diff --git a/cores/stb_image.h b/cores/stb_image.h new file mode 100644 index 0000000000..a43313c10c --- /dev/null +++ b/cores/stb_image.h @@ -0,0 +1,3705 @@ +#ifndef STBI_INCLUDE_STB_IMAGE_H +#define STBI_INCLUDE_STB_IMAGE_H + +#include + +#define STBI_VERSION 1 + +enum +{ + STBI_default = 0, + + STBI_grey = 1, + STBI_grey_alpha = 2, + STBI_rgb = 3, + STBI_rgb_alpha = 4 +}; + +#ifdef __cplusplus +extern "C" { +#endif + +#ifdef STB_IMAGE_STATIC +#define STBIDEF static +#else +#define STBIDEF extern +#endif + +typedef struct +{ + int (*read) (void *user,char *data,int size); + void (*skip) (void *user,int n); + int (*eof) (void *user); +} stbi_io_callbacks; + +STBIDEF uint8_t *stbi_load (char const *filename, int *x, int *y, int *comp, int req_comp); + +#ifndef STBI_NO_STDIO +STBIDEF uint8_t *stbi_load_from_file (FILE *f, int *x, int *y, int *comp, int req_comp); +#endif + +STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header); + +#ifdef __cplusplus +} +#endif + +#endif + +#ifdef STB_IMAGE_IMPLEMENTATION + +#if defined(STBI_ONLY_JPEG) || defined(STBI_ONLY_PNG) || defined(STBI_ONLY_BMP) \ + || defined(STBI_ONLY_TGA) || defined(STBI_ONLY_GIF) || defined(STBI_ONLY_PSD) \ + || defined(STBI_ONLY_HDR) || defined(STBI_ONLY_PIC) || defined(STBI_ONLY_PNM) \ + || defined(STBI_ONLY_ZLIB) + #ifndef STBI_ONLY_JPEG + #define STBI_NO_JPEG + #endif + #ifndef STBI_ONLY_PNG + #define STBI_NO_PNG + #endif + #ifndef STBI_ONLY_BMP + #define STBI_NO_BMP + #endif + #ifndef STBI_ONLY_TGA + #define STBI_NO_TGA + #endif +#endif + +#if defined(STBI_NO_PNG) && !defined(STBI_SUPPORT_ZLIB) && !defined(STBI_NO_ZLIB) +#define STBI_NO_ZLIB +#endif + + +#include +#include +#include +#include + +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +#include +#endif + +#ifndef STBI_NO_STDIO +#include +#endif + +#ifndef STBI_ASSERT +#include +#define STBI_ASSERT(x) assert(x) +#endif + +#include + +typedef unsigned char validate_uint32[sizeof(uint32_t)==4 ? 1 : -1]; + +#ifdef _MSC_VER +#define STBI_NOTUSED(v) (void)(v) +#else +#define STBI_NOTUSED(v) (void)sizeof(v) +#endif + +#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 + +#if defined(__x86_64__) || defined(_M_X64) +#define STBI__X64_TARGET +#elif defined(__i386) || defined(_M_IX86) +#define STBI__X86_TARGET +#endif + +#if defined(__GNUC__) && (defined(STBI__X86_TARGET) || defined(STBI__X64_TARGET)) && !defined(__SSE2__) && !defined(STBI_NO_SIMD) +#define STBI_NO_SIMD +#endif + +#if defined(__MINGW32__) && defined(STBI__X86_TARGET) && !defined(STBI_MINGW_ENABLE_SSE2) && !defined(STBI_NO_SIMD) +#define STBI_NO_SIMD +#endif + +#if !defined(STBI_NO_SIMD) && defined(STBI__X86_TARGET) +#define STBI_SSE2 +#include + +#ifdef _MSC_VER + +#if _MSC_VER >= 1400 +#include +static int stbi__cpuid3(void) +{ + int info[4]; + __cpuid(info,1); + return info[3]; +} +#else +static int stbi__cpuid3(void) +{ + int res; + __asm { + mov eax,1 + cpuid + mov res,edx + } + return res; +} +#endif + +#define STBI_SIMD_ALIGN(type, name) __declspec(align(16)) type name + +static int stbi__sse2_available() +{ + int info3 = stbi__cpuid3(); + return ((info3 >> 26) & 1) != 0; +} +#else +#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) + +static int stbi__sse2_available() +{ +#if defined(__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__) >= 408 + return __builtin_cpu_supports("sse2"); +#else + return 0; +#endif +} +#endif +#endif + +#if defined(STBI_NO_SIMD) && defined(STBI_NEON) +#undef STBI_NEON +#endif + +#ifdef STBI_NEON +#include +#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) +#endif + +#ifndef STBI_SIMD_ALIGN +#define STBI_SIMD_ALIGN(type, name) type name +#endif + +typedef struct +{ + uint32_t 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_t buffer_start[128]; + + uint8_t *img_buffer, *img_buffer_end; + uint8_t *img_buffer_original; +} stbi__context; + + +static void stbi__refill_buffer(stbi__context *s); + +static void stbi__start_callbacks(stbi__context *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; + stbi__refill_buffer(s); +} + +#ifndef STBI_NO_STDIO + +static int stbi__stdio_read(void *user, char *data, int size) +{ + return (int) fread(data,1,size,(FILE*) user); +} + +static void stbi__stdio_skip(void *user, int n) +{ + fseek((FILE*) user, n, SEEK_CUR); +} + +static int stbi__stdio_eof(void *user) +{ + return feof((FILE*) user); +} + +static stbi_io_callbacks stbi__stdio_callbacks = +{ + stbi__stdio_read, + stbi__stdio_skip, + stbi__stdio_eof, +}; + +static void stbi__start_file(stbi__context *s, FILE *f) +{ + stbi__start_callbacks(s, &stbi__stdio_callbacks, (void *) f); +} + +#endif + +static void stbi__rewind(stbi__context *s) +{ + s->img_buffer = s->img_buffer_original; +} + +#ifndef STBI_NO_JPEG +static int stbi__jpeg_test(stbi__context *s); +static uint8_t *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, int req_comp); +#endif + +#ifndef STBI_NO_PNG +static int stbi__png_test(stbi__context *s); +static uint8_t *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, int req_comp); +#endif + +#ifndef STBI_NO_BMP +static int stbi__bmp_test(stbi__context *s); +static uint8_t *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp); +#endif + +#ifndef STBI_NO_TGA +static int stbi__tga_test(stbi__context *s); +static uint8_t *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp); +#endif + +static const char *stbi__g_failure_reason; + +static int stbi__err(const char *str) +{ + stbi__g_failure_reason = str; + return 0; +} + +#ifdef STBI_NO_FAILURE_STRINGS + #define stbi__err(x,y) 0 +#elif defined(STBI_FAILURE_USERMSG) + #define stbi__err(x,y) stbi__err(y) +#else + #define stbi__err(x,y) stbi__err(x) +#endif + +#define stbi__errpf(x,y) ((float *) (stbi__err(x,y)?NULL:NULL)) +#define stbi__errpuc(x,y) ((unsigned char *) (stbi__err(x,y)?NULL:NULL)) + +static int stbi__vertically_flip_on_load = 0; + +static unsigned char *stbi__load_main(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + #ifndef STBI_NO_JPEG + if (stbi__jpeg_test(s)) return stbi__jpeg_load(s,x,y,comp,req_comp); + #endif + #ifndef STBI_NO_PNG + if (stbi__png_test(s)) return stbi__png_load(s,x,y,comp,req_comp); + #endif + #ifndef STBI_NO_BMP + if (stbi__bmp_test(s)) return stbi__bmp_load(s,x,y,comp,req_comp); + #endif + + #ifndef STBI_NO_TGA + if (stbi__tga_test(s)) + return stbi__tga_load(s,x,y,comp,req_comp); + #endif + + return stbi__errpuc("unknown image type", "Image not of any known type, or corrupt"); +} + +static unsigned char *stbi__load_flip(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + unsigned char *result = stbi__load_main(s, x, y, comp, req_comp); + + if (stbi__vertically_flip_on_load && result != NULL) + { + int row,col,z; + uint8_t temp; + int w = *x, h = *y; + int depth = req_comp ? req_comp : *comp; + + for (row = 0; row < (h>>1); row++) + { + for (col = 0; col < w; col++) + { + for (z = 0; z < depth; z++) + { + temp = result[(row * w + col) * depth + z]; + result[(row * w + col) * depth + z] = result[((h - row - 1) * w + col) * depth + z]; + result[((h - row - 1) * w + col) * depth + z] = temp; + } + } + } + } + + return result; +} + +#ifndef STBI_NO_STDIO + +static FILE *stbi__fopen(char const *filename, char const *mode) +{ + FILE *f; +#if defined(_MSC_VER) && _MSC_VER >= 1400 + if (0 != fopen_s(&f, filename, mode)) + f=0; +#else + f = fopen(filename, mode); +#endif + return f; +} + + +STBIDEF uint8_t *stbi_load(char const *filename, int *x, int *y, int *comp, int req_comp) +{ + unsigned char *result; + FILE *f = stbi__fopen(filename, "rb"); + if (!f) + return stbi__errpuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file(f,x,y,comp,req_comp); + fclose(f); + return result; +} + +STBIDEF uint8_t *stbi_load_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) +{ + unsigned char *result; + stbi__context s; + stbi__start_file(&s,f); + result = stbi__load_flip(&s,x,y,comp,req_comp); + if (result) + fseek(f, - (int) (s.img_buffer_end - s.img_buffer), SEEK_CUR); + return result; +} +#endif + +static float stbi__h2l_gamma_i=1.0f/2.2f, stbi__h2l_scale_i=1.0f; +static float stbi__l2h_gamma=2.2f, stbi__l2h_scale=1.0f; + +enum +{ + STBI__SCAN_load=0, + STBI__SCAN_type, + STBI__SCAN_header +}; + +static void stbi__refill_buffer(stbi__context *s) +{ + int n = (s->io.read)(s->io_user_data,(char*)s->buffer_start,s->buflen); + if (n == 0) + { + s->read_from_callbacks = 0; + s->img_buffer = s->buffer_start; + s->img_buffer_end = s->buffer_start+1; + *s->img_buffer = 0; + } else { + s->img_buffer = s->buffer_start; + s->img_buffer_end = s->buffer_start + n; + } +} + +static INLINE uint8_t stbi__get8(stbi__context *s) +{ + if (s->img_buffer < s->img_buffer_end) + return *s->img_buffer++; + if (s->read_from_callbacks) { + stbi__refill_buffer(s); + return *s->img_buffer++; + } + return 0; +} + +static INLINE int stbi__at_eof(stbi__context *s) +{ + if (s->io.read) + { + if (!(s->io.eof)(s->io_user_data)) + return 0; + if (s->read_from_callbacks == 0) return 1; + } + + return s->img_buffer >= s->img_buffer_end; +} + +static void stbi__skip(stbi__context *s, int n) +{ + if (n < 0) { + s->img_buffer = s->img_buffer_end; + return; + } + if (s->io.read) { + int blen = (int) (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 stbi__getn(stbi__context *s, uint8_t *buffer, int n) +{ + if (s->io.read) { + int blen = (int) (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 stbi__get16be(stbi__context *s) +{ + int z = stbi__get8(s); + return (z << 8) + stbi__get8(s); +} + +static uint32_t stbi__get32be(stbi__context *s) +{ + uint32_t z = stbi__get16be(s); + return (z << 16) + stbi__get16be(s); +} + +static int stbi__get16le(stbi__context *s) +{ + int z = stbi__get8(s); + return z + (stbi__get8(s) << 8); +} + +static uint32_t stbi__get32le(stbi__context *s) +{ + uint32_t z = stbi__get16le(s); + return z + (stbi__get16le(s) << 16); +} + +#define STBI__BYTECAST(x) ((uint8_t) ((x) & 255)) + +static uint8_t stbi__compute_y(int r, int g, int b) +{ + return (uint8_t) (((r*77) + (g*150) + (29*b)) >> 8); +} + +static unsigned char *stbi__convert_format(unsigned char *data, int img_n, int req_comp, unsigned int x, unsigned int y) +{ + int i,j; + unsigned char *good; + + if (req_comp == img_n) return data; + STBI_ASSERT(req_comp >= 1 && req_comp <= 4); + + good = (unsigned char *) malloc(req_comp * x * y); + + if (good == NULL) + { + free(data); + return stbi__errpuc("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) + 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]=stbi__compute_y(src[0],src[1],src[2]); break; + CASE(3,2) dest[0]=stbi__compute_y(src[0],src[1],src[2]), dest[1] = 255; break; + CASE(4,1) dest[0]=stbi__compute_y(src[0],src[1],src[2]); break; + CASE(4,2) dest[0]=stbi__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: STBI_ASSERT(0); + } + #undef CASE + } + + free(data); + return good; +} + +#ifndef STBI_NO_JPEG + +#define FAST_BITS 9 + +typedef struct +{ + uint8_t fast[1 << FAST_BITS]; + uint16_t code[256]; + uint8_t values[256]; + uint8_t size[257]; + unsigned int maxcode[18]; + int delta[17]; +} stbi__huffman; + +typedef struct +{ + stbi__context *s; + stbi__huffman huff_dc[4]; + stbi__huffman huff_ac[4]; + uint8_t dequant[4][64]; + int16_t fast_ac[4][1 << FAST_BITS]; + + int img_h_max, img_v_max; + int img_mcu_x, img_mcu_y; + int img_mcu_w, img_mcu_h; + + struct + { + int id; + int h,v; + int tq; + int hd,ha; + int dc_pred; + + int x,y,w2,h2; + uint8_t *data; + void *raw_data, *raw_coeff; + uint8_t *linebuf; + short *coeff; + int coeff_w, coeff_h; + } img_comp[4]; + + uint32_t code_buffer; + int code_bits; + unsigned char marker; + int nomore; + + int progressive; + int spec_start; + int spec_end; + int succ_high; + int succ_low; + int eob_run; + + int scan_n, order[4]; + int restart_interval, todo; + + void (*idct_block_kernel)(uint8_t *out, int out_stride, short data[64]); + void (*YCbCr_to_RGB_kernel)(uint8_t *out, const uint8_t *y, const uint8_t *pcb, const uint8_t *pcr, int count, int step); + uint8_t *(*resample_row_hv_2_kernel)(uint8_t *out, uint8_t *in_near, uint8_t *in_far, int w, int hs); +} stbi__jpeg; + +static int stbi__build_huffman(stbi__huffman *h, int *count) +{ + int i,j,k=0,code; + + for (i=0; i < 16; ++i) + for (j=0; j < count[i]; ++j) + h->size[k++] = (uint8_t) (i+1); + h->size[k] = 0; + + code = 0; + k = 0; + for(j=1; j <= 16; ++j) { + h->delta[j] = k - code; + if (h->size[k] == j) { + while (h->size[k] == j) + h->code[k++] = (uint16_t) (code++); + if (code-1 >= (1 << j)) return stbi__err("bad code lengths","Corrupt JPEG"); + } + h->maxcode[j] = code << (16-j); + code <<= 1; + } + h->maxcode[j] = 0xffffffff; + + 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_t) i; + } + } + } + return 1; +} + +static void stbi__build_fast_ac(int16_t *fast_ac, stbi__huffman *h) +{ + int i; + for (i=0; i < (1 << FAST_BITS); ++i) { + uint8_t fast = h->fast[i]; + fast_ac[i] = 0; + if (fast < 255) { + int rs = h->values[fast]; + int run = (rs >> 4) & 15; + int magbits = rs & 15; + int len = h->size[fast]; + + if (magbits && len + magbits <= FAST_BITS) + { + int k = ((i << len) & ((1 << FAST_BITS) - 1)) >> (FAST_BITS - magbits); + int m = 1 << (magbits - 1); + if (k < m) k += (-1 << magbits) + 1; + if (k >= -128 && k <= 127) + fast_ac[i] = (int16_t) ((k << 8) + (run << 4) + (len + magbits)); + } + } + } +} + +static void stbi__grow_buffer_unsafe(stbi__jpeg *j) +{ + do { + int b = j->nomore ? 0 : stbi__get8(j->s); + if (b == 0xff) { + int c = stbi__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); +} + +static uint32_t stbi__bmask[17]={0,1,3,7,15,31,63,127,255,511,1023,2047,4095,8191,16383,32767,65535}; + +static INLINE int stbi__jpeg_huff_decode(stbi__jpeg *j, stbi__huffman *h) +{ + unsigned int temp; + int c,k; + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + + 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]; + } + + temp = j->code_buffer >> 16; + for (k=FAST_BITS+1 ; ; ++k) + if (temp < h->maxcode[k]) + break; + if (k == 17) { + j->code_bits -= 16; + return -1; + } + + if (k > j->code_bits) + return -1; + + c = ((j->code_buffer >> (32 - k)) & stbi__bmask[k]) + h->delta[k]; + STBI_ASSERT((((j->code_buffer) >> (32 - h->size[c])) & stbi__bmask[h->size[c]]) == h->code[c]); + + j->code_bits -= k; + j->code_buffer <<= k; + return h->values[c]; +} + +static int const stbi__jbias[16] = {0,-1,-3,-7,-15,-31,-63,-127,-255,-511,-1023,-2047,-4095,-8191,-16383,-32767}; + +static INLINE int stbi__extend_receive(stbi__jpeg *j, int n) +{ + unsigned int k; + int sgn; + if (j->code_bits < n) stbi__grow_buffer_unsafe(j); + + sgn = (int32_t)j->code_buffer >> 31; + k = stbi_lrot(j->code_buffer, n); + STBI_ASSERT(n >= 0 && n < sizeof(stbi__bmask)/sizeof(*stbi__bmask)); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k + (stbi__jbias[n] & ~sgn); +} + +static INLINE int stbi__jpeg_get_bits(stbi__jpeg *j, int n) +{ + unsigned int k; + if (j->code_bits < n) stbi__grow_buffer_unsafe(j); + k = stbi_lrot(j->code_buffer, n); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k; +} + +static INLINE int stbi__jpeg_get_bit(stbi__jpeg *j) +{ + unsigned int k; + if (j->code_bits < 1) stbi__grow_buffer_unsafe(j); + k = j->code_buffer; + j->code_buffer <<= 1; + --j->code_bits; + return k & 0x80000000; +} + +static uint8_t stbi__jpeg_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, + 63, 63, 63, 63, 63, 63, 63, 63, + 63, 63, 63, 63, 63, 63, 63 +}; + +static int stbi__jpeg_decode_block(stbi__jpeg *j, short data[64], stbi__huffman *hdc, stbi__huffman *hac, int16_t *fac, int b, uint8_t *dequant) +{ + int diff,dc,k; + int t; + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + t = stbi__jpeg_huff_decode(j, hdc); + if (t < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + + memset(data,0,64*sizeof(data[0])); + + diff = t ? stbi__extend_receive(j, t) : 0; + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + data[0] = (short) (dc * dequant[0]); + + k = 1; + do { + unsigned int zig; + int c,r,s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); + r = fac[c]; + if (r) + { + k += (r >> 4) & 15; + s = r & 15; + j->code_buffer <<= s; + j->code_bits -= s; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) ((r >> 8) * dequant[zig]); + } else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (rs != 0xf0) break; + k += 16; + } else { + k += r; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) (stbi__extend_receive(j,s) * dequant[zig]); + } + } + } while (k < 64); + return 1; +} + +static int stbi__jpeg_decode_block_prog_dc(stbi__jpeg *j, short data[64], stbi__huffman *hdc, int b) +{ + int diff,dc; + int t; + if (j->spec_end != 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + + if (j->succ_high == 0) { + memset(data,0,64*sizeof(data[0])); + t = stbi__jpeg_huff_decode(j, hdc); + diff = t ? stbi__extend_receive(j, t) : 0; + + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + data[0] = (short) (dc << j->succ_low); + } else { + if (stbi__jpeg_get_bit(j)) + data[0] += (short) (1 << j->succ_low); + } + return 1; +} + +static int stbi__jpeg_decode_block_prog_ac(stbi__jpeg *j, short data[64], stbi__huffman *hac, int16_t *fac) +{ + int k; + if (j->spec_start == 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + + if (j->succ_high == 0) { + int shift = j->succ_low; + + if (j->eob_run) { + --j->eob_run; + return 1; + } + + k = j->spec_start; + do { + unsigned int zig; + int c,r,s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); + r = fac[c]; + if (r) + { + k += (r >> 4) & 15; + s = r & 15; + j->code_buffer <<= s; + j->code_bits -= s; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) ((r >> 8) << shift); + } else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r); + if (r) + j->eob_run += stbi__jpeg_get_bits(j, r); + --j->eob_run; + break; + } + k += 16; + } else { + k += r; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short) (stbi__extend_receive(j,s) << shift); + } + } + } while (k <= j->spec_end); + } + else + { + short bit = (short) (1 << j->succ_low); + + if (j->eob_run) { + --j->eob_run; + for (k = j->spec_start; k <= j->spec_end; ++k) { + short *p = &data[stbi__jpeg_dezigzag[k]]; + if (*p != 0) + if (stbi__jpeg_get_bit(j)) + if ((*p & bit)==0) { + if (*p > 0) + *p += bit; + else + *p -= bit; + } + } + } else { + k = j->spec_start; + do { + int r,s; + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r) - 1; + if (r) + j->eob_run += stbi__jpeg_get_bits(j, r); + r = 64; + } else + r = 16; + } else { + if (s != 1) return stbi__err("bad huffman code", "Corrupt JPEG"); + if (stbi__jpeg_get_bit(j)) + s = bit; + else + s = -bit; + } + + while (k <= j->spec_end) { + short *p = &data[stbi__jpeg_dezigzag[k]]; + if (*p != 0) { + if (stbi__jpeg_get_bit(j)) + if ((*p & bit)==0) { + if (*p > 0) + *p += bit; + else + *p -= bit; + } + ++k; + } else { + if (r == 0) { + if (s) + data[stbi__jpeg_dezigzag[k++]] = (short) s; + break; + } + --r; + ++k; + } + } + } while (k <= j->spec_end); + } + } + return 1; +} + +static INLINE uint8_t stbi__clamp(int x) +{ + if ((unsigned int) x > 255) { + if (x < 0) return 0; + if (x > 255) return 255; + } + return (uint8_t) x; +} + +#define stbi__f2f(x) ((int) (((x) * 4096 + 0.5))) +#define stbi__fsh(x) ((x) << 12) + +#define STBI__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) * stbi__f2f(0.5411961f); \ + t2 = p1 + p3*stbi__f2f(-1.847759065f); \ + t3 = p1 + p2*stbi__f2f( 0.765366865f); \ + p2 = s0; \ + p3 = s4; \ + t0 = stbi__fsh(p2+p3); \ + t1 = stbi__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)*stbi__f2f( 1.175875602f); \ + t0 = t0*stbi__f2f( 0.298631336f); \ + t1 = t1*stbi__f2f( 2.053119869f); \ + t2 = t2*stbi__f2f( 3.072711026f); \ + t3 = t3*stbi__f2f( 1.501321110f); \ + p1 = p5 + p1*stbi__f2f(-0.899976223f); \ + p2 = p5 + p2*stbi__f2f(-2.562915447f); \ + p3 = p3*stbi__f2f(-1.961570560f); \ + p4 = p4*stbi__f2f(-0.390180644f); \ + t3 += p1+p4; \ + t2 += p2+p3; \ + t1 += p2+p4; \ + t0 += p1+p3; + +static void stbi__idct_block(uint8_t *out, int out_stride, short data[64]) +{ + int i,val[64],*v=val; + uint8_t *o; + short *d = data; + + for (i=0; i < 8; ++i,++d, ++v) { + if (d[ 8]==0 && d[16]==0 && d[24]==0 && d[32]==0 + && d[40]==0 && d[48]==0 && d[56]==0) { + int dcterm = d[0] << 2; + v[0] = v[8] = v[16] = v[24] = v[32] = v[40] = v[48] = v[56] = dcterm; + } else { + STBI__IDCT_1D(d[ 0],d[ 8],d[16],d[24],d[32],d[40],d[48],d[56]) + 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) { + STBI__IDCT_1D(v[0],v[1],v[2],v[3],v[4],v[5],v[6],v[7]) + x0 += 65536 + (128<<17); + x1 += 65536 + (128<<17); + x2 += 65536 + (128<<17); + x3 += 65536 + (128<<17); + o[0] = stbi__clamp((x0+t3) >> 17); + o[7] = stbi__clamp((x0-t3) >> 17); + o[1] = stbi__clamp((x1+t2) >> 17); + o[6] = stbi__clamp((x1-t2) >> 17); + o[2] = stbi__clamp((x2+t1) >> 17); + o[5] = stbi__clamp((x2-t1) >> 17); + o[3] = stbi__clamp((x3+t0) >> 17); + o[4] = stbi__clamp((x3-t0) >> 17); + } +} + +#ifdef STBI_SSE2 +static void stbi__idct_simd(uint8_t *out, int out_stride, short data[64]) +{ + __m128i row0, row1, row2, row3, row4, row5, row6, row7; + __m128i tmp; + + #define dct_const(x,y) _mm_setr_epi16((x),(y),(x),(y),(x),(y),(x),(y)) + + #define dct_rot(out0,out1, x,y,c0,c1) \ + __m128i c0##lo = _mm_unpacklo_epi16((x),(y)); \ + __m128i c0##hi = _mm_unpackhi_epi16((x),(y)); \ + __m128i out0##_l = _mm_madd_epi16(c0##lo, c0); \ + __m128i out0##_h = _mm_madd_epi16(c0##hi, c0); \ + __m128i out1##_l = _mm_madd_epi16(c0##lo, c1); \ + __m128i out1##_h = _mm_madd_epi16(c0##hi, c1) + + #define dct_widen(out, in) \ + __m128i out##_l = _mm_srai_epi32(_mm_unpacklo_epi16(_mm_setzero_si128(), (in)), 4); \ + __m128i out##_h = _mm_srai_epi32(_mm_unpackhi_epi16(_mm_setzero_si128(), (in)), 4) + + #define dct_wadd(out, a, b) \ + __m128i out##_l = _mm_add_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_add_epi32(a##_h, b##_h) + + #define dct_wsub(out, a, b) \ + __m128i out##_l = _mm_sub_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_sub_epi32(a##_h, b##_h) + + #define dct_bfly32o(out0, out1, a,b,bias,s) \ + { \ + __m128i abiased_l = _mm_add_epi32(a##_l, bias); \ + __m128i abiased_h = _mm_add_epi32(a##_h, bias); \ + dct_wadd(sum, abiased, b); \ + dct_wsub(dif, abiased, b); \ + out0 = _mm_packs_epi32(_mm_srai_epi32(sum_l, s), _mm_srai_epi32(sum_h, s)); \ + out1 = _mm_packs_epi32(_mm_srai_epi32(dif_l, s), _mm_srai_epi32(dif_h, s)); \ + } + + #define dct_interleave8(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi8(a, b); \ + b = _mm_unpackhi_epi8(tmp, b) + + #define dct_interleave16(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi16(a, b); \ + b = _mm_unpackhi_epi16(tmp, b) + + #define dct_pass(bias,shift) \ + { \ + /* even part */ \ + dct_rot(t2e,t3e, row2,row6, rot0_0,rot0_1); \ + __m128i sum04 = _mm_add_epi16(row0, row4); \ + __m128i dif04 = _mm_sub_epi16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + dct_rot(y0o,y2o, row7,row3, rot2_0,rot2_1); \ + dct_rot(y1o,y3o, row5,row1, rot3_0,rot3_1); \ + __m128i sum17 = _mm_add_epi16(row1, row7); \ + __m128i sum35 = _mm_add_epi16(row3, row5); \ + dct_rot(y4o,y5o, sum17,sum35, rot1_0,rot1_1); \ + dct_wadd(x4, y0o, y4o); \ + dct_wadd(x5, y1o, y5o); \ + dct_wadd(x6, y2o, y5o); \ + dct_wadd(x7, y3o, y4o); \ + dct_bfly32o(row0,row7, x0,x7,bias,shift); \ + dct_bfly32o(row1,row6, x1,x6,bias,shift); \ + dct_bfly32o(row2,row5, x2,x5,bias,shift); \ + dct_bfly32o(row3,row4, x3,x4,bias,shift); \ + } + + __m128i rot0_0 = dct_const(stbi__f2f(0.5411961f), stbi__f2f(0.5411961f) + stbi__f2f(-1.847759065f)); + __m128i rot0_1 = dct_const(stbi__f2f(0.5411961f) + stbi__f2f( 0.765366865f), stbi__f2f(0.5411961f)); + __m128i rot1_0 = dct_const(stbi__f2f(1.175875602f) + stbi__f2f(-0.899976223f), stbi__f2f(1.175875602f)); + __m128i rot1_1 = dct_const(stbi__f2f(1.175875602f), stbi__f2f(1.175875602f) + stbi__f2f(-2.562915447f)); + __m128i rot2_0 = dct_const(stbi__f2f(-1.961570560f) + stbi__f2f( 0.298631336f), stbi__f2f(-1.961570560f)); + __m128i rot2_1 = dct_const(stbi__f2f(-1.961570560f), stbi__f2f(-1.961570560f) + stbi__f2f( 3.072711026f)); + __m128i rot3_0 = dct_const(stbi__f2f(-0.390180644f) + stbi__f2f( 2.053119869f), stbi__f2f(-0.390180644f)); + __m128i rot3_1 = dct_const(stbi__f2f(-0.390180644f), stbi__f2f(-0.390180644f) + stbi__f2f( 1.501321110f)); + + __m128i bias_0 = _mm_set1_epi32(512); + __m128i bias_1 = _mm_set1_epi32(65536 + (128<<17)); + + row0 = _mm_load_si128((const __m128i *) (data + 0*8)); + row1 = _mm_load_si128((const __m128i *) (data + 1*8)); + row2 = _mm_load_si128((const __m128i *) (data + 2*8)); + row3 = _mm_load_si128((const __m128i *) (data + 3*8)); + row4 = _mm_load_si128((const __m128i *) (data + 4*8)); + row5 = _mm_load_si128((const __m128i *) (data + 5*8)); + row6 = _mm_load_si128((const __m128i *) (data + 6*8)); + row7 = _mm_load_si128((const __m128i *) (data + 7*8)); + + dct_pass(bias_0, 10); + + { + dct_interleave16(row0, row4); + dct_interleave16(row1, row5); + dct_interleave16(row2, row6); + dct_interleave16(row3, row7); + + dct_interleave16(row0, row2); + dct_interleave16(row1, row3); + dct_interleave16(row4, row6); + dct_interleave16(row5, row7); + + dct_interleave16(row0, row1); + dct_interleave16(row2, row3); + dct_interleave16(row4, row5); + dct_interleave16(row6, row7); + } + + dct_pass(bias_1, 17); + + { + __m128i p0 = _mm_packus_epi16(row0, row1); + __m128i p1 = _mm_packus_epi16(row2, row3); + __m128i p2 = _mm_packus_epi16(row4, row5); + __m128i p3 = _mm_packus_epi16(row6, row7); + + dct_interleave8(p0, p2); + dct_interleave8(p1, p3); + + dct_interleave8(p0, p1); + dct_interleave8(p2, p3); + + dct_interleave8(p0, p2); + dct_interleave8(p1, p3); + + _mm_storel_epi64((__m128i *) out, p0); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p0, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i *) out, p2); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p2, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i *) out, p1); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p1, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i *) out, p3); out += out_stride; + _mm_storel_epi64((__m128i *) out, _mm_shuffle_epi32(p3, 0x4e)); + } + +#undef dct_const +#undef dct_rot +#undef dct_widen +#undef dct_wadd +#undef dct_wsub +#undef dct_bfly32o +#undef dct_interleave8 +#undef dct_interleave16 +#undef dct_pass +} + +#endif + +#ifdef STBI_NEON + +static void stbi__idct_simd(uint8_t *out, int out_stride, short data[64]) +{ + int16x8_t row0, row1, row2, row3, row4, row5, row6, row7; + + int16x4_t rot0_0 = vdup_n_s16(stbi__f2f(0.5411961f)); + int16x4_t rot0_1 = vdup_n_s16(stbi__f2f(-1.847759065f)); + int16x4_t rot0_2 = vdup_n_s16(stbi__f2f( 0.765366865f)); + int16x4_t rot1_0 = vdup_n_s16(stbi__f2f( 1.175875602f)); + int16x4_t rot1_1 = vdup_n_s16(stbi__f2f(-0.899976223f)); + int16x4_t rot1_2 = vdup_n_s16(stbi__f2f(-2.562915447f)); + int16x4_t rot2_0 = vdup_n_s16(stbi__f2f(-1.961570560f)); + int16x4_t rot2_1 = vdup_n_s16(stbi__f2f(-0.390180644f)); + int16x4_t rot3_0 = vdup_n_s16(stbi__f2f( 0.298631336f)); + int16x4_t rot3_1 = vdup_n_s16(stbi__f2f( 2.053119869f)); + int16x4_t rot3_2 = vdup_n_s16(stbi__f2f( 3.072711026f)); + int16x4_t rot3_3 = vdup_n_s16(stbi__f2f( 1.501321110f)); + +#define dct_long_mul(out, inq, coeff) \ + int32x4_t out##_l = vmull_s16(vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmull_s16(vget_high_s16(inq), coeff) + +#define dct_long_mac(out, acc, inq, coeff) \ + int32x4_t out##_l = vmlal_s16(acc##_l, vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmlal_s16(acc##_h, vget_high_s16(inq), coeff) + +#define dct_widen(out, inq) \ + int32x4_t out##_l = vshll_n_s16(vget_low_s16(inq), 12); \ + int32x4_t out##_h = vshll_n_s16(vget_high_s16(inq), 12) + +#define dct_wadd(out, a, b) \ + int32x4_t out##_l = vaddq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vaddq_s32(a##_h, b##_h) + +#define dct_wsub(out, a, b) \ + int32x4_t out##_l = vsubq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vsubq_s32(a##_h, b##_h) + +#define dct_bfly32o(out0,out1, a,b,shiftop,s) \ + { \ + dct_wadd(sum, a, b); \ + dct_wsub(dif, a, b); \ + out0 = vcombine_s16(shiftop(sum_l, s), shiftop(sum_h, s)); \ + out1 = vcombine_s16(shiftop(dif_l, s), shiftop(dif_h, s)); \ + } + +#define dct_pass(shiftop, shift) \ + { \ + /* even part */ \ + int16x8_t sum26 = vaddq_s16(row2, row6); \ + dct_long_mul(p1e, sum26, rot0_0); \ + dct_long_mac(t2e, p1e, row6, rot0_1); \ + dct_long_mac(t3e, p1e, row2, rot0_2); \ + int16x8_t sum04 = vaddq_s16(row0, row4); \ + int16x8_t dif04 = vsubq_s16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + int16x8_t sum15 = vaddq_s16(row1, row5); \ + int16x8_t sum17 = vaddq_s16(row1, row7); \ + int16x8_t sum35 = vaddq_s16(row3, row5); \ + int16x8_t sum37 = vaddq_s16(row3, row7); \ + int16x8_t sumodd = vaddq_s16(sum17, sum35); \ + dct_long_mul(p5o, sumodd, rot1_0); \ + dct_long_mac(p1o, p5o, sum17, rot1_1); \ + dct_long_mac(p2o, p5o, sum35, rot1_2); \ + dct_long_mul(p3o, sum37, rot2_0); \ + dct_long_mul(p4o, sum15, rot2_1); \ + dct_wadd(sump13o, p1o, p3o); \ + dct_wadd(sump24o, p2o, p4o); \ + dct_wadd(sump23o, p2o, p3o); \ + dct_wadd(sump14o, p1o, p4o); \ + dct_long_mac(x4, sump13o, row7, rot3_0); \ + dct_long_mac(x5, sump24o, row5, rot3_1); \ + dct_long_mac(x6, sump23o, row3, rot3_2); \ + dct_long_mac(x7, sump14o, row1, rot3_3); \ + dct_bfly32o(row0,row7, x0,x7,shiftop,shift); \ + dct_bfly32o(row1,row6, x1,x6,shiftop,shift); \ + dct_bfly32o(row2,row5, x2,x5,shiftop,shift); \ + dct_bfly32o(row3,row4, x3,x4,shiftop,shift); \ + } + + row0 = vld1q_s16(data + 0*8); + row1 = vld1q_s16(data + 1*8); + row2 = vld1q_s16(data + 2*8); + row3 = vld1q_s16(data + 3*8); + row4 = vld1q_s16(data + 4*8); + row5 = vld1q_s16(data + 5*8); + row6 = vld1q_s16(data + 6*8); + row7 = vld1q_s16(data + 7*8); + + row0 = vaddq_s16(row0, vsetq_lane_s16(1024, vdupq_n_s16(0), 0)); + + dct_pass(vrshrn_n_s32, 10); + + { +#define dct_trn16(x, y) { int16x8x2_t t = vtrnq_s16(x, y); x = t.val[0]; y = t.val[1]; } +#define dct_trn32(x, y) { int32x4x2_t t = vtrnq_s32(vreinterpretq_s32_s16(x), vreinterpretq_s32_s16(y)); x = vreinterpretq_s16_s32(t.val[0]); y = vreinterpretq_s16_s32(t.val[1]); } +#define dct_trn64(x, y) { int16x8_t x0 = x; int16x8_t y0 = y; x = vcombine_s16(vget_low_s16(x0), vget_low_s16(y0)); y = vcombine_s16(vget_high_s16(x0), vget_high_s16(y0)); } + + dct_trn16(row0, row1); + dct_trn16(row2, row3); + dct_trn16(row4, row5); + dct_trn16(row6, row7); + + dct_trn32(row0, row2); + dct_trn32(row1, row3); + dct_trn32(row4, row6); + dct_trn32(row5, row7); + + dct_trn64(row0, row4); + dct_trn64(row1, row5); + dct_trn64(row2, row6); + dct_trn64(row3, row7); + +#undef dct_trn16 +#undef dct_trn32 +#undef dct_trn64 + } + + dct_pass(vshrn_n_s32, 16); + + { + uint8x8_t p0 = vqrshrun_n_s16(row0, 1); + uint8x8_t p1 = vqrshrun_n_s16(row1, 1); + uint8x8_t p2 = vqrshrun_n_s16(row2, 1); + uint8x8_t p3 = vqrshrun_n_s16(row3, 1); + uint8x8_t p4 = vqrshrun_n_s16(row4, 1); + uint8x8_t p5 = vqrshrun_n_s16(row5, 1); + uint8x8_t p6 = vqrshrun_n_s16(row6, 1); + uint8x8_t p7 = vqrshrun_n_s16(row7, 1); + +#define dct_trn8_8(x, y) { uint8x8x2_t t = vtrn_u8(x, y); x = t.val[0]; y = t.val[1]; } +#define dct_trn8_16(x, y) { uint16x4x2_t t = vtrn_u16(vreinterpret_u16_u8(x), vreinterpret_u16_u8(y)); x = vreinterpret_u8_u16(t.val[0]); y = vreinterpret_u8_u16(t.val[1]); } +#define dct_trn8_32(x, y) { uint32x2x2_t t = vtrn_u32(vreinterpret_u32_u8(x), vreinterpret_u32_u8(y)); x = vreinterpret_u8_u32(t.val[0]); y = vreinterpret_u8_u32(t.val[1]); } + + dct_trn8_8(p0, p1); + dct_trn8_8(p2, p3); + dct_trn8_8(p4, p5); + dct_trn8_8(p6, p7); + + dct_trn8_16(p0, p2); + dct_trn8_16(p1, p3); + dct_trn8_16(p4, p6); + dct_trn8_16(p5, p7); + + dct_trn8_32(p0, p4); + dct_trn8_32(p1, p5); + dct_trn8_32(p2, p6); + dct_trn8_32(p3, p7); + + vst1_u8(out, p0); out += out_stride; + vst1_u8(out, p1); out += out_stride; + vst1_u8(out, p2); out += out_stride; + vst1_u8(out, p3); out += out_stride; + vst1_u8(out, p4); out += out_stride; + vst1_u8(out, p5); out += out_stride; + vst1_u8(out, p6); out += out_stride; + vst1_u8(out, p7); + +#undef dct_trn8_8 +#undef dct_trn8_16 +#undef dct_trn8_32 + } + +#undef dct_long_mul +#undef dct_long_mac +#undef dct_widen +#undef dct_wadd +#undef dct_wsub +#undef dct_bfly32o +#undef dct_pass +} + +#endif + +#define STBI__MARKER_none 0xff +static uint8_t stbi__get_marker(stbi__jpeg *j) +{ + uint8_t x; + if (j->marker != STBI__MARKER_none) { x = j->marker; j->marker = STBI__MARKER_none; return x; } + x = stbi__get8(j->s); + if (x != 0xff) return STBI__MARKER_none; + while (x == 0xff) + x = stbi__get8(j->s); + return x; +} + +#define STBI__RESTART(x) ((x) >= 0xd0 && (x) <= 0xd7) + +static void stbi__jpeg_reset(stbi__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 = STBI__MARKER_none; + j->todo = j->restart_interval ? j->restart_interval : 0x7fffffff; + j->eob_run = 0; +} + +static int stbi__parse_entropy_coded_data(stbi__jpeg *z) +{ + stbi__jpeg_reset(z); + if (!z->progressive) + { + if (z->scan_n == 1) + { + int i,j; + STBI_SIMD_ALIGN(short, data[64]); + int n = z->order[0]; + 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) { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; + z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data); + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + else + { + int i,j,k,x,y; + STBI_SIMD_ALIGN(short, data[64]); + for (j=0; j < z->img_mcu_y; ++j) + { + for (i=0; i < z->img_mcu_x; ++i) + { + for (k=0; k < z->scan_n; ++k) + { + int n = z->order[k]; + 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; + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; + z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*y2+x2, z->img_comp[n].w2, data); + } + } + } + + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } + else + { + if (z->scan_n == 1) + { + int i,j; + int n = z->order[0]; + 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) { + short *data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + if (z->spec_start == 0) { + if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } else { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block_prog_ac(z, data, &z->huff_ac[ha], z->fast_ac[ha])) + return 0; + } + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + else + { + int i,j,k,x,y; + for (j=0; j < z->img_mcu_y; ++j) { + for (i=0; i < z->img_mcu_x; ++i) { + for (k=0; k < z->scan_n; ++k) { + int n = z->order[k]; + 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); + int y2 = (j*z->img_comp[n].v + y); + short *data = z->img_comp[n].coeff + 64 * (x2 + y2 * z->img_comp[n].coeff_w); + if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } + } + } + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } +} + +static void stbi__jpeg_dequantize(short *data, uint8_t *dequant) +{ + int i; + for (i=0; i < 64; ++i) + data[i] *= dequant[i]; +} + +static void stbi__jpeg_finish(stbi__jpeg *z) +{ + if (z->progressive) { + int i,j,n; + for (n=0; n < z->s->img_n; ++n) { + 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) { + short *data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + stbi__jpeg_dequantize(data, z->dequant[z->img_comp[n].tq]); + z->idct_block_kernel(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data); + } + } + } + } +} + +static int stbi__process_marker(stbi__jpeg *z, int m) +{ + int L; + switch (m) { + case STBI__MARKER_none: + return stbi__err("expected marker","Corrupt JPEG"); + + case 0xDD: + if (stbi__get16be(z->s) != 4) return stbi__err("bad DRI len","Corrupt JPEG"); + z->restart_interval = stbi__get16be(z->s); + return 1; + + case 0xDB: + L = stbi__get16be(z->s)-2; + while (L > 0) { + int q = stbi__get8(z->s); + int p = q >> 4; + int t = q & 15,i; + if (p != 0) return stbi__err("bad DQT type","Corrupt JPEG"); + if (t > 3) return stbi__err("bad DQT table","Corrupt JPEG"); + for (i=0; i < 64; ++i) + z->dequant[t][stbi__jpeg_dezigzag[i]] = stbi__get8(z->s); + L -= 65; + } + return L==0; + + case 0xC4: + L = stbi__get16be(z->s)-2; + while (L > 0) { + uint8_t *v; + int sizes[16],i,n=0; + int q = stbi__get8(z->s); + int tc = q >> 4; + int th = q & 15; + if (tc > 1 || th > 3) return stbi__err("bad DHT header","Corrupt JPEG"); + for (i=0; i < 16; ++i) { + sizes[i] = stbi__get8(z->s); + n += sizes[i]; + } + L -= 17; + if (tc == 0) { + if (!stbi__build_huffman(z->huff_dc+th, sizes)) return 0; + v = z->huff_dc[th].values; + } else { + if (!stbi__build_huffman(z->huff_ac+th, sizes)) return 0; + v = z->huff_ac[th].values; + } + for (i=0; i < n; ++i) + v[i] = stbi__get8(z->s); + if (tc != 0) + stbi__build_fast_ac(z->fast_ac[th], z->huff_ac + th); + L -= n; + } + return L==0; + } + if ((m >= 0xE0 && m <= 0xEF) || m == 0xFE) { + stbi__skip(z->s, stbi__get16be(z->s)-2); + return 1; + } + return 0; +} + +static int stbi__process_scan_header(stbi__jpeg *z) +{ + int i; + int Ls = stbi__get16be(z->s); + z->scan_n = stbi__get8(z->s); + if (z->scan_n < 1 || z->scan_n > 4 || z->scan_n > (int) z->s->img_n) return stbi__err("bad SOS component count","Corrupt JPEG"); + if (Ls != 6+2*z->scan_n) return stbi__err("bad SOS len","Corrupt JPEG"); + for (i=0; i < z->scan_n; ++i) { + int id = stbi__get8(z->s), which; + int q = stbi__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 stbi__err("bad DC huff","Corrupt JPEG"); + z->img_comp[which].ha = q & 15; if (z->img_comp[which].ha > 3) return stbi__err("bad AC huff","Corrupt JPEG"); + z->order[i] = which; + } + + { + int aa; + z->spec_start = stbi__get8(z->s); + z->spec_end = stbi__get8(z->s); + aa = stbi__get8(z->s); + z->succ_high = (aa >> 4); + z->succ_low = (aa & 15); + if (z->progressive) { + if (z->spec_start > 63 || z->spec_end > 63 || z->spec_start > z->spec_end || z->succ_high > 13 || z->succ_low > 13) + return stbi__err("bad SOS", "Corrupt JPEG"); + } else { + if (z->spec_start != 0) return stbi__err("bad SOS","Corrupt JPEG"); + if (z->succ_high != 0 || z->succ_low != 0) return stbi__err("bad SOS","Corrupt JPEG"); + z->spec_end = 63; + } + } + + return 1; +} + +static int stbi__process_frame_header(stbi__jpeg *z, int scan) +{ + stbi__context *s = z->s; + int Lf,p,i,q, h_max=1,v_max=1,c; + Lf = stbi__get16be(s); if (Lf < 11) return stbi__err("bad SOF len","Corrupt JPEG"); + p = stbi__get8(s); if (p != 8) return stbi__err("only 8-bit","JPEG format not supported: 8-bit only"); + s->img_y = stbi__get16be(s); if (s->img_y == 0) return stbi__err("no header height", "JPEG format not supported: delayed height"); + s->img_x = stbi__get16be(s); if (s->img_x == 0) return stbi__err("0 width","Corrupt JPEG"); + c = stbi__get8(s); + if (c != 3 && c != 1) return stbi__err("bad component count","Corrupt JPEG"); + 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 stbi__err("bad SOF len","Corrupt JPEG"); + + for (i=0; i < s->img_n; ++i) { + z->img_comp[i].id = stbi__get8(s); + if (z->img_comp[i].id != i+1) + if (z->img_comp[i].id != i) + return stbi__err("bad component ID","Corrupt JPEG"); + q = stbi__get8(s); + z->img_comp[i].h = (q >> 4); if (!z->img_comp[i].h || z->img_comp[i].h > 4) return stbi__err("bad H","Corrupt JPEG"); + z->img_comp[i].v = q & 15; if (!z->img_comp[i].v || z->img_comp[i].v > 4) return stbi__err("bad V","Corrupt JPEG"); + z->img_comp[i].tq = stbi__get8(s); if (z->img_comp[i].tq > 3) return stbi__err("bad TQ","Corrupt JPEG"); + } + + if (scan != STBI__SCAN_load) return 1; + + if ((1 << 30) / s->img_x / s->img_n < s->img_y) return stbi__err("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; + } + + 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) { + 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; + 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 stbi__err("outofmem", "Out of memory"); + } + z->img_comp[i].data = (uint8_t*) (((size_t) z->img_comp[i].raw_data + 15) & ~15); + z->img_comp[i].linebuf = NULL; + if (z->progressive) { + z->img_comp[i].coeff_w = (z->img_comp[i].w2 + 7) >> 3; + z->img_comp[i].coeff_h = (z->img_comp[i].h2 + 7) >> 3; + z->img_comp[i].raw_coeff = malloc(z->img_comp[i].coeff_w * z->img_comp[i].coeff_h * 64 * sizeof(short) + 15); + z->img_comp[i].coeff = (short*) (((size_t) z->img_comp[i].raw_coeff + 15) & ~15); + } else { + z->img_comp[i].coeff = 0; + z->img_comp[i].raw_coeff = 0; + } + } + + return 1; +} + +#define stbi__DNL(x) ((x) == 0xdc) +#define stbi__SOI(x) ((x) == 0xd8) +#define stbi__EOI(x) ((x) == 0xd9) +#define stbi__SOF(x) ((x) == 0xc0 || (x) == 0xc1 || (x) == 0xc2) +#define stbi__SOS(x) ((x) == 0xda) + +#define stbi__SOF_progressive(x) ((x) == 0xc2) + +static int stbi__decode_jpeg_header(stbi__jpeg *z, int scan) +{ + int m; + z->marker = STBI__MARKER_none; + m = stbi__get_marker(z); + if (!stbi__SOI(m)) return stbi__err("no SOI","Corrupt JPEG"); + if (scan == STBI__SCAN_type) return 1; + m = stbi__get_marker(z); + while (!stbi__SOF(m)) + { + if (!stbi__process_marker(z,m)) return 0; + m = stbi__get_marker(z); + while (m == STBI__MARKER_none) + { + if (stbi__at_eof(z->s)) return stbi__err("no SOF", "Corrupt JPEG"); + m = stbi__get_marker(z); + } + } + z->progressive = stbi__SOF_progressive(m); + if (!stbi__process_frame_header(z, scan)) return 0; + return 1; +} + +static int stbi__decode_jpeg_image(stbi__jpeg *j) +{ + int m; + for (m = 0; m < 4; m++) { + j->img_comp[m].raw_data = NULL; + j->img_comp[m].raw_coeff = NULL; + } + j->restart_interval = 0; + if (!stbi__decode_jpeg_header(j, STBI__SCAN_load)) return 0; + m = stbi__get_marker(j); + while (!stbi__EOI(m)) { + if (stbi__SOS(m)) { + if (!stbi__process_scan_header(j)) return 0; + if (!stbi__parse_entropy_coded_data(j)) return 0; + if (j->marker == STBI__MARKER_none ) { + while (!stbi__at_eof(j->s)) { + int x = stbi__get8(j->s); + if (x == 255) { + j->marker = stbi__get8(j->s); + break; + } else if (x != 0) { + return stbi__err("junk before marker", "Corrupt JPEG"); + } + } + } + } else { + if (!stbi__process_marker(j, m)) return 0; + } + m = stbi__get_marker(j); + } + if (j->progressive) + stbi__jpeg_finish(j); + return 1; +} + + +typedef uint8_t *(*resample_row_func)(uint8_t *out, uint8_t *in0, uint8_t *in1, + int w, int hs); + +#define stbi__div4(x) ((uint8_t) ((x) >> 2)) + +static uint8_t *resample_row_1(uint8_t *out, uint8_t *in_near, uint8_t *in_far, int w, int hs) +{ + STBI_NOTUSED(out); + STBI_NOTUSED(in_far); + STBI_NOTUSED(w); + STBI_NOTUSED(hs); + return in_near; +} + +static uint8_t* stbi__resample_row_v_2(uint8_t *out, uint8_t *in_near, uint8_t *in_far, int w, int hs) +{ + int i; + STBI_NOTUSED(hs); + for (i=0; i < w; ++i) + out[i] = stbi__div4(3*in_near[i] + in_far[i] + 2); + return out; +} + +static uint8_t* stbi__resample_row_h_2(uint8_t *out, uint8_t *in_near, uint8_t *in_far, int w, int hs) +{ + int i; + uint8_t *input = in_near; + + if (w == 1) { + out[0] = out[1] = input[0]; + return out; + } + + out[0] = input[0]; + out[1] = stbi__div4(input[0]*3 + input[1] + 2); + for (i=1; i < w-1; ++i) { + int n = 3*input[i]+2; + out[i*2+0] = stbi__div4(n+input[i-1]); + out[i*2+1] = stbi__div4(n+input[i+1]); + } + out[i*2+0] = stbi__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 stbi__div16(x) ((uint8_t) ((x) >> 4)) + +static uint8_t *stbi__resample_row_hv_2(uint8_t *out, uint8_t *in_near, uint8_t *in_far, int w, int hs) +{ + int i,t0,t1; + if (w == 1) { + out[0] = out[1] = stbi__div4(3*in_near[0] + in_far[0] + 2); + return out; + } + + t1 = 3*in_near[0] + in_far[0]; + out[0] = stbi__div4(t1+2); + for (i=1; i < w; ++i) { + t0 = t1; + t1 = 3*in_near[i]+in_far[i]; + out[i*2-1] = stbi__div16(3*t0 + t1 + 8); + out[i*2 ] = stbi__div16(3*t1 + t0 + 8); + } + out[w*2-1] = stbi__div4(t1+2); + + STBI_NOTUSED(hs); + + return out; +} + +#if defined(STBI_SSE2) || defined(STBI_NEON) +static uint8_t *stbi__resample_row_hv_2_simd(uint8_t *out, uint8_t *in_near, uint8_t *in_far, int w, int hs) +{ + int i=0,t0,t1; + + if (w == 1) { + out[0] = out[1] = stbi__div4(3*in_near[0] + in_far[0] + 2); + return out; + } + + t1 = 3*in_near[0] + in_far[0]; + for (; i < ((w-1) & ~7); i += 8) + { +#if defined(STBI_SSE2) + __m128i zero = _mm_setzero_si128(); + __m128i farb = _mm_loadl_epi64((__m128i *) (in_far + i)); + __m128i nearb = _mm_loadl_epi64((__m128i *) (in_near + i)); + __m128i farw = _mm_unpacklo_epi8(farb, zero); + __m128i nearw = _mm_unpacklo_epi8(nearb, zero); + __m128i diff = _mm_sub_epi16(farw, nearw); + __m128i nears = _mm_slli_epi16(nearw, 2); + __m128i curr = _mm_add_epi16(nears, diff); + __m128i prv0 = _mm_slli_si128(curr, 2); + __m128i nxt0 = _mm_srli_si128(curr, 2); + __m128i prev = _mm_insert_epi16(prv0, t1, 0); + __m128i next = _mm_insert_epi16(nxt0, 3*in_near[i+8] + in_far[i+8], 7); + __m128i bias = _mm_set1_epi16(8); + __m128i curs = _mm_slli_epi16(curr, 2); + __m128i prvd = _mm_sub_epi16(prev, curr); + __m128i nxtd = _mm_sub_epi16(next, curr); + __m128i curb = _mm_add_epi16(curs, bias); + __m128i even = _mm_add_epi16(prvd, curb); + __m128i odd = _mm_add_epi16(nxtd, curb); + __m128i int0 = _mm_unpacklo_epi16(even, odd); + __m128i int1 = _mm_unpackhi_epi16(even, odd); + __m128i de0 = _mm_srli_epi16(int0, 4); + __m128i de1 = _mm_srli_epi16(int1, 4); + __m128i outv = _mm_packus_epi16(de0, de1); + _mm_storeu_si128((__m128i *) (out + i*2), outv); +#elif defined(STBI_NEON) + uint8x8x2_t o; + uint8x8_t farb = vld1_u8(in_far + i); + uint8x8_t nearb = vld1_u8(in_near + i); + int16x8_t diff = vreinterpretq_s16_u16(vsubl_u8(farb, nearb)); + int16x8_t nears = vreinterpretq_s16_u16(vshll_n_u8(nearb, 2)); + int16x8_t curr = vaddq_s16(nears, diff); + int16x8_t prv0 = vextq_s16(curr, curr, 7); + int16x8_t nxt0 = vextq_s16(curr, curr, 1); + int16x8_t prev = vsetq_lane_s16(t1, prv0, 0); + int16x8_t next = vsetq_lane_s16(3*in_near[i+8] + in_far[i+8], nxt0, 7); + int16x8_t curs = vshlq_n_s16(curr, 2); + int16x8_t prvd = vsubq_s16(prev, curr); + int16x8_t nxtd = vsubq_s16(next, curr); + int16x8_t even = vaddq_s16(curs, prvd); + int16x8_t odd = vaddq_s16(curs, nxtd); + o.val[0] = vqrshrun_n_s16(even, 4); + o.val[1] = vqrshrun_n_s16(odd, 4); + vst2_u8(out + i*2, o); +#endif + + t1 = 3*in_near[i+7] + in_far[i+7]; + } + + t0 = t1; + t1 = 3*in_near[i] + in_far[i]; + out[i*2] = stbi__div16(3*t1 + t0 + 8); + + for (++i; i < w; ++i) { + t0 = t1; + t1 = 3*in_near[i]+in_far[i]; + out[i*2-1] = stbi__div16(3*t0 + t1 + 8); + out[i*2 ] = stbi__div16(3*t1 + t0 + 8); + } + out[w*2-1] = stbi__div4(t1+2); + + STBI_NOTUSED(hs); + + return out; +} +#endif + +static uint8_t *stbi__resample_row_generic(uint8_t *out, uint8_t *in_near, uint8_t *in_far, int w, int hs) +{ + int i,j; + STBI_NOTUSED(in_far); + for (i=0; i < w; ++i) + for (j=0; j < hs; ++j) + out[i*hs+j] = in_near[i]; + return out; +} + +#ifdef STBI_JPEG_OLD +#define float2fixed(x) ((int) ((x) * 65536 + 0.5)) +static void stbi__YCbCr_to_RGB_row(uint8_t *out, const uint8_t *y, const uint8_t *pcb, const uint8_t *pcr, int count, int step) +{ + int i; + for (i=0; i < count; ++i) + { + int y_fixed = (y[i] << 16) + 32768; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + int r = y_fixed + cr*float2fixed(1.40200f); + int g = y_fixed - cr*float2fixed(0.71414f) - cb*float2fixed(0.34414f); + int 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_t)r; + out[1] = (uint8_t)g; + out[2] = (uint8_t)b; + out[3] = 255; + out += step; + } +} +#else +#define float2fixed(x) (((int) ((x) * 4096.0f + 0.5f)) << 8) +static void stbi__YCbCr_to_RGB_row(uint8_t *out, const uint8_t *y, const uint8_t *pcb, const uint8_t *pcr, int count, int step) +{ + int i; + for (i=0; i < count; ++i) { + int y_fixed = (y[i] << 20) + (1<<19); + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + int r = y_fixed + cr* float2fixed(1.40200f); + int g = y_fixed + (cr*-float2fixed(0.71414f)) + ((cb*-float2fixed(0.34414f)) & 0xffff0000); + int b = y_fixed + cb* float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + 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_t)r; + out[1] = (uint8_t)g; + out[2] = (uint8_t)b; + out[3] = 255; + out += step; + } +} +#endif + +#if defined(STBI_SSE2) || defined(STBI_NEON) +static void stbi__YCbCr_to_RGB_simd(uint8_t *out, uint8_t const *y, uint8_t const *pcb, uint8_t const *pcr, int count, int step) +{ + int i = 0; + +#ifdef STBI_SSE2 + if (step == 4) + { + __m128i signflip = _mm_set1_epi8(-0x80); + __m128i cr_const0 = _mm_set1_epi16( (short) ( 1.40200f*4096.0f+0.5f)); + __m128i cr_const1 = _mm_set1_epi16( - (short) ( 0.71414f*4096.0f+0.5f)); + __m128i cb_const0 = _mm_set1_epi16( - (short) ( 0.34414f*4096.0f+0.5f)); + __m128i cb_const1 = _mm_set1_epi16( (short) ( 1.77200f*4096.0f+0.5f)); + __m128i y_bias = _mm_set1_epi8((char) (unsigned char) 128); + __m128i xw = _mm_set1_epi16(255); + + for (; i+7 < count; i += 8) { + __m128i y_bytes = _mm_loadl_epi64((__m128i *) (y+i)); + __m128i cr_bytes = _mm_loadl_epi64((__m128i *) (pcr+i)); + __m128i cb_bytes = _mm_loadl_epi64((__m128i *) (pcb+i)); + __m128i cr_biased = _mm_xor_si128(cr_bytes, signflip); + __m128i cb_biased = _mm_xor_si128(cb_bytes, signflip); + + __m128i yw = _mm_unpacklo_epi8(y_bias, y_bytes); + __m128i crw = _mm_unpacklo_epi8(_mm_setzero_si128(), cr_biased); + __m128i cbw = _mm_unpacklo_epi8(_mm_setzero_si128(), cb_biased); + + __m128i yws = _mm_srli_epi16(yw, 4); + __m128i cr0 = _mm_mulhi_epi16(cr_const0, crw); + __m128i cb0 = _mm_mulhi_epi16(cb_const0, cbw); + __m128i cb1 = _mm_mulhi_epi16(cbw, cb_const1); + __m128i cr1 = _mm_mulhi_epi16(crw, cr_const1); + __m128i rws = _mm_add_epi16(cr0, yws); + __m128i gwt = _mm_add_epi16(cb0, yws); + __m128i bws = _mm_add_epi16(yws, cb1); + __m128i gws = _mm_add_epi16(gwt, cr1); + + __m128i rw = _mm_srai_epi16(rws, 4); + __m128i bw = _mm_srai_epi16(bws, 4); + __m128i gw = _mm_srai_epi16(gws, 4); + + __m128i brb = _mm_packus_epi16(rw, bw); + __m128i gxb = _mm_packus_epi16(gw, xw); + + __m128i t0 = _mm_unpacklo_epi8(brb, gxb); + __m128i t1 = _mm_unpackhi_epi8(brb, gxb); + __m128i o0 = _mm_unpacklo_epi16(t0, t1); + __m128i o1 = _mm_unpackhi_epi16(t0, t1); + + _mm_storeu_si128((__m128i *) (out + 0), o0); + _mm_storeu_si128((__m128i *) (out + 16), o1); + out += 32; + } + } +#endif + +#ifdef STBI_NEON + if (step == 4) + { + uint8x8_t signflip = vdup_n_u8(0x80); + int16x8_t cr_const0 = vdupq_n_s16( (short) ( 1.40200f*4096.0f+0.5f)); + int16x8_t cr_const1 = vdupq_n_s16( - (short) ( 0.71414f*4096.0f+0.5f)); + int16x8_t cb_const0 = vdupq_n_s16( - (short) ( 0.34414f*4096.0f+0.5f)); + int16x8_t cb_const1 = vdupq_n_s16( (short) ( 1.77200f*4096.0f+0.5f)); + + for (; i+7 < count; i += 8) + { + uint8x8x4_t o; + uint8x8_t y_bytes = vld1_u8(y + i); + uint8x8_t cr_bytes = vld1_u8(pcr + i); + uint8x8_t cb_bytes = vld1_u8(pcb + i); + int8x8_t cr_biased = vreinterpret_s8_u8(vsub_u8(cr_bytes, signflip)); + int8x8_t cb_biased = vreinterpret_s8_u8(vsub_u8(cb_bytes, signflip)); + + int16x8_t yws = vreinterpretq_s16_u16(vshll_n_u8(y_bytes, 4)); + int16x8_t crw = vshll_n_s8(cr_biased, 7); + int16x8_t cbw = vshll_n_s8(cb_biased, 7); + + int16x8_t cr0 = vqdmulhq_s16(crw, cr_const0); + int16x8_t cb0 = vqdmulhq_s16(cbw, cb_const0); + int16x8_t cr1 = vqdmulhq_s16(crw, cr_const1); + int16x8_t cb1 = vqdmulhq_s16(cbw, cb_const1); + int16x8_t rws = vaddq_s16(yws, cr0); + int16x8_t gws = vaddq_s16(vaddq_s16(yws, cb0), cr1); + int16x8_t bws = vaddq_s16(yws, cb1); + + o.val[0] = vqrshrun_n_s16(rws, 4); + o.val[1] = vqrshrun_n_s16(gws, 4); + o.val[2] = vqrshrun_n_s16(bws, 4); + o.val[3] = vdup_n_u8(255); + + vst4_u8(out, o); + out += 8*4; + } + } +#endif + + for (; i < count; ++i) + { + int y_fixed = (y[i] << 20) + (1<<19); + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + int r = y_fixed + cr* float2fixed(1.40200f); + int g = y_fixed + cr*-float2fixed(0.71414f) + ((cb*-float2fixed(0.34414f)) & 0xffff0000); + int b = y_fixed + cb* float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + 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_t)r; + out[1] = (uint8_t)g; + out[2] = (uint8_t)b; + out[3] = 255; + out += step; + } +} +#endif + +static void stbi__setup_jpeg(stbi__jpeg *j) +{ + j->idct_block_kernel = stbi__idct_block; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_row; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2; + +#ifdef STBI_SSE2 + if (stbi__sse2_available()) + { + j->idct_block_kernel = stbi__idct_simd; +#ifndef STBI_JPEG_OLD + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; +#endif + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; + } +#endif + +#ifdef STBI_NEON + j->idct_block_kernel = stbi__idct_simd; +#ifndef STBI_JPEG_OLD + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; +#endif + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; +#endif +} + +static void stbi__cleanup_jpeg(stbi__jpeg *j) +{ + int i; + for (i=0; i < j->s->img_n; ++i) + { + if (j->img_comp[i].raw_data) + { + free(j->img_comp[i].raw_data); + j->img_comp[i].raw_data = NULL; + j->img_comp[i].data = NULL; + } + if (j->img_comp[i].raw_coeff) + { + free(j->img_comp[i].raw_coeff); + j->img_comp[i].raw_coeff = 0; + j->img_comp[i].coeff = 0; + } + 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_t *line0,*line1; + int hs,vs; + int w_lores; + int ystep; + int ypos; +} stbi__resample; + +static uint8_t *load_jpeg_image(stbi__jpeg *z, int *out_x, int *out_y, int *comp, int req_comp) +{ + int n, decode_n; + z->s->img_n = 0; + + if (req_comp < 0 || req_comp > 4) + return stbi__errpuc("bad req_comp", "Internal error"); + + if (!stbi__decode_jpeg_image(z)) { stbi__cleanup_jpeg(z); return NULL; } + + 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; + + { + int k; + unsigned int i,j; + uint8_t *output; + uint8_t *coutput[4]; + + stbi__resample res_comp[4]; + + for (k=0; k < decode_n; ++k) { + stbi__resample *r = &res_comp[k]; + + z->img_comp[k].linebuf = (uint8_t *) malloc(z->s->img_x + 3); + if (!z->img_comp[k].linebuf) { stbi__cleanup_jpeg(z); return stbi__errpuc("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 = stbi__resample_row_v_2; + else if (r->hs == 2 && r->vs == 1) r->resample = stbi__resample_row_h_2; + else if (r->hs == 2 && r->vs == 2) r->resample = z->resample_row_hv_2_kernel; + else r->resample = stbi__resample_row_generic; + } + + output = (uint8_t *) malloc(n * z->s->img_x * z->s->img_y + 1); + if (!output) { stbi__cleanup_jpeg(z); return stbi__errpuc("outofmem", "Out of memory"); } + + for (j=0; j < z->s->img_y; ++j) { + uint8_t *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_t *y = coutput[0]; + if (z->s->img_n == 3) { + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + } else + for (i=0; i < z->s->img_x; ++i) { + out[0] = out[1] = out[2] = y[i]; + out[3] = 255; + out += n; + } + } else { + uint8_t *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; + } + } + stbi__cleanup_jpeg(z); + *out_x = z->s->img_x; + *out_y = z->s->img_y; + if (comp) *comp = z->s->img_n; + return output; + } +} + +static unsigned char *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + stbi__jpeg j; + j.s = s; + stbi__setup_jpeg(&j); + return load_jpeg_image(&j, x,y,comp,req_comp); +} + +static int stbi__jpeg_test(stbi__context *s) +{ + int r; + stbi__jpeg j; + j.s = s; + stbi__setup_jpeg(&j); + r = stbi__decode_jpeg_header(&j, STBI__SCAN_type); + stbi__rewind(s); + return r; +} +#endif + +#ifndef STBI_NO_ZLIB + +#define STBI__ZFAST_BITS 9 +#define STBI__ZFAST_MASK ((1 << STBI__ZFAST_BITS) - 1) + +typedef struct +{ + uint16_t fast[1 << STBI__ZFAST_BITS]; + uint16_t firstcode[16]; + int maxcode[17]; + uint16_t firstsymbol[16]; + uint8_t size[288]; + uint16_t value[288]; +} stbi__zhuffman; + +static INLINE int stbi__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; +} + +static INLINE int stbi__bit_reverse(int v, int bits) +{ + STBI_ASSERT(bits <= 16); + return stbi__bitreverse16(v) >> (16-bits); +} + +static int stbi__zbuild_huffman(stbi__zhuffman *z, uint8_t *sizelist, int num) +{ + int i,k=0; + int code, next_code[16], sizes[17]; + + memset(sizes, 0, sizeof(sizes)); + memset(z->fast, 0, sizeof(z->fast)); + for (i=0; i < num; ++i) + ++sizes[sizelist[i]]; + sizes[0] = 0; + for (i=1; i < 16; ++i) + if (sizes[i] > (1 << i)) + return stbi__err("bad sizes", "Corrupt PNG"); + code = 0; + for (i=1; i < 16; ++i) { + next_code[i] = code; + z->firstcode[i] = (uint16_t) code; + z->firstsymbol[i] = (uint16_t) k; + code = (code + sizes[i]); + if (sizes[i]) + if (code-1 >= (1 << i)) return stbi__err("bad codelengths","Corrupt PNG"); + z->maxcode[i] = code << (16-i); + code <<= 1; + k += sizes[i]; + } + z->maxcode[16] = 0x10000; + for (i=0; i < num; ++i) + { + int s = sizelist[i]; + if (s) + { + int c = next_code[s] - z->firstcode[s] + z->firstsymbol[s]; + uint16_t fastv = (uint16_t) ((s << 9) | i); + z->size [c] = (uint8_t ) s; + z->value[c] = (uint16_t) i; + if (s <= STBI__ZFAST_BITS) { + int k = stbi__bit_reverse(next_code[s],s); + while (k < (1 << STBI__ZFAST_BITS)) { + z->fast[k] = fastv; + k += (1 << s); + } + } + ++next_code[s]; + } + } + return 1; +} + +typedef struct +{ + uint8_t *zbuffer, *zbuffer_end; + int num_bits; + uint32_t code_buffer; + + char *zout; + char *zout_start; + char *zout_end; + int z_expandable; + + stbi__zhuffman z_length, z_distance; +} stbi__zbuf; + +static INLINE uint8_t stbi__zget8(stbi__zbuf *z) +{ + if (z->zbuffer >= z->zbuffer_end) return 0; + return *z->zbuffer++; +} + +static void stbi__fill_bits(stbi__zbuf *z) +{ + do { + STBI_ASSERT(z->code_buffer < (1U << z->num_bits)); + z->code_buffer |= stbi__zget8(z) << z->num_bits; + z->num_bits += 8; + } while (z->num_bits <= 24); +} + +static INLINE unsigned int stbi__zreceive(stbi__zbuf *z, int n) +{ + unsigned int k; + if (z->num_bits < n) stbi__fill_bits(z); + k = z->code_buffer & ((1 << n) - 1); + z->code_buffer >>= n; + z->num_bits -= n; + return k; +} + +static int stbi__zhuffman_decode_slowpath(stbi__zbuf *a, stbi__zhuffman *z) +{ + int b,s,k; + k = stbi__bit_reverse(a->code_buffer, 16); + for (s=STBI__ZFAST_BITS+1; ; ++s) + if (k < z->maxcode[s]) + break; + if (s == 16) return -1; + b = (k >> (16-s)) - z->firstcode[s] + z->firstsymbol[s]; + STBI_ASSERT(z->size[b] == s); + a->code_buffer >>= s; + a->num_bits -= s; + return z->value[b]; +} + +static INLINE int stbi__zhuffman_decode(stbi__zbuf *a, stbi__zhuffman *z) +{ + int b,s; + if (a->num_bits < 16) stbi__fill_bits(a); + b = z->fast[a->code_buffer & STBI__ZFAST_MASK]; + if (b) { + s = b >> 9; + a->code_buffer >>= s; + a->num_bits -= s; + return b & 511; + } + return stbi__zhuffman_decode_slowpath(a, z); +} + +static int stbi__zexpand(stbi__zbuf *z, char *zout, int n) +{ + char *q; + int cur, limit; + z->zout = zout; + if (!z->z_expandable) return stbi__err("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 stbi__err("outofmem", "Out of memory"); + z->zout_start = q; + z->zout = q + cur; + z->zout_end = q + limit; + return 1; +} + +static int stbi__zlength_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 stbi__zlength_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 stbi__zdist_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 stbi__zdist_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 stbi__parse_huffman_block(stbi__zbuf *a) +{ + char *zout = a->zout; + for(;;) { + int z = stbi__zhuffman_decode(a, &a->z_length); + if (z < 256) { + if (z < 0) return stbi__err("bad huffman code","Corrupt PNG"); + if (zout >= a->zout_end) { + if (!stbi__zexpand(a, zout, 1)) return 0; + zout = a->zout; + } + *zout++ = (char) z; + } else { + uint8_t *p; + int len,dist; + if (z == 256) { + a->zout = zout; + return 1; + } + z -= 257; + len = stbi__zlength_base[z]; + if (stbi__zlength_extra[z]) len += stbi__zreceive(a, stbi__zlength_extra[z]); + z = stbi__zhuffman_decode(a, &a->z_distance); + if (z < 0) return stbi__err("bad huffman code","Corrupt PNG"); + dist = stbi__zdist_base[z]; + if (stbi__zdist_extra[z]) dist += stbi__zreceive(a, stbi__zdist_extra[z]); + if (zout - a->zout_start < dist) return stbi__err("bad dist","Corrupt PNG"); + if (zout + len > a->zout_end) { + if (!stbi__zexpand(a, zout, len)) return 0; + zout = a->zout; + } + p = (uint8_t *) (zout - dist); + if (dist == 1) + { + uint8_t v = *p; + if (len) { do *zout++ = v; while (--len); } + } else { + if (len) { do *zout++ = *p++; while (--len); } + } + } + } +} + +static int stbi__compute_huffman_codes(stbi__zbuf *a) +{ + static uint8_t length_dezigzag[19] = { 16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15 }; + stbi__zhuffman z_codelength; + uint8_t lencodes[286+32+137]; + uint8_t codelength_sizes[19]; + int i,n; + + int hlit = stbi__zreceive(a,5) + 257; + int hdist = stbi__zreceive(a,5) + 1; + int hclen = stbi__zreceive(a,4) + 4; + + memset(codelength_sizes, 0, sizeof(codelength_sizes)); + for (i=0; i < hclen; ++i) { + int s = stbi__zreceive(a,3); + codelength_sizes[length_dezigzag[i]] = (uint8_t) s; + } + if (!stbi__zbuild_huffman(&z_codelength, codelength_sizes, 19)) return 0; + + n = 0; + while (n < hlit + hdist) { + int c = stbi__zhuffman_decode(a, &z_codelength); + if (c < 0 || c >= 19) return stbi__err("bad codelengths", "Corrupt PNG"); + if (c < 16) + lencodes[n++] = (uint8_t) c; + else if (c == 16) { + c = stbi__zreceive(a,2)+3; + memset(lencodes+n, lencodes[n-1], c); + n += c; + } else if (c == 17) { + c = stbi__zreceive(a,3)+3; + memset(lencodes+n, 0, c); + n += c; + } else { + STBI_ASSERT(c == 18); + c = stbi__zreceive(a,7)+11; + memset(lencodes+n, 0, c); + n += c; + } + } + if (n != hlit+hdist) return stbi__err("bad codelengths","Corrupt PNG"); + if (!stbi__zbuild_huffman(&a->z_length, lencodes, hlit)) return 0; + if (!stbi__zbuild_huffman(&a->z_distance, lencodes+hlit, hdist)) return 0; + return 1; +} + +static int stbi__parse_uncomperssed_block(stbi__zbuf *a) +{ + uint8_t header[4]; + int len,nlen,k; + if (a->num_bits & 7) + stbi__zreceive(a, a->num_bits & 7); + k = 0; + while (a->num_bits > 0) { + header[k++] = (uint8_t) (a->code_buffer & 255); + a->code_buffer >>= 8; + a->num_bits -= 8; + } + STBI_ASSERT(a->num_bits == 0); + while (k < 4) + header[k++] = stbi__zget8(a); + len = header[1] * 256 + header[0]; + nlen = header[3] * 256 + header[2]; + if (nlen != (len ^ 0xffff)) return stbi__err("zlib corrupt","Corrupt PNG"); + if (a->zbuffer + len > a->zbuffer_end) return stbi__err("read past buffer","Corrupt PNG"); + if (a->zout + len > a->zout_end) + if (!stbi__zexpand(a, a->zout, len)) return 0; + memcpy(a->zout, a->zbuffer, len); + a->zbuffer += len; + a->zout += len; + return 1; +} + +static int stbi__parse_zlib_header(stbi__zbuf *a) +{ + int cmf = stbi__zget8(a); + int cm = cmf & 15; + /* int cinfo = cmf >> 4; */ + int flg = stbi__zget8(a); + if ((cmf*256+flg) % 31 != 0) return stbi__err("bad zlib header","Corrupt PNG"); + if (flg & 32) return stbi__err("no preset dict","Corrupt PNG"); + if (cm != 8) return stbi__err("bad compression","Corrupt PNG"); + return 1; +} + +static uint8_t stbi__zdefault_length[288], stbi__zdefault_distance[32]; +static void stbi__init_zdefaults(void) +{ + int i; + for (i=0; i <= 143; ++i) stbi__zdefault_length[i] = 8; + for ( ; i <= 255; ++i) stbi__zdefault_length[i] = 9; + for ( ; i <= 279; ++i) stbi__zdefault_length[i] = 7; + for ( ; i <= 287; ++i) stbi__zdefault_length[i] = 8; + + for (i=0; i <= 31; ++i) stbi__zdefault_distance[i] = 5; +} + +static int stbi__parse_zlib(stbi__zbuf *a, int parse_header) +{ + int final, type; + if (parse_header) + if (!stbi__parse_zlib_header(a)) return 0; + a->num_bits = 0; + a->code_buffer = 0; + do { + final = stbi__zreceive(a,1); + type = stbi__zreceive(a,2); + if (type == 0) { + if (!stbi__parse_uncomperssed_block(a)) return 0; + } else if (type == 3) { + return 0; + } else { + if (type == 1) { + if (!stbi__zdefault_distance[31]) stbi__init_zdefaults(); + if (!stbi__zbuild_huffman(&a->z_length , stbi__zdefault_length , 288)) return 0; + if (!stbi__zbuild_huffman(&a->z_distance, stbi__zdefault_distance, 32)) return 0; + } else { + if (!stbi__compute_huffman_codes(a)) return 0; + } + if (!stbi__parse_huffman_block(a)) return 0; + } + } while (!final); + return 1; +} + +static int stbi__do_zlib(stbi__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 stbi__parse_zlib(a, parse_header); +} + +STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header) +{ + stbi__zbuf a; + char *p = (char *) malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (uint8_t *) buffer; + a.zbuffer_end = (uint8_t *) buffer + len; + if (stbi__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; + } +} +#endif + +#ifndef STBI_NO_PNG +typedef struct +{ + uint32_t length; + uint32_t type; +} stbi__pngchunk; + +static stbi__pngchunk stbi__get_chunk_header(stbi__context *s) +{ + stbi__pngchunk c; + c.length = stbi__get32be(s); + c.type = stbi__get32be(s); + return c; +} + +static int stbi__check_png_header(stbi__context *s) +{ + static uint8_t png_sig[8] = { 137,80,78,71,13,10,26,10 }; + int i; + for (i=0; i < 8; ++i) + if (stbi__get8(s) != png_sig[i]) return stbi__err("bad png sig","Not a PNG"); + return 1; +} + +typedef struct +{ + stbi__context *s; + uint8_t *idata, *expanded, *out; +} stbi__png; + + +enum { + STBI__F_none=0, + STBI__F_sub=1, + STBI__F_up=2, + STBI__F_avg=3, + STBI__F_paeth=4, + STBI__F_avg_first, + STBI__F_paeth_first +}; + +static uint8_t first_row_filter[5] = +{ + STBI__F_none, + STBI__F_sub, + STBI__F_none, + STBI__F_avg_first, + STBI__F_paeth_first +}; + +static int stbi__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; +} + +static uint8_t stbi__depth_scale_table[9] = { 0, 0xff, 0x55, 0, 0x11, 0,0,0, 0x01 }; + +static int stbi__create_png_image_raw(stbi__png *a, uint8_t *raw, uint32_t raw_len, int out_n, uint32_t x, uint32_t y, int depth, int color) +{ + int k; + stbi__context *s = a->s; + uint32_t i,j,stride = x*out_n; + uint32_t img_len, img_width_bytes; + int img_n = s->img_n; + + STBI_ASSERT(out_n == s->img_n || out_n == s->img_n+1); + a->out = (uint8_t *) malloc(x * y * out_n); + if (!a->out) return stbi__err("outofmem", "Out of memory"); + + img_width_bytes = (((img_n * x * depth) + 7) >> 3); + img_len = (img_width_bytes + 1) * y; + if (s->img_x == x && s->img_y == y) + { + if (raw_len != img_len) return stbi__err("not enough pixels","Corrupt PNG"); + } + else + { + if (raw_len < img_len) return stbi__err("not enough pixels","Corrupt PNG"); + } + + for (j=0; j < y; ++j) { + uint8_t *cur = a->out + stride*j; + uint8_t *prior = cur - stride; + int filter = *raw++; + int filter_bytes = img_n; + int width = x; + if (filter > 4) + return stbi__err("invalid filter","Corrupt PNG"); + + if (depth < 8) { + STBI_ASSERT(img_width_bytes <= x); + cur += x*out_n - img_width_bytes; + filter_bytes = 1; + width = img_width_bytes; + } + + if (j == 0) filter = first_row_filter[filter]; + + for (k=0; k < filter_bytes; ++k) + { + switch (filter) + { + case STBI__F_none : cur[k] = raw[k]; break; + case STBI__F_sub : cur[k] = raw[k]; break; + case STBI__F_up : cur[k] = STBI__BYTECAST(raw[k] + prior[k]); break; + case STBI__F_avg : cur[k] = STBI__BYTECAST(raw[k] + (prior[k]>>1)); break; + case STBI__F_paeth : cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(0,prior[k],0)); break; + case STBI__F_avg_first : cur[k] = raw[k]; break; + case STBI__F_paeth_first: cur[k] = raw[k]; break; + } + } + + if (depth == 8) { + if (img_n != out_n) + cur[img_n] = 255; + raw += img_n; + cur += out_n; + prior += out_n; + } else { + raw += 1; + cur += 1; + prior += 1; + } + + if (depth < 8 || img_n == out_n) + { + int nk = (width - 1)*img_n; + #define CASE(f) \ + case f: \ + for (k=0; k < nk; ++k) + switch (filter) { + case STBI__F_none: memcpy(cur, raw, nk); break; + CASE(STBI__F_sub) cur[k] = STBI__BYTECAST(raw[k] + cur[k-filter_bytes]); break; + CASE(STBI__F_up) cur[k] = STBI__BYTECAST(raw[k] + prior[k]); break; + CASE(STBI__F_avg) cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k-filter_bytes])>>1)); break; + CASE(STBI__F_paeth) cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-filter_bytes],prior[k],prior[k-filter_bytes])); break; + CASE(STBI__F_avg_first) cur[k] = STBI__BYTECAST(raw[k] + (cur[k-filter_bytes] >> 1)); break; + CASE(STBI__F_paeth_first) cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-filter_bytes],0,0)); break; + } + #undef CASE + raw += nk; + } else { + STBI_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(STBI__F_none) cur[k] = raw[k]; break; + CASE(STBI__F_sub) cur[k] = STBI__BYTECAST(raw[k] + cur[k-out_n]); break; + CASE(STBI__F_up) cur[k] = STBI__BYTECAST(raw[k] + prior[k]); break; + CASE(STBI__F_avg) cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k-out_n])>>1)); break; + CASE(STBI__F_paeth) cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-out_n],prior[k],prior[k-out_n])); break; + CASE(STBI__F_avg_first) cur[k] = STBI__BYTECAST(raw[k] + (cur[k-out_n] >> 1)); break; + CASE(STBI__F_paeth_first) cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-out_n],0,0)); break; + } + #undef CASE + } + } + + if (depth < 8) + { + for (j=0; j < y; ++j) + { + uint8_t *cur = a->out + stride*j; + uint8_t *in = a->out + stride*j + x*out_n - img_width_bytes; + uint8_t scale = (color == 0) ? stbi__depth_scale_table[depth] : 1; + + if (depth == 4) + { + for (k=x*img_n; k >= 2; k-=2, ++in) { + *cur++ = scale * ((*in >> 4) ); + *cur++ = scale * ((*in ) & 0x0f); + } + if (k > 0) *cur++ = scale * ((*in >> 4) ); + } else if (depth == 2) { + for (k=x*img_n; k >= 4; k-=4, ++in) { + *cur++ = scale * ((*in >> 6) ); + *cur++ = scale * ((*in >> 4) & 0x03); + *cur++ = scale * ((*in >> 2) & 0x03); + *cur++ = scale * ((*in ) & 0x03); + } + if (k > 0) *cur++ = scale * ((*in >> 6) ); + if (k > 1) *cur++ = scale * ((*in >> 4) & 0x03); + if (k > 2) *cur++ = scale * ((*in >> 2) & 0x03); + } else if (depth == 1) { + for (k=x*img_n; k >= 8; k-=8, ++in) { + *cur++ = scale * ((*in >> 7) ); + *cur++ = scale * ((*in >> 6) & 0x01); + *cur++ = scale * ((*in >> 5) & 0x01); + *cur++ = scale * ((*in >> 4) & 0x01); + *cur++ = scale * ((*in >> 3) & 0x01); + *cur++ = scale * ((*in >> 2) & 0x01); + *cur++ = scale * ((*in >> 1) & 0x01); + *cur++ = scale * ((*in ) & 0x01); + } + if (k > 0) *cur++ = scale * ((*in >> 7) ); + if (k > 1) *cur++ = scale * ((*in >> 6) & 0x01); + if (k > 2) *cur++ = scale * ((*in >> 5) & 0x01); + if (k > 3) *cur++ = scale * ((*in >> 4) & 0x01); + if (k > 4) *cur++ = scale * ((*in >> 3) & 0x01); + if (k > 5) *cur++ = scale * ((*in >> 2) & 0x01); + if (k > 6) *cur++ = scale * ((*in >> 1) & 0x01); + } + if (img_n != out_n) { + uint8_t *cur = a->out + stride*j; + int i; + if (img_n == 1) { + for (i=x-1; i >= 0; --i) { + cur[i*2+1] = 255; + cur[i*2+0] = cur[i]; + } + } else { + STBI_ASSERT(img_n == 3); + for (i=x-1; i >= 0; --i) { + cur[i*4+3] = 255; + cur[i*4+2] = cur[i*3+2]; + cur[i*4+1] = cur[i*3+1]; + cur[i*4+0] = cur[i*3+0]; + } + } + } + } + } + + return 1; +} + +static int stbi__create_png_image(stbi__png *a, uint8_t *image_data, uint32_t image_data_len, int out_n, int depth, int color, int interlaced) +{ + uint8_t *final; + int p; + if (!interlaced) + return stbi__create_png_image_raw(a, image_data, image_data_len, out_n, a->s->img_x, a->s->img_y, depth, color); + + final = (uint8_t *) 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; + 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) { + uint32_t img_len = ((((a->s->img_n * x * depth) + 7) >> 3) + 1) * y; + if (!stbi__create_png_image_raw(a, image_data, image_data_len, out_n, x, y, depth, color)) { + free(final); + return 0; + } + for (j=0; j < y; ++j) { + for (i=0; i < x; ++i) { + int out_y = j*yspc[p]+yorig[p]; + int out_x = i*xspc[p]+xorig[p]; + memcpy(final + out_y*a->s->img_x*out_n + out_x*out_n, + a->out + (j*x+i)*out_n, out_n); + } + } + free(a->out); + image_data += img_len; + image_data_len -= img_len; + } + } + a->out = final; + + return 1; +} + +static int stbi__compute_transparency(stbi__png *z, uint8_t tc[3], int out_n) +{ + stbi__context *s = z->s; + uint32_t i, pixel_count = s->img_x * s->img_y; + uint8_t *p = z->out; + + STBI_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 stbi__expand_png_palette(stbi__png *a, uint8_t *palette, int len, int pal_img_n) +{ + uint32_t i, pixel_count = a->s->img_x * a->s->img_y; + uint8_t *p, *temp_out, *orig = a->out; + + p = (uint8_t *) malloc(pixel_count * pal_img_n); + if (p == NULL) return stbi__err("outofmem", "Out of memory"); + + 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; + +static void stbi__de_iphone(stbi__png *z) +{ + stbi__context *s = z->s; + uint32_t i, pixel_count = s->img_x * s->img_y; + uint8_t *p = z->out; + + if (s->img_out_n == 3) + { + for (i=0; i < pixel_count; ++i) { + uint8_t t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 3; + } + } + else + { + STBI_ASSERT(s->img_out_n == 4); + if (stbi__unpremultiply_on_load) + { + for (i=0; i < pixel_count; ++i) { + uint8_t a = p[3]; + uint8_t 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 { + for (i=0; i < pixel_count; ++i) { + uint8_t t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 4; + } + } + } +} + +#define STBI__PNG_TYPE(a,b,c,d) (((a) << 24) + ((b) << 16) + ((c) << 8) + (d)) + +static int stbi__parse_png_file(stbi__png *z, int scan, int req_comp) +{ + uint8_t palette[1024], pal_img_n=0; + uint8_t has_trans=0, tc[3]; + uint32_t ioff=0, idata_limit=0, i, pal_len=0; + int first=1,k,interlace=0, color=0, depth=0, is_iphone=0; + stbi__context *s = z->s; + + z->expanded = NULL; + z->idata = NULL; + z->out = NULL; + + if (!stbi__check_png_header(s)) return 0; + + if (scan == STBI__SCAN_type) return 1; + + for (;;) { + stbi__pngchunk c = stbi__get_chunk_header(s); + switch (c.type) { + case STBI__PNG_TYPE('C','g','B','I'): + is_iphone = 1; + stbi__skip(s, c.length); + break; + case STBI__PNG_TYPE('I','H','D','R'): { + int comp,filter; + if (!first) return stbi__err("multiple IHDR","Corrupt PNG"); + first = 0; + if (c.length != 13) return stbi__err("bad IHDR len","Corrupt PNG"); + s->img_x = stbi__get32be(s); if (s->img_x > (1 << 24)) return stbi__err("too large","Very large image (corrupt?)"); + s->img_y = stbi__get32be(s); if (s->img_y > (1 << 24)) return stbi__err("too large","Very large image (corrupt?)"); + depth = stbi__get8(s); if (depth != 1 && depth != 2 && depth != 4 && depth != 8) return stbi__err("1/2/4/8-bit only","PNG not supported: 1/2/4/8-bit only"); + color = stbi__get8(s); if (color > 6) return stbi__err("bad ctype","Corrupt PNG"); + if (color == 3) pal_img_n = 3; else if (color & 1) return stbi__err("bad ctype","Corrupt PNG"); + comp = stbi__get8(s); if (comp) return stbi__err("bad comp method","Corrupt PNG"); + filter= stbi__get8(s); if (filter) return stbi__err("bad filter method","Corrupt PNG"); + interlace = stbi__get8(s); if (interlace>1) return stbi__err("bad interlace method","Corrupt PNG"); + if (!s->img_x || !s->img_y) return stbi__err("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 stbi__err("too large", "Image too large to decode"); + if (scan == STBI__SCAN_header) return 1; + } else { + s->img_n = 1; + if ((1 << 30) / s->img_x / 4 < s->img_y) return stbi__err("too large","Corrupt PNG"); + } + break; + } + + case STBI__PNG_TYPE('P','L','T','E'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (c.length > 256*3) return stbi__err("invalid PLTE","Corrupt PNG"); + pal_len = c.length / 3; + if (pal_len * 3 != c.length) return stbi__err("invalid PLTE","Corrupt PNG"); + for (i=0; i < pal_len; ++i) { + palette[i*4+0] = stbi__get8(s); + palette[i*4+1] = stbi__get8(s); + palette[i*4+2] = stbi__get8(s); + palette[i*4+3] = 255; + } + break; + } + + case STBI__PNG_TYPE('t','R','N','S'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (z->idata) return stbi__err("tRNS after IDAT","Corrupt PNG"); + if (pal_img_n) { + if (scan == STBI__SCAN_header) { s->img_n = 4; return 1; } + if (pal_len == 0) return stbi__err("tRNS before PLTE","Corrupt PNG"); + if (c.length > pal_len) return stbi__err("bad tRNS len","Corrupt PNG"); + pal_img_n = 4; + for (i=0; i < c.length; ++i) + palette[i*4+3] = stbi__get8(s); + } else { + if (!(s->img_n & 1)) return stbi__err("tRNS with alpha","Corrupt PNG"); + if (c.length != (uint32_t) s->img_n*2) return stbi__err("bad tRNS len","Corrupt PNG"); + has_trans = 1; + for (k=0; k < s->img_n; ++k) + tc[k] = (uint8_t) (stbi__get16be(s) & 255) * stbi__depth_scale_table[depth]; + } + break; + } + + case STBI__PNG_TYPE('I','D','A','T'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (pal_img_n && !pal_len) return stbi__err("no PLTE","Corrupt PNG"); + if (scan == STBI__SCAN_header) { s->img_n = pal_img_n; return 1; } + if ((int)(ioff + c.length) < (int)ioff) return 0; + if (ioff + c.length > idata_limit) { + uint8_t *p; + if (idata_limit == 0) idata_limit = c.length > 4096 ? c.length : 4096; + while (ioff + c.length > idata_limit) + idata_limit *= 2; + p = (uint8_t *) realloc(z->idata, idata_limit); if (p == NULL) return stbi__err("outofmem", "Out of memory"); + z->idata = p; + } + if (!stbi__getn(s, z->idata+ioff,c.length)) return stbi__err("outofdata","Corrupt PNG"); + ioff += c.length; + break; + } + + case STBI__PNG_TYPE('I','E','N','D'): { + uint32_t raw_len, bpl; + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (scan != STBI__SCAN_load) return 1; + if (z->idata == NULL) return stbi__err("no IDAT","Corrupt PNG"); + bpl = (s->img_x * depth + 7) / 8; + raw_len = bpl * s->img_y * s->img_n /* pixels */ + s->img_y /* filter mode per row */; + z->expanded = (uint8_t *) stbi_zlib_decode_malloc_guesssize_headerflag((char *) z->idata, ioff, raw_len, (int *) &raw_len, !is_iphone); + if (z->expanded == NULL) return 0; + 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 (!stbi__create_png_image(z, z->expanded, raw_len, s->img_out_n, depth, color, interlace)) return 0; + if (has_trans) + if (!stbi__compute_transparency(z, tc, s->img_out_n)) return 0; + if (is_iphone && stbi__de_iphone_flag && s->img_out_n > 2) + stbi__de_iphone(z); + if (pal_img_n) { + s->img_n = pal_img_n; + s->img_out_n = pal_img_n; + if (req_comp >= 3) s->img_out_n = req_comp; + if (!stbi__expand_png_palette(z, palette, pal_len, s->img_out_n)) + return 0; + } + free(z->expanded); z->expanded = NULL; + return 1; + } + + default: + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if ((c.type & (1 << 29)) == 0) { + #ifndef STBI_NO_FAILURE_STRINGS + static char invalid_chunk[] = "XXXX PNG chunk not known"; + invalid_chunk[0] = STBI__BYTECAST(c.type >> 24); + invalid_chunk[1] = STBI__BYTECAST(c.type >> 16); + invalid_chunk[2] = STBI__BYTECAST(c.type >> 8); + invalid_chunk[3] = STBI__BYTECAST(c.type >> 0); + #endif + return stbi__err(invalid_chunk, "PNG not supported: unknown PNG chunk type"); + } + stbi__skip(s, c.length); + break; + } + stbi__get32be(s); + } +} + +static unsigned char *stbi__do_png(stbi__png *p, int *x, int *y, int *n, int req_comp) +{ + unsigned char *result=NULL; + if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); + if (stbi__parse_png_file(p, STBI__SCAN_load, req_comp)) { + result = p->out; + p->out = NULL; + if (req_comp && req_comp != p->s->img_out_n) { + result = stbi__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_out_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__context *s, int *x, int *y, int *comp, int req_comp) +{ + stbi__png p; + p.s = s; + return stbi__do_png(&p, x,y,comp,req_comp); +} + +static int stbi__png_test(stbi__context *s) +{ + int r; + r = stbi__check_png_header(s); + stbi__rewind(s); + return r; +} +#endif + +#ifndef STBI_NO_BMP +static int stbi__bmp_test_raw(stbi__context *s) +{ + int r; + int sz; + if (stbi__get8(s) != 'B') return 0; + if (stbi__get8(s) != 'M') return 0; + stbi__get32le(s); + stbi__get16le(s); + stbi__get16le(s); + stbi__get32le(s); + sz = stbi__get32le(s); + r = (sz == 12 || sz == 40 || sz == 56 || sz == 108 || sz == 124); + return r; +} + +static int stbi__bmp_test(stbi__context *s) +{ + int r = stbi__bmp_test_raw(s); + stbi__rewind(s); + return r; +} + + +static int stbi__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 stbi__bitcount(unsigned int a) +{ + a = (a & 0x55555555) + ((a >> 1) & 0x55555555); + a = (a & 0x33333333) + ((a >> 2) & 0x33333333); + a = (a + (a >> 4)) & 0x0f0f0f0f; + a = (a + (a >> 8)); + a = (a + (a >> 16)); + return a & 0xff; +} + +static int stbi__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 uint8_t *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + uint8_t *out; + unsigned int mr=0,mg=0,mb=0,ma=0, fake_a=0; + uint8_t pal[256][4]; + int psize=0,i,j,compress=0,width; + int bpp, flip_vertically, pad, target, offset, hsz; + if (stbi__get8(s) != 'B' || stbi__get8(s) != 'M') return stbi__errpuc("not BMP", "Corrupt BMP"); + stbi__get32le(s); + stbi__get16le(s); + stbi__get16le(s); + offset = stbi__get32le(s); + hsz = stbi__get32le(s); + if (hsz != 12 && hsz != 40 && hsz != 56 && hsz != 108 && hsz != 124) return stbi__errpuc("unknown BMP", "BMP type not supported: unknown"); + if (hsz == 12) { + s->img_x = stbi__get16le(s); + s->img_y = stbi__get16le(s); + } else { + s->img_x = stbi__get32le(s); + s->img_y = stbi__get32le(s); + } + if (stbi__get16le(s) != 1) return stbi__errpuc("bad BMP", "bad BMP"); + bpp = stbi__get16le(s); + if (bpp == 1) return stbi__errpuc("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 = stbi__get32le(s); + if (compress == 1 || compress == 2) return stbi__errpuc("BMP RLE", "BMP type not supported: RLE"); + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + if (hsz == 40 || hsz == 56) { + if (hsz == 56) { + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + stbi__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; + STBI_NOTUSED(fake_a); + } else { + mr = 31u << 10; + mg = 31u << 5; + mb = 31u << 0; + } + } else if (compress == 3) { + mr = stbi__get32le(s); + mg = stbi__get32le(s); + mb = stbi__get32le(s); + if (mr == mg && mg == mb) { + return stbi__errpuc("bad BMP", "bad BMP"); + } + } else + return stbi__errpuc("bad BMP", "bad BMP"); + } + } else { + STBI_ASSERT(hsz == 108 || hsz == 124); + mr = stbi__get32le(s); + mg = stbi__get32le(s); + mb = stbi__get32le(s); + ma = stbi__get32le(s); + stbi__get32le(s); + for (i=0; i < 12; ++i) + stbi__get32le(s); + if (hsz == 124) { + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + stbi__get32le(s); + } + } + if (bpp < 16) + psize = (offset - 14 - hsz) >> 2; + } + s->img_n = ma ? 4 : 3; + if (req_comp && req_comp >= 3) + target = req_comp; + else + target = s->img_n; + out = (uint8_t *) malloc(target * s->img_x * s->img_y); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + if (bpp < 16) { + int z=0; + if (psize == 0 || psize > 256) { free(out); return stbi__errpuc("invalid", "Corrupt BMP"); } + for (i=0; i < psize; ++i) { + pal[i][2] = stbi__get8(s); + pal[i][1] = stbi__get8(s); + pal[i][0] = stbi__get8(s); + if (hsz != 12) stbi__get8(s); + pal[i][3] = 255; + } + stbi__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 stbi__errpuc("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=stbi__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) ? stbi__get8(s) : v2; + out[z++] = pal[v][0]; + out[z++] = pal[v][1]; + out[z++] = pal[v][2]; + if (target == 4) out[z++] = 255; + } + stbi__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; + stbi__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 stbi__errpuc("bad masks", "Corrupt BMP"); } + rshift = stbi__high_bit(mr)-7; rcount = stbi__bitcount(mr); + gshift = stbi__high_bit(mg)-7; gcount = stbi__bitcount(mg); + bshift = stbi__high_bit(mb)-7; bcount = stbi__bitcount(mb); + ashift = stbi__high_bit(ma)-7; acount = stbi__bitcount(ma); + } + for (j=0; j < (int) s->img_y; ++j) { + if (easy) { + for (i=0; i < (int) s->img_x; ++i) { + unsigned char a; + out[z+2] = stbi__get8(s); + out[z+1] = stbi__get8(s); + out[z+0] = stbi__get8(s); + z += 3; + a = (easy == 2 ? stbi__get8(s) : 255); + if (target == 4) out[z++] = a; + } + } else { + for (i=0; i < (int) s->img_x; ++i) { + uint32_t v = (bpp == 16 ? (uint32_t) stbi__get16le(s) : stbi__get32le(s)); + int a; + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mr, rshift, rcount)); + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mg, gshift, gcount)); + out[z++] = STBI__BYTECAST(stbi__shiftsigned(v & mb, bshift, bcount)); + a = (ma ? stbi__shiftsigned(v & ma, ashift, acount) : 255); + if (target == 4) out[z++] = STBI__BYTECAST(a); + } + } + stbi__skip(s, pad); + } + } + if (flip_vertically) { + uint8_t t; + for (j=0; j < (int) s->img_y>>1; ++j) { + uint8_t *p1 = out + j *s->img_x*target; + uint8_t *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 = stbi__convert_format(out, target, req_comp, s->img_x, s->img_y); + if (out == NULL) return out; + } + + *x = s->img_x; + *y = s->img_y; + if (comp) *comp = s->img_n; + return out; +} +#endif + +#ifndef STBI_NO_TGA +static int stbi__tga_test(stbi__context *s) +{ + int res; + int sz; + stbi__get8(s); + sz = stbi__get8(s); + if ( sz > 1 ) return 0; + sz = stbi__get8(s); + if ( (sz != 1) && (sz != 2) && (sz != 3) && (sz != 9) && (sz != 10) && (sz != 11) ) return 0; + stbi__get16be(s); + stbi__get16be(s); + stbi__get8(s); + stbi__get16be(s); + stbi__get16be(s); + if ( stbi__get16be(s) < 1 ) return 0; + if ( stbi__get16be(s) < 1 ) return 0; + sz = stbi__get8(s); + if ( (sz != 8) && (sz != 16) && (sz != 24) && (sz != 32) ) + res = 0; + else + res = 1; + stbi__rewind(s); + return res; +} + +static uint8_t *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + unsigned char *tga_data; + int i, j; + int tga_offset = stbi__get8(s); + int tga_indexed = stbi__get8(s); + int tga_image_type = stbi__get8(s); + int tga_is_RLE = 0; + int tga_palette_start = stbi__get16le(s); + int tga_palette_len = stbi__get16le(s); + int tga_palette_bits = stbi__get8(s); + int tga_x_origin = stbi__get16le(s); + int tga_y_origin = stbi__get16le(s); + int tga_width = stbi__get16le(s); + int tga_height = stbi__get16le(s); + int tga_bits_per_pixel = stbi__get8(s); + int tga_comp = tga_bits_per_pixel / 8; + int tga_inverted = stbi__get8(s); + unsigned char *tga_palette = NULL; + unsigned char raw_data[4] = {0}; + int RLE_count = 0; + int RLE_repeating = 0; + int read_next_pixel = 1; + + if ( tga_image_type >= 8 ) + { + tga_image_type -= 8; + tga_is_RLE = 1; + } + + tga_inverted = 1 - ((tga_inverted >> 5) & 1); + + if ( (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; + + if ( tga_indexed ) + tga_comp = tga_palette_bits / 8; + + *x = tga_width; + *y = tga_height; + if (comp) *comp = tga_comp; + + tga_data = (unsigned char*)malloc( (size_t)tga_width * tga_height * tga_comp ); + if (!tga_data) return stbi__errpuc("outofmem", "Out of memory"); + + stbi__skip(s, tga_offset ); + + if ( !tga_indexed && !tga_is_RLE) { + for (i=0; i < tga_height; ++i) { + int y = tga_inverted ? tga_height -i - 1 : i; + uint8_t *tga_row = tga_data + y*tga_width*tga_comp; + stbi__getn(s, tga_row, tga_width * tga_comp); + } + } + else + { + if ( tga_indexed) + { + stbi__skip(s, tga_palette_start ); + tga_palette = (unsigned char*)malloc( tga_palette_len * tga_palette_bits / 8 ); + if (!tga_palette) { + free(tga_data); + return stbi__errpuc("outofmem", "Out of memory"); + } + if (!stbi__getn(s, tga_palette, tga_palette_len * tga_palette_bits / 8 )) { + free(tga_data); + free(tga_palette); + return stbi__errpuc("bad palette", "Corrupt TGA"); + } + } + for (i=0; i < tga_width * tga_height; ++i) + { + if ( tga_is_RLE ) + { + if ( RLE_count == 0 ) + { + int RLE_cmd = stbi__get8(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; + } + if ( read_next_pixel ) + { + if ( tga_indexed ) + { + int pal_idx = stbi__get8(s); + if ( pal_idx >= tga_palette_len ) + { + 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 + { + for (j = 0; j*8 < tga_bits_per_pixel; ++j) + raw_data[j] = stbi__get8(s); + } + read_next_pixel = 0; + } + + for (j = 0; j < tga_comp; ++j) + tga_data[i*tga_comp+j] = raw_data[j]; + + --RLE_count; + } + if ( tga_inverted ) + { + for (j = 0; j*2 < tga_height; ++j) + { + int index1 = j * tga_width * tga_comp; + int index2 = (tga_height - 1 - j) * tga_width * tga_comp; + for (i = tga_width * tga_comp; i > 0; --i) + { + unsigned char temp = tga_data[index1]; + tga_data[index1] = tga_data[index2]; + tga_data[index2] = temp; + ++index1; + ++index2; + } + } + } + if ( tga_palette != NULL ) + { + free( tga_palette ); + } + } + + if (tga_comp >= 3) + { + unsigned char* tga_pixel = tga_data; + for (i=0; i < tga_width * tga_height; ++i) + { + unsigned char temp = tga_pixel[0]; + tga_pixel[0] = tga_pixel[2]; + tga_pixel[2] = temp; + tga_pixel += tga_comp; + } + } + + if (req_comp && req_comp != tga_comp) + tga_data = stbi__convert_format(tga_data, tga_comp, req_comp, tga_width, tga_height); + + tga_palette_start = tga_palette_len = tga_palette_bits = + tga_x_origin = tga_y_origin = 0; + return tga_data; +} +#endif + +#endif