//C-  -*- C++ -*-
//C- -------------------------------------------------------------------
//C- DjVuLibre-3.5
//C- Copyright (c) 2002  Leon Bottou and Yann Le Cun.
//C- Copyright (c) 2001  AT&T
//C-
//C- This software is subject to, and may be distributed under, the
//C- GNU General Public License, either Version 2 of the license,
//C- or (at your option) any later version. The license should have
//C- accompanied the software or you may obtain a copy of the license
//C- from the Free Software Foundation at http://www.fsf.org .
//C-
//C- This program is distributed in the hope that it will be useful,
//C- but WITHOUT ANY WARRANTY; without even the implied warranty of
//C- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
//C- GNU General Public License for more details.
//C- 
//C- DjVuLibre-3.5 is derived from the DjVu(r) Reference Library from
//C- Lizardtech Software.  Lizardtech Software has authorized us to
//C- replace the original DjVu(r) Reference Library notice by the following
//C- text (see doc/lizard2002.djvu and doc/lizardtech2007.djvu):
//C-
//C-  ------------------------------------------------------------------
//C- | DjVu (r) Reference Library (v. 3.5)
//C- | Copyright (c) 1999-2001 LizardTech, Inc. All Rights Reserved.
//C- | The DjVu Reference Library is protected by U.S. Pat. No.
//C- | 6,058,214 and patents pending.
//C- |
//C- | This software is subject to, and may be distributed under, the
//C- | GNU General Public License, either Version 2 of the license,
//C- | or (at your option) any later version. The license should have
//C- | accompanied the software or you may obtain a copy of the license
//C- | from the Free Software Foundation at http://www.fsf.org .
//C- |
//C- | The computer code originally released by LizardTech under this
//C- | license and unmodified by other parties is deemed "the LIZARDTECH
//C- | ORIGINAL CODE."  Subject to any third party intellectual property
//C- | claims, LizardTech grants recipient a worldwide, royalty-free, 
//C- | non-exclusive license to make, use, sell, or otherwise dispose of 
//C- | the LIZARDTECH ORIGINAL CODE or of programs derived from the 
//C- | LIZARDTECH ORIGINAL CODE in compliance with the terms of the GNU 
//C- | General Public License.   This grant only confers the right to 
//C- | infringe patent claims underlying the LIZARDTECH ORIGINAL CODE to 
//C- | the extent such infringement is reasonably necessary to enable 
//C- | recipient to make, have made, practice, sell, or otherwise dispose 
//C- | of the LIZARDTECH ORIGINAL CODE (or portions thereof) and not to 
//C- | any greater extent that may be necessary to utilize further 
//C- | modifications or combinations.
//C- |
//C- | The LIZARDTECH ORIGINAL CODE is provided "AS IS" WITHOUT WARRANTY
//C- | OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED
//C- | TO ANY WARRANTY OF NON-INFRINGEMENT, OR ANY IMPLIED WARRANTY OF
//C- | MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
//C- +------------------------------------------------------------------

#ifdef HAVE_CONFIG_H
# include "config.h"
#endif
#if NEED_GNUG_PRAGMAS
# pragma implementation
#endif

/** @name cpaldjvu

    {\bf Synopsis}
    \begin{verbatim}
        cpaldjvu [options] <inputppmfile> <outputdjvufile>
    \end{verbatim}

    {\bf Description}
    
    File #"cpaldjvu.cpp"# demonstrates a simple quasi-lossless compressor for
    low resolution, low color, images with a reduced number of colors (e.g
    screendumps).  It simply quantizes the image on a limited number of
    colors, uses the dominant color to construct a uniform background, then
    performs lossless jb2 compression for all remaining objects.  
    
    Options
    \begin{description}
    \item[-colors n]  Maximum number of colors during quantization (default 256)
    \item[-dpi n]     Resolution written into the output file (default 100).
    \item[-verbose]   Displays additional messages.
    \end{description}

    {\bf Remarks}

    This is an interesting alternative to GIF. It performs especially well on
    screendumps.  Compression ratios can get hurt when there are continuous
    tone segment in the image.  Demoting such segments from foreground to
    background is a pretty interesting project.  Dithered segments behave
    surprisingly well.

    @memo
    Simple encoder for low resolution, low color images.
    @author
    L\'eon Bottou <leonb@research.att.com>
*/
//@{
//@}


#include "DjVuGlobal.h"
#include "GException.h"
#include "GSmartPointer.h"
#include "GContainer.h"
#include "ByteStream.h"
#include "IFFByteStream.h"
#include "GRect.h"
#include "GBitmap.h"
#include "JB2Image.h"
#include "DjVuPalette.h"
#include "IW44Image.h"
#include "DjVuInfo.h"
#include "GOS.h"
#include "GURL.h"
#include "DjVuMessage.h"
#include "jb2tune.h"
#include "common.h"

#ifdef MIN
#undef MIN
#endif
#ifdef MAX
#undef MAX
#endif
inline int MIN(int a, int b) { return ( a<b ?a :b); }
inline int MAX(int a, int b) { return ( a>b ?a :b); }


// --------------------------------------------------
// COLOR CONNECTED COMPONENT ANALYSIS
// --------------------------------------------------

// -- A run of pixels with the same color
struct Run    
{ 
  short y;       // vertical coordinate
  short x1;      // first horizontal coordinate
  short x2;      // last horizontal coordinate
  short color;   // color id
  int ccid;      // component id
};


// -- A component descriptor
struct CC    
{
  GRect bb;      // bounding box
  int npix;      // number of black pixels
  int nrun;      // number of runs
  int frun;      // first run in cc ordered array of runs
  int color;     // color id
};


// -- An image composed of runs
class CCImage 
{
public:
  int height;            // Height of the image in pixels
  int width;             // Width of the image in pixels
  GTArray<Run> runs;     // Array of runs
  GTArray<CC>  ccs;      // Array of component descriptors
  int nregularccs;       // Number of regular ccs (set by merge_and_split_ccs)
  CCImage(int width, int height);
  void add_single_run(int y, int x1, int x2, int color, int ccid=0);
  GP<GBitmap> get_bitmap_for_cc(int ccid) const;
  void make_ccids_by_analysis();
  void make_ccs_from_ccids();
  void merge_and_split_ccs(int smallsize, int largesize);
  void sort_in_reading_order(); 
  void erase_cc(int ccid);
};


// -- Compares runs
static inline bool
operator <= (const Run &a, const Run &b)
{
  return (a.y<b.y) || (a.y==b.y && a.x1<=b.x1);
}


// -- Constructs CCImage and provide defaults
CCImage::CCImage(int width, int height)
  : height(height), width(width), nregularccs(0)
{
}


// -- Adds a run to the CCImage
inline void 
CCImage::add_single_run(int y, int x1, int x2, int color, int ccid)
{
  int index = runs.hbound();
  runs.touch(++index);
  Run& run = runs[index];
  run.y = y;
  run.x1 = x1;
  run.x2 = x2;
  run.color = color;
  run.ccid = ccid;
  
}


// -- Performs color connected component analysis
void
CCImage::make_ccids_by_analysis()
{
  // Sort runs
  runs.sort();
  // Single Pass Connected Component Analysis (with unodes)
  int n;
  int p=0;
  GTArray<int> umap;
  for (n=0; n<=runs.hbound(); n++)
    {
      int y = runs[n].y;
      int x1 = runs[n].x1 - 1;
      int x2 = runs[n].x2 + 1;
      int color = runs[n].color;
      int id = (umap.hbound() + 1);
      // iterate over previous line runs
      if (p>0) p--;
      for(;runs[p].y < y-1;p++);
      for(;(runs[p].y < y) && (runs[p].x1 <= x2);p++ )
        {
          if ( runs[p].x2 >= x1 )
            {
              if (runs[p].color == color)
                {
                  // previous run touches current run and has same color
                  int oid = runs[p].ccid;
                  while (umap[oid] < oid)
                    oid = umap[oid];
                  if ((int)id > umap.hbound()) {
                    id = oid;
                  } else if (id < oid) {
                    umap[oid] = id;
                  } else {
                    umap[id] = oid;
                    id = oid;
                  }
                  // freshen previous run id
                  runs[p].ccid = id;
                }
              // stop if previous run goes past current run
              if (runs[p].x2 >= x2)
                break;
            }
        }
      // create new entry in umap
      runs[n].ccid = id;
      if (id > umap.hbound())
        {
          umap.touch(id);
          umap[id] = id;
        }
    }
  // Update umap and ccid
  for (n=0; n<=runs.hbound(); n++)
    {
      Run &run = runs[n];
      int ccid = run.ccid;
      while (umap[ccid] < ccid)
        ccid = umap[ccid];
      umap[run.ccid] = ccid;
      run.ccid = ccid;
    }
}


// -- Constructs the ``ccs'' array from run's ccids.
void
CCImage::make_ccs_from_ccids()
{
  int n;
  Run *pruns = runs;
  // Find maximal ccid
  int maxccid = -1;
  for (n=0; n<=runs.hbound(); n++)
    if (pruns[n].ccid > maxccid)
      maxccid = runs[n].ccid;
  GTArray<int> armap(0,maxccid);
  int *rmap = armap;
  // Renumber ccs 
  for (n=0; n<=maxccid; n++)
    armap[n] = -1;
  for (n=0; n<=runs.hbound(); n++)
    if (pruns[n].ccid >= 0)
      rmap[ pruns[n].ccid ] = 1;
  int nid = 0;
  for (n=0; n<=maxccid; n++)
    if (rmap[n] > 0)
      rmap[n] = nid++;
  // Adjust nregularccs (since ccs are renumbered)
  while (nregularccs>0 && rmap[nregularccs-1]<0)
    nregularccs -= 1;
  if (nregularccs>0)
    nregularccs = 1 + rmap[nregularccs-1];
  // Prepare cc descriptors
  ccs.resize(0,nid-1);
  for (n=0; n<nid; n++)
    ccs[n].nrun = 0;
  // Relabel runs
  for (n=0; n<=runs.hbound(); n++)
    {
      Run &run = pruns[n];
      if (run.ccid < 0) continue;  // runs with negative ccids are destroyed
      int oldccid = run.ccid;
      int newccid = rmap[oldccid];
      CC &cc = ccs[newccid];
      run.ccid = newccid;
      cc.nrun += 1;
    }
  // Compute positions for runs of cc
  int frun = 0;
  for (n=0; n<nid; n++) 
    {
      ccs[n].frun = rmap[n] = frun;
      frun += ccs[n].nrun;
    }
  // Copy runs
  GTArray<Run> rtmp;
  rtmp.steal(runs);
  Run *ptmp = rtmp;
  runs.resize(0,frun-1);
  pruns = runs;
  for (n=0; n<=rtmp.hbound(); n++)
    {
      int id = ptmp[n].ccid;
      if (id < 0) continue;
      int pos = rmap[id]++;
      pruns[pos] = ptmp[n];
    }
  // Finalize ccs
  for (n=0; n<nid; n++)
    {
      CC &cc = ccs[n];
      int npix = 0;
      runs.sort(cc.frun, cc.frun+cc.nrun-1);
      Run *run = &runs[cc.frun];
      int xmin = run->x1;
      int xmax = run->x2;
      int ymin = run->y;
      int ymax = run->y;
      cc.color = run->color;
      for (int i=0; i<cc.nrun; i++, run++)
        {
          if (run->x1 < xmin)  xmin = run->x1;
          if (run->x2 > xmax)  xmax = run->x2;
          if (run->y  < ymin)  ymin = run->y;
          if (run->y  > ymax)  ymax = run->y;
          npix += run->x2 - run->x1 + 1;
        }
      cc.npix = npix;
      cc.bb.xmin = xmin;
      cc.bb.ymin = ymin;
      cc.bb.xmax = xmax + 1;
      cc.bb.ymax = ymax + 1;
    }
}


// -- Helper for merge_and_split_ccs
struct Grid_x_Color 
{
  short gridi;
  short gridj;
  int color;
};


// -- Helper for merge_and_split_ccs
static inline unsigned int
hash(const Grid_x_Color &x) 
{
  return (x.gridi<<16) ^ (x.gridj<<8) ^ x.color;
}


// -- Helper for merge_and_split_ccs
static inline bool
operator==(const Grid_x_Color &x, const Grid_x_Color &y)
{
  return (x.gridi==y.gridi) && (x.gridj==y.gridj) && (x.color==y.color);
}


// -- Helper for merge_and_split_ccs
static int
makeccid(const Grid_x_Color &x, GMap<Grid_x_Color,int> &map, int &ncc)
{
  GPosition p = map.contains(x);
  if (p) return map[p];
  return map[x] = ncc++;
}


// -- Merges small ccs of similar color and splits large ccs
void
CCImage::merge_and_split_ccs(int smallsize, int largesize)
{
  int ncc = ccs.size();
  int nruns = runs.size();
  int splitsize = largesize;
  if (ncc <= 0) return;
  // Associative map for storing merged ccids
  GMap<Grid_x_Color,int> map;
  nregularccs = ncc;
  // Set the correct ccids for the runs
  for (int ccid=0; ccid<ccs.size(); ccid++)
    {
      CC* cc = &ccs[ccid];
      if (cc->nrun <= 0) continue;
      Grid_x_Color key;
      key.color = cc->color;
      int ccheight = cc->bb.height();
      int ccwidth = cc->bb.width();
      if (ccheight<=smallsize && ccwidth<=smallsize)
        {
          key.gridi = (cc->bb.ymin+cc->bb.ymax)/splitsize/2;
          key.gridj = (cc->bb.xmin+cc->bb.xmax)/splitsize/2;
          int newccid = makeccid(key, map, ncc);
          for(int runid=cc->frun; runid<cc->frun+cc->nrun; runid++)
            runs[runid].ccid = newccid;
        }
      else if (ccheight>=largesize || ccwidth>=largesize)
        {
          for(int runid=cc->frun; runid<cc->frun+cc->nrun; runid++)
            {
              Run *r = & runs[runid];
              key.gridi = r->y/splitsize;
              key.gridj = r->x1/splitsize;
              int gridj_end = r->x2/splitsize;
              int gridj_span = gridj_end - key.gridj;
              r->ccid = makeccid(key, map, ncc);
              if (gridj_span>0)
                {
                  // truncate current run 
                  runs.touch(nruns+gridj_span-1);
                  r = &runs[runid];
                  int x = key.gridj*splitsize + splitsize;
                  int x_end = r->x2;
                  r->x2 = x-1;
                  // append additional runs to the runs array
                  while (++key.gridj < gridj_end)
                    {
                      Run& newrun = runs[nruns++];
                      newrun.y = r->y;
                      newrun.x1 = x;
                      x += splitsize;
                      newrun.x2 = x-1;
                      newrun.color = key.color;
                      newrun.ccid = makeccid(key, map, ncc);
                    }
                  // append last run to the run array
                  Run& newrun = runs[nruns++];
                  newrun.y = r->y;
                  newrun.x1 = x;
                  newrun.x2 = x_end;
                  newrun.color = key.color;
                  newrun.ccid = makeccid(key, map, ncc);
                }
            }
        }
    }
  // Recompute cc descriptors
  make_ccs_from_ccids();
}


// -- Helps sorting cc
static int 
top_edges_descending (const void *pa, const void *pb)
{
  if (((CC*) pa)->bb.ymax != ((CC*) pb)->bb.ymax)
    return (((CC*) pb)->bb.ymax - ((CC*) pa)->bb.ymax);
  if (((CC*) pa)->bb.xmin != ((CC*) pb)->bb.xmin)
    return (((CC*) pa)->bb.xmin - ((CC*) pb)->bb.xmin);
  return (((CC*) pa)->frun - ((CC*) pb)->frun);
}


// -- Helps sorting cc
static int 
left_edges_ascending (const void *pa, const void *pb)
{
  if (((CC*) pa)->bb.xmin != ((CC*) pb)->bb.xmin)
    return (((CC*) pa)->bb.xmin - ((CC*) pb)->bb.xmin);
  if (((CC*) pb)->bb.ymax != ((CC*) pa)->bb.ymax)
    return (((CC*) pb)->bb.ymax - ((CC*) pa)->bb.ymax);
  return (((CC*) pa)->frun - ((CC*) pb)->frun);
}


// -- Helps sorting cc
static int 
integer_ascending (const void *pa, const void *pb)
{
  return ( *(int*)pb - *(int*)pa );
}


// -- Sort ccs in approximate reading order
void 
CCImage::sort_in_reading_order()
{
  if (nregularccs<2) return;
  CC *ccarray = new CC[nregularccs];
  // Copy existing ccarray (but segregate special ccs)
  int ccid;
  for(ccid=0; ccid<nregularccs; ccid++)
    ccarray[ccid] = ccs[ccid];
  // Sort the ccarray list into top-to-bottom order.
  qsort (ccarray, nregularccs, sizeof(CC), top_edges_descending);
  // Subdivide the ccarray list roughly into text lines
  int maxtopchange = width / 40;
  if (maxtopchange < 32) 
    maxtopchange = 32;
  // - Loop until processing all ccs
  int ccno = 0;
  int *bottoms = new int[nregularccs];
  while (ccno < nregularccs)
    {
      // - Gather first line approximation
      int nccno;
      int sublist_top = ccarray[ccno].bb.ymax-1;
      int sublist_bottom = ccarray[ccno].bb.ymin;
      for (nccno=ccno; nccno < nregularccs; nccno++)
        {
          if (ccarray[nccno].bb.ymax-1 < sublist_bottom) break;
          if (ccarray[nccno].bb.ymax-1 < sublist_top - maxtopchange) break;
          int bottom = ccarray[nccno].bb.ymin;
          bottoms[nccno-ccno] = bottom;
          if (bottom < sublist_bottom)
            sublist_bottom = bottom;
        }
      // - If more than one candidate cc for the line
      if (nccno > ccno + 1)
        {
          // - Compute median bottom
          qsort(bottoms, nccno-ccno, sizeof(int), integer_ascending);
          int bottom = bottoms[ (nccno-ccno-1)/2 ];
          // - Compose final line
          for (nccno=ccno; nccno < nregularccs; nccno++)
            if (ccarray[nccno].bb.ymax-1 < bottom)
              break;
          // - Sort final line
          qsort (ccarray+ccno, nccno-ccno, sizeof(CC), left_edges_ascending);
        }
      // - Next line
      ccno = nccno;
    }
  // Copy ccarray back and renumber the runs
  for(ccid=0; ccid<nregularccs; ccid++)
    {
      CC& cc = ccarray[ccid];
      ccs[ccid] = cc;
      for(int r=cc.frun; r<cc.frun+cc.nrun; r++)
        runs[r].ccid = ccid;
    }
  // Free memory
  delete [] bottoms;
  delete[] ccarray;
}


// -- Creates a bitmap for a particular component
GP<GBitmap>   
CCImage::get_bitmap_for_cc(const int ccid) const
{
  const CC &cc = ccs[ccid];
  const GRect &bb = cc.bb;
  GP<GBitmap> bits = GBitmap::create(bb.height(), bb.width());
  const Run *prun = & runs[(int)cc.frun];
  for (int i=0; i<cc.nrun; i++,prun++)
    {
      if (prun->y<bb.ymin || prun->y>=bb.ymax)
        G_THROW("Internal error (y bounds)");
      if (prun->x1<bb.xmin || prun->x2>=bb.xmax)
        G_THROW("Internal error (x bounds)");
      unsigned char *row = (*bits)[prun->y - bb.ymin];
      for (int x=prun->x1; x<=prun->x2; x++)
        row[x - bb.xmin] = 1;
    }
  return bits;
}


// -- Marks cc for deletion
void 
CCImage::erase_cc(int ccid)
{
  CC &cc = ccs[ccid];
  Run *r = &runs[cc.frun];
  int nr = cc.nrun;
  cc.nrun = 0;
  cc.npix = 0;
  while (--nr >= 0)
    (r++)->ccid = -1;  // will be deleted by make_ccs_from_ccids()
}




// --------------------------------------------------
// DEMOTION OF FOREGROUND CCS TO BACKGROUND STATUS
// --------------------------------------------------


// ISSUE: DEMOTION OF CCS (UNIMPLEMENTED) 

// The current code uses a single color for the background layer.  Many large,
// non matching, ccs however may be better encoded as part of the background
// layer.  A way to achieve this could be to consider each cc and evaluate the
// costs of coding it as foreground (does it match other ccs, does it comes
// with a complex geometry) or background (does its color blend smoothly with
// the surrounding parts of the background image).  One just needs then to
// remove the demoted ccs from the mask using CCImage::erase_cc() and
// recompute the ccs using CCImage::make_ccs_from_ccids().  Defining the
// compilation symbols BACKGROUND_SUBSAMPLING_FACTOR and
// PROGRESSIVE_BACKGROUND will enable code for computing the background from
// the input image and the provided mask.


// --------------------------------------------------
// MAIN COMPRESSION ROUTINE
// --------------------------------------------------


// -- Options for low color compression
struct cpaldjvuopts
{
  int ncolors;
  int dpi;
  bool verbose;
  bool bgwhite;
};


// -- Compresses low color pixmap.
void 
cpaldjvu(ByteStream *ibs, GURL &urlout, const cpaldjvuopts &opts)
{
  GP<GPixmap> ginput=GPixmap::create(*ibs);
  int w = ginput->columns();
  int h = ginput->rows();
  int dpi = MAX(200, MIN(900, opts.dpi));
  int largesize = MIN(500, MAX(64, dpi));
  int smallsize = MAX(2, dpi/150);

  // Compute optimal palette and quantize input pixmap
  GP<DjVuPalette> gpal=DjVuPalette::create();
  DjVuPalette &pal=*gpal;
  GPixel bgcolor;
  int bgindex = -1;
  if (! opts.bgwhite)
    {
      bgindex = pal.compute_pixmap_palette(*ginput, opts.ncolors);
      pal.index_to_color(bgindex, bgcolor);
    }
  else
    {
      bgcolor = GPixel::WHITE;
      pal.histogram_clear();
      for (int j=0; j<h; j++)
        {
          const GPixel *p = (*ginput)[j];
          for (int i=0; i<w; i++)
            if (p[i] != GPixel::WHITE)
              pal.histogram_add(p[i], 1);
        }
      pal.compute_palette(opts.ncolors);
    }
  if (opts.verbose)
    DjVuFormatErrorUTF8( "%s\t%d\t%d\t%d",
                         ERR_MSG("cpaldjvu.quantizied"), 
                         w, h, pal.size());
  if (opts.verbose && !opts.bgwhite)
    DjVuPrintErrorUTF8("cpaldjvu: background color is #%02x%02x%02x.\n", 
                       bgcolor.r, bgcolor.g, bgcolor.b);
  
  // Fill CCImage with color runs
  int xruncount=0,yruncount=0;
  CCImage rimg(w, h);
  int *line;
  GPBuffer<int> gline(line,w);
  int *prevline;
  GPBuffer<int> gprevline(prevline,w);
  for (int x=0;x<w;x++)
  {
    prevline[x]=bgindex;
  }
  for (int y=0; y<h; y++)
    {
      int x;
      const GPixel *row = (*ginput)[y];
      for(x=0;x<w;x++)
        {
          line[x] = pal.color_to_index(row[x]);
          if (opts.bgwhite && row[x]==GPixel::WHITE)
            line[x] = bgindex;
        }
      for(x=0;x<w;)
        {
          int x1 = x;
          int index = line[x++];
          while (x<w && line[x]==index) { x++; }
          if (index != bgindex)
            {
              xruncount++;
              rimg.add_single_run(y, x1, x-1, index);
            }
        }
      for(x=0;x<w;x++)
        if(prevline[x] != line[x]) yruncount++;
      gprevline.swap(gline);
    }
  ginput = 0; //save memory
  if (opts.verbose)
    DjVuFormatErrorUTF8( "%s\t%d", ERR_MSG("cpaldjvu.color_runs"), 
                         rimg.runs.size());

  // Perform Color Connected Component Analysis
  rimg.make_ccids_by_analysis();                  // Obtain ccids
  rimg.make_ccs_from_ccids();                     // Compute cc descriptors
  if (opts.verbose)
    DjVuFormatErrorUTF8( "%s\t%d", ERR_MSG("cpaldjvu.ccs_before"), 
                         rimg.ccs.size());
  rimg.merge_and_split_ccs(smallsize,largesize);  // Eliminates gross ccs
  if (opts.verbose)
    DjVuFormatErrorUTF8( "%s\t%d", ERR_MSG("cpaldjvu.ccs_after"), 
                         rimg.ccs.size());
  rimg.sort_in_reading_order();                   // Sort cc descriptors
  
  // Create JB2Image and fill colordata
  GP<JB2Image> gjimg=JB2Image::create();
  JB2Image &jimg=*gjimg;
  jimg.set_dimension(w, h);
  int nccs = rimg.ccs.size();
  for (int ccid=0; ccid<nccs; ccid++)
    {
      JB2Shape shape;
      JB2Blit  blit;
      shape.parent = -1;
      shape.userdata = 0;
      if (ccid >= rimg.nregularccs)
        shape.userdata |= JB2SHAPE_SPECIAL;
      shape.bits = rimg.get_bitmap_for_cc(ccid);
      shape.bits->compress();
      CC& cc = rimg.ccs[ccid];
      blit.shapeno = jimg.add_shape(shape);
      blit.left = cc.bb.xmin;
      blit.bottom = cc.bb.ymin;
      int blitno = jimg.add_blit(blit);
      pal.colordata.touch(blitno);
      pal.colordata[blitno] = cc.color;
    }
  
  // Organize JB2Image
  tune_jb2image_lossless(&jimg);
  if (opts.verbose)
    {
      int nshape=0, nrefine=0;
      for (int i=0; i<jimg.get_shape_count(); i++) {
        if (!jimg.get_shape(i).bits) continue;
        if (jimg.get_shape(i).parent >= 0) nrefine++; 
        nshape++; 
      }
      DjVuFormatErrorUTF8( "%s\t%d\t%d",
                       ERR_MSG("cpaldjvu.cross_code"), 
                       nshape, nrefine);
    }
  
  // Create background image
#ifdef BACKGROUND_SUBSAMPLING_FACTOR
  // -- we may create the background by masking and subsampling
  GP<GPixmap> ginputsub=GPixmap::create();
  GPixmap &inputsub=*ginputsub;
  GP<GBitmap> mask = jimg.get_bitmap(BACKGROUND_SUBSAMPLING_FACTOR);
  inputsub.downsample(&input, BACKGROUND_SUBSAMPLING_FACTOR);
  GP<IW44Image> iwimage=IW44Image::create(inputsub, mask);
#else
  // -- but who cares since the background is uniform.
  GP<GPixmap> ginputsub=GPixmap::create((h+11)/12, (w+11)/12, &bgcolor);
  GPixmap &inputsub=*ginputsub;
  GP<IW44Image> iwimage=IW44Image::create_encode(inputsub);
#endif

  // Assemble DJVU file
  GP<ByteStream> obs=ByteStream::create(urlout, "wb");
  GP<IFFByteStream> giff=IFFByteStream::create(obs);
  IFFByteStream &iff=*giff;
  // -- main composite chunk
  iff.put_chunk("FORM:DJVU", 1);
  // -- ``INFO'' chunk
  iff.put_chunk("INFO");
  GP<DjVuInfo> ginfo=DjVuInfo::create();
  DjVuInfo info=*ginfo;
  info.height = h;
  info.width = w;
  info.dpi = opts.dpi;
  info.encode(*iff.get_bytestream());
  iff.close_chunk();
  // -- ``Sjbz'' chunk
  iff.put_chunk("Sjbz");
  jimg.encode(iff.get_bytestream());
  iff.close_chunk();
  // -- ``FGbz'' chunk
  iff.put_chunk("FGbz");
  pal.encode(iff.get_bytestream());
  iff.close_chunk();
  // -- ``BG44'' chunk
  IWEncoderParms iwparms;
#ifdef PROGRESSIVE_BACKGROUND
  // ----- we may use several chunks to enable progressive rendering ...
  iff.put_chunk("BG44");
  iwparms.slices = 74;
  iwimage->encode_chunk(iff, iwparms);
  iff.close_chunk();
  iff.put_chunk("BG44");
  iwparms.slices = 87;
  iwimage->encode_chunk(iff, iwparms);
  iff.close_chunk();
#endif
  // ----- but who cares when the background is so small.
  iff.put_chunk("BG44");
  iwparms.slices = 97;
  iwimage->encode_chunk(iff.get_bytestream(), iwparms);
  iff.close_chunk();
  // -- terminate main composite chunk
  iff.close_chunk();
  // Finished!
}  



// --------------------------------------------------
// MAIN
// --------------------------------------------------


void
usage()
{
  DjVuPrintErrorUTF8(
#ifdef DJVULIBRE_VERSION
         "CPALDJVU --- DjVuLibre-" DJVULIBRE_VERSION "\n"
#endif
         "DjVu encoder for images with few colors\n\n"
         "Usage: cpaldjvu [options] <inputppmfile> <outputdjvufile>\n"
         "Options are:\n"
         "   -colors [2-4096] Maximum number of colors during quantization (default 256).\n"
         "   -dpi [25-6000]   Resolution written into the output file (default 100).\n"
         "   -verbose         Displays additional messages.\n"
         "   -bgwhite         Use the lightest color for background (usually white).\n"
         );
  exit(10);
}


int 
main(int argc, const char **argv)
{
  DJVU_LOCALE;
  GArray<GUTF8String> dargv(0,argc-1);
  for(int i=0;i<argc;++i)
    dargv[i]=GNativeString(argv[i]);
  G_TRY
    {
      GURL inputppmurl;
      GURL outputdjvuurl;
      // Defaults
      cpaldjvuopts opts;
      opts.dpi = 100;
      opts.ncolors = 256;
      opts.verbose = false;
      opts.bgwhite = false;
      // Parse options
      for (int i=1; i<argc; i++)
        {
          GUTF8String arg = dargv[i];
          if (arg == "-colors" && i+1<argc)
            {
              char *end;
              opts.ncolors = strtol(dargv[++i], &end, 10);
              if (*end || opts.ncolors<2 || opts.ncolors>4096)
                usage();
            }
          else if (arg == "-dpi" && i+1<argc)
            {
              char *end;
              opts.dpi = strtol(dargv[++i], &end, 10);
              if (*end || opts.dpi<25 || opts.dpi>6000)
                usage();
            }
          else if (arg == "-verbose" || arg == "-v")
            opts.verbose = true;
          else if (arg == "-bgwhite")
            opts.bgwhite = true;
          else if (arg[0] == '-' && arg[1])
            usage();
          else if (inputppmurl.is_empty())
            inputppmurl = GURL::Filename::UTF8(arg);
          else if (outputdjvuurl.is_empty())
            outputdjvuurl = GURL::Filename::UTF8(arg);
          else
            usage();
        }
      if (inputppmurl.is_empty() || outputdjvuurl.is_empty())
        usage();
      // Load and run
      GP<ByteStream> ibs=ByteStream::create(inputppmurl,"rb");
      cpaldjvu(ibs, outputdjvuurl, opts);
    }
  G_CATCH(ex)
    {
      ex.perror();
      exit(1);
    }
  G_ENDCATCH;
  return 0;
}