alac

# # Original C(++) version by Apple, http://alac.macosforge.org # # Javascript port by Jens Nockert and Devon Govett of OFMLabs, https://github.com/ofmlabs/alac.js # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # AV = require 'av' Aglib = require './ag_dec' Dplib = require './dp_dec' Matrixlib = require './matrix_dec' class ALACDecoder extends AV.Decoder AV.Decoder.register('alac', ALACDecoder) ID_SCE = 0 # Single Channel Element ID_CPE = 1 # Channel Pair Element ID_CCE = 2 # Coupling Channel Element ID_LFE = 3 # LFE Channel Element ID_DSE = 4 # not yet supported ID_PCE = 5 ID_FIL = 6 ID_END = 7 setCookie: (cookie) -> data = AV.Stream.fromBuffer(cookie) # For historical reasons the decoder needs to be resilient to magic cookies vended by older encoders. # There may be additional data encapsulating the ALACSpecificConfig. # This would consist of format ('frma') and 'alac' atoms which precede the ALACSpecificConfig. # See ALACMagicCookieDescription.txt in the original Apple decoder for additional documentation # concerning the 'magic cookie' # skip format ('frma') atom if present if data.peekString(4, 4) is 'frma' data.advance(12) # skip 'alac' atom header if present if data.peekString(4, 4) is 'alac' data.advance(12) # read the ALACSpecificConfig @config = frameLength: data.readUInt32() compatibleVersion: data.readUInt8() bitDepth: data.readUInt8() pb: data.readUInt8() mb: data.readUInt8() kb: data.readUInt8() numChannels: data.readUInt8() maxRun: data.readUInt16() maxFrameBytes: data.readUInt32() avgBitRate: data.readUInt32() sampleRate: data.readUInt32() # CAF files don't encode the bitsPerChannel @format.bitsPerChannel ||= @config.bitDepth # allocate mix buffers @mixBuffers = [ new Int32Array(@config.frameLength) # left channel new Int32Array(@config.frameLength) # right channel ] # allocate dynamic predictor buffer predictorBuffer = new ArrayBuffer(@config.frameLength * 4) @predictor = new Int32Array(predictorBuffer) # "shift off" buffer shares memory with predictor buffer @shiftBuffer = new Int16Array(predictorBuffer) readChunk: (data) -> return unless @stream.available(4) data = @bitstream samples = @config.frameLength numChannels = @config.numChannels channelIndex = 0 output = new ArrayBuffer(samples * numChannels * @config.bitDepth / 8) end = false while not end # read element tag tag = data.read(3) switch tag when ID_SCE, ID_LFE, ID_CPE channels = if tag is ID_CPE then 2 else 1 # if decoding this would take us over the max channel limit, bail if channelIndex + channels > numChannels throw new Error 'Too many channels!' # no idea what this is for... doesn't seem used anywhere elementInstanceTag = data.read(4) # read the 12 unused header bits unused = data.read(12) unless unused is 0 throw new Error 'Unused part of header does not contain 0, it should' # read the 1-bit "partial frame" flag, 2-bit "shift-off" flag & 1-bit "escape" flag partialFrame = data.read(1) bytesShifted = data.read(2) escapeFlag = data.read(1) if bytesShifted is 3 throw new Error "Bytes are shifted by 3, they shouldn't be" # check for partial frame to override requested samples if partialFrame samples = data.read(32) if escapeFlag is 0 shift = bytesShifted * 8 chanBits = @config.bitDepth - shift + channels - 1 # compressed frame, read rest of parameters mixBits = data.read(8) mixRes = data.read(8) mode = [] denShift = [] pbFactor = [] num = [] coefs = [] for ch in [0...channels] by 1 mode[ch] = data.read(4) denShift[ch] = data.read(4) pbFactor[ch] = data.read(3) num[ch] = data.read(5) table = coefs[ch] = new Int16Array(32) for i in [0...num[ch]] by 1 table[i] = data.read(16) # if shift active, skip the the shift buffer but remember where it starts if bytesShifted shiftbits = data.copy() data.advance(shift * channels * samples) # decompress and run predictors {mb, pb, kb, maxRun} = @config for ch in [0...channels] by 1 params = Aglib.ag_params(mb, (pb * pbFactor[ch]) / 4, kb, samples, samples, maxRun) status = Aglib.dyn_decomp(params, data, @predictor, samples, chanBits) unless status throw new Error 'Error in Aglib.dyn_decomp' if mode[ch] is 0 Dplib.unpc_block(@predictor, @mixBuffers[ch], samples, coefs[ch], num[ch], chanBits, denShift[ch]) else # the special "numActive == 31" mode can be done in-place Dplib.unpc_block(@predictor, @predictor, samples, null, 31, chanBits, 0) Dplib.unpc_block(@predictor, @mixBuffers[ch], samples, coefs[ch], num[ch], chanBits, denShift[ch]) else # uncompressed frame, copy data into the mix buffer to use common output code chanBits = @config.bitDepth shift = 32 - chanBits for i in [0...samples] by 1 for ch in [0...channels] by 1 val = (data.read(chanBits) << shift) >> shift @mixBuffers[ch][i] = val mixBits = mixRes = 0 bytesShifted = 0 # now read the shifted values into the shift buffer if bytesShifted shift = bytesShifted * 8 for i in [0...samples * channels] by 1 @shiftBuffer[i] = shiftbits.read(shift) # un-mix the data and convert to output format # - note that mixRes = 0 means just interleave so we use that path for uncompressed frames switch @config.bitDepth when 16 out16 = new Int16Array(output, channelIndex) if channels is 2 Matrixlib.unmix16(@mixBuffers[0], @mixBuffers[1], out16, numChannels, samples, mixBits, mixRes) else j = 0 buf = @mixBuffers[0] for i in [0...samples] by 1 out16[j] = buf[i] j += numChannels else throw new Error 'Only supports 16-bit samples right now' channelIndex += channels when ID_CCE, ID_PCE throw new Error "Unsupported element: #{tag}" when ID_DSE # the tag associates this data stream element with a given audio element elementInstanceTag = data.read(4) dataByteAlignFlag = data.read(1) # 8-bit count or (8-bit + 8-bit count) if 8-bit count == 255 count = data.read(8) if count is 255 count += data.read(8) # the align flag means the bitstream should be byte-aligned before reading the following data bytes if dataByteAlignFlag data.align() # skip the data bytes data.advance(count * 8) unless data.pos < data.length throw new Error 'buffer overrun' when ID_FIL # 4-bit count or (4-bit + 8-bit count) if 4-bit count == 15 # - plus this weird -1 thing I still don't fully understand count = data.read(4) if count is 15 count += data.read(8) - 1 data.advance(count * 8) unless data.pos < data.length throw new Error 'buffer overrun' when ID_END data.align() end = true else throw new Error "Unknown element: #{tag}" if channelIndex > numChannels throw new Error 'Channel index too large.' return new Int16Array(output) module.exports = ALACDecoder