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
|
/* ScummVM - Graphic Adventure Engine
*
* ScummVM is the legal property of its developers, whose names
* are too numerous to list here. Please refer to the COPYRIGHT
* file distributed with this source distribution.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This program 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 General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
*
*/
#include "common/stream.h"
#include "common/textconsole.h"
#include "common/util.h"
#include "audio/decoders/adpcm.h"
#include "audio/decoders/adpcm_intern.h"
namespace Audio {
// Routines to convert 12 bit linear samples to the
// Dialogic or Oki ADPCM coding format aka VOX.
// See also <http://www.comptek.ru/telephony/tnotes/tt1-13.html>
//
// IMA ADPCM support is based on
// <http://wiki.multimedia.cx/index.php?title=IMA_ADPCM>
//
// In addition, also MS IMA ADPCM is supported. See
// <http://wiki.multimedia.cx/index.php?title=Microsoft_IMA_ADPCM>.
ADPCMStream::ADPCMStream(Common::SeekableReadStream *stream, DisposeAfterUse::Flag disposeAfterUse, uint32 size, int rate, int channels, uint32 blockAlign)
: _stream(stream, disposeAfterUse),
_startpos(stream->pos()),
_endpos(_startpos + size),
_channels(channels),
_blockAlign(blockAlign),
_rate(rate) {
reset();
}
void ADPCMStream::reset() {
memset(&_status, 0, sizeof(_status));
_blockPos[0] = _blockPos[1] = _blockAlign; // To make sure first header is read
}
bool ADPCMStream::rewind() {
// TODO: Error checking.
reset();
_stream->seek(_startpos);
return true;
}
#pragma mark -
int Oki_ADPCMStream::readBuffer(int16 *buffer, const int numSamples) {
int samples;
byte data;
for (samples = 0; samples < numSamples && !endOfData(); samples++) {
if (_decodedSampleCount == 0) {
data = _stream->readByte();
_decodedSamples[0] = decodeOKI((data >> 4) & 0x0f);
_decodedSamples[1] = decodeOKI((data >> 0) & 0x0f);
_decodedSampleCount = 2;
}
// (1 - (count - 1)) ensures that _decodedSamples acts as a FIFO of depth 2
buffer[samples] = _decodedSamples[1 - (_decodedSampleCount - 1)];
_decodedSampleCount--;
}
return samples;
}
static const int16 okiStepSize[49] = {
16, 17, 19, 21, 23, 25, 28, 31,
34, 37, 41, 45, 50, 55, 60, 66,
73, 80, 88, 97, 107, 118, 130, 143,
157, 173, 190, 209, 230, 253, 279, 307,
337, 371, 408, 449, 494, 544, 598, 658,
724, 796, 876, 963, 1060, 1166, 1282, 1411,
1552
};
// Decode Linear to ADPCM
int16 Oki_ADPCMStream::decodeOKI(byte code) {
int16 diff, E, samp;
E = (2 * (code & 0x7) + 1) * okiStepSize[_status.ima_ch[0].stepIndex] / 8;
diff = (code & 0x08) ? -E : E;
samp = _status.ima_ch[0].last + diff;
// Clip the values to +/- 2^11 (supposed to be 12 bits)
samp = CLIP<int16>(samp, -2048, 2047);
_status.ima_ch[0].last = samp;
_status.ima_ch[0].stepIndex += _stepAdjustTable[code];
_status.ima_ch[0].stepIndex = CLIP<int32>(_status.ima_ch[0].stepIndex, 0, ARRAYSIZE(okiStepSize) - 1);
// * 16 effectively converts 12-bit input to 16-bit output
return samp * 16;
}
#pragma mark -
int DVI_ADPCMStream::readBuffer(int16 *buffer, const int numSamples) {
int samples;
byte data;
for (samples = 0; samples < numSamples && !endOfData(); samples++) {
if (_decodedSampleCount == 0) {
data = _stream->readByte();
_decodedSamples[0] = decodeIMA((data >> 4) & 0x0f, 0);
_decodedSamples[1] = decodeIMA((data >> 0) & 0x0f, _channels == 2 ? 1 : 0);
_decodedSampleCount = 2;
}
// (1 - (count - 1)) ensures that _decodedSamples acts as a FIFO of depth 2
buffer[samples] = _decodedSamples[1 - (_decodedSampleCount - 1)];
_decodedSampleCount--;
}
return samples;
}
#pragma mark -
int Apple_ADPCMStream::readBuffer(int16 *buffer, const int numSamples) {
// Need to write at least one samples per channel
assert((numSamples % _channels) == 0);
// Current sample positions
int samples[2] = { 0, 0};
// Number of samples per channel
int chanSamples = numSamples / _channels;
for (int i = 0; i < _channels; i++) {
_stream->seek(_streamPos[i]);
while ((samples[i] < chanSamples) &&
// Last byte read and a new one needed
!((_stream->eos() || (_stream->pos() >= _endpos)) && (_chunkPos[i] == 0))) {
if (_blockPos[i] == _blockAlign) {
// 2 byte header per block
uint16 temp = _stream->readUint16BE();
// First 9 bits are the upper bits of the predictor
_status.ima_ch[i].last = (int16) (temp & 0xFF80);
// Lower 7 bits are the step index
_status.ima_ch[i].stepIndex = temp & 0x007F;
// Clip the step index
_status.ima_ch[i].stepIndex = CLIP<int32>(_status.ima_ch[i].stepIndex, 0, 88);
_blockPos[i] = 2;
}
if (_chunkPos[i] == 0) {
// Decode data
byte data = _stream->readByte();
_buffer[i][0] = decodeIMA(data & 0x0F, i);
_buffer[i][1] = decodeIMA(data >> 4, i);
}
// The original is interleaved block-wise, we want it sample-wise
buffer[_channels * samples[i] + i] = _buffer[i][_chunkPos[i]];
if (++_chunkPos[i] > 1) {
// We're about to decode the next byte, so advance the block position
_chunkPos[i] = 0;
_blockPos[i]++;
}
samples[i]++;
if (_channels == 2)
if (_blockPos[i] == _blockAlign)
// We're at the end of the block.
// Since the channels are interleaved, skip the next block
_stream->skip(MIN<uint32>(_blockAlign, _endpos - _stream->pos()));
_streamPos[i] = _stream->pos();
}
}
return samples[0] + samples[1];
}
#pragma mark -
int MSIma_ADPCMStream::readBuffer(int16 *buffer, const int numSamples) {
// Need to write at least one sample per channel
assert((numSamples % _channels) == 0);
int samples = 0;
while (samples < numSamples && !_stream->eos() && _stream->pos() < _endpos) {
if (_blockPos[0] == _blockAlign) {
for (int i = 0; i < _channels; i++) {
// read block header
_status.ima_ch[i].last = _stream->readSint16LE();
_status.ima_ch[i].stepIndex = _stream->readSint16LE();
}
_blockPos[0] = _channels * 4;
}
// Decode a set of samples
for (int i = 0; i < _channels; i++) {
// The stream encodes four bytes per channel at a time
for (int j = 0; j < 4; j++) {
byte data = _stream->readByte();
_blockPos[0]++;
_buffer[i][j * 2] = decodeIMA(data & 0x0f, i);
_buffer[i][j * 2 + 1] = decodeIMA((data >> 4) & 0x0f, i);
_samplesLeft[i] += 2;
}
}
while (samples < numSamples && _samplesLeft[0] != 0) {
for (int i = 0; i < _channels; i++) {
buffer[samples + i] = _buffer[i][8 - _samplesLeft[i]];
_samplesLeft[i]--;
}
samples += _channels;
}
}
return samples;
}
#pragma mark -
static const int MSADPCMAdaptCoeff1[] = {
256, 512, 0, 192, 240, 460, 392
};
static const int MSADPCMAdaptCoeff2[] = {
0, -256, 0, 64, 0, -208, -232
};
static const int MSADPCMAdaptationTable[] = {
230, 230, 230, 230, 307, 409, 512, 614,
768, 614, 512, 409, 307, 230, 230, 230
};
int16 MS_ADPCMStream::decodeMS(ADPCMChannelStatus *c, byte code) {
int32 predictor;
predictor = (((c->sample1) * (c->coeff1)) + ((c->sample2) * (c->coeff2))) / 256;
predictor += (signed)((code & 0x08) ? (code - 0x10) : (code)) * c->delta;
predictor = CLIP<int32>(predictor, -32768, 32767);
c->sample2 = c->sample1;
c->sample1 = predictor;
c->delta = (MSADPCMAdaptationTable[(int)code] * c->delta) >> 8;
if (c->delta < 16)
c->delta = 16;
return (int16)predictor;
}
int MS_ADPCMStream::readBuffer(int16 *buffer, const int numSamples) {
int samples;
byte data;
int i;
for (samples = 0; samples < numSamples && !endOfData(); samples++) {
if (_decodedSampleCount == 0) {
if (_blockPos[0] == _blockAlign) {
// read block header
for (i = 0; i < _channels; i++) {
_status.ch[i].predictor = CLIP(_stream->readByte(), (byte)0, (byte)6);
_status.ch[i].coeff1 = MSADPCMAdaptCoeff1[_status.ch[i].predictor];
_status.ch[i].coeff2 = MSADPCMAdaptCoeff2[_status.ch[i].predictor];
}
for (i = 0; i < _channels; i++)
_status.ch[i].delta = _stream->readSint16LE();
for (i = 0; i < _channels; i++)
_status.ch[i].sample1 = _stream->readSint16LE();
for (i = 0; i < _channels; i++)
_decodedSamples[_decodedSampleCount++] = _status.ch[i].sample2 = _stream->readSint16LE();
for (i = 0; i < _channels; i++)
_decodedSamples[_decodedSampleCount++] = _status.ch[i].sample1;
_blockPos[0] = _channels * 7;
} else {
data = _stream->readByte();
_blockPos[0]++;
_decodedSamples[_decodedSampleCount++] = decodeMS(&_status.ch[0], (data >> 4) & 0x0f);
_decodedSamples[_decodedSampleCount++] = decodeMS(&_status.ch[_channels - 1], data & 0x0f);
}
_decodedSampleIndex = 0;
}
// _decodedSamples acts as a FIFO of depth 2 or 4;
buffer[samples] = _decodedSamples[_decodedSampleIndex++];
_decodedSampleCount--;
}
return samples;
}
#pragma mark -
#define DK3_READ_NIBBLE(channelNo) \
do { \
if (_topNibble) { \
_nibble = _lastByte >> 4; \
_topNibble = false; \
} else { \
_lastByte = _stream->readByte(); \
_nibble = _lastByte & 0xf; \
_topNibble = true; \
--blockBytesLeft; \
--audioBytesLeft; \
} \
decodeIMA(_nibble, channelNo); \
} while(0)
int DK3_ADPCMStream::readBuffer(int16 *buffer, const int numSamples) {
assert((numSamples % 4) == 0);
const uint32 startOffset = _stream->pos() % _blockAlign;
uint32 audioBytesLeft = _endpos - _stream->pos();
uint32 blockBytesLeft;
if (startOffset != 0) {
blockBytesLeft = _blockAlign - startOffset;
} else {
blockBytesLeft = 0;
}
int samples = 0;
while (samples < numSamples && audioBytesLeft) {
if (blockBytesLeft == 0) {
blockBytesLeft = MIN(_blockAlign, audioBytesLeft);
_topNibble = false;
if (blockBytesLeft < 16) {
warning("Truncated DK3 ADPCM block header");
break;
}
_stream->skip(2);
const uint16 rate = _stream->readUint16LE();
assert(rate == getRate());
_stream->skip(6);
// Get predictor for both sum/diff channels
_status.ima_ch[0].last = _stream->readSint16LE();
_status.ima_ch[1].last = _stream->readSint16LE();
// Get index for both sum/diff channels
_status.ima_ch[0].stepIndex = _stream->readByte();
_status.ima_ch[1].stepIndex = _stream->readByte();
assert(_status.ima_ch[0].stepIndex < ARRAYSIZE(_imaTable));
assert(_status.ima_ch[1].stepIndex < ARRAYSIZE(_imaTable));
blockBytesLeft -= 16;
audioBytesLeft -= 16;
}
DK3_READ_NIBBLE(0);
DK3_READ_NIBBLE(1);
*buffer++ = _status.ima_ch[0].last + _status.ima_ch[1].last;
*buffer++ = _status.ima_ch[0].last - _status.ima_ch[1].last;
DK3_READ_NIBBLE(0);
*buffer++ = _status.ima_ch[0].last + _status.ima_ch[1].last;
*buffer++ = _status.ima_ch[0].last - _status.ima_ch[1].last;
samples += 4;
// if the last sample of a block ends on an odd byte, the encoder adds
// an extra alignment byte
if (!_topNibble && blockBytesLeft == 1) {
_stream->skip(1);
--blockBytesLeft;
--audioBytesLeft;
}
}
return samples;
}
#undef DK3_READ_NIBBLE
#pragma mark -
// This table is used to adjust the step for use on the next sample.
// We could half the table, but since the lookup index used is always
// a 4-bit nibble, it's more efficient to just keep it as it is.
const int16 ADPCMStream::_stepAdjustTable[16] = {
-1, -1, -1, -1, 2, 4, 6, 8,
-1, -1, -1, -1, 2, 4, 6, 8
};
const int16 Ima_ADPCMStream::_imaTable[89] = {
7, 8, 9, 10, 11, 12, 13, 14,
16, 17, 19, 21, 23, 25, 28, 31,
34, 37, 41, 45, 50, 55, 60, 66,
73, 80, 88, 97, 107, 118, 130, 143,
157, 173, 190, 209, 230, 253, 279, 307,
337, 371, 408, 449, 494, 544, 598, 658,
724, 796, 876, 963, 1060, 1166, 1282, 1411,
1552, 1707, 1878, 2066, 2272, 2499, 2749, 3024,
3327, 3660, 4026, 4428, 4871, 5358, 5894, 6484,
7132, 7845, 8630, 9493,10442,11487,12635,13899,
15289,16818,18500,20350,22385,24623,27086,29794,
32767
};
int16 Ima_ADPCMStream::decodeIMA(byte code, int channel) {
int32 E = (2 * (code & 0x7) + 1) * _imaTable[_status.ima_ch[channel].stepIndex] / 8;
int32 diff = (code & 0x08) ? -E : E;
int32 samp = CLIP<int32>(_status.ima_ch[channel].last + diff, -32768, 32767);
_status.ima_ch[channel].last = samp;
_status.ima_ch[channel].stepIndex += _stepAdjustTable[code];
_status.ima_ch[channel].stepIndex = CLIP<int32>(_status.ima_ch[channel].stepIndex, 0, ARRAYSIZE(_imaTable) - 1);
return samp;
}
SeekableAudioStream *makeADPCMStream(Common::SeekableReadStream *stream, DisposeAfterUse::Flag disposeAfterUse, uint32 size, ADPCMType type, int rate, int channels, uint32 blockAlign) {
// If size is 0, report the entire size of the stream
if (!size)
size = stream->size();
switch (type) {
case kADPCMOki:
return new Oki_ADPCMStream(stream, disposeAfterUse, size, rate, channels, blockAlign);
case kADPCMMSIma:
return new MSIma_ADPCMStream(stream, disposeAfterUse, size, rate, channels, blockAlign);
case kADPCMMS:
return new MS_ADPCMStream(stream, disposeAfterUse, size, rate, channels, blockAlign);
case kADPCMDVI:
return new DVI_ADPCMStream(stream, disposeAfterUse, size, rate, channels, blockAlign);
case kADPCMApple:
return new Apple_ADPCMStream(stream, disposeAfterUse, size, rate, channels, blockAlign);
case kADPCMDK3:
return new DK3_ADPCMStream(stream, disposeAfterUse, size, rate, channels, blockAlign);
default:
error("Unsupported ADPCM encoding");
break;
}
}
class PacketizedADPCMStream : public StatelessPacketizedAudioStream {
public:
PacketizedADPCMStream(ADPCMType type, int rate, int channels, uint32 blockAlign) :
StatelessPacketizedAudioStream(rate, channels), _type(type), _blockAlign(blockAlign) {}
protected:
AudioStream *makeStream(Common::SeekableReadStream *data);
private:
ADPCMType _type;
uint32 _blockAlign;
};
AudioStream *PacketizedADPCMStream::makeStream(Common::SeekableReadStream *data) {
return makeADPCMStream(data, DisposeAfterUse::YES, data->size(), _type, getRate(), getChannels(), _blockAlign);
}
PacketizedAudioStream *makePacketizedADPCMStream(ADPCMType type, int rate, int channels, uint32 blockAlign) {
// Filter out types we can't support (they're not fully stateless)
switch (type) {
case kADPCMOki:
case kADPCMDVI:
return 0;
default:
break;
}
return new PacketizedADPCMStream(type, rate, channels, blockAlign);
}
} // End of namespace Audio
|