default:
            rice->sum0 += value - (rice->sum0 >> 4);
            if (rice->k0 > 0 && rice->sum0 < shift_16[rice->k0])
                rice->k0--;
            else if(rice->sum0 > shift_16[rice->k0 + 1])
                rice->k0++;
        }

        // extract coded value
#define UNFOLD(x) (((x)&1) ? (++(x)>>1) : (-(x)>>1))
        *p = UNFOLD(value);

        // run hybrid filter
        ttafilter_process(filter, p, 0);

        // fixed order prediction
#define PRED(x, k) (int32_t)((((uint64_t)x << k) - x) >> k)
        switch (s->bps) {
            case 1: *p += PRED(*predictor, 4); break;
            case 2:
            case 3: *p += PRED(*predictor, 5); break;
            case 4: *p += *predictor; break;
        }
        *predictor = *p;

        // flip channels
        if (cur_chan < (s->channels-1))
            cur_chan++;
        else {
            // decorrelate in case of stereo integer
            if (s->channels > 1) {
                int32_t *r = p - 1;
                for (*p += *r / 2; r > p - s->channels; r--)
                    *r = *(r + 1) - *r;
            }
            cur_chan = 0;
        }
    }

    if (get_bits_left(&s->gb) < 32)
        return -1;
    skip_bits(&s->gb, 32); // frame crc

        // convert to output buffer
        switch(s->bps) {
            case 1: {
                uint8_t *samples = (uint8_t *)s->frame.data[0];
                for (p = s->decode_buffer; p < s->decode_buffer + (framelen * s->channels); p++)
                    *samples++ = *p + 0x80;
                break;
            }
            case 2: {
                uint16_t *samples = (int16_t *)s->frame.data[0];
                for (p = s->decode_buffer; p < s->decode_buffer + (framelen * s->channels); p++)
                    *samples++ = *p;
                break;
            }
            case 3: {
                // shift samples for 24-bit sample format
                int32_t *samples = (int32_t *)s->frame.data[0];
                for (p = s->decode_buffer; p < s->decode_buffer + (framelen * s->channels); p++)
                    *samples++ <<= 8;
                // reset decode buffer
                s->decode_buffer = NULL;
                break;
            }
        }

    *got_frame_ptr   = 1;
    *(AVFrame *)data = s->frame;

    return buf_size;
}

static av_cold int tta_decode_close(AVCodecContext *avctx) {
    TTAContext *s = avctx->priv_data;

    av_free(s->decode_buffer);
    av_freep(&s->ch_ctx);

    return 0;
}

AVCodec ff_tta_decoder = {
    .name           = "tta",
    .type           = AVMEDIA_TYPE_AUDIO,
    .id             = CODEC_ID_TTA,
    .priv_data_size = sizeof(TTAContext),
    .init           = tta_decode_init,
    .close          = tta_decode_close,
    .decode         = tta_decode_frame,
    .capabilities   = CODEC_CAP_DR1,
    .long_name = NULL_IF_CONFIG_SMALL("True Audio (TTA)"),
};