/*
 * Copyright (c) 2003 The FFmpeg Project.
 *
 * 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
 */

/*
 * How to use this decoder:
 * SVQ3 data is transported within Apple Quicktime files. Quicktime files
 * have stsd atoms to describe media trak properties. A stsd atom for a
 * video trak contains 1 or more ImageDescription atoms. These atoms begin
 * with the 4-byte length of the atom followed by the codec fourcc. Some
 * decoders need information in this atom to operate correctly. Such
 * is the case with SVQ3. In order to get the best use out of this decoder,
 * the calling app must make the SVQ3 ImageDescription atom available
 * via the AVCodecContext's extradata[_size] field:
 *
 * AVCodecContext.extradata = pointer to ImageDescription, first characters
 * are expected to be 'S', 'V', 'Q', and '3', NOT the 4-byte atom length
 * AVCodecContext.extradata_size = size of ImageDescription atom memory
 * buffer (which will be the same as the ImageDescription atom size field
 * from the QT file, minus 4 bytes since the length is missing)
 *
 * You will know you have these parameters passed correctly when the decoder
 * correctly decodes this file:
 *  ftp://ftp.mplayerhq.hu/MPlayer/samples/V-codecs/SVQ3/Vertical400kbit.sorenson3.mov
 */

/**
 * @file svq3.c
 * svq3 decoder.
 */

#include <stdio.h>
#include <stdlib.h>
#include <memory.h>
#include <assert.h>
#include "mem.h"
#include "common.h"
#include "bswap.h"
#include "intreadwrite.h"
#include "eval.h"
#include "rational.h"
#include "log.h"
#include "avutil.h"
#include "cabac.h"
#include "avcodec.h"
#include "dsputil.h"
#include "ratecontrol.h"
#include "parser.h"
#include "bitstream.h"
#include "rl.h"
#include "mpegvideo.h"
#include "h264pred.h"
#include "h264.h"
#include "h264data.h"
#include "golomb.h"
#include "rectangle.h"

#define FULLPEL_MODE  1
#define HALFPEL_MODE  2
#define THIRDPEL_MODE 3
#define PREDICT_MODE  4

/* dual scan (from some older h264 draft)
 o-->o-->o   o
         |  /|
 o   o   o / o
 | / |   |/  |
 o   o   o   o
   /
 o-->o-->o-->o
*/
static const uint8_t svq3_scan[16]={
 0+0*4, 1+0*4, 2+0*4, 2+1*4,
 2+2*4, 3+0*4, 3+1*4, 3+2*4,
 0+1*4, 0+2*4, 1+1*4, 1+2*4,
 0+3*4, 1+3*4, 2+3*4, 3+3*4,
};

static const uint8_t svq3_pred_0[25][2] = {
  { 0, 0 },
  { 1, 0 }, { 0, 1 },
  { 0, 2 }, { 1, 1 }, { 2, 0 },
  { 3, 0 }, { 2, 1 }, { 1, 2 }, { 0, 3 },
  { 0, 4 }, { 1, 3 }, { 2, 2 }, { 3, 1 }, { 4, 0 },
  { 4, 1 }, { 3, 2 }, { 2, 3 }, { 1, 4 },
  { 2, 4 }, { 3, 3 }, { 4, 2 },
  { 4, 3 }, { 3, 4 },
  { 4, 4 }
};

static const int8_t svq3_pred_1[6][6][5] = {
  { { 2,-1,-1,-1,-1 }, { 2, 1,-1,-1,-1 }, { 1, 2,-1,-1,-1 },
    { 2, 1,-1,-1,-1 }, { 1, 2,-1,-1,-1 }, { 1, 2,-1,-1,-1 } },
  { { 0, 2,-1,-1,-1 }, { 0, 2, 1, 4, 3 }, { 0, 1, 2, 4, 3 },
    { 0, 2, 1, 4, 3 }, { 2, 0, 1, 3, 4 }, { 0, 4, 2, 1, 3 } },
  { { 2, 0,-1,-1,-1 }, { 2, 1, 0, 4, 3 }, { 1, 2, 4, 0, 3 },
    { 2, 1, 0, 4, 3 }, { 2, 1, 4, 3, 0 }, { 1, 2, 4, 0, 3 } },
  { { 2, 0,-1,-1,-1 }, { 2, 0, 1, 4, 3 }, { 1, 2, 0, 4, 3 },
    { 2, 1, 0, 4, 3 }, { 2, 1, 3, 4, 0 }, { 2, 4, 1, 0, 3 } },
  { { 0, 2,-1,-1,-1 }, { 0, 2, 1, 3, 4 }, { 1, 2, 3, 0, 4 },
    { 2, 0, 1, 3, 4 }, { 2, 1, 3, 0, 4 }, { 2, 0, 4, 3, 1 } },
  { { 0, 2,-1,-1,-1 }, { 0, 2, 4, 1, 3 }, { 1, 4, 2, 0, 3 },
    { 4, 2, 0, 1, 3 }, { 2, 0, 1, 4, 3 }, { 4, 2, 1, 0, 3 } },
};

static const struct { uint8_t run; uint8_t level; } svq3_dct_tables[2][16] = {
  { { 0, 0 }, { 0, 1 }, { 1, 1 }, { 2, 1 }, { 0, 2 }, { 3, 1 }, { 4, 1 }, { 5, 1 },
    { 0, 3 }, { 1, 2 }, { 2, 2 }, { 6, 1 }, { 7, 1 }, { 8, 1 }, { 9, 1 }, { 0, 4 } },
  { { 0, 0 }, { 0, 1 }, { 1, 1 }, { 0, 2 }, { 2, 1 }, { 0, 3 }, { 0, 4 }, { 0, 5 },
    { 3, 1 }, { 4, 1 }, { 1, 2 }, { 1, 3 }, { 0, 6 }, { 0, 7 }, { 0, 8 }, { 0, 9 } }
};

static const uint32_t svq3_dequant_coeff[32] = {
   3881,  4351,  4890,  5481,  6154,  6914,  7761,  8718,
   9781, 10987, 12339, 13828, 15523, 17435, 19561, 21873,
  24552, 27656, 30847, 34870, 38807, 43747, 49103, 54683,
  61694, 68745, 77615, 89113,100253,109366,126635,141533
};


void svq3_luma_dc_dequant_idct_c(DCTELEM *block, int qp){
    const int qmul= svq3_dequant_coeff[qp];
#define stride 16
    int i;
    int temp[16];
    static const int x_offset[4]={0, 1*stride, 4* stride,  5*stride};
    static const int y_offset[4]={0, 2*stride, 8* stride, 10*stride};

    for(i=0; i<4; i++){
        const int offset= y_offset[i];
        const int z0= 13*(block[offset+stride*0] +    block[offset+stride*4]);
        const int z1= 13*(block[offset+stride*0] -    block[offset+stride*4]);
        const int z2=  7* block[offset+stride*1] - 17*block[offset+stride*5];
        const int z3= 17* block[offset+stride*1] +  7*block[offset+stride*5];

        temp[4*i+0]= z0+z3;
        temp[4*i+1]= z1+z2;
        temp[4*i+2]= z1-z2;
        temp[4*i+3]= z0-z3;
    }

    for(i=0; i<4; i++){
        const int offset= x_offset[i];
        const int z0= 13*(temp[4*0+i] +    temp[4*2+i]);
        const int z1= 13*(temp[4*0+i] -    temp[4*2+i]);
        const int z2=  7* temp[4*1+i] - 17*temp[4*3+i];
        const int z3= 17* temp[4*1+i] +  7*temp[4*3+i];

        block[stride*0 +offset]= ((z0 + z3)*qmul + 0x80000)>>20;
        block[stride*2 +offset]= ((z1 + z2)*qmul + 0x80000)>>20;
        block[stride*8 +offset]= ((z1 - z2)*qmul + 0x80000)>>20;
        block[stride*10+offset]= ((z0 - z3)*qmul + 0x80000)>>20;
    }
}
#undef stride

void svq3_add_idct_c (uint8_t *dst, DCTELEM *block, int stride, int qp, int dc){
    const int qmul= svq3_dequant_coeff[qp];
    int i;
    uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;

    if (dc) {
        dc = 13*13*((dc == 1) ? 1538*block[0] : ((qmul*(block[0] >> 3)) / 2));
        block[0] = 0;
    }

    for (i=0; i < 4; i++) {
        const int z0= 13*(block[0 + 4*i] +    block[2 + 4*i]);
        const int z1= 13*(block[0 + 4*i] -    block[2 + 4*i]);
        const int z2=  7* block[1 + 4*i] - 17*block[3 + 4*i];
        const int z3= 17* block[1 + 4*i] +  7*block[3 + 4*i];

        block[0 + 4*i]= z0 + z3;
        block[1 + 4*i]= z1 + z2;
        block[2 + 4*i]= z1 - z2;
        block[3 + 4*i]= z0 - z3;
    }

    for (i=0; i < 4; i++) {
        const int z0= 13*(block[i + 4*0] +    block[i + 4*2]);
        const int z1= 13*(block[i + 4*0] -    block[i + 4*2]);
        const int z2=  7* block[i + 4*1] - 17*block[i + 4*3];
        const int z3= 17* block[i + 4*1] +  7*block[i + 4*3];
        const int rr= (dc + 0x80000);

        dst[i + stride*0]= cm[ dst[i + stride*0] + (((z0 + z3)*qmul + rr) >> 20) ];
        dst[i + stride*1]= cm[ dst[i + stride*1] + (((z1 + z2)*qmul + rr) >> 20) ];
        dst[i + stride*2]= cm[ dst[i + stride*2] + (((z1 - z2)*qmul + rr) >> 20) ];
        dst[i + stride*3]= cm[ dst[i + stride*3] + (((z0 - z3)*qmul + rr) >> 20) ];
    }
}

static  int svq3_decode_block (GetBitContext *gb, 
									 DCTELEM *block,
                                     int index, const int type) {

  static const uint8_t *const scan_patterns[4] =
  { luma_dc_zigzag_scan, zigzag_scan, svq3_scan, chroma_dc_scan };

  int run, level, sign, vlc, limit;
  const int intra = (3 * type) >> 2;
  const uint8_t *const scan = scan_patterns[type];

  for (limit=(16 >> intra); index < 16; index=limit, limit+=8) {
    for (; (vlc = svq3_get_ue_golomb (gb)) != 0; index++) {

      if (vlc == INVALID_VLC)
        return -1;

      sign = (vlc & 0x1) - 1;
      vlc  = (vlc + 1) >> 1;

      if (type == 3) {
        if (vlc < 3) {
          run   = 0;
          level = vlc;
        } else if (vlc < 4) {
          run   = 1;
          level = 1;
        } else {
          run   = (vlc & 0x3);
          level = ((vlc + 9) >> 2) - run;
        }
      } else {
        if (vlc < 16) {
          run   = svq3_dct_tables[intra][vlc].run;
          level = svq3_dct_tables[intra][vlc].level;
        } else if (intra) {
          run   = (vlc & 0x7);
          level = (vlc >> 3) + ((run == 0) ? 8 : ((run < 2) ? 2 : ((run < 5) ? 0 : -1)));
        } else {
          run   = (vlc & 0xF);
          level = (vlc >> 4) + ((run == 0) ? 4 : ((run < 3) ? 2 : ((run < 10) ? 1 : 0)));
        }
      }

      if ((index += run) >= limit)
        return -1;

      block[scan[index]] = (level ^ sign) - sign;
    }

    if (type != 2) {
      break;
    }
  }

  return 0;
}

static  void svq3_mc_dir_part (MpegEncContext *s,
                                     int x, int y, int width, int height,
                                     int mx, int my, int dxy,
                                     int thirdpel, int dir, int avg) {

  const Picture *pic = (dir == 0) ? &s->last_picture : &s->next_picture;
  uint8_t *src, *dest;
  int i, emu = 0;
  int blocksize= 2 - (width>>3); //16->0, 8->1, 4->2

  mx += x;
  my += y;

  if (mx < 0 || mx >= (s->h_edge_pos - width  - 1) ||
      my < 0 || my >= (s->v_edge_pos - height - 1)) {

    if ((s->flags & CODEC_FLAG_EMU_EDGE)) {
      emu = 1;
    }

    mx = av_clip (mx, -16, (s->h_edge_pos - width  + 15));
    my = av_clip (my, -16, (s->v_edge_pos - height + 15));
  }

  /* form component predictions */
  dest = s->current_picture.data[0] + x + y*s->linesize;
  src  = pic->data[0] + mx + my*s->linesize;

  if (emu) {
    ff_emulated_edge_mc (s->edge_emu_buffer, src, s->linesize, (width + 1), (height + 1),
                         mx, my, s->h_edge_pos, s->v_edge_pos);
    src = s->edge_emu_buffer;
  }
  if(thirdpel)
    (avg ? s->dsp.avg_tpel_pixels_tab : s->dsp.put_tpel_pixels_tab)[dxy](dest, src, s->linesize, width, height);
  else
    (avg ? s->dsp.avg_pixels_tab : s->dsp.put_pixels_tab)[blocksize][dxy](dest, src, s->linesize, height);

  if (!(s->flags & CODEC_FLAG_GRAY)) {
    mx     = (mx + (mx < (int) x)) >> 1;
    my     = (my + (my < (int) y)) >> 1;
    width  = (width  >> 1);
    height = (height >> 1);
    blocksize++;

    for (i=1; i < 3; i++) {
      dest = s->current_picture.data[i] + (x >> 1) + (y >> 1)*s->uvlinesize;
      src  = pic->data[i] + mx + my*s->uvlinesize;

      if (emu) {
        ff_emulated_edge_mc (s->edge_emu_buffer, src, s->uvlinesize, (width + 1), (height + 1),
                             mx, my, (s->h_edge_pos >> 1), (s->v_edge_pos >> 1));
        src = s->edge_emu_buffer;
      }
      if(thirdpel)
        (avg ? s->dsp.avg_tpel_pixels_tab : s->dsp.put_tpel_pixels_tab)[dxy](dest, src, s->uvlinesize, width, height);
      else
        (avg ? s->dsp.avg_pixels_tab : s->dsp.put_pixels_tab)[blocksize][dxy](dest, src, s->uvlinesize, height);
    }
  }
}

static  int svq3_mc_dir (H264Context *h, int size, int mode, int dir, int avg) {

  int i, j, k, mx, my, dx, dy, x, y;
  MpegEncContext *const s = (MpegEncContext *) h;
  const int part_width  = ((size & 5) == 4) ? 4 : 16 >> (size & 1);
  const int part_height = 16 >> ((unsigned) (size + 1) / 3);
  const int extra_width = (mode == PREDICT_MODE) ? -16*6 : 0;