流媒体/Mpeg4/MP4

开发平台：
Visual C++

decode_i386.cpp：源码内容
							/**************************************************************************************
 *                                                                                    *
 * This application contains code from OpenDivX and is released as a "Larger Work"    *
 * under that license. Consistant with that license, this application is released     *
 * under the GNU General Public License.                                              *
 *                                                                                    *
 * The OpenDivX license can be found at: http://www.projectmayo.com/opendivx/docs.php *
 * The GPL can be found at: http://www.gnu.org/copyleft/gpl.html                      *
 *                                                                                    *
 * Authors: Damien Chavarria <roy204 at projectmayo.com>                              *
 *                                                                                    *
 **************************************************************************************/
 /* 
 * Mpeg Layer-1,2,3 audio decoder 
 * ------------------------------
 * copyright (c) 1995,1996,1997 by Michael Hipp, All rights reserved.
 *
 * slighlty optimized for machines without autoincrement/decrement.
 * The performance is highly compiler dependend. Maybe
 * the decode.c version for 'normal' processor may be faster
 * even for Intel processors.
 */
#include <stdlib.h>
#include <math.h>
#include <string.h>
#include "mpg123.h"
#include "mpglib.h"
extern struct mpstr *gmp;
 /* old WRITE_SAMPLE */
#define WRITE_SAMPLE(samples,sum,clip) 
  if( (sum) > 32767.0) { *(samples) = 0x7fff; (clip)++; } 
  else if( (sum) < -32768.0) { *(samples) = -0x8000; (clip)++; } 
  else { *(samples) = sum; }
int synth_1to1_mono(real *bandPtr,unsigned char *samples,int *pnt)
{
  short samples_tmp[64];
  short *tmp1 = samples_tmp;
  int i,ret;
  int pnt1 = 0;
  ret = synth_1to1(bandPtr,0,(unsigned char *) samples_tmp,&pnt1);
  samples += *pnt;
  for(i=0;i<32;i++) {
    *( (short *) samples) = *tmp1;
    samples += 2;
    tmp1 += 2;
  }
  *pnt += 64;
  return ret;
}
int synth_1to1(real *bandPtr,int channel,unsigned char *out,int *pnt)
{
  static const int step = 2;
  int bo;
  short *samples = (short *) (out + *pnt);
  real *b0,(*buf)[0x110];
  int clip = 0; 
  int bo1;
  bo = gmp->synth_bo;
  if(!channel) {
    bo--;
    bo &= 0xf;
    buf = gmp->synth_buffs[0];
  }
  else {
    samples++;
    buf = gmp->synth_buffs[1];
  }
  if(bo & 0x1) {
    b0 = buf[0];
    bo1 = bo;
    dct64(buf[1]+((bo+1)&0xf),buf[0]+bo,bandPtr);
  }
  else {
    b0 = buf[1];
    bo1 = bo+1;
    dct64(buf[0]+bo,buf[1]+bo+1,bandPtr);
  }
  gmp->synth_bo = bo;
  
  {
    register int j;
    real *window = decwin + 16 - bo1;
    for (j=16;j;j--,b0+=0x10,window+=0x20,samples+=step)
    {
      real sum;
      sum  = window[0x0] * b0[0x0];
      sum -= window[0x1] * b0[0x1];
      sum += window[0x2] * b0[0x2];
      sum -= window[0x3] * b0[0x3];
      sum += window[0x4] * b0[0x4];
      sum -= window[0x5] * b0[0x5];
      sum += window[0x6] * b0[0x6];
      sum -= window[0x7] * b0[0x7];
      sum += window[0x8] * b0[0x8];
      sum -= window[0x9] * b0[0x9];
      sum += window[0xA] * b0[0xA];
      sum -= window[0xB] * b0[0xB];
      sum += window[0xC] * b0[0xC];
      sum -= window[0xD] * b0[0xD];
      sum += window[0xE] * b0[0xE];
      sum -= window[0xF] * b0[0xF];
      WRITE_SAMPLE(samples,sum,clip);
    }
    {
      real sum;
      sum  = window[0x0] * b0[0x0];
      sum += window[0x2] * b0[0x2];
      sum += window[0x4] * b0[0x4];
      sum += window[0x6] * b0[0x6];
      sum += window[0x8] * b0[0x8];
      sum += window[0xA] * b0[0xA];
      sum += window[0xC] * b0[0xC];
      sum += window[0xE] * b0[0xE];
      WRITE_SAMPLE(samples,sum,clip);
      b0-=0x10,window-=0x20,samples+=step;
    }
    window += bo1<<1;
    for (j=15;j;j--,b0-=0x10,window-=0x20,samples+=step)
    {
      real sum;
      sum = -window[-0x1] * b0[0x0];
      sum -= window[-0x2] * b0[0x1];
      sum -= window[-0x3] * b0[0x2];
      sum -= window[-0x4] * b0[0x3];
      sum -= window[-0x5] * b0[0x4];
      sum -= window[-0x6] * b0[0x5];
      sum -= window[-0x7] * b0[0x6];
      sum -= window[-0x8] * b0[0x7];
      sum -= window[-0x9] * b0[0x8];
      sum -= window[-0xA] * b0[0x9];
      sum -= window[-0xB] * b0[0xA];
      sum -= window[-0xC] * b0[0xB];
      sum -= window[-0xD] * b0[0xC];
      sum -= window[-0xE] * b0[0xD];
      sum -= window[-0xF] * b0[0xE];
      sum -= window[-0x0] * b0[0xF];
      WRITE_SAMPLE(samples,sum,clip);
    }
  }
  *pnt += 128;
  return clip;
}

						