Some more updates.

Started to enter huffman tables.
General reorganisation as xine_decoder.c was getting too big.

CVS patchset: 5245
CVS date: 2003/08/05 11:30:56

1 files changed, 849 insertions, 0 deletions

diff --git a/src/libdts/decoder.c b/src/libdts/decoder.c
new file mode 100644
index 000000000..724d61927
--- /dev/null
+++ b/src/libdts/decoder.c
@@ -0,0 +1,849 @@
+/*
+ * Copyright (C) 2000-2003 the xine project
+ *
+ * This file is part of xine, a unix video player.
+ *
+ * xine 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.
+ *
+ * xine 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., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA
+ *
+ * $Id: decoder.c,v 1.1 2003/08/05 11:30:56 jcdutton Exp $
+ *
+ * 04-08-2003 DTS software decode (C) James Courtier-Dutton
+ *
+ */
+
+#ifndef __sun
+/* required for swab() */
+#define _XOPEN_SOURCE 500
+#endif
+
+#include <stdlib.h>
+#include <unistd.h>
+#include <string.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <netinet/in.h> /* ntohs */
+#include <assert.h>
+
+#include "xine_internal.h"
+#include "xineutils.h"
+#include "audio_out.h"
+#include "buffer.h"
+
+#include "dts_debug.h"
+#include "decoder.h"
+#include "decoder_internal.h"
+#include "print_info.h"
+
+#ifdef ENABLE_DTS_PARSE
+
+typedef struct {
+  uint8_t *start;
+  uint32_t byte_position;
+  uint32_t bit_position;
+  uint8_t byte;
+} getbits_state_t;
+
+static float AdjTable[] = {
+  1.0000,
+  1.1250,
+  1.2500,
+  1.4375
+};
+
+#include "huffman_tables.h"
+
+static int32_t getbits_init(getbits_state_t *state, uint8_t *start) {
+  if ((state == NULL) || (start == NULL)) return -1;
+  state->start = start;
+  state->bit_position = 0;
+  state->byte_position = 0;
+  state->byte = start[0];
+  return 0;
+}
+/* Non-optimized getbits. */
+/* This can easily be optimized for particular platforms. */
+static uint32_t getbits(getbits_state_t *state, uint32_t number_of_bits) {
+  uint32_t result=0;
+  uint8_t byte=0;
+  if (number_of_bits > 32) {
+    printf("Number of bits > 32 in getbits\n");
+    assert(0);
+  }
+
+  if ((state->bit_position) > 0) {  /* Last getbits left us in the middle of a byte. */
+    if (number_of_bits > (8-state->bit_position)) { /* this getbits will span 2 or more bytes. */
+      byte = state->byte;
+      byte = byte >> (state->bit_position);
+      result = byte;
+      number_of_bits -= (8-state->bit_position);
+      state->bit_position = 0;
+      state->byte_position++;
+      state->byte = state->start[state->byte_position];
+    } else {
+      byte=state->byte;
+      state->byte = state->byte << number_of_bits;
+      byte = byte >> (8 - number_of_bits);
+      result = byte;
+      state->bit_position += number_of_bits; /* Here it is impossible for bit_position > 8 */
+      if (state->bit_position == 8) {
+        state->bit_position = 0;
+        state->byte_position++;
+        state->byte = state->start[state->byte_position];
+      }
+      number_of_bits = 0;
+    }
+  }
+  if ((state->bit_position) == 0)
+    while (number_of_bits > 7) {
+      result = (result << 8) + state->byte;
+      state->byte_position++;
+      state->byte = state->start[state->byte_position];
+      number_of_bits -= 8;
+    }
+    if (number_of_bits > 0) { /* number_of_bits < 8 */
+      byte = state->byte;
+      state->byte = state->byte << number_of_bits;
+      state->bit_position += number_of_bits; /* Here it is impossible for bit_position > 7 */
+      if (state->bit_position > 7) printf ("bit_pos2 too large: %d\n",state->bit_position);
+      byte = byte >> (8 - number_of_bits);
+      result = (result << number_of_bits) + byte;
+      number_of_bits = 0;
+    }
+
+  return result;
+}
+
+static int32_t huff_lookup(getbits_state_t *state, int32_t HuffTable[][2] ) {
+  int32_t n=1;
+  int32_t bit;
+
+  {
+    bit = getbits(state, 1);
+    n = HuffTable[n][bit];
+  } while (n > 0);
+  /* printf("returning %d\n", n + HuffTable[0][0]); */
+  return n + HuffTable[0][0];
+}
+
+
+static int32_t qscales(int32_t nQSelect, getbits_state_t *state, int32_t *nScale) {
+/* FIXME: IMPLEMENT */
+return 0;
+}
+
+/* Used by dts.wav files, only 14 bits of the 16 possible are used in the CD. */
+static void squash14to16(uint8_t *buf_from, uint8_t *buf_to, uint32_t number_of_bytes) {
+  int32_t from;
+  int32_t to=0;
+  uint16_t sample1;
+  uint16_t sample2;
+  uint16_t sample3;
+  uint16_t sample4;
+  uint16_t sample16bit;
+  /* This should convert the 14bit sync word into a 16bit one. */  
+  printf("libdts: squashing %d bytes.\n", number_of_bytes);
+  for(from=0;from<number_of_bytes;from+=8) {
+    sample1 = buf_from[from+0] | buf_from[from+1] << 8;
+    sample1 = (sample1 & 0x1fff) | ((sample1 & 0x8000) >> 2);
+    sample2 = buf_from[from+2] | buf_from[from+3] << 8;
+    sample2 = (sample2 & 0x1fff) | ((sample2 & 0x8000) >> 2);
+    sample16bit = (sample1 << 2) | (sample2 >> 12);
+    buf_to[to++] = sample16bit >> 8; /* Add some swabbing in as well */
+    buf_to[to++] = sample16bit & 0xff;
+    sample3 = buf_from[from+4] | buf_from[from+5] << 8;
+    sample3 = (sample3 & 0x1fff) | ((sample3 & 0x8000) >> 2);
+    sample16bit = ((sample2 & 0xfff) << 4) | (sample3 >> 10);
+    buf_to[to++] = sample16bit >> 8; /* Add some swabbing in as well */
+    buf_to[to++] = sample16bit & 0xff;
+    sample4 = buf_from[from+6] | buf_from[from+7] << 8;
+    sample4 = (sample4 & 0x1fff) | ((sample4 & 0x8000) >> 2);
+    sample16bit = ((sample3 & 0x3ff) << 6) | (sample4 >> 8);
+    buf_to[to++] = sample16bit >> 8; /* Add some swabbing in as well */
+    buf_to[to++] = sample16bit & 0xff;
+    buf_to[to++] = sample4 & 0xff;
+  }
+
+}
+
+#if 0
+/* FIXME: Make this re-entrant */
+static void InverseADPCM(void) {
+/*
+ * NumADPCMCoeff =4, the number of ADPCM coefficients.
+ * raADPCMcoeff[] are the ADPCM coefficients extracted
+ * from the bit stream.
+ * raSample[NumADPCMCoeff], ..., raSample[-1] are the
+ * history from last subframe or subsubframe. It must
+ * updated each time before reverse ADPCM is run for a
+ * block of samples for each subband.
+ */
+for (m=0; m<nNumSample; m++)
+for (n=0; n<NumADPCMCoeff; n++)
+raSample[m] += raADPCMcoeff[n]*raSample[m-n-1];
+}
+#endif
+
+
+void dts_parse_data (dts_decoder_t *this, buf_element_t *buf) {
+  uint8_t        *data_in = (uint8_t *)buf->content;
+  getbits_state_t state;
+  decoder_data_t decoder_data;
+  decoder_data.sync_type=0;
+  decoder_data.header_crc_check_bytes=0;
+
+  int32_t        n, ch, i;
+  printf("libdts: buf->size = %d\n", buf->size);
+  printf("libdts: parse1: ");
+  for(i=0;i<16;i++) {
+    printf("%02x ",data_in[i]);
+  }
+  printf("\n");
+  
+  if ((data_in[0] == 0x7f) && 
+      (data_in[1] == 0xfe) &&
+      (data_in[2] == 0x80) &&
+      (data_in[3] == 0x01)) {
+    decoder_data.sync_type=1;
+  }
+  if (data_in[0] == 0xff &&
+      data_in[1] == 0x1f &&
+      data_in[2] == 0x00 &&
+      data_in[3] == 0xe8 &&
+      data_in[4] == 0xf1 &&    /* DTS standard document was wrong here! */
+      data_in[5] == 0x07 ) {   /* DTS standard document was wrong here! */
+    squash14to16(&data_in[0], &data_in[0], buf->size);
+    buf->size = buf->size - (buf->size / 8); /* size = size * 7 / 8; */
+    decoder_data.sync_type=2;
+  }
+  if (decoder_data.sync_type == 0) {
+    printf("libdts: DTS Sync bad\n");
+    return;
+  }
+  printf("libdts: DTS Sync OK. type=%d\n", decoder_data.sync_type);
+  printf("libdts: parse2: ");
+  for(i=0;i<16;i++) {
+    printf("%02x ",data_in[i]);
+  }
+  printf("\n");
+
+  getbits_init(&state, &data_in[4]);
+
+  /* B.2 Unpack Frame Header Routine */
+  /* Frame Type V FTYPE 1 bit */
+  decoder_data.frame_type = getbits(&state, 1); /* 1: Normal Frame, 2:Termination Frame */
+  /* Deficit Sample Count V SHORT 5 bits */
+  decoder_data.deficit_sample_count = getbits(&state, 5);
+  /* CRC Present Flag V CPF 1 bit */
+  decoder_data.crc_present_flag = getbits(&state, 1);
+  /* Number of PCM Sample Blocks V NBLKS 7 bits */
+  decoder_data.number_of_pcm_blocks = getbits(&state, 7);
+  /* Primary Frame Byte Size V FSIZE 14 bits */
+  decoder_data.primary_frame_byte_size = getbits(&state, 14);
+  /* Audio Channel Arrangement ACC AMODE 6 bits */
+  decoder_data.audio_channel_arrangement = getbits(&state, 6);
+  /* Core Audio Sampling Frequency ACC SFREQ 4 bits */
+  decoder_data.core_audio_sampling_frequency = getbits(&state, 4);
+  /* Transmission Bit Rate ACC RATE 5 bits */
+  decoder_data.transmission_bit_rate = getbits(&state, 5);
+  /* Embedded Down Mix Enabled V MIX 1 bit */
+  decoder_data.embedded_down_mix_enabled = getbits(&state, 1);
+  /* Embedded Dynamic Range Flag V DYNF 1 bit */
+  decoder_data.embedded_dynamic_range_flag = getbits(&state, 1);
+  /* Embedded Time Stamp Flag V TIMEF 1 bit */
+  decoder_data.embedded_time_stamp_flag = getbits(&state, 1);
+  /* Auxiliary Data Flag V AUXF 1 bit */
+  decoder_data.auxiliary_data_flag = getbits(&state, 1);
+  /* HDCD NV HDCD 1 bits */
+  decoder_data.hdcd = getbits(&state, 1);
+  /* Extension Audio Descriptor Flag ACC EXT_AUDIO_ID 3 bits */
+  decoder_data.extension_audio_descriptor_flag = getbits(&state, 3);
+  /* Extended Coding Flag ACC EXT_AUDIO 1 bit */
+  decoder_data.extended_coding_flag = getbits(&state, 1);
+  /* Audio Sync Word Insertion Flag ACC ASPF 1 bit */
+  decoder_data.audio_sync_word_insertion_flag = getbits(&state, 1);
+  /* Low Frequency Effects Flag V LFF 2 bits */
+  decoder_data.low_frequency_effects_flag = getbits(&state, 2);
+  /* Predictor History Flag Switch V HFLAG 1 bit */
+  decoder_data.predictor_history_flag_switch = getbits(&state, 1);
+  /* Header CRC Check Bytes V HCRC 16 bits */
+  if (decoder_data.crc_present_flag == 1) 
+    decoder_data.header_crc_check_bytes  = getbits(&state, 16);
+  /* Multirate Interpolator Switch NV FILTS 1 bit */
+  decoder_data.multirate_interpolator_switch = getbits(&state, 1);
+  /* Encoder Software Revision ACC/NV VERNUM 4 bits */
+  decoder_data.encoder_software_revision = getbits(&state, 4);
+  /* Copy History NV CHIST 2 bits */
+  decoder_data.copy_history = getbits(&state, 2);
+  /* Source PCM Resolution ACC/NV PCMR 3 bits */
+  decoder_data.source_pcm_resolution = getbits(&state, 3);
+  /* Front Sum/Difference Flag V SUMF 1 bit */
+  decoder_data.front_sum_difference_flag = getbits(&state, 1);
+  /* Surrounds Sum/Difference Flag V SUMS 1 bit */
+  decoder_data.surrounds_sum_difference_flag = getbits(&state, 1);
+  /* Dialog Normalisation Parameter/Unspecified V DIALNORM/UNSPEC 4 bits */
+  switch (decoder_data.encoder_software_revision) {
+  case 6:
+    decoder_data.dialog_normalisation_unspecified = 0;
+    decoder_data.dialog_normalisation_parameter = getbits(&state, 4);
+    decoder_data.dialog_normalisation_gain = - (16+decoder_data.dialog_normalisation_parameter);
+    break;
+  case 7:
+    decoder_data.dialog_normalisation_unspecified = 0;
+    decoder_data.dialog_normalisation_parameter = getbits(&state, 4);
+    decoder_data.dialog_normalisation_gain = - (decoder_data.dialog_normalisation_parameter);
+    break;
+  default:
+    decoder_data.dialog_normalisation_unspecified = getbits(&state, 4);
+    decoder_data.dialog_normalisation_gain = decoder_data.dialog_normalisation_parameter = 0;
+    break;
+  }
+
+  /* B.3 Audio Decoding */
+  /* B.3.1 Primary Audio Coding Header */
+
+  /* Number of Subframes V SUBFS 4 bits */
+  decoder_data.number_of_subframes = getbits(&state, 4) + 1 ;
+  /* Number of Primary Audio Channels V PCHS 3 bits */
+  decoder_data.number_of_primary_audio_channels = getbits(&state, 3) + 1 ;
+  /* Subband Activity Count V SUBS 5 bits per channel */
+  for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++) {
+    decoder_data.subband_activity_count[ch] = getbits(&state, 5) + 2 ;
+  }
+  /* High Frequency VQ Start Subband V VQSUB 5 bits per channel */
+  for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++) {
+    decoder_data.high_frequency_VQ_start_subband[ch] = getbits(&state, 5) + 1 ;
+  }
+  /* Joint Intensity Coding Index V JOINX 3 bits per channel */
+  for (n=0; ch<decoder_data.number_of_primary_audio_channels; ch++) {
+    decoder_data.joint_intensity_coding_index[ch] = getbits(&state, 3) ;
+  }
+  /* Transient Mode Code Book V THUFF 2 bits per channel */
+  for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++) {
+    decoder_data.transient_mode_code_book[ch] = getbits(&state, 2) ;
+  }
+  /* Scale Factor Code Book V SHUFF 3 bits per channel */
+  for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++) {
+    decoder_data.scales_factor_code_book[ch] = getbits(&state, 3) ;
+  }
+  /* Bit Allocation Quantizer Select BHUFF V 3 bits per channel */
+  for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++) {
+    decoder_data.bit_allocation_quantizer_select[ch] = getbits(&state, 3) ;
+  }
+  /* Quantization Index Codebook Select V SEL variable bits */
+  /* ABITS=1: */
+  n=0;
+  for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++)
+    decoder_data.quantization_index_codebook_select[ch][n] = getbits(&state, 1);
+  /* ABITS = 2 to 5: */
+  for (n=1; n<5; n++)
+    for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++)
+      decoder_data.quantization_index_codebook_select[ch][n] = getbits(&state, 2);
+  /* ABITS = 6 to 10: */
+  for (n=5; n<10; n++)
+    for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++)
+      decoder_data.quantization_index_codebook_select[ch][n] = getbits(&state, 3);
+  /* ABITS = 11 to 26: */
+  for (n=10; n<26; n++)
+    for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++)
+      decoder_data.quantization_index_codebook_select[ch][n] = 0; /* Not transmitted, set to zero. */
+
+  /* Scale Factor Adjustment Index V ADJ 2 bits per occasion */
+  /* ABITS = 1: */
+  n = 0;
+  for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++) {
+    int32_t adj;
+    if ( decoder_data.quantization_index_codebook_select[ch][n] == 0 ) { /* Transmitted only if quantization_index_codebook_select=0 (Huffman code used) */
+      /* Extract ADJ index */
+      adj = getbits(&state, 2);
+      /* Look up ADJ table */
+      decoder_data.scale_factor_adjustment_index[ch][n] = AdjTable[adj];
+    }
+  }
+  /* ABITS = 2 to 5: */
+  for (n=1; n<5; n++){
+    for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++){
+      int32_t adj;
+      if ( decoder_data.quantization_index_codebook_select[ch][n] < 3 ) { /* Transmitted only when quantization_index_codebook_select<3 */
+        /* Extract ADJ index */
+        adj = getbits(&state, 2);
+        /* Look up ADJ table */
+        decoder_data.scale_factor_adjustment_index[ch][n] = AdjTable[adj];
+      }
+    }
+  }
+  /* ABITS = 6 to 10: */
+  for (n=5; n<10; n++){
+    for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++){
+      int32_t adj;
+      if ( decoder_data.quantization_index_codebook_select[ch][n] < 7 ) { /* Transmitted only when quantization_index_codebook_select<7 */
+        /* Extract ADJ index */
+        adj = getbits(&state, 2);
+        /* Look up ADJ table */
+        decoder_data.scale_factor_adjustment_index[ch][n] = AdjTable[adj];
+      }
+    }
+  }
+
+  if (decoder_data.crc_present_flag == 1) { /* Present only if CPF=1. */
+    decoder_data.audio_header_crc_check_word = getbits(&state, 16);
+  }
+
+/* B.3.2          Unpack Subframes */
+/* B.3.2.1 Primary Audio Coding Side Information */
+
+/* Subsubframe Count V SSC 2 bit */
+  decoder_data.subsubframe_count = getbits(&state, 2) + 1;
+/* Partial Subsubframe Sample Count V PSC 3 bit */
+  decoder_data.partial_subsubframe_sample_count = getbits(&state, 3);
+/* Prediction Mode V PMODE 1 bit per subband */
+  for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++) {
+    for (n=0; n<decoder_data.subband_activity_count[ch]; n++) {
+      decoder_data.prediction_mode[ch][n] = getbits(&state, 1);
+    }
+  }
+
+/* Prediction Coefficients VQ Address V PVQ 12 bits per occurrence */
+  for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++) {
+    for (n=0; n<decoder_data.subband_activity_count[ch]; n++) {
+      decoder_data.PVQIndex[ch][n] = 0;
+      if ( decoder_data.prediction_mode[ch][n]>0 ) { /* Transmitted only when ADPCM active */
+        /* Extract the VQindex */
+        decoder_data.nVQIndex = getbits(&state,12);
+        /* Look up the VQ table for prediction coefficients. */
+        /* FIXME: How to implement LookUp? */
+        decoder_data.PVQIndex[ch][n] = decoder_data.nVQIndex;
+        /* FIXME: We don't have the ADPCMCoeff table. */
+        /* ADPCMCoeffVQ.LookUp(nVQIndex, PVQ[ch][n]);*/ /*  4 coefficients  FIXME: Need to work out what this does. */
+      }
+    }
+  }
+
+
+  /* Bit Allocation Index V ABITS variable bits */
+  /* FIXME: No getbits here InverseQ does the getbits */
+  for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++) {
+    /* Bit Allocation Quantizer Select tells which codebook was used */
+    decoder_data.nQSelect = decoder_data.bit_allocation_quantizer_select[ch]; 
+    /* Use this codebook to decode the bit stream for bit_allocation_index[ch][n] */
+    for (n=0; n<decoder_data.high_frequency_VQ_start_subband[ch]; n++) {
+      /* Not for VQ encoded subbands. */
+      /* FIXME: What is Inverse Quantization(InverseQ) ? */
+      /* This basically selects a huffman table number nQSelect, */
+      /* and uses it to read a variable amount of bits and does a huffman search to find the value. */
+      /* FIXME: Need to implement InverseQ, so we can uncomment this line */
+      if (decoder_data.nQSelect == 6) {
+          decoder_data.bit_allocation_index[ch][n] = getbits(&state,5);
+      } else {
+        XINE_ASSERT(0, "bit_alloc parse failed, (nQSelect != 6) not implemented yet.");
+      }
+
+      /*QABITS.ppQ[nQSelect]->InverseQ(&state, bit_allocation_index[ch][n]); */
+    }
+  }
+
+  /* Transition Mode V TMODE variable bits */
+
+  /* Always assume no transition unless told */
+  for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++){
+    for (n=0; n<decoder_data.subband_activity_count[ch]; n++) {
+      decoder_data.transition_mode[ch][n] = 0;
+    } 
+    /* Decode transition_mode[ch][n] */
+    if ( decoder_data.subsubframe_count>1 ) {
+      /* Transient possible only if more than one subsubframe. */
+      for (ch=0; ch<decoder_data.number_of_primary_audio_channels; ch++) {
+        /* transition_mode[ch][n] is encoded by a codebook indexed by transient_mode_code_book[ch] */
+        decoder_data.nQSelect = decoder_data.transient_mode_code_book[ch];
+        for (n=0; n<decoder_data.high_frequency_VQ_start_subband[ch]; n++) {
+          /* No VQ encoded subbands */
+          if ( decoder_data.bit_allocation_index[ch][n] >0 ) {
+            /* Present only if bits allocated */
+            /* Use codebook nQSelect to decode transition_mode from the bit stream */
+            /* FIXME: What is Inverse Quantization(InverseQ) ? */
+            if (decoder_data.nQSelect == 0) {
+            decoder_data.transition_mode[ch][n] = huff_lookup(&state, HuffA4);
+            } else {
+              XINE_ASSERT(0, "transition mod parse failed, (nQSelect != 0) not implemented yet.");
+            }
+
+            /* QTMODE.ppQ[nQSelect]->InverseQ(&state,transition_mode[ch][n]); */
+          } else {
+            decoder_data.transition_mode[ch][n] = 0;
+          }
+        }
+      }
+    }
+  }
+
+/* WORKING ON THIS BIT */
+
+
+#if 0
+  /* Scale Factors V SCALES variable bits */
+  for (ch=0; ch<number_of_primary_audio_channels; ch++) {
+    /* Clear scale_factors */
+    for (n=0; n<subband_activity_count[ch]; n++) {
+      scale_factors[ch][n][0] = 0;
+      scale_factors[ch][n][1] = 0;
+    }
+    /* scales_factor_code_book indicates which codebook was used to encode scale_factors */
+    nQSelect = scales_factor_code_book[ch];
+    /* Select the root square table (scale_factors were nonlinearly */
+    /* quantized). */
+    /* Assume nQSelect != 6 */
+    /* So RMS is always 6 bit. */
+    if ( nQSelect == 6 ) {
+      /* pScaleTable = &RMS7Bit;*/  /* 7-bit root square table */
+    } else {
+      /* pScaleTable = &RMS6Bit;*/ /* 6-bit root square table */
+    }
+    /*
+     * Clear accumulation (if Huffman code was used, the difference
+     * of scale_factors was encoded).
+     */
+    nScaleSum = 0;
+    /*
+     * Extract scale_factors for Subbands up to high_frequency_VQ_start_subband[ch]
+     */
+    for (n=0; n<high_frequency_VQ_start_subband[ch]; n++) {
+      if ( bit_allocation_index[ch][n] >0 ) { /* Not present if no bit allocated */
+        /*
+         * First scale factor
+         */
+        /* Use the (Huffman) code indicated by nQSelect to decode */
+        /* the quantization index of scale_factors from the bit stream */
+        /* FIXME: What is Inverse Quantization(InverseQ) ? */
+        qscales(nQSelect, &state, &nScale);
+        /* QSCALES.ppQ[nQSelect]->InverseQ(InputFrame, nScale); */
+        /* Take care of difference encoding */
+        if ( nQSelect < 5 ) { /* Huffman encoded, nScale is the difference */
+          nScaleSum += nScale; /* of the quantization indexes of scale_factors. */
+        } else { /* Otherwise, nScale is the quantization */
+          nScaleSum = nScale; /* level of scale_factors. */
+        }
+        /* Look up scale_factors from the root square table */
+        /* FIXME: How to implement LookUp? */
+        pScaleTable->LookUp(nScaleSum, scale_factors[ch][n][0])
+        /*
+         * Two scale factors transmitted if there is a transient
+         */
+        if (transition_mode[ch][n]>0) {
+          /* Use the (Huffman) code indicated by nQSelect to decode */
+          /* the quantization index of scale_factors from the bit stream */
+          /* FIXME: What is Inverse Quantization(InverseQ) ? */
+          QSCALES.ppQ[nQSelect]->InverseQ(InputFrame, nScale);
+          /* Take care of difference encoding */
+          if ( nQSelect < 5 ) /* Huffman encoded, nScale is the difference */
+            nScaleSum += nScale; /* of the quantization indexes of scale_factors. */
+          else /* Otherwise, nScale is the quantization */
+            nScaleSum = nScale; /* level of scale_factors. */
+          /* Look up scale_factors from the root square table */
+          /* FIXME: How to implement LookUp? */
+          pScaleTable->LookUp(nScaleSum, scale_factors[ch][n][1]);
+        }
+      }
+    }
+    /*
+     * High frequency VQ subbands
+     */
+    for (n=high_frequency_VQ_start_subband[ch]; n<subband_activity_count[ch]; n++) {
+      /* Use the code book indicated by nQSelect to decode */
+      /* the quantization index of scale_factors from the bit stream */
+      /* FIXME: What is Inverse Quantization(InverseQ) ? */
+      QSCALES.ppQ[nQSelect]->InverseQ(InputFrame, nScale);
+      /* Take care of difference encoding */
+      if ( nQSelect < 5 ) /* Huffman encoded, nScale is the difference */
+        nScaleSum += nScale; /* of the quantization indexes of scale_factors. */
+      else /* Otherwise, nScale is the quantization */
+        nScaleSum = nScale; /* level of scale_factors. */
+      /* Look up scale_factors from the root square table */
+      /* FIXME: How to implement LookUp? */
+      pScaleTable->LookUp(nScaleSum, scale_factors[ch][n][0])
+    }
+  }
+
+/* #if 0 */
+/* FIXME: ALL CODE BELOW HERE does not compile yet. */
+
+
+  /* Joint Subband Scale Factor Codebook Select V JOIN SHUFF 3 bits per channel */
+  for (ch=0; ch<number_of_primary_audio_channels; ch++)
+    if (joint_intensity_coding_index[ch]>0 ) /* Transmitted only if joint subband coding enabled. */
+      joint_subband_scale_factor_codebook_select[ch] = getbits(&state,3);
+
+  /* Scale Factors for Joint Subband Coding V JOIN SCALES variable bits */
+  int nSourceCh;
+  for (ch=0; ch<number_of_primary_audio_channels; ch++) {
+    if (joint_intensity_coding_index[ch]>0 ) { /* Only if joint subband coding enabled. */
+      nSourceCh = joint_intensity_coding_index[ch]-1; /* Get source channel. joint_intensity_coding_index counts */
+      /* channels as 1,2,3,4,5, so minus 1. */
+      nQSelect = joint_subband_scale_factor_codebook_select[ch]; /* Select code book. */
+      for (n=subband_activity_count[ch]; n<subband_activity_count[nSourceCh]; n++) {
+        /* Use the code book indicated by nQSelect to decode */
+        /* the quantization index of scale_factors_for_joint_subband_coding */
+        /* FIXME: What is Inverse Quantization(InverseQ) ? */
+        QSCALES.ppQ[nQSelect]->InverseQ(InputFrame, nJScale);
+        /* Bias by 64 */
+        nJScale = nJScale + 64;
+        /* Look up scale_factors_for_joint_subband_coding from the joint scale table */
+        /* FIXME: How to implement LookUp? */
+        JScaleTbl.LookUp(nJScale, scale_factors_for_joint_subband_coding[ch][n]);
+      }
+    }
+  }
+
+  /* Stereo Down-Mix Coefficients NV DOWN 7 bits per coefficient */
+  if ( (MIX!=0) && (number_of_primary_audio_channels>2) ) {
+    /* Extract down mix indexes */
+    for (ch=0; ch<number_of_primary_audio_channels; ch++) { /* Each primary channel */
+      stereo_down_mix_coefficients[ch][0] = getbits(&state,7);
+      stereo_down_mix_coefficients[ch][1] = getbits(&state,7);
+    }
+  }
+  /* Look up down mix coefficients */
+  for (n=0; n<subband_activity_count; n++) { /* Each active subbands */
+    LeftChannel = 0;
+    RightChannel = 0;
+    for (ch=0; ch<number_of_primary_audio_channels; ch++) { /* Each primary channels */
+      LeftChannel += stereo_down_mix_coefficients[ch][0]*Sample[Ch];
+      RightChannel += stereo_down_mix_coefficients[ch][1]*Sample[Ch];
+    }
+  }
+  /* Down mixing may also be performed on the PCM samples after the filterbank reconstruction. */
+
+  /* Dynamic Range Coefficient NV RANGE 8 bits */
+  if ( embedded_dynamic_range_flag != 0 ) {
+    nIndex = getbits(&state,8);
+    /* FIXME: How to implement LookUp? */
+    RANGEtbl.LookUp(nIndex,dynamic_range_coefficient);
+    /* The following range adjustment is to be performed */
+    /* after QMF reconstruction */
+    for (ch=0; ch<number_of_primary_audio_channels; ch++)
+      for (n=0; n<nNumSamples; n++)
+        AudioCh[ch].ReconstructedSamples[n] *= dynamic_range_coefficient;
+  }
+
+  /* Side Information CRC Check Word V SICRC 16 bits */
+  if ( CPF==1 ) /* Present only if CPF=1. */
+    SICRC = getbits(&state,16);
+
+  /* B.3.3 Primary Audio Data Arrays */
+
+  /* VQ Encoded High Frequency Subbands NV HFREQ 10 bits per applicable subbands */
+  for (ch=0; ch<number_of_primary_audio_channels; ch++) {
+    for (n=high_frequency_VQ_start_subband[ch]; n<subband_activity_count[ch]; n++) {
+      /* Extract the VQ address from the bit stream */
+      nVQIndex = getbits(&state,10);
+      /* Look up the VQ code book for 32 subband samples. */
+      /* FIXME: How to implement LookUp? */
+      HFreqVQ.LookUp(nVQIndex, VQ_encoded_high_frequency_subbands[ch][n])
+      /* Scale and take the samples */
+      Scale = (real)scale_factors[ch][n][0]; /* Get the scale factor */
+      for (m=0; m<subsubframe_count*8; m++, nSample++) {
+        aPrmCh[ch].aSubband[n].raSample[m] = rScale*VQ_encoded_high_frequency_subbands[ch][n][m];
+      }
+    }
+  }
+
+  /* Low Frequency Effect Data V LFE 8 bits per sample */
+  if ( low_frequency_effects_flag>0 ) { /* Present only if flagged by low_frequency_effects_flag */
+    /* extract low_frequency_effect_data samples from the bit stream */
+    for (n=0; n<2*low_frequency_effects_flag*subsubframe_count; n++) {
+      low_frequency_effect_data[n] = (signed int)(signed char)getbits(&state,8);
+      /* Use char to get sign extension because it */
+      /* is 8-bit 2's compliment. */
+      /* Extract scale factor index from the bit stream */
+    }
+    LFEscaleIndex = getbits(&state,8);
+    /* Look up the 7-bit root square quantization table */
+    /* FIXME: How to implement LookUp? */
+    pLFE_RMS->LookUp(LFEscaleIndex,nScale);
+    /* Account for the quantizer step size which is 0.035 */
+    rScale = nScale*0.035;
+    /* Get the actual low_frequency_effect_data samples */
+    for (n=0; n<2*low_frequency_effects_flag*subsubframe_count; n++) {
+      LFECh.rLFE[k] = low_frequency_effect_data[n]*rScale;
+    }
+    /* Interpolation low_frequency_effect_data samples */
+    LFECh.InterpolationFIR(low_frequency_effects_flag); /* low_frequency_effects_flag indicates which */
+    /* interpolation filter to use */
+  }
+
+  /* Audio Data V AUDIO variable bits */
+  /*
+   * Select quantization step size table
+   */
+  if ( RATE == 0x1f ) {
+    pStepSizeTable = &StepSizeLossLess; /* Lossless quantization */
+  } else {
+    pStepSizeTable = &StepSizeLossy; /* Lossy */
+  }
+  /*
+   * Unpack the subband samples
+   */
+  for (nSubSubFrame=0; nSubSubFrame<subsubframe_count; nSubSubFrame++) {
+    for (ch=0; ch<number_of_primary_audio_channels; ch++) {
+      for (n=0; n<high_frequency_VQ_start_subband[ch]; n++) { /* Not high frequency VQ subbands */
+      /*
+       * Select the mid-tread linear quantizer
+       */
+      nABITS = bit_allocation_index[ch][n]; /* Select the mid-tread quantizer */
+      pCQGroup = &pCQGroupAUDIO[nABITS-1];/* Select the group of */
+      /* code books corresponding to the */
+      /* the mid-tread linear quantizer. */
+      nNumQ = pCQGroupAUDIO[nABITS-1].nNumQ-1;/* Number of code */
+      /* books in this group */
+      /*
+       * Determine quantization index code book and its type
+       */
+      /* Select quantization index code book */
+      nSEL = quantization_index_codebook_select[ch][nABITS-1];
+      /* Determine its type */
+      nQType = 1; /* Assume Huffman type by default */
+      if ( nSEL==nNumQ ) { /* Not Huffman type */
+        if ( nABITS<=7 ) {
+          nQType = 3; /* Block code */
+        } else {
+          nQType = 2; /* No further encoding */
+        }
+      }
+      if ( nABITS==0 ) { /* No bits allocated */
+        nQType = 0;
+      }
+      /*
+       * Extract bits from the bit stream
+       * This retrieves 8 AUDIO values
+       */
+      switch ( nQType ) {
+        case 0: /* No bits allocated */
+          for (m=0; m<8; m++)
+            AUDIO[m] = 0;
+          break;
+        case 1: /* Huffman code */
+          for (m=0; m<8; m++)
+            /* FIXME: What is Inverse Quantization(InverseQ) ? */
+            pCQGroup->ppQ[nSEL]->InverseQ(InputFrame,AUDIO[m]);
+          break;
+        case 2: /* No further encoding */
+          for (m=0; m<8; m++) {
+            /* Extract quantization index from the bit stream */
+            /* FIXME: What is Inverse Quantization(InverseQ) ? */
+            pCQGroup->ppQ[nSEL]->InverseQ(InputFrame, nCode)
+            /* Take care of 2's compliment */
+            AUDIO[m] = pCQGroup->ppQ[nSEL]->SignExtension(nCode);
+          }
+          break;
+        case 3: /* Block code */
+          /* Block code is just 1 value with 4 samples derived from it.
+           * with each sample a digit from the number (using a base derived from nABITS via a table)
+           * E.g. nABITS = 10, base = 5 (Base value taken from table.)
+           * 1st sample = (value % 5) - (int(5/2); (Values between -2 and +2 )
+           * 2st sample = ((value / 5) % 5) - (int(5/2);
+           * 3rd sample = ((value / 25) % 5) - (int(5/2);
+           * 4th sample = ((value / 125) % 5) - (int(5/2);
+           * 
+           */
+          pCBQ = &pCBlockQ[nABITS-1]; /* Select block code book */
+          m = 0;
+          for (nBlock=0; nBlock<2; nBlock++) {
+            /* Extract the block code index from the bit stream */
+            /* FIXME: What is Inverse Quantization(InverseQ) ? */
+            pCQGroup->ppQ[nSEL]->InverseQ(InputFrame, nCode)
+            /* Look up 4 samples from the block code book */
+            /* FIXME: How to implement LookUp? */
+            pCBQ->LookUp(nCode,&AUDIO[m])
+            m += 4;
+          }
+          break;
+        default: /* Undefined */
+          printf("ERROR: Unknown AUDIO quantization index code book.");
+      }
+      /*
+       * Account for quantization step size and scale factor
+       */
+      /* Look up quantization step size */
+      nABITS = bit_allocation_index[ch][n];
+      /* FIXME: How to implement LookUp? */
+      pStepSizeTable->LookUp(nABITS, rStepSize);
+      /* Identify transient location */
+      nTmode = transition_mode[ch][n];
+      if ( nTmode == 0 ) /* No transient */
+        nTmode = subsubframe_count;
+      /* Determine proper scale factor */
+      if (nSubSubFrame<nTmode) /* Pre-transient */
+        rScale = rStepSize * scale_factors[ch][n][0]; /* Use first scale factor */
+      else /* After-transient */
+        rScale = rStepSize * scale_factors[ch][n][1]; /* Use second scale factor */
+      /* Adjustmemt of scale factor */
+      rScale *= scale_factor_adjustment_index[ch][quantization_index_codebook_select[ch][nABITS-1]]; /* scale_factor_adjustment_index[ ][ ] are assumed 1 */
+      /* unless changed by bit */
+      /* stream when quantization_index_codebook_select indicates */
+      /* Huffman code. */
+      /* Scale the samples */
+      nSample = 8*nSubSubFrame; /* Set sample index */
+      for (m=0; m<8; m++, nSample++)
+        aPrmCh[ch].aSubband[n].aSample[nSample] = rScale*AUDIO[m];
+      /*
+       * Inverse ADPCM
+       */
+      if ( PMODE[ch][n] != 0 ) /* Only when prediction mode is on. */
+        aPrmCh[ch].aSubband[n].InverseADPCM();
+      /*
+       * Check for DSYNC
+       */
+      if ( (nSubSubFrame==(subsubframe_count-1)) || (ASPF==1) ) {
+        DSYNC = getbits(&state,16);
+        if ( DSYNC != 0xffff )
+          printf("DSYNC error at end of subsubframe #%d", nSubSubFrame);
+      }
+    }
+  }
+/* B.3.4 Unpack Optional Information */
+/* TODO ^^^ */
+
+#endif
+/* CODE BELOW here does compile */
+
+  printf("getbits status: byte_pos = %d, bit_pos = %d\n", 
+          state.byte_position,
+          state.bit_position);
+#if 0
+  for(n=0;n<2016;n++) {
+    if((n % 32) == 0) printf("\n");
+    printf("%02X ",state.start[state.byte_position+n]);
+  }
+  printf("\n");
+#endif
+
+#if 0
+  if ((extension_audio_descriptor_flag == 0)
+     || (extension_audio_descriptor_flag == 3)) {
+    printf("libdts:trying extension...\n");
+    channel_extension_sync_word = getbits(&state, 32);
+    extension_primary_frame_byte_size = getbits(&state, 10); 
+    extension_channel_arrangement = getbits(&state, 4);
+  }
+#endif
+
+#if 0
+    extension_sync_word_SYNC96 = getbits(&state, 32);
+    extension_frame_byte_data_size_FSIZE96 = getbits(&state, 12);
+    revision_number = getbits(&state, 4);
+#endif
+dts_print_decoded_data(&decoder_data);
+}
+
+#endif