stb_image.cpp

#ifndef STB_IMAGE_H
#define STB_IMAGE_H

#ifndef STBI_HEADER_FILE_ONLY

#include "stb_image.h"

#ifndef STBI_NO_HDR
#include <math.h>  // ldexp
#include <string.h> // strcmp, strtok
#endif

#ifndef STBI_NO_STDIO
#include <stdio.h>
#endif
#include <stdlib.h>
#include <memory.h>
#include <assert.h>
#include <stdarg.h>

#ifndef _MSC_VER
   #ifdef __cplusplus
   #define stbi_inline inline
   #else
   #define stbi_inline
   #endif
#else
   #define stbi_inline __forceinline
#endif


// implementation:
typedef unsigned char  uint8;
typedef unsigned short uint16;
typedef   signed short  int16;
typedef unsigned int   uint32;
typedef   signed int    int32;
typedef unsigned int   uint;

// should produce compiler error if size is wrong
typedef unsigned char validate_uint32[sizeof(uint32)==4 ? 1 : -1];

#if defined(STBI_NO_STDIO) && !defined(STBI_NO_WRITE)
#define STBI_NO_WRITE
#endif

#define STBI_NOTUSED(v)  (void)sizeof(v)

#ifdef _MSC_VER
#define STBI_HAS_LROTL
#endif

#ifdef STBI_HAS_LROTL
   #define stbi_lrot(x,y)  _lrotl(x,y)
#else
   #define stbi_lrot(x,y)  (((x) << (y)) | ((x) >> (32 - (y))))
#endif

///////////////////////////////////////////////
//
//  stbi struct and start_xxx functions

// stbi structure is our basic context used by all images, so it
// contains all the IO context, plus some basic image information
typedef struct
{
   uint32 img_x, img_y;
   int img_n, img_out_n;

   stbi_io_callbacks io;
   void *io_user_data;

   int read_from_callbacks;
   int buflen;
   uint8 buffer_start[128];

   uint8 *img_buffer, *img_buffer_end;
   uint8 *img_buffer_original;
} stbi;


static void refill_buffer(stbi *s);

// initialize a memory-decode context
static void start_mem(stbi *s, uint8 const *buffer, int len)
{
   s->io.read = NULL;
   s->read_from_callbacks = 0;
   s->img_buffer = s->img_buffer_original = (uint8 *) buffer;
   s->img_buffer_end = (uint8 *) buffer+len;
}

// initialize a callback-based context
static void start_callbacks(stbi *s, stbi_io_callbacks *c, void *user)
{
   s->io = *c;
   s->io_user_data = user;
   s->buflen = sizeof(s->buffer_start);
   s->read_from_callbacks = 1;
   s->img_buffer_original = s->buffer_start;
   refill_buffer(s);
}

#ifndef STBI_NO_STDIO

static int stdio_read(void *user, char *data, int size)
{
   return (int) fread(data,1,size,(FILE*) user);
}

static void stdio_skip(void *user, unsigned n)
{
   fseek((FILE*) user, n, SEEK_CUR);
}

static int stdio_eof(void *user)
{
   return feof((FILE*) user);
}

static stbi_io_callbacks stbi_stdio_callbacks =
{
   stdio_read,
   stdio_skip,
   stdio_eof,
};

static void start_file(stbi *s, FILE *f)
{
   start_callbacks(s, &stbi_stdio_callbacks, (void *) f);
}

//static void stop_file(stbi *s) { }

#endif // !STBI_NO_STDIO

static void stbi_rewind(stbi *s)
{
   // conceptually rewind SHOULD rewind to the beginning of the stream,
   // but we just rewind to the beginning of the initial buffer, because
   // we only use it after doing 'test', which only ever looks at at most 92 bytes
   s->img_buffer = s->img_buffer_original;
}

static int      stbi_jpeg_test(stbi *s);
static stbi_uc *stbi_jpeg_load(stbi *s, int *x, int *y, int *comp, int req_comp);
static int      stbi_jpeg_info(stbi *s, int *x, int *y, int *comp);
static int      stbi_png_test(stbi *s);
static stbi_uc *stbi_png_load(stbi *s, int *x, int *y, int *comp, int req_comp);
static int      stbi_png_info(stbi *s, int *x, int *y, int *comp);
static int      stbi_bmp_test(stbi *s);
static stbi_uc *stbi_bmp_load(stbi *s, int *x, int *y, int *comp, int req_comp);
static int      stbi_tga_test(stbi *s);
static stbi_uc *stbi_tga_load(stbi *s, int *x, int *y, int *comp, int req_comp);
static int      stbi_tga_info(stbi *s, int *x, int *y, int *comp);
static int      stbi_psd_test(stbi *s);
static stbi_uc *stbi_psd_load(stbi *s, int *x, int *y, int *comp, int req_comp);
static int      stbi_hdr_test(stbi *s);
static float   *stbi_hdr_load(stbi *s, int *x, int *y, int *comp, int req_comp);
static int      stbi_pic_test(stbi *s);
static stbi_uc *stbi_pic_load(stbi *s, int *x, int *y, int *comp, int req_comp);
static int      stbi_gif_test(stbi *s);
static stbi_uc *stbi_gif_load(stbi *s, int *x, int *y, int *comp, int req_comp);
static int      stbi_gif_info(stbi *s, int *x, int *y, int *comp);


// this is not threadsafe
static const char *failure_reason;

const char *stbi_failure_reason(void)
{
   return failure_reason;
}

static int e(const char *str)
{
   failure_reason = str;
   return 0;
}

// e - error
// epf - error returning pointer to float
// epuc - error returning pointer to unsigned char

#ifdef STBI_NO_FAILURE_STRINGS
   #define e(x,y)  0
#elif defined(STBI_FAILURE_USERMSG)
   #define e(x,y)  e(y)
#else
   #define e(x,y)  e(x)
#endif

#define epf(x,y)   ((float *) (e(x,y)?NULL:NULL))
#define epuc(x,y)  ((unsigned char *) (e(x,y)?NULL:NULL))

void stbi_image_free(void *retval_from_stbi_load)
{
   free(retval_from_stbi_load);
}

#ifndef STBI_NO_HDR
static float   *ldr_to_hdr(stbi_uc *data, int x, int y, int comp);
static stbi_uc *hdr_to_ldr(float   *data, int x, int y, int comp);
#endif

static unsigned char *stbi_load_main(stbi *s, int *x, int *y, int *comp, int req_comp)
{
   if (stbi_jpeg_test(s)) return stbi_jpeg_load(s,x,y,comp,req_comp);
   if (stbi_png_test(s))  return stbi_png_load(s,x,y,comp,req_comp);
   if (stbi_bmp_test(s))  return stbi_bmp_load(s,x,y,comp,req_comp);
   if (stbi_gif_test(s))  return stbi_gif_load(s,x,y,comp,req_comp);
   if (stbi_psd_test(s))  return stbi_psd_load(s,x,y,comp,req_comp);
   if (stbi_pic_test(s))  return stbi_pic_load(s,x,y,comp,req_comp);

   #ifndef STBI_NO_HDR
   if (stbi_hdr_test(s)) {
      float *hdr = stbi_hdr_load(s, x,y,comp,req_comp);
      return hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp);
   }
   #endif

   // test tga last because it's a crappy test!
   if (stbi_tga_test(s))
      return stbi_tga_load(s,x,y,comp,req_comp);
   return epuc("unknown image type", "Image not of any known type, or corrupt");
}

#ifndef STBI_NO_STDIO
unsigned char *stbi_load(char const *filename, int *x, int *y, int *comp, int req_comp)
{
   FILE *f = fopen(filename, "rb");
   unsigned char *result;
   if (!f) return epuc("can't fopen", "Unable to open file");
   result = stbi_load_from_file(f,x,y,comp,req_comp);
   fclose(f);
   return result;
}

unsigned char *stbi_load_from_file(FILE *f, int *x, int *y, int *comp, int req_comp)
{
   stbi s;
   start_file(&s,f);
   return stbi_load_main(&s,x,y,comp,req_comp);
}
#endif //!STBI_NO_STDIO

unsigned char *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp)
{
   stbi s;
   start_mem(&s,buffer,len);
   return stbi_load_main(&s,x,y,comp,req_comp);
}

unsigned char *stbi_load_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp)
{
   stbi s;
   start_callbacks(&s, (stbi_io_callbacks *) clbk, user);
   return stbi_load_main(&s,x,y,comp,req_comp);
}

#ifndef STBI_NO_HDR

float *stbi_loadf_main(stbi *s, int *x, int *y, int *comp, int req_comp)
{
   unsigned char *data;
   #ifndef STBI_NO_HDR
   if (stbi_hdr_test(s))
      return stbi_hdr_load(s,x,y,comp,req_comp);
   #endif
   data = stbi_load_main(s, x, y, comp, req_comp);
   if (data)
      return ldr_to_hdr(data, *x, *y, req_comp ? req_comp : *comp);
   return epf("unknown image type", "Image not of any known type, or corrupt");
}

float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp)
{
   stbi s;
   start_mem(&s,buffer,len);
   return stbi_loadf_main(&s,x,y,comp,req_comp);
}

float *stbi_loadf_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp)
{
   stbi s;
   start_callbacks(&s, (stbi_io_callbacks *) clbk, user);
   return stbi_loadf_main(&s,x,y,comp,req_comp);
}

#ifndef STBI_NO_STDIO
float *stbi_loadf(char const *filename, int *x, int *y, int *comp, int req_comp)
{
   FILE *f = fopen(filename, "rb");
   float *result;
   if (!f) return epf("can't fopen", "Unable to open file");
   result = stbi_loadf_from_file(f,x,y,comp,req_comp);
   fclose(f);
   return result;
}

float *stbi_loadf_from_file(FILE *f, int *x, int *y, int *comp, int req_comp)
{
   stbi s;
   start_file(&s,f);
   return stbi_loadf_main(&s,x,y,comp,req_comp);
}
#endif // !STBI_NO_STDIO

#endif // !STBI_NO_HDR

// these is-hdr-or-not is defined independent of whether STBI_NO_HDR is
// defined, for API simplicity; if STBI_NO_HDR is defined, it always
// reports false!

int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len)
{
   #ifndef STBI_NO_HDR
   stbi s;
   start_mem(&s,buffer,len);
   return stbi_hdr_test(&s);
   #else
   STBI_NOTUSED(buffer);
   STBI_NOTUSED(len);
   return 0;
   #endif
}

#ifndef STBI_NO_STDIO
extern int      stbi_is_hdr          (char const *filename)
{
   FILE *f = fopen(filename, "rb");
   int result=0;
   if (f) {
      result = stbi_is_hdr_from_file(f);
      fclose(f);
   }
   return result;
}

extern int      stbi_is_hdr_from_file(FILE *f)
{
   #ifndef STBI_NO_HDR
   stbi s;
   start_file(&s,f);
   return stbi_hdr_test(&s);
   #else
   return 0;
   #endif
}
#endif // !STBI_NO_STDIO

extern int      stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user)
{
   #ifndef STBI_NO_HDR
   stbi s;
   start_callbacks(&s, (stbi_io_callbacks *) clbk, user);
   return stbi_hdr_test(&s);
   #else
   return 0;
   #endif
}

#ifndef STBI_NO_HDR
static float h2l_gamma_i=1.0f/2.2f, h2l_scale_i=1.0f;
static float l2h_gamma=2.2f, l2h_scale=1.0f;

void   stbi_hdr_to_ldr_gamma(float gamma) { h2l_gamma_i = 1/gamma; }
void   stbi_hdr_to_ldr_scale(float scale) { h2l_scale_i = 1/scale; }

void   stbi_ldr_to_hdr_gamma(float gamma) { l2h_gamma = gamma; }
void   stbi_ldr_to_hdr_scale(float scale) { l2h_scale = scale; }
#endif


//////////////////////////////////////////////////////////////////////////////
//
// Common code used by all image loaders
//

enum
{
   SCAN_load=0,
   SCAN_type,
   SCAN_header
};

static void refill_buffer(stbi *s)
{
   int n = (s->io.read)(s->io_user_data,(char*)s->buffer_start,s->buflen);
   if (n == 0) {
      // at end of file, treat same as if from memory
      s->read_from_callbacks = 0;
      s->img_buffer = s->img_buffer_end-1;
      *s->img_buffer = 0;
   } else {
      s->img_buffer = s->buffer_start;
      s->img_buffer_end = s->buffer_start + n;
   }
}

stbi_inline static int get8(stbi *s)
{
   if (s->img_buffer < s->img_buffer_end)
      return *s->img_buffer++;
   if (s->read_from_callbacks) {
      refill_buffer(s);
      return *s->img_buffer++;
   }
   return 0;
}

stbi_inline static int at_eof(stbi *s)
{
   if (s->io.read) {
      if (!(s->io.eof)(s->io_user_data)) return 0;
      // if feof() is true, check if buffer = end
      // special case: we've only got the special 0 character at the end
      if (s->read_from_callbacks == 0) return 1;
   }

   return s->img_buffer >= s->img_buffer_end;
}

stbi_inline static uint8 get8u(stbi *s)
{
   return (uint8) get8(s);
}

static void skip(stbi *s, int n)
{
   if (s->io.read) {
      int blen = s->img_buffer_end - s->img_buffer;
      if (blen < n) {
         s->img_buffer = s->img_buffer_end;
         (s->io.skip)(s->io_user_data, n - blen);
         return;
      }
   }
   s->img_buffer += n;
}

static int getn(stbi *s, stbi_uc *buffer, int n)
{
   if (s->io.read) {
      int blen = s->img_buffer_end - s->img_buffer;
      if (blen < n) {
         int res, count;

         memcpy(buffer, s->img_buffer, blen);

         count = (s->io.read)(s->io_user_data, (char*) buffer + blen, n - blen);
         res = (count == (n-blen));
         s->img_buffer = s->img_buffer_end;
         return res;
      }
   }

   if (s->img_buffer+n <= s->img_buffer_end) {
      memcpy(buffer, s->img_buffer, n);
      s->img_buffer += n;
      return 1;
   } else
      return 0;
}

static int get16(stbi *s)
{
   int z = get8(s);
   return (z << 8) + get8(s);
}

static uint32 get32(stbi *s)
{
   uint32 z = get16(s);
   return (z << 16) + get16(s);
}

static int get16le(stbi *s)
{
   int z = get8(s);
   return z + (get8(s) << 8);
}

static uint32 get32le(stbi *s)
{
   uint32 z = get16le(s);
   return z + (get16le(s) << 16);
}

//////////////////////////////////////////////////////////////////////////////
//
//  generic converter from built-in img_n to req_comp
//    individual types do this automatically as much as possible (e.g. jpeg
//    does all cases internally since it needs to colorspace convert anyway,
//    and it never has alpha, so very few cases ). png can automatically
//    interleave an alpha=255 channel, but falls back to this for other cases
//
//  assume data buffer is malloced, so malloc a new one and free that one
//  only failure mode is malloc failing

static uint8 compute_y(int r, int g, int b)
{
   return (uint8) (((r*77) + (g*150) +  (29*b)) >> 8);
}

static unsigned char *convert_format(unsigned char *data, int img_n, int req_comp, uint x, uint y)
{
   int i,j;
   unsigned char *good;

   if (req_comp == img_n) return data;
   assert(req_comp >= 1 && req_comp <= 4);

   good = (unsigned char *) malloc(req_comp * x * y);
   if (good == NULL) {
      free(data);
      return epuc("outofmem", "Out of memory");
   }

   for (j=0; j < (int) y; ++j) {
      unsigned char *src  = data + j * x * img_n   ;
      unsigned char *dest = good + j * x * req_comp;

      #define COMBO(a,b)  ((a)*8+(b))
      #define CASE(a,b)   case COMBO(a,b): for(i=x-1; i >= 0; --i, src += a, dest += b)
      // convert source image with img_n components to one with req_comp components;
      // avoid switch per pixel, so use switch per scanline and massive macros
      switch (COMBO(img_n, req_comp)) {
         CASE(1,2) dest[0]=src[0], dest[1]=255; break;
         CASE(1,3) dest[0]=dest[1]=dest[2]=src[0]; break;
         CASE(1,4) dest[0]=dest[1]=dest[2]=src[0], dest[3]=255; break;
         CASE(2,1) dest[0]=src[0]; break;
         CASE(2,3) dest[0]=dest[1]=dest[2]=src[0]; break;
         CASE(2,4) dest[0]=dest[1]=dest[2]=src[0], dest[3]=src[1]; break;
         CASE(3,4) dest[0]=src[0],dest[1]=src[1],dest[2]=src[2],dest[3]=255; break;
         CASE(3,1) dest[0]=compute_y(src[0],src[1],src[2]); break;
         CASE(3,2) dest[0]=compute_y(src[0],src[1],src[2]), dest[1] = 255; break;
         CASE(4,1) dest[0]=compute_y(src[0],src[1],src[2]); break;
         CASE(4,2) dest[0]=compute_y(src[0],src[1],src[2]), dest[1] = src[3]; break;
         CASE(4,3) dest[0]=src[0],dest[1]=src[1],dest[2]=src[2]; break;
         default: assert(0);
      }
      #undef CASE
   }

   free(data);
   return good;
}

#ifndef STBI_NO_HDR
static float   *ldr_to_hdr(stbi_uc *data, int x, int y, int comp)
{
   int i,k,n;
   float *output = (float *) malloc(x * y * comp * sizeof(float));
   if (output == NULL) { free(data); return epf("outofmem", "Out of memory"); }
   // compute number of non-alpha components
   if (comp & 1) n = comp; else n = comp-1;
   for (i=0; i < x*y; ++i) {
      for (k=0; k < n; ++k) {
         output[i*comp + k] = (float) pow(data[i*comp+k]/255.0f, l2h_gamma) * l2h_scale;
      }
      if (k < comp) output[i*comp + k] = data[i*comp+k]/255.0f;
   }
   free(data);
   return output;
}

#define float2int(x)   ((int) (x))
static stbi_uc *hdr_to_ldr(float   *data, int x, int y, int comp)
{
   int i,k,n;
   stbi_uc *output = (stbi_uc *) malloc(x * y * comp);
   if (output == NULL) { free(data); return epuc("outofmem", "Out of memory"); }
   // compute number of non-alpha components
   if (comp & 1) n = comp; else n = comp-1;
   for (i=0; i < x*y; ++i) {
      for (k=0; k < n; ++k) {
         float z = (float) pow(data[i*comp+k]*h2l_scale_i, h2l_gamma_i) * 255 + 0.5f;
         if (z < 0) z = 0;
         if (z > 255) z = 255;
         output[i*comp + k] = (uint8) float2int(z);
      }
      if (k < comp) {
         float z = data[i*comp+k] * 255 + 0.5f;
         if (z < 0) z = 0;
         if (z > 255) z = 255;
         output[i*comp + k] = (uint8) float2int(z);
      }
   }
   free(data);
   return output;
}
#endif

//////////////////////////////////////////////////////////////////////////////
//
//  "baseline" JPEG/JFIF decoder (not actually fully baseline implementation)
//
//    simple implementation
//      - channel subsampling of at most 2 in each dimension
//      - doesn't support delayed output of y-dimension
//      - simple interface (only one output format: 8-bit interleaved RGB)
//      - doesn't try to recover corrupt jpegs
//      - doesn't allow partial loading, loading multiple at once
//      - still fast on x86 (copying globals into locals doesn't help x86)
//      - allocates lots of intermediate memory (full size of all components)
//        - non-interleaved case requires this anyway
//        - allows good upsampling (see next)
//    high-quality
//      - upsampled channels are bilinearly interpolated, even across blocks
//      - quality integer IDCT derived from IJG's 'slow'
//    performance
//      - fast huffman; reasonable integer IDCT
//      - uses a lot of intermediate memory, could cache poorly
//      - load http://nothings.org/remote/anemones.jpg 3 times on 2.8Ghz P4
//          stb_jpeg:   1.34 seconds (MSVC6, default release build)
//          stb_jpeg:   1.06 seconds (MSVC6, processor = Pentium Pro)
//          IJL11.dll:  1.08 seconds (compiled by intel)
//          IJG 1998:   0.98 seconds (MSVC6, makefile provided by IJG)
//          IJG 1998:   0.95 seconds (MSVC6, makefile + proc=PPro)

// huffman decoding acceleration
#define FAST_BITS   9  // larger handles more cases; smaller stomps less cache

typedef struct
{
   uint8  fast[1 << FAST_BITS];
   // weirdly, repacking this into AoS is a 10% speed loss, instead of a win
   uint16 code[256];
   uint8  values[256];
   uint8  size[257];
   unsigned int maxcode[18];
   int    delta[17];   // old 'firstsymbol' - old 'firstcode'
} huffman;

typedef struct
{
   #ifdef STBI_SIMD
   unsigned short dequant2[4][64];
   #endif
   stbi *s;
   huffman huff_dc[4];
   huffman huff_ac[4];
   uint8 dequant[4][64];

// sizes for components, interleaved MCUs
   int img_h_max, img_v_max;
   int img_mcu_x, img_mcu_y;
   int img_mcu_w, img_mcu_h;

// definition of jpeg image component
   struct
   {
      int id;
      int h,v;
      int tq;
      int hd,ha;
      int dc_pred;

      int x,y,w2,h2;
      uint8 *data;
      void *raw_data;
      uint8 *linebuf;
   } img_comp[4];

   uint32         code_buffer; // jpeg entropy-coded buffer
   int            code_bits;   // number of valid bits
   unsigned char  marker;      // marker seen while filling entropy buffer
   int            nomore;      // flag if we saw a marker so must stop

   int scan_n, order[4];
   int restart_interval, todo;
} jpeg;

static int build_huffman(huffman *h, int *count)
{
   int i,j,k=0,code;
   // build size list for each symbol (from JPEG spec)
   for (i=0; i < 16; ++i)
      for (j=0; j < count[i]; ++j)
         h->size[k++] = (uint8) (i+1);
   h->size[k] = 0;

   // compute actual symbols (from jpeg spec)
   code = 0;
   k = 0;
   for(j=1; j <= 16; ++j) {
      // compute delta to add to code to compute symbol id
      h->delta[j] = k - code;
      if (h->size[k] == j) {
         while (h->size[k] == j)
            h->code[k++] = (uint16) (code++);
         if (code-1 >= (1 << j)) return e("bad code lengths","Corrupt JPEG");
      }
      // compute largest code + 1 for this size, preshifted as needed later
      h->maxcode[j] = code << (16-j);
      code <<= 1;
   }
   h->maxcode[j] = 0xffffffff;

   // build non-spec acceleration table; 255 is flag for not-accelerated
   memset(h->fast, 255, 1 << FAST_BITS);
   for (i=0; i < k; ++i) {
      int s = h->size[i];
      if (s <= FAST_BITS) {
         int c = h->code[i] << (FAST_BITS-s);
         int m = 1 << (FAST_BITS-s);
         for (j=0; j < m; ++j) {
            h->fast[c+j] = (uint8) i;
         }
      }
   }
   return 1;
}

static void grow_buffer_unsafe(jpeg *j)
{
   do {
      int b = j->nomore ? 0 : get8(j->s);
      if (b == 0xff) {
         int c = get8(j->s);
         if (c != 0) {
            j->marker = (unsigned char) c;
            j->nomore = 1;
            return;
         }
      }
      j->code_buffer |= b << (24 - j->code_bits);
      j->code_bits += 8;
   } while (j->code_bits <= 24);
}

// (1 << n) - 1
static uint32 bmask[17]={0,1,3,7,15,31,63,127,255,511,1023,2047,4095,8191,16383,32767,65535};

// decode a jpeg huffman value from the bitstream
stbi_inline static int decode(jpeg *j, huffman *h)
{
   unsigned int temp;
   int c,k;

   if (j->code_bits < 16) grow_buffer_unsafe(j);

   // look at the top FAST_BITS and determine what symbol ID it is,
   // if the code is <= FAST_BITS
   c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1);
   k = h->fast[c];
   if (k < 255) {
      int s = h->size[k];
      if (s > j->code_bits)
         return -1;
      j->code_buffer <<= s;
      j->code_bits -= s;
      return h->values[k];
   }

   // naive test is to shift the code_buffer down so k bits are
   // valid, then test against maxcode. To speed this up, we've
   // preshifted maxcode left so that it has (16-k) 0s at the
   // end; in other words, regardless of the number of bits, it
   // wants to be compared against something shifted to have 16;
   // that way we don't need to shift inside the loop.
   temp = j->code_buffer >> 16;
   for (k=FAST_BITS+1 ; ; ++k)
      if (temp < h->maxcode[k])
         break;
   if (k == 17) {
      // error! code not found
      j->code_bits -= 16;
      return -1;
   }

   if (k > j->code_bits)
      return -1;

   // convert the huffman code to the symbol id
   c = ((j->code_buffer >> (32 - k)) & bmask[k]) + h->delta[k];
   assert((((j->code_buffer) >> (32 - h->size[c])) & bmask[h->size[c]]) == h->code[c]);

   // convert the id to a symbol
   j->code_bits -= k;
   j->code_buffer <<= k;
   return h->values[c];
}

// combined JPEG 'receive' and JPEG 'extend', since baseline
// always extends everything it receives.
stbi_inline static int extend_receive(jpeg *j, int n)
{
   unsigned int m = 1 << (n-1);
   unsigned int k;
   if (j->code_bits < n) grow_buffer_unsafe(j);

   #if 1
   k = stbi_lrot(j->code_buffer, n);
   j->code_buffer = k & ~bmask[n];
   k &= bmask[n];
   j->code_bits -= n;
   #else
   k = (j->code_buffer >> (32 - n)) & bmask[n];
   j->code_bits -= n;
   j->code_buffer <<= n;
   #endif
   // the following test is probably a random branch that won't
   // predict well. I tried to table accelerate it but failed.
   // maybe it's compiling as a conditional move?
   if (k < m)
      return (-1 << n) + k + 1;
   else
      return k;
}

// given a value that's at position X in the zigzag stream,
// where does it appear in the 8x8 matrix coded as row-major?
static uint8 dezigzag[64+15] =
{
    0,  1,  8, 16,  9,  2,  3, 10,
   17, 24, 32, 25, 18, 11,  4,  5,
   12, 19, 26, 33, 40, 48, 41, 34,
   27, 20, 13,  6,  7, 14, 21, 28,
   35, 42, 49, 56, 57, 50, 43, 36,
   29, 22, 15, 23, 30, 37, 44, 51,
   58, 59, 52, 45, 38, 31, 39, 46,
   53, 60, 61, 54, 47, 55, 62, 63,
   // let corrupt input sample past end
   63, 63, 63, 63, 63, 63, 63, 63,
   63, 63, 63, 63, 63, 63, 63
};

// decode one 64-entry block--
static int decode_block(jpeg *j, short data[64], huffman *hdc, huffman *hac, int b)
{
   int diff,dc,k;
   int t = decode(j, hdc);
   if (t < 0) return e("bad huffman code","Corrupt JPEG");

   // 0 all the ac values now so we can do it 32-bits at a time
   memset(data,0,64*sizeof(data[0]));

   diff = t ? extend_receive(j, t) : 0;
   dc = j->img_comp[b].dc_pred + diff;
   j->img_comp[b].dc_pred = dc;
   data[0] = (short) dc;

   // decode AC components, see JPEG spec
   k = 1;
   do {
      int r,s;
      int rs = decode(j, hac);
      if (rs < 0) return e("bad huffman code","Corrupt JPEG");
      s = rs & 15;
      r = rs >> 4;
      if (s == 0) {
         if (rs != 0xf0) break; // end block
         k += 16;
      } else {
         k += r;
         // decode into unzigzag'd location
         data[dezigzag[k++]] = (short) extend_receive(j,s);
      }
   } while (k < 64);
   return 1;
}

// take a -128..127 value and clamp it and convert to 0..255
stbi_inline static uint8 clamp(int x)
{
   // trick to use a single test to catch both cases
   if ((unsigned int) x > 255) {
      if (x < 0) return 0;
      if (x > 255) return 255;
   }
   return (uint8) x;
}

#define f2f(x)  (int) (((x) * 4096 + 0.5))
#define fsh(x)  ((x) << 12)

// derived from jidctint -- DCT_ISLOW
#define IDCT_1D(s0,s1,s2,s3,s4,s5,s6,s7)       \
   int t0,t1,t2,t3,p1,p2,p3,p4,p5,x0,x1,x2,x3; \
   p2 = s2;                                    \
   p3 = s6;                                    \
   p1 = (p2+p3) * f2f(0.5411961f);             \
   t2 = p1 + p3*f2f(-1.847759065f);            \
   t3 = p1 + p2*f2f( 0.765366865f);            \
   p2 = s0;                                    \
   p3 = s4;                                    \
   t0 = fsh(p2+p3);                            \
   t1 = fsh(p2-p3);                            \
   x0 = t0+t3;                                 \
   x3 = t0-t3;                                 \
   x1 = t1+t2;                                 \
   x2 = t1-t2;                                 \
   t0 = s7;                                    \
   t1 = s5;                                    \
   t2 = s3;                                    \
   t3 = s1;                                    \
   p3 = t0+t2;                                 \
   p4 = t1+t3;                                 \
   p1 = t0+t3;                                 \
   p2 = t1+t2;                                 \
   p5 = (p3+p4)*f2f( 1.175875602f);            \
   t0 = t0*f2f( 0.298631336f);                 \
   t1 = t1*f2f( 2.053119869f);                 \
   t2 = t2*f2f( 3.072711026f);                 \
   t3 = t3*f2f( 1.501321110f);                 \
   p1 = p5 + p1*f2f(-0.899976223f);            \
   p2 = p5 + p2*f2f(-2.562915447f);            \
   p3 = p3*f2f(-1.961570560f);                 \
   p4 = p4*f2f(-0.390180644f);                 \
   t3 += p1+p4;                                \
   t2 += p2+p3;                                \
   t1 += p2+p4;                                \
   t0 += p1+p3;

#ifdef STBI_SIMD
typedef unsigned short stbi_dequantize_t;
#else
typedef uint8 stbi_dequantize_t;
#endif

// .344 seconds on 3*anemones.jpg
static void idct_block(uint8 *out, int out_stride, short data[64], stbi_dequantize_t *dequantize)
{
   int i,val[64],*v=val;
   stbi_dequantize_t *dq = dequantize;
   uint8 *o;
   short *d = data;

   // columns
   for (i=0; i < 8; ++i,++d,++dq, ++v) {
      // if all zeroes, shortcut -- this avoids dequantizing 0s and IDCTing
      if (d[ 8]==0 && d[16]==0 && d[24]==0 && d[32]==0
           && d[40]==0 && d[48]==0 && d[56]==0) {
         //    no shortcut                 0     seconds
         //    (1|2|3|4|5|6|7)==0          0     seconds
         //    all separate               -0.047 seconds
         //    1 && 2|3 && 4|5 && 6|7:    -0.047 seconds
         int dcterm = d[0] * dq[0] << 2;
         v[0] = v[8] = v[16] = v[24] = v[32] = v[40] = v[48] = v[56] = dcterm;
      } else {
         IDCT_1D(d[ 0]*dq[ 0],d[ 8]*dq[ 8],d[16]*dq[16],d[24]*dq[24],
                 d[32]*dq[32],d[40]*dq[40],d[48]*dq[48],d[56]*dq[56])
         // constants scaled things up by 1<<12; let's bring them back
         // down, but keep 2 extra bits of precision
         x0 += 512; x1 += 512; x2 += 512; x3 += 512;
         v[ 0] = (x0+t3) >> 10;
         v[56] = (x0-t3) >> 10;
         v[ 8] = (x1+t2) >> 10;
         v[48] = (x1-t2) >> 10;
         v[16] = (x2+t1) >> 10;
         v[40] = (x2-t1) >> 10;
         v[24] = (x3+t0) >> 10;
         v[32] = (x3-t0) >> 10;
      }
   }

   for (i=0, v=val, o=out; i < 8; ++i,v+=8,o+=out_stride) {
      // no fast case since the first 1D IDCT spread components out
      IDCT_1D(v[0],v[1],v[2],v[3],v[4],v[5],v[6],v[7])
      // constants scaled things up by 1<<12, plus we had 1<<2 from first
      // loop, plus horizontal and vertical each scale by sqrt(8) so together
      // we've got an extra 1<<3, so 1<<17 total we need to remove.
      // so we want to round that, which means adding 0.5 * 1<<17,
      // aka 65536. Also, we'll end up with -128 to 127 that we want
      // to encode as 0..255 by adding 128, so we'll add that before the shift
      x0 += 65536 + (128<<17);
      x1 += 65536 + (128<<17);
      x2 += 65536 + (128<<17);
      x3 += 65536 + (128<<17);
      // tried computing the shifts into temps, or'ing the temps to see
      // if any were out of range, but that was slower
      o[0] = clamp((x0+t3) >> 17);
      o[7] = clamp((x0-t3) >> 17);
      o[1] = clamp((x1+t2) >> 17);
      o[6] = clamp((x1-t2) >> 17);
      o[2] = clamp((x2+t1) >> 17);
      o[5] = clamp((x2-t1) >> 17);
      o[3] = clamp((x3+t0) >> 17);
      o[4] = clamp((x3-t0) >> 17);
   }
}

#ifdef STBI_SIMD
static stbi_idct_8x8 stbi_idct_installed = idct_block;

void stbi_install_idct(stbi_idct_8x8 func)
{
   stbi_idct_installed = func;
}
#endif

#define MARKER_none  0xff
// if there's a pending marker from the entropy stream, return that
// otherwise, fetch from the stream and get a marker. if there's no
// marker, return 0xff, which is never a valid marker value
static uint8 get_marker(jpeg *j)
{
   uint8 x;
   if (j->marker != MARKER_none) { x = j->marker; j->marker = MARKER_none; return x; }
   x = get8u(j->s);
   if (x != 0xff) return MARKER_none;
   while (x == 0xff)
      x = get8u(j->s);
   return x;
}

// in each scan, we'll have scan_n components, and the order
// of the components is specified by order[]
#define RESTART(x)     ((x) >= 0xd0 && (x) <= 0xd7)

// after a restart interval, reset the entropy decoder and
// the dc prediction
static void reset(jpeg *j)
{
   j->code_bits = 0;
   j->code_buffer = 0;
   j->nomore = 0;
   j->img_comp[0].dc_pred = j->img_comp[1].dc_pred = j->img_comp[2].dc_pred = 0;
   j->marker = MARKER_none;
   j->todo = j->restart_interval ? j->restart_interval : 0x7fffffff;
   // no more than 1<<31 MCUs if no restart_interal? that's plenty safe,
   // since we don't even allow 1<<30 pixels
}

static int parse_entropy_coded_data(jpeg *z)
{
   reset(z);
   if (z->scan_n == 1) {
      int i,j;
      #ifdef STBI_SIMD
      __declspec(align(16))
      #endif
      short data[64];
      int n = z->order[0];
      // non-interleaved data, we just need to process one block at a time,
      // in trivial scanline order
      // number of blocks to do just depends on how many actual "pixels" this
      // component has, independent of interleaved MCU blocking and such
      int w = (z->img_comp[n].x+7) >> 3;
      int h = (z->img_comp[n].y+7) >> 3;
      for (j=0; j < h; ++j) {
         for (i=0; i < w; ++i) {
            if (!decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+z->img_comp[n].ha, n)) return 0;
            #ifdef STBI_SIMD
            stbi_idct_installed(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data, z->dequant2[z->img_comp[n].tq]);
            #else
            idct_block(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data, z->dequant[z->img_comp[n].tq]);
            #endif
            // every data block is an MCU, so countdown the restart interval
            if (--z->todo <= 0) {
               if (z->code_bits < 24) grow_buffer_unsafe(z);
               // if it's NOT a restart, then just bail, so we get corrupt data
               // rather than no data
               if (!RESTART(z->marker)) return 1;
               reset(z);
            }
         }
      }
   } else { // interleaved!
      int i,j,k,x,y;
      short data[64];
      for (j=0; j < z->img_mcu_y; ++j) {
         for (i=0; i < z->img_mcu_x; ++i) {
            // scan an interleaved mcu... process scan_n components in order
            for (k=0; k < z->scan_n; ++k) {
               int n = z->order[k];
               // scan out an mcu's worth of this component; that's just determined
               // by the basic H and V specified for the component
               for (y=0; y < z->img_comp[n].v; ++y) {
                  for (x=0; x < z->img_comp[n].h; ++x) {
                     int x2 = (i*z->img_comp[n].h + x)*8;
                     int y2 = (j*z->img_comp[n].v + y)*8;
                     if (!decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+z->img_comp[n].ha, n)) return 0;
                     #ifdef STBI_SIMD
                     stbi_idct_installed(z->img_comp[n].data+z->img_comp[n].w2*y2+x2, z->img_comp[n].w2, data, z->dequant2[z->img_comp[n].tq]);
                     #else
                     idct_block(z->img_comp[n].data+z->img_comp[n].w2*y2+x2, z->img_comp[n].w2, data, z->dequant[z->img_comp[n].tq]);
                     #endif
                  }
               }
            }
            // after all interleaved components, that's an interleaved MCU,
            // so now count down the restart interval
            if (--z->todo <= 0) {
               if (z->code_bits < 24) grow_buffer_unsafe(z);
               // if it's NOT a restart, then just bail, so we get corrupt data
               // rather than no data
               if (!RESTART(z->marker)) return 1;
               reset(z);
            }
         }
      }
   }
   return 1;
}

static int process_marker(jpeg *z, int m)
{
   int L;
   switch (m) {
      case MARKER_none: // no marker found
         return e("expected marker","Corrupt JPEG");

      case 0xC2: // SOF - progressive
         return e("progressive jpeg","JPEG format not supported (progressive)");

      case 0xDD: // DRI - specify restart interval
         if (get16(z->s) != 4) return e("bad DRI len","Corrupt JPEG");
         z->restart_interval = get16(z->s);
         return 1;

      case 0xDB: // DQT - define quantization table
         L = get16(z->s)-2;
         while (L > 0) {
            int q = get8(z->s);
            int p = q >> 4;
            int t = q & 15,i;
            if (p != 0) return e("bad DQT type","Corrupt JPEG");
            if (t > 3) return e("bad DQT table","Corrupt JPEG");
            for (i=0; i < 64; ++i)
               z->dequant[t][dezigzag[i]] = get8u(z->s);
            #ifdef STBI_SIMD
            for (i=0; i < 64; ++i)
               z->dequant2[t][i] = z->dequant[t][i];
            #endif
            L -= 65;
         }
         return L==0;

      case 0xC4: // DHT - define huffman table
         L = get16(z->s)-2;
         while (L > 0) {
            uint8 *v;
            int sizes[16],i,m=0;
            int q = get8(z->s);
            int tc = q >> 4;
            int th = q & 15;
            if (tc > 1 || th > 3) return e("bad DHT header","Corrupt JPEG");
            for (i=0; i < 16; ++i) {
               sizes[i] = get8(z->s);
               m += sizes[i];
            }
            L -= 17;
            if (tc == 0) {
               if (!build_huffman(z->huff_dc+th, sizes)) return 0;
               v = z->huff_dc[th].values;
            } else {
               if (!build_huffman(z->huff_ac+th, sizes)) return 0;
               v = z->huff_ac[th].values;
            }
            for (i=0; i < m; ++i)
               v[i] = get8u(z->s);
            L -= m;
         }
         return L==0;
   }
   // check for comment block or APP blocks
   if ((m >= 0xE0 && m <= 0xEF) || m == 0xFE) {
      skip(z->s, get16(z->s)-2);
      return 1;
   }
   return 0;
}

// after we see SOS
static int process_scan_header(jpeg *z)
{
   int i;
   int Ls = get16(z->s);
   z->scan_n = get8(z->s);
   if (z->scan_n < 1 || z->scan_n > 4 || z->scan_n > (int) z->s->img_n) return e("bad SOS component count","Corrupt JPEG");
   if (Ls != 6+2*z->scan_n) return e("bad SOS len","Corrupt JPEG");
   for (i=0; i < z->scan_n; ++i) {
      int id = get8(z->s), which;
      int q = get8(z->s);
      for (which = 0; which < z->s->img_n; ++which)
         if (z->img_comp[which].id == id)
            break;
      if (which == z->s->img_n) return 0;
      z->img_comp[which].hd = q >> 4;   if (z->img_comp[which].hd > 3) return e("bad DC huff","Corrupt JPEG");
      z->img_comp[which].ha = q & 15;   if (z->img_comp[which].ha > 3) return e("bad AC huff","Corrupt JPEG");
      z->order[i] = which;
   }
   if (get8(z->s) != 0) return e("bad SOS","Corrupt JPEG");
   get8(z->s); // should be 63, but might be 0
   if (get8(z->s) != 0) return e("bad SOS","Corrupt JPEG");

   return 1;
}

static int process_frame_header(jpeg *z, int scan)
{
   stbi *s = z->s;
   int Lf,p,i,q, h_max=1,v_max=1,c;
   Lf = get16(s);         if (Lf < 11) return e("bad SOF len","Corrupt JPEG"); // JPEG
   p  = get8(s);          if (p != 8) return e("only 8-bit","JPEG format not supported: 8-bit only"); // JPEG baseline
   s->img_y = get16(s);   if (s->img_y == 0) return e("no header height", "JPEG format not supported: delayed height"); // Legal, but we don't handle it--but neither does IJG
   s->img_x = get16(s);   if (s->img_x == 0) return e("0 width","Corrupt JPEG"); // JPEG requires
   c = get8(s);
   if (c != 3 && c != 1) return e("bad component count","Corrupt JPEG");    // JFIF requires
   s->img_n = c;
   for (i=0; i < c; ++i) {
      z->img_comp[i].data = NULL;
      z->img_comp[i].linebuf = NULL;
   }

   if (Lf != 8+3*s->img_n) return e("bad SOF len","Corrupt JPEG");

   for (i=0; i < s->img_n; ++i) {
      z->img_comp[i].id = get8(s);
      if (z->img_comp[i].id != i+1)   // JFIF requires
         if (z->img_comp[i].id != i)  // some version of jpegtran outputs non-JFIF-compliant files!
            return e("bad component ID","Corrupt JPEG");
      q = get8(s);
      z->img_comp[i].h = (q >> 4);  if (!z->img_comp[i].h || z->img_comp[i].h > 4) return e("bad H","Corrupt JPEG");
      z->img_comp[i].v = q & 15;    if (!z->img_comp[i].v || z->img_comp[i].v > 4) return e("bad V","Corrupt JPEG");
      z->img_comp[i].tq = get8(s);  if (z->img_comp[i].tq > 3) return e("bad TQ","Corrupt JPEG");
   }

   if (scan != SCAN_load) return 1;

   if ((1 << 30) / s->img_x / s->img_n < s->img_y) return e("too large", "Image too large to decode");

   for (i=0; i < s->img_n; ++i) {
      if (z->img_comp[i].h > h_max) h_max = z->img_comp[i].h;
      if (z->img_comp[i].v > v_max) v_max = z->img_comp[i].v;
   }

   // compute interleaved mcu info
   z->img_h_max = h_max;
   z->img_v_max = v_max;
   z->img_mcu_w = h_max * 8;
   z->img_mcu_h = v_max * 8;
   z->img_mcu_x = (s->img_x + z->img_mcu_w-1) / z->img_mcu_w;
   z->img_mcu_y = (s->img_y + z->img_mcu_h-1) / z->img_mcu_h;

   for (i=0; i < s->img_n; ++i) {
      // number of effective pixels (e.g. for non-interleaved MCU)
      z->img_comp[i].x = (s->img_x * z->img_comp[i].h + h_max-1) / h_max;
      z->img_comp[i].y = (s->img_y * z->img_comp[i].v + v_max-1) / v_max;
      // to simplify generation, we'll allocate enough memory to decode
      // the bogus oversized data from using interleaved MCUs and their
      // big blocks (e.g. a 16x16 iMCU on an image of width 33); we won't
      // discard the extra data until colorspace conversion
      z->img_comp[i].w2 = z->img_mcu_x * z->img_comp[i].h * 8;
      z->img_comp[i].h2 = z->img_mcu_y * z->img_comp[i].v * 8;
      z->img_comp[i].raw_data = malloc(z->img_comp[i].w2 * z->img_comp[i].h2+15);
      if (z->img_comp[i].raw_data == NULL) {
         for(--i; i >= 0; --i) {
            free(z->img_comp[i].raw_data);
            z->img_comp[i].data = NULL;
         }
         return e("outofmem", "Out of memory");
      }
      // align blocks for installable-idct using mmx/sse
      z->img_comp[i].data = (uint8*) (((size_t) z->img_comp[i].raw_data + 15) & ~15);
      z->img_comp[i].linebuf = NULL;
   }

   return 1;
}

// use comparisons since in some cases we handle more than one case (e.g. SOF)
#define DNL(x)         ((x) == 0xdc)
#define SOI(x)         ((x) == 0xd8)
#define EOI(x)         ((x) == 0xd9)
#define SOF(x)         ((x) == 0xc0 || (x) == 0xc1)
#define SOS(x)         ((x) == 0xda)

static int decode_jpeg_header(jpeg *z, int scan)
{
   int m;
   z->marker = MARKER_none; // initialize cached marker to empty
   m = get_marker(z);
   if (!SOI(m)) return e("no SOI","Corrupt JPEG");
   if (scan == SCAN_type) return 1;
   m = get_marker(z);
   while (!SOF(m)) {
      if (!process_marker(z,m)) return 0;
      m = get_marker(z);
      while (m == MARKER_none) {
         // some files have extra padding after their blocks, so ok, we'll scan
         if (at_eof(z->s)) return e("no SOF", "Corrupt JPEG");
         m = get_marker(z);
      }
   }
   if (!process_frame_header(z, scan)) return 0;
   return 1;
}

static int decode_jpeg_image(jpeg *j)
{
   int m;
   j->restart_interval = 0;
   if (!decode_jpeg_header(j, SCAN_load)) return 0;
   m = get_marker(j);
   while (!EOI(m)) {
      if (SOS(m)) {
         if (!process_scan_header(j)) return 0;
         if (!parse_entropy_coded_data(j)) return 0;
         if (j->marker == MARKER_none ) {
            // handle 0s at the end of image data from IP Kamera 9060
            while (!at_eof(j->s)) {
               int x = get8(j->s);
               if (x == 255) {
                  j->marker = get8u(j->s);
                  break;
               } else if (x != 0) {
                  return 0;
               }
            }
            // if we reach eof without hitting a marker, get_marker() below will fail and we'll eventually return 0
         }
      } else {
         if (!process_marker(j, m)) return 0;
      }
      m = get_marker(j);
   }
   return 1;
}

// static jfif-centered resampling (across block boundaries)

typedef uint8 *(*resample_row_func)(uint8 *out, uint8 *in0, uint8 *in1,
                                    int w, int hs);

#define div4(x) ((uint8) ((x) >> 2))

static uint8 *resample_row_1(uint8 *out, uint8 *in_near, uint8 *in_far, int w, int hs)
{
   STBI_NOTUSED(out);
   STBI_NOTUSED(in_far);
   STBI_NOTUSED(w);
   STBI_NOTUSED(hs);
   return in_near;
}

static uint8* resample_row_v_2(uint8 *out, uint8 *in_near, uint8 *in_far, int w, int hs)
{
   // need to generate two samples vertically for every one in input
   int i;
   STBI_NOTUSED(hs);
   for (i=0; i < w; ++i)
      out[i] = div4(3*in_near[i] + in_far[i] + 2);
   return out;
}

static uint8*  resample_row_h_2(uint8 *out, uint8 *in_near, uint8 *in_far, int w, int hs)
{
   // need to generate two samples horizontally for every one in input
   int i;
   uint8 *input = in_near;

   if (w == 1) {
      // if only one sample, can't do any interpolation
      out[0] = out[1] = input[0];
      return out;
   }

   out[0] = input[0];
   out[1] = div4(input[0]*3 + input[1] + 2);
   for (i=1; i < w-1; ++i) {
      int n = 3*input[i]+2;
      out[i*2+0] = div4(n+input[i-1]);
      out[i*2+1] = div4(n+input[i+1]);
   }
   out[i*2+0] = div4(input[w-2]*3 + input[w-1] + 2);
   out[i*2+1] = input[w-1];

   STBI_NOTUSED(in_far);
   STBI_NOTUSED(hs);

   return out;
}

#define div16(x) ((uint8) ((x) >> 4))

static uint8 *resample_row_hv_2(uint8 *out, uint8 *in_near, uint8 *in_far, int w, int hs)
{
   // need to generate 2x2 samples for every one in input
   int i,t0,t1;
   if (w == 1) {
      out[0] = out[1] = div4(3*in_near[0] + in_far[0] + 2);
      return out;
   }

   t1 = 3*in_near[0] + in_far[0];
   out[0] = div4(t1+2);
   for (i=1; i < w; ++i) {
      t0 = t1;
      t1 = 3*in_near[i]+in_far[i];
      out[i*2-1] = div16(3*t0 + t1 + 8);
      out[i*2  ] = div16(3*t1 + t0 + 8);
   }
   out[w*2-1] = div4(t1+2);

   STBI_NOTUSED(hs);

   return out;
}

static uint8 *resample_row_generic(uint8 *out, uint8 *in_near, uint8 *in_far, int w, int hs)
{
   // resample with nearest-neighbor
   int i,j;
   in_far = in_far;
   for (i=0; i < w; ++i)
      for (j=0; j < hs; ++j)
         out[i*hs+j] = in_near[i];
   return out;
}

#define float2fixed(x)  ((int) ((x) * 65536 + 0.5))

// 0.38 seconds on 3*anemones.jpg   (0.25 with processor = Pro)
// VC6 without processor=Pro is generating multiple LEAs per multiply!
static void YCbCr_to_RGB_row(uint8 *out, const uint8 *y, const uint8 *pcb, const uint8 *pcr, int count, int step)
{
   int i;
   for (i=0; i < count; ++i) {
      int y_fixed = (y[i] << 16) + 32768; // rounding
      int r,g,b;
      int cr = pcr[i] - 128;
      int cb = pcb[i] - 128;
      r = y_fixed + cr*float2fixed(1.40200f);
      g = y_fixed - cr*float2fixed(0.71414f) - cb*float2fixed(0.34414f);
      b = y_fixed                            + cb*float2fixed(1.77200f);
      r >>= 16;
      g >>= 16;
      b >>= 16;
      if ((unsigned) r > 255) { if (r < 0) r = 0; else r = 255; }
      if ((unsigned) g > 255) { if (g < 0) g = 0; else g = 255; }
      if ((unsigned) b > 255) { if (b < 0) b = 0; else b = 255; }
      out[0] = (uint8)r;
      out[1] = (uint8)g;
      out[2] = (uint8)b;
      out[3] = 255;
      out += step;
   }
}

#ifdef STBI_SIMD
static stbi_YCbCr_to_RGB_run stbi_YCbCr_installed = YCbCr_to_RGB_row;

void stbi_install_YCbCr_to_RGB(stbi_YCbCr_to_RGB_run func)
{
   stbi_YCbCr_installed = func;
}
#endif


// clean up the temporary component buffers
static void cleanup_jpeg(jpeg *j)
{
   int i;
   for (i=0; i < j->s->img_n; ++i) {
      if (j->img_comp[i].data) {
         free(j->img_comp[i].raw_data);
         j->img_comp[i].data = NULL;
      }
      if (j->img_comp[i].linebuf) {
         free(j->img_comp[i].linebuf);
         j->img_comp[i].linebuf = NULL;
      }
   }
}

typedef struct
{
   resample_row_func resample;
   uint8 *line0,*line1;
   int hs,vs;   // expansion factor in each axis
   int w_lores; // horizontal pixels pre-expansion
   int ystep;   // how far through vertical expansion we are
   int ypos;    // which pre-expansion row we're on
} stbi_resample;

static uint8 *load_jpeg_image(jpeg *z, int *out_x, int *out_y, int *comp, int req_comp)
{
   int n, decode_n;
   // validate req_comp
   if (req_comp < 0 || req_comp > 4) return epuc("bad req_comp", "Internal error");
   z->s->img_n = 0;

   // load a jpeg image from whichever source
   if (!decode_jpeg_image(z)) { cleanup_jpeg(z); return NULL; }

   // determine actual number of components to generate
   n = req_comp ? req_comp : z->s->img_n;

   if (z->s->img_n == 3 && n < 3)
      decode_n = 1;
   else
      decode_n = z->s->img_n;

   // resample and color-convert
   {
      int k;
      uint i,j;
      uint8 *output;
      uint8 *coutput[4];

      stbi_resample res_comp[4];

      for (k=0; k < decode_n; ++k) {
         stbi_resample *r = &res_comp[k];

         // allocate line buffer big enough for upsampling off the edges
         // with upsample factor of 4
         z->img_comp[k].linebuf = (uint8 *) malloc(z->s->img_x + 3);
         if (!z->img_comp[k].linebuf) { cleanup_jpeg(z); return epuc("outofmem", "Out of memory"); }

         r->hs      = z->img_h_max / z->img_comp[k].h;
         r->vs      = z->img_v_max / z->img_comp[k].v;
         r->ystep   = r->vs >> 1;
         r->w_lores = (z->s->img_x + r->hs-1) / r->hs;
         r->ypos    = 0;
         r->line0   = r->line1 = z->img_comp[k].data;

         if      (r->hs == 1 && r->vs == 1) r->resample = resample_row_1;
         else if (r->hs == 1 && r->vs == 2) r->resample = resample_row_v_2;
         else if (r->hs == 2 && r->vs == 1) r->resample = resample_row_h_2;
         else if (r->hs == 2 && r->vs == 2) r->resample = resample_row_hv_2;
         else                               r->resample = resample_row_generic;
      }

      // can't error after this so, this is safe
      output = (uint8 *) malloc(n * z->s->img_x * z->s->img_y + 1);
      if (!output) { cleanup_jpeg(z); return epuc("outofmem", "Out of memory"); }

      // now go ahead and resample
      for (j=0; j < z->s->img_y; ++j) {
         uint8 *out = output + n * z->s->img_x * j;
         for (k=0; k < decode_n; ++k) {
            stbi_resample *r = &res_comp[k];
            int y_bot = r->ystep >= (r->vs >> 1);
            coutput[k] = r->resample(z->img_comp[k].linebuf,
                                     y_bot ? r->line1 : r->line0,
                                     y_bot ? r->line0 : r->line1,
                                     r->w_lores, r->hs);
            if (++r->ystep >= r->vs) {
               r->ystep = 0;
               r->line0 = r->line1;
               if (++r->ypos < z->img_comp[k].y)
                  r->line1 += z->img_comp[k].w2;
            }
         }
         if (n >= 3) {
            uint8 *y = coutput[0];
            if (z->s->img_n == 3) {
               #ifdef STBI_SIMD
               stbi_YCbCr_installed(out, y, coutput[1], coutput[2], z->s.img_x, n);
               #else
               YCbCr_to_RGB_row(out, y, coutput[1], coutput[2], z->s->img_x, n);
               #endif
            } else
               for (i=0; i < z->s->img_x; ++i) {
                  out[0] = out[1] = out[2] = y[i];
                  out[3] = 255; // not used if n==3
                  out += n;
               }
         } else {
            uint8 *y = coutput[0];
            if (n == 1)
               for (i=0; i < z->s->img_x; ++i) out[i] = y[i];
            else
               for (i=0; i < z->s->img_x; ++i) *out++ = y[i], *out++ = 255;
         }
      }
      cleanup_jpeg(z);
      *out_x = z->s->img_x;
      *out_y = z->s->img_y;
      if (comp) *comp  = z->s->img_n; // report original components, not output
      return output;
   }
}

static unsigned char *stbi_jpeg_load(stbi *s, int *x, int *y, int *comp, int req_comp)
{
   jpeg j;
   j.s = s;
   return load_jpeg_image(&j, x,y,comp,req_comp);
}

static int stbi_jpeg_test(stbi *s)
{
   int r;
   jpeg j;
   j.s = s;
   r = decode_jpeg_header(&j, SCAN_type);
   stbi_rewind(s);
   return r;
}

static int stbi_jpeg_info_raw(jpeg *j, int *x, int *y, int *comp)
{
   if (!decode_jpeg_header(j, SCAN_header)) {
      stbi_rewind( j->s );
      return 0;
   }
   if (x) *x = j->s->img_x;
   if (y) *y = j->s->img_y;
   if (comp) *comp = j->s->img_n;
   return 1;
}

static int stbi_jpeg_info(stbi *s, int *x, int *y, int *comp)
{
   jpeg j;
   j.s = s;
   return stbi_jpeg_info_raw(&j, x, y, comp);
}

// public domain zlib decode    v0.2  Sean Barrett 2006-11-18
//    simple implementation
//      - all input must be provided in an upfront buffer
//      - all output is written to a single output buffer (can malloc/realloc)
//    performance
//      - fast huffman

// fast-way is faster to check than jpeg huffman, but slow way is slower
#define ZFAST_BITS  9 // accelerate all cases in default tables
#define ZFAST_MASK  ((1 << ZFAST_BITS) - 1)

// zlib-style huffman encoding
// (jpegs packs from left, zlib from right, so can't share code)
typedef struct
{
   uint16 fast[1 << ZFAST_BITS];
   uint16 firstcode[16];
   int maxcode[17];
   uint16 firstsymbol[16];
   uint8  size[288];
   uint16 value[288];
} zhuffman;

stbi_inline static int bitreverse16(int n)
{
  n = ((n & 0xAAAA) >>  1) | ((n & 0x5555) << 1);
  n = ((n & 0xCCCC) >>  2) | ((n & 0x3333) << 2);
  n = ((n & 0xF0F0) >>  4) | ((n & 0x0F0F) << 4);
  n = ((n & 0xFF00) >>  8) | ((n & 0x00FF) << 8);
  return n;
}

stbi_inline static int bit_reverse(int v, int bits)
{
   assert(bits <= 16);
   // to bit reverse n bits, reverse 16 and shift
   // e.g. 11 bits, bit reverse and shift away 5
   return bitreverse16(v) >> (16-bits);
}

static int zbuild_huffman(zhuffman *z, uint8 *sizelist, int num)
{
   int i,k=0;
   int code, next_code[16], sizes[17];

   // DEFLATE spec for generating codes
   memset(sizes, 0, sizeof(sizes));
   memset(z->fast, 255, sizeof(z->fast));
   for (i=0; i < num; ++i)
      ++sizes[sizelist[i]];
   sizes[0] = 0;
   for (i=1; i < 16; ++i)
      assert(sizes[i] <= (1 << i));
   code = 0;
   for (i=1; i < 16; ++i) {
      next_code[i] = code;
      z->firstcode[i] = (uint16) code;
      z->firstsymbol[i] = (uint16) k;
      code = (code + sizes[i]);
      if (sizes[i])
         if (code-1 >= (1 << i)) return e("bad codelengths","Corrupt JPEG");
      z->maxcode[i] = code << (16-i); // preshift for inner loop
      code <<= 1;
      k += sizes[i];
   }
   z->maxcode[16] = 0x10000; // sentinel
   for (i=0; i < num; ++i) {
      int s = sizelist[i];
      if (s) {
         int c = next_code[s] - z->firstcode[s] + z->firstsymbol[s];
         z->size[c] = (uint8)s;
         z->value[c] = (uint16)i;
         if (s <= ZFAST_BITS) {
            int k = bit_reverse(next_code[s],s);
            while (k < (1 << ZFAST_BITS)) {
               z->fast[k] = (uint16) c;
               k += (1 << s);
            }
         }
         ++next_code[s];
      }
   }
   return 1;
}

// zlib-from-memory implementation for PNG reading
//    because PNG allows splitting the zlib stream arbitrarily,
//    and it's annoying structurally to have PNG call ZLIB call PNG,
//    we require PNG read all the IDATs and combine them into a single
//    memory buffer

typedef struct
{
   uint8 *zbuffer, *zbuffer_end;
   int num_bits;
   uint32 code_buffer;

   char *zout;
   char *zout_start;
   char *zout_end;
   int   z_expandable;

   zhuffman z_length, z_distance;
} zbuf;

stbi_inline static int zget8(zbuf *z)
{
   if (z->zbuffer >= z->zbuffer_end) return 0;
   return *z->zbuffer++;
}

static void fill_bits(zbuf *z)
{
   do {
      assert(z->code_buffer < (1U << z->num_bits));
      z->code_buffer |= zget8(z) << z->num_bits;
      z->num_bits += 8;
   } while (z->num_bits <= 24);
}

stbi_inline static unsigned int zreceive(zbuf *z, int n)
{
   unsigned int k;
   if (z->num_bits < n) fill_bits(z);
   k = z->code_buffer & ((1 << n) - 1);
   z->code_buffer >>= n;
   z->num_bits -= n;
   return k;
}

stbi_inline static int zhuffman_decode(zbuf *a, zhuffman *z)
{
   int b,s,k;
   if (a->num_bits < 16) fill_bits(a);
   b = z->fast[a->code_buffer & ZFAST_MASK];
   if (b < 0xffff) {
      s = z->size[b];
      a->code_buffer >>= s;
      a->num_bits -= s;
      return z->value[b];
   }

   // not resolved by fast table, so compute it the slow way
   // use jpeg approach, which requires MSbits at top
   k = bit_reverse(a->code_buffer, 16);
   for (s=ZFAST_BITS+1; ; ++s)
      if (k < z->maxcode[s])
         break;
   if (s == 16) return -1; // invalid code!
   // code size is s, so:
   b = (k >> (16-s)) - z->firstcode[s] + z->firstsymbol[s];
   assert(z->size[b] == s);
   a->code_buffer >>= s;
   a->num_bits -= s;
   return z->value[b];
}

static int expand(zbuf *z, int n)  // need to make room for n bytes
{
   char *q;
   int cur, limit;
   if (!z->z_expandable) return e("output buffer limit","Corrupt PNG");
   cur   = (int) (z->zout     - z->zout_start);
   limit = (int) (z->zout_end - z->zout_start);
   while (cur + n > limit)
      limit *= 2;
   q = (char *) realloc(z->zout_start, limit);
   if (q == NULL) return e("outofmem", "Out of memory");
   z->zout_start = q;
   z->zout       = q + cur;
   z->zout_end   = q + limit;
   return 1;
}

static int length_base[31] = {
   3,4,5,6,7,8,9,10,11,13,
   15,17,19,23,27,31,35,43,51,59,
   67,83,99,115,131,163,195,227,258,0,0 };

static int length_extra[31]=
{ 0,0,0,0,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,5,5,5,5,0,0,0 };

static int dist_base[32] = { 1,2,3,4,5,7,9,13,17,25,33,49,65,97,129,193,
257,385,513,769,1025,1537,2049,3073,4097,6145,8193,12289,16385,24577,0,0};

static int dist_extra[32] =
{ 0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13};

static int parse_huffman_block(zbuf *a)
{
   for(;;) {
      int z = zhuffman_decode(a, &a->z_length);
      if (z < 256) {
         if (z < 0) return e("bad huffman code","Corrupt PNG"); // error in huffman codes
         if (a->zout >= a->zout_end) if (!expand(a, 1)) return 0;
         *a->zout++ = (char) z;
      } else {
         uint8 *p;
         int len,dist;
         if (z == 256) return 1;
         z -= 257;
         len = length_base[z];
         if (length_extra[z]) len += zreceive(a, length_extra[z]);
         z = zhuffman_decode(a, &a->z_distance);
         if (z < 0) return e("bad huffman code","Corrupt PNG");
         dist = dist_base[z];
         if (dist_extra[z]) dist += zreceive(a, dist_extra[z]);
         if (a->zout - a->zout_start < dist) return e("bad dist","Corrupt PNG");
         if (a->zout + len > a->zout_end) if (!expand(a, len)) return 0;
         p = (uint8 *) (a->zout - dist);
         while (len--)
            *a->zout++ = *p++;
      }
   }
}

static int compute_huffman_codes(zbuf *a)
{
   static uint8 length_dezigzag[19] = { 16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15 };
   zhuffman z_codelength;
   uint8 lencodes[286+32+137];//padding for maximum single op
   uint8 codelength_sizes[19];
   int i,n;

   int hlit  = zreceive(a,5) + 257;
   int hdist = zreceive(a,5) + 1;
   int hclen = zreceive(a,4) + 4;

   memset(codelength_sizes, 0, sizeof(codelength_sizes));
   for (i=0; i < hclen; ++i) {
      int s = zreceive(a,3);
      codelength_sizes[length_dezigzag[i]] = (uint8) s;
   }
   if (!zbuild_huffman(&z_codelength, codelength_sizes, 19)) return 0;

   n = 0;
   while (n < hlit + hdist) {
      int c = zhuffman_decode(a, &z_codelength);
      assert(c >= 0 && c < 19);
      if (c < 16)
         lencodes[n++] = (uint8) c;
      else if (c == 16) {
         c = zreceive(a,2)+3;
         memset(lencodes+n, lencodes[n-1], c);
         n += c;
      } else if (c == 17) {
         c = zreceive(a,3)+3;
         memset(lencodes+n, 0, c);
         n += c;
      } else {
         assert(c == 18);
         c = zreceive(a,7)+11;
         memset(lencodes+n, 0, c);
         n += c;
      }
   }
   if (n != hlit+hdist) return e("bad codelengths","Corrupt PNG");
   if (!zbuild_huffman(&a->z_length, lencodes, hlit)) return 0;
   if (!zbuild_huffman(&a->z_distance, lencodes+hlit, hdist)) return 0;
   return 1;
}

static int parse_uncompressed_block(zbuf *a)
{
   uint8 header[4];
   int len,nlen,k;
   if (a->num_bits & 7)
      zreceive(a, a->num_bits & 7); // discard
   // drain the bit-packed data into header
   k = 0;
   while (a->num_bits > 0) {
      header[k++] = (uint8) (a->code_buffer & 255); // wtf this warns?
      a->code_buffer >>= 8;
      a->num_bits -= 8;
   }
   assert(a->num_bits == 0);
   // now fill header the normal way
   while (k < 4)
      header[k++] = (uint8) zget8(a);
   len  = header[1] * 256 + header[0];
   nlen = header[3] * 256 + header[2];
   if (nlen != (len ^ 0xffff)) return e("zlib corrupt","Corrupt PNG");
   if (a->zbuffer + len > a->zbuffer_end) return e("read past buffer","Corrupt PNG");
   if (a->zout + len > a->zout_end)
      if (!expand(a, len)) return 0;
   memcpy(a->zout, a->zbuffer, len);
   a->zbuffer += len;
   a->zout += len;
   return 1;
}

static int parse_zlib_header(zbuf *a)
{
   int cmf   = zget8(a);
   int cm    = cmf & 15;
   /* int cinfo = cmf >> 4; */
   int flg   = zget8(a);
   if ((cmf*256+flg) % 31 != 0) return e("bad zlib header","Corrupt PNG"); // zlib spec
   if (flg & 32) return e("no preset dict","Corrupt PNG"); // preset dictionary not allowed in png
   if (cm != 8) return e("bad compression","Corrupt PNG"); // DEFLATE required for png
   // window = 1 << (8 + cinfo)... but who cares, we fully buffer output
   return 1;
}

// @TODO: should statically initialize these for optimal thread safety
static uint8 default_length[288], default_distance[32];
static void init_defaults(void)
{
   int i;   // use <= to match clearly with spec
   for (i=0; i <= 143; ++i)     default_length[i]   = 8;
   for (   ; i <= 255; ++i)     default_length[i]   = 9;
   for (   ; i <= 279; ++i)     default_length[i]   = 7;
   for (   ; i <= 287; ++i)     default_length[i]   = 8;

   for (i=0; i <=  31; ++i)     default_distance[i] = 5;
}

int stbi_png_partial; // a quick hack to only allow decoding some of a PNG... I should implement real streaming support instead
static int parse_zlib(zbuf *a, int parse_header)
{
   int final, type;
   if (parse_header)
      if (!parse_zlib_header(a)) return 0;
   a->num_bits = 0;
   a->code_buffer = 0;
   do {
      final = zreceive(a,1);
      type = zreceive(a,2);
      if (type == 0) {
         if (!parse_uncompressed_block(a)) return 0;
      } else if (type == 3) {
         return 0;
      } else {
         if (type == 1) {
            // use fixed code lengths
            if (!default_distance[31]) init_defaults();
            if (!zbuild_huffman(&a->z_length  , default_length  , 288)) return 0;
            if (!zbuild_huffman(&a->z_distance, default_distance,  32)) return 0;
         } else {
            if (!compute_huffman_codes(a)) return 0;
         }
         if (!parse_huffman_block(a)) return 0;
      }
      if (stbi_png_partial && a->zout - a->zout_start > 65536)
         break;
   } while (!final);
   return 1;
}

static int do_zlib(zbuf *a, char *obuf, int olen, int exp, int parse_header)
{
   a->zout_start = obuf;
   a->zout       = obuf;
   a->zout_end   = obuf + olen;
   a->z_expandable = exp;

   return parse_zlib(a, parse_header);
}

char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen)
{
   zbuf a;
   char *p = (char *) malloc(initial_size);
   if (p == NULL) return NULL;
   a.zbuffer = (uint8 *) buffer;
   a.zbuffer_end = (uint8 *) buffer + len;
   if (do_zlib(&a, p, initial_size, 1, 1)) {
      if (outlen) *outlen = (int) (a.zout - a.zout_start);
      return a.zout_start;
   } else {
      free(a.zout_start);
      return NULL;
   }
}

char *stbi_zlib_decode_malloc(char const *buffer, int len, int *outlen)
{
   return stbi_zlib_decode_malloc_guesssize(buffer, len, 16384, outlen);
}

char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header)
{
   zbuf a;
   char *p = (char *) malloc(initial_size);
   if (p == NULL) return NULL;
   a.zbuffer = (uint8 *) buffer;
   a.zbuffer_end = (uint8 *) buffer + len;
   if (do_zlib(&a, p, initial_size, 1, parse_header)) {
      if (outlen) *outlen = (int) (a.zout - a.zout_start);
      return a.zout_start;
   } else {
      free(a.zout_start);
      return NULL;
   }
}

int stbi_zlib_decode_buffer(char *obuffer, int olen, char const *ibuffer, int ilen)
{
   zbuf a;
   a.zbuffer = (uint8 *) ibuffer;
   a.zbuffer_end = (uint8 *) ibuffer + ilen;
   if (do_zlib(&a, obuffer, olen, 0, 1))
      return (int) (a.zout - a.zout_start);
   else
      return -1;
}

char *stbi_zlib_decode_noheader_malloc(char const *buffer, int len, int *outlen)
{
   zbuf a;
   char *p = (char *) malloc(16384);
   if (p == NULL) return NULL;
   a.zbuffer = (uint8 *) buffer;
   a.zbuffer_end = (uint8 *) buffer+len;
   if (do_zlib(&a, p, 16384, 1, 0)) {
      if (outlen) *outlen = (int) (a.zout - a.zout_start);
      return a.zout_start;
   } else {
      free(a.zout_start);
      return NULL;
   }
}

int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen)
{
   zbuf a;
   a.zbuffer = (uint8 *) ibuffer;
   a.zbuffer_end = (uint8 *) ibuffer + ilen;
   if (do_zlib(&a, obuffer, olen, 0, 0))
      return (int) (a.zout - a.zout_start);
   else
      return -1;
}

// public domain "baseline" PNG decoder   v0.10  Sean Barrett 2006-11-18
//    simple implementation
//      - only 8-bit samples
//      - no CRC checking
//      - allocates lots of intermediate memory
//        - avoids problem of streaming data between subsystems
//        - avoids explicit window management
//    performance
//      - uses stb_zlib, a PD zlib implementation with fast huffman decoding


typedef struct
{
   uint32 length;
   uint32 type;
} chunk;

#define PNG_TYPE(a,b,c,d)  (((a) << 24) + ((b) << 16) + ((c) << 8) + (d))

static chunk get_chunk_header(stbi *s)
{
   chunk c;
   c.length = get32(s);
   c.type   = get32(s);
   return c;
}

static int check_png_header(stbi *s)
{
   static uint8 png_sig[8] = { 137,80,78,71,13,10,26,10 };
   int i;
   for (i=0; i < 8; ++i)
      if (get8u(s) != png_sig[i]) return e("bad png sig","Not a PNG");
   return 1;
}

typedef struct
{
   stbi *s;
   uint8 *idata, *expanded, *out;
} png;


enum {
   F_none=0, F_sub=1, F_up=2, F_avg=3, F_paeth=4,
   F_avg_first, F_paeth_first
};

static uint8 first_row_filter[5] =
{
   F_none, F_sub, F_none, F_avg_first, F_paeth_first
};

static int paeth(int a, int b, int c)
{
   int p = a + b - c;
   int pa = abs(p-a);
   int pb = abs(p-b);
   int pc = abs(p-c);
   if (pa <= pb && pa <= pc) return a;
   if (pb <= pc) return b;
   return c;
}

// create the png data from post-deflated data
static int create_png_image_raw(png *a, uint8 *raw, uint32 raw_len, int out_n, uint32 x, uint32 y)
{
   stbi *s = a->s;
   uint32 i,j,stride = x*out_n;
   int k;
   int img_n = s->img_n; // copy it into a local for later
   assert(out_n == s->img_n || out_n == s->img_n+1);
   if (stbi_png_partial) y = 1;
   a->out = (uint8 *) malloc(x * y * out_n);
   if (!a->out) return e("outofmem", "Out of memory");
   if (!stbi_png_partial) {
      if (s->img_x == x && s->img_y == y) {
         if (raw_len != (img_n * x + 1) * y) return e("not enough pixels","Corrupt PNG");
      } else { // interlaced:
         if (raw_len < (img_n * x + 1) * y) return e("not enough pixels","Corrupt PNG");
      }
   }
   for (j=0; j < y; ++j) {
      uint8 *cur = a->out + stride*j;
      uint8 *prior = cur - stride;
      int filter = *raw++;
      if (filter > 4) return e("invalid filter","Corrupt PNG");
      // if first row, use special filter that doesn't sample previous row
      if (j == 0) filter = first_row_filter[filter];
      // handle first pixel explicitly
      for (k=0; k < img_n; ++k) {
         switch (filter) {
            case F_none       : cur[k] = raw[k]; break;
            case F_sub        : cur[k] = raw[k]; break;
            case F_up         : cur[k] = raw[k] + prior[k]; break;
            case F_avg        : cur[k] = raw[k] + (prior[k]>>1); break;
            case F_paeth      : cur[k] = (uint8) (raw[k] + paeth(0,prior[k],0)); break;
            case F_avg_first  : cur[k] = raw[k]; break;
            case F_paeth_first: cur[k] = raw[k]; break;
         }
      }
      if (img_n != out_n) cur[img_n] = 255;
      raw += img_n;
      cur += out_n;
      prior += out_n;
      // this is a little gross, so that we don't switch per-pixel or per-component
      if (img_n == out_n) {
         #define CASE(f) \
             case f:     \
                for (i=x-1; i >= 1; --i, raw+=img_n,cur+=img_n,prior+=img_n) \
                   for (k=0; k < img_n; ++k)
         switch (filter) {
            CASE(F_none)  cur[k] = raw[k]; break;
            CASE(F_sub)   cur[k] = raw[k] + cur[k-img_n]; break;
            CASE(F_up)    cur[k] = raw[k] + prior[k]; break;
            CASE(F_avg)   cur[k] = raw[k] + ((prior[k] + cur[k-img_n])>>1); break;
            CASE(F_paeth)  cur[k] = (uint8) (raw[k] + paeth(cur[k-img_n],prior[k],prior[k-img_n])); break;
            CASE(F_avg_first)    cur[k] = raw[k] + (cur[k-img_n] >> 1); break;
            CASE(F_paeth_first)  cur[k] = (uint8) (raw[k] + paeth(cur[k-img_n],0,0)); break;
         }
         #undef CASE
      } else {
         assert(img_n+1 == out_n);
         #define CASE(f) \
             case f:     \
                for (i=x-1; i >= 1; --i, cur[img_n]=255,raw+=img_n,cur+=out_n,prior+=out_n) \
                   for (k=0; k < img_n; ++k)
         switch (filter) {
            CASE(F_none)  cur[k] = raw[k]; break;
            CASE(F_sub)   cur[k] = raw[k] + cur[k-out_n]; break;
            CASE(F_up)    cur[k] = raw[k] + prior[k]; break;
            CASE(F_avg)   cur[k] = raw[k] + ((prior[k] + cur[k-out_n])>>1); break;
            CASE(F_paeth)  cur[k] = (uint8) (raw[k] + paeth(cur[k-out_n],prior[k],prior[k-out_n])); break;
            CASE(F_avg_first)    cur[k] = raw[k] + (cur[k-out_n] >> 1); break;
            CASE(F_paeth_first)  cur[k] = (uint8) (raw[k] + paeth(cur[k-out_n],0,0)); break;
         }
         #undef CASE
      }
   }
   return 1;
}

static int create_png_image(png *a, uint8 *raw, uint32 raw_len, int out_n, int interlaced)
{
   uint8 *final;
   int p;
   int save;
   if (!interlaced)
      return create_png_image_raw(a, raw, raw_len, out_n, a->s->img_x, a->s->img_y);
   save = stbi_png_partial;
   stbi_png_partial = 0;

   // de-interlacing
   final = (uint8 *) malloc(a->s->img_x * a->s->img_y * out_n);
   for (p=0; p < 7; ++p) {
      int xorig[] = { 0,4,0,2,0,1,0 };
      int yorig[] = { 0,0,4,0,2,0,1 };
      int xspc[]  = { 8,8,4,4,2,2,1 };
      int yspc[]  = { 8,8,8,4,4,2,2 };
      int i,j,x,y;
      // pass1_x[4] = 0, pass1_x[5] = 1, pass1_x[12] = 1
      x = (a->s->img_x - xorig[p] + xspc[p]-1) / xspc[p];
      y = (a->s->img_y - yorig[p] + yspc[p]-1) / yspc[p];
      if (x && y) {
         if (!create_png_image_raw(a, raw, raw_len, out_n, x, y)) {
            free(final);
            return 0;
         }
         for (j=0; j < y; ++j)
            for (i=0; i < x; ++i)
               memcpy(final + (j*yspc[p]+yorig[p])*a->s->img_x*out_n + (i*xspc[p]+xorig[p])*out_n,
                      a->out + (j*x+i)*out_n, out_n);
         free(a->out);
         raw += (x*out_n+1)*y;
         raw_len -= (x*out_n+1)*y;
      }
   }
   a->out = final;

   stbi_png_partial = save;
   return 1;
}

static int compute_transparency(png *z, uint8 tc[3], int out_n)
{
   stbi *s = z->s;
   uint32 i, pixel_count = s->img_x * s->img_y;
   uint8 *p = z->out;

   // compute color-based transparency, assuming we've
   // already got 255 as the alpha value in the output
   assert(out_n == 2 || out_n == 4);

   if (out_n == 2) {
      for (i=0; i < pixel_count; ++i) {
         p[1] = (p[0] == tc[0] ? 0 : 255);
         p += 2;
      }
   } else {
      for (i=0; i < pixel_count; ++i) {
         if (p[0] == tc[0] && p[1] == tc[1] && p[2] == tc[2])
            p[3] = 0;
         p += 4;
      }
   }
   return 1;
}

static int expand_palette(png *a, uint8 *palette, int len, int pal_img_n)
{
   uint32 i, pixel_count = a->s->img_x * a->s->img_y;
   uint8 *p, *temp_out, *orig = a->out;

   p = (uint8 *) malloc(pixel_count * pal_img_n);
   if (p == NULL) return e("outofmem", "Out of memory");

   // between here and free(out) below, exitting would leak
   temp_out = p;

   if (pal_img_n == 3) {
      for (i=0; i < pixel_count; ++i) {
         int n = orig[i]*4;
         p[0] = palette[n  ];
         p[1] = palette[n+1];
         p[2] = palette[n+2];
         p += 3;
      }
   } else {
      for (i=0; i < pixel_count; ++i) {
         int n = orig[i]*4;
         p[0] = palette[n  ];
         p[1] = palette[n+1];
         p[2] = palette[n+2];
         p[3] = palette[n+3];
         p += 4;
      }
   }
   free(a->out);
   a->out = temp_out;

   STBI_NOTUSED(len);

   return 1;
}

static int stbi_unpremultiply_on_load = 0;
static int stbi_de_iphone_flag = 0;

void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply)
{
   stbi_unpremultiply_on_load = flag_true_if_should_unpremultiply;
}
void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert)
{
   stbi_de_iphone_flag = flag_true_if_should_convert;
}

static void stbi_de_iphone(png *z)
{
   stbi *s = z->s;
   uint32 i, pixel_count = s->img_x * s->img_y;
   uint8 *p = z->out;

   if (s->img_out_n == 3) {  // convert bgr to rgb
      for (i=0; i < pixel_count; ++i) {
         uint8 t = p[0];
         p[0] = p[2];
         p[2] = t;
         p += 3;
      }
   } else {
      assert(s->img_out_n == 4);
      if (stbi_unpremultiply_on_load) {
         // convert bgr to rgb and unpremultiply
         for (i=0; i < pixel_count; ++i) {
            uint8 a = p[3];
            uint8 t = p[0];
            if (a) {
               p[0] = p[2] * 255 / a;
               p[1] = p[1] * 255 / a;
               p[2] =  t   * 255 / a;
            } else {
               p[0] = p[2];
               p[2] = t;
            }
            p += 4;
         }
      } else {
         // convert bgr to rgb
         for (i=0; i < pixel_count; ++i) {
            uint8 t = p[0];
            p[0] = p[2];
            p[2] = t;
            p += 4;
         }
      }
   }
}

static int parse_png_file(png *z, int scan, int req_comp)
{
   uint8 palette[1024], pal_img_n=0;
   uint8 has_trans=0, tc[3];
   uint32 ioff=0, idata_limit=0, i, pal_len=0;
   int first=1,k,interlace=0, iphone=0;
   stbi *s = z->s;

   z->expanded = NULL;
   z->idata = NULL;
   z->out = NULL;

   if (!check_png_header(s)) return 0;

   if (scan == SCAN_type) return 1;

   for (;;) {
      chunk c = get_chunk_header(s);
      switch (c.type) {
         case PNG_TYPE('C','g','B','I'):
            iphone = stbi_de_iphone_flag;
            skip(s, c.length);
            break;
         case PNG_TYPE('I','H','D','R'): {
            int depth,color,comp,filter;
            if (!first) return e("multiple IHDR","Corrupt PNG");
            first = 0;
            if (c.length != 13) return e("bad IHDR len","Corrupt PNG");
            s->img_x = get32(s); if (s->img_x > (1 << 24)) return e("too large","Very large image (corrupt?)");
            s->img_y = get32(s); if (s->img_y > (1 << 24)) return e("too large","Very large image (corrupt?)");
            depth = get8(s);  if (depth != 8)        return e("8bit only","PNG not supported: 8-bit only");
            color = get8(s);  if (color > 6)         return e("bad ctype","Corrupt PNG");
            if (color == 3) pal_img_n = 3; else if (color & 1) return e("bad ctype","Corrupt PNG");
            comp  = get8(s);  if (comp) return e("bad comp method","Corrupt PNG");
            filter= get8(s);  if (filter) return e("bad filter method","Corrupt PNG");
            interlace = get8(s); if (interlace>1) return e("bad interlace method","Corrupt PNG");
            if (!s->img_x || !s->img_y) return e("0-pixel image","Corrupt PNG");
            if (!pal_img_n) {
               s->img_n = (color & 2 ? 3 : 1) + (color & 4 ? 1 : 0);
               if ((1 << 30) / s->img_x / s->img_n < s->img_y) return e("too large", "Image too large to decode");
               if (scan == SCAN_header) return 1;
            } else {
               // if paletted, then pal_n is our final components, and
               // img_n is # components to decompress/filter.
               s->img_n = 1;
               if ((1 << 30) / s->img_x / 4 < s->img_y) return e("too large","Corrupt PNG");
               // if SCAN_header, have to scan to see if we have a tRNS
            }
            break;
         }

         case PNG_TYPE('P','L','T','E'):  {
            if (first) return e("first not IHDR", "Corrupt PNG");
            if (c.length > 256*3) return e("invalid PLTE","Corrupt PNG");
            pal_len = c.length / 3;
            if (pal_len * 3 != c.length) return e("invalid PLTE","Corrupt PNG");
            for (i=0; i < pal_len; ++i) {
               palette[i*4+0] = get8u(s);
               palette[i*4+1] = get8u(s);
               palette[i*4+2] = get8u(s);
               palette[i*4+3] = 255;
            }
            break;
         }

         case PNG_TYPE('t','R','N','S'): {
            if (first) return e("first not IHDR", "Corrupt PNG");
            if (z->idata) return e("tRNS after IDAT","Corrupt PNG");
            if (pal_img_n) {
               if (scan == SCAN_header) { s->img_n = 4; return 1; }
               if (pal_len == 0) return e("tRNS before PLTE","Corrupt PNG");
               if (c.length > pal_len) return e("bad tRNS len","Corrupt PNG");
               pal_img_n = 4;
               for (i=0; i < c.length; ++i)
                  palette[i*4+3] = get8u(s);
            } else {
               if (!(s->img_n & 1)) return e("tRNS with alpha","Corrupt PNG");
               if (c.length != (uint32) s->img_n*2) return e("bad tRNS len","Corrupt PNG");
               has_trans = 1;
               for (k=0; k < s->img_n; ++k)
                  tc[k] = (uint8) get16(s); // non 8-bit images will be larger
            }
            break;
         }

         case PNG_TYPE('I','D','A','T'): {
            if (first) return e("first not IHDR", "Corrupt PNG");
            if (pal_img_n && !pal_len) return e("no PLTE","Corrupt PNG");
            if (scan == SCAN_header) { s->img_n = pal_img_n; return 1; }
            if (ioff + c.length > idata_limit) {
               uint8 *p;
               if (idata_limit == 0) idata_limit = c.length > 4096 ? c.length : 4096;
               while (ioff + c.length > idata_limit)
                  idata_limit *= 2;
               p = (uint8 *) realloc(z->idata, idata_limit); if (p == NULL) return e("outofmem", "Out of memory");
               z->idata = p;
            }
            if (!getn(s, z->idata+ioff,c.length)) return e("outofdata","Corrupt PNG");
            ioff += c.length;
            break;
         }

         case PNG_TYPE('I','E','N','D'): {
            uint32 raw_len;
            if (first) return e("first not IHDR", "Corrupt PNG");
            if (scan != SCAN_load) return 1;
            if (z->idata == NULL) return e("no IDAT","Corrupt PNG");
            z->expanded = (uint8 *) stbi_zlib_decode_malloc_guesssize_headerflag((char *) z->idata, ioff, 16384, (int *) &raw_len, !iphone);
            if (z->expanded == NULL) return 0; // zlib should set error
            free(z->idata); z->idata = NULL;
            if ((req_comp == s->img_n+1 && req_comp != 3 && !pal_img_n) || has_trans)
               s->img_out_n = s->img_n+1;
            else
               s->img_out_n = s->img_n;
            if (!create_png_image(z, z->expanded, raw_len, s->img_out_n, interlace)) return 0;
            if (has_trans)
               if (!compute_transparency(z, tc, s->img_out_n)) return 0;
            if (iphone && s->img_out_n > 2)
               stbi_de_iphone(z);
            if (pal_img_n) {
               // pal_img_n == 3 or 4
               s->img_n = pal_img_n; // record the actual colors we had
               s->img_out_n = pal_img_n;
               if (req_comp >= 3) s->img_out_n = req_comp;
               if (!expand_palette(z, palette, pal_len, s->img_out_n))
                  return 0;
            }
            free(z->expanded); z->expanded = NULL;
            return 1;
         }

         default:
            // if critical, fail
            if (first) return e("first not IHDR", "Corrupt PNG");
            if ((c.type & (1 << 29)) == 0) {
               #ifndef STBI_NO_FAILURE_STRINGS
               // not threadsafe
               static char invalid_chunk[] = "XXXX chunk not known";
               invalid_chunk[0] = (uint8) (c.type >> 24);
               invalid_chunk[1] = (uint8) (c.type >> 16);
               invalid_chunk[2] = (uint8) (c.type >>  8);
               invalid_chunk[3] = (uint8) (c.type >>  0);
               #endif
               return e(invalid_chunk, "PNG not supported: unknown chunk type");
            }
            skip(s, c.length);
            break;
      }
      // end of chunk, read and skip CRC
      get32(s);
   }
}

static unsigned char *do_png(png *p, int *x, int *y, int *n, int req_comp)
{
   unsigned char *result=NULL;
   if (req_comp < 0 || req_comp > 4) return epuc("bad req_comp", "Internal error");
   if (parse_png_file(p, SCAN_load, req_comp)) {
      result = p->out;
      p->out = NULL;
      if (req_comp && req_comp != p->s->img_out_n) {
         result = convert_format(result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y);
         p->s->img_out_n = req_comp;
         if (result == NULL) return result;
      }
      *x = p->s->img_x;
      *y = p->s->img_y;
      if (n) *n = p->s->img_n;
   }
   free(p->out);      p->out      = NULL;
   free(p->expanded); p->expanded = NULL;
   free(p->idata);    p->idata    = NULL;

   return result;
}

static unsigned char *stbi_png_load(stbi *s, int *x, int *y, int *comp, int req_comp)
{
   png p;
   p.s = s;
   return do_png(&p, x,y,comp,req_comp);
}

static int stbi_png_test(stbi *s)
{
   int r;
   r = check_png_header(s);
   stbi_rewind(s);
   return r;
}

static int stbi_png_info_raw(png *p, int *x, int *y, int *comp)
{
   if (!parse_png_file(p, SCAN_header, 0)) {
      stbi_rewind( p->s );
      return 0;
   }
   if (x) *x = p->s->img_x;
   if (y) *y = p->s->img_y;
   if (comp) *comp = p->s->img_n;
   return 1;
}

static int      stbi_png_info(stbi *s, int *x, int *y, int *comp)
{
   png p;
   p.s = s;
   return stbi_png_info_raw(&p, x, y, comp);
}

// Microsoft/Windows BMP image

static int bmp_test(stbi *s)
{
   int sz;
   if (get8(s) != 'B') return 0;
   if (get8(s) != 'M') return 0;
   get32le(s); // discard filesize
   get16le(s); // discard reserved
   get16le(s); // discard reserved
   get32le(s); // discard data offset
   sz = get32le(s);
   if (sz == 12 || sz == 40 || sz == 56 || sz == 108) return 1;
   return 0;
}

static int stbi_bmp_test(stbi *s)
{
   int r = bmp_test(s);
   stbi_rewind(s);
   return r;
}


// returns 0..31 for the highest set bit
static int high_bit(unsigned int z)
{
   int n=0;
   if (z == 0) return -1;
   if (z >= 0x10000) n += 16, z >>= 16;
   if (z >= 0x00100) n +=  8, z >>=  8;
   if (z >= 0x00010) n +=  4, z >>=  4;
   if (z >= 0x00004) n +=  2, z >>=  2;
   if (z >= 0x00002) n +=  1, z >>=  1;
   return n;
}

static int bitcount(unsigned int a)
{
   a = (a & 0x55555555) + ((a >>  1) & 0x55555555); // max 2
   a = (a & 0x33333333) + ((a >>  2) & 0x33333333); // max 4
   a = (a + (a >> 4)) & 0x0f0f0f0f; // max 8 per 4, now 8 bits
   a = (a + (a >> 8)); // max 16 per 8 bits
   a = (a + (a >> 16)); // max 32 per 8 bits
   return a & 0xff;
}

static int shiftsigned(int v, int shift, int bits)
{
   int result;
   int z=0;

   if (shift < 0) v <<= -shift;
   else v >>= shift;
   result = v;

   z = bits;
   while (z < 8) {
      result += v >> z;
      z += bits;
   }
   return result;
}

static stbi_uc *bmp_load(stbi *s, int *x, int *y, int *comp, int req_comp)
{
   uint8 *out;
   unsigned int mr=0,mg=0,mb=0,ma=0, fake_a=0;
   mr = fake_a;
   stbi_uc pal[256][4];
   int psize=0,i,j,compress=0,width;
   int bpp, flip_vertically, pad, target, offset, hsz;
   if (get8(s) != 'B' || get8(s) != 'M') return epuc("not BMP", "Corrupt BMP");
   get32le(s); // discard filesize
   get16le(s); // discard reserved
   get16le(s); // discard reserved
   offset = get32le(s);
   hsz = get32le(s);
   if (hsz != 12 && hsz != 40 && hsz != 56 && hsz != 108) return epuc("unknown BMP", "BMP type not supported: unknown");
   if (hsz == 12) {
      s->img_x = get16le(s);
      s->img_y = get16le(s);
   } else {
      s->img_x = get32le(s);
      s->img_y = get32le(s);
   }
   if (get16le(s) != 1) return epuc("bad BMP", "bad BMP");
   bpp = get16le(s);
   if (bpp == 1) return epuc("monochrome", "BMP type not supported: 1-bit");
   flip_vertically = ((int) s->img_y) > 0;
   s->img_y = abs((int) s->img_y);
   if (hsz == 12) {
      if (bpp < 24)
         psize = (offset - 14 - 24) / 3;
   } else {
      compress = get32le(s);
      if (compress == 1 || compress == 2) return epuc("BMP RLE", "BMP type not supported: RLE");
      get32le(s); // discard sizeof
      get32le(s); // discard hres
      get32le(s); // discard vres
      get32le(s); // discard colorsused
      get32le(s); // discard max important
      if (hsz == 40 || hsz == 56) {
         if (hsz == 56) {
            get32le(s);
            get32le(s);
            get32le(s);
            get32le(s);
         }
         if (bpp == 16 || bpp == 32) {
            mr = mg = mb = 0;
            if (compress == 0) {
               if (bpp == 32) {
                  mr = 0xffu << 16;
                  mg = 0xffu <<  8;
                  mb = 0xffu <<  0;
                  ma = 0xffu << 24;
                  fake_a = 1; // @TODO: check for cases like alpha value is all 0 and switch it to 255
               } else {
                  mr = 31u << 10;
                  mg = 31u <<  5;
                  mb = 31u <<  0;
               }
            } else if (compress == 3) {
               mr = get32le(s);
               mg = get32le(s);
               mb = get32le(s);
               // not documented, but generated by photoshop and handled by mspaint
               if (mr == mg && mg == mb) {
                  // ?!?!?
                  return epuc("bad BMP", "bad BMP");
               }
            } else
               return epuc("bad BMP", "bad BMP");
         }
      } else {
         assert(hsz == 108);
         mr = get32le(s);
         mg = get32le(s);
         mb = get32le(s);
         ma = get32le(s);
         get32le(s); // discard color space
         for (i=0; i < 12; ++i)
            get32le(s); // discard color space parameters
      }
      if (bpp < 16)
         psize = (offset - 14 - hsz) >> 2;
   }
   s->img_n = ma ? 4 : 3;
   if (req_comp && req_comp >= 3) // we can directly decode 3 or 4
      target = req_comp;
   else
      target = s->img_n; // if they want monochrome, we'll post-convert
   out = (stbi_uc *) malloc(target * s->img_x * s->img_y);
   if (!out) return epuc("outofmem", "Out of memory");
   if (bpp < 16) {
      int z=0;
      if (psize == 0 || psize > 256) { free(out); return epuc("invalid", "Corrupt BMP"); }
      for (i=0; i < psize; ++i) {
         pal[i][2] = get8u(s);
         pal[i][1] = get8u(s);
         pal[i][0] = get8u(s);
         if (hsz != 12) get8(s);
         pal[i][3] = 255;
      }
      skip(s, offset - 14 - hsz - psize * (hsz == 12 ? 3 : 4));
      if (bpp == 4) width = (s->img_x + 1) >> 1;
      else if (bpp == 8) width = s->img_x;
      else { free(out); return epuc("bad bpp", "Corrupt BMP"); }
      pad = (-width)&3;
      for (j=0; j < (int) s->img_y; ++j) {
         for (i=0; i < (int) s->img_x; i += 2) {
            int v=get8(s),v2=0;
            if (bpp == 4) {
               v2 = v & 15;
               v >>= 4;
            }
            out[z++] = pal[v][0];
            out[z++] = pal[v][1];
            out[z++] = pal[v][2];
            if (target == 4) out[z++] = 255;
            if (i+1 == (int) s->img_x) break;
            v = (bpp == 8) ? get8(s) : v2;
            out[z++] = pal[v][0];
            out[z++] = pal[v][1];
            out[z++] = pal[v][2];
            if (target == 4) out[z++] = 255;
         }
         skip(s, pad);
      }
   } else {
      int rshift=0,gshift=0,bshift=0,ashift=0,rcount=0,gcount=0,bcount=0,acount=0;
      int z = 0;
      int easy=0;
      skip(s, offset - 14 - hsz);
      if (bpp == 24) width = 3 * s->img_x;
      else if (bpp == 16) width = 2*s->img_x;
      else /* bpp = 32 and pad = 0 */ width=0;
      pad = (-width) & 3;
      if (bpp == 24) {
         easy = 1;
      } else if (bpp == 32) {
         if (mb == 0xff && mg == 0xff00 && mr == 0x00ff0000 && ma == 0xff000000)
            easy = 2;
      }
      if (!easy) {
         if (!mr || !mg || !mb) { free(out); return epuc("bad masks", "Corrupt BMP"); }
         // right shift amt to put high bit in position #7
         rshift = high_bit(mr)-7; rcount = bitcount(mr);
         gshift = high_bit(mg)-7; gcount = bitcount(mr);
         bshift = high_bit(mb)-7; bcount = bitcount(mr);
         ashift = high_bit(ma)-7; acount = bitcount(mr);
      }
      for (j=0; j < (int) s->img_y; ++j) {
         if (easy) {
            for (i=0; i < (int) s->img_x; ++i) {
               int a;
               out[z+2] = get8u(s);
               out[z+1] = get8u(s);
               out[z+0] = get8u(s);
               z += 3;
               a = (easy == 2 ? get8(s) : 255);
               if (target == 4) out[z++] = (uint8) a;
            }
         } else {
            for (i=0; i < (int) s->img_x; ++i) {
               uint32 v = (bpp == 16 ? get16le(s) : get32le(s));
               int a;
               out[z++] = (uint8) shiftsigned(v & mr, rshift, rcount);
               out[z++] = (uint8) shiftsigned(v & mg, gshift, gcount);
               out[z++] = (uint8) shiftsigned(v & mb, bshift, bcount);
               a = (ma ? shiftsigned(v & ma, ashift, acount) : 255);
               if (target == 4) out[z++] = (uint8) a;
            }
         }
         skip(s, pad);
      }
   }
   if (flip_vertically) {
      stbi_uc t;
      for (j=0; j < (int) s->img_y>>1; ++j) {
         stbi_uc *p1 = out +      j     *s->img_x*target;
         stbi_uc *p2 = out + (s->img_y-1-j)*s->img_x*target;
         for (i=0; i < (int) s->img_x*target; ++i) {
            t = p1[i], p1[i] = p2[i], p2[i] = t;
         }
      }
   }

   if (req_comp && req_comp != target) {
      out = convert_format(out, target, req_comp, s->img_x, s->img_y);
      if (out == NULL) return out; // convert_format frees input on failure
   }

   *x = s->img_x;
   *y = s->img_y;
   if (comp) *comp = s->img_n;
   return out;
}

static stbi_uc *stbi_bmp_load(stbi *s,int *x, int *y, int *comp, int req_comp)
{
   return bmp_load(s, x,y,comp,req_comp);
}


// Targa Truevision - TGA
// by Jonathan Dummer

static int tga_info(stbi *s, int *x, int *y, int *comp)
{
    int tga_w, tga_h, tga_comp;
    int sz;
    get8u(s);                   // discard Offset
    sz = get8u(s);              // color type
    if( sz > 1 ) {
        stbi_rewind(s);
        return 0;      // only RGB or indexed allowed
    }
    sz = get8u(s);              // image type
    // only RGB or grey allowed, +/- RLE
    if ((sz != 1) && (sz != 2) && (sz != 3) && (sz != 9) && (sz != 10) && (sz != 11)) return 0;
    skip(s,9);
    tga_w = get16le(s);
    if( tga_w < 1 ) {
        stbi_rewind(s);
        return 0;   // test width
    }
    tga_h = get16le(s);
    if( tga_h < 1 ) {
        stbi_rewind(s);
        return 0;   // test height
    }
    sz = get8(s);               // bits per pixel
    // only RGB or RGBA or grey allowed
    if ((sz != 8) && (sz != 16) && (sz != 24) && (sz != 32)) {
        stbi_rewind(s);
        return 0;
    }
    tga_comp = sz;
    if (x) *x = tga_w;
    if (y) *y = tga_h;
    if (comp) *comp = tga_comp / 8;
    return 1;                   // seems to have passed everything
}

int stbi_tga_info(stbi *s, int *x, int *y, int *comp)
{
    return tga_info(s, x, y, comp);
}

static int tga_test(stbi *s)
{
   int sz;
   get8u(s);      //   discard Offset
   sz = get8u(s);   //   color type
   if ( sz > 1 ) return 0;   //   only RGB or indexed allowed
   sz = get8u(s);   //   image type
   if ( (sz != 1) && (sz != 2) && (sz != 3) && (sz != 9) && (sz != 10) && (sz != 11) ) return 0;   //   only RGB or grey allowed, +/- RLE
   get16(s);      //   discard palette start
   get16(s);      //   discard palette length
   get8(s);         //   discard bits per palette color entry
   get16(s);      //   discard x origin
   get16(s);      //   discard y origin
   if ( get16(s) < 1 ) return 0;      //   test width
   if ( get16(s) < 1 ) return 0;      //   test height
   sz = get8(s);   //   bits per pixel
   if ( (sz != 8) && (sz != 16) && (sz != 24) && (sz != 32) ) return 0;   //   only RGB or RGBA or grey allowed
   return 1;      //   seems to have passed everything
}

static int stbi_tga_test(stbi *s)
{
   int res = tga_test(s);
   stbi_rewind(s);
   return res;
}

static stbi_uc *tga_load(stbi *s, int *x, int *y, int *comp, int req_comp)
{
   //   read in the TGA header stuff
   int tga_offset = get8u(s);
   int tga_indexed = get8u(s);
   int tga_image_type = get8u(s);
   int tga_is_RLE = 0;
   int tga_palette_start = get16le(s);
   int tga_palette_len = get16le(s);
   int tga_palette_bits = get8u(s);
   int tga_x_origin = get16le(s);
   int tga_y_origin = get16le(s);
   int tga_width = get16le(s);
   int tga_height = get16le(s);
   int tga_bits_per_pixel = get8u(s);
   int tga_inverted = get8u(s);
   //   image data
   unsigned char *tga_data;
   unsigned char *tga_palette = NULL;
   int i, j;
   unsigned char raw_data[4];
   unsigned char trans_data[4];
   int RLE_count = 0;
   int RLE_repeating = 0;
   int read_next_pixel = 1;

   //   do a tiny bit of precessing
   if ( tga_image_type >= 8 )
   {
      tga_image_type -= 8;
      tga_is_RLE = 1;
   }
   /* int tga_alpha_bits = tga_inverted & 15; */
   tga_inverted = 1 - ((tga_inverted >> 5) & 1);

   //   error check
   if ( //(tga_indexed) ||
      (tga_width < 1) || (tga_height < 1) ||
      (tga_image_type < 1) || (tga_image_type > 3) ||
      ((tga_bits_per_pixel != 8) && (tga_bits_per_pixel != 16) &&
      (tga_bits_per_pixel != 24) && (tga_bits_per_pixel != 32))
      )
   {
      return NULL; // we don't report this as a bad TGA because we don't even know if it's TGA
   }

   //   If I'm paletted, then I'll use the number of bits from the palette
   if ( tga_indexed )
   {
      tga_bits_per_pixel = tga_palette_bits;
   }

   //   tga info
   *x = tga_width;
   *y = tga_height;
   if ( (req_comp < 1) || (req_comp > 4) )
   {
      //   just use whatever the file was
      req_comp = tga_bits_per_pixel / 8;
      *comp = req_comp;
   } else
   {
      //   force a new number of components
      *comp = tga_bits_per_pixel/8;
   }
   tga_data = (unsigned char*)malloc( tga_width * tga_height * req_comp );
   if (!tga_data) return epuc("outofmem", "Out of memory");

   //   skip to the data's starting position (offset usually = 0)
   skip(s, tga_offset );
   //   do I need to load a palette?
   if ( tga_indexed )
   {
      //   any data to skip? (offset usually = 0)
      skip(s, tga_palette_start );
      //   load the palette
      tga_palette = (unsigned char*)malloc( tga_palette_len * tga_palette_bits / 8 );
      if (!tga_palette) return epuc("outofmem", "Out of memory");
      if (!getn(s, tga_palette, tga_palette_len * tga_palette_bits / 8 )) {
         free(tga_data);
         free(tga_palette);
         return epuc("bad palette", "Corrupt TGA");
      }
   }
   //   load the data
   trans_data[0] = trans_data[1] = trans_data[2] = trans_data[3] = 0;
   for (i=0; i < tga_width * tga_height; ++i)
   {
      //   if I'm in RLE mode, do I need to get a RLE chunk?
      if ( tga_is_RLE )
      {
         if ( RLE_count == 0 )
         {
            //   yep, get the next byte as a RLE command
            int RLE_cmd = get8u(s);
            RLE_count = 1 + (RLE_cmd & 127);
            RLE_repeating = RLE_cmd >> 7;
            read_next_pixel = 1;
         } else if ( !RLE_repeating )
         {
            read_next_pixel = 1;
         }
      } else
      {
         read_next_pixel = 1;
      }
      //   OK, if I need to read a pixel, do it now
      if ( read_next_pixel )
      {
         //   load however much data we did have
         if ( tga_indexed )
         {
            //   read in 1 byte, then perform the lookup
            int pal_idx = get8u(s);
            if ( pal_idx >= tga_palette_len )
            {
               //   invalid index
               pal_idx = 0;
            }
            pal_idx *= tga_bits_per_pixel / 8;
            for (j = 0; j*8 < tga_bits_per_pixel; ++j)
            {
               raw_data[j] = tga_palette[pal_idx+j];
            }
         } else
         {
            //   read in the data raw
            for (j = 0; j*8 < tga_bits_per_pixel; ++j)
            {
               raw_data[j] = get8u(s);
            }
         }
         //   convert raw to the intermediate format
         switch (tga_bits_per_pixel)
         {
         case 8:
            //   Luminous => RGBA
            trans_data[0] = raw_data[0];
            trans_data[1] = raw_data[0];
            trans_data[2] = raw_data[0];
            trans_data[3] = 255;
            break;
         case 16:
            //   Luminous,Alpha => RGBA
            trans_data[0] = raw_data[0];
            trans_data[1] = raw_data[0];
            trans_data[2] = raw_data[0];
            trans_data[3] = raw_data[1];
            break;
         case 24:
            //   BGR => RGBA
            trans_data[0] = raw_data[2];
            trans_data[1] = raw_data[1];
            trans_data[2] = raw_data[0];
            trans_data[3] = 255;
            break;
         case 32:
            //   BGRA => RGBA
            trans_data[0] = raw_data[2];
            trans_data[1] = raw_data[1];
            trans_data[2] = raw_data[0];
            trans_data[3] = raw_data[3];
            break;
         }
         //   clear the reading flag for the next pixel
         read_next_pixel = 0;
      } // end of reading a pixel
      //   convert to final format
      switch (req_comp)
      {
      case 1:
         //   RGBA => Luminance
         tga_data[i*req_comp+0] = compute_y(trans_data[0],trans_data[1],trans_data[2]);
         break;
      case 2:
         //   RGBA => Luminance,Alpha
         tga_data[i*req_comp+0] = compute_y(trans_data[0],trans_data[1],trans_data[2]);
         tga_data[i*req_comp+1] = trans_data[3];
         break;
      case 3:
         //   RGBA => RGB
         tga_data[i*req_comp+0] = trans_data[0];
         tga_data[i*req_comp+1] = trans_data[1];
         tga_data[i*req_comp+2] = trans_data[2];
         break;
      case 4:
         //   RGBA => RGBA
         tga_data[i*req_comp+0] = trans_data[0];
         tga_data[i*req_comp+1] = trans_data[1];
         tga_data[i*req_comp+2] = trans_data[2];
         tga_data[i*req_comp+3] = trans_data[3];
         break;
      }
      //   in case we're in RLE mode, keep counting down
      --RLE_count;
   }
   //   do I need to invert the image?
   if ( tga_inverted )
   {
      for (j = 0; j*2 < tga_height; ++j)
      {
         int index1 = j * tga_width * req_comp;
         int index2 = (tga_height - 1 - j) * tga_width * req_comp;
         for (i = tga_width * req_comp; i > 0; --i)
         {
            unsigned char temp = tga_data[index1];
            tga_data[index1] = tga_data[index2];
            tga_data[index2] = temp;
            ++index1;
            ++index2;
         }
      }
   }
   //   clear my palette, if I had one
   if ( tga_palette != NULL )
   {
      free( tga_palette );
   }
   //   the things I do to get rid of an error message, and yet keep
   //   Microsoft's C compilers happy... [8^(
   tga_palette_start = tga_palette_len = tga_palette_bits =
         tga_x_origin = tga_y_origin = 0;
   //   OK, done
   return tga_data;
}

static stbi_uc *stbi_tga_load(stbi *s, int *x, int *y, int *comp, int req_comp)
{
   return tga_load(s,x,y,comp,req_comp);
}


// *************************************************************************************************
// Photoshop PSD loader -- PD by Thatcher Ulrich, integration by Nicolas Schulz, tweaked by STB

static int psd_test(stbi *s)
{
   if (get32(s) != 0x38425053) return 0;   // "8BPS"
   else return 1;
}

static int stbi_psd_test(stbi *s)
{
   int r = psd_test(s);
   stbi_rewind(s);
   return r;
}

static stbi_uc *psd_load(stbi *s, int *x, int *y, int *comp, int req_comp)
{
   int   pixelCount;
   int channelCount, compression;
   int channel, i, count, len;
   int w,h;
   uint8 *out;

   // Check identifier
   if (get32(s) != 0x38425053)   // "8BPS"
      return epuc("not PSD", "Corrupt PSD image");

   // Check file type version.
   if (get16(s) != 1)
      return epuc("wrong version", "Unsupported version of PSD image");

   // Skip 6 reserved bytes.
   skip(s, 6 );

   // Read the number of channels (R, G, B, A, etc).
   channelCount = get16(s);
   if (channelCount < 0 || channelCount > 16)
      return epuc("wrong channel count", "Unsupported number of channels in PSD image");

   // Read the rows and columns of the image.
   h = get32(s);
   w = get32(s);

   // Make sure the depth is 8 bits.
   if (get16(s) != 8)
      return epuc("unsupported bit depth", "PSD bit depth is not 8 bit");

   // Make sure the color mode is RGB.
   // Valid options are:
   //   0: Bitmap
   //   1: Grayscale
   //   2: Indexed color
   //   3: RGB color
   //   4: CMYK color
   //   7: Multichannel
   //   8: Duotone
   //   9: Lab color
   if (get16(s) != 3)
      return epuc("wrong color format", "PSD is not in RGB color format");

   // Skip the Mode Data.  (It's the palette for indexed color; other info for other modes.)
   skip(s,get32(s) );

   // Skip the image resources.  (resolution, pen tool paths, etc)
   skip(s, get32(s) );

   // Skip the reserved data.
   skip(s, get32(s) );

   // Find out if the data is compressed.
   // Known values:
   //   0: no compression
   //   1: RLE compressed
   compression = get16(s);
   if (compression > 1)
      return epuc("bad compression", "PSD has an unknown compression format");

   // Create the destination image.
   out = (stbi_uc *) malloc(4 * w*h);
   if (!out) return epuc("outofmem", "Out of memory");
   pixelCount = w*h;

   // Initialize the data to zero.
   //memset( out, 0, pixelCount * 4 );

   // Finally, the image data.
   if (compression) {
      // RLE as used by .PSD and .TIFF
      // Loop until you get the number of unpacked bytes you are expecting:
      //     Read the next source byte into n.
      //     If n is between 0 and 127 inclusive, copy the next n+1 bytes literally.
      //     Else if n is between -127 and -1 inclusive, copy the next byte -n+1 times.
      //     Else if n is 128, noop.
      // Endloop

      // The RLE-compressed data is preceeded by a 2-byte data count for each row in the data,
      // which we're going to just skip.
      skip(s, h * channelCount * 2 );

      // Read the RLE data by channel.
      for (channel = 0; channel < 4; channel++) {
         uint8 *p;

         p = out+channel;
         if (channel >= channelCount) {
            // Fill this channel with default data.
            for (i = 0; i < pixelCount; i++) *p = (channel == 3 ? 255 : 0), p += 4;
         } else {
            // Read the RLE data.
            count = 0;
            while (count < pixelCount) {
               len = get8(s);
               if (len == 128) {
                  // No-op.
               } else if (len < 128) {
                  // Copy next len+1 bytes literally.
                  len++;
                  count += len;
                  while (len) {
                     *p = get8u(s);
                     p += 4;
                     len--;
                  }
               } else if (len > 128) {
                  uint8   val;
                  // Next -len+1 bytes in the dest are replicated from next source byte.
                  // (Interpret len as a negative 8-bit int.)
                  len ^= 0x0FF;
                  len += 2;
                  val = get8u(s);
                  count += len;
                  while (len) {
                     *p = val;
                     p += 4;
                     len--;
                  }
               }
            }
         }
      }

   } else {
      // We're at the raw image data.  It's each channel in order (Red, Green, Blue, Alpha, ...)
      // where each channel consists of an 8-bit value for each pixel in the image.

      // Read the data by channel.
      for (channel = 0; channel < 4; channel++) {
         uint8 *p;

         p = out + channel;
         if (channel > channelCount) {
            // Fill this channel with default data.
            for (i = 0; i < pixelCount; i++) *p = channel == 3 ? 255 : 0, p += 4;
         } else {
            // Read the data.
            for (i = 0; i < pixelCount; i++)
               *p = get8u(s), p += 4;
         }
      }
   }

   if (req_comp && req_comp != 4) {
      out = convert_format(out, 4, req_comp, w, h);
      if (out == NULL) return out; // convert_format frees input on failure
   }

   if (comp) *comp = channelCount;
   *y = h;
   *x = w;

   return out;
}

static stbi_uc *stbi_psd_load(stbi *s, int *x, int *y, int *comp, int req_comp)
{
   return psd_load(s,x,y,comp,req_comp);
}

// *************************************************************************************************
// Softimage PIC loader
// by Tom Seddon
//
// See http://softimage.wiki.softimage.com/index.php/INFO:_PIC_file_format
// See http://ozviz.wasp.uwa.edu.au/~pbourke/dataformats/softimagepic/

static int pic_is4(stbi *s,const char *str)
{
   int i;
   for (i=0; i<4; ++i)
      if (get8(s) != (stbi_uc)str[i])
         return 0;

   return 1;
}

static int pic_test(stbi *s)
{
   int i;

   if (!pic_is4(s,"\x53\x80\xF6\x34"))
      return 0;

   for(i=0;i<84;++i)
      get8(s);

   if (!pic_is4(s,"PICT"))
      return 0;

   return 1;
}

typedef struct
{
   stbi_uc size,type,channel;
} pic_packet_t;

static stbi_uc *pic_readval(stbi *s, int channel, stbi_uc *dest)
{
   int mask=0x80, i;

   for (i=0; i<4; ++i, mask>>=1) {
      if (channel & mask) {
         if (at_eof(s)) return epuc("bad file","PIC file too short");
         dest[i]=get8u(s);
      }
   }

   return dest;
}

static void pic_copyval(int channel,stbi_uc *dest,const stbi_uc *src)
{
   int mask=0x80,i;

   for (i=0;i<4; ++i, mask>>=1)
      if (channel&mask)
         dest[i]=src[i];
}

static stbi_uc *pic_load2(stbi *s,int width,int height,int *comp, stbi_uc *result)
{
   int act_comp=0,num_packets=0,y,chained;
   pic_packet_t packets[10];

   // this will (should...) cater for even some bizarre stuff like having data
    // for the same channel in multiple packets.
   do {
      pic_packet_t *packet;

      if (num_packets==sizeof(packets)/sizeof(packets[0]))
         return epuc("bad format","too many packets");

      packet = &packets[num_packets++];

      chained = get8(s);
      packet->size    = get8u(s);
      packet->type    = get8u(s);
      packet->channel = get8u(s);

      act_comp |= packet->channel;

      if (at_eof(s))          return epuc("bad file","file too short (reading packets)");
      if (packet->size != 8)  return epuc("bad format","packet isn't 8bpp");
   } while (chained);

   *comp = (act_comp & 0x10 ? 4 : 3); // has alpha channel?

   for(y=0; y<height; ++y) {
      int packet_idx;

      for(packet_idx=0; packet_idx < num_packets; ++packet_idx) {
         pic_packet_t *packet = &packets[packet_idx];
         stbi_uc *dest = result+y*width*4;

         switch (packet->type) {
            default:
               return epuc("bad format","packet has bad compression type");

            case 0: {//uncompressed
               int x;

               for(x=0;x<width;++x, dest+=4)
                  if (!pic_readval(s,packet->channel,dest))
                     return 0;
               break;
            }

            case 1://Pure RLE
               {
                  int left=width, i;

                  while (left>0) {
                     stbi_uc count,value[4];

                     count=get8u(s);
                     if (at_eof(s))   return epuc("bad file","file too short (pure read count)");

                     if (count > left)
                        count = (uint8) left;

                     if (!pic_readval(s,packet->channel,value))  return 0;

                     for(i=0; i<count; ++i,dest+=4)
                        pic_copyval(packet->channel,dest,value);
                     left -= count;
                  }
               }
               break;

            case 2: {//Mixed RLE
               int left=width;
               while (left>0) {
                  int count = get8(s), i;
                  if (at_eof(s))  return epuc("bad file","file too short (mixed read count)");

                  if (count >= 128) { // Repeated
                     stbi_uc value[4];
                     int i;

                     if (count==128)
                        count = get16(s);
                     else
                        count -= 127;
                     if (count > left)
                        return epuc("bad file","scanline overrun");

                     if (!pic_readval(s,packet->channel,value))
                        return 0;

                     for(i=0;i<count;++i, dest += 4)
                        pic_copyval(packet->channel,dest,value);
                  } else { // Raw
                     ++count;
                     if (count>left) return epuc("bad file","scanline overrun");

                     for(i=0;i<count;++i, dest+=4)
                        if (!pic_readval(s,packet->channel,dest))
                           return 0;
                  }
                  left-=count;
               }
               break;
            }
         }
      }
   }

   return result;
}

static stbi_uc *pic_load(stbi *s,int *px,int *py,int *comp,int req_comp)
{
   stbi_uc *result;
   int i, x,y;

   for (i=0; i<92; ++i)
      get8(s);

   x = get16(s);
   y = get16(s);
   if (at_eof(s))  return epuc("bad file","file too short (pic header)");
   if ((1 << 28) / x < y) return epuc("too large", "Image too large to decode");

   get32(s); //skip `ratio'
   get16(s); //skip `fields'
   get16(s); //skip `pad'

   // intermediate buffer is RGBA
   result = (stbi_uc *) malloc(x*y*4);
   memset(result, 0xff, x*y*4);

   if (!pic_load2(s,x,y,comp, result)) {
      free(result);
      result=0;
   }
   *px = x;
   *py = y;
   if (req_comp == 0) req_comp = *comp;
   result=convert_format(result,4,req_comp,x,y);

   return result;
}

static int stbi_pic_test(stbi *s)
{
   int r = pic_test(s);
   stbi_rewind(s);
   return r;
}

static stbi_uc *stbi_pic_load(stbi *s, int *x, int *y, int *comp, int req_comp)
{
   return pic_load(s,x,y,comp,req_comp);
}

// *************************************************************************************************
// GIF loader -- public domain by Jean-Marc Lienher -- simplified/shrunk by stb
typedef struct stbi_gif_lzw_struct {
   int16 prefix;
   uint8 first;
   uint8 suffix;
} stbi_gif_lzw;

typedef struct stbi_gif_struct
{
   int w,h;
   stbi_uc *out;                 // output buffer (always 4 components)
   int flags, bgindex, ratio, transparent, eflags;
   uint8  pal[256][4];
   uint8 lpal[256][4];
   stbi_gif_lzw codes[4096];
   uint8 *color_table;
   int parse, step;
   int lflags;
   int start_x, start_y;
   int max_x, max_y;
   int cur_x, cur_y;
   int line_size;
} stbi_gif;

static int gif_test(stbi *s)
{
   int sz;
   if (get8(s) != 'G' || get8(s) != 'I' || get8(s) != 'F' || get8(s) != '8') return 0;
   sz = get8(s);
   if (sz != '9' && sz != '7') return 0;
   if (get8(s) != 'a') return 0;
   return 1;
}

static int stbi_gif_test(stbi *s)
{
   int r = gif_test(s);
   stbi_rewind(s);
   return r;
}

static void stbi_gif_parse_colortable(stbi *s, uint8 pal[256][4], int num_entries, int transp)
{
   int i;
   for (i=0; i < num_entries; ++i) {
      pal[i][2] = get8u(s);
      pal[i][1] = get8u(s);
      pal[i][0] = get8u(s);
      pal[i][3] = transp ? 0 : 255;
   }
}

static int stbi_gif_header(stbi *s, stbi_gif *g, int *comp, int is_info)
{
   uint8 version;
   if (get8(s) != 'G' || get8(s) != 'I' || get8(s) != 'F' || get8(s) != '8')
      return e("not GIF", "Corrupt GIF");

   version = get8u(s);
   if (version != '7' && version != '9')    return e("not GIF", "Corrupt GIF");
   if (get8(s) != 'a')                      return e("not GIF", "Corrupt GIF");

   failure_reason = "";
   g->w = get16le(s);
   g->h = get16le(s);
   g->flags = get8(s);
   g->bgindex = get8(s);
   g->ratio = get8(s);
   g->transparent = -1;

   if (comp != 0) *comp = 4;  // can't actually tell whether it's 3 or 4 until we parse the comments

   if (is_info) return 1;

   if (g->flags & 0x80)
      stbi_gif_parse_colortable(s,g->pal, 2 << (g->flags & 7), -1);

   return 1;
}

static int stbi_gif_info_raw(stbi *s, int *x, int *y, int *comp)
{
   stbi_gif g;
   if (!stbi_gif_header(s, &g, comp, 1)) {
      stbi_rewind( s );
      return 0;
   }
   if (x) *x = g.w;
   if (y) *y = g.h;
   return 1;
}

static void stbi_out_gif_code(stbi_gif *g, uint16 code)
{
   uint8 *p, *c;

   // recurse to decode the prefixes, since the linked-list is backwards,
   // and working backwards through an interleaved image would be nasty
   if (g->codes[code].prefix >= 0)
      stbi_out_gif_code(g, g->codes[code].prefix);

   if (g->cur_y >= g->max_y) return;

   p = &g->out[g->cur_x + g->cur_y];
   c = &g->color_table[g->codes[code].suffix * 4];

   if (c[3] >= 128) {
      p[0] = c[2];
      p[1] = c[1];
      p[2] = c[0];
      p[3] = c[3];
   }
   g->cur_x += 4;

   if (g->cur_x >= g->max_x) {
      g->cur_x = g->start_x;
      g->cur_y += g->step;

      while (g->cur_y >= g->max_y && g->parse > 0) {
         g->step = (1 << g->parse) * g->line_size;
         g->cur_y = g->start_y + (g->step >> 1);
         --g->parse;
      }
   }
}

static uint8 *stbi_process_gif_raster(stbi *s, stbi_gif *g)
{
   uint8 lzw_cs;
   int32 len, code;
   uint32 first;
   int32 codesize, codemask, avail, oldcode, bits, valid_bits, clear;
   stbi_gif_lzw *p;

   lzw_cs = get8u(s);
   clear = 1 << lzw_cs;
   first = 1;
   codesize = lzw_cs + 1;
   codemask = (1 << codesize) - 1;
   bits = 0;
   valid_bits = 0;
   for (code = 0; code < clear; code++) {
      g->codes[code].prefix = -1;
      g->codes[code].first = (uint8) code;
      g->codes[code].suffix = (uint8) code;
   }

   // support no starting clear code
   avail = clear+2;
   oldcode = -1;

   len = 0;
   for(;;) {
      if (valid_bits < codesize) {
         if (len == 0) {
            len = get8(s); // start new block
            if (len == 0)
               return g->out;
         }
         --len;
         bits |= (int32) get8(s) << valid_bits;
         valid_bits += 8;
      } else {
         int32 code = bits & codemask;
         bits >>= codesize;
         valid_bits -= codesize;
         // @OPTIMIZE: is there some way we can accelerate the non-clear path?
         if (code == clear) {  // clear code
            codesize = lzw_cs + 1;
            codemask = (1 << codesize) - 1;
            avail = clear + 2;
            oldcode = -1;
            first = 0;
         } else if (code == clear + 1) { // end of stream code
            skip(s, len);
            while ((len = get8(s)) > 0)
               skip(s,len);
            return g->out;
         } else if (code <= avail) {
            if (first) return epuc("no clear code", "Corrupt GIF");

            if (oldcode >= 0) {
               p = &g->codes[avail++];
               if (avail > 4096)        return epuc("too many codes", "Corrupt GIF");
               p->prefix = (int16) oldcode;
               p->first = g->codes[oldcode].first;
               p->suffix = (code == avail) ? p->first : g->codes[code].first;
            } else if (code == avail)
               return epuc("illegal code in raster", "Corrupt GIF");

            stbi_out_gif_code(g, (uint16) code);

            if ((avail & codemask) == 0 && avail <= 0x0FFF) {
               codesize++;
               codemask = (1 << codesize) - 1;
            }

            oldcode = code;
         } else {
            return epuc("illegal code in raster", "Corrupt GIF");
         }
      }
   }
}

static void stbi_fill_gif_background(stbi_gif *g)
{
   int i;
   uint8 *c = g->pal[g->bgindex];
   // @OPTIMIZE: write a dword at a time
   for (i = 0; i < g->w * g->h * 4; i += 4) {
      uint8 *p  = &g->out[i];
      p[0] = c[2];
      p[1] = c[1];
      p[2] = c[0];
      p[3] = c[3];
   }
}

// this function is designed to support animated gifs, although stb_image doesn't support it
static uint8 *stbi_gif_load_next(stbi *s, stbi_gif *g, int *comp, int req_comp)
{
   int i;
   uint8 *old_out = 0;

   if (g->out == 0) {
      if (!stbi_gif_header(s, g, comp,0))     return 0; // failure_reason set by stbi_gif_header
      g->out = (uint8 *) malloc(4 * g->w * g->h);
      if (g->out == 0)                      return epuc("outofmem", "Out of memory");
      stbi_fill_gif_background(g);
   } else {
      // animated-gif-only path
      if (((g->eflags & 0x1C) >> 2) == 3) {
         old_out = g->out;
         g->out = (uint8 *) malloc(4 * g->w * g->h);
         if (g->out == 0)                   return epuc("outofmem", "Out of memory");
         memcpy(g->out, old_out, g->w*g->h*4);
      }
   }

   for (;;) {
      switch (get8(s)) {
         case 0x2C: /* Image Descriptor */
         {
            int32 x, y, w, h;
            uint8 *o;

            x = get16le(s);
            y = get16le(s);
            w = get16le(s);
            h = get16le(s);
            if (((x + w) > (g->w)) || ((y + h) > (g->h)))
               return epuc("bad Image Descriptor", "Corrupt GIF");

            g->line_size = g->w * 4;
            g->start_x = x * 4;
            g->start_y = y * g->line_size;
            g->max_x   = g->start_x + w * 4;
            g->max_y   = g->start_y + h * g->line_size;
            g->cur_x   = g->start_x;
            g->cur_y   = g->start_y;

            g->lflags = get8(s);

            if (g->lflags & 0x40) {
               g->step = 8 * g->line_size; // first interlaced spacing
               g->parse = 3;
            } else {
               g->step = g->line_size;
               g->parse = 0;
            }

            if (g->lflags & 0x80) {
               stbi_gif_parse_colortable(s,g->lpal, 2 << (g->lflags & 7), g->eflags & 0x01 ? g->transparent : -1);
               g->color_table = (uint8 *) g->lpal;
            } else if (g->flags & 0x80) {
               for (i=0; i < 256; ++i)  // @OPTIMIZE: reset only the previous transparent
                  g->pal[i][3] = 255;
               if (g->transparent >= 0 && (g->eflags & 0x01))
                  g->pal[g->transparent][3] = 0;
               g->color_table = (uint8 *) g->pal;
            } else
               return epuc("missing color table", "Corrupt GIF");

            o = stbi_process_gif_raster(s, g);
            if (o == NULL) return NULL;

            if (req_comp && req_comp != 4)
               o = convert_format(o, 4, req_comp, g->w, g->h);
            return o;
         }

         case 0x21: // Comment Extension.
         {
            int len;
            if (get8(s) == 0xF9) { // Graphic Control Extension.
               len = get8(s);
               if (len == 4) {
                  g->eflags = get8(s);
                  get16le(s); // delay
                  g->transparent = get8(s);
               } else {
                  skip(s, len);
                  break;
               }
            }
            while ((len = get8(s)) != 0)
               skip(s, len);
            break;
         }

         case 0x3B: // gif stream termination code
            return (uint8 *) 1;

         default:
            return epuc("unknown code", "Corrupt GIF");
      }
   }
}

static stbi_uc *stbi_gif_load(stbi *s, int *x, int *y, int *comp, int req_comp)
{
   uint8 *u = 0;
   stbi_gif g={0};

   u = stbi_gif_load_next(s, &g, comp, req_comp);
   if (u == (void *) 1) u = 0;  // end of animated gif marker
   if (u) {
      *x = g.w;
      *y = g.h;
   }

   return u;
}

static int stbi_gif_info(stbi *s, int *x, int *y, int *comp)
{
   return stbi_gif_info_raw(s,x,y,comp);
}


// *************************************************************************************************
// Radiance RGBE HDR loader
// originally by Nicolas Schulz
#ifndef STBI_NO_HDR
static int hdr_test(stbi *s)
{
   const char *signature = "#?RADIANCE\n";
   int i;
   for (i=0; signature[i]; ++i)
      if (get8(s) != signature[i])
         return 0;
   return 1;
}

static int stbi_hdr_test(stbi* s)
{
   int r = hdr_test(s);
   stbi_rewind(s);
   return r;
}

#define HDR_BUFLEN  1024
static char *hdr_gettoken(stbi *z, char *buffer)
{
   int len=0;
   char c = '\0';

   c = (char) get8(z);

   while (!at_eof(z) && c != '\n') {
      buffer[len++] = c;
      if (len == HDR_BUFLEN-1) {
         // flush to end of line
         while (!at_eof(z) && get8(z) != '\n')
            ;
         break;
      }
      c = (char) get8(z);
   }

   buffer[len] = 0;
   return buffer;
}

static void hdr_convert(float *output, stbi_uc *input, int req_comp)
{
   if ( input[3] != 0 ) {
      float f1;
      // Exponent
      f1 = (float) ldexp(1.0f, input[3] - (int)(128 + 8));
      if (req_comp <= 2)
         output[0] = (input[0] + input[1] + input[2]) * f1 / 3;
      else {
         output[0] = input[0] * f1;
         output[1] = input[1] * f1;
         output[2] = input[2] * f1;
      }
      if (req_comp == 2) output[1] = 1;
      if (req_comp == 4) output[3] = 1;
   } else {
      switch (req_comp) {
         case 4: output[3] = 1; /* fallthrough */
         case 3: output[0] = output[1] = output[2] = 0;
                 break;
         case 2: output[1] = 1; /* fallthrough */
         case 1: output[0] = 0;
                 break;
      }
   }
}

static float *hdr_load(stbi *s, int *x, int *y, int *comp, int req_comp)
{
   char buffer[HDR_BUFLEN];
   char *token;
   int valid = 0;
   int width, height;
   stbi_uc *scanline;
   float *hdr_data;
   int len;
   unsigned char count, value;
   int i, j, k, c1,c2, z;


   // Check identifier
   if (strcmp(hdr_gettoken(s,buffer), "#?RADIANCE") != 0)
      return epf("not HDR", "Corrupt HDR image");

   // Parse header
   for(;;) {
      token = hdr_gettoken(s,buffer);
      if (token[0] == 0) break;
      if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1;
   }

   if (!valid)    return epf("unsupported format", "Unsupported HDR format");

   // Parse width and height
   // can't use sscanf() if we're not using stdio!
   token = hdr_gettoken(s,buffer);
   if (strncmp(token, "-Y ", 3))  return epf("unsupported data layout", "Unsupported HDR format");
   token += 3;
   height = strtol(token, &token, 10);
   while (*token == ' ') ++token;
   if (strncmp(token, "+X ", 3))  return epf("unsupported data layout", "Unsupported HDR format");
   token += 3;
   width = strtol(token, NULL, 10);

   *x = width;
   *y = height;

   *comp = 3;
   if (req_comp == 0) req_comp = 3;

   // Read data
   hdr_data = (float *) malloc(height * width * req_comp * sizeof(float));

   // Load image data
   // image data is stored as some number of sca
   if ( width < 8 || width >= 32768) {
      // Read flat data
      for (j=0; j < height; ++j) {
         for (i=0; i < width; ++i) {
            stbi_uc rgbe[4];
           main_decode_loop:
            getn(s, rgbe, 4);
            hdr_convert(hdr_data + j * width * req_comp + i * req_comp, rgbe, req_comp);
         }
      }
   } else {
      // Read RLE-encoded data
      scanline = NULL;

      for (j = 0; j < height; ++j) {
         c1 = get8(s);
         c2 = get8(s);
         len = get8(s);
         if (c1 != 2 || c2 != 2 || (len & 0x80)) {
            // not run-length encoded, so we have to actually use THIS data as a decoded
            // pixel (note this can't be a valid pixel--one of RGB must be >= 128)
            uint8 rgbe[4];
            rgbe[0] = (uint8) c1;
            rgbe[1] = (uint8) c2;
            rgbe[2] = (uint8) len;
            rgbe[3] = (uint8) get8u(s);
            hdr_convert(hdr_data, rgbe, req_comp);
            i = 1;
            j = 0;
            free(scanline);
            goto main_decode_loop; // yes, this makes no sense
         }
         len <<= 8;
         len |= get8(s);
         if (len != width) { free(hdr_data); free(scanline); return epf("invalid decoded scanline length", "corrupt HDR"); }
         if (scanline == NULL) scanline = (stbi_uc *) malloc(width * 4);

         for (k = 0; k < 4; ++k) {
            i = 0;
            while (i < width) {
               count = get8u(s);
               if (count > 128) {
                  // Run
                  value = get8u(s);
                  count -= 128;
                  for (z = 0; z < count; ++z)
                     scanline[i++ * 4 + k] = value;
               } else {
                  // Dump
                  for (z = 0; z < count; ++z)
                     scanline[i++ * 4 + k] = get8u(s);
               }
            }
         }
         for (i=0; i < width; ++i)
            hdr_convert(hdr_data+(j*width + i)*req_comp, scanline + i*4, req_comp);
      }
      free(scanline);
   }

   return hdr_data;
}

static float *stbi_hdr_load(stbi *s, int *x, int *y, int *comp, int req_comp)
{
   return hdr_load(s,x,y,comp,req_comp);
}

static int stbi_hdr_info(stbi *s, int *x, int *y, int *comp)
{
   char buffer[HDR_BUFLEN];
   char *token;
   int valid = 0;

   if (strcmp(hdr_gettoken(s,buffer), "#?RADIANCE") != 0) {
       stbi_rewind( s );
       return 0;
   }

   for(;;) {
      token = hdr_gettoken(s,buffer);
      if (token[0] == 0) break;
      if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1;
   }

   if (!valid) {
       stbi_rewind( s );
       return 0;
   }
   token = hdr_gettoken(s,buffer);
   if (strncmp(token, "-Y ", 3)) {
       stbi_rewind( s );
       return 0;
   }
   token += 3;
   *y = strtol(token, &token, 10);
   while (*token == ' ') ++token;
   if (strncmp(token, "+X ", 3)) {
       stbi_rewind( s );
       return 0;
   }
   token += 3;
   *x = strtol(token, NULL, 10);
   *comp = 3;
   return 1;
}
#endif // STBI_NO_HDR

static int stbi_bmp_info(stbi *s, int *x, int *y, int *comp)
{
   int hsz;
   if (get8(s) != 'B' || get8(s) != 'M') {
       stbi_rewind( s );
       return 0;
   }
   skip(s,12);
   hsz = get32le(s);
   if (hsz != 12 && hsz != 40 && hsz != 56 && hsz != 108) {
       stbi_rewind( s );
       return 0;
   }
   if (hsz == 12) {
      *x = get16le(s);
      *y = get16le(s);
   } else {
      *x = get32le(s);
      *y = get32le(s);
   }
   if (get16le(s) != 1) {
       stbi_rewind( s );
       return 0;
   }
   *comp = get16le(s) / 8;
   return 1;
}

static int stbi_psd_info(stbi *s, int *x, int *y, int *comp)
{
   int channelCount;
   if (get32(s) != 0x38425053) {
       stbi_rewind( s );
       return 0;
   }
   if (get16(s) != 1) {
       stbi_rewind( s );
       return 0;
   }
   skip(s, 6);
   channelCount = get16(s);
   if (channelCount < 0 || channelCount > 16) {
       stbi_rewind( s );
       return 0;
   }
   *y = get32(s);
   *x = get32(s);
   if (get16(s) != 8) {
       stbi_rewind( s );
       return 0;
   }
   if (get16(s) != 3) {
       stbi_rewind( s );
       return 0;
   }
   *comp = 4;
   return 1;
}

static int stbi_pic_info(stbi *s, int *x, int *y, int *comp)
{
   int act_comp=0,num_packets=0,chained;
   pic_packet_t packets[10];

   skip(s, 92);

   *x = get16(s);
   *y = get16(s);
   if (at_eof(s))  return 0;
   if ( (*x) != 0 && (1 << 28) / (*x) < (*y)) {
       stbi_rewind( s );
       return 0;
   }

   skip(s, 8);

   do {
      pic_packet_t *packet;

      if (num_packets==sizeof(packets)/sizeof(packets[0]))
         return 0;

      packet = &packets[num_packets++];
      chained = get8(s);
      packet->size    = get8u(s);
      packet->type    = get8u(s);
      packet->channel = get8u(s);
      act_comp |= packet->channel;

      if (at_eof(s)) {
          stbi_rewind( s );
          return 0;
      }
      if (packet->size != 8) {
          stbi_rewind( s );
          return 0;
      }
   } while (chained);

   *comp = (act_comp & 0x10 ? 4 : 3);

   return 1;
}

static int stbi_info_main(stbi *s, int *x, int *y, int *comp)
{
   if (stbi_jpeg_info(s, x, y, comp))
       return 1;
   if (stbi_png_info(s, x, y, comp))
       return 1;
   if (stbi_gif_info(s, x, y, comp))
       return 1;
   if (stbi_bmp_info(s, x, y, comp))
       return 1;
   if (stbi_psd_info(s, x, y, comp))
       return 1;
   if (stbi_pic_info(s, x, y, comp))
       return 1;
   #ifndef STBI_NO_HDR
   if (stbi_hdr_info(s, x, y, comp))
       return 1;
   #endif
   // test tga last because it's a crappy test!
   if (stbi_tga_info(s, x, y, comp))
       return 1;
   return e("unknown image type", "Image not of any known type, or corrupt");
}

#ifndef STBI_NO_STDIO
int stbi_info(char const *filename, int *x, int *y, int *comp)
{
    FILE *f = fopen(filename, "rb");
    int result;
    if (!f) return e("can't fopen", "Unable to open file");
    result = stbi_info_from_file(f, x, y, comp);
    fclose(f);
    return result;
}

int stbi_info_from_file(FILE *f, int *x, int *y, int *comp)
{
   int r;
   stbi s;
   long pos = ftell(f);
   start_file(&s, f);
   r = stbi_info_main(&s,x,y,comp);
   fseek(f,pos,SEEK_SET);
   return r;
}
#endif // !STBI_NO_STDIO

int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp)
{
   stbi s;
   start_mem(&s,buffer,len);
   return stbi_info_main(&s,x,y,comp);
}

int stbi_info_from_callbacks(stbi_io_callbacks const *c, void *user, int *x, int *y, int *comp)
{
   stbi s;
   start_callbacks(&s, (stbi_io_callbacks *) c, user);
   return stbi_info_main(&s,x,y,comp);
}

#endif // STBI_HEADER_FILE_ONLY

/*
   revision history:
      1.33 (2011-07-14)
             make stbi_is_hdr work in STBI_NO_HDR (as specified), minor compiler-friendly improvements
      1.32 (2011-07-13)
             support for "info" function for all supported filetypes (SpartanJ)
      1.31 (2011-06-20)
             a few more leak fixes, bug in PNG handling (SpartanJ)
      1.30 (2011-06-11)
             added ability to load files via callbacks to accomidate custom input streams (Ben Wenger)
             removed deprecated format-specific test/load functions
             removed support for installable file formats (stbi_loader) -- would have been broken for IO callbacks anyway
             error cases in bmp and tga give messages and don't leak (Raymond Barbiero, grisha)
             fix inefficiency in decoding 32-bit BMP (David Woo)
      1.29 (2010-08-16)
             various warning fixes from Aurelien Pocheville
      1.28 (2010-08-01)
             fix bug in GIF palette transparency (SpartanJ)
      1.27 (2010-08-01)
             cast-to-uint8 to fix warnings
      1.26 (2010-07-24)
             fix bug in file buffering for PNG reported by SpartanJ
      1.25 (2010-07-17)
             refix trans_data warning (Won Chun)
      1.24 (2010-07-12)
             perf improvements reading from files on platforms with lock-heavy fgetc()
             minor perf improvements for jpeg
             deprecated type-specific functions so we'll get feedback if they're needed
             attempt to fix trans_data warning (Won Chun)
      1.23   fixed bug in iPhone support
      1.22 (2010-07-10)
             removed image *writing* support
             stbi_info support from Jetro Lauha
             GIF support from Jean-Marc Lienher
             iPhone PNG-extensions from James Brown
             warning-fixes from Nicolas Schulz and Janez Zemva (i.e. Janez (U+017D)emva)
      1.21   fix use of 'uint8' in header (reported by jon blow)
      1.20   added support for Softimage PIC, by Tom Seddon
      1.19   bug in interlaced PNG corruption check (found by ryg)
      1.18 2008-08-02
             fix a threading bug (local mutable static)
      1.17   support interlaced PNG
      1.16   major bugfix - convert_format converted one too many pixels
      1.15   initialize some fields for thread safety
      1.14   fix threadsafe conversion bug
             header-file-only version (#define STBI_HEADER_FILE_ONLY before including)
      1.13   threadsafe
      1.12   const qualifiers in the API
      1.11   Support installable IDCT, colorspace conversion routines
      1.10   Fixes for 64-bit (don't use "unsigned long")
             optimized upsampling by Fabian "ryg" Giesen
      1.09   Fix format-conversion for PSD code (bad global variables!)
      1.08   Thatcher Ulrich's PSD code integrated by Nicolas Schulz
      1.07   attempt to fix C++ warning/errors again
      1.06   attempt to fix C++ warning/errors again
      1.05   fix TGA loading to return correct *comp and use good luminance calc
      1.04   default float alpha is 1, not 255; use 'void *' for stbi_image_free
      1.03   bugfixes to STBI_NO_STDIO, STBI_NO_HDR
      1.02   support for (subset of) HDR files, float interface for preferred access to them
      1.01   fix bug: possible bug in handling right-side up bmps... not sure
             fix bug: the stbi_bmp_load() and stbi_tga_load() functions didn't work at all
      1.00   interface to zlib that skips zlib header
      0.99   correct handling of alpha in palette
      0.98   TGA loader by lonesock; dynamically add loaders (untested)
      0.97   jpeg errors on too large a file; also catch another malloc failure
      0.96   fix detection of invalid v value - particleman@mollyrocket forum
      0.95   during header scan, seek to markers in case of padding
      0.94   STBI_NO_STDIO to disable stdio usage; rename all #defines the same
      0.93   handle jpegtran output; verbose errors
      0.92   read 4,8,16,24,32-bit BMP files of several formats
      0.91   output 24-bit Windows 3.0 BMP files
      0.90   fix a few more warnings; bump version number to approach 1.0
      0.61   bugfixes due to Marc LeBlanc, Christopher Lloyd
      0.60   fix compiling as c++
      0.59   fix warnings: merge Dave Moore's -Wall fixes
      0.58   fix bug: zlib uncompressed mode len/nlen was wrong endian
      0.57   fix bug: jpg last huffman symbol before marker was >9 bits but less than 16 available
      0.56   fix bug: zlib uncompressed mode len vs. nlen
      0.55   fix bug: restart_interval not initialized to 0
      0.54   allow NULL for 'int *comp'
      0.53   fix bug in png 3->4; speedup png decoding
      0.52   png handles req_comp=3,4 directly; minor cleanup; jpeg comments
      0.51   obey req_comp requests, 1-component jpegs return as 1-component,
             on 'test' only check type, not whether we support this variant
      0.50   first released version
*/


#endif // STB_IMAGE_H