aboutsummaryrefslogtreecommitdiffstats
path: root/libavcodec/audiotoolboxdec.c
blob: f840a6ba421c0c4c0808c5f3258283eb361d0557 (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
/*
 * Audio Toolbox system codecs
 *
 * copyright (c) 2016 Rodger Combs
 *
 * 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
 */

#include <AudioToolbox/AudioToolbox.h>

#include "config.h"
#include "avcodec.h"
#include "bytestream.h"
#include "internal.h"
#include "libavutil/avassert.h"
#include "libavutil/opt.h"
#include "libavutil/log.h"

#ifndef __MAC_10_11
#define kAudioFormatEnhancedAC3 'ec-3'
#endif

typedef struct ATDecodeContext {
    AVClass *av_class;

    AudioConverterRef converter;
    AudioStreamPacketDescription pkt_desc;
    AVPacket in_pkt;
    AVPacket new_in_pkt;
    AVBitStreamFilterContext *bsf;
    char *decoded_data;
    int channel_map[64];

    int64_t last_pts;
    int eof;
} ATDecodeContext;

static UInt32 ffat_get_format_id(enum AVCodecID codec, int profile)
{
    switch (codec) {
    case AV_CODEC_ID_AAC:
        return kAudioFormatMPEG4AAC;
    case AV_CODEC_ID_AC3:
        return kAudioFormatAC3;
    case AV_CODEC_ID_ADPCM_IMA_QT:
        return kAudioFormatAppleIMA4;
    case AV_CODEC_ID_ALAC:
        return kAudioFormatAppleLossless;
    case AV_CODEC_ID_AMR_NB:
        return kAudioFormatAMR;
    case AV_CODEC_ID_EAC3:
        return kAudioFormatEnhancedAC3;
    case AV_CODEC_ID_GSM_MS:
        return kAudioFormatMicrosoftGSM;
    case AV_CODEC_ID_ILBC:
        return kAudioFormatiLBC;
    case AV_CODEC_ID_MP1:
        return kAudioFormatMPEGLayer1;
    case AV_CODEC_ID_MP2:
        return kAudioFormatMPEGLayer2;
    case AV_CODEC_ID_MP3:
        return kAudioFormatMPEGLayer3;
    case AV_CODEC_ID_PCM_ALAW:
        return kAudioFormatALaw;
    case AV_CODEC_ID_PCM_MULAW:
        return kAudioFormatULaw;
    case AV_CODEC_ID_QDMC:
        return kAudioFormatQDesign;
    case AV_CODEC_ID_QDM2:
        return kAudioFormatQDesign2;
    default:
        av_assert0(!"Invalid codec ID!");
        return 0;
    }
}

static int ffat_get_channel_id(AudioChannelLabel label)
{
    if (label == 0)
        return -1;
    else if (label <= kAudioChannelLabel_LFEScreen)
        return label - 1;
    else if (label <= kAudioChannelLabel_RightSurround)
        return label + 4;
    else if (label <= kAudioChannelLabel_CenterSurround)
        return label + 1;
    else if (label <= kAudioChannelLabel_RightSurroundDirect)
        return label + 23;
    else if (label <= kAudioChannelLabel_TopBackRight)
        return label - 1;
    else if (label < kAudioChannelLabel_RearSurroundLeft)
        return -1;
    else if (label <= kAudioChannelLabel_RearSurroundRight)
        return label - 29;
    else if (label <= kAudioChannelLabel_RightWide)
        return label - 4;
    else if (label == kAudioChannelLabel_LFE2)
        return ff_ctzll(AV_CH_LOW_FREQUENCY_2);
    else if (label == kAudioChannelLabel_Mono)
        return ff_ctzll(AV_CH_FRONT_CENTER);
    else
        return -1;
}

static int ffat_compare_channel_descriptions(const void* a, const void* b)
{
    const AudioChannelDescription* da = a;
    const AudioChannelDescription* db = b;
    return ffat_get_channel_id(da->mChannelLabel) - ffat_get_channel_id(db->mChannelLabel);
}

static AudioChannelLayout *ffat_convert_layout(AudioChannelLayout *layout, UInt32* size)
{
    AudioChannelLayoutTag tag = layout->mChannelLayoutTag;
    AudioChannelLayout *new_layout;
    if (tag == kAudioChannelLayoutTag_UseChannelDescriptions)
        return layout;
    else if (tag == kAudioChannelLayoutTag_UseChannelBitmap)
        AudioFormatGetPropertyInfo(kAudioFormatProperty_ChannelLayoutForBitmap,
                                   sizeof(UInt32), &layout->mChannelBitmap, size);
    else
        AudioFormatGetPropertyInfo(kAudioFormatProperty_ChannelLayoutForTag,
                                   sizeof(AudioChannelLayoutTag), &tag, size);
    new_layout = av_malloc(*size);
    if (!new_layout) {
        av_free(layout);
        return NULL;
    }
    if (tag == kAudioChannelLayoutTag_UseChannelBitmap)
        AudioFormatGetProperty(kAudioFormatProperty_ChannelLayoutForBitmap,
                               sizeof(UInt32), &layout->mChannelBitmap, size, new_layout);
    else
        AudioFormatGetProperty(kAudioFormatProperty_ChannelLayoutForTag,
                               sizeof(AudioChannelLayoutTag), &tag, size, new_layout);
    new_layout->mChannelLayoutTag = kAudioChannelLayoutTag_UseChannelDescriptions;
    av_free(layout);
    return new_layout;
}

static int ffat_update_ctx(AVCodecContext *avctx)
{
    ATDecodeContext *at = avctx->priv_data;
    AudioStreamBasicDescription format;
    UInt32 size = sizeof(format);
    if (!AudioConverterGetProperty(at->converter,
                                   kAudioConverterCurrentInputStreamDescription,
                                   &size, &format)) {
        if (format.mSampleRate)
            avctx->sample_rate = format.mSampleRate;
        avctx->channels = format.mChannelsPerFrame;
        avctx->channel_layout = av_get_default_channel_layout(avctx->channels);
        avctx->frame_size = format.mFramesPerPacket;
    }

    if (!AudioConverterGetProperty(at->converter,
                                   kAudioConverterCurrentOutputStreamDescription,
                                   &size, &format)) {
        format.mSampleRate = avctx->sample_rate;
        format.mChannelsPerFrame = avctx->channels;
        AudioConverterSetProperty(at->converter,
                                  kAudioConverterCurrentOutputStreamDescription,
                                  size, &format);
    }

    if (!AudioConverterGetPropertyInfo(at->converter, kAudioConverterOutputChannelLayout,
                                       &size, NULL) && size) {
        AudioChannelLayout *layout = av_malloc(size);
        uint64_t layout_mask = 0;
        int i;
        if (!layout)
            return AVERROR(ENOMEM);
        AudioConverterGetProperty(at->converter, kAudioConverterOutputChannelLayout,
                                  &size, layout);
        if (!(layout = ffat_convert_layout(layout, &size)))
            return AVERROR(ENOMEM);
        for (i = 0; i < layout->mNumberChannelDescriptions; i++) {
            int id = ffat_get_channel_id(layout->mChannelDescriptions[i].mChannelLabel);
            if (id < 0)
                goto done;
            if (layout_mask & (1 << id))
                goto done;
            layout_mask |= 1 << id;
            layout->mChannelDescriptions[i].mChannelFlags = i; // Abusing flags as index
        }
        avctx->channel_layout = layout_mask;
        qsort(layout->mChannelDescriptions, layout->mNumberChannelDescriptions,
              sizeof(AudioChannelDescription), &ffat_compare_channel_descriptions);
        for (i = 0; i < layout->mNumberChannelDescriptions; i++)
            at->channel_map[i] = layout->mChannelDescriptions[i].mChannelFlags;
done:
        av_free(layout);
    }

    if (!avctx->frame_size)
        avctx->frame_size = 2048;

    return 0;
}

static void put_descr(PutByteContext *pb, int tag, unsigned int size)
{
    int i = 3;
    bytestream2_put_byte(pb, tag);
    for (; i > 0; i--)
        bytestream2_put_byte(pb, (size >> (7 * i)) | 0x80);
    bytestream2_put_byte(pb, size & 0x7F);
}

static int ffat_set_extradata(AVCodecContext *avctx)
{
    ATDecodeContext *at = avctx->priv_data;
    if (avctx->extradata_size) {
        OSStatus status;
        char *extradata = avctx->extradata;
        int extradata_size = avctx->extradata_size;
        if (avctx->codec_id == AV_CODEC_ID_AAC) {
            PutByteContext pb;
            extradata_size = 5 + 3 + 5+13 + 5+avctx->extradata_size;
            if (!(extradata = av_malloc(extradata_size)))
                return AVERROR(ENOMEM);

            bytestream2_init_writer(&pb, extradata, extradata_size);

            // ES descriptor
            put_descr(&pb, 0x03, 3 + 5+13 + 5+avctx->extradata_size);
            bytestream2_put_be16(&pb, 0);
            bytestream2_put_byte(&pb, 0x00); // flags (= no flags)

            // DecoderConfig descriptor
            put_descr(&pb, 0x04, 13 + 5+avctx->extradata_size);

            // Object type indication
            bytestream2_put_byte(&pb, 0x40);

            bytestream2_put_byte(&pb, 0x15); // flags (= Audiostream)

            bytestream2_put_be24(&pb, 0); // Buffersize DB

            bytestream2_put_be32(&pb, 0); // maxbitrate
            bytestream2_put_be32(&pb, 0); // avgbitrate

            // DecoderSpecific info descriptor
            put_descr(&pb, 0x05, avctx->extradata_size);
            bytestream2_put_buffer(&pb, avctx->extradata, avctx->extradata_size);
        }

        status = AudioConverterSetProperty(at->converter,
                                           kAudioConverterDecompressionMagicCookie,
                                           extradata_size, extradata);
        if (status != 0)
            av_log(avctx, AV_LOG_WARNING, "AudioToolbox cookie error: %i\n", (int)status);

        if (avctx->codec_id == AV_CODEC_ID_AAC)
            av_free(extradata);
    }
    return 0;
}

static av_cold int ffat_create_decoder(AVCodecContext *avctx)
{
    ATDecodeContext *at = avctx->priv_data;
    OSStatus status;
    int i;

    enum AVSampleFormat sample_fmt = (avctx->bits_per_raw_sample == 32) ?
                                     AV_SAMPLE_FMT_S32 : AV_SAMPLE_FMT_S16;

    AudioStreamBasicDescription in_format = {
        .mSampleRate = avctx->sample_rate ? avctx->sample_rate : 44100,
        .mFormatID = ffat_get_format_id(avctx->codec_id, avctx->profile),
        .mBytesPerPacket = avctx->block_align,
        .mChannelsPerFrame = avctx->channels ? avctx->channels : 1,
    };
    AudioStreamBasicDescription out_format = {
        .mSampleRate = in_format.mSampleRate,
        .mFormatID = kAudioFormatLinearPCM,
        .mFormatFlags = kAudioFormatFlagIsSignedInteger | kAudioFormatFlagIsPacked,
        .mFramesPerPacket = 1,
        .mChannelsPerFrame = in_format.mChannelsPerFrame,
        .mBitsPerChannel = av_get_bytes_per_sample(sample_fmt) * 8,
    };

    avctx->sample_fmt = sample_fmt;

    if (avctx->codec_id == AV_CODEC_ID_ADPCM_IMA_QT)
        in_format.mFramesPerPacket = 64;

    status = AudioConverterNew(&in_format, &out_format, &at->converter);

    if (status != 0) {
        av_log(avctx, AV_LOG_ERROR, "AudioToolbox init error: %i\n", (int)status);
        return AVERROR_UNKNOWN;
    }

    if ((status = ffat_set_extradata(avctx)) < 0)
        return status;

    for (i = 0; i < (sizeof(at->channel_map) / sizeof(at->channel_map[0])); i++)
        at->channel_map[i] = i;

    ffat_update_ctx(avctx);

    if(!(at->decoded_data = av_malloc(av_get_bytes_per_sample(avctx->sample_fmt)
                                      * avctx->frame_size * avctx->channels)))
        return AVERROR(ENOMEM);

    at->last_pts = AV_NOPTS_VALUE;

    return 0;
}

static av_cold int ffat_init_decoder(AVCodecContext *avctx)
{
    if (avctx->channels || avctx->extradata_size)
        return ffat_create_decoder(avctx);
    else
        return 0;
}

static OSStatus ffat_decode_callback(AudioConverterRef converter, UInt32 *nb_packets,
                                     AudioBufferList *data,
                                     AudioStreamPacketDescription **packets,
                                     void *inctx)
{
    AVCodecContext *avctx = inctx;
    ATDecodeContext *at = avctx->priv_data;

    if (at->eof) {
        *nb_packets = 0;
        if (packets) {
            *packets = &at->pkt_desc;
            at->pkt_desc.mDataByteSize = 0;
        }
        return 0;
    }

    av_packet_move_ref(&at->in_pkt, &at->new_in_pkt);
    at->new_in_pkt.data = 0;
    at->new_in_pkt.size = 0;

    if (!at->in_pkt.data) {
        *nb_packets = 0;
        return 1;
    }

    data->mNumberBuffers              = 1;
    data->mBuffers[0].mNumberChannels = 0;
    data->mBuffers[0].mDataByteSize   = at->in_pkt.size;
    data->mBuffers[0].mData           = at->in_pkt.data;
    *nb_packets = 1;

    if (packets) {
        *packets = &at->pkt_desc;
        at->pkt_desc.mDataByteSize = at->in_pkt.size;
    }

    return 0;
}

#define COPY_SAMPLES(type) \
    type *in_ptr = (type*)at->decoded_data; \
    type *end_ptr = in_ptr + frame->nb_samples * avctx->channels; \
    type *out_ptr = (type*)frame->data[0]; \
    for (; in_ptr < end_ptr; in_ptr += avctx->channels, out_ptr += avctx->channels) { \
        int c; \
        for (c = 0; c < avctx->channels; c++) \
            out_ptr[c] = in_ptr[at->channel_map[c]]; \
    }

static void ffat_copy_samples(AVCodecContext *avctx, AVFrame *frame)
{
    ATDecodeContext *at = avctx->priv_data;
    if (avctx->sample_fmt == AV_SAMPLE_FMT_S32) {
        COPY_SAMPLES(int32_t);
    } else {
        COPY_SAMPLES(int16_t);
    }
}

static int ffat_decode(AVCodecContext *avctx, void *data,
                       int *got_frame_ptr, AVPacket *avpkt)
{
    ATDecodeContext *at = avctx->priv_data;
    AVFrame *frame = data;
    int pkt_size = avpkt->size;
    AVPacket filtered_packet;
    OSStatus ret;
    AudioBufferList out_buffers;

    if (avctx->codec_id == AV_CODEC_ID_AAC && avpkt->size > 2 &&
        (AV_RB16(avpkt->data) & 0xfff0) == 0xfff0) {
        uint8_t *p_filtered = NULL;
        int      n_filtered = 0;
        if (!at->bsf) {
            if(!(at->bsf = av_bitstream_filter_init("aac_adtstoasc")))
                return AVERROR(ENOMEM);
        }

        ret = av_bitstream_filter_filter(at->bsf, avctx, NULL, &p_filtered, &n_filtered,
                                         avpkt->data, avpkt->size, 0);
        if (ret >= 0 && p_filtered != avpkt->data) {
            filtered_packet = *avpkt;
            avpkt = &filtered_packet;
            avpkt->data = p_filtered;
            avpkt->size = n_filtered;
        }
    }

    if (!at->converter) {
        if ((ret = ffat_create_decoder(avctx)) < 0)
            return ret;
    }

    out_buffers = (AudioBufferList){
        .mNumberBuffers = 1,
        .mBuffers = {
            {
                .mNumberChannels = avctx->channels,
                .mDataByteSize = av_get_bytes_per_sample(avctx->sample_fmt) * avctx->frame_size
                                 * avctx->channels,
            }
        }
    };

    av_packet_unref(&at->new_in_pkt);

    if (avpkt->size) {
        if ((ret = av_packet_ref(&at->new_in_pkt, avpkt)) < 0)
            return ret;
        at->new_in_pkt.data = avpkt->data;
    } else {
        at->eof = 1;
    }

    frame->sample_rate = avctx->sample_rate;

    frame->nb_samples = avctx->frame_size;

    out_buffers.mBuffers[0].mData = at->decoded_data;

    ret = AudioConverterFillComplexBuffer(at->converter, ffat_decode_callback, avctx,
                                          &frame->nb_samples, &out_buffers, NULL);
    if ((!ret || ret == 1) && frame->nb_samples) {
        if ((ret = ff_get_buffer(avctx, frame, 0)) < 0)
            return ret;
        ffat_copy_samples(avctx, frame);
        *got_frame_ptr = 1;
        if (at->last_pts != AV_NOPTS_VALUE) {
            frame->pkt_pts = at->last_pts;
            at->last_pts = avpkt->pts;
        }
    } else if (ret && ret != 1) {
        av_log(avctx, AV_LOG_WARNING, "Decode error: %i\n", ret);
    } else {
        at->last_pts = avpkt->pts;
    }

    return pkt_size;
}

static av_cold void ffat_decode_flush(AVCodecContext *avctx)
{
    ATDecodeContext *at = avctx->priv_data;
    AudioConverterReset(at->converter);
    av_packet_unref(&at->new_in_pkt);
    av_packet_unref(&at->in_pkt);
}

static av_cold int ffat_close_decoder(AVCodecContext *avctx)
{
    ATDecodeContext *at = avctx->priv_data;
    AudioConverterDispose(at->converter);
    av_packet_unref(&at->new_in_pkt);
    av_packet_unref(&at->in_pkt);
    av_free(at->decoded_data);
    return 0;
}

#define FFAT_DEC_CLASS(NAME) \
    static const AVClass ffat_##NAME##_dec_class = { \
        .class_name = "at_" #NAME "_dec", \
        .version    = LIBAVUTIL_VERSION_INT, \
    };

#define FFAT_DEC(NAME, ID) \
    FFAT_DEC_CLASS(NAME) \
    AVCodec ff_##NAME##_at_decoder = { \
        .name           = #NAME "_at", \
        .long_name      = NULL_IF_CONFIG_SMALL(#NAME " (AudioToolbox)"), \
        .type           = AVMEDIA_TYPE_AUDIO, \
        .id             = ID, \
        .priv_data_size = sizeof(ATDecodeContext), \
        .init           = ffat_init_decoder, \
        .close          = ffat_close_decoder, \
        .decode         = ffat_decode, \
        .flush          = ffat_decode_flush, \
        .priv_class     = &ffat_##NAME##_dec_class, \
        .capabilities   = AV_CODEC_CAP_DR1 | AV_CODEC_CAP_DELAY, \
        .caps_internal  = FF_CODEC_CAP_INIT_THREADSAFE, \
    };

FFAT_DEC(aac,          AV_CODEC_ID_AAC)
FFAT_DEC(ac3,          AV_CODEC_ID_AC3)
FFAT_DEC(adpcm_ima_qt, AV_CODEC_ID_ADPCM_IMA_QT)
FFAT_DEC(alac,         AV_CODEC_ID_ALAC)
FFAT_DEC(amr_nb,       AV_CODEC_ID_AMR_NB)
FFAT_DEC(eac3,         AV_CODEC_ID_EAC3)
FFAT_DEC(gsm_ms,       AV_CODEC_ID_GSM_MS)
FFAT_DEC(ilbc,         AV_CODEC_ID_ILBC)
FFAT_DEC(mp1,          AV_CODEC_ID_MP1)
FFAT_DEC(mp2,          AV_CODEC_ID_MP2)
FFAT_DEC(mp3,          AV_CODEC_ID_MP3)
FFAT_DEC(pcm_alaw,     AV_CODEC_ID_PCM_ALAW)
FFAT_DEC(pcm_mulaw,    AV_CODEC_ID_PCM_MULAW)
FFAT_DEC(qdmc,         AV_CODEC_ID_QDMC)
FFAT_DEC(qdm2,         AV_CODEC_ID_QDM2)