commit 05f68bc40969656b193b14ae968f5280f10b8da8 Author: gered Date: Sun Jan 4 18:43:57 2015 -0500 initial commit (just stuffing all my test project files into a git repo) diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..e1b3e04 --- /dev/null +++ b/.gitignore @@ -0,0 +1,9 @@ +.DS_Store +/Makefile +CMakeFiles +CMakeCache.txt +cmake_install.cmake +.idea +/*.log +/SoftwareRasterizer +/*.exe diff --git a/CMakeLists.txt b/CMakeLists.txt new file mode 100644 index 0000000..393e707 --- /dev/null +++ b/CMakeLists.txt @@ -0,0 +1,21 @@ +cmake_minimum_required(VERSION 2.8.4) + +set(PROJECT_NAME SoftwareRasterizer) +set(PROJECT_SRC_DIR ${CMAKE_CURRENT_SOURCE_DIR}) + +project(SoftwareRasterizer) + +set(CMAKE_MODULE_PATH ${CMAKE_MODULE_PATH} "${PROJECT_SRC_DIR}/cmake") +find_package(SDL2 REQUIRED) + +file(GLOB_RECURSE SOURCE_FILES "." + "${PROJECT_SRC_DIR}/src/*.cpp" + "${PROJECT_SRC_DIR}/src/*.c" + "${PROJECT_SRC_DIR}/src/*.h" + "${PROJECT_SRC_DIR}/include/*.h" + ) +include_directories(${SDL2_INCLUDE_DIR} "${PROJECT_SRC_DIR}/include") +add_executable(${PROJECT_NAME} ${SOURCE_FILES}) + +target_link_libraries(${PROJECT_NAME} ${SDL2_LIBRARY}) + diff --git a/README.md b/README.md new file mode 100644 index 0000000..f445f7c --- /dev/null +++ b/README.md @@ -0,0 +1,22 @@ +# Software Rasterizer + +A 2D and 3D graphics software rasterizer implemented in C. Includes SDL backend to abstract OS-specifics (window management, input, etc). + +This is basically a "for fun" toy project of mine. Probably not too noteworthy and definitely has been done better before by people far smarter then me! + +**In early development! Pre-alpha quality code!** + +2D features: + +* 24 bit (RGB) and 32 bit (RGBA) surface support +* drawing primitives: pixel, line, rectangle (outline and filled) +* sprite drawing: solid, transparent, blending, color tinting +* clipping support on all drawing operations as well as "fast" variants which skip clipping +* image -> surface loading via stb_image +* TTF -> sprite font texture atlas via stb_truetype (TODO) + +3D features: + +* TODO + + diff --git a/cmake/FindSDL2.cmake b/cmake/FindSDL2.cmake new file mode 100644 index 0000000..f24af55 --- /dev/null +++ b/cmake/FindSDL2.cmake @@ -0,0 +1,249 @@ +# Locate SDL2 library +# This module defines +# SDL2_LIBRARY, the name of the library to link against +# SDL2_FOUND, if false, do not try to link to SDL2 +# SDL2_INCLUDE_DIR, where to find SDL.h +# +# This module responds to the the flag: +# SDL2_BUILDING_LIBRARY +# If this is defined, then no SDL2_main will be linked in because +# only applications need main(). +# Otherwise, it is assumed you are building an application and this +# module will attempt to locate and set the the proper link flags +# as part of the returned SDL2_LIBRARY variable. +# +# Don't forget to include SDL2main.h and SDL2main.m your project for the +# OS X framework based version. (Other versions link to -lSDL2main which +# this module will try to find on your behalf.) Also for OS X, this +# module will automatically add the -framework Cocoa on your behalf. +# +# +# Additional Note: If you see an empty SDL2_LIBRARY_TEMP in your configuration +# and no SDL2_LIBRARY, it means CMake did not find your SDL2 library +# (SDL2.dll, libsdl2.so, SDL2.framework, etc). +# Set SDL2_LIBRARY_TEMP to point to your SDL2 library, and configure again. +# Similarly, if you see an empty SDL2MAIN_LIBRARY, you should set this value +# as appropriate. These values are used to generate the final SDL2_LIBRARY +# variable, but when these values are unset, SDL2_LIBRARY does not get created. +# +# +# $SDL2 is an environment variable that would +# correspond to the ./configure --prefix=$SDL2 +# used in building SDL2. +# l.e.galup 9-20-02 +# +# Modified by Eric Wing. +# Added code to assist with automated building by using environmental variables +# and providing a more controlled/consistent search behavior. +# Added new modifications to recognize OS X frameworks and +# additional Unix paths (FreeBSD, etc). +# Also corrected the header search path to follow "proper" SDL2 guidelines. +# Added a search for SDL2main which is needed by some platforms. +# Added a search for threads which is needed by some platforms. +# Added needed compile switches for MinGW. +# +# On OSX, this will prefer the Framework version (if found) over others. +# People will have to manually change the cache values of +# SDL2_LIBRARY to override this selection or set the CMake environment +# CMAKE_INCLUDE_PATH to modify the search paths. +# +# Note that the header path has changed from SDL2/SDL.h to just SDL.h +# This needed to change because "proper" SDL2 convention +# is #include "SDL.h", not . This is done for portability +# reasons because not all systems place things in SDL2/ (see FreeBSD). +# +# Ported by Johnny Patterson. This is a literal port for SDL2 of the FindSDL.cmake +# module with the minor edit of changing "SDL" to "SDL2" where necessary. This +# was not created for redistribution, and exists temporarily pending official +# SDL2 CMake modules. +# +# Note that on windows this will only search for the 32bit libraries, to search +# for 64bit change x86/i686-w64 to x64/x86_64-w64 + +#============================================================================= +# Copyright 2003-2009 Kitware, Inc. +# +# CMake - Cross Platform Makefile Generator +# Copyright 2000-2014 Kitware, Inc. +# Copyright 2000-2011 Insight Software Consortium +# All rights reserved. +# +# Redistribution and use in source and binary forms, with or without +# modification, are permitted provided that the following conditions +# are met: +# +# * Redistributions of source code must retain the above copyright +# notice, this list of conditions and the following disclaimer. +# +# * Redistributions in binary form must reproduce the above copyright +# notice, this list of conditions and the following disclaimer in the +# documentation and/or other materials provided with the distribution. +# +# * Neither the names of Kitware, Inc., the Insight Software Consortium, +# nor the names of their contributors may be used to endorse or promote +# products derived from this software without specific prior written +# permission. +# +# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +# HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. +# +# This software is distributed WITHOUT ANY WARRANTY; without even the +# implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. +# See the License for more information. +#============================================================================= +# (To distribute this file outside of CMake, substitute the full +# License text for the above reference.) + +FIND_PATH(SDL2_INCLUDE_DIR SDL.h + HINTS + $ENV{SDL2} + PATH_SUFFIXES include/SDL2 include SDL2 + i686-w64-mingw32/include/SDL2 + x86_64-w64-mingw32/include/SDL2 + PATHS + ~/Library/Frameworks + /Library/Frameworks + /usr/local/include/SDL2 + /usr/include/SDL2 + /sw # Fink + /opt/local # DarwinPorts + /opt/csw # Blastwave + /opt +) + +# Lookup the 64 bit libs on x64 +IF(CMAKE_SIZEOF_VOID_P EQUAL 8) + FIND_LIBRARY(SDL2_LIBRARY_TEMP SDL2 + HINTS + $ENV{SDL2} + PATH_SUFFIXES lib64 lib + lib/x64 + x86_64-w64-mingw32/lib + PATHS + /sw + /opt/local + /opt/csw + /opt + ) +# On 32bit build find the 32bit libs +ELSE(CMAKE_SIZEOF_VOID_P EQUAL 8) + FIND_LIBRARY(SDL2_LIBRARY_TEMP SDL2 + HINTS + $ENV{SDL2} + PATH_SUFFIXES lib + lib/x86 + i686-w64-mingw32/lib + PATHS + /sw + /opt/local + /opt/csw + /opt + ) +ENDIF(CMAKE_SIZEOF_VOID_P EQUAL 8) + +IF(NOT SDL2_BUILDING_LIBRARY) + IF(NOT ${SDL2_INCLUDE_DIR} MATCHES ".framework") + # Non-OS X framework versions expect you to also dynamically link to + # SDL2main. This is mainly for Windows and OS X. Other (Unix) platforms + # seem to provide SDL2main for compatibility even though they don't + # necessarily need it. + # Lookup the 64 bit libs on x64 + IF(CMAKE_SIZEOF_VOID_P EQUAL 8) + FIND_LIBRARY(SDL2MAIN_LIBRARY + NAMES SDL2main + HINTS + $ENV{SDL2} + PATH_SUFFIXES lib64 lib + lib/x64 + x86_64-w64-mingw32/lib + PATHS + /sw + /opt/local + /opt/csw + /opt + ) + # On 32bit build find the 32bit libs + ELSE(CMAKE_SIZEOF_VOID_P EQUAL 8) + FIND_LIBRARY(SDL2MAIN_LIBRARY + NAMES SDL2main + HINTS + $ENV{SDL2} + PATH_SUFFIXES lib + lib/x86 + i686-w64-mingw32/lib + PATHS + /sw + /opt/local + /opt/csw + /opt + ) + ENDIF(CMAKE_SIZEOF_VOID_P EQUAL 8) + ENDIF(NOT ${SDL2_INCLUDE_DIR} MATCHES ".framework") +ENDIF(NOT SDL2_BUILDING_LIBRARY) + +# SDL2 may require threads on your system. +# The Apple build may not need an explicit flag because one of the +# frameworks may already provide it. +# But for non-OSX systems, I will use the CMake Threads package. +IF(NOT APPLE) + FIND_PACKAGE(Threads) +ENDIF(NOT APPLE) + +# MinGW needs an additional library, mwindows +# It's total link flags should look like -lmingw32 -lSDL2main -lSDL2 -lmwindows +# (Actually on second look, I think it only needs one of the m* libraries.) +IF(MINGW) + SET(MINGW32_LIBRARY mingw32 CACHE STRING "mwindows for MinGW") +ENDIF(MINGW) + +SET(SDL2_FOUND "NO") + IF(SDL2_LIBRARY_TEMP) + # For SDL2main + IF(NOT SDL2_BUILDING_LIBRARY) + IF(SDL2MAIN_LIBRARY) + SET(SDL2_LIBRARY_TEMP ${SDL2MAIN_LIBRARY} ${SDL2_LIBRARY_TEMP}) + ENDIF(SDL2MAIN_LIBRARY) + ENDIF(NOT SDL2_BUILDING_LIBRARY) + + # For OS X, SDL2 uses Cocoa as a backend so it must link to Cocoa. + # CMake doesn't display the -framework Cocoa string in the UI even + # though it actually is there if I modify a pre-used variable. + # I think it has something to do with the CACHE STRING. + # So I use a temporary variable until the end so I can set the + # "real" variable in one-shot. + IF(APPLE) + SET(SDL2_LIBRARY_TEMP ${SDL2_LIBRARY_TEMP} "-framework Cocoa") + ENDIF(APPLE) + + # For threads, as mentioned Apple doesn't need this. + # In fact, there seems to be a problem if I used the Threads package + # and try using this line, so I'm just skipping it entirely for OS X. + IF(NOT APPLE) + SET(SDL2_LIBRARY_TEMP ${SDL2_LIBRARY_TEMP} ${CMAKE_THREAD_LIBS_INIT}) + ENDIF(NOT APPLE) + + # For MinGW library + IF(MINGW) + SET(SDL2_LIBRARY_TEMP ${MINGW32_LIBRARY} ${SDL2_LIBRARY_TEMP}) + ENDIF(MINGW) + + # Set the final string here so the GUI reflects the final state. + SET(SDL2_LIBRARY ${SDL2_LIBRARY_TEMP} CACHE STRING "Where the SDL2 Library can be found") + # Set the temp variable to INTERNAL so it is not seen in the CMake GUI + SET(SDL2_LIBRARY_TEMP "${SDL2_LIBRARY_TEMP}" CACHE INTERNAL "") + + SET(SDL2_FOUND "YES") +ENDIF(SDL2_LIBRARY_TEMP) + +INCLUDE(FindPackageHandleStandardArgs) + +FIND_PACKAGE_HANDLE_STANDARD_ARGS(SDL2 REQUIRED_VARS SDL2_LIBRARY SDL2_INCLUDE_DIR) diff --git a/include/assets/image.h b/include/assets/image.h new file mode 100644 index 0000000..1a2c0e1 --- /dev/null +++ b/include/assets/image.h @@ -0,0 +1,15 @@ +#pragma once + +#ifdef __cplusplus +extern "C" { +#endif + +#include "../fbgfx/surface.h" +#include "../types.h" + +SURFACE* image_load_file(const char *filename); +SURFACE* image_load_memory(const void *image_file_bin, uint32_t size); + +#ifdef __cplusplus +} +#endif diff --git a/include/fbgfx/clipping.h b/include/fbgfx/clipping.h new file mode 100644 index 0000000..c86d72d --- /dev/null +++ b/include/fbgfx/clipping.h @@ -0,0 +1,17 @@ +#pragma once + +#ifdef __cplusplus +extern "C" { +#endif + +#include "../types.h" +#include "../math/rect.h" + +bool point_in_bounds(const RECT *clip_region, int x, int y); +bool line_in_bounds(const RECT *clip_region, int x1, int y1, int x2, int y2); +bool clamp_to_region(const RECT *clip_region, int *x1, int *y1, int *x2, int *y2); +bool clip_blit_region(const RECT *dest_clip_region, RECT *source_blit_region, int *dest_x, int *dest_y); + +#ifdef __cplusplus +} +#endif diff --git a/include/fbgfx/color.h b/include/fbgfx/color.h new file mode 100644 index 0000000..c512fec --- /dev/null +++ b/include/fbgfx/color.h @@ -0,0 +1,179 @@ +#pragma once + +#ifdef __cplusplus +extern "C" { +#endif + +#include "../types.h" + +typedef uint32_t COLOR; +typedef uint8_t COLOR_COMPONENT; + +static const COLOR_COMPONENT TRANSPARENT = 0; +static const COLOR_COMPONENT OPAQUE = 255; + +static inline COLOR_COMPONENT color_get_a(COLOR color) { + return (color & 0xff000000) >> 24; +} + +static inline COLOR_COMPONENT color_get_r(COLOR color) { + return (color & 0x00ff0000) >> 16; +} + +static inline COLOR_COMPONENT color_get_g(COLOR color) { + return (color & 0x0000ff00) >> 8; +} + +static inline COLOR_COMPONENT color_get_b(COLOR color) { + return (color & 0x000000ff); +} + +static inline void color_extract(COLOR color, COLOR_COMPONENT *r, COLOR_COMPONENT *g, COLOR_COMPONENT *b, COLOR_COMPONENT *a) { + if (a) + *a = color_get_a(color); + if (r) + *r = color_get_r(color); + if (g) + *g = color_get_g(color); + if (b) + *b = color_get_b(color); +} + +static inline COLOR color_create_rgba(COLOR_COMPONENT r, COLOR_COMPONENT g, COLOR_COMPONENT b, COLOR_COMPONENT a) { + return (a << 24) | (r << 16) | (g << 8) | b; +} + +static inline COLOR color_create_rgb(COLOR_COMPONENT r, COLOR_COMPONENT g, COLOR_COMPONENT b) { + return color_create_rgba(r, g, b, OPAQUE); +} + + +/* floating-point color components */ + +typedef float COLORF_COMPONENT; + +static const COLORF_COMPONENT TRANSPARENT_F = 0.0f; +static const COLORF_COMPONENT OPAQUE_F = 1.0f; + +static inline COLORF_COMPONENT colorf_get_a(COLOR color) { + return (float)((color & 0xff000000) >> 24) / 255; +} + +static inline COLORF_COMPONENT colorf_get_r(COLOR color) { + return (float)((color & 0x00ff0000) >> 16) / 255; +} + +static inline COLORF_COMPONENT colorf_get_g(COLOR color) { + return (float)((color & 0x0000ff00) >> 8) / 255; +} + +static inline COLORF_COMPONENT colorf_get_b(COLOR color) { + return (float)(color & 0x000000ff) / 255; +} + +static inline void colorf_extract(COLOR color, COLORF_COMPONENT *r, COLORF_COMPONENT *g, COLORF_COMPONENT *b, COLORF_COMPONENT *a) { + if (a) + *a = colorf_get_a(color); + if (r) + *r = colorf_get_r(color); + if (g) + *g = colorf_get_g(color); + if (b) + *b = colorf_get_b(color); +} + +static inline COLOR colorf_create_rgba(COLORF_COMPONENT r, COLORF_COMPONENT g, COLORF_COMPONENT b, COLORF_COMPONENT a) { + return ((COLOR_COMPONENT)(a * 255) << 24) | + ((COLOR_COMPONENT)(r * 255) << 16) | + ((COLOR_COMPONENT)(g * 255) << 8) | + (COLOR_COMPONENT)(b * 255); +} + +static inline COLOR colorf_create_rgb(COLORF_COMPONENT r, COLORF_COMPONENT g, COLORF_COMPONENT b, COLORF_COMPONENT a) { + return colorf_create_rgba(r, g, b, OPAQUE_F); +} + +static inline COLOR color_blend(COLOR source, COLOR destination) { + int sR = color_get_r(source); + int sG = color_get_g(source); + int sB = color_get_b(source); + int sA = color_get_a(source); + int dR = color_get_r(destination); + int dG = color_get_g(destination); + int dB = color_get_b(destination); + int dA = color_get_a(destination); + + // this is equivalent to "glBlendFunc(GL_SRC_ALPHA, GL_ONE_MINUS_SRC_ALPHA)" + return color_create_rgba( + (COLOR_COMPONENT)(((sR * sA) + (dR * (255 - sA))) / 255), + (COLOR_COMPONENT)(((sG * sA) + (dG * (255 - sA))) / 255), + (COLOR_COMPONENT)(((sB * sA) + (dB * (255 - sA))) / 255), + (COLOR_COMPONENT)(((sA * sA) + (dA * (255 - sA))) / 255) + ); +} + +static inline COLOR color_tint(COLOR color, COLOR tint) { + int cR = color_get_r(color); + int cG = color_get_g(color); + int cB = color_get_b(color); + int cA = color_get_a(color); + int tR = color_get_r(tint); + int tG = color_get_g(tint); + int tB = color_get_b(tint); + int tA = color_get_a(tint); + + return color_create_rgba( + (COLOR_COMPONENT)((cR * tR) / 255), + (COLOR_COMPONENT)((cG * tG) / 255), + (COLOR_COMPONENT)((cB * tB) / 255), + (COLOR_COMPONENT)((cA * tA) / 255) + ); +} + +static inline COLOR color_tint_alpha(COLOR_COMPONENT alpha, COLOR tint) { + return color_create_rgba( + color_get_r(tint), + color_get_g(tint), + color_get_b(tint), + (COLOR_COMPONENT)(((uint32_t)alpha * color_get_a(tint)) / 255) + ); +} + +static inline void color_convert_rgba_to_rgb(COLOR rgba, COLOR_COMPONENT *r, COLOR_COMPONENT *g, COLOR_COMPONENT *b) { + COLOR_COMPONENT a = color_get_a(rgba); + *r = (COLOR_COMPONENT)(((uint32_t)color_get_r(rgba) * a) / 255); + *g = (COLOR_COMPONENT)(((uint32_t)color_get_g(rgba) * a) / 255); + *b = (COLOR_COMPONENT)(((uint32_t)color_get_b(rgba) * a) / 255); +} + +/* -- */ + +#define COLOR_WHITE (color_create_rgb(255, 255, 255)) +#define COLOR_RED (color_create_rgb(255, 0, 0)) +#define COLOR_GREEN (color_create_rgb(0, 255, 0)) +#define COLOR_BLUE (color_create_rgb(0, 0, 255)) +#define COLOR_YELLOW (color_create_rgb(255, 255, 0)) +#define COLOR_CYAN (color_create_rgb(0, 255, 255)) +#define COLOR_MAGENTA (color_create_rgb(255, 0, 255)) +#define COLOR_BLACK (color_create_rgb(0, 0, 0)) + +#define COLOR_DOS_BLACK (color_create_rgb(0, 0, 0)) +#define COLOR_DOS_BLUE (color_create_rgb(0, 0, 170)) +#define COLOR_DOS_GREEN (color_create_rgb(0, 170, 0)) +#define COLOR_DOS_CYAN (color_create_rgb(0, 170, 170)) +#define COLOR_DOS_RED (color_create_rgb(170, 0, 0)) +#define COLOR_DOS_MAGENTA (color_create_rgb(170, 0, 170)) +#define COLOR_DOS_BROWN (color_create_rgb(170, 85, 0)) +#define COLOR_DOS_WHITE (color_create_rgb(170, 170, 170)) +#define COLOR_DOS_GRAY (color_create_rgb(85, 85, 85)) +#define COLOR_DOS_LIGHT_BLUE (color_create_rgb(85, 85, 255)) +#define COLOR_DOS_LIGHT_GREEN (color_create_rgb(85, 255, 64)) +#define COLOR_DOS_LIGHT_CYAN (color_create_rgb(85, 255, 255)) +#define COLOR_DOS_LIGHT_RED (color_create_rgb(255, 85, 85)) +#define COLOR_DOS_LIGHT_MAGENTA (color_create_rgb(255, 85, 255)) +#define COLOR_DOS_YELLOW (color_create_rgb(255, 255, 85)) +#define COLOR_DOS_BRIGHT_WHITE (color_create_rgb(255, 255, 255)) + +#ifdef __cplusplus +} +#endif \ No newline at end of file diff --git a/include/fbgfx/surface.h b/include/fbgfx/surface.h new file mode 100644 index 0000000..f9c77c9 --- /dev/null +++ b/include/fbgfx/surface.h @@ -0,0 +1,188 @@ +#pragma once + +#ifdef __cplusplus +extern "C" { +#endif + +#include "../types.h" +#include "../math/rect.h" +#include "color.h" + +typedef enum { + SURFACE_FORMAT_ALPHA = 8, + SURFACE_FORMAT_RGB = 24, + SURFACE_FORMAT_RGBA = 32 +} SURFACE_FORMAT; + +typedef enum { + SURFACE_FLAGS_NONE = 0, + SURFACE_FLAGS_SIDEWAYS_BUFFER = 1 +} SURFACE_FLAGS; + +typedef struct SURFACE { + SURFACE_FLAGS flags; + COLOR_COMPONENT *pixels; + bool wrapped_buffer; + SURFACE_FORMAT format; + int bytes_per_pixel; + int width; + int height; + int x_inc; + int y_inc; + RECT clip_region; +} SURFACE; + +extern SURFACE *top_screen_surface; +extern SURFACE *bottom_screen_surface; + +void surface_init(); +void surface_exit(); + +void surface_swap_framebuffers(); + +SURFACE* surface_create(int width, int height, SURFACE_FORMAT format, SURFACE_FLAGS flags); +SURFACE* surface_wrap_existing(void *framebuffer, int width, int height, SURFACE_FORMAT format, SURFACE_FLAGS flags); +void surface_resize(SURFACE *surface, int width, int height); +void surface_destroy(SURFACE *surface); + +void surface_convert_rgb_to_bgr(SURFACE *surface); + +void surface_clear(SURFACE *surface, COLOR color); + +void surface_set_pixel(SURFACE *surface, int x, int y, COLOR color); +void surface_set_pixel_fast(SURFACE *surface, int x, int y, COLOR color); +COLOR surface_get_pixel(const SURFACE *surface, int x, int y); +COLOR surface_get_pixel_fast(SURFACE *surface, int x, int y); + +void surface_line(SURFACE *surface, int x1, int y1, int x2, int y2, COLOR color); +void surface_line_fast(SURFACE *surface, int x1, int y1, int x2, int y2, COLOR color); +void surface_hline(SURFACE *surface, int x1, int x2, int y, COLOR color); +void surface_hline_fast(SURFACE *surface, int x1, int x2, int y, COLOR color); +void surface_vline(SURFACE *surface, int x, int y1, int y2, COLOR color); +void surface_vline_fast(SURFACE *surface, int x, int y1, int y2, COLOR color); +void surface_rect(SURFACE *surface, int x1, int y1, int x2, int y2, COLOR color); +void surface_rect_fast(SURFACE *surface, int x1, int y1, int x2, int y2, COLOR color); +void surface_rect_filled(SURFACE *surface, int x1, int y1, int x2, int y2, COLOR color); +void surface_rect_filled_fast(SURFACE *surface, int x1, int y1, int x2, int y2, COLOR color); + +void surface_blit_region(SURFACE *source, SURFACE *destination, int source_x, int source_y, int source_width, int source_height, int dest_x, int dest_y); +void surface_blit_blend_region(SURFACE *source, SURFACE *destination, int source_x, int source_y, int source_width, int source_height, int dest_x, int dest_y); +void surface_blit_trans_region(SURFACE *source, SURFACE *destination, int source_x, int source_y, int source_width, int source_height, int dest_x, int dest_y); +void surface_blit_tint_region(SURFACE *source, SURFACE *destination, int source_x, int source_y, int source_width, int source_height, int dest_x, int dest_y, COLOR tint); +void surface_blit_blend_tint_region(SURFACE *source, SURFACE *destination, int source_x, int source_y, int source_width, int source_height, int dest_x, int dest_y, COLOR tint); + +#define GET_PIXEL_FAST(fmt, ...) surface_get_pixel_ ## fmt ## _fast(__VA_ARGS__) +#define SET_PIXEL_FAST(fmt, ...) surface_set_pixel_ ## fmt ## _fast(__VA_ARGS__) +#define GET_PIXEL_IDX_FAST(fmt, ...) surface_get_pixel_idx_ ## fmt ## _fast(__VA_ARGS__) +#define SET_PIXEL_IDX_FAST(fmt, ...) surface_set_pixel_idx_ ## fmt ## _fast(__VA_ARGS__) + +//////////////// + +static inline uint32_t surface_get_buffer_size(const SURFACE *surface) { + return (uint32_t)(surface->width * surface->height) * surface->bytes_per_pixel; +} + +static inline uint32_t surface_get_index_of(const SURFACE *surface, int x, int y) { + if (surface->flags & SURFACE_FLAGS_SIDEWAYS_BUFFER) + return (uint32_t)((x * surface->height) + (surface->height - y - 1)) * surface->bytes_per_pixel; + else + return (uint32_t)(x + (y * surface->width)) * surface->bytes_per_pixel; +} + +static inline COLOR_COMPONENT* surface_get_pointer_to(const SURFACE *surface, int x, int y) { + return surface->pixels + surface_get_index_of(surface, x, y); +} + +//////////////// + +static inline void surface_set_pixel_idx_alpha_fast(SURFACE *surface, uint32_t index, COLOR color) { + COLOR_COMPONENT alpha = color_get_a(color); + COLOR_COMPONENT *p = (surface->pixels + index); + *p = alpha; +} + +static inline void surface_set_pixel_idx_rgb_fast(SURFACE *surface, uint32_t index, COLOR color) { + COLOR_COMPONENT *p = (surface->pixels + index); + color_convert_rgba_to_rgb(color, (p + 2), (p + 1), p); +} + +static inline void surface_set_pixel_idx_rgba_fast(SURFACE *surface, uint32_t index, COLOR color) { + COLOR_COMPONENT *p = (surface->pixels + index); + color_extract(color, (p + 2), (p + 1), p, (p + 3)); +} + +static inline COLOR surface_get_pixel_idx_alpha_fast(const SURFACE *surface, uint32_t index) { + COLOR_COMPONENT *p = (surface->pixels + index); + return color_create_rgba(0, 0, 0, *p); +} + +static inline COLOR surface_get_pixel_idx_rgb_fast(const SURFACE *surface, uint32_t index) { + COLOR_COMPONENT *p = (surface->pixels + index); + COLOR_COMPONENT r = p[2]; + COLOR_COMPONENT g = p[1]; + COLOR_COMPONENT b = p[0]; + return color_create_rgb(r, g, b); +} + +static inline COLOR surface_get_pixel_idx_rgba_fast(const SURFACE *surface, uint32_t index) { + COLOR_COMPONENT *p = (surface->pixels + index); + COLOR_COMPONENT a = p[3]; + COLOR_COMPONENT r = p[2]; + COLOR_COMPONENT g = p[1]; + COLOR_COMPONENT b = p[0]; + return color_create_rgba(r, g, b, a); +} + +static inline void surface_set_pixel_alpha_fast(SURFACE *surface, int x, int y, COLOR color) { + uint32_t index = surface_get_index_of(surface, x, y); + surface_set_pixel_idx_alpha_fast(surface, index, color); +} + +static inline void surface_set_pixel_rgb_fast(SURFACE *surface, int x, int y, COLOR color) { + uint32_t index = surface_get_index_of(surface, x, y); + surface_set_pixel_idx_rgb_fast(surface, index, color); +} + +static inline void surface_set_pixel_rgba_fast(SURFACE *surface, int x, int y, COLOR color) { + uint32_t index = surface_get_index_of(surface, x, y); + surface_set_pixel_idx_rgba_fast(surface, index, color); +} + +static inline COLOR surface_get_pixel_alpha_fast(const SURFACE *surface, int x, int y) { + uint32_t index = surface_get_index_of(surface, x, y); + return surface_get_pixel_idx_alpha_fast(surface, index); +} + +static inline COLOR surface_get_pixel_rgb_fast(const SURFACE *surface, int x, int y) { + uint32_t index = surface_get_index_of(surface, x, y); + return surface_get_pixel_idx_rgb_fast(surface, index); +} + +static inline COLOR surface_get_pixel_rgba_fast(const SURFACE *surface, int x, int y) { + uint32_t index = surface_get_index_of(surface, x, y); + return surface_get_pixel_idx_rgba_fast(surface, index); +} + +static inline void surface_blit(SURFACE *source, SURFACE *destination, int x, int y) { + surface_blit_region(source, destination, 0, 0, source->width, source->height, x, y); +} + +static inline void surface_blit_blend(SURFACE *source, SURFACE *destination, int x, int y) { + surface_blit_blend_region(source, destination, 0, 0, source->width, source->height, x, y); +} + +static inline void surface_blit_trans(SURFACE *source, SURFACE *destination, int x, int y) { + surface_blit_trans_region(source, destination, 0, 0, source->width, source->height, x, y); +} + +static inline void surface_blit_tint(SURFACE *source, SURFACE *destination, int x, int y, COLOR tint) { + surface_blit_tint_region(source, destination, 0, 0, source->width, source->height, x, y, tint); +} + +static inline void surface_blit_blend_tint(SURFACE *source, SURFACE *destination, int x, int y, COLOR tint) { + surface_blit_blend_tint_region(source, destination, 0, 0, source->width, source->height, x, y, tint); +} + +#ifdef __cplusplus +} +#endif diff --git a/include/math/rect.h b/include/math/rect.h new file mode 100644 index 0000000..8518e01 --- /dev/null +++ b/include/math/rect.h @@ -0,0 +1,39 @@ +#pragma once + +#ifdef __cplusplus +extern "C" { +#endif + +typedef struct { + int x; + int y; + int width; + int height; +} RECT; + +static inline RECT rect_create(int x, int y, int width, int height) { + RECT rect; + rect.x = x; + rect.y = y; + rect.width = width; + rect.height = height; + return rect; +} + +static inline int rect_get_right(const RECT *rect) { + if (rect->width > 0) + return rect->x + rect->width - 1; + else + return rect->x; +} + +static inline int rect_get_bottom(const RECT *rect) { + if (rect->height > 0) + return rect->y + rect->height - 1; + else + return rect->y; +} + +#ifdef __cplusplus +} +#endif diff --git a/include/types.h b/include/types.h new file mode 100644 index 0000000..8399bfe --- /dev/null +++ b/include/types.h @@ -0,0 +1,13 @@ +#pragma once + +#include +#include +#include + +#define SWAP(T, a, b) \ + do { \ + T tmp = a; \ + a = b; \ + b = tmp; \ + } while (0) + diff --git a/src/assets/image.c b/src/assets/image.c new file mode 100644 index 0000000..520c936 --- /dev/null +++ b/src/assets/image.c @@ -0,0 +1,63 @@ +#include "assets/image.h" +#include "../vendor/stb_image.h" +#include "fbgfx/surface.h" +#include "types.h" +#include + +SURFACE* create_surface_from_pixels(uint8_t *pixels, int width, int height, int bpp) { + SURFACE *surface = NULL; + SURFACE_FORMAT format; + + switch (bpp) { + case 1: format = SURFACE_FORMAT_ALPHA; break; + case 3: format = SURFACE_FORMAT_RGB; break; + case 4: format = SURFACE_FORMAT_RGBA; break; + + default: + stbi_image_free(pixels); + return NULL; + } + + surface = surface_wrap_existing(pixels, width, height, format, 0); + if (!surface) { + stbi_image_free(pixels); + return NULL; + } + + // HACK: this one little time, we're going go manually toggle this setting in a way we shouldn't be! + // (all that stbi_image_free() does is call free(), so it's fine if we leave cleanup of the + // stb_image pixel buffer to surface_destroy() ...) + surface->wrapped_buffer = false; + + surface_convert_rgb_to_bgr(surface); + return surface; +} + +SURFACE* image_load_file(const char *filename) { + if (!filename) + return NULL; + + int width; + int height; + int bpp; + uint8_t *pixels = stbi_load(filename, &width, &height, &bpp, 0); + if (!pixels) + return NULL; + else + return create_surface_from_pixels(pixels, width, height, bpp); +} + +SURFACE* image_load_memory(const void *image_file_bin, uint32_t size) { + if (!image_file_bin || !size) + return NULL; + + int width; + int height; + int bpp; + uint8_t *pixels = stbi_load_from_memory(image_file_bin, size, &width, &height, &bpp, 0); + if (!pixels) + return NULL; + else + return create_surface_from_pixels(pixels, width, height, bpp); +} + diff --git a/src/fbgfx/clipping.c b/src/fbgfx/clipping.c new file mode 100644 index 0000000..777f340 --- /dev/null +++ b/src/fbgfx/clipping.c @@ -0,0 +1,115 @@ +#include "fbgfx/clipping.h" + +bool point_in_bounds(const RECT *clip_region, int x, int y) { + return ( + x >= clip_region->x && + y >= clip_region->y && + x <= rect_get_right(clip_region) && + y <= rect_get_bottom(clip_region) + ); +} + +bool line_in_bounds(const RECT *clip_region, int x1, int y1, int x2, int y2) { + int clip_region_right = rect_get_right(clip_region); + int clip_region_bottom = rect_get_bottom(clip_region); + + if (y1 < clip_region->y && + y2 < clip_region->y) + return false; + if (y1 > clip_region_bottom && + y2 > clip_region_bottom) + return false; + if (x1 < clip_region->x && + x2 < clip_region->x) + return false; + if (x1 > clip_region_right && + x2 > clip_region_right) + return false; + + return true; +} + +bool clamp_to_region(const RECT *clip_region, int *x1, int *y1, int *x2, int *y2) { + int clip_region_right = rect_get_right(clip_region); + int clip_region_bottom = rect_get_bottom(clip_region); + + if (*y1 < clip_region->y && + *y2 < clip_region->y) + return false; + if (*y1 > clip_region_bottom && + *y2 > clip_region_bottom) + return false; + if (*x1 < clip_region->x && + *x2 < clip_region->x) + return false; + if (*x1 > clip_region_right && + *x2 > clip_region_right) + return false; + + if (*y1 < clip_region->y) + *y1 = clip_region->y; + if (*y1 > clip_region_bottom) + *y1 = clip_region_bottom; + + if (*y2 < clip_region->y) + *y2 = clip_region->y; + if (*y2 > clip_region_bottom) + *y2 = clip_region_bottom; + + if (*x1 < clip_region->x) + *x1 = clip_region->x; + if (*x1 > clip_region_right) + *x1 = clip_region_right; + + if (*x2 < clip_region->x) + *x2 = clip_region->x; + if (*x2 > clip_region_right) + *x2 = clip_region_right; + + return true; +} + +bool clip_blit_region(const RECT *dest_clip_region, RECT *source_blit_region, int *dest_x, int *dest_y) { + // check for "completely out of bounds" scenarios first to return early + if (*dest_x > rect_get_right(dest_clip_region)) + return false; + if (*dest_x + source_blit_region->width < dest_clip_region->x) + return false; + if (*dest_y > rect_get_bottom(dest_clip_region)) + return false; + if (*dest_y + source_blit_region->height < dest_clip_region->y) + return false; + + // destination region is fully / partially within the clip region + + // off the left side? + if (*dest_x < dest_clip_region->x) { + int offset = dest_clip_region->x - *dest_x; + source_blit_region->x += offset; + source_blit_region->width -= offset; + *dest_x = dest_clip_region->x; + } + + // off the right side? + if (*dest_x > (dest_clip_region->width - source_blit_region->width)) { + int offset = *dest_x + source_blit_region->width - dest_clip_region->width; + source_blit_region->width -= offset; + } + + // off the top edge? + if (*dest_y < dest_clip_region->y) { + int offset = dest_clip_region->y - *dest_y; + source_blit_region->y += offset; + source_blit_region->height -= offset; + *dest_y = dest_clip_region->y; + } + + // off the bottom edge? + if (*dest_y > (dest_clip_region->height - source_blit_region->height)) { + int offset = *dest_y + source_blit_region->height - dest_clip_region->height; + source_blit_region->height -= offset; + } + + return true; +} + diff --git a/src/fbgfx/surface.c b/src/fbgfx/surface.c new file mode 100644 index 0000000..b829683 --- /dev/null +++ b/src/fbgfx/surface.c @@ -0,0 +1,161 @@ +#include "fbgfx/surface.h" +#include "types.h" +#include +#include + +SURFACE *top_screen_surface = NULL; +SURFACE *bottom_screen_surface = NULL; + +void surface_init() { + if (top_screen_surface || bottom_screen_surface) + return; // TODO: assert + + /* + top_screen_surface = surface_wrap_existing( + gfxGetFramebuffer(GFX_TOP, GFX_LEFT, NULL, NULL), + 400, 240, + SURFACE_FORMAT_RGB, + SURFACE_FLAGS_SIDEWAYS_BUFFER + ); + + bottom_screen_surface = surface_wrap_existing( + gfxGetFramebuffer(GFX_BOTTOM, GFX_LEFT, NULL, NULL), + 320, 240, + SURFACE_FORMAT_RGB, + SURFACE_FLAGS_SIDEWAYS_BUFFER + ); + */ +} + +void surface_exit() { + if (top_screen_surface) + surface_destroy(top_screen_surface); + if (bottom_screen_surface) + surface_destroy(bottom_screen_surface); + + top_screen_surface = NULL; + bottom_screen_surface = NULL; +} + +void surface_swap_framebuffers() { + /* + top_screen_surface->pixels = gfxGetFramebuffer(GFX_TOP, GFX_LEFT, NULL, NULL); + bottom_screen_surface->pixels = gfxGetFramebuffer(GFX_BOTTOM, GFX_LEFT, NULL, NULL); + */ +} + +void setup_initial_surface_properties(SURFACE *surface, int width, int height, SURFACE_FORMAT format, SURFACE_FLAGS flags) { + surface->width = width; + surface->height = height; + surface->format = format; + surface->flags = flags; + + surface->bytes_per_pixel = surface->format / 8; + + surface->clip_region = rect_create(0, 0, surface->width, surface->height); + + // lazy calculation for the win + uint32_t reference_index = surface_get_index_of(surface, 0, 0); + surface->x_inc = surface_get_index_of(surface, 1, 0) - reference_index; + surface->y_inc = surface_get_index_of(surface, 0, 1) - reference_index; +} + +SURFACE* surface_create(int width, int height, SURFACE_FORMAT format, SURFACE_FLAGS flags) { + SURFACE *surface = malloc(sizeof(SURFACE)); + if (surface == NULL) + return NULL; + + setup_initial_surface_properties(surface, width, height, format, flags); + + size_t pixels_buffer_size = surface_get_buffer_size(surface); + + surface->pixels = malloc(pixels_buffer_size); + if (surface->pixels == NULL) { + free(surface); + return NULL; + } + + surface->wrapped_buffer = false; + + memset(surface->pixels, 0, pixels_buffer_size); + + return surface; +} + +SURFACE* surface_wrap_existing(void *framebuffer, int width, int height, SURFACE_FORMAT format, SURFACE_FLAGS flags) { + SURFACE *surface = malloc(sizeof(SURFACE)); + if (surface == NULL) + return NULL; + + setup_initial_surface_properties(surface, width, height, format, flags); + + surface->pixels = (COLOR_COMPONENT*)framebuffer; + surface->wrapped_buffer = true; + + // note: not clearing framebuffer. since we're using existing memory, maybe the + // calling code wants it the way it is now? don't want to assume anything! + + return surface; +} + +void surface_resize(SURFACE *surface, int width, int height) { + if (!surface) + return; // TODO: assert + if (surface->wrapped_buffer) + return; // TODO: assert + + setup_initial_surface_properties(surface, width, height, surface->format, surface->flags); + size_t new_pixels_buffer_size = surface_get_buffer_size(surface); + + COLOR_COMPONENT *pixels = malloc(new_pixels_buffer_size); + if (pixels == NULL) + return; // TODO: assert + + free(surface->pixels); + surface->pixels = pixels; + + memset(surface->pixels, 0, new_pixels_buffer_size); +} + +void surface_destroy(SURFACE *surface) { + if (!surface) + return; // TODO: assert + + if (!surface->wrapped_buffer) + free(surface->pixels); + free(surface); +} + +////////////// + +void surface_clear(SURFACE *surface, COLOR color) { + size_t pixels_buffer_size = surface_get_buffer_size(surface); + COLOR_COMPONENT *p = surface->pixels; + COLOR_COMPONENT *end = p + pixels_buffer_size; + + COLOR_COMPONENT r, g, b, a; + color_extract(color, &r, &g, &b, &a); + + switch (surface->format) { + case SURFACE_FORMAT_ALPHA: + memset(surface->pixels, a, pixels_buffer_size); + break; + + case SURFACE_FORMAT_RGB: + for (; p < end; p += 3) { + p[0] = b; + p[1] = g; + p[2] = r; + } + break; + + case SURFACE_FORMAT_RGBA: + for (; p < end; p += 4) { + p[0] = b; + p[1] = g; + p[2] = r; + p[3] = a; + } + break; + } +} diff --git a/src/fbgfx/surface_blit.c b/src/fbgfx/surface_blit.c new file mode 100644 index 0000000..a285d62 --- /dev/null +++ b/src/fbgfx/surface_blit.c @@ -0,0 +1,192 @@ +#include +#include +#include "fbgfx/surface.h" +#include "fbgfx/clipping.h" +#include "math/rect.h" +#include "types.h" + +typedef struct { + COLOR tint; +} BLIT_PARAMS; + +#define BLIT_SOLID_PIXEL(src, dst, src_index, dst_index, params, src_fmt, dst_fmt) \ + source_pixel = GET_PIXEL_IDX_FAST(src_fmt, src, src_index); \ + SET_PIXEL_IDX_FAST(dst_fmt, dst, dst_index, source_pixel); + +#define BLIT_BLEND_PIXEL(src, dst, src_index, dst_index, params, src_fmt, dst_fmt) \ + source_pixel = GET_PIXEL_IDX_FAST(src_fmt, src, src_index); \ + dest_pixel = GET_PIXEL_IDX_FAST(dst_fmt, dst, dst_index); \ + SET_PIXEL_IDX_FAST(dst_fmt, dst, dst_index, color_blend(source_pixel, dest_pixel)); + +#define BLIT_TRANS_PIXEL(src, dst, src_index, dst_index, params, src_fmt, dst_fmt) \ + source_pixel = GET_PIXEL_IDX_FAST(src_fmt, src, src_index); \ + if (color_get_a(source_pixel) > 0) \ + SET_PIXEL_IDX_FAST(dst_fmt, dst, dst_index, source_pixel); + +#define BLIT_TINT_PIXEL(src, dst, src_index, dst_index, params, src_fmt, dst_fmt) \ + source_pixel = GET_PIXEL_IDX_FAST(src_fmt, src, src_index); \ + SET_PIXEL_IDX_FAST(dst_fmt, dst, dst_index, color_tint(source_pixel, params.tint)); + +#define BLIT_TINTALPHA_PIXEL(src, dst, src_index, dst_index, params, src_fmt, dst_fmt) \ + source_pixel = GET_PIXEL_IDX_FAST(src_fmt, src, src_index); \ + source_pixel = color_tint_alpha(color_get_a(source_pixel), params.tint); \ + SET_PIXEL_IDX_FAST(dst_fmt, dst, dst_index, source_pixel); + +#define BLIT_BLEND_TINT_PIXEL(src, dst, src_index, dst_index, params, src_fmt, dst_fmt) \ + source_pixel = GET_PIXEL_IDX_FAST(src_fmt, src, src_index); \ + dest_pixel = GET_PIXEL_IDX_FAST(dst_fmt, dst, dst_index); \ + source_pixel = color_tint(source_pixel, params.tint); \ + SET_PIXEL_IDX_FAST(dst_fmt, dst, dst_index, color_blend(source_pixel, dest_pixel)); + +#define BLIT_BLEND_TINTALPHA_PIXEL(src, dst, src_index, dst_index, params, src_fmt, dst_fmt) \ + source_pixel = GET_PIXEL_IDX_FAST(src_fmt, src, src_index); \ + dest_pixel = GET_PIXEL_IDX_FAST(dst_fmt, dst, dst_index); \ + source_pixel = color_tint_alpha(color_get_a(source_pixel), params.tint); \ + SET_PIXEL_IDX_FAST(dst_fmt, dst, dst_index, color_blend(source_pixel, dest_pixel)); + +#define BLIT_INNER_LOOP(src, dst, blit_pixel, params, src_fmt, dst_fmt) \ + blit_pixel(src, dst, src_x_index, dst_x_index, params, src_fmt, dst_fmt); \ + src_x_index += src->x_inc; \ + dst_x_index += dst->x_inc; \ + +#define BLIT(src, dst, src_region, dst_x, dst_y, blit_pixel, params, src_fmt, dst_fmt) \ + do { \ + uint32_t src_x_index; \ + uint32_t src_y_index; \ + uint32_t dst_x_index; \ + uint32_t dst_y_index; \ + COLOR source_pixel; \ + COLOR dest_pixel; \ + \ + src_y_index = surface_get_index_of(src, src_region.x, src_region.y); \ + dst_y_index = surface_get_index_of(dst, dst_x, dst_y); \ + \ + int x; \ + int y; \ + for (y = 0; y < src_region.height; ++y) { \ + src_x_index = src_y_index; \ + dst_x_index = dst_y_index; \ + \ + for (x = 0; x < src_region.width; ++x) { \ + BLIT_INNER_LOOP(src, dst, blit_pixel, params, src_fmt, dst_fmt); \ + } \ + \ + src_y_index += src->y_inc; \ + dst_y_index += dst->y_inc; \ + } \ + } while(0) + +#define BLIT_UNROLLED8(src, dst, src_region, dst_x, dst_y, blit_pixel, params, src_fmt, dst_fmt) \ + do { \ + uint32_t src_x_index; \ + uint32_t src_y_index; \ + uint32_t dst_x_index; \ + uint32_t dst_y_index; \ + COLOR source_pixel; \ + COLOR dest_pixel; \ + \ + src_y_index = surface_get_index_of(src, src_region.x, src_region.y); \ + dst_y_index = surface_get_index_of(dst, dst_x, dst_y); \ + \ + int x; \ + int y; \ + for (y = 0; y < src_region.height; ++y) { \ + src_x_index = src_y_index; \ + dst_x_index = dst_y_index; \ + \ + for (x = 0; x < src_region.width; x += 8) { \ + BLIT_INNER_LOOP(src, dst, blit_pixel, params, src_fmt, dst_fmt); \ + BLIT_INNER_LOOP(src, dst, blit_pixel, params, src_fmt, dst_fmt); \ + BLIT_INNER_LOOP(src, dst, blit_pixel, params, src_fmt, dst_fmt); \ + BLIT_INNER_LOOP(src, dst, blit_pixel, params, src_fmt, dst_fmt); \ + BLIT_INNER_LOOP(src, dst, blit_pixel, params, src_fmt, dst_fmt); \ + BLIT_INNER_LOOP(src, dst, blit_pixel, params, src_fmt, dst_fmt); \ + BLIT_INNER_LOOP(src, dst, blit_pixel, params, src_fmt, dst_fmt); \ + BLIT_INNER_LOOP(src, dst, blit_pixel, params, src_fmt, dst_fmt); \ + } \ + \ + src_y_index += src->y_inc; \ + dst_y_index += dst->y_inc; \ + } \ + } while(0) + +#define BLIT_FMT(src_fmt, dst_fmt) ((src_fmt << 16) | dst_fmt) + +#define SELECT_BLIT(src, dst, src_region, dst_x, dst_y, blit, blit_pixel, params) \ + switch (BLIT_FMT(src->format, dst->format)) { \ + case BLIT_FMT(SURFACE_FORMAT_ALPHA, SURFACE_FORMAT_ALPHA): blit(src, dst, src_region, dst_x, dst_y, blit_pixel, params, alpha, alpha); break; \ + case BLIT_FMT(SURFACE_FORMAT_ALPHA, SURFACE_FORMAT_RGB): blit(src, dst, src_region, dst_x, dst_y, blit_pixel, params, alpha, rgb); break; \ + case BLIT_FMT(SURFACE_FORMAT_ALPHA, SURFACE_FORMAT_RGBA): blit(src, dst, src_region, dst_x, dst_y, blit_pixel, params, alpha, rgba); break; \ + case BLIT_FMT(SURFACE_FORMAT_RGB, SURFACE_FORMAT_ALPHA): blit(src, dst, src_region, dst_x, dst_y, blit_pixel, params, rgb, alpha); break; \ + case BLIT_FMT(SURFACE_FORMAT_RGB, SURFACE_FORMAT_RGB): blit(src, dst, src_region, dst_x, dst_y, blit_pixel, params, rgb, rgb); break; \ + case BLIT_FMT(SURFACE_FORMAT_RGB, SURFACE_FORMAT_RGBA): blit(src, dst, src_region, dst_x, dst_y, blit_pixel, params, rgb, rgba); break; \ + case BLIT_FMT(SURFACE_FORMAT_RGBA, SURFACE_FORMAT_ALPHA): blit(src, dst, src_region, dst_x, dst_y, blit_pixel, params, rgba, alpha); break; \ + case BLIT_FMT(SURFACE_FORMAT_RGBA, SURFACE_FORMAT_RGB): blit(src, dst, src_region, dst_x, dst_y, blit_pixel, params, rgba, rgb); break; \ + case BLIT_FMT(SURFACE_FORMAT_RGBA, SURFACE_FORMAT_RGBA): blit(src, dst, src_region, dst_x, dst_y, blit_pixel, params, rgba, rgba); break; \ + } + +#define DO_BLIT(src, dst, src_region, dst_x, dst_y, blit_pixel, params) \ + if (src_region.width % 8 == 0) { \ + SELECT_BLIT(src, dst, src_region, dst_x, dst_y, BLIT_UNROLLED8, blit_pixel, params); \ + } else { \ + SELECT_BLIT(src, dst, src_region, dst_x, dst_y, BLIT, blit_pixel, params); \ + } + +void surface_blit_region(SURFACE *source, SURFACE *destination, int source_x, int source_y, int source_width, int source_height, int dest_x, int dest_y) { + RECT source_region = rect_create(source_x, source_y, source_width, source_height); + bool on_screen = clip_blit_region(&destination->clip_region, &source_region, &dest_x, &dest_y); + if (!on_screen) + return; + + DO_BLIT(source, destination, source_region, dest_x, dest_y, BLIT_SOLID_PIXEL, NULL); +} + +void surface_blit_blend_region(SURFACE *source, SURFACE *destination, int source_x, int source_y, int source_width, int source_height, int dest_x, int dest_y) { + RECT source_region = rect_create(source_x, source_y, source_width, source_height); + bool on_screen = clip_blit_region(&destination->clip_region, &source_region, &dest_x, &dest_y); + if (!on_screen) + return; + + DO_BLIT(source, destination, source_region, dest_x, dest_y, BLIT_BLEND_PIXEL, NULL); +} + +void surface_blit_trans_region(SURFACE *source, SURFACE *destination, int source_x, int source_y, int source_width, int source_height, int dest_x, int dest_y) { + RECT source_region = rect_create(source_x, source_y, source_width, source_height); + bool on_screen = clip_blit_region(&destination->clip_region, &source_region, &dest_x, &dest_y); + if (!on_screen) + return; + + DO_BLIT(source, destination, source_region, dest_x, dest_y, BLIT_TRANS_PIXEL, NULL); +} + +void surface_blit_tint_region(SURFACE *source, SURFACE *destination, int source_x, int source_y, int source_width, int source_height, int dest_x, int dest_y, COLOR tint) { + RECT source_region = rect_create(source_x, source_y, source_width, source_height); + bool on_screen = clip_blit_region(&destination->clip_region, &source_region, &dest_x, &dest_y); + if (!on_screen) + return; + + BLIT_PARAMS params; + params.tint = tint; + + if (source->format == SURFACE_FORMAT_ALPHA) { + DO_BLIT(source, destination, source_region, dest_x, dest_y, BLIT_TINTALPHA_PIXEL, params); + } else { + DO_BLIT(source, destination, source_region, dest_x, dest_y, BLIT_TINT_PIXEL, params); + } +} + +void surface_blit_blend_tint_region(SURFACE *source, SURFACE *destination, int source_x, int source_y, int source_width, int source_height, int dest_x, int dest_y, COLOR tint) { + RECT source_region = rect_create(source_x, source_y, source_width, source_height); + bool on_screen = clip_blit_region(&destination->clip_region, &source_region, &dest_x, &dest_y); + if (!on_screen) + return; + + BLIT_PARAMS params; + params.tint = tint; + + if (source->format == SURFACE_FORMAT_ALPHA) { + DO_BLIT(source, destination, source_region, dest_x, dest_y, BLIT_BLEND_TINTALPHA_PIXEL, params); + } else { + DO_BLIT(source, destination, source_region, dest_x, dest_y, BLIT_BLEND_TINT_PIXEL, params); + } +} diff --git a/src/fbgfx/surface_convert.c b/src/fbgfx/surface_convert.c new file mode 100644 index 0000000..06e355e --- /dev/null +++ b/src/fbgfx/surface_convert.c @@ -0,0 +1,16 @@ +#include "fbgfx/surface.h" +#include "types.h" + +void surface_convert_rgb_to_bgr(SURFACE *surface) { + if (surface->format != SURFACE_FORMAT_RGB && surface->format != SURFACE_FORMAT_RGBA) + return; + + uint32_t buffer_size = surface_get_buffer_size(surface); + + COLOR_COMPONENT *p = surface->pixels; + COLOR_COMPONENT *end = p + buffer_size; + for (; p < end; p += surface->bytes_per_pixel) { + SWAP(COLOR_COMPONENT, p[0], p[2]); + } +} + diff --git a/src/fbgfx/surface_primitives.c b/src/fbgfx/surface_primitives.c new file mode 100644 index 0000000..c09158a --- /dev/null +++ b/src/fbgfx/surface_primitives.c @@ -0,0 +1,280 @@ +#include "fbgfx/surface.h" +#include "fbgfx/clipping.h" +#include "math/rect.h" +#include "types.h" +#include + +#define SIGN(x) ((x < 0) ? -1 : ((x > 0) ? 1 : 0)) + + +/** + * PIXEL PLOTTING + */ + +static inline void set_pixel_fast(SURFACE *surface, int x, int y, COLOR color) { + switch (surface->format) { + case SURFACE_FORMAT_ALPHA: surface_set_pixel_alpha_fast(surface, x, y, color); break; + case SURFACE_FORMAT_RGB: surface_set_pixel_rgb_fast(surface, x, y, color); break; + case SURFACE_FORMAT_RGBA: surface_set_pixel_rgba_fast(surface, x, y, color); break; + } +} + +static inline COLOR get_pixel_fast(const SURFACE *surface, int x, int y) { + switch (surface->format) { + case SURFACE_FORMAT_ALPHA: return surface_get_pixel_alpha_fast(surface, x, y); + case SURFACE_FORMAT_RGB: return surface_get_pixel_rgb_fast(surface, x, y); + case SURFACE_FORMAT_RGBA: return surface_get_pixel_rgba_fast(surface, x, y); + } +} + +void surface_set_pixel(SURFACE *surface, int x, int y, COLOR color) { + if (!point_in_bounds(&surface->clip_region, x, y)) + return; + + set_pixel_fast(surface, x, y, color); +} + +void surface_set_pixel_fast(SURFACE *surface, int x, int y, COLOR color) { + set_pixel_fast(surface, x, y, color); +} + +COLOR surface_get_pixel(const SURFACE *surface, int x, int y) { + if (!point_in_bounds(&surface->clip_region, x, y)) + return 0; + + return get_pixel_fast(surface, x, y); +} + +COLOR surface_get_pixel_fast(SURFACE *surface, int x, int y) { + return get_pixel_fast(surface, x, y); +} + + + +/** + * LINE DRAWING + */ + +#define DRAW_LINE(surface, x1, y1, x2, y2, color, draw_pixel) \ + do { \ + int delta_x = x2 - x1; \ + int delta_y = y2 - y1; \ + int delta_x_abs = abs(delta_x); \ + int delta_y_abs = abs(delta_y); \ + int delta_x_sign = SIGN(delta_x); \ + int delta_y_sign = SIGN(delta_y); \ + int x = delta_y_abs / 2; \ + int y = delta_x_abs / 2; \ + int px = x1; \ + int py = y1; \ + \ + draw_pixel(surface, px, py, color); \ + \ + if (delta_x_abs >= delta_y_abs) { \ + int i; \ + for(i = 0; i < delta_x_abs; ++i) { \ + y += delta_y_abs; \ + \ + if (y >= delta_x_abs) { \ + y -= delta_x_abs; \ + py += delta_y_sign; \ + } \ + \ + px += delta_x_sign; \ + draw_pixel(surface, px, py, color); \ + } \ + \ + } else { \ + int i; \ + for(i = 0; i < delta_y_abs; ++i) { \ + x += delta_x_abs; \ + \ + if (x >= delta_y_abs) { \ + x -= delta_y_abs; \ + px += delta_x_sign; \ + } \ + \ + py += delta_y_sign; \ + draw_pixel(surface, px, py, color); \ + } \ + } \ + } while (0) + +void surface_line_fast(SURFACE *surface, int x1, int y1, int x2, int y2, COLOR color) { + switch (surface->format) { + case SURFACE_FORMAT_ALPHA: DRAW_LINE(surface, x1, y1, x2, y2, color, surface_set_pixel_alpha_fast); break; + case SURFACE_FORMAT_RGB: DRAW_LINE(surface, x1, y1, x2, y2, color, surface_set_pixel_rgb_fast); break; + case SURFACE_FORMAT_RGBA: DRAW_LINE(surface, x1, y1, x2, y2, color, surface_set_pixel_rgba_fast); break; + } +} + +void surface_line(SURFACE *surface, int x1, int y1, int x2, int y2, COLOR color) { + if (!line_in_bounds(&surface->clip_region, x1, y1, x2, y2)) + return; + + DRAW_LINE(surface, x1, y1, x2, y2, color, surface_set_pixel); +} + +#define DRAW_FAST_HLINE(surface, x1, x2, y, color, fmt) \ + do { \ + if (x2 < x1) \ + SWAP(int, x1, x2); \ + \ + int x; \ + uint32_t i = surface_get_index_of(surface, x1, y); \ + for (x = x1; x <= x2; ++x, i += surface->x_inc) \ + SET_PIXEL_IDX_FAST(fmt, surface, i, color); \ + } while(0) + +void surface_hline_fast(SURFACE *surface, int x1, int x2, int y, COLOR color) { + switch (surface->format) { + case SURFACE_FORMAT_ALPHA: DRAW_FAST_HLINE(surface, x1, x2, y, color, alpha); break; + case SURFACE_FORMAT_RGB: DRAW_FAST_HLINE(surface, x1, x2, y, color, rgb); break; + case SURFACE_FORMAT_RGBA: DRAW_FAST_HLINE(surface, x1, x2, y, color, rgba); break; + } +} + +void surface_hline(SURFACE *surface, int x1, int x2, int y, COLOR color) { + if (!clamp_to_region(&surface->clip_region, &x1, &y, &x2, &y)) + return; + + surface_hline_fast(surface, x1, x2, y, color); +} + +#define DRAW_FAST_VLINE(surface, x, y1, y2, color, fmt) \ + do { \ + if (y2 < y1) \ + SWAP(int, y1, y2); \ + \ + int y; \ + uint32_t i = surface_get_index_of(surface, x, y1); \ + for (y = y1; y <= y2; ++y, i += surface->y_inc) \ + SET_PIXEL_IDX_FAST(fmt, surface, i, color); \ + } while (0) + +void surface_vline_fast(SURFACE *surface, int x, int y1, int y2, COLOR color) { + switch (surface->format) { + case SURFACE_FORMAT_ALPHA: DRAW_FAST_VLINE(surface, x, y1, y2, color, alpha); break; + case SURFACE_FORMAT_RGB: DRAW_FAST_VLINE(surface, x, y1, y2, color, rgb); break; + case SURFACE_FORMAT_RGBA: DRAW_FAST_VLINE(surface, x, y1, y2, color, rgba); break; + } +} + +void surface_vline(SURFACE *surface, int x, int y1, int y2, COLOR color) { + if (!clamp_to_region(&surface->clip_region, &x, &y1, &x, &y2)) + return; + + surface_vline_fast(surface, x, y1, y2, color); +} + + + +/** + * RECTANGLE DRAWING + */ + +void surface_rect(SURFACE *surface, int x1, int y1, int x2, int y2, COLOR color) { + if (!line_in_bounds(&surface->clip_region, x1, y1, x2, y2)) + return; + + if (x2 < x1) + SWAP(int, x1, x2); + if (y2 < y1) + SWAP(int, y1, y2); + + int x; + for (x = x1; x <= x2; ++x) { + surface_set_pixel(surface, x, y1, color); + surface_set_pixel(surface, x, y2, color); + } + + int y; + for (y = y1; y <= y2; ++y) { + surface_set_pixel(surface, x1, y, color); + surface_set_pixel(surface, x2, y, color); + } +} + +#define DRAW_FAST_RECT(surface, x1, y1, x2, y2, color, fmt) \ + do { \ + if (x2 < x1) \ + SWAP(int, x1, x2); \ + if (y2 < y1) \ + SWAP(int, y1, y2); \ + \ + uint32_t a; \ + uint32_t b; \ + \ + int x = x1; \ + a = surface_get_index_of(surface, x, y1); \ + b = surface_get_index_of(surface, x, y2); \ + for (; x <= x2; ++x) { \ + SET_PIXEL_IDX_FAST(fmt, surface, a, color); \ + SET_PIXEL_IDX_FAST(fmt, surface, b, color); \ + a += surface->x_inc; \ + b += surface->x_inc; \ + } \ + \ + int y = y1; \ + a = surface_get_index_of(surface, x1, y); \ + b = surface_get_index_of(surface, x2, y); \ + for (; y <= y2; ++y) { \ + SET_PIXEL_IDX_FAST(fmt, surface, a, color); \ + SET_PIXEL_IDX_FAST(fmt, surface, b, color); \ + a += surface->y_inc; \ + b += surface->y_inc; \ + } \ + } while (0) + +void surface_rect_fast(SURFACE *surface, int x1, int y1, int x2, int y2, COLOR color) { + switch (surface->format) { + case SURFACE_FORMAT_ALPHA: DRAW_FAST_RECT(surface, x1, y1, x2, y2, color, alpha); break; + case SURFACE_FORMAT_RGB: DRAW_FAST_RECT(surface, x1, y1, x2, y2, color, rgb); break; + case SURFACE_FORMAT_RGBA: DRAW_FAST_RECT(surface, x1, y1, x2, y2, color, rgba); break; + } +} + + + +/** + * FILLED RECTANGLE DRAWING + */ + +#define DRAW_FAST_RECT_FILLED(surface, x1, y1, x2, y2, color, fmt) \ + do { \ + if (x2 < x1) \ + SWAP(int, x1, x2); \ + if (y2 < y1) \ + SWAP(int, y1, y2); \ + \ + uint32_t x_index; \ + uint32_t y_index; \ + int x; \ + int y; \ + \ + y_index = surface_get_index_of(surface, x1, y1); \ + for (y = y1; y <= y2; ++y) { \ + x_index = y_index; \ + for (x = x1; x <= x2; ++x) { \ + SET_PIXEL_IDX_FAST(fmt, surface, x_index, color); \ + x_index += surface->x_inc; \ + } \ + \ + y_index += surface->y_inc; \ + } \ + } while (0) + +void surface_rect_filled_fast(SURFACE *surface, int x1, int y1, int x2, int y2, COLOR color) { + switch (surface->format) { + case SURFACE_FORMAT_ALPHA: DRAW_FAST_RECT_FILLED(surface, x1, y1, x2, y2, color, alpha); break; + case SURFACE_FORMAT_RGB: DRAW_FAST_RECT_FILLED(surface, x1, y1, x2, y2, color, rgb); break; + case SURFACE_FORMAT_RGBA: DRAW_FAST_RECT_FILLED(surface, x1, y1, x2, y2, color, rgba); break; + } +} + +void surface_rect_filled(SURFACE *surface, int x1, int y1, int x2, int y2, COLOR color) { + if (!clamp_to_region(&surface->clip_region, &x1, &y1, &x2, &y2)) + return; + + surface_rect_filled_fast(surface, x1, y1, x2, y2, color); +} diff --git a/src/main.cpp b/src/main.cpp new file mode 100644 index 0000000..0684ff6 --- /dev/null +++ b/src/main.cpp @@ -0,0 +1,69 @@ +#include +#include +#include +#include "vm/vm.h" +#include "fbgfx/surface.h" +#include "fbgfx/clipping.h" +#include "assets/image.h" +#include "vm/vm_window.h" + +static inline double get_seconds() { + return (double)clock() / CLOCKS_PER_SEC; +} + +int main(int argc, char** argv) { + log_init(); + input_init(); + WINDOW *window = window_init("Rasterizer Test", 1280, 960, 320, 240); + if (!window) + return 1; + + SURFACE *sprite = image_load_file("/Users/gered/resources/grass01.bmp"); + SURFACE *sprite2 = surface_create(16, 16, SURFACE_FORMAT_ALPHA, SURFACE_FLAGS_NONE); + surface_clear(sprite2, color_create_rgba(0, 0, 0, 0)); + surface_rect_filled(sprite2, 4, 4, 11, 11, color_create_rgba(0, 0, 0, 128)); + surface_line(sprite2, 0, 0, 15, 15, color_create_rgba(0, 0, 0, 255)); + + double before; + double after; + int i; + + before = get_seconds(); + + for (i = 0; i < 500000; ++i) { + surface_blit(sprite, window->surface, 0, 0); + //surface_blit_blend(sprite, window->surface, 0, 0); + } + + after = get_seconds(); + + printf("time taken: %f\n", after - before); + + while (true) { + if (!window_do_events(window)) + break; + if (input_is_key_pressed(KSYM_ESCAPE)) + break; + + //surface_clear(window->surface, color_create_rgba(192, 192, 192, 255)); + int x, y; + for (y = 0; y < window->surface->height; y += 16) { + for (x = 0; x < window->surface->width; x += 16) { + surface_blit(sprite, window->surface, x, y); + } + } + + surface_line(window->surface, window->viewport_width / 2, window->viewport_height / 2, input_get_mouse_x(), input_get_mouse_y(), COLOR_YELLOW); + //surface_blit_tint(sprite, window->surface, INPUT_GetMouseX(), INPUT_GetMouseY(), COLOR_DOS_BROWN); + surface_blit_blend(sprite2, window->surface, 0, 0); + surface_blit_blend_tint(sprite2, window->surface, 16, 0, color_create_rgba(255, 0, 0, 64)); + + window_render(window); + } + + window_destroy(window); + input_destroy(); + log_end(); + + return 0; +} diff --git a/src/vendor/stb_image.c b/src/vendor/stb_image.c new file mode 100644 index 0000000..8d04cad --- /dev/null +++ b/src/vendor/stb_image.c @@ -0,0 +1,8 @@ +/* + * This source file obviously only exists to provide a spot + * for the stb_image implementation to exist. + */ + +#define STB_IMAGE_IMPLEMENTATION +#include "stb_image.h" + diff --git a/src/vendor/stb_image.h b/src/vendor/stb_image.h new file mode 100644 index 0000000..90d0d15 --- /dev/null +++ b/src/vendor/stb_image.h @@ -0,0 +1,4687 @@ +/* stb_image - v1.46 - public domain JPEG/PNG reader - http://nothings.org/stb_image.c + when you control the images you're loading + no warranty implied; use at your own risk + + Do this: + #define STB_IMAGE_IMPLEMENTATION + before you include this file in *one* C or C++ file to create the implementation. + + #define STBI_ASSERT(x) to avoid using assert.h. + + QUICK NOTES: + Primarily of interest to game developers and other people who can + avoid problematic images and only need the trivial interface + + JPEG baseline (no JPEG progressive) + PNG 8-bit-per-channel only + + TGA (not sure what subset, if a subset) + BMP non-1bpp, non-RLE + PSD (composited view only, no extra channels) + + GIF (*comp always reports as 4-channel) + HDR (radiance rgbE format) + PIC (Softimage PIC) + + - decode from memory or through FILE (define STBI_NO_STDIO to remove code) + - decode from arbitrary I/O callbacks + - overridable dequantizing-IDCT, YCbCr-to-RGB conversion (define STBI_SIMD) + + Latest revisions: + 1.46 (2014-08-26) fix broken tRNS chunk in non-paletted PNG + 1.45 (2014-08-16) workaround MSVC-ARM internal compiler error by wrapping malloc + 1.44 (2014-08-07) warnings + 1.43 (2014-07-15) fix MSVC-only bug in 1.42 + 1.42 (2014-07-09) no _CRT_SECURE_NO_WARNINGS; error-path fixes; STBI_ASSERT + 1.41 (2014-06-25) fix search&replace that messed up comments/error messages + 1.40 (2014-06-22) gcc warning + 1.39 (2014-06-15) TGA optimization bugfix, multiple BMP fixes + 1.38 (2014-06-06) suppress MSVC run-time warnings, fix accidental rename of 'skip' + 1.37 (2014-06-04) remove duplicate typedef + 1.36 (2014-06-03) converted to header file, allow reading incorrect iphoned-images without iphone flag + 1.35 (2014-05-27) warnings, bugfixes, TGA optimization, etc + + See end of file for full revision history. + + TODO: + stbi_info support for BMP,PSD,HDR,PIC + + + ============================ Contributors ========================= + + Image formats Bug fixes & warning fixes + Sean Barrett (jpeg, png, bmp) Marc LeBlanc + Nicolas Schulz (hdr, psd) Christpher Lloyd + Jonathan Dummer (tga) Dave Moore + Jean-Marc Lienher (gif) Won Chun + Tom Seddon (pic) the Horde3D community + Thatcher Ulrich (psd) Janez Zemva + Jonathan Blow + Laurent Gomila + Extensions, features Aruelien Pocheville + Jetro Lauha (stbi_info) Ryamond Barbiero + James "moose2000" Brown (iPhone PNG) David Woo + Ben "Disch" Wenger (io callbacks) Roy Eltham + Martin "SpartanJ" Golini Luke Graham + Thomas Ruf + John Bartholomew + Optimizations & bugfixes Ken Hamada + Fabian "ryg" Giesen Cort Stratton + Arseny Kapoulkine Blazej Dariusz Roszkowski + Thibault Reuille + Paul Du Bois + Guillaume George + Jerry Jansson + If your name should be here but Hayaki Saito + isn't, let Sean know. Johan Duparc + Ronny Chevalier + Michal Cichon +*/ + +#ifndef STBI_INCLUDE_STB_IMAGE_H +#define STBI_INCLUDE_STB_IMAGE_H + +// Limitations: +// - no jpeg progressive support +// - non-HDR formats support 8-bit samples only (jpeg, png) +// - no delayed line count (jpeg) -- IJG doesn't support either +// - no 1-bit BMP +// - GIF always returns *comp=4 +// +// Basic usage (see HDR discussion below): +// int x,y,n; +// unsigned char *data = stbi_load(filename, &x, &y, &n, 0); +// // ... process data if not NULL ... +// // ... x = width, y = height, n = # 8-bit components per pixel ... +// // ... replace '0' with '1'..'4' to force that many components per pixel +// // ... but 'n' will always be the number that it would have been if you said 0 +// stbi_image_free(data) +// +// Standard parameters: +// int *x -- outputs image width in pixels +// int *y -- outputs image height in pixels +// int *comp -- outputs # of image components in image file +// int req_comp -- if non-zero, # of image components requested in result +// +// The return value from an image loader is an 'unsigned char *' which points +// to the pixel data. The pixel data consists of *y scanlines of *x pixels, +// with each pixel consisting of N interleaved 8-bit components; the first +// pixel pointed to is top-left-most in the image. There is no padding between +// image scanlines or between pixels, regardless of format. The number of +// components N is 'req_comp' if req_comp is non-zero, or *comp otherwise. +// If req_comp is non-zero, *comp has the number of components that _would_ +// have been output otherwise. E.g. if you set req_comp to 4, you will always +// get RGBA output, but you can check *comp to easily see if it's opaque. +// +// An output image with N components has the following components interleaved +// in this order in each pixel: +// +// N=#comp components +// 1 grey +// 2 grey, alpha +// 3 red, green, blue +// 4 red, green, blue, alpha +// +// If image loading fails for any reason, the return value will be NULL, +// and *x, *y, *comp will be unchanged. The function stbi_failure_reason() +// can be queried for an extremely brief, end-user unfriendly explanation +// of why the load failed. Define STBI_NO_FAILURE_STRINGS to avoid +// compiling these strings at all, and STBI_FAILURE_USERMSG to get slightly +// more user-friendly ones. +// +// Paletted PNG, BMP, GIF, and PIC images are automatically depalettized. +// +// =========================================================================== +// +// iPhone PNG support: +// +// By default we convert iphone-formatted PNGs back to RGB; nominally they +// would silently load as BGR, except the existing code should have just +// failed on such iPhone PNGs. But you can disable this conversion by +// by calling stbi_convert_iphone_png_to_rgb(0), in which case +// you will always just get the native iphone "format" through. +// +// Call stbi_set_unpremultiply_on_load(1) as well to force a divide per +// pixel to remove any premultiplied alpha *only* if the image file explicitly +// says there's premultiplied data (currently only happens in iPhone images, +// and only if iPhone convert-to-rgb processing is on). +// +// =========================================================================== +// +// HDR image support (disable by defining STBI_NO_HDR) +// +// stb_image now supports loading HDR images in general, and currently +// the Radiance .HDR file format, although the support is provided +// generically. You can still load any file through the existing interface; +// if you attempt to load an HDR file, it will be automatically remapped to +// LDR, assuming gamma 2.2 and an arbitrary scale factor defaulting to 1; +// both of these constants can be reconfigured through this interface: +// +// stbi_hdr_to_ldr_gamma(2.2f); +// stbi_hdr_to_ldr_scale(1.0f); +// +// (note, do not use _inverse_ constants; stbi_image will invert them +// appropriately). +// +// Additionally, there is a new, parallel interface for loading files as +// (linear) floats to preserve the full dynamic range: +// +// float *data = stbi_loadf(filename, &x, &y, &n, 0); +// +// If you load LDR images through this interface, those images will +// be promoted to floating point values, run through the inverse of +// constants corresponding to the above: +// +// stbi_ldr_to_hdr_scale(1.0f); +// stbi_ldr_to_hdr_gamma(2.2f); +// +// Finally, given a filename (or an open file or memory block--see header +// file for details) containing image data, you can query for the "most +// appropriate" interface to use (that is, whether the image is HDR or +// not), using: +// +// stbi_is_hdr(char *filename); +// +// =========================================================================== +// +// I/O callbacks +// +// I/O callbacks allow you to read from arbitrary sources, like packaged +// files or some other source. Data read from callbacks are processed +// through a small internal buffer (currently 128 bytes) to try to reduce +// overhead. +// +// The three functions you must define are "read" (reads some bytes of data), +// "skip" (skips some bytes of data), "eof" (reports if the stream is at the end). + + +#ifndef STBI_NO_STDIO +#include +#endif // STBI_NO_STDIO + +#define STBI_VERSION 1 + +enum +{ + STBI_default = 0, // only used for req_comp + + STBI_grey = 1, + STBI_grey_alpha = 2, + STBI_rgb = 3, + STBI_rgb_alpha = 4 +}; + +typedef unsigned char stbi_uc; + +#ifdef __cplusplus +extern "C" { +#endif + +#ifdef STB_IMAGE_STATIC +#define STBIDEF static +#else +#define STBIDEF extern +#endif + +////////////////////////////////////////////////////////////////////////////// +// +// PRIMARY API - works on images of any type +// + +// +// load image by filename, open file, or memory buffer +// + +STBIDEF stbi_uc *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp); + +#ifndef STBI_NO_STDIO +STBIDEF stbi_uc *stbi_load (char const *filename, int *x, int *y, int *comp, int req_comp); +STBIDEF stbi_uc *stbi_load_from_file (FILE *f, int *x, int *y, int *comp, int req_comp); +// for stbi_load_from_file, file pointer is left pointing immediately after image +#endif + +typedef struct +{ + int (*read) (void *user,char *data,int size); // fill 'data' with 'size' bytes. return number of bytes actually read + void (*skip) (void *user,int n); // skip the next 'n' bytes, or 'unget' the last -n bytes if negative + int (*eof) (void *user); // returns nonzero if we are at end of file/data +} stbi_io_callbacks; + +STBIDEF stbi_uc *stbi_load_from_callbacks (stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp); + +#ifndef STBI_NO_HDR + STBIDEF float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp); + + #ifndef STBI_NO_STDIO + STBIDEF float *stbi_loadf (char const *filename, int *x, int *y, int *comp, int req_comp); + STBIDEF float *stbi_loadf_from_file (FILE *f, int *x, int *y, int *comp, int req_comp); + #endif + + STBIDEF float *stbi_loadf_from_callbacks (stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp); + + STBIDEF void stbi_hdr_to_ldr_gamma(float gamma); + STBIDEF void stbi_hdr_to_ldr_scale(float scale); + + STBIDEF void stbi_ldr_to_hdr_gamma(float gamma); + STBIDEF void stbi_ldr_to_hdr_scale(float scale); +#endif // STBI_NO_HDR + +// stbi_is_hdr is always defined +STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user); +STBIDEF int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len); +#ifndef STBI_NO_STDIO +STBIDEF int stbi_is_hdr (char const *filename); +STBIDEF int stbi_is_hdr_from_file(FILE *f); +#endif // STBI_NO_STDIO + + +// get a VERY brief reason for failure +// NOT THREADSAFE +STBIDEF const char *stbi_failure_reason (void); + +// free the loaded image -- this is just free() +STBIDEF void stbi_image_free (void *retval_from_stbi_load); + +// get image dimensions & components without fully decoding +STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp); +STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp); + +#ifndef STBI_NO_STDIO +STBIDEF int stbi_info (char const *filename, int *x, int *y, int *comp); +STBIDEF int stbi_info_from_file (FILE *f, int *x, int *y, int *comp); + +#endif + + + +// for image formats that explicitly notate that they have premultiplied alpha, +// we just return the colors as stored in the file. set this flag to force +// unpremultiplication. results are undefined if the unpremultiply overflow. +STBIDEF void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply); + +// indicate whether we should process iphone images back to canonical format, +// or just pass them through "as-is" +STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert); + + +// ZLIB client - used by PNG, available for other purposes + +STBIDEF char *stbi_zlib_decode_malloc_guesssize(const char *buffer, int len, int initial_size, int *outlen); +STBIDEF char *stbi_zlib_decode_malloc_guesssize_headerflag(const char *buffer, int len, int initial_size, int *outlen, int parse_header); +STBIDEF char *stbi_zlib_decode_malloc(const char *buffer, int len, int *outlen); +STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); + +STBIDEF char *stbi_zlib_decode_noheader_malloc(const char *buffer, int len, int *outlen); +STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen); + + +// define faster low-level operations (typically SIMD support) +#ifdef STBI_SIMD +typedef void (*stbi_idct_8x8)(stbi_uc *out, int out_stride, short data[64], unsigned short *dequantize); +// compute an integer IDCT on "input" +// input[x] = data[x] * dequantize[x] +// write results to 'out': 64 samples, each run of 8 spaced by 'out_stride' +// CLAMP results to 0..255 +typedef void (*stbi_YCbCr_to_RGB_run)(stbi_uc *output, stbi_uc const *y, stbi_uc const *cb, stbi_uc const *cr, int count, int step); +// compute a conversion from YCbCr to RGB +// 'count' pixels +// write pixels to 'output'; each pixel is 'step' bytes (either 3 or 4; if 4, write '255' as 4th), order R,G,B +// y: Y input channel +// cb: Cb input channel; scale/biased to be 0..255 +// cr: Cr input channel; scale/biased to be 0..255 + +STBIDEF void stbi_install_idct(stbi_idct_8x8 func); +STBIDEF void stbi_install_YCbCr_to_RGB(stbi_YCbCr_to_RGB_run func); +#endif // STBI_SIMD + + +#ifdef __cplusplus +} +#endif + +// +// +//// end header file ///////////////////////////////////////////////////// +#endif // STBI_INCLUDE_STB_IMAGE_H + +#ifdef STB_IMAGE_IMPLEMENTATION + +#ifndef STBI_NO_HDR +#include // ldexp +#include // strcmp, strtok +#endif + +#ifndef STBI_NO_STDIO +#include +#endif +#include +#include +#ifndef STBI_ASSERT +#include +#define STBI_ASSERT(x) assert(x) +#endif +#include +#include // ptrdiff_t on osx + +#ifndef _MSC_VER + #ifdef __cplusplus + #define stbi_inline inline + #else + #define stbi_inline + #endif +#else + #define stbi_inline __forceinline +#endif + + +#ifdef _MSC_VER +typedef unsigned short stbi__uint16; +typedef signed short stbi__int16; +typedef unsigned int stbi__uint32; +typedef signed int stbi__int32; +#else +#include +typedef uint16_t stbi__uint16; +typedef int16_t stbi__int16; +typedef uint32_t stbi__uint32; +typedef int32_t stbi__int32; +#endif + +// should produce compiler error if size is wrong +typedef unsigned char validate_uint32[sizeof(stbi__uint32)==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 + +/////////////////////////////////////////////// +// +// stbi__context struct and start_xxx functions + +// stbi__context structure is our basic context used by all images, so it +// contains all the IO context, plus some basic image information +typedef struct +{ + stbi__uint32 img_x, img_y; + int img_n, img_out_n; + + stbi_io_callbacks io; + void *io_user_data; + + int read_from_callbacks; + int buflen; + stbi_uc buffer_start[128]; + + stbi_uc *img_buffer, *img_buffer_end; + stbi_uc *img_buffer_original; +} stbi__context; + + +static void stbi__refill_buffer(stbi__context *s); + +// initialize a memory-decode context +static void stbi__start_mem(stbi__context *s, stbi_uc const *buffer, int len) +{ + s->io.read = NULL; + s->read_from_callbacks = 0; + s->img_buffer = s->img_buffer_original = (stbi_uc *) buffer; + s->img_buffer_end = (stbi_uc *) buffer+len; +} + +// initialize a callback-based context +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); +} + +//static void stop_file(stbi__context *s) { } + +#endif // !STBI_NO_STDIO + +static void stbi__rewind(stbi__context *s) +{ + // conceptually rewind SHOULD rewind to the beginning of the stream, + // but we just rewind to the beginning of the initial buffer, because + // we only use it after doing 'test', which only ever looks at at most 92 bytes + s->img_buffer = s->img_buffer_original; +} + +static int stbi__jpeg_test(stbi__context *s); +static stbi_uc *stbi__jpeg_load(stbi__context *s, int *x, int *y, int *comp, int req_comp); +static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__png_test(stbi__context *s); +static stbi_uc *stbi__png_load(stbi__context *s, int *x, int *y, int *comp, int req_comp); +static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__bmp_test(stbi__context *s); +static stbi_uc *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp); +static int stbi__tga_test(stbi__context *s); +static stbi_uc *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp); +static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp); +static int stbi__psd_test(stbi__context *s); +static stbi_uc *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req_comp); +#ifndef STBI_NO_HDR +static int stbi__hdr_test(stbi__context *s); +static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, int req_comp); +#endif +static int stbi__pic_test(stbi__context *s); +static stbi_uc *stbi__pic_load(stbi__context *s, int *x, int *y, int *comp, int req_comp); +static int stbi__gif_test(stbi__context *s); +static stbi_uc *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, int req_comp); +static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp); + + +// this is not threadsafe +static const char *stbi__g_failure_reason; + +STBIDEF const char *stbi_failure_reason(void) +{ + return stbi__g_failure_reason; +} + +static int stbi__err(const char *str) +{ + stbi__g_failure_reason = str; + return 0; +} + +static void *stbi__malloc(size_t size) +{ + return malloc(size); +} + +// stbi__err - error +// stbi__errpf - error returning pointer to float +// stbi__errpuc - error returning pointer to unsigned char + +#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)) + +STBIDEF void stbi_image_free(void *retval_from_stbi_load) +{ + free(retval_from_stbi_load); +} + +#ifndef STBI_NO_HDR +static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp); +static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp); +#endif + +static unsigned char *stbi_load_main(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + if (stbi__jpeg_test(s)) return stbi__jpeg_load(s,x,y,comp,req_comp); + if (stbi__png_test(s)) return stbi__png_load(s,x,y,comp,req_comp); + if (stbi__bmp_test(s)) return stbi__bmp_load(s,x,y,comp,req_comp); + if (stbi__gif_test(s)) return stbi__gif_load(s,x,y,comp,req_comp); + if (stbi__psd_test(s)) return stbi__psd_load(s,x,y,comp,req_comp); + if (stbi__pic_test(s)) return stbi__pic_load(s,x,y,comp,req_comp); + + #ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) { + float *hdr = stbi__hdr_load(s, x,y,comp,req_comp); + return stbi__hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp); + } + #endif + + // test tga last because it's a crappy test! + if (stbi__tga_test(s)) + return stbi__tga_load(s,x,y,comp,req_comp); + return stbi__errpuc("unknown image type", "Image not of any known type, or corrupt"); +} + +#ifndef STBI_NO_STDIO + +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 unsigned char *stbi_load(char const *filename, int *x, int *y, int *comp, int req_comp) +{ + FILE *f = stbi__fopen(filename, "rb"); + unsigned char *result; + 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 unsigned char *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_main(&s,x,y,comp,req_comp); + if (result) { + // need to 'unget' all the characters in the IO buffer + fseek(f, - (int) (s.img_buffer_end - s.img_buffer), SEEK_CUR); + } + return result; +} +#endif //!STBI_NO_STDIO + +STBIDEF unsigned char *stbi_load_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi_load_main(&s,x,y,comp,req_comp); +} + +unsigned char *stbi_load_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); + return stbi_load_main(&s,x,y,comp,req_comp); +} + +#ifndef STBI_NO_HDR + +float *stbi_loadf_main(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + unsigned char *data; + #ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) + return stbi__hdr_load(s,x,y,comp,req_comp); + #endif + data = stbi_load_main(s, x, y, comp, req_comp); + if (data) + return stbi__ldr_to_hdr(data, *x, *y, req_comp ? req_comp : *comp); + return stbi__errpf("unknown image type", "Image not of any known type, or corrupt"); +} + +float *stbi_loadf_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi_loadf_main(&s,x,y,comp,req_comp); +} + +float *stbi_loadf_from_callbacks(stbi_io_callbacks const *clbk, void *user, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); + return stbi_loadf_main(&s,x,y,comp,req_comp); +} + +#ifndef STBI_NO_STDIO +float *stbi_loadf(char const *filename, int *x, int *y, int *comp, int req_comp) +{ + float *result; + FILE *f = stbi__fopen(filename, "rb"); + if (!f) return stbi__errpf("can't fopen", "Unable to open file"); + result = stbi_loadf_from_file(f,x,y,comp,req_comp); + fclose(f); + return result; +} + +float *stbi_loadf_from_file(FILE *f, int *x, int *y, int *comp, int req_comp) +{ + stbi__context s; + stbi__start_file(&s,f); + return stbi_loadf_main(&s,x,y,comp,req_comp); +} +#endif // !STBI_NO_STDIO + +#endif // !STBI_NO_HDR + +// these is-hdr-or-not is defined independent of whether STBI_NO_HDR is +// defined, for API simplicity; if STBI_NO_HDR is defined, it always +// reports false! + +int stbi_is_hdr_from_memory(stbi_uc const *buffer, int len) +{ + #ifndef STBI_NO_HDR + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__hdr_test(&s); + #else + STBI_NOTUSED(buffer); + STBI_NOTUSED(len); + return 0; + #endif +} + +#ifndef STBI_NO_STDIO +STBIDEF int stbi_is_hdr (char const *filename) +{ + FILE *f = stbi__fopen(filename, "rb"); + int result=0; + if (f) { + result = stbi_is_hdr_from_file(f); + fclose(f); + } + return result; +} + +STBIDEF int stbi_is_hdr_from_file(FILE *f) +{ + #ifndef STBI_NO_HDR + stbi__context s; + stbi__start_file(&s,f); + return stbi__hdr_test(&s); + #else + return 0; + #endif +} +#endif // !STBI_NO_STDIO + +STBIDEF int stbi_is_hdr_from_callbacks(stbi_io_callbacks const *clbk, void *user) +{ + #ifndef STBI_NO_HDR + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) clbk, user); + return stbi__hdr_test(&s); + #else + return 0; + #endif +} + +#ifndef STBI_NO_HDR +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; + +void stbi_hdr_to_ldr_gamma(float gamma) { stbi__h2l_gamma_i = 1/gamma; } +void stbi_hdr_to_ldr_scale(float scale) { stbi__h2l_scale_i = 1/scale; } + +void stbi_ldr_to_hdr_gamma(float gamma) { stbi__l2h_gamma = gamma; } +void stbi_ldr_to_hdr_scale(float scale) { stbi__l2h_scale = scale; } +#endif + + +////////////////////////////////////////////////////////////////////////////// +// +// Common code used by all image loaders +// + +enum +{ + SCAN_load=0, + SCAN_type, + 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) { + // at end of file, treat same as if from memory, but need to handle case + // where s->img_buffer isn't pointing to safe memory, e.g. 0-byte file + 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; + } +} + +stbi_inline static stbi_uc 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; +} + +stbi_inline static int stbi__at_eof(stbi__context *s) +{ + if (s->io.read) { + if (!(s->io.eof)(s->io_user_data)) return 0; + // if feof() is true, check if buffer = end + // special case: we've only got the special 0 character at the end + if (s->read_from_callbacks == 0) return 1; + } + + return s->img_buffer >= s->img_buffer_end; +} + +static void stbi__skip(stbi__context *s, int n) +{ + 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, stbi_uc *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 stbi__uint32 stbi__get32be(stbi__context *s) +{ + stbi__uint32 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 stbi__uint32 stbi__get32le(stbi__context *s) +{ + stbi__uint32 z = stbi__get16le(s); + return z + (stbi__get16le(s) << 16); +} + +////////////////////////////////////////////////////////////////////////////// +// +// generic converter from built-in img_n to req_comp +// individual types do this automatically as much as possible (e.g. jpeg +// does all cases internally since it needs to colorspace convert anyway, +// and it never has alpha, so very few cases ). png can automatically +// interleave an alpha=255 channel, but falls back to this for other cases +// +// assume data buffer is malloced, so malloc a new one and free that one +// only failure mode is malloc failing + +static stbi_uc stbi__compute_y(int r, int g, int b) +{ + return (stbi_uc) (((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 *) stbi__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) + // convert source image with img_n components to one with req_comp components; + // avoid switch per pixel, so use switch per scanline and massive macros + switch (COMBO(img_n, req_comp)) { + CASE(1,2) dest[0]=src[0], dest[1]=255; break; + CASE(1,3) dest[0]=dest[1]=dest[2]=src[0]; break; + CASE(1,4) dest[0]=dest[1]=dest[2]=src[0], dest[3]=255; break; + CASE(2,1) dest[0]=src[0]; break; + CASE(2,3) dest[0]=dest[1]=dest[2]=src[0]; break; + CASE(2,4) dest[0]=dest[1]=dest[2]=src[0], dest[3]=src[1]; break; + CASE(3,4) dest[0]=src[0],dest[1]=src[1],dest[2]=src[2],dest[3]=255; break; + CASE(3,1) dest[0]=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_HDR +static float *stbi__ldr_to_hdr(stbi_uc *data, int x, int y, int comp) +{ + int i,k,n; + float *output = (float *) stbi__malloc(x * y * comp * sizeof(float)); + if (output == NULL) { free(data); return stbi__errpf("outofmem", "Out of memory"); } + // compute number of non-alpha components + if (comp & 1) n = comp; else n = comp-1; + for (i=0; i < x*y; ++i) { + for (k=0; k < n; ++k) { + output[i*comp + k] = (float) (pow(data[i*comp+k]/255.0f, stbi__l2h_gamma) * stbi__l2h_scale); + } + if (k < comp) output[i*comp + k] = data[i*comp+k]/255.0f; + } + free(data); + return output; +} + +#define stbi__float2int(x) ((int) (x)) +static stbi_uc *stbi__hdr_to_ldr(float *data, int x, int y, int comp) +{ + int i,k,n; + stbi_uc *output = (stbi_uc *) stbi__malloc(x * y * comp); + if (output == NULL) { free(data); return stbi__errpuc("outofmem", "Out of memory"); } + // compute number of non-alpha components + if (comp & 1) n = comp; else n = comp-1; + for (i=0; i < x*y; ++i) { + for (k=0; k < n; ++k) { + float z = (float) pow(data[i*comp+k]*stbi__h2l_scale_i, stbi__h2l_gamma_i) * 255 + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i*comp + k] = (stbi_uc) stbi__float2int(z); + } + if (k < comp) { + float z = data[i*comp+k] * 255 + 0.5f; + if (z < 0) z = 0; + if (z > 255) z = 255; + output[i*comp + k] = (stbi_uc) stbi__float2int(z); + } + } + free(data); + return output; +} +#endif + +////////////////////////////////////////////////////////////////////////////// +// +// "baseline" JPEG/JFIF decoder (not actually fully baseline implementation) +// +// simple implementation +// - channel subsampling of at most 2 in each dimension +// - doesn't support delayed output of y-dimension +// - simple interface (only one output format: 8-bit interleaved RGB) +// - doesn't try to recover corrupt jpegs +// - doesn't allow partial loading, loading multiple at once +// - still fast on x86 (copying globals into locals doesn't help x86) +// - allocates lots of intermediate memory (full size of all components) +// - non-interleaved case requires this anyway +// - allows good upsampling (see next) +// high-quality +// - upsampled channels are bilinearly interpolated, even across blocks +// - quality integer IDCT derived from IJG's 'slow' +// performance +// - fast huffman; reasonable integer IDCT +// - uses a lot of intermediate memory, could cache poorly +// - load http://nothings.org/remote/anemones.jpg 3 times on 2.8Ghz P4 +// stb_jpeg: 1.34 seconds (MSVC6, default release build) +// stb_jpeg: 1.06 seconds (MSVC6, processor = Pentium Pro) +// IJL11.dll: 1.08 seconds (compiled by intel) +// IJG 1998: 0.98 seconds (MSVC6, makefile provided by IJG) +// IJG 1998: 0.95 seconds (MSVC6, makefile + proc=PPro) + +// huffman decoding acceleration +#define FAST_BITS 9 // larger handles more cases; smaller stomps less cache + +typedef struct +{ + stbi_uc fast[1 << FAST_BITS]; + // weirdly, repacking this into AoS is a 10% speed loss, instead of a win + stbi__uint16 code[256]; + stbi_uc values[256]; + stbi_uc size[257]; + unsigned int maxcode[18]; + int delta[17]; // old 'firstsymbol' - old 'firstcode' +} stbi__huffman; + +typedef struct +{ + #ifdef STBI_SIMD + unsigned short dequant2[4][64]; + #endif + stbi__context *s; + stbi__huffman huff_dc[4]; + stbi__huffman huff_ac[4]; + stbi_uc dequant[4][64]; + +// sizes for components, interleaved MCUs + int img_h_max, img_v_max; + int img_mcu_x, img_mcu_y; + int img_mcu_w, img_mcu_h; + +// definition of jpeg image component + struct + { + int id; + int h,v; + int tq; + int hd,ha; + int dc_pred; + + int x,y,w2,h2; + stbi_uc *data; + void *raw_data; + stbi_uc *linebuf; + } img_comp[4]; + + stbi__uint32 code_buffer; // jpeg entropy-coded buffer + int code_bits; // number of valid bits + unsigned char marker; // marker seen while filling entropy buffer + int nomore; // flag if we saw a marker so must stop + + int scan_n, order[4]; + int restart_interval, todo; +} stbi__jpeg; + +static int stbi__build_huffman(stbi__huffman *h, int *count) +{ + int i,j,k=0,code; + // build size list for each symbol (from JPEG spec) + for (i=0; i < 16; ++i) + for (j=0; j < count[i]; ++j) + h->size[k++] = (stbi_uc) (i+1); + h->size[k] = 0; + + // compute actual symbols (from jpeg spec) + code = 0; + k = 0; + for(j=1; j <= 16; ++j) { + // compute delta to add to code to compute symbol id + h->delta[j] = k - code; + if (h->size[k] == j) { + while (h->size[k] == j) + h->code[k++] = (stbi__uint16) (code++); + if (code-1 >= (1 << j)) return stbi__err("bad code lengths","Corrupt JPEG"); + } + // compute largest code + 1 for this size, preshifted as needed later + h->maxcode[j] = code << (16-j); + code <<= 1; + } + h->maxcode[j] = 0xffffffff; + + // build non-spec acceleration table; 255 is flag for not-accelerated + memset(h->fast, 255, 1 << FAST_BITS); + for (i=0; i < k; ++i) { + int s = h->size[i]; + if (s <= FAST_BITS) { + int c = h->code[i] << (FAST_BITS-s); + int m = 1 << (FAST_BITS-s); + for (j=0; j < m; ++j) { + h->fast[c+j] = (stbi_uc) i; + } + } + } + return 1; +} + +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); +} + +// (1 << n) - 1 +static stbi__uint32 stbi__bmask[17]={0,1,3,7,15,31,63,127,255,511,1023,2047,4095,8191,16383,32767,65535}; + +// decode a jpeg huffman value from the bitstream +stbi_inline static int 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); + + // look at the top FAST_BITS and determine what symbol ID it is, + // if the code is <= FAST_BITS + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS)-1); + k = h->fast[c]; + if (k < 255) { + int s = h->size[k]; + if (s > j->code_bits) + return -1; + j->code_buffer <<= s; + j->code_bits -= s; + return h->values[k]; + } + + // naive test is to shift the code_buffer down so k bits are + // valid, then test against maxcode. To speed this up, we've + // preshifted maxcode left so that it has (16-k) 0s at the + // end; in other words, regardless of the number of bits, it + // wants to be compared against something shifted to have 16; + // that way we don't need to shift inside the loop. + temp = j->code_buffer >> 16; + for (k=FAST_BITS+1 ; ; ++k) + if (temp < h->maxcode[k]) + break; + if (k == 17) { + // error! code not found + j->code_bits -= 16; + return -1; + } + + if (k > j->code_bits) + return -1; + + // convert the huffman code to the symbol id + c = ((j->code_buffer >> (32 - k)) & stbi__bmask[k]) + h->delta[k]; + STBI_ASSERT((((j->code_buffer) >> (32 - h->size[c])) & stbi__bmask[h->size[c]]) == h->code[c]); + + // convert the id to a symbol + j->code_bits -= k; + j->code_buffer <<= k; + return h->values[c]; +} + +// combined JPEG 'receive' and JPEG 'extend', since baseline +// always extends everything it receives. +stbi_inline static int stbi__extend_receive(stbi__jpeg *j, int n) +{ + unsigned int m = 1 << (n-1); + unsigned int k; + if (j->code_bits < n) stbi__grow_buffer_unsafe(j); + + #if 1 + k = stbi_lrot(j->code_buffer, n); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + #else + k = (j->code_buffer >> (32 - n)) & stbi__bmask[n]; + j->code_bits -= n; + j->code_buffer <<= n; + #endif + // the following test is probably a random branch that won't + // predict well. I tried to table accelerate it but failed. + // maybe it's compiling as a conditional move? + if (k < m) + return (-1 << n) + k + 1; + else + return k; +} + +// given a value that's at position X in the zigzag stream, +// where does it appear in the 8x8 matrix coded as row-major? +static stbi_uc 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, + // let corrupt input sample past end + 63, 63, 63, 63, 63, 63, 63, 63, + 63, 63, 63, 63, 63, 63, 63 +}; + +// decode one 64-entry block-- +static int stbi__jpeg_decode_block(stbi__jpeg *j, short data[64], stbi__huffman *hdc, stbi__huffman *hac, int b) +{ + int diff,dc,k; + int t = stbi__jpeg_huff_decode(j, hdc); + if (t < 0) return stbi__err("bad huffman code","Corrupt JPEG"); + + // 0 all the ac values now so we can do it 32-bits at a time + memset(data,0,64*sizeof(data[0])); + + diff = t ? 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; + + // decode AC components, see JPEG spec + k = 1; + 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 (rs != 0xf0) break; // end block + k += 16; + } else { + k += r; + // decode into unzigzag'd location + data[stbi__jpeg_dezigzag[k++]] = (short) stbi__extend_receive(j,s); + } + } while (k < 64); + return 1; +} + +// take a -128..127 value and stbi__clamp it and convert to 0..255 +stbi_inline static stbi_uc stbi__clamp(int x) +{ + // trick to use a single test to catch both cases + if ((unsigned int) x > 255) { + if (x < 0) return 0; + if (x > 255) return 255; + } + return (stbi_uc) x; +} + +#define stbi__f2f(x) (int) (((x) * 4096 + 0.5)) +#define stbi__fsh(x) ((x) << 12) + +// derived from jidctint -- DCT_ISLOW +#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; + +#ifdef STBI_SIMD +typedef unsigned short stbi_dequantize_t; +#else +typedef stbi_uc stbi_dequantize_t; +#endif + +// .344 seconds on 3*anemones.jpg +static void stbi__idct_block(stbi_uc *out, int out_stride, short data[64], stbi_dequantize_t *dequantize) +{ + int i,val[64],*v=val; + stbi_dequantize_t *dq = dequantize; + stbi_uc *o; + short *d = data; + + // columns + for (i=0; i < 8; ++i,++d,++dq, ++v) { + // if all zeroes, shortcut -- this avoids dequantizing 0s and IDCTing + if (d[ 8]==0 && d[16]==0 && d[24]==0 && d[32]==0 + && d[40]==0 && d[48]==0 && d[56]==0) { + // no shortcut 0 seconds + // (1|2|3|4|5|6|7)==0 0 seconds + // all separate -0.047 seconds + // 1 && 2|3 && 4|5 && 6|7: -0.047 seconds + int dcterm = d[0] * dq[0] << 2; + v[0] = v[8] = v[16] = v[24] = v[32] = v[40] = v[48] = v[56] = dcterm; + } else { + STBI__IDCT_1D(d[ 0]*dq[ 0],d[ 8]*dq[ 8],d[16]*dq[16],d[24]*dq[24], + d[32]*dq[32],d[40]*dq[40],d[48]*dq[48],d[56]*dq[56]) + // constants scaled things up by 1<<12; let's bring them back + // down, but keep 2 extra bits of precision + x0 += 512; x1 += 512; x2 += 512; x3 += 512; + v[ 0] = (x0+t3) >> 10; + v[56] = (x0-t3) >> 10; + v[ 8] = (x1+t2) >> 10; + v[48] = (x1-t2) >> 10; + v[16] = (x2+t1) >> 10; + v[40] = (x2-t1) >> 10; + v[24] = (x3+t0) >> 10; + v[32] = (x3-t0) >> 10; + } + } + + for (i=0, v=val, o=out; i < 8; ++i,v+=8,o+=out_stride) { + // no fast case since the first 1D IDCT spread components out + STBI__IDCT_1D(v[0],v[1],v[2],v[3],v[4],v[5],v[6],v[7]) + // constants scaled things up by 1<<12, plus we had 1<<2 from first + // loop, plus horizontal and vertical each scale by sqrt(8) so together + // we've got an extra 1<<3, so 1<<17 total we need to remove. + // so we want to round that, which means adding 0.5 * 1<<17, + // aka 65536. Also, we'll end up with -128 to 127 that we want + // to encode as 0..255 by adding 128, so we'll add that before the shift + x0 += 65536 + (128<<17); + x1 += 65536 + (128<<17); + x2 += 65536 + (128<<17); + x3 += 65536 + (128<<17); + // tried computing the shifts into temps, or'ing the temps to see + // if any were out of range, but that was slower + o[0] = 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_SIMD +static stbi_idct_8x8 stbi__idct_installed = stbi__idct_block; + +STBIDEF void stbi_install_idct(stbi_idct_8x8 func) +{ + stbi__idct_installed = func; +} +#endif + +#define STBI__MARKER_none 0xff +// if there's a pending marker from the entropy stream, return that +// otherwise, fetch from the stream and get a marker. if there's no +// marker, return 0xff, which is never a valid marker value +static stbi_uc stbi__get_marker(stbi__jpeg *j) +{ + stbi_uc 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; +} + +// in each scan, we'll have scan_n components, and the order +// of the components is specified by order[] +#define STBI__RESTART(x) ((x) >= 0xd0 && (x) <= 0xd7) + +// after a restart interval, stbi__jpeg_reset the entropy decoder and +// the dc prediction +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; + // no more than 1<<31 MCUs if no restart_interal? that's plenty safe, + // since we don't even allow 1<<30 pixels +} + +static int stbi__parse_entropy_coded_data(stbi__jpeg *z) +{ + stbi__jpeg_reset(z); + if (z->scan_n == 1) { + int i,j; + #ifdef STBI_SIMD + __declspec(align(16)) + #endif + short data[64]; + int n = z->order[0]; + // non-interleaved data, we just need to process one block at a time, + // in trivial scanline order + // number of blocks to do just depends on how many actual "pixels" this + // component has, independent of interleaved MCU blocking and such + int w = (z->img_comp[n].x+7) >> 3; + int h = (z->img_comp[n].y+7) >> 3; + for (j=0; j < h; ++j) { + for (i=0; i < w; ++i) { + if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+z->img_comp[n].ha, n)) return 0; + #ifdef STBI_SIMD + stbi__idct_installed(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data, z->dequant2[z->img_comp[n].tq]); + #else + stbi__idct_block(z->img_comp[n].data+z->img_comp[n].w2*j*8+i*8, z->img_comp[n].w2, data, z->dequant[z->img_comp[n].tq]); + #endif + // every data block is an MCU, so countdown the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + // if it's NOT a restart, then just bail, so we get corrupt data + // rather than no data + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + } else { // interleaved! + int i,j,k,x,y; + short data[64]; + for (j=0; j < z->img_mcu_y; ++j) { + for (i=0; i < z->img_mcu_x; ++i) { + // scan an interleaved mcu... process scan_n components in order + for (k=0; k < z->scan_n; ++k) { + int n = z->order[k]; + // scan out an mcu's worth of this component; that's just determined + // by the basic H and V specified for the component + for (y=0; y < z->img_comp[n].v; ++y) { + for (x=0; x < z->img_comp[n].h; ++x) { + int x2 = (i*z->img_comp[n].h + x)*8; + int y2 = (j*z->img_comp[n].v + y)*8; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc+z->img_comp[n].hd, z->huff_ac+z->img_comp[n].ha, n)) return 0; + #ifdef STBI_SIMD + stbi__idct_installed(z->img_comp[n].data+z->img_comp[n].w2*y2+x2, z->img_comp[n].w2, data, z->dequant2[z->img_comp[n].tq]); + #else + stbi__idct_block(z->img_comp[n].data+z->img_comp[n].w2*y2+x2, z->img_comp[n].w2, data, z->dequant[z->img_comp[n].tq]); + #endif + } + } + } + // after all interleaved components, that's an interleaved MCU, + // so now count down the restart interval + if (--z->todo <= 0) { + if (z->code_bits < 24) stbi__grow_buffer_unsafe(z); + // if it's NOT a restart, then just bail, so we get corrupt data + // rather than no data + if (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + } + return 1; +} + +static int stbi__process_marker(stbi__jpeg *z, int m) +{ + int L; + switch (m) { + case STBI__MARKER_none: // no marker found + return stbi__err("expected marker","Corrupt JPEG"); + + case 0xC2: // stbi__SOF - progressive + return stbi__err("progressive jpeg","JPEG format not supported (progressive)"); + + case 0xDD: // DRI - specify restart interval + 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: // DQT - define quantization table + 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); + #ifdef STBI_SIMD + for (i=0; i < 64; ++i) + z->dequant2[t][i] = z->dequant[t][i]; + #endif + L -= 65; + } + return L==0; + + case 0xC4: // DHT - define huffman table + L = stbi__get16be(z->s)-2; + while (L > 0) { + stbi_uc *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); + L -= n; + } + return L==0; + } + // check for comment block or APP blocks + if ((m >= 0xE0 && m <= 0xEF) || m == 0xFE) { + stbi__skip(z->s, stbi__get16be(z->s)-2); + return 1; + } + return 0; +} + +// after we see stbi__SOS +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 stbi__SOS component count","Corrupt JPEG"); + if (Ls != 6+2*z->scan_n) return stbi__err("bad stbi__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; + } + if (stbi__get8(z->s) != 0) return stbi__err("bad stbi__SOS","Corrupt JPEG"); + stbi__get8(z->s); // should be 63, but might be 0 + if (stbi__get8(z->s) != 0) return stbi__err("bad stbi__SOS","Corrupt JPEG"); + + 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 stbi__SOF len","Corrupt JPEG"); // JPEG + p = stbi__get8(s); if (p != 8) return stbi__err("only 8-bit","JPEG format not supported: 8-bit only"); // JPEG baseline + s->img_y = stbi__get16be(s); if (s->img_y == 0) return stbi__err("no header height", "JPEG format not supported: delayed height"); // Legal, but we don't handle it--but neither does IJG + s->img_x = stbi__get16be(s); if (s->img_x == 0) return stbi__err("0 width","Corrupt JPEG"); // JPEG requires + c = stbi__get8(s); + if (c != 3 && c != 1) return stbi__err("bad component count","Corrupt JPEG"); // JFIF requires + s->img_n = c; + for (i=0; i < c; ++i) { + z->img_comp[i].data = NULL; + z->img_comp[i].linebuf = NULL; + } + + if (Lf != 8+3*s->img_n) return stbi__err("bad stbi__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) // JFIF requires + if (z->img_comp[i].id != i) // some version of jpegtran outputs non-JFIF-compliant files! + 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 != 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; + } + + // compute interleaved mcu info + z->img_h_max = h_max; + z->img_v_max = v_max; + z->img_mcu_w = h_max * 8; + z->img_mcu_h = v_max * 8; + z->img_mcu_x = (s->img_x + z->img_mcu_w-1) / z->img_mcu_w; + z->img_mcu_y = (s->img_y + z->img_mcu_h-1) / z->img_mcu_h; + + for (i=0; i < s->img_n; ++i) { + // number of effective pixels (e.g. for non-interleaved MCU) + z->img_comp[i].x = (s->img_x * z->img_comp[i].h + h_max-1) / h_max; + z->img_comp[i].y = (s->img_y * z->img_comp[i].v + v_max-1) / v_max; + // to simplify generation, we'll allocate enough memory to decode + // the bogus oversized data from using interleaved MCUs and their + // big blocks (e.g. a 16x16 iMCU on an image of width 33); we won't + // discard the extra data until colorspace conversion + z->img_comp[i].w2 = z->img_mcu_x * z->img_comp[i].h * 8; + z->img_comp[i].h2 = z->img_mcu_y * z->img_comp[i].v * 8; + z->img_comp[i].raw_data = stbi__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"); + } + // align blocks for installable-idct using mmx/sse + z->img_comp[i].data = (stbi_uc*) (((size_t) z->img_comp[i].raw_data + 15) & ~15); + z->img_comp[i].linebuf = NULL; + } + + return 1; +} + +// use comparisons since in some cases we handle more than one case (e.g. stbi__SOF) +#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) +#define stbi__SOS(x) ((x) == 0xda) + +static int decode_jpeg_header(stbi__jpeg *z, int scan) +{ + int m; + z->marker = STBI__MARKER_none; // initialize cached marker to empty + m = stbi__get_marker(z); + if (!stbi__SOI(m)) return stbi__err("no stbi__SOI","Corrupt JPEG"); + if (scan == 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) { + // some files have extra padding after their blocks, so ok, we'll scan + if (stbi__at_eof(z->s)) return stbi__err("no stbi__SOF", "Corrupt JPEG"); + m = stbi__get_marker(z); + } + } + if (!stbi__process_frame_header(z, scan)) return 0; + return 1; +} + +static int decode_jpeg_image(stbi__jpeg *j) +{ + int m; + j->restart_interval = 0; + if (!decode_jpeg_header(j, 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 ) { + // handle 0s at the end of image data from IP Kamera 9060 + 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 0; + } + } + // if we reach eof without hitting a marker, stbi__get_marker() below will fail and we'll eventually return 0 + } + } else { + if (!stbi__process_marker(j, m)) return 0; + } + m = stbi__get_marker(j); + } + return 1; +} + +// static jfif-centered resampling (across block boundaries) + +typedef stbi_uc *(*resample_row_func)(stbi_uc *out, stbi_uc *in0, stbi_uc *in1, + int w, int hs); + +#define stbi__div4(x) ((stbi_uc) ((x) >> 2)) + +static stbi_uc *resample_row_1(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + STBI_NOTUSED(out); + STBI_NOTUSED(in_far); + STBI_NOTUSED(w); + STBI_NOTUSED(hs); + return in_near; +} + +static stbi_uc* stbi__resample_row_v_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // need to generate two samples vertically for every one in input + int i; + STBI_NOTUSED(hs); + for (i=0; i < w; ++i) + out[i] = stbi__div4(3*in_near[i] + in_far[i] + 2); + return out; +} + +static stbi_uc* stbi__resample_row_h_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // need to generate two samples horizontally for every one in input + int i; + stbi_uc *input = in_near; + + if (w == 1) { + // if only one sample, can't do any interpolation + out[0] = out[1] = input[0]; + return out; + } + + out[0] = input[0]; + out[1] = 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) ((stbi_uc) ((x) >> 4)) + +static stbi_uc *stbi__resample_row_hv_2(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // need to generate 2x2 samples for every one in input + int i,t0,t1; + if (w == 1) { + out[0] = out[1] = 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; +} + +static stbi_uc *stbi__resample_row_generic(stbi_uc *out, stbi_uc *in_near, stbi_uc *in_far, int w, int hs) +{ + // resample with nearest-neighbor + 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; +} + +#define float2fixed(x) ((int) ((x) * 65536 + 0.5)) + +// 0.38 seconds on 3*anemones.jpg (0.25 with processor = Pro) +// VC6 without processor=Pro is generating multiple LEAs per multiply! +static void stbi__YCbCr_to_RGB_row(stbi_uc *out, const stbi_uc *y, const stbi_uc *pcb, const stbi_uc *pcr, int count, int step) +{ + int i; + for (i=0; i < count; ++i) { + int y_fixed = (y[i] << 16) + 32768; // rounding + int r,g,b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr*float2fixed(1.40200f); + g = y_fixed - cr*float2fixed(0.71414f) - cb*float2fixed(0.34414f); + b = y_fixed + cb*float2fixed(1.77200f); + r >>= 16; + g >>= 16; + b >>= 16; + if ((unsigned) r > 255) { if (r < 0) r = 0; else r = 255; } + if ((unsigned) g > 255) { if (g < 0) g = 0; else g = 255; } + if ((unsigned) b > 255) { if (b < 0) b = 0; else b = 255; } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } +} + +#ifdef STBI_SIMD +static stbi_YCbCr_to_RGB_run stbi__YCbCr_installed = stbi__YCbCr_to_RGB_row; + +STBIDEF void stbi_install_YCbCr_to_RGB(stbi_YCbCr_to_RGB_run func) +{ + stbi__YCbCr_installed = func; +} +#endif + + +// clean up the temporary component buffers +static void 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].linebuf) { + free(j->img_comp[i].linebuf); + j->img_comp[i].linebuf = NULL; + } + } +} + +typedef struct +{ + resample_row_func resample; + stbi_uc *line0,*line1; + int hs,vs; // expansion factor in each axis + int w_lores; // horizontal pixels pre-expansion + int ystep; // how far through vertical expansion we are + int ypos; // which pre-expansion row we're on +} stbi__resample; + +static stbi_uc *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; // make stbi__cleanup_jpeg safe + + // validate req_comp + if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); + + // load a jpeg image from whichever source + if (!decode_jpeg_image(z)) { stbi__cleanup_jpeg(z); return NULL; } + + // determine actual number of components to generate + n = req_comp ? req_comp : z->s->img_n; + + if (z->s->img_n == 3 && n < 3) + decode_n = 1; + else + decode_n = z->s->img_n; + + // resample and color-convert + { + int k; + unsigned int i,j; + stbi_uc *output; + stbi_uc *coutput[4]; + + stbi__resample res_comp[4]; + + for (k=0; k < decode_n; ++k) { + stbi__resample *r = &res_comp[k]; + + // allocate line buffer big enough for upsampling off the edges + // with upsample factor of 4 + z->img_comp[k].linebuf = (stbi_uc *) stbi__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 = stbi__resample_row_hv_2; + else r->resample = stbi__resample_row_generic; + } + + // can't error after this so, this is safe + output = (stbi_uc *) stbi__malloc(n * z->s->img_x * z->s->img_y + 1); + if (!output) { stbi__cleanup_jpeg(z); return stbi__errpuc("outofmem", "Out of memory"); } + + // now go ahead and resample + for (j=0; j < z->s->img_y; ++j) { + stbi_uc *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) { + stbi_uc *y = coutput[0]; + if (z->s->img_n == 3) { + #ifdef STBI_SIMD + stbi__YCbCr_installed(out, y, coutput[1], coutput[2], z->s->img_x, n); + #else + stbi__YCbCr_to_RGB_row(out, y, coutput[1], coutput[2], z->s->img_x, n); + #endif + } else + for (i=0; i < z->s->img_x; ++i) { + out[0] = out[1] = out[2] = y[i]; + out[3] = 255; // not used if n==3 + out += n; + } + } else { + stbi_uc *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; // report original components, not output + 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; + 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; + r = decode_jpeg_header(&j, SCAN_type); + stbi__rewind(s); + return r; +} + +static int stbi__jpeg_info_raw(stbi__jpeg *j, int *x, int *y, int *comp) +{ + if (!decode_jpeg_header(j, SCAN_header)) { + stbi__rewind( j->s ); + return 0; + } + if (x) *x = j->s->img_x; + if (y) *y = j->s->img_y; + if (comp) *comp = j->s->img_n; + return 1; +} + +static int stbi__jpeg_info(stbi__context *s, int *x, int *y, int *comp) +{ + stbi__jpeg j; + j.s = s; + return stbi__jpeg_info_raw(&j, x, y, comp); +} + +// public domain zlib decode v0.2 Sean Barrett 2006-11-18 +// simple implementation +// - all input must be provided in an upfront buffer +// - all output is written to a single output buffer (can malloc/realloc) +// performance +// - fast huffman + +// fast-way is faster to check than jpeg huffman, but slow way is slower +#define STBI__ZFAST_BITS 9 // accelerate all cases in default tables +#define STBI__ZFAST_MASK ((1 << STBI__ZFAST_BITS) - 1) + +// zlib-style huffman encoding +// (jpegs packs from left, zlib from right, so can't share code) +typedef struct +{ + stbi__uint16 fast[1 << STBI__ZFAST_BITS]; + stbi__uint16 firstcode[16]; + int maxcode[17]; + stbi__uint16 firstsymbol[16]; + stbi_uc size[288]; + stbi__uint16 value[288]; +} stbi__zhuffman; + +stbi_inline static 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; +} + +stbi_inline static int stbi__bit_reverse(int v, int bits) +{ + STBI_ASSERT(bits <= 16); + // to bit reverse n bits, reverse 16 and shift + // e.g. 11 bits, bit reverse and shift away 5 + return stbi__bitreverse16(v) >> (16-bits); +} + +static int stbi__zbuild_huffman(stbi__zhuffman *z, stbi_uc *sizelist, int num) +{ + int i,k=0; + int code, next_code[16], sizes[17]; + + // DEFLATE spec for generating codes + memset(sizes, 0, sizeof(sizes)); + memset(z->fast, 255, sizeof(z->fast)); + for (i=0; i < num; ++i) + ++sizes[sizelist[i]]; + sizes[0] = 0; + for (i=1; i < 16; ++i) + STBI_ASSERT(sizes[i] <= (1 << i)); + code = 0; + for (i=1; i < 16; ++i) { + next_code[i] = code; + z->firstcode[i] = (stbi__uint16) code; + z->firstsymbol[i] = (stbi__uint16) k; + code = (code + sizes[i]); + if (sizes[i]) + if (code-1 >= (1 << i)) return stbi__err("bad codelengths","Corrupt JPEG"); + z->maxcode[i] = code << (16-i); // preshift for inner loop + code <<= 1; + k += sizes[i]; + } + z->maxcode[16] = 0x10000; // sentinel + for (i=0; i < num; ++i) { + int s = sizelist[i]; + if (s) { + int c = next_code[s] - z->firstcode[s] + z->firstsymbol[s]; + z->size [c] = (stbi_uc ) s; + z->value[c] = (stbi__uint16) i; + if (s <= STBI__ZFAST_BITS) { + int k = stbi__bit_reverse(next_code[s],s); + while (k < (1 << STBI__ZFAST_BITS)) { + z->fast[k] = (stbi__uint16) c; + k += (1 << s); + } + } + ++next_code[s]; + } + } + return 1; +} + +// zlib-from-memory implementation for PNG reading +// because PNG allows splitting the zlib stream arbitrarily, +// and it's annoying structurally to have PNG call ZLIB call PNG, +// we require PNG read all the IDATs and combine them into a single +// memory buffer + +typedef struct +{ + stbi_uc *zbuffer, *zbuffer_end; + int num_bits; + stbi__uint32 code_buffer; + + char *zout; + char *zout_start; + char *zout_end; + int z_expandable; + + stbi__zhuffman z_length, z_distance; +} stbi__zbuf; + +stbi_inline static stbi_uc 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); +} + +stbi_inline static 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; +} + +stbi_inline static int stbi__zhuffman_decode(stbi__zbuf *a, stbi__zhuffman *z) +{ + int b,s,k; + if (a->num_bits < 16) stbi__fill_bits(a); + b = z->fast[a->code_buffer & STBI__ZFAST_MASK]; + if (b < 0xffff) { + s = z->size[b]; + a->code_buffer >>= s; + a->num_bits -= s; + return z->value[b]; + } + + // not resolved by fast table, so compute it the slow way + // use jpeg approach, which requires MSbits at top + k = 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; // invalid code! + // code size is s, so: + 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 int stbi__zexpand(stbi__zbuf *z, int n) // need to make room for n bytes +{ + char *q; + int cur, limit; + 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) +{ + for(;;) { + int z = stbi__zhuffman_decode(a, &a->z_length); + if (z < 256) { + if (z < 0) return stbi__err("bad huffman code","Corrupt PNG"); // error in huffman codes + if (a->zout >= a->zout_end) if (!stbi__zexpand(a, 1)) return 0; + *a->zout++ = (char) z; + } else { + stbi_uc *p; + int len,dist; + if (z == 256) 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 (a->zout - a->zout_start < dist) return stbi__err("bad dist","Corrupt PNG"); + if (a->zout + len > a->zout_end) if (!stbi__zexpand(a, len)) return 0; + p = (stbi_uc *) (a->zout - dist); + while (len--) + *a->zout++ = *p++; + } + } +} + +static int stbi__compute_huffman_codes(stbi__zbuf *a) +{ + static stbi_uc 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; + stbi_uc lencodes[286+32+137];//padding for maximum single op + stbi_uc 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]] = (stbi_uc) 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); + STBI_ASSERT(c >= 0 && c < 19); + if (c < 16) + lencodes[n++] = (stbi_uc) 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) +{ + stbi_uc header[4]; + int len,nlen,k; + if (a->num_bits & 7) + stbi__zreceive(a, a->num_bits & 7); // discard + // drain the bit-packed data into header + k = 0; + while (a->num_bits > 0) { + header[k++] = (stbi_uc) (a->code_buffer & 255); // suppress MSVC run-time check + a->code_buffer >>= 8; + a->num_bits -= 8; + } + STBI_ASSERT(a->num_bits == 0); + // now fill header the normal way + 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, 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"); // zlib spec + if (flg & 32) return stbi__err("no preset dict","Corrupt PNG"); // preset dictionary not allowed in png + if (cm != 8) return stbi__err("bad compression","Corrupt PNG"); // DEFLATE required for png + // window = 1 << (8 + cinfo)... but who cares, we fully buffer output + return 1; +} + +// @TODO: should statically initialize these for optimal thread safety +static stbi_uc stbi__zdefault_length[288], stbi__zdefault_distance[32]; +static void stbi__init_zdefaults(void) +{ + int i; // use <= to match clearly with spec + 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) { + // use fixed code lengths + 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(const char *buffer, int len, int initial_size, int *outlen) +{ + stbi__zbuf a; + char *p = (char *) stbi__malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *) buffer; + a.zbuffer_end = (stbi_uc *) buffer + len; + if (stbi__do_zlib(&a, p, initial_size, 1, 1)) { + if (outlen) *outlen = (int) (a.zout - a.zout_start); + return a.zout_start; + } else { + free(a.zout_start); + return NULL; + } +} + +STBIDEF char *stbi_zlib_decode_malloc(char const *buffer, int len, int *outlen) +{ + return stbi_zlib_decode_malloc_guesssize(buffer, len, 16384, outlen); +} + +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 *) stbi__malloc(initial_size); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *) buffer; + a.zbuffer_end = (stbi_uc *) 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; + } +} + +STBIDEF int stbi_zlib_decode_buffer(char *obuffer, int olen, char const *ibuffer, int ilen) +{ + stbi__zbuf a; + a.zbuffer = (stbi_uc *) ibuffer; + a.zbuffer_end = (stbi_uc *) ibuffer + ilen; + if (stbi__do_zlib(&a, obuffer, olen, 0, 1)) + return (int) (a.zout - a.zout_start); + else + return -1; +} + +STBIDEF char *stbi_zlib_decode_noheader_malloc(char const *buffer, int len, int *outlen) +{ + stbi__zbuf a; + char *p = (char *) stbi__malloc(16384); + if (p == NULL) return NULL; + a.zbuffer = (stbi_uc *) buffer; + a.zbuffer_end = (stbi_uc *) buffer+len; + if (stbi__do_zlib(&a, p, 16384, 1, 0)) { + if (outlen) *outlen = (int) (a.zout - a.zout_start); + return a.zout_start; + } else { + free(a.zout_start); + return NULL; + } +} + +STBIDEF int stbi_zlib_decode_noheader_buffer(char *obuffer, int olen, const char *ibuffer, int ilen) +{ + stbi__zbuf a; + a.zbuffer = (stbi_uc *) ibuffer; + a.zbuffer_end = (stbi_uc *) ibuffer + ilen; + if (stbi__do_zlib(&a, obuffer, olen, 0, 0)) + return (int) (a.zout - a.zout_start); + else + return -1; +} + +// public domain "baseline" PNG decoder v0.10 Sean Barrett 2006-11-18 +// simple implementation +// - only 8-bit samples +// - no CRC checking +// - allocates lots of intermediate memory +// - avoids problem of streaming data between subsystems +// - avoids explicit window management +// performance +// - uses stb_zlib, a PD zlib implementation with fast huffman decoding + + +typedef struct +{ + stbi__uint32 length; + stbi__uint32 type; +} stbi__pngchunk; + +#define PNG_TYPE(a,b,c,d) (((a) << 24) + ((b) << 16) + ((c) << 8) + (d)) + +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 stbi_uc 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; + stbi_uc *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 stbi_uc 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; +} + +#define STBI__BYTECAST(x) ((stbi_uc) ((x) & 255)) // truncate int to byte without warnings + +// create the png data from post-deflated data +static int stbi__create_png_image_raw(stbi__png *a, stbi_uc *raw, stbi__uint32 raw_len, int out_n, stbi__uint32 x, stbi__uint32 y) +{ + stbi__context *s = a->s; + stbi__uint32 i,j,stride = x*out_n; + int k; + int img_n = s->img_n; // copy it into a local for later + STBI_ASSERT(out_n == s->img_n || out_n == s->img_n+1); + a->out = (stbi_uc *) stbi__malloc(x * y * out_n); + if (!a->out) return stbi__err("outofmem", "Out of memory"); + if (s->img_x == x && s->img_y == y) { + if (raw_len != (img_n * x + 1) * y) return stbi__err("not enough pixels","Corrupt PNG"); + } else { // interlaced: + if (raw_len < (img_n * x + 1) * y) return stbi__err("not enough pixels","Corrupt PNG"); + } + for (j=0; j < y; ++j) { + stbi_uc *cur = a->out + stride*j; + stbi_uc *prior = cur - stride; + int filter = *raw++; + if (filter > 4) return stbi__err("invalid filter","Corrupt PNG"); + // if first row, use special filter that doesn't sample previous row + if (j == 0) filter = first_row_filter[filter]; + // handle first pixel explicitly + for (k=0; k < img_n; ++k) { + switch (filter) { + case 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 (img_n != out_n) cur[img_n] = 255; + raw += img_n; + cur += out_n; + prior += out_n; + // this is a little gross, so that we don't switch per-pixel or per-component + if (img_n == out_n) { + #define CASE(f) \ + case f: \ + for (i=x-1; i >= 1; --i, raw+=img_n,cur+=img_n,prior+=img_n) \ + for (k=0; k < img_n; ++k) + switch (filter) { + CASE(STBI__F_none) cur[k] = raw[k]; break; + CASE(STBI__F_sub) cur[k] = STBI__BYTECAST(raw[k] + cur[k-img_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-img_n])>>1)); break; + CASE(STBI__F_paeth) cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-img_n],prior[k],prior[k-img_n])); break; + CASE(STBI__F_avg_first) cur[k] = STBI__BYTECAST(raw[k] + (cur[k-img_n] >> 1)); break; + CASE(STBI__F_paeth_first) cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-img_n],0,0)); break; + } + #undef CASE + } 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 + } + } + return 1; +} + +static int stbi__create_png_image(stbi__png *a, stbi_uc *raw, stbi__uint32 raw_len, int out_n, int interlaced) +{ + stbi_uc *final; + int p; + if (!interlaced) + return stbi__create_png_image_raw(a, raw, raw_len, out_n, a->s->img_x, a->s->img_y); + + // de-interlacing + final = (stbi_uc *) stbi__malloc(a->s->img_x * a->s->img_y * out_n); + for (p=0; p < 7; ++p) { + int xorig[] = { 0,4,0,2,0,1,0 }; + int yorig[] = { 0,0,4,0,2,0,1 }; + int xspc[] = { 8,8,4,4,2,2,1 }; + int yspc[] = { 8,8,8,4,4,2,2 }; + int i,j,x,y; + // pass1_x[4] = 0, pass1_x[5] = 1, pass1_x[12] = 1 + x = (a->s->img_x - xorig[p] + xspc[p]-1) / xspc[p]; + y = (a->s->img_y - yorig[p] + yspc[p]-1) / yspc[p]; + if (x && y) { + if (!stbi__create_png_image_raw(a, raw, raw_len, out_n, x, y)) { + free(final); + return 0; + } + for (j=0; j < y; ++j) + for (i=0; i < x; ++i) + memcpy(final + (j*yspc[p]+yorig[p])*a->s->img_x*out_n + (i*xspc[p]+xorig[p])*out_n, + a->out + (j*x+i)*out_n, out_n); + free(a->out); + raw += (x*out_n+1)*y; + raw_len -= (x*out_n+1)*y; + } + } + a->out = final; + + return 1; +} + +static int stbi__compute_transparency(stbi__png *z, stbi_uc tc[3], int out_n) +{ + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi_uc *p = z->out; + + // compute color-based transparency, assuming we've + // already got 255 as the alpha value in the output + 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, stbi_uc *palette, int len, int pal_img_n) +{ + stbi__uint32 i, pixel_count = a->s->img_x * a->s->img_y; + stbi_uc *p, *temp_out, *orig = a->out; + + p = (stbi_uc *) stbi__malloc(pixel_count * pal_img_n); + if (p == NULL) return stbi__err("outofmem", "Out of memory"); + + // between here and free(out) below, exitting would leak + temp_out = p; + + if (pal_img_n == 3) { + for (i=0; i < pixel_count; ++i) { + int n = orig[i]*4; + p[0] = palette[n ]; + p[1] = palette[n+1]; + p[2] = palette[n+2]; + p += 3; + } + } else { + for (i=0; i < pixel_count; ++i) { + int n = orig[i]*4; + p[0] = palette[n ]; + p[1] = palette[n+1]; + p[2] = palette[n+2]; + p[3] = palette[n+3]; + p += 4; + } + } + free(a->out); + a->out = temp_out; + + STBI_NOTUSED(len); + + return 1; +} + +static int stbi__unpremultiply_on_load = 0; +static int stbi__de_iphone_flag = 0; + +STBIDEF void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply) +{ + stbi__unpremultiply_on_load = flag_true_if_should_unpremultiply; +} + +STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert) +{ + stbi__de_iphone_flag = flag_true_if_should_convert; +} + +static void stbi__de_iphone(stbi__png *z) +{ + stbi__context *s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi_uc *p = z->out; + + if (s->img_out_n == 3) { // convert bgr to rgb + for (i=0; i < pixel_count; ++i) { + stbi_uc 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) { + // convert bgr to rgb and unpremultiply + for (i=0; i < pixel_count; ++i) { + stbi_uc a = p[3]; + stbi_uc t = p[0]; + if (a) { + p[0] = p[2] * 255 / a; + p[1] = p[1] * 255 / a; + p[2] = t * 255 / a; + } else { + p[0] = p[2]; + p[2] = t; + } + p += 4; + } + } else { + // convert bgr to rgb + for (i=0; i < pixel_count; ++i) { + stbi_uc t = p[0]; + p[0] = p[2]; + p[2] = t; + p += 4; + } + } + } +} + +static int stbi__parse_png_file(stbi__png *z, int scan, int req_comp) +{ + stbi_uc palette[1024], pal_img_n=0; + stbi_uc has_trans=0, tc[3]; + stbi__uint32 ioff=0, idata_limit=0, i, pal_len=0; + int first=1,k,interlace=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 == SCAN_type) return 1; + + for (;;) { + stbi__pngchunk c = stbi__get_chunk_header(s); + switch (c.type) { + case PNG_TYPE('C','g','B','I'): + is_iphone = 1; + stbi__skip(s, c.length); + break; + case PNG_TYPE('I','H','D','R'): { + int depth,color,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 != 8) return stbi__err("8bit only","PNG not supported: 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 == SCAN_header) return 1; + } else { + // if paletted, then pal_n is our final components, and + // img_n is # components to decompress/filter. + s->img_n = 1; + if ((1 << 30) / s->img_x / 4 < s->img_y) return stbi__err("too large","Corrupt PNG"); + // if SCAN_header, have to scan to see if we have a tRNS + } + break; + } + + case PNG_TYPE('P','L','T','E'): { + if (first) return 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 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 == 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 != (stbi__uint32) 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] = (stbi_uc) (stbi__get16be(s) & 255); // non 8-bit images will be larger + } + break; + } + + case 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 == SCAN_header) { s->img_n = pal_img_n; return 1; } + if (ioff + c.length > idata_limit) { + stbi_uc *p; + if (idata_limit == 0) idata_limit = c.length > 4096 ? c.length : 4096; + while (ioff + c.length > idata_limit) + idata_limit *= 2; + p = (stbi_uc *) 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 PNG_TYPE('I','E','N','D'): { + stbi__uint32 raw_len; + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (scan != SCAN_load) return 1; + if (z->idata == NULL) return stbi__err("no IDAT","Corrupt PNG"); + z->expanded = (stbi_uc *) stbi_zlib_decode_malloc_guesssize_headerflag((char *) z->idata, ioff, 16384, (int *) &raw_len, !is_iphone); + if (z->expanded == NULL) return 0; // zlib should set error + free(z->idata); z->idata = NULL; + if ((req_comp == s->img_n+1 && req_comp != 3 && !pal_img_n) || has_trans) + s->img_out_n = s->img_n+1; + else + s->img_out_n = s->img_n; + if (!stbi__create_png_image(z, z->expanded, raw_len, s->img_out_n, 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) { + // pal_img_n == 3 or 4 + s->img_n = pal_img_n; // record the actual colors we had + s->img_out_n = pal_img_n; + if (req_comp >= 3) s->img_out_n = req_comp; + if (!stbi__expand_png_palette(z, palette, pal_len, s->img_out_n)) + return 0; + } + free(z->expanded); z->expanded = NULL; + return 1; + } + + default: + // if critical, fail + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if ((c.type & (1 << 29)) == 0) { + #ifndef STBI_NO_FAILURE_STRINGS + // not threadsafe + 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; + } + // end of PNG chunk, read and skip CRC + 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, 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; +} + +static int stbi__png_info_raw(stbi__png *p, int *x, int *y, int *comp) +{ + if (!stbi__parse_png_file(p, SCAN_header, 0)) { + stbi__rewind( p->s ); + return 0; + } + if (x) *x = p->s->img_x; + if (y) *y = p->s->img_y; + if (comp) *comp = p->s->img_n; + return 1; +} + +static int stbi__png_info(stbi__context *s, int *x, int *y, int *comp) +{ + stbi__png p; + p.s = s; + return stbi__png_info_raw(&p, x, y, comp); +} + +// Microsoft/Windows BMP image +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); // discard filesize + stbi__get16le(s); // discard reserved + stbi__get16le(s); // discard reserved + stbi__get32le(s); // discard data offset + 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; +} + + +// returns 0..31 for the highest set bit +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); // max 2 + a = (a & 0x33333333) + ((a >> 2) & 0x33333333); // max 4 + a = (a + (a >> 4)) & 0x0f0f0f0f; // max 8 per 4, now 8 bits + a = (a + (a >> 8)); // max 16 per 8 bits + a = (a + (a >> 16)); // max 32 per 8 bits + return a & 0xff; +} + +static int 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 stbi_uc *stbi__bmp_load(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + stbi_uc *out; + unsigned int mr=0,mg=0,mb=0,ma=0, fake_a=0; + stbi_uc 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); // discard filesize + stbi__get16le(s); // discard reserved + stbi__get16le(s); // discard reserved + 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); // discard sizeof + stbi__get32le(s); // discard hres + stbi__get32le(s); // discard vres + stbi__get32le(s); // discard colorsused + stbi__get32le(s); // discard max important + 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; // @TODO: check for cases like alpha value is all 0 and switch it to 255 + 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); + // not documented, but generated by photoshop and handled by mspaint + 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); // discard color space + for (i=0; i < 12; ++i) + stbi__get32le(s); // discard color space parameters + if (hsz == 124) { + stbi__get32le(s); // discard rendering intent + stbi__get32le(s); // discard offset of profile data + stbi__get32le(s); // discard size of profile data + stbi__get32le(s); // discard reserved + } + } + if (bpp < 16) + psize = (offset - 14 - hsz) >> 2; + } + s->img_n = ma ? 4 : 3; + if (req_comp && req_comp >= 3) // we can directly decode 3 or 4 + target = req_comp; + else + target = s->img_n; // if they want monochrome, we'll post-convert + out = (stbi_uc *) stbi__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"); } + // right shift amt to put high bit in position #7 + 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) { + stbi__uint32 v = (stbi__uint32) (bpp == 16 ? 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) { + stbi_uc t; + for (j=0; j < (int) s->img_y>>1; ++j) { + stbi_uc *p1 = out + j *s->img_x*target; + stbi_uc *p2 = out + (s->img_y-1-j)*s->img_x*target; + for (i=0; i < (int) s->img_x*target; ++i) { + t = p1[i], p1[i] = p2[i], p2[i] = t; + } + } + } + + if (req_comp && req_comp != target) { + out = stbi__convert_format(out, target, req_comp, s->img_x, s->img_y); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + + *x = s->img_x; + *y = s->img_y; + if (comp) *comp = s->img_n; + return out; +} + +// Targa Truevision - TGA +// by Jonathan Dummer + +static int stbi__tga_info(stbi__context *s, int *x, int *y, int *comp) +{ + int tga_w, tga_h, tga_comp; + int sz; + stbi__get8(s); // discard Offset + sz = stbi__get8(s); // color type + if( sz > 1 ) { + stbi__rewind(s); + return 0; // only RGB or indexed allowed + } + sz = stbi__get8(s); // image type + // only RGB or grey allowed, +/- RLE + if ((sz != 1) && (sz != 2) && (sz != 3) && (sz != 9) && (sz != 10) && (sz != 11)) return 0; + stbi__skip(s,9); + tga_w = stbi__get16le(s); + if( tga_w < 1 ) { + stbi__rewind(s); + return 0; // test width + } + tga_h = stbi__get16le(s); + if( tga_h < 1 ) { + stbi__rewind(s); + return 0; // test height + } + sz = stbi__get8(s); // bits per pixel + // only RGB or RGBA or grey allowed + if ((sz != 8) && (sz != 16) && (sz != 24) && (sz != 32)) { + stbi__rewind(s); + return 0; + } + tga_comp = sz; + if (x) *x = tga_w; + if (y) *y = tga_h; + if (comp) *comp = tga_comp / 8; + return 1; // seems to have passed everything +} + +static int stbi__tga_test(stbi__context *s) +{ + int res; + int sz; + stbi__get8(s); // discard Offset + sz = stbi__get8(s); // color type + if ( sz > 1 ) return 0; // only RGB or indexed allowed + sz = stbi__get8(s); // image type + if ( (sz != 1) && (sz != 2) && (sz != 3) && (sz != 9) && (sz != 10) && (sz != 11) ) return 0; // only RGB or grey allowed, +/- RLE + stbi__get16be(s); // discard palette start + stbi__get16be(s); // discard palette length + stbi__get8(s); // discard bits per palette color entry + stbi__get16be(s); // discard x origin + stbi__get16be(s); // discard y origin + if ( stbi__get16be(s) < 1 ) return 0; // test width + if ( stbi__get16be(s) < 1 ) return 0; // test height + sz = stbi__get8(s); // bits per pixel + if ( (sz != 8) && (sz != 16) && (sz != 24) && (sz != 32) ) + res = 0; + else + res = 1; + stbi__rewind(s); + return res; +} + +static stbi_uc *stbi__tga_load(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + // read in the TGA header stuff + 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); + // image data + unsigned char *tga_data; + unsigned char *tga_palette = NULL; + int i, j; + unsigned char raw_data[4]; + int RLE_count = 0; + int RLE_repeating = 0; + int read_next_pixel = 1; + + // do a tiny bit of precessing + if ( tga_image_type >= 8 ) + { + tga_image_type -= 8; + tga_is_RLE = 1; + } + /* int tga_alpha_bits = tga_inverted & 15; */ + tga_inverted = 1 - ((tga_inverted >> 5) & 1); + + // error check + if ( //(tga_indexed) || + (tga_width < 1) || (tga_height < 1) || + (tga_image_type < 1) || (tga_image_type > 3) || + ((tga_bits_per_pixel != 8) && (tga_bits_per_pixel != 16) && + (tga_bits_per_pixel != 24) && (tga_bits_per_pixel != 32)) + ) + { + return NULL; // we don't report this as a bad TGA because we don't even know if it's TGA + } + + // If I'm paletted, then I'll use the number of bits from the palette + if ( tga_indexed ) + { + tga_comp = tga_palette_bits / 8; + } + + // tga info + *x = tga_width; + *y = tga_height; + if (comp) *comp = tga_comp; + + tga_data = (unsigned char*)stbi__malloc( tga_width * tga_height * tga_comp ); + if (!tga_data) return stbi__errpuc("outofmem", "Out of memory"); + + // skip to the data's starting position (offset usually = 0) + 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; + stbi_uc *tga_row = tga_data + y*tga_width*tga_comp; + stbi__getn(s, tga_row, tga_width * tga_comp); + } + } else { + // do I need to load a palette? + if ( tga_indexed) + { + // any data to skip? (offset usually = 0) + stbi__skip(s, tga_palette_start ); + // load the palette + tga_palette = (unsigned char*)stbi__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"); + } + } + // load the data + for (i=0; i < tga_width * tga_height; ++i) + { + // if I'm in RLE mode, do I need to get a RLE stbi__pngchunk? + if ( tga_is_RLE ) + { + if ( RLE_count == 0 ) + { + // yep, get the next byte as a RLE command + 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; + } + // OK, if I need to read a pixel, do it now + if ( read_next_pixel ) + { + // load however much data we did have + if ( tga_indexed ) + { + // read in 1 byte, then perform the lookup + int pal_idx = stbi__get8(s); + if ( pal_idx >= tga_palette_len ) + { + // invalid index + pal_idx = 0; + } + pal_idx *= tga_bits_per_pixel / 8; + for (j = 0; j*8 < tga_bits_per_pixel; ++j) + { + raw_data[j] = tga_palette[pal_idx+j]; + } + } else + { + // read in the data raw + for (j = 0; j*8 < tga_bits_per_pixel; ++j) + { + raw_data[j] = stbi__get8(s); + } + } + // clear the reading flag for the next pixel + read_next_pixel = 0; + } // end of reading a pixel + + // copy data + for (j = 0; j < tga_comp; ++j) + tga_data[i*tga_comp+j] = raw_data[j]; + + // in case we're in RLE mode, keep counting down + --RLE_count; + } + // do I need to invert the image? + if ( tga_inverted ) + { + for (j = 0; j*2 < tga_height; ++j) + { + int index1 = j * tga_width * 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; + } + } + } + // clear my palette, if I had one + if ( tga_palette != NULL ) + { + free( tga_palette ); + } + } + + // swap RGB + 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; + } + } + + // convert to target component count + if (req_comp && req_comp != tga_comp) + tga_data = stbi__convert_format(tga_data, tga_comp, req_comp, tga_width, tga_height); + + // the things I do to get rid of an error message, and yet keep + // Microsoft's C compilers happy... [8^( + tga_palette_start = tga_palette_len = tga_palette_bits = + tga_x_origin = tga_y_origin = 0; + // OK, done + return tga_data; +} + +// ************************************************************************************************* +// Photoshop PSD loader -- PD by Thatcher Ulrich, integration by Nicolas Schulz, tweaked by STB + +static int stbi__psd_test(stbi__context *s) +{ + int r = (stbi__get32be(s) == 0x38425053); + stbi__rewind(s); + return r; +} + +static stbi_uc *stbi__psd_load(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + int pixelCount; + int channelCount, compression; + int channel, i, count, len; + int w,h; + stbi_uc *out; + + // Check identifier + if (stbi__get32be(s) != 0x38425053) // "8BPS" + return stbi__errpuc("not PSD", "Corrupt PSD image"); + + // Check file type version. + if (stbi__get16be(s) != 1) + return stbi__errpuc("wrong version", "Unsupported version of PSD image"); + + // Skip 6 reserved bytes. + stbi__skip(s, 6 ); + + // Read the number of channels (R, G, B, A, etc). + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) + return stbi__errpuc("wrong channel count", "Unsupported number of channels in PSD image"); + + // Read the rows and columns of the image. + h = stbi__get32be(s); + w = stbi__get32be(s); + + // Make sure the depth is 8 bits. + if (stbi__get16be(s) != 8) + return stbi__errpuc("unsupported bit depth", "PSD bit depth is not 8 bit"); + + // Make sure the color mode is RGB. + // Valid options are: + // 0: Bitmap + // 1: Grayscale + // 2: Indexed color + // 3: RGB color + // 4: CMYK color + // 7: Multichannel + // 8: Duotone + // 9: Lab color + if (stbi__get16be(s) != 3) + return stbi__errpuc("wrong color format", "PSD is not in RGB color format"); + + // Skip the Mode Data. (It's the palette for indexed color; other info for other modes.) + stbi__skip(s,stbi__get32be(s) ); + + // Skip the image resources. (resolution, pen tool paths, etc) + stbi__skip(s, stbi__get32be(s) ); + + // Skip the reserved data. + stbi__skip(s, stbi__get32be(s) ); + + // Find out if the data is compressed. + // Known values: + // 0: no compression + // 1: RLE compressed + compression = stbi__get16be(s); + if (compression > 1) + return stbi__errpuc("bad compression", "PSD has an unknown compression format"); + + // Create the destination image. + out = (stbi_uc *) stbi__malloc(4 * w*h); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + pixelCount = w*h; + + // Initialize the data to zero. + //memset( out, 0, pixelCount * 4 ); + + // Finally, the image data. + if (compression) { + // RLE as used by .PSD and .TIFF + // Loop until you get the number of unpacked bytes you are expecting: + // Read the next source byte into n. + // If n is between 0 and 127 inclusive, copy the next n+1 bytes literally. + // Else if n is between -127 and -1 inclusive, copy the next byte -n+1 times. + // Else if n is 128, noop. + // Endloop + + // The RLE-compressed data is preceeded by a 2-byte data count for each row in the data, + // which we're going to just skip. + stbi__skip(s, h * channelCount * 2 ); + + // Read the RLE data by channel. + for (channel = 0; channel < 4; channel++) { + stbi_uc *p; + + p = out+channel; + if (channel >= channelCount) { + // Fill this channel with default data. + for (i = 0; i < pixelCount; i++) *p = (channel == 3 ? 255 : 0), p += 4; + } else { + // Read the RLE data. + count = 0; + while (count < pixelCount) { + len = stbi__get8(s); + if (len == 128) { + // No-op. + } else if (len < 128) { + // Copy next len+1 bytes literally. + len++; + count += len; + while (len) { + *p = stbi__get8(s); + p += 4; + len--; + } + } else if (len > 128) { + stbi_uc val; + // Next -len+1 bytes in the dest are replicated from next source byte. + // (Interpret len as a negative 8-bit int.) + len ^= 0x0FF; + len += 2; + val = stbi__get8(s); + count += len; + while (len) { + *p = val; + p += 4; + len--; + } + } + } + } + } + + } else { + // We're at the raw image data. It's each channel in order (Red, Green, Blue, Alpha, ...) + // where each channel consists of an 8-bit value for each pixel in the image. + + // Read the data by channel. + for (channel = 0; channel < 4; channel++) { + stbi_uc *p; + + p = out + channel; + if (channel > channelCount) { + // Fill this channel with default data. + for (i = 0; i < pixelCount; i++) *p = channel == 3 ? 255 : 0, p += 4; + } else { + // Read the data. + for (i = 0; i < pixelCount; i++) + *p = stbi__get8(s), p += 4; + } + } + } + + if (req_comp && req_comp != 4) { + out = stbi__convert_format(out, 4, req_comp, w, h); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + + if (comp) *comp = channelCount; + *y = h; + *x = w; + + return out; +} + +// ************************************************************************************************* +// Softimage PIC loader +// by Tom Seddon +// +// See http://softimage.wiki.softimage.com/index.php/INFO:_PIC_file_format +// See http://ozviz.wasp.uwa.edu.au/~pbourke/dataformats/softimagepic/ + +static int stbi__pic_is4(stbi__context *s,const char *str) +{ + int i; + for (i=0; i<4; ++i) + if (stbi__get8(s) != (stbi_uc)str[i]) + return 0; + + return 1; +} + +static int stbi__pic_test_core(stbi__context *s) +{ + int i; + + if (!stbi__pic_is4(s,"\x53\x80\xF6\x34")) + return 0; + + for(i=0;i<84;++i) + stbi__get8(s); + + if (!stbi__pic_is4(s,"PICT")) + return 0; + + return 1; +} + +typedef struct +{ + stbi_uc size,type,channel; +} stbi__pic_packet; + +static stbi_uc *stbi__readval(stbi__context *s, int channel, stbi_uc *dest) +{ + int mask=0x80, i; + + for (i=0; i<4; ++i, mask>>=1) { + if (channel & mask) { + if (stbi__at_eof(s)) return stbi__errpuc("bad file","PIC file too short"); + dest[i]=stbi__get8(s); + } + } + + return dest; +} + +static void stbi__copyval(int channel,stbi_uc *dest,const stbi_uc *src) +{ + int mask=0x80,i; + + for (i=0;i<4; ++i, mask>>=1) + if (channel&mask) + dest[i]=src[i]; +} + +static stbi_uc *stbi__pic_load_core(stbi__context *s,int width,int height,int *comp, stbi_uc *result) +{ + int act_comp=0,num_packets=0,y,chained; + stbi__pic_packet packets[10]; + + // this will (should...) cater for even some bizarre stuff like having data + // for the same channel in multiple packets. + do { + stbi__pic_packet *packet; + + if (num_packets==sizeof(packets)/sizeof(packets[0])) + return stbi__errpuc("bad format","too many packets"); + + packet = &packets[num_packets++]; + + chained = stbi__get8(s); + packet->size = stbi__get8(s); + packet->type = stbi__get8(s); + packet->channel = stbi__get8(s); + + act_comp |= packet->channel; + + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (reading packets)"); + if (packet->size != 8) return stbi__errpuc("bad format","packet isn't 8bpp"); + } while (chained); + + *comp = (act_comp & 0x10 ? 4 : 3); // has alpha channel? + + for(y=0; ytype) { + default: + return stbi__errpuc("bad format","packet has bad compression type"); + + case 0: {//uncompressed + int x; + + for(x=0;xchannel,dest)) + return 0; + break; + } + + case 1://Pure RLE + { + int left=width, i; + + while (left>0) { + stbi_uc count,value[4]; + + count=stbi__get8(s); + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (pure read count)"); + + if (count > left) + count = (stbi_uc) left; + + if (!stbi__readval(s,packet->channel,value)) return 0; + + for(i=0; ichannel,dest,value); + left -= count; + } + } + break; + + case 2: {//Mixed RLE + int left=width; + while (left>0) { + int count = stbi__get8(s), i; + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (mixed read count)"); + + if (count >= 128) { // Repeated + stbi_uc value[4]; + int i; + + if (count==128) + count = stbi__get16be(s); + else + count -= 127; + if (count > left) + return stbi__errpuc("bad file","scanline overrun"); + + if (!stbi__readval(s,packet->channel,value)) + return 0; + + for(i=0;ichannel,dest,value); + } else { // Raw + ++count; + if (count>left) return stbi__errpuc("bad file","scanline overrun"); + + for(i=0;ichannel,dest)) + return 0; + } + left-=count; + } + break; + } + } + } + } + + return result; +} + +static stbi_uc *stbi__pic_load(stbi__context *s,int *px,int *py,int *comp,int req_comp) +{ + stbi_uc *result; + int i, x,y; + + for (i=0; i<92; ++i) + stbi__get8(s); + + x = stbi__get16be(s); + y = stbi__get16be(s); + if (stbi__at_eof(s)) return stbi__errpuc("bad file","file too short (pic header)"); + if ((1 << 28) / x < y) return stbi__errpuc("too large", "Image too large to decode"); + + stbi__get32be(s); //skip `ratio' + stbi__get16be(s); //skip `fields' + stbi__get16be(s); //skip `pad' + + // intermediate buffer is RGBA + result = (stbi_uc *) stbi__malloc(x*y*4); + memset(result, 0xff, x*y*4); + + if (!stbi__pic_load_core(s,x,y,comp, result)) { + free(result); + result=0; + } + *px = x; + *py = y; + if (req_comp == 0) req_comp = *comp; + result=stbi__convert_format(result,4,req_comp,x,y); + + return result; +} + +static int stbi__pic_test(stbi__context *s) +{ + int r = stbi__pic_test_core(s); + stbi__rewind(s); + return r; +} + +// ************************************************************************************************* +// GIF loader -- public domain by Jean-Marc Lienher -- simplified/shrunk by stb +typedef struct +{ + stbi__int16 prefix; + stbi_uc first; + stbi_uc suffix; +} stbi__gif_lzw; + +typedef struct +{ + int w,h; + stbi_uc *out; // output buffer (always 4 components) + int flags, bgindex, ratio, transparent, eflags; + stbi_uc pal[256][4]; + stbi_uc lpal[256][4]; + stbi__gif_lzw codes[4096]; + stbi_uc *color_table; + int parse, step; + int lflags; + int start_x, start_y; + int max_x, max_y; + int cur_x, cur_y; + int line_size; +} stbi__gif; + +static int stbi__gif_test_raw(stbi__context *s) +{ + int sz; + if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || stbi__get8(s) != '8') return 0; + sz = stbi__get8(s); + if (sz != '9' && sz != '7') return 0; + if (stbi__get8(s) != 'a') return 0; + return 1; +} + +static int stbi__gif_test(stbi__context *s) +{ + int r = stbi__gif_test_raw(s); + stbi__rewind(s); + return r; +} + +static void stbi__gif_parse_colortable(stbi__context *s, stbi_uc pal[256][4], int num_entries, int transp) +{ + int i; + for (i=0; i < num_entries; ++i) { + pal[i][2] = stbi__get8(s); + pal[i][1] = stbi__get8(s); + pal[i][0] = stbi__get8(s); + pal[i][3] = transp ? 0 : 255; + } +} + +static int stbi__gif_header(stbi__context *s, stbi__gif *g, int *comp, int is_info) +{ + stbi_uc version; + if (stbi__get8(s) != 'G' || stbi__get8(s) != 'I' || stbi__get8(s) != 'F' || stbi__get8(s) != '8') + return stbi__err("not GIF", "Corrupt GIF"); + + version = stbi__get8(s); + if (version != '7' && version != '9') return stbi__err("not GIF", "Corrupt GIF"); + if (stbi__get8(s) != 'a') return stbi__err("not GIF", "Corrupt GIF"); + + stbi__g_failure_reason = ""; + g->w = stbi__get16le(s); + g->h = stbi__get16le(s); + g->flags = stbi__get8(s); + g->bgindex = stbi__get8(s); + g->ratio = stbi__get8(s); + g->transparent = -1; + + if (comp != 0) *comp = 4; // can't actually tell whether it's 3 or 4 until we parse the comments + + if (is_info) return 1; + + if (g->flags & 0x80) + stbi__gif_parse_colortable(s,g->pal, 2 << (g->flags & 7), -1); + + return 1; +} + +static int stbi__gif_info_raw(stbi__context *s, int *x, int *y, int *comp) +{ + stbi__gif g; + if (!stbi__gif_header(s, &g, comp, 1)) { + stbi__rewind( s ); + return 0; + } + if (x) *x = g.w; + if (y) *y = g.h; + return 1; +} + +static void stbi__out_gif_code(stbi__gif *g, stbi__uint16 code) +{ + stbi_uc *p, *c; + + // recurse to decode the prefixes, since the linked-list is backwards, + // and working backwards through an interleaved image would be nasty + if (g->codes[code].prefix >= 0) + stbi__out_gif_code(g, g->codes[code].prefix); + + if (g->cur_y >= g->max_y) return; + + p = &g->out[g->cur_x + g->cur_y]; + c = &g->color_table[g->codes[code].suffix * 4]; + + if (c[3] >= 128) { + p[0] = c[2]; + p[1] = c[1]; + p[2] = c[0]; + p[3] = c[3]; + } + g->cur_x += 4; + + if (g->cur_x >= g->max_x) { + g->cur_x = g->start_x; + g->cur_y += g->step; + + while (g->cur_y >= g->max_y && g->parse > 0) { + g->step = (1 << g->parse) * g->line_size; + g->cur_y = g->start_y + (g->step >> 1); + --g->parse; + } + } +} + +static stbi_uc *stbi__process_gif_raster(stbi__context *s, stbi__gif *g) +{ + stbi_uc lzw_cs; + stbi__int32 len, code; + stbi__uint32 first; + stbi__int32 codesize, codemask, avail, oldcode, bits, valid_bits, clear; + stbi__gif_lzw *p; + + lzw_cs = stbi__get8(s); + clear = 1 << lzw_cs; + first = 1; + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + bits = 0; + valid_bits = 0; + for (code = 0; code < clear; code++) { + g->codes[code].prefix = -1; + g->codes[code].first = (stbi_uc) code; + g->codes[code].suffix = (stbi_uc) code; + } + + // support no starting clear code + avail = clear+2; + oldcode = -1; + + len = 0; + for(;;) { + if (valid_bits < codesize) { + if (len == 0) { + len = stbi__get8(s); // start new block + if (len == 0) + return g->out; + } + --len; + bits |= (stbi__int32) stbi__get8(s) << valid_bits; + valid_bits += 8; + } else { + stbi__int32 code = bits & codemask; + bits >>= codesize; + valid_bits -= codesize; + // @OPTIMIZE: is there some way we can accelerate the non-clear path? + if (code == clear) { // clear code + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + avail = clear + 2; + oldcode = -1; + first = 0; + } else if (code == clear + 1) { // end of stream code + stbi__skip(s, len); + while ((len = stbi__get8(s)) > 0) + stbi__skip(s,len); + return g->out; + } else if (code <= avail) { + if (first) return stbi__errpuc("no clear code", "Corrupt GIF"); + + if (oldcode >= 0) { + p = &g->codes[avail++]; + if (avail > 4096) return stbi__errpuc("too many codes", "Corrupt GIF"); + p->prefix = (stbi__int16) oldcode; + p->first = g->codes[oldcode].first; + p->suffix = (code == avail) ? p->first : g->codes[code].first; + } else if (code == avail) + return stbi__errpuc("illegal code in raster", "Corrupt GIF"); + + stbi__out_gif_code(g, (stbi__uint16) code); + + if ((avail & codemask) == 0 && avail <= 0x0FFF) { + codesize++; + codemask = (1 << codesize) - 1; + } + + oldcode = code; + } else { + return stbi__errpuc("illegal code in raster", "Corrupt GIF"); + } + } + } +} + +static void stbi__fill_gif_background(stbi__gif *g) +{ + int i; + stbi_uc *c = g->pal[g->bgindex]; + // @OPTIMIZE: write a dword at a time + for (i = 0; i < g->w * g->h * 4; i += 4) { + stbi_uc *p = &g->out[i]; + p[0] = c[2]; + p[1] = c[1]; + p[2] = c[0]; + p[3] = c[3]; + } +} + +// this function is designed to support animated gifs, although stb_image doesn't support it +static stbi_uc *stbi__gif_load_next(stbi__context *s, stbi__gif *g, int *comp, int req_comp) +{ + int i; + stbi_uc *old_out = 0; + + if (g->out == 0) { + if (!stbi__gif_header(s, g, comp,0)) return 0; // stbi__g_failure_reason set by stbi__gif_header + g->out = (stbi_uc *) stbi__malloc(4 * g->w * g->h); + if (g->out == 0) return stbi__errpuc("outofmem", "Out of memory"); + stbi__fill_gif_background(g); + } else { + // animated-gif-only path + if (((g->eflags & 0x1C) >> 2) == 3) { + old_out = g->out; + g->out = (stbi_uc *) stbi__malloc(4 * g->w * g->h); + if (g->out == 0) return stbi__errpuc("outofmem", "Out of memory"); + memcpy(g->out, old_out, g->w*g->h*4); + } + } + + for (;;) { + switch (stbi__get8(s)) { + case 0x2C: /* Image Descriptor */ + { + stbi__int32 x, y, w, h; + stbi_uc *o; + + x = stbi__get16le(s); + y = stbi__get16le(s); + w = stbi__get16le(s); + h = stbi__get16le(s); + if (((x + w) > (g->w)) || ((y + h) > (g->h))) + return stbi__errpuc("bad Image Descriptor", "Corrupt GIF"); + + g->line_size = g->w * 4; + g->start_x = x * 4; + g->start_y = y * g->line_size; + g->max_x = g->start_x + w * 4; + g->max_y = g->start_y + h * g->line_size; + g->cur_x = g->start_x; + g->cur_y = g->start_y; + + g->lflags = stbi__get8(s); + + if (g->lflags & 0x40) { + g->step = 8 * g->line_size; // first interlaced spacing + g->parse = 3; + } else { + g->step = g->line_size; + g->parse = 0; + } + + if (g->lflags & 0x80) { + stbi__gif_parse_colortable(s,g->lpal, 2 << (g->lflags & 7), g->eflags & 0x01 ? g->transparent : -1); + g->color_table = (stbi_uc *) g->lpal; + } else if (g->flags & 0x80) { + for (i=0; i < 256; ++i) // @OPTIMIZE: stbi__jpeg_reset only the previous transparent + g->pal[i][3] = 255; + if (g->transparent >= 0 && (g->eflags & 0x01)) + g->pal[g->transparent][3] = 0; + g->color_table = (stbi_uc *) g->pal; + } else + return stbi__errpuc("missing color table", "Corrupt GIF"); + + o = stbi__process_gif_raster(s, g); + if (o == NULL) return NULL; + + if (req_comp && req_comp != 4) + o = stbi__convert_format(o, 4, req_comp, g->w, g->h); + return o; + } + + case 0x21: // Comment Extension. + { + int len; + if (stbi__get8(s) == 0xF9) { // Graphic Control Extension. + len = stbi__get8(s); + if (len == 4) { + g->eflags = stbi__get8(s); + stbi__get16le(s); // delay + g->transparent = stbi__get8(s); + } else { + stbi__skip(s, len); + break; + } + } + while ((len = stbi__get8(s)) != 0) + stbi__skip(s, len); + break; + } + + case 0x3B: // gif stream termination code + return (stbi_uc *) s; // using '1' causes warning on some compilers + + default: + return stbi__errpuc("unknown code", "Corrupt GIF"); + } + } +} + +static stbi_uc *stbi__gif_load(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + stbi_uc *u = 0; + stbi__gif g; + memset(&g, 0, sizeof(g)); + + u = stbi__gif_load_next(s, &g, comp, req_comp); + if (u == (stbi_uc *) s) u = 0; // end of animated gif marker + if (u) { + *x = g.w; + *y = g.h; + } + + return u; +} + +static int stbi__gif_info(stbi__context *s, int *x, int *y, int *comp) +{ + return stbi__gif_info_raw(s,x,y,comp); +} + + +// ************************************************************************************************* +// Radiance RGBE HDR loader +// originally by Nicolas Schulz +#ifndef STBI_NO_HDR +static int stbi__hdr_test_core(stbi__context *s) +{ + const char *signature = "#?RADIANCE\n"; + int i; + for (i=0; signature[i]; ++i) + if (stbi__get8(s) != signature[i]) + return 0; + return 1; +} + +static int stbi__hdr_test(stbi__context* s) +{ + int r = stbi__hdr_test_core(s); + stbi__rewind(s); + return r; +} + +#define STBI__HDR_BUFLEN 1024 +static char *stbi__hdr_gettoken(stbi__context *z, char *buffer) +{ + int len=0; + char c = '\0'; + + c = (char) stbi__get8(z); + + while (!stbi__at_eof(z) && c != '\n') { + buffer[len++] = c; + if (len == STBI__HDR_BUFLEN-1) { + // flush to end of line + while (!stbi__at_eof(z) && stbi__get8(z) != '\n') + ; + break; + } + c = (char) stbi__get8(z); + } + + buffer[len] = 0; + return buffer; +} + +static void stbi__hdr_convert(float *output, stbi_uc *input, int req_comp) +{ + if ( input[3] != 0 ) { + float f1; + // Exponent + f1 = (float) ldexp(1.0f, input[3] - (int)(128 + 8)); + if (req_comp <= 2) + output[0] = (input[0] + input[1] + input[2]) * f1 / 3; + else { + output[0] = input[0] * f1; + output[1] = input[1] * f1; + output[2] = input[2] * f1; + } + if (req_comp == 2) output[1] = 1; + if (req_comp == 4) output[3] = 1; + } else { + switch (req_comp) { + case 4: output[3] = 1; /* fallthrough */ + case 3: output[0] = output[1] = output[2] = 0; + break; + case 2: output[1] = 1; /* fallthrough */ + case 1: output[0] = 0; + break; + } + } +} + +static float *stbi__hdr_load(stbi__context *s, int *x, int *y, int *comp, int req_comp) +{ + char buffer[STBI__HDR_BUFLEN]; + char *token; + int valid = 0; + int width, height; + stbi_uc *scanline; + float *hdr_data; + int len; + unsigned char count, value; + int i, j, k, c1,c2, z; + + + // Check identifier + if (strcmp(stbi__hdr_gettoken(s,buffer), "#?RADIANCE") != 0) + return stbi__errpf("not HDR", "Corrupt HDR image"); + + // Parse header + for(;;) { + token = stbi__hdr_gettoken(s,buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) return stbi__errpf("unsupported format", "Unsupported HDR format"); + + // Parse width and height + // can't use sscanf() if we're not using stdio! + token = stbi__hdr_gettoken(s,buffer); + if (strncmp(token, "-Y ", 3)) return stbi__errpf("unsupported data layout", "Unsupported HDR format"); + token += 3; + height = (int) strtol(token, &token, 10); + while (*token == ' ') ++token; + if (strncmp(token, "+X ", 3)) return stbi__errpf("unsupported data layout", "Unsupported HDR format"); + token += 3; + width = (int) strtol(token, NULL, 10); + + *x = width; + *y = height; + + if (comp) *comp = 3; + if (req_comp == 0) req_comp = 3; + + // Read data + hdr_data = (float *) stbi__malloc(height * width * req_comp * sizeof(float)); + + // Load image data + // image data is stored as some number of sca + if ( width < 8 || width >= 32768) { + // Read flat data + for (j=0; j < height; ++j) { + for (i=0; i < width; ++i) { + stbi_uc rgbe[4]; + main_decode_loop: + stbi__getn(s, rgbe, 4); + stbi__hdr_convert(hdr_data + j * width * req_comp + i * req_comp, rgbe, req_comp); + } + } + } else { + // Read RLE-encoded data + scanline = NULL; + + for (j = 0; j < height; ++j) { + c1 = stbi__get8(s); + c2 = stbi__get8(s); + len = stbi__get8(s); + if (c1 != 2 || c2 != 2 || (len & 0x80)) { + // not run-length encoded, so we have to actually use THIS data as a decoded + // pixel (note this can't be a valid pixel--one of RGB must be >= 128) + stbi_uc rgbe[4]; + rgbe[0] = (stbi_uc) c1; + rgbe[1] = (stbi_uc) c2; + rgbe[2] = (stbi_uc) len; + rgbe[3] = (stbi_uc) stbi__get8(s); + stbi__hdr_convert(hdr_data, rgbe, req_comp); + i = 1; + j = 0; + free(scanline); + goto main_decode_loop; // yes, this makes no sense + } + len <<= 8; + len |= stbi__get8(s); + if (len != width) { free(hdr_data); free(scanline); return stbi__errpf("invalid decoded scanline length", "corrupt HDR"); } + if (scanline == NULL) scanline = (stbi_uc *) stbi__malloc(width * 4); + + for (k = 0; k < 4; ++k) { + i = 0; + while (i < width) { + count = stbi__get8(s); + if (count > 128) { + // Run + value = stbi__get8(s); + count -= 128; + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = value; + } else { + // Dump + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = stbi__get8(s); + } + } + } + for (i=0; i < width; ++i) + stbi__hdr_convert(hdr_data+(j*width + i)*req_comp, scanline + i*4, req_comp); + } + free(scanline); + } + + return hdr_data; +} + +static int stbi__hdr_info(stbi__context *s, int *x, int *y, int *comp) +{ + char buffer[STBI__HDR_BUFLEN]; + char *token; + int valid = 0; + + if (strcmp(stbi__hdr_gettoken(s,buffer), "#?RADIANCE") != 0) { + stbi__rewind( s ); + return 0; + } + + for(;;) { + token = stbi__hdr_gettoken(s,buffer); + if (token[0] == 0) break; + if (strcmp(token, "FORMAT=32-bit_rle_rgbe") == 0) valid = 1; + } + + if (!valid) { + stbi__rewind( s ); + return 0; + } + token = stbi__hdr_gettoken(s,buffer); + if (strncmp(token, "-Y ", 3)) { + stbi__rewind( s ); + return 0; + } + token += 3; + *y = (int) strtol(token, &token, 10); + while (*token == ' ') ++token; + if (strncmp(token, "+X ", 3)) { + stbi__rewind( s ); + return 0; + } + token += 3; + *x = (int) strtol(token, NULL, 10); + *comp = 3; + return 1; +} +#endif // STBI_NO_HDR + +static int stbi__bmp_info(stbi__context *s, int *x, int *y, int *comp) +{ + int hsz; + if (stbi__get8(s) != 'B' || stbi__get8(s) != 'M') { + stbi__rewind( s ); + return 0; + } + stbi__skip(s,12); + hsz = stbi__get32le(s); + if (hsz != 12 && hsz != 40 && hsz != 56 && hsz != 108 && hsz != 124) { + stbi__rewind( s ); + return 0; + } + if (hsz == 12) { + *x = stbi__get16le(s); + *y = stbi__get16le(s); + } else { + *x = stbi__get32le(s); + *y = stbi__get32le(s); + } + if (stbi__get16le(s) != 1) { + stbi__rewind( s ); + return 0; + } + *comp = stbi__get16le(s) / 8; + return 1; +} + +static int stbi__psd_info(stbi__context *s, int *x, int *y, int *comp) +{ + int channelCount; + if (stbi__get32be(s) != 0x38425053) { + stbi__rewind( s ); + return 0; + } + if (stbi__get16be(s) != 1) { + stbi__rewind( s ); + return 0; + } + stbi__skip(s, 6); + channelCount = stbi__get16be(s); + if (channelCount < 0 || channelCount > 16) { + stbi__rewind( s ); + return 0; + } + *y = stbi__get32be(s); + *x = stbi__get32be(s); + if (stbi__get16be(s) != 8) { + stbi__rewind( s ); + return 0; + } + if (stbi__get16be(s) != 3) { + stbi__rewind( s ); + return 0; + } + *comp = 4; + return 1; +} + +static int stbi__pic_info(stbi__context *s, int *x, int *y, int *comp) +{ + int act_comp=0,num_packets=0,chained; + stbi__pic_packet packets[10]; + + stbi__skip(s, 92); + + *x = stbi__get16be(s); + *y = stbi__get16be(s); + if (stbi__at_eof(s)) return 0; + if ( (*x) != 0 && (1 << 28) / (*x) < (*y)) { + stbi__rewind( s ); + return 0; + } + + stbi__skip(s, 8); + + do { + stbi__pic_packet *packet; + + if (num_packets==sizeof(packets)/sizeof(packets[0])) + return 0; + + packet = &packets[num_packets++]; + chained = stbi__get8(s); + packet->size = stbi__get8(s); + packet->type = stbi__get8(s); + packet->channel = stbi__get8(s); + act_comp |= packet->channel; + + if (stbi__at_eof(s)) { + stbi__rewind( s ); + return 0; + } + if (packet->size != 8) { + stbi__rewind( s ); + return 0; + } + } while (chained); + + *comp = (act_comp & 0x10 ? 4 : 3); + + return 1; +} + +static int stbi__info_main(stbi__context *s, int *x, int *y, int *comp) +{ + if (stbi__jpeg_info(s, x, y, comp)) + return 1; + if (stbi__png_info(s, x, y, comp)) + return 1; + if (stbi__gif_info(s, x, y, comp)) + return 1; + if (stbi__bmp_info(s, x, y, comp)) + return 1; + if (stbi__psd_info(s, x, y, comp)) + return 1; + if (stbi__pic_info(s, x, y, comp)) + return 1; + #ifndef STBI_NO_HDR + if (stbi__hdr_info(s, x, y, comp)) + return 1; + #endif + // test tga last because it's a crappy test! + if (stbi__tga_info(s, x, y, comp)) + return 1; + return stbi__err("unknown image type", "Image not of any known type, or corrupt"); +} + +#ifndef STBI_NO_STDIO +STBIDEF int stbi_info(char const *filename, int *x, int *y, int *comp) +{ + FILE *f = stbi__fopen(filename, "rb"); + int result; + if (!f) return stbi__err("can't fopen", "Unable to open file"); + result = stbi_info_from_file(f, x, y, comp); + fclose(f); + return result; +} + +STBIDEF int stbi_info_from_file(FILE *f, int *x, int *y, int *comp) +{ + int r; + stbi__context s; + long pos = ftell(f); + stbi__start_file(&s, f); + r = stbi__info_main(&s,x,y,comp); + fseek(f,pos,SEEK_SET); + return r; +} +#endif // !STBI_NO_STDIO + +STBIDEF int stbi_info_from_memory(stbi_uc const *buffer, int len, int *x, int *y, int *comp) +{ + stbi__context s; + stbi__start_mem(&s,buffer,len); + return stbi__info_main(&s,x,y,comp); +} + +STBIDEF int stbi_info_from_callbacks(stbi_io_callbacks const *c, void *user, int *x, int *y, int *comp) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks *) c, user); + return stbi__info_main(&s,x,y,comp); +} + +#endif // STB_IMAGE_IMPLEMENTATION + +/* + revision history: + 1.46 (2014-08-26) + fix broken tRNS chunk (colorkey-style transparency) in non-paletted PNG + 1.45 (2014-08-16) + fix MSVC-ARM internal compiler error by wrapping malloc + 1.44 (2014-08-07) + various warning fixes from Ronny Chevalier + 1.43 (2014-07-15) + fix MSVC-only compiler problem in code changed in 1.42 + 1.42 (2014-07-09) + don't define _CRT_SECURE_NO_WARNINGS (affects user code) + fixes to stbi__cleanup_jpeg path + added STBI_ASSERT to avoid requiring assert.h + 1.41 (2014-06-25) + fix search&replace from 1.36 that messed up comments/error messages + 1.40 (2014-06-22) + fix gcc struct-initialization warning + 1.39 (2014-06-15) + fix to TGA optimization when req_comp != number of components in TGA; + fix to GIF loading because BMP wasn't rewinding (whoops, no GIFs in my test suite) + add support for BMP version 5 (more ignored fields) + 1.38 (2014-06-06) + suppress MSVC warnings on integer casts truncating values + fix accidental rename of 'skip' field of I/O + 1.37 (2014-06-04) + remove duplicate typedef + 1.36 (2014-06-03) + convert to header file single-file library + if de-iphone isn't set, load iphone images color-swapped instead of returning NULL + 1.35 (2014-05-27) + various warnings + fix broken STBI_SIMD path + fix bug where stbi_load_from_file no longer left file pointer in correct place + fix broken non-easy path for 32-bit BMP (possibly never used) + TGA optimization by Arseny Kapoulkine + 1.34 (unknown) + use STBI_NOTUSED in stbi__resample_row_generic(), fix one more leak in tga failure case + 1.33 (2011-07-14) + make stbi_is_hdr work in STBI_NO_HDR (as specified), minor compiler-friendly improvements + 1.32 (2011-07-13) + support for "info" function for all supported filetypes (SpartanJ) + 1.31 (2011-06-20) + a few more leak fixes, bug in PNG handling (SpartanJ) + 1.30 (2011-06-11) + added ability to load files via callbacks to accomidate custom input streams (Ben Wenger) + removed deprecated format-specific test/load functions + removed support for installable file formats (stbi_loader) -- would have been broken for IO callbacks anyway + error cases in bmp and tga give messages and don't leak (Raymond Barbiero, grisha) + fix inefficiency in decoding 32-bit BMP (David Woo) + 1.29 (2010-08-16) + various warning fixes from Aurelien Pocheville + 1.28 (2010-08-01) + fix bug in GIF palette transparency (SpartanJ) + 1.27 (2010-08-01) + cast-to-stbi_uc to fix warnings + 1.26 (2010-07-24) + fix bug in file buffering for PNG reported by SpartanJ + 1.25 (2010-07-17) + refix trans_data warning (Won Chun) + 1.24 (2010-07-12) + perf improvements reading from files on platforms with lock-heavy fgetc() + minor perf improvements for jpeg + deprecated type-specific functions so we'll get feedback if they're needed + attempt to fix trans_data warning (Won Chun) + 1.23 fixed bug in iPhone support + 1.22 (2010-07-10) + removed image *writing* support + stbi_info support from Jetro Lauha + GIF support from Jean-Marc Lienher + iPhone PNG-extensions from James Brown + warning-fixes from Nicolas Schulz and Janez Zemva (i.stbi__err. Janez (U+017D)emva) + 1.21 fix use of 'stbi_uc' in header (reported by jon blow) + 1.20 added support for Softimage PIC, by Tom Seddon + 1.19 bug in interlaced PNG corruption check (found by ryg) + 1.18 2008-08-02 + fix a threading bug (local mutable static) + 1.17 support interlaced PNG + 1.16 major bugfix - stbi__convert_format converted one too many pixels + 1.15 initialize some fields for thread safety + 1.14 fix threadsafe conversion bug + header-file-only version (#define STBI_HEADER_FILE_ONLY before including) + 1.13 threadsafe + 1.12 const qualifiers in the API + 1.11 Support installable IDCT, colorspace conversion routines + 1.10 Fixes for 64-bit (don't use "unsigned long") + optimized upsampling by Fabian "ryg" Giesen + 1.09 Fix format-conversion for PSD code (bad global variables!) + 1.08 Thatcher Ulrich's PSD code integrated by Nicolas Schulz + 1.07 attempt to fix C++ warning/errors again + 1.06 attempt to fix C++ warning/errors again + 1.05 fix TGA loading to return correct *comp and use good luminance calc + 1.04 default float alpha is 1, not 255; use 'void *' for stbi_image_free + 1.03 bugfixes to STBI_NO_STDIO, STBI_NO_HDR + 1.02 support for (subset of) HDR files, float interface for preferred access to them + 1.01 fix bug: possible bug in handling right-side up bmps... not sure + fix bug: the stbi__bmp_load() and stbi__tga_load() functions didn't work at all + 1.00 interface to zlib that skips zlib header + 0.99 correct handling of alpha in palette + 0.98 TGA loader by lonesock; dynamically add loaders (untested) + 0.97 jpeg errors on too large a file; also catch another malloc failure + 0.96 fix detection of invalid v value - particleman@mollyrocket forum + 0.95 during header scan, seek to markers in case of padding + 0.94 STBI_NO_STDIO to disable stdio usage; rename all #defines the same + 0.93 handle jpegtran output; verbose errors + 0.92 read 4,8,16,24,32-bit BMP files of several formats + 0.91 output 24-bit Windows 3.0 BMP files + 0.90 fix a few more warnings; bump version number to approach 1.0 + 0.61 bugfixes due to Marc LeBlanc, Christopher Lloyd + 0.60 fix compiling as c++ + 0.59 fix warnings: merge Dave Moore's -Wall fixes + 0.58 fix bug: zlib uncompressed mode len/nlen was wrong endian + 0.57 fix bug: jpg last huffman symbol before marker was >9 bits but less than 16 available + 0.56 fix bug: zlib uncompressed mode len vs. nlen + 0.55 fix bug: restart_interval not initialized to 0 + 0.54 allow NULL for 'int *comp' + 0.53 fix bug in png 3->4; speedup png decoding + 0.52 png handles req_comp=3,4 directly; minor cleanup; jpeg comments + 0.51 obey req_comp requests, 1-component jpegs return as 1-component, + on 'test' only check type, not whether we support this variant + 0.50 first released version +*/ diff --git a/src/vendor/stb_truetype.c b/src/vendor/stb_truetype.c new file mode 100644 index 0000000..4d05372 --- /dev/null +++ b/src/vendor/stb_truetype.c @@ -0,0 +1,8 @@ +/* + * This source file obviously only exists to provide a spot + * for the stb_typetype implementation to exist. + */ + +#define STB_TRUETYPE_IMPLEMENTATION +#include "stb_truetype.h" + diff --git a/src/vendor/stb_truetype.h b/src/vendor/stb_truetype.h new file mode 100644 index 0000000..3cd936f --- /dev/null +++ b/src/vendor/stb_truetype.h @@ -0,0 +1,2100 @@ +// stb_truetype.h - v0.99 - public domain +// authored from 2009-2014 by Sean Barrett / RAD Game Tools +// +// This library processes TrueType files: +// parse files +// extract glyph metrics +// extract glyph shapes +// render glyphs to one-channel bitmaps with antialiasing (box filter) +// +// Todo: +// non-MS cmaps +// crashproof on bad data +// hinting? (no longer patented) +// cleartype-style AA? +// optimize: use simple memory allocator for intermediates +// optimize: build edge-list directly from curves +// optimize: rasterize directly from curves? +// +// ADDITIONAL CONTRIBUTORS +// +// Mikko Mononen: compound shape support, more cmap formats +// Tor Andersson: kerning, subpixel rendering +// +// Bug/warning reports/fixes: +// "Zer" on mollyrocket (with fix) +// Cass Everitt +// stoiko (Haemimont Games) +// Brian Hook +// Walter van Niftrik +// David Gow +// David Given +// Ivan-Assen Ivanov +// Anthony Pesch +// Johan Duparc +// Hou Qiming +// Fabian "ryg" Giesen +// +// VERSION HISTORY +// +// 0.99 (2014-09-18) fix multiple bugs with subpixel rendering (ryg) +// 0.9 (2014-08-07) support certain mac/iOS fonts without an MS platformID +// 0.8b (2014-07-07) fix a warning +// 0.8 (2014-05-25) fix a few more warnings +// 0.7 (2013-09-25) bugfix: subpixel glyph bug fixed in 0.5 had come back +// 0.6c (2012-07-24) improve documentation +// 0.6b (2012-07-20) fix a few more warnings +// 0.6 (2012-07-17) fix warnings; added stbtt_ScaleForMappingEmToPixels, +// stbtt_GetFontBoundingBox, stbtt_IsGlyphEmpty +// 0.5 (2011-12-09) bugfixes: +// subpixel glyph renderer computed wrong bounding box +// first vertex of shape can be off-curve (FreeSans) +// 0.4b (2011-12-03) fixed an error in the font baking example +// 0.4 (2011-12-01) kerning, subpixel rendering (tor) +// bugfixes for: +// codepoint-to-glyph conversion using table fmt=12 +// codepoint-to-glyph conversion using table fmt=4 +// stbtt_GetBakedQuad with non-square texture (Zer) +// updated Hello World! sample to use kerning and subpixel +// fixed some warnings +// 0.3 (2009-06-24) cmap fmt=12, compound shapes (MM) +// userdata, malloc-from-userdata, non-zero fill (STB) +// 0.2 (2009-03-11) Fix unsigned/signed char warnings +// 0.1 (2009-03-09) First public release +// +// LICENSE +// +// This software is in the public domain. Where that dedication is not +// recognized, you are granted a perpetual, irrevokable license to copy +// and modify this file as you see fit. +// +// USAGE +// +// Include this file in whatever places neeed to refer to it. In ONE C/C++ +// file, write: +// #define STB_TRUETYPE_IMPLEMENTATION +// before the #include of this file. This expands out the actual +// implementation into that C/C++ file. +// +// Simple 3D API (don't ship this, but it's fine for tools and quick start, +// and you can cut and paste from it to move to more advanced) +// stbtt_BakeFontBitmap() -- bake a font to a bitmap for use as texture +// stbtt_GetBakedQuad() -- compute quad to draw for a given char +// +// "Load" a font file from a memory buffer (you have to keep the buffer loaded) +// stbtt_InitFont() +// stbtt_GetFontOffsetForIndex() -- use for TTC font collections +// +// Render a unicode codepoint to a bitmap +// stbtt_GetCodepointBitmap() -- allocates and returns a bitmap +// stbtt_MakeCodepointBitmap() -- renders into bitmap you provide +// stbtt_GetCodepointBitmapBox() -- how big the bitmap must be +// +// Character advance/positioning +// stbtt_GetCodepointHMetrics() +// stbtt_GetFontVMetrics() +// stbtt_GetCodepointKernAdvance() +// +// ADDITIONAL DOCUMENTATION +// +// Immediately after this block comment are a series of sample programs. +// +// After the sample programs is the "header file" section. This section +// includes documentation for each API function. +// +// Some important concepts to understand to use this library: +// +// Codepoint +// Characters are defined by unicode codepoints, e.g. 65 is +// uppercase A, 231 is lowercase c with a cedilla, 0x7e30 is +// the hiragana for "ma". +// +// Glyph +// A visual character shape (every codepoint is rendered as +// some glyph) +// +// Glyph index +// A font-specific integer ID representing a glyph +// +// Baseline +// Glyph shapes are defined relative to a baseline, which is the +// bottom of uppercase characters. Characters extend both above +// and below the baseline. +// +// Current Point +// As you draw text to the screen, you keep track of a "current point" +// which is the origin of each character. The current point's vertical +// position is the baseline. Even "baked fonts" use this model. +// +// Vertical Font Metrics +// The vertical qualities of the font, used to vertically position +// and space the characters. See docs for stbtt_GetFontVMetrics. +// +// Font Size in Pixels or Points +// The preferred interface for specifying font sizes in stb_truetype +// is to specify how tall the font's vertical extent should be in pixels. +// If that sounds good enough, skip the next paragraph. +// +// Most font APIs instead use "points", which are a common typographic +// measurement for describing font size, defined as 72 points per inch. +// stb_truetype provides a point API for compatibility. However, true +// "per inch" conventions don't make much sense on computer displays +// since they different monitors have different number of pixels per +// inch. For example, Windows traditionally uses a convention that +// there are 96 pixels per inch, thus making 'inch' measurements have +// nothing to do with inches, and thus effectively defining a point to +// be 1.333 pixels. Additionally, the TrueType font data provides +// an explicit scale factor to scale a given font's glyphs to points, +// but the author has observed that this scale factor is often wrong +// for non-commercial fonts, thus making fonts scaled in points +// according to the TrueType spec incoherently sized in practice. +// +// ADVANCED USAGE +// +// Quality: +// +// - Use the functions with Subpixel at the end to allow your characters +// to have subpixel positioning. Since the font is anti-aliased, not +// hinted, this is very import for quality. (This is not possible with +// baked fonts.) +// +// - Kerning is now supported, and if you're supporting subpixel rendering +// then kerning is worth using to give your text a polished look. +// +// Performance: +// +// - Convert Unicode codepoints to glyph indexes and operate on the glyphs; +// if you don't do this, stb_truetype is forced to do the conversion on +// every call. +// +// - There are a lot of memory allocations. We should modify it to take +// a temp buffer and allocate from the temp buffer (without freeing), +// should help performance a lot. +// +// NOTES +// +// The system uses the raw data found in the .ttf file without changing it +// and without building auxiliary data structures. This is a bit inefficient +// on little-endian systems (the data is big-endian), but assuming you're +// caching the bitmaps or glyph shapes this shouldn't be a big deal. +// +// It appears to be very hard to programmatically determine what font a +// given file is in a general way. I provide an API for this, but I don't +// recommend it. +// +// +// SOURCE STATISTICS (based on v0.6c, 2050 LOC) +// +// Documentation & header file 520 LOC \___ 660 LOC documentation +// Sample code 140 LOC / +// Truetype parsing 620 LOC ---- 620 LOC TrueType +// Software rasterization 240 LOC \ . +// Curve tesselation 120 LOC \__ 550 LOC Bitmap creation +// Bitmap management 100 LOC / +// Baked bitmap interface 70 LOC / +// Font name matching & access 150 LOC ---- 150 +// C runtime library abstraction 60 LOC ---- 60 + + +////////////////////////////////////////////////////////////////////////////// +////////////////////////////////////////////////////////////////////////////// +//// +//// SAMPLE PROGRAMS +//// +// +// Incomplete text-in-3d-api example, which draws quads properly aligned to be lossless +// +#if 0 +#define STB_TRUETYPE_IMPLEMENTATION // force following include to generate implementation +#include "stb_truetype.h" + +char ttf_buffer[1<<20]; +unsigned char temp_bitmap[512*512]; + +stbtt_bakedchar cdata[96]; // ASCII 32..126 is 95 glyphs +GLstbtt_uint ftex; + +void my_stbtt_initfont(void) +{ + fread(ttf_buffer, 1, 1<<20, fopen("c:/windows/fonts/times.ttf", "rb")); + stbtt_BakeFontBitmap(data,0, 32.0, temp_bitmap,512,512, 32,96, cdata); // no guarantee this fits! + // can free ttf_buffer at this point + glGenTextures(1, &ftex); + glBindTexture(GL_TEXTURE_2D, ftex); + glTexImage2D(GL_TEXTURE_2D, 0, GL_ALPHA, 512,512, 0, GL_ALPHA, GL_UNSIGNED_BYTE, temp_bitmap); + // can free temp_bitmap at this point + glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR); +} + +void my_stbtt_print(float x, float y, char *text) +{ + // assume orthographic projection with units = screen pixels, origin at top left + glBindTexture(GL_TEXTURE_2D, ftex); + glBegin(GL_QUADS); + while (*text) { + if (*text >= 32 && *text < 128) { + stbtt_aligned_quad q; + stbtt_GetBakedQuad(cdata, 512,512, *text-32, &x,&y,&q,1);//1=opengl & d3d10+,0=d3d9 + glTexCoord2f(q.s0,q.t1); glVertex2f(q.x0,q.y0); + glTexCoord2f(q.s1,q.t1); glVertex2f(q.x1,q.y0); + glTexCoord2f(q.s1,q.t0); glVertex2f(q.x1,q.y1); + glTexCoord2f(q.s0,q.t0); glVertex2f(q.x0,q.y1); + } + ++text; + } + glEnd(); +} +#endif +// +// +////////////////////////////////////////////////////////////////////////////// +// +// Complete program (this compiles): get a single bitmap, print as ASCII art +// +#if 0 +#include +#define STB_TRUETYPE_IMPLEMENTATION // force following include to generate implementation +#include "stb_truetype.h" + +char ttf_buffer[1<<25]; + +int main(int argc, char **argv) +{ + stbtt_fontinfo font; + unsigned char *bitmap; + int w,h,i,j,c = (argc > 1 ? atoi(argv[1]) : 'a'), s = (argc > 2 ? atoi(argv[2]) : 20); + + fread(ttf_buffer, 1, 1<<25, fopen(argc > 3 ? argv[3] : "c:/windows/fonts/arialbd.ttf", "rb")); + + stbtt_InitFont(&font, ttf_buffer, stbtt_GetFontOffsetForIndex(ttf_buffer,0)); + bitmap = stbtt_GetCodepointBitmap(&font, 0,stbtt_ScaleForPixelHeight(&font, s), c, &w, &h, 0,0); + + for (j=0; j < h; ++j) { + for (i=0; i < w; ++i) + putchar(" .:ioVM@"[bitmap[j*w+i]>>5]); + putchar('\n'); + } + return 0; +} +#endif +// +// Output: +// +// .ii. +// @@@@@@. +// V@Mio@@o +// :i. V@V +// :oM@@M +// :@@@MM@M +// @@o o@M +// :@@. M@M +// @@@o@@@@ +// :M@@V:@@. +// +////////////////////////////////////////////////////////////////////////////// +// +// Complete program: print "Hello World!" banner, with bugs +// +#if 0 +char buffer[24<<20]; +unsigned char screen[20][79]; + +int main(int arg, char **argv) +{ + stbtt_fontinfo font; + int i,j,ascent,baseline,ch=0; + float scale, xpos=2; // leave a little padding in case the character extends left + char *text = "Heljo World!"; + + fread(buffer, 1, 1000000, fopen("c:/windows/fonts/arialbd.ttf", "rb")); + stbtt_InitFont(&font, buffer, 0); + + scale = stbtt_ScaleForPixelHeight(&font, 15); + stbtt_GetFontVMetrics(&font, &ascent,0,0); + baseline = (int) (ascent*scale); + + while (text[ch]) { + int advance,lsb,x0,y0,x1,y1; + float x_shift = xpos - (float) floor(xpos); + stbtt_GetCodepointHMetrics(&font, text[ch], &advance, &lsb); + stbtt_GetCodepointBitmapBoxSubpixel(&font, text[ch], scale,scale,x_shift,0, &x0,&y0,&x1,&y1); + stbtt_MakeCodepointBitmapSubpixel(&font, &screen[baseline + y0][(int) xpos + x0], x1-x0,y1-y0, 79, scale,scale,x_shift,0, text[ch]); + // note that this stomps the old data, so where character boxes overlap (e.g. 'lj') it's wrong + // because this API is really for baking character bitmaps into textures. if you want to render + // a sequence of characters, you really need to render each bitmap to a temp buffer, then + // "alpha blend" that into the working buffer + xpos += (advance * scale); + if (text[ch+1]) + xpos += scale*stbtt_GetCodepointKernAdvance(&font, text[ch],text[ch+1]); + ++ch; + } + + for (j=0; j < 20; ++j) { + for (i=0; i < 78; ++i) + putchar(" .:ioVM@"[screen[j][i]>>5]); + putchar('\n'); + } + + return 0; +} +#endif + + +////////////////////////////////////////////////////////////////////////////// +////////////////////////////////////////////////////////////////////////////// +//// +//// INTEGRATION WITH YOUR CODEBASE +//// +//// The following sections allow you to supply alternate definitions +//// of C library functions used by stb_truetype. + +#ifdef STB_TRUETYPE_IMPLEMENTATION + // #define your own (u)stbtt_int8/16/32 before including to override this + #ifndef stbtt_uint8 + typedef unsigned char stbtt_uint8; + typedef signed char stbtt_int8; + typedef unsigned short stbtt_uint16; + typedef signed short stbtt_int16; + typedef unsigned int stbtt_uint32; + typedef signed int stbtt_int32; + #endif + + typedef char stbtt__check_size32[sizeof(stbtt_int32)==4 ? 1 : -1]; + typedef char stbtt__check_size16[sizeof(stbtt_int16)==2 ? 1 : -1]; + + // #define your own STBTT_sort() to override this to avoid qsort + #ifndef STBTT_sort + #include + #define STBTT_sort(data,num_items,item_size,compare_func) qsort(data,num_items,item_size,compare_func) + #endif + + // #define your own STBTT_ifloor/STBTT_iceil() to avoid math.h + #ifndef STBTT_ifloor + #include + #define STBTT_ifloor(x) ((int) floor(x)) + #define STBTT_iceil(x) ((int) ceil(x)) + #endif + + #ifndef STBTT_sqrt + #include + #define STBTT_sqrt(x) sqrt(x) + #endif + + // #define your own functions "STBTT_malloc" / "STBTT_free" to avoid malloc.h + #ifndef STBTT_malloc + #include + #define STBTT_malloc(x,u) ((void)(u),malloc(x)) + #define STBTT_free(x,u) ((void)(u),free(x)) + #endif + + #ifndef STBTT_assert + #include + #define STBTT_assert(x) assert(x) + #endif + + #ifndef STBTT_strlen + #include + #define STBTT_strlen(x) strlen(x) + #endif + + #ifndef STBTT_memcpy + //#include + #include + #define STBTT_memcpy memcpy + #define STBTT_memset memset + #endif +#endif + +/////////////////////////////////////////////////////////////////////////////// +/////////////////////////////////////////////////////////////////////////////// +//// +//// INTERFACE +//// +//// + +#ifndef __STB_INCLUDE_STB_TRUETYPE_H__ +#define __STB_INCLUDE_STB_TRUETYPE_H__ + +#ifdef __cplusplus +extern "C" { +#endif + +////////////////////////////////////////////////////////////////////////////// +// +// TEXTURE BAKING API +// +// If you use this API, you only have to call two functions ever. +// + +typedef struct +{ + unsigned short x0,y0,x1,y1; // coordinates of bbox in bitmap + float xoff,yoff,xadvance; +} stbtt_bakedchar; + +extern int stbtt_BakeFontBitmap(const unsigned char *data, int offset, // font location (use offset=0 for plain .ttf) + float pixel_height, // height of font in pixels + unsigned char *pixels, int pw, int ph, // bitmap to be filled in + int first_char, int num_chars, // characters to bake + stbtt_bakedchar *chardata); // you allocate this, it's num_chars long +// if return is positive, the first unused row of the bitmap +// if return is negative, returns the negative of the number of characters that fit +// if return is 0, no characters fit and no rows were used +// This uses a very crappy packing. + +typedef struct +{ + float x0,y0,s0,t0; // top-left + float x1,y1,s1,t1; // bottom-right +} stbtt_aligned_quad; + +extern void stbtt_GetBakedQuad(stbtt_bakedchar *chardata, int pw, int ph, // same data as above + int char_index, // character to display + float *xpos, float *ypos, // pointers to current position in screen pixel space + stbtt_aligned_quad *q, // output: quad to draw + int opengl_fillrule); // true if opengl fill rule; false if DX9 or earlier +// Call GetBakedQuad with char_index = 'character - first_char', and it +// creates the quad you need to draw and advances the current position. +// +// The coordinate system used assumes y increases downwards. +// +// Characters will extend both above and below the current position; +// see discussion of "BASELINE" above. +// +// It's inefficient; you might want to c&p it and optimize it. + + +////////////////////////////////////////////////////////////////////////////// +// +// FONT LOADING +// +// + +extern int stbtt_GetFontOffsetForIndex(const unsigned char *data, int index); +// Each .ttf/.ttc file may have more than one font. Each font has a sequential +// index number starting from 0. Call this function to get the font offset for +// a given index; it returns -1 if the index is out of range. A regular .ttf +// file will only define one font and it always be at offset 0, so it will +// return '0' for index 0, and -1 for all other indices. You can just skip +// this step if you know it's that kind of font. + + +// The following structure is defined publically so you can declare one on +// the stack or as a global or etc, but you should treat it as opaque. +typedef struct stbtt_fontinfo +{ + void * userdata; + unsigned char * data; // pointer to .ttf file + int fontstart; // offset of start of font + + int numGlyphs; // number of glyphs, needed for range checking + + int loca,head,glyf,hhea,hmtx,kern; // table locations as offset from start of .ttf + int index_map; // a cmap mapping for our chosen character encoding + int indexToLocFormat; // format needed to map from glyph index to glyph +} stbtt_fontinfo; + +extern int stbtt_InitFont(stbtt_fontinfo *info, const unsigned char *data, int offset); +// Given an offset into the file that defines a font, this function builds +// the necessary cached info for the rest of the system. You must allocate +// the stbtt_fontinfo yourself, and stbtt_InitFont will fill it out. You don't +// need to do anything special to free it, because the contents are pure +// value data with no additional data structures. Returns 0 on failure. + + +////////////////////////////////////////////////////////////////////////////// +// +// CHARACTER TO GLYPH-INDEX CONVERSIOn + +int stbtt_FindGlyphIndex(const stbtt_fontinfo *info, int unicode_codepoint); +// If you're going to perform multiple operations on the same character +// and you want a speed-up, call this function with the character you're +// going to process, then use glyph-based functions instead of the +// codepoint-based functions. + + +////////////////////////////////////////////////////////////////////////////// +// +// CHARACTER PROPERTIES +// + +extern float stbtt_ScaleForPixelHeight(const stbtt_fontinfo *info, float pixels); +// computes a scale factor to produce a font whose "height" is 'pixels' tall. +// Height is measured as the distance from the highest ascender to the lowest +// descender; in other words, it's equivalent to calling stbtt_GetFontVMetrics +// and computing: +// scale = pixels / (ascent - descent) +// so if you prefer to measure height by the ascent only, use a similar calculation. + +extern float stbtt_ScaleForMappingEmToPixels(const stbtt_fontinfo *info, float pixels); +// computes a scale factor to produce a font whose EM size is mapped to +// 'pixels' tall. This is probably what traditional APIs compute, but +// I'm not positive. + +extern void stbtt_GetFontVMetrics(const stbtt_fontinfo *info, int *ascent, int *descent, int *lineGap); +// ascent is the coordinate above the baseline the font extends; descent +// is the coordinate below the baseline the font extends (i.e. it is typically negative) +// lineGap is the spacing between one row's descent and the next row's ascent... +// so you should advance the vertical position by "*ascent - *descent + *lineGap" +// these are expressed in unscaled coordinates, so you must multiply by +// the scale factor for a given size + +extern void stbtt_GetFontBoundingBox(const stbtt_fontinfo *info, int *x0, int *y0, int *x1, int *y1); +// the bounding box around all possible characters + +extern void stbtt_GetCodepointHMetrics(const stbtt_fontinfo *info, int codepoint, int *advanceWidth, int *leftSideBearing); +// leftSideBearing is the offset from the current horizontal position to the left edge of the character +// advanceWidth is the offset from the current horizontal position to the next horizontal position +// these are expressed in unscaled coordinates + +extern int stbtt_GetCodepointKernAdvance(const stbtt_fontinfo *info, int ch1, int ch2); +// an additional amount to add to the 'advance' value between ch1 and ch2 + +extern int stbtt_GetCodepointBox(const stbtt_fontinfo *info, int codepoint, int *x0, int *y0, int *x1, int *y1); +// Gets the bounding box of the visible part of the glyph, in unscaled coordinates + +extern void stbtt_GetGlyphHMetrics(const stbtt_fontinfo *info, int glyph_index, int *advanceWidth, int *leftSideBearing); +extern int stbtt_GetGlyphKernAdvance(const stbtt_fontinfo *info, int glyph1, int glyph2); +extern int stbtt_GetGlyphBox(const stbtt_fontinfo *info, int glyph_index, int *x0, int *y0, int *x1, int *y1); +// as above, but takes one or more glyph indices for greater efficiency + + +////////////////////////////////////////////////////////////////////////////// +// +// GLYPH SHAPES (you probably don't need these, but they have to go before +// the bitmaps for C declaration-order reasons) +// + +#ifndef STBTT_vmove // you can predefine these to use different values (but why?) + enum { + STBTT_vmove=1, + STBTT_vline, + STBTT_vcurve + }; +#endif + +#ifndef stbtt_vertex // you can predefine this to use different values + // (we share this with other code at RAD) + #define stbtt_vertex_type short // can't use stbtt_int16 because that's not visible in the header file + typedef struct + { + stbtt_vertex_type x,y,cx,cy; + unsigned char type,padding; + } stbtt_vertex; +#endif + +extern int stbtt_IsGlyphEmpty(const stbtt_fontinfo *info, int glyph_index); +// returns non-zero if nothing is drawn for this glyph + +extern int stbtt_GetCodepointShape(const stbtt_fontinfo *info, int unicode_codepoint, stbtt_vertex **vertices); +extern int stbtt_GetGlyphShape(const stbtt_fontinfo *info, int glyph_index, stbtt_vertex **vertices); +// returns # of vertices and fills *vertices with the pointer to them +// these are expressed in "unscaled" coordinates +// +// The shape is a series of countours. Each one starts with +// a STBTT_moveto, then consists of a series of mixed +// STBTT_lineto and STBTT_curveto segments. A lineto +// draws a line from previous endpoint to its x,y; a curveto +// draws a quadratic bezier from previous endpoint to +// its x,y, using cx,cy as the bezier control point. + +extern void stbtt_FreeShape(const stbtt_fontinfo *info, stbtt_vertex *vertices); +// frees the data allocated above + +////////////////////////////////////////////////////////////////////////////// +// +// BITMAP RENDERING +// + +extern void stbtt_FreeBitmap(unsigned char *bitmap, void *userdata); +// frees the bitmap allocated below + +extern unsigned char *stbtt_GetCodepointBitmap(const stbtt_fontinfo *info, float scale_x, float scale_y, int codepoint, int *width, int *height, int *xoff, int *yoff); +// allocates a large-enough single-channel 8bpp bitmap and renders the +// specified character/glyph at the specified scale into it, with +// antialiasing. 0 is no coverage (transparent), 255 is fully covered (opaque). +// *width & *height are filled out with the width & height of the bitmap, +// which is stored left-to-right, top-to-bottom. +// +// xoff/yoff are the offset it pixel space from the glyph origin to the top-left of the bitmap + +extern unsigned char *stbtt_GetCodepointBitmapSubpixel(const stbtt_fontinfo *info, float scale_x, float scale_y, float shift_x, float shift_y, int codepoint, int *width, int *height, int *xoff, int *yoff); +// the same as stbtt_GetCodepoitnBitmap, but you can specify a subpixel +// shift for the character + +extern void stbtt_MakeCodepointBitmap(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, int codepoint); +// the same as stbtt_GetCodepointBitmap, but you pass in storage for the bitmap +// in the form of 'output', with row spacing of 'out_stride' bytes. the bitmap +// is clipped to out_w/out_h bytes. Call stbtt_GetCodepointBitmapBox to get the +// width and height and positioning info for it first. + +extern void stbtt_MakeCodepointBitmapSubpixel(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, float shift_x, float shift_y, int codepoint); +// same as stbtt_MakeCodepointBitmap, but you can specify a subpixel +// shift for the character + +extern void stbtt_GetCodepointBitmapBox(const stbtt_fontinfo *font, int codepoint, float scale_x, float scale_y, int *ix0, int *iy0, int *ix1, int *iy1); +// get the bbox of the bitmap centered around the glyph origin; so the +// bitmap width is ix1-ix0, height is iy1-iy0, and location to place +// the bitmap top left is (leftSideBearing*scale,iy0). +// (Note that the bitmap uses y-increases-down, but the shape uses +// y-increases-up, so CodepointBitmapBox and CodepointBox are inverted.) + +extern void stbtt_GetCodepointBitmapBoxSubpixel(const stbtt_fontinfo *font, int codepoint, float scale_x, float scale_y, float shift_x, float shift_y, int *ix0, int *iy0, int *ix1, int *iy1); +// same as stbtt_GetCodepointBitmapBox, but you can specify a subpixel +// shift for the character + +// the following functions are equivalent to the above functions, but operate +// on glyph indices instead of Unicode codepoints (for efficiency) +extern unsigned char *stbtt_GetGlyphBitmap(const stbtt_fontinfo *info, float scale_x, float scale_y, int glyph, int *width, int *height, int *xoff, int *yoff); +extern unsigned char *stbtt_GetGlyphBitmapSubpixel(const stbtt_fontinfo *info, float scale_x, float scale_y, float shift_x, float shift_y, int glyph, int *width, int *height, int *xoff, int *yoff); +extern void stbtt_MakeGlyphBitmap(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, int glyph); +extern void stbtt_MakeGlyphBitmapSubpixel(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, float shift_x, float shift_y, int glyph); +extern void stbtt_GetGlyphBitmapBox(const stbtt_fontinfo *font, int glyph, float scale_x, float scale_y, int *ix0, int *iy0, int *ix1, int *iy1); +extern void stbtt_GetGlyphBitmapBoxSubpixel(const stbtt_fontinfo *font, int glyph, float scale_x, float scale_y,float shift_x, float shift_y, int *ix0, int *iy0, int *ix1, int *iy1); + + +// @TODO: don't expose this structure +typedef struct +{ + int w,h,stride; + unsigned char *pixels; +} stbtt__bitmap; + +extern void stbtt_Rasterize(stbtt__bitmap *result, float flatness_in_pixels, stbtt_vertex *vertices, int num_verts, float scale_x, float scale_y, float shift_x, float shift_y, int x_off, int y_off, int invert, void *userdata); + +////////////////////////////////////////////////////////////////////////////// +// +// Finding the right font... +// +// You should really just solve this offline, keep your own tables +// of what font is what, and don't try to get it out of the .ttf file. +// That's because getting it out of the .ttf file is really hard, because +// the names in the file can appear in many possible encodings, in many +// possible languages, and e.g. if you need a case-insensitive comparison, +// the details of that depend on the encoding & language in a complex way +// (actually underspecified in truetype, but also gigantic). +// +// But you can use the provided functions in two possible ways: +// stbtt_FindMatchingFont() will use *case-sensitive* comparisons on +// unicode-encoded names to try to find the font you want; +// you can run this before calling stbtt_InitFont() +// +// stbtt_GetFontNameString() lets you get any of the various strings +// from the file yourself and do your own comparisons on them. +// You have to have called stbtt_InitFont() first. + + +extern int stbtt_FindMatchingFont(const unsigned char *fontdata, const char *name, int flags); +// returns the offset (not index) of the font that matches, or -1 if none +// if you use STBTT_MACSTYLE_DONTCARE, use a font name like "Arial Bold". +// if you use any other flag, use a font name like "Arial"; this checks +// the 'macStyle' header field; i don't know if fonts set this consistently +#define STBTT_MACSTYLE_DONTCARE 0 +#define STBTT_MACSTYLE_BOLD 1 +#define STBTT_MACSTYLE_ITALIC 2 +#define STBTT_MACSTYLE_UNDERSCORE 4 +#define STBTT_MACSTYLE_NONE 8 // <= not same as 0, this makes us check the bitfield is 0 + +extern int stbtt_CompareUTF8toUTF16_bigendian(const char *s1, int len1, const char *s2, int len2); +// returns 1/0 whether the first string interpreted as utf8 is identical to +// the second string interpreted as big-endian utf16... useful for strings from next func + +extern const char *stbtt_GetFontNameString(const stbtt_fontinfo *font, int *length, int platformID, int encodingID, int languageID, int nameID); +// returns the string (which may be big-endian double byte, e.g. for unicode) +// and puts the length in bytes in *length. +// +// some of the values for the IDs are below; for more see the truetype spec: +// http://developer.apple.com/textfonts/TTRefMan/RM06/Chap6name.html +// http://www.microsoft.com/typography/otspec/name.htm + +enum { // platformID + STBTT_PLATFORM_ID_UNICODE =0, + STBTT_PLATFORM_ID_MAC =1, + STBTT_PLATFORM_ID_ISO =2, + STBTT_PLATFORM_ID_MICROSOFT =3 +}; + +enum { // encodingID for STBTT_PLATFORM_ID_UNICODE + STBTT_UNICODE_EID_UNICODE_1_0 =0, + STBTT_UNICODE_EID_UNICODE_1_1 =1, + STBTT_UNICODE_EID_ISO_10646 =2, + STBTT_UNICODE_EID_UNICODE_2_0_BMP=3, + STBTT_UNICODE_EID_UNICODE_2_0_FULL=4 +}; + +enum { // encodingID for STBTT_PLATFORM_ID_MICROSOFT + STBTT_MS_EID_SYMBOL =0, + STBTT_MS_EID_UNICODE_BMP =1, + STBTT_MS_EID_SHIFTJIS =2, + STBTT_MS_EID_UNICODE_FULL =10 +}; + +enum { // encodingID for STBTT_PLATFORM_ID_MAC; same as Script Manager codes + STBTT_MAC_EID_ROMAN =0, STBTT_MAC_EID_ARABIC =4, + STBTT_MAC_EID_JAPANESE =1, STBTT_MAC_EID_HEBREW =5, + STBTT_MAC_EID_CHINESE_TRAD =2, STBTT_MAC_EID_GREEK =6, + STBTT_MAC_EID_KOREAN =3, STBTT_MAC_EID_RUSSIAN =7 +}; + +enum { // languageID for STBTT_PLATFORM_ID_MICROSOFT; same as LCID... + // problematic because there are e.g. 16 english LCIDs and 16 arabic LCIDs + STBTT_MS_LANG_ENGLISH =0x0409, STBTT_MS_LANG_ITALIAN =0x0410, + STBTT_MS_LANG_CHINESE =0x0804, STBTT_MS_LANG_JAPANESE =0x0411, + STBTT_MS_LANG_DUTCH =0x0413, STBTT_MS_LANG_KOREAN =0x0412, + STBTT_MS_LANG_FRENCH =0x040c, STBTT_MS_LANG_RUSSIAN =0x0419, + STBTT_MS_LANG_GERMAN =0x0407, STBTT_MS_LANG_SPANISH =0x0409, + STBTT_MS_LANG_HEBREW =0x040d, STBTT_MS_LANG_SWEDISH =0x041D +}; + +enum { // languageID for STBTT_PLATFORM_ID_MAC + STBTT_MAC_LANG_ENGLISH =0 , STBTT_MAC_LANG_JAPANESE =11, + STBTT_MAC_LANG_ARABIC =12, STBTT_MAC_LANG_KOREAN =23, + STBTT_MAC_LANG_DUTCH =4 , STBTT_MAC_LANG_RUSSIAN =32, + STBTT_MAC_LANG_FRENCH =1 , STBTT_MAC_LANG_SPANISH =6 , + STBTT_MAC_LANG_GERMAN =2 , STBTT_MAC_LANG_SWEDISH =5 , + STBTT_MAC_LANG_HEBREW =10, STBTT_MAC_LANG_CHINESE_SIMPLIFIED =33, + STBTT_MAC_LANG_ITALIAN =3 , STBTT_MAC_LANG_CHINESE_TRAD =19 +}; + +#ifdef __cplusplus +} +#endif + +#endif // __STB_INCLUDE_STB_TRUETYPE_H__ + +/////////////////////////////////////////////////////////////////////////////// +/////////////////////////////////////////////////////////////////////////////// +//// +//// IMPLEMENTATION +//// +//// + +#ifdef STB_TRUETYPE_IMPLEMENTATION + +////////////////////////////////////////////////////////////////////////// +// +// accessors to parse data from file +// + +// on platforms that don't allow misaligned reads, if we want to allow +// truetype fonts that aren't padded to alignment, define ALLOW_UNALIGNED_TRUETYPE + +#define ttBYTE(p) (* (stbtt_uint8 *) (p)) +#define ttCHAR(p) (* (stbtt_int8 *) (p)) +#define ttFixed(p) ttLONG(p) + +#if defined(STB_TRUETYPE_BIGENDIAN) && !defined(ALLOW_UNALIGNED_TRUETYPE) + + #define ttUSHORT(p) (* (stbtt_uint16 *) (p)) + #define ttSHORT(p) (* (stbtt_int16 *) (p)) + #define ttULONG(p) (* (stbtt_uint32 *) (p)) + #define ttLONG(p) (* (stbtt_int32 *) (p)) + +#else + + stbtt_uint16 ttUSHORT(const stbtt_uint8 *p) { return p[0]*256 + p[1]; } + stbtt_int16 ttSHORT(const stbtt_uint8 *p) { return p[0]*256 + p[1]; } + stbtt_uint32 ttULONG(const stbtt_uint8 *p) { return (p[0]<<24) + (p[1]<<16) + (p[2]<<8) + p[3]; } + stbtt_int32 ttLONG(const stbtt_uint8 *p) { return (p[0]<<24) + (p[1]<<16) + (p[2]<<8) + p[3]; } + +#endif + +#define stbtt_tag4(p,c0,c1,c2,c3) ((p)[0] == (c0) && (p)[1] == (c1) && (p)[2] == (c2) && (p)[3] == (c3)) +#define stbtt_tag(p,str) stbtt_tag4(p,str[0],str[1],str[2],str[3]) + +static int stbtt__isfont(const stbtt_uint8 *font) +{ + // check the version number + if (stbtt_tag4(font, '1',0,0,0)) return 1; // TrueType 1 + if (stbtt_tag(font, "typ1")) return 1; // TrueType with type 1 font -- we don't support this! + if (stbtt_tag(font, "OTTO")) return 1; // OpenType with CFF + if (stbtt_tag4(font, 0,1,0,0)) return 1; // OpenType 1.0 + return 0; +} + +// @OPTIMIZE: binary search +static stbtt_uint32 stbtt__find_table(stbtt_uint8 *data, stbtt_uint32 fontstart, const char *tag) +{ + stbtt_int32 num_tables = ttUSHORT(data+fontstart+4); + stbtt_uint32 tabledir = fontstart + 12; + stbtt_int32 i; + for (i=0; i < num_tables; ++i) { + stbtt_uint32 loc = tabledir + 16*i; + if (stbtt_tag(data+loc+0, tag)) + return ttULONG(data+loc+8); + } + return 0; +} + +int stbtt_GetFontOffsetForIndex(const unsigned char *font_collection, int index) +{ + // if it's just a font, there's only one valid index + if (stbtt__isfont(font_collection)) + return index == 0 ? 0 : -1; + + // check if it's a TTC + if (stbtt_tag(font_collection, "ttcf")) { + // version 1? + if (ttULONG(font_collection+4) == 0x00010000 || ttULONG(font_collection+4) == 0x00020000) { + stbtt_int32 n = ttLONG(font_collection+8); + if (index >= n) + return -1; + return ttULONG(font_collection+12+index*14); + } + } + return -1; +} + +int stbtt_InitFont(stbtt_fontinfo *info, const unsigned char *data2, int fontstart) +{ + stbtt_uint8 *data = (stbtt_uint8 *) data2; + stbtt_uint32 cmap, t; + stbtt_int32 i,numTables; + + info->data = data; + info->fontstart = fontstart; + + cmap = stbtt__find_table(data, fontstart, "cmap"); // required + info->loca = stbtt__find_table(data, fontstart, "loca"); // required + info->head = stbtt__find_table(data, fontstart, "head"); // required + info->glyf = stbtt__find_table(data, fontstart, "glyf"); // required + info->hhea = stbtt__find_table(data, fontstart, "hhea"); // required + info->hmtx = stbtt__find_table(data, fontstart, "hmtx"); // required + info->kern = stbtt__find_table(data, fontstart, "kern"); // not required + if (!cmap || !info->loca || !info->head || !info->glyf || !info->hhea || !info->hmtx) + return 0; + + t = stbtt__find_table(data, fontstart, "maxp"); + if (t) + info->numGlyphs = ttUSHORT(data+t+4); + else + info->numGlyphs = 0xffff; + + // find a cmap encoding table we understand *now* to avoid searching + // later. (todo: could make this installable) + // the same regardless of glyph. + numTables = ttUSHORT(data + cmap + 2); + info->index_map = 0; + for (i=0; i < numTables; ++i) { + stbtt_uint32 encoding_record = cmap + 4 + 8 * i; + // find an encoding we understand: + switch(ttUSHORT(data+encoding_record)) { + case STBTT_PLATFORM_ID_MICROSOFT: + switch (ttUSHORT(data+encoding_record+2)) { + case STBTT_MS_EID_UNICODE_BMP: + case STBTT_MS_EID_UNICODE_FULL: + // MS/Unicode + info->index_map = cmap + ttULONG(data+encoding_record+4); + break; + } + break; + case STBTT_PLATFORM_ID_UNICODE: + // Mac/iOS has these + // all the encodingIDs are unicode, so we don't bother to check it + info->index_map = cmap + ttULONG(data+encoding_record+4); + break; + } + } + if (info->index_map == 0) + return 0; + + info->indexToLocFormat = ttUSHORT(data+info->head + 50); + return 1; +} + +int stbtt_FindGlyphIndex(const stbtt_fontinfo *info, int unicode_codepoint) +{ + stbtt_uint8 *data = info->data; + stbtt_uint32 index_map = info->index_map; + + stbtt_uint16 format = ttUSHORT(data + index_map + 0); + if (format == 0) { // apple byte encoding + stbtt_int32 bytes = ttUSHORT(data + index_map + 2); + if (unicode_codepoint < bytes-6) + return ttBYTE(data + index_map + 6 + unicode_codepoint); + return 0; + } else if (format == 6) { + stbtt_uint32 first = ttUSHORT(data + index_map + 6); + stbtt_uint32 count = ttUSHORT(data + index_map + 8); + if ((stbtt_uint32) unicode_codepoint >= first && (stbtt_uint32) unicode_codepoint < first+count) + return ttUSHORT(data + index_map + 10 + (unicode_codepoint - first)*2); + return 0; + } else if (format == 2) { + STBTT_assert(0); // @TODO: high-byte mapping for japanese/chinese/korean + return 0; + } else if (format == 4) { // standard mapping for windows fonts: binary search collection of ranges + stbtt_uint16 segcount = ttUSHORT(data+index_map+6) >> 1; + stbtt_uint16 searchRange = ttUSHORT(data+index_map+8) >> 1; + stbtt_uint16 entrySelector = ttUSHORT(data+index_map+10); + stbtt_uint16 rangeShift = ttUSHORT(data+index_map+12) >> 1; + stbtt_uint16 item, offset, start, end; + + // do a binary search of the segments + stbtt_uint32 endCount = index_map + 14; + stbtt_uint32 search = endCount; + + if (unicode_codepoint > 0xffff) + return 0; + + // they lie from endCount .. endCount + segCount + // but searchRange is the nearest power of two, so... + if (unicode_codepoint >= ttUSHORT(data + search + rangeShift*2)) + search += rangeShift*2; + + // now decrement to bias correctly to find smallest + search -= 2; + while (entrySelector) { + searchRange >>= 1; + start = ttUSHORT(data + search + searchRange*2 + segcount*2 + 2); + end = ttUSHORT(data + search + searchRange*2); + if (unicode_codepoint > end) + search += searchRange*2; + --entrySelector; + } + search += 2; + + item = (stbtt_uint16) ((search - endCount) >> 1); + + STBTT_assert(unicode_codepoint <= ttUSHORT(data + endCount + 2*item)); + start = ttUSHORT(data + index_map + 14 + segcount*2 + 2 + 2*item); + end = ttUSHORT(data + index_map + 14 + 2 + 2*item); + if (unicode_codepoint < start) + return 0; + + offset = ttUSHORT(data + index_map + 14 + segcount*6 + 2 + 2*item); + if (offset == 0) + return (stbtt_uint16) (unicode_codepoint + ttSHORT(data + index_map + 14 + segcount*4 + 2 + 2*item)); + + return ttUSHORT(data + offset + (unicode_codepoint-start)*2 + index_map + 14 + segcount*6 + 2 + 2*item); + } else if (format == 12 || format == 13) { + stbtt_uint32 ngroups = ttULONG(data+index_map+12); + stbtt_int32 low,high; + low = 0; high = (stbtt_int32)ngroups; + // Binary search the right group. + while (low < high) { + stbtt_int32 mid = low + ((high-low) >> 1); // rounds down, so low <= mid < high + stbtt_uint32 start_char = ttULONG(data+index_map+16+mid*12); + stbtt_uint32 end_char = ttULONG(data+index_map+16+mid*12+4); + if ((stbtt_uint32) unicode_codepoint < start_char) + high = mid; + else if ((stbtt_uint32) unicode_codepoint > end_char) + low = mid+1; + else { + stbtt_uint32 start_glyph = ttULONG(data+index_map+16+mid*12+8); + if (format == 12) + return start_glyph + unicode_codepoint-start_char; + else // format == 13 + return start_glyph; + } + } + return 0; // not found + } + // @TODO + STBTT_assert(0); + return 0; +} + +int stbtt_GetCodepointShape(const stbtt_fontinfo *info, int unicode_codepoint, stbtt_vertex **vertices) +{ + return stbtt_GetGlyphShape(info, stbtt_FindGlyphIndex(info, unicode_codepoint), vertices); +} + +static void stbtt_setvertex(stbtt_vertex *v, stbtt_uint8 type, stbtt_int32 x, stbtt_int32 y, stbtt_int32 cx, stbtt_int32 cy) +{ + v->type = type; + v->x = (stbtt_int16) x; + v->y = (stbtt_int16) y; + v->cx = (stbtt_int16) cx; + v->cy = (stbtt_int16) cy; +} + +static int stbtt__GetGlyfOffset(const stbtt_fontinfo *info, int glyph_index) +{ + int g1,g2; + + if (glyph_index >= info->numGlyphs) return -1; // glyph index out of range + if (info->indexToLocFormat >= 2) return -1; // unknown index->glyph map format + + if (info->indexToLocFormat == 0) { + g1 = info->glyf + ttUSHORT(info->data + info->loca + glyph_index * 2) * 2; + g2 = info->glyf + ttUSHORT(info->data + info->loca + glyph_index * 2 + 2) * 2; + } else { + g1 = info->glyf + ttULONG (info->data + info->loca + glyph_index * 4); + g2 = info->glyf + ttULONG (info->data + info->loca + glyph_index * 4 + 4); + } + + return g1==g2 ? -1 : g1; // if length is 0, return -1 +} + +int stbtt_GetGlyphBox(const stbtt_fontinfo *info, int glyph_index, int *x0, int *y0, int *x1, int *y1) +{ + int g = stbtt__GetGlyfOffset(info, glyph_index); + if (g < 0) return 0; + + if (x0) *x0 = ttSHORT(info->data + g + 2); + if (y0) *y0 = ttSHORT(info->data + g + 4); + if (x1) *x1 = ttSHORT(info->data + g + 6); + if (y1) *y1 = ttSHORT(info->data + g + 8); + return 1; +} + +int stbtt_GetCodepointBox(const stbtt_fontinfo *info, int codepoint, int *x0, int *y0, int *x1, int *y1) +{ + return stbtt_GetGlyphBox(info, stbtt_FindGlyphIndex(info,codepoint), x0,y0,x1,y1); +} + +int stbtt_IsGlyphEmpty(const stbtt_fontinfo *info, int glyph_index) +{ + stbtt_int16 numberOfContours; + int g = stbtt__GetGlyfOffset(info, glyph_index); + if (g < 0) return 1; + numberOfContours = ttSHORT(info->data + g); + return numberOfContours == 0; +} + +static int stbtt__close_shape(stbtt_vertex *vertices, int num_vertices, int was_off, int start_off, + stbtt_int32 sx, stbtt_int32 sy, stbtt_int32 scx, stbtt_int32 scy, stbtt_int32 cx, stbtt_int32 cy) +{ + if (start_off) { + if (was_off) + stbtt_setvertex(&vertices[num_vertices++], STBTT_vcurve, (cx+scx)>>1, (cy+scy)>>1, cx,cy); + stbtt_setvertex(&vertices[num_vertices++], STBTT_vcurve, sx,sy,scx,scy); + } else { + if (was_off) + stbtt_setvertex(&vertices[num_vertices++], STBTT_vcurve,sx,sy,cx,cy); + else + stbtt_setvertex(&vertices[num_vertices++], STBTT_vline,sx,sy,0,0); + } + return num_vertices; +} + +int stbtt_GetGlyphShape(const stbtt_fontinfo *info, int glyph_index, stbtt_vertex **pvertices) +{ + stbtt_int16 numberOfContours; + stbtt_uint8 *endPtsOfContours; + stbtt_uint8 *data = info->data; + stbtt_vertex *vertices=0; + int num_vertices=0; + int g = stbtt__GetGlyfOffset(info, glyph_index); + + *pvertices = NULL; + + if (g < 0) return 0; + + numberOfContours = ttSHORT(data + g); + + if (numberOfContours > 0) { + stbtt_uint8 flags=0,flagcount; + stbtt_int32 ins, i,j=0,m,n, next_move, was_off=0, off, start_off=0; + stbtt_int32 x,y,cx,cy,sx,sy, scx,scy; + stbtt_uint8 *points; + endPtsOfContours = (data + g + 10); + ins = ttUSHORT(data + g + 10 + numberOfContours * 2); + points = data + g + 10 + numberOfContours * 2 + 2 + ins; + + n = 1+ttUSHORT(endPtsOfContours + numberOfContours*2-2); + + m = n + 2*numberOfContours; // a loose bound on how many vertices we might need + vertices = (stbtt_vertex *) STBTT_malloc(m * sizeof(vertices[0]), info->userdata); + if (vertices == 0) + return 0; + + next_move = 0; + flagcount=0; + + // in first pass, we load uninterpreted data into the allocated array + // above, shifted to the end of the array so we won't overwrite it when + // we create our final data starting from the front + + off = m - n; // starting offset for uninterpreted data, regardless of how m ends up being calculated + + // first load flags + + for (i=0; i < n; ++i) { + if (flagcount == 0) { + flags = *points++; + if (flags & 8) + flagcount = *points++; + } else + --flagcount; + vertices[off+i].type = flags; + } + + // now load x coordinates + x=0; + for (i=0; i < n; ++i) { + flags = vertices[off+i].type; + if (flags & 2) { + stbtt_int16 dx = *points++; + x += (flags & 16) ? dx : -dx; // ??? + } else { + if (!(flags & 16)) { + x = x + (stbtt_int16) (points[0]*256 + points[1]); + points += 2; + } + } + vertices[off+i].x = (stbtt_int16) x; + } + + // now load y coordinates + y=0; + for (i=0; i < n; ++i) { + flags = vertices[off+i].type; + if (flags & 4) { + stbtt_int16 dy = *points++; + y += (flags & 32) ? dy : -dy; // ??? + } else { + if (!(flags & 32)) { + y = y + (stbtt_int16) (points[0]*256 + points[1]); + points += 2; + } + } + vertices[off+i].y = (stbtt_int16) y; + } + + // now convert them to our format + num_vertices=0; + sx = sy = cx = cy = scx = scy = 0; + for (i=0; i < n; ++i) { + flags = vertices[off+i].type; + x = (stbtt_int16) vertices[off+i].x; + y = (stbtt_int16) vertices[off+i].y; + + if (next_move == i) { + if (i != 0) + num_vertices = stbtt__close_shape(vertices, num_vertices, was_off, start_off, sx,sy,scx,scy,cx,cy); + + // now start the new one + start_off = !(flags & 1); + if (start_off) { + // if we start off with an off-curve point, then when we need to find a point on the curve + // where we can start, and we need to save some state for when we wraparound. + scx = x; + scy = y; + if (!(vertices[off+i+1].type & 1)) { + // next point is also a curve point, so interpolate an on-point curve + sx = (x + (stbtt_int32) vertices[off+i+1].x) >> 1; + sy = (y + (stbtt_int32) vertices[off+i+1].y) >> 1; + } else { + // otherwise just use the next point as our start point + sx = (stbtt_int32) vertices[off+i+1].x; + sy = (stbtt_int32) vertices[off+i+1].y; + ++i; // we're using point i+1 as the starting point, so skip it + } + } else { + sx = x; + sy = y; + } + stbtt_setvertex(&vertices[num_vertices++], STBTT_vmove,sx,sy,0,0); + was_off = 0; + next_move = 1 + ttUSHORT(endPtsOfContours+j*2); + ++j; + } else { + if (!(flags & 1)) { // if it's a curve + if (was_off) // two off-curve control points in a row means interpolate an on-curve midpoint + stbtt_setvertex(&vertices[num_vertices++], STBTT_vcurve, (cx+x)>>1, (cy+y)>>1, cx, cy); + cx = x; + cy = y; + was_off = 1; + } else { + if (was_off) + stbtt_setvertex(&vertices[num_vertices++], STBTT_vcurve, x,y, cx, cy); + else + stbtt_setvertex(&vertices[num_vertices++], STBTT_vline, x,y,0,0); + was_off = 0; + } + } + } + num_vertices = stbtt__close_shape(vertices, num_vertices, was_off, start_off, sx,sy,scx,scy,cx,cy); + } else if (numberOfContours == -1) { + // Compound shapes. + int more = 1; + stbtt_uint8 *comp = data + g + 10; + num_vertices = 0; + vertices = 0; + while (more) { + stbtt_uint16 flags, gidx; + int comp_num_verts = 0, i; + stbtt_vertex *comp_verts = 0, *tmp = 0; + float mtx[6] = {1,0,0,1,0,0}, m, n; + + flags = ttSHORT(comp); comp+=2; + gidx = ttSHORT(comp); comp+=2; + + if (flags & 2) { // XY values + if (flags & 1) { // shorts + mtx[4] = ttSHORT(comp); comp+=2; + mtx[5] = ttSHORT(comp); comp+=2; + } else { + mtx[4] = ttCHAR(comp); comp+=1; + mtx[5] = ttCHAR(comp); comp+=1; + } + } + else { + // @TODO handle matching point + STBTT_assert(0); + } + if (flags & (1<<3)) { // WE_HAVE_A_SCALE + mtx[0] = mtx[3] = ttSHORT(comp)/16384.0f; comp+=2; + mtx[1] = mtx[2] = 0; + } else if (flags & (1<<6)) { // WE_HAVE_AN_X_AND_YSCALE + mtx[0] = ttSHORT(comp)/16384.0f; comp+=2; + mtx[1] = mtx[2] = 0; + mtx[3] = ttSHORT(comp)/16384.0f; comp+=2; + } else if (flags & (1<<7)) { // WE_HAVE_A_TWO_BY_TWO + mtx[0] = ttSHORT(comp)/16384.0f; comp+=2; + mtx[1] = ttSHORT(comp)/16384.0f; comp+=2; + mtx[2] = ttSHORT(comp)/16384.0f; comp+=2; + mtx[3] = ttSHORT(comp)/16384.0f; comp+=2; + } + + // Find transformation scales. + m = (float) STBTT_sqrt(mtx[0]*mtx[0] + mtx[1]*mtx[1]); + n = (float) STBTT_sqrt(mtx[2]*mtx[2] + mtx[3]*mtx[3]); + + // Get indexed glyph. + comp_num_verts = stbtt_GetGlyphShape(info, gidx, &comp_verts); + if (comp_num_verts > 0) { + // Transform vertices. + for (i = 0; i < comp_num_verts; ++i) { + stbtt_vertex* v = &comp_verts[i]; + stbtt_vertex_type x,y; + x=v->x; y=v->y; + v->x = (stbtt_vertex_type)(m * (mtx[0]*x + mtx[2]*y + mtx[4])); + v->y = (stbtt_vertex_type)(n * (mtx[1]*x + mtx[3]*y + mtx[5])); + x=v->cx; y=v->cy; + v->cx = (stbtt_vertex_type)(m * (mtx[0]*x + mtx[2]*y + mtx[4])); + v->cy = (stbtt_vertex_type)(n * (mtx[1]*x + mtx[3]*y + mtx[5])); + } + // Append vertices. + tmp = (stbtt_vertex*)STBTT_malloc((num_vertices+comp_num_verts)*sizeof(stbtt_vertex), info->userdata); + if (!tmp) { + if (vertices) STBTT_free(vertices, info->userdata); + if (comp_verts) STBTT_free(comp_verts, info->userdata); + return 0; + } + if (num_vertices > 0) STBTT_memcpy(tmp, vertices, num_vertices*sizeof(stbtt_vertex)); + STBTT_memcpy(tmp+num_vertices, comp_verts, comp_num_verts*sizeof(stbtt_vertex)); + if (vertices) STBTT_free(vertices, info->userdata); + vertices = tmp; + STBTT_free(comp_verts, info->userdata); + num_vertices += comp_num_verts; + } + // More components ? + more = flags & (1<<5); + } + } else if (numberOfContours < 0) { + // @TODO other compound variations? + STBTT_assert(0); + } else { + // numberOfCounters == 0, do nothing + } + + *pvertices = vertices; + return num_vertices; +} + +void stbtt_GetGlyphHMetrics(const stbtt_fontinfo *info, int glyph_index, int *advanceWidth, int *leftSideBearing) +{ + stbtt_uint16 numOfLongHorMetrics = ttUSHORT(info->data+info->hhea + 34); + if (glyph_index < numOfLongHorMetrics) { + if (advanceWidth) *advanceWidth = ttSHORT(info->data + info->hmtx + 4*glyph_index); + if (leftSideBearing) *leftSideBearing = ttSHORT(info->data + info->hmtx + 4*glyph_index + 2); + } else { + if (advanceWidth) *advanceWidth = ttSHORT(info->data + info->hmtx + 4*(numOfLongHorMetrics-1)); + if (leftSideBearing) *leftSideBearing = ttSHORT(info->data + info->hmtx + 4*numOfLongHorMetrics + 2*(glyph_index - numOfLongHorMetrics)); + } +} + +int stbtt_GetGlyphKernAdvance(const stbtt_fontinfo *info, int glyph1, int glyph2) +{ + stbtt_uint8 *data = info->data + info->kern; + stbtt_uint32 needle, straw; + int l, r, m; + + // we only look at the first table. it must be 'horizontal' and format 0. + if (!info->kern) + return 0; + if (ttUSHORT(data+2) < 1) // number of tables, need at least 1 + return 0; + if (ttUSHORT(data+8) != 1) // horizontal flag must be set in format + return 0; + + l = 0; + r = ttUSHORT(data+10) - 1; + needle = glyph1 << 16 | glyph2; + while (l <= r) { + m = (l + r) >> 1; + straw = ttULONG(data+18+(m*6)); // note: unaligned read + if (needle < straw) + r = m - 1; + else if (needle > straw) + l = m + 1; + else + return ttSHORT(data+22+(m*6)); + } + return 0; +} + +int stbtt_GetCodepointKernAdvance(const stbtt_fontinfo *info, int ch1, int ch2) +{ + if (!info->kern) // if no kerning table, don't waste time looking up both codepoint->glyphs + return 0; + return stbtt_GetGlyphKernAdvance(info, stbtt_FindGlyphIndex(info,ch1), stbtt_FindGlyphIndex(info,ch2)); +} + +void stbtt_GetCodepointHMetrics(const stbtt_fontinfo *info, int codepoint, int *advanceWidth, int *leftSideBearing) +{ + stbtt_GetGlyphHMetrics(info, stbtt_FindGlyphIndex(info,codepoint), advanceWidth, leftSideBearing); +} + +void stbtt_GetFontVMetrics(const stbtt_fontinfo *info, int *ascent, int *descent, int *lineGap) +{ + if (ascent ) *ascent = ttSHORT(info->data+info->hhea + 4); + if (descent) *descent = ttSHORT(info->data+info->hhea + 6); + if (lineGap) *lineGap = ttSHORT(info->data+info->hhea + 8); +} + +void stbtt_GetFontBoundingBox(const stbtt_fontinfo *info, int *x0, int *y0, int *x1, int *y1) +{ + *x0 = ttSHORT(info->data + info->head + 36); + *y0 = ttSHORT(info->data + info->head + 38); + *x1 = ttSHORT(info->data + info->head + 40); + *y1 = ttSHORT(info->data + info->head + 42); +} + +float stbtt_ScaleForPixelHeight(const stbtt_fontinfo *info, float height) +{ + int fheight = ttSHORT(info->data + info->hhea + 4) - ttSHORT(info->data + info->hhea + 6); + return (float) height / fheight; +} + +float stbtt_ScaleForMappingEmToPixels(const stbtt_fontinfo *info, float pixels) +{ + int unitsPerEm = ttUSHORT(info->data + info->head + 18); + return pixels / unitsPerEm; +} + +void stbtt_FreeShape(const stbtt_fontinfo *info, stbtt_vertex *v) +{ + STBTT_free(v, info->userdata); +} + +////////////////////////////////////////////////////////////////////////////// +// +// antialiasing software rasterizer +// + +void stbtt_GetGlyphBitmapBoxSubpixel(const stbtt_fontinfo *font, int glyph, float scale_x, float scale_y,float shift_x, float shift_y, int *ix0, int *iy0, int *ix1, int *iy1) +{ + int x0,y0,x1,y1; + if (!stbtt_GetGlyphBox(font, glyph, &x0,&y0,&x1,&y1)) { + // e.g. space character + if (ix0) *ix0 = 0; + if (iy0) *iy0 = 0; + if (ix1) *ix1 = 0; + if (iy1) *iy1 = 0; + } else { + // move to integral bboxes (treating pixels as little squares, what pixels get touched)? + if (ix0) *ix0 = STBTT_ifloor( x0 * scale_x + shift_x); + if (iy0) *iy0 = STBTT_ifloor(-y1 * scale_y + shift_y); + if (ix1) *ix1 = STBTT_iceil ( x1 * scale_x + shift_x); + if (iy1) *iy1 = STBTT_iceil (-y0 * scale_y + shift_y); + } +} + +void stbtt_GetGlyphBitmapBox(const stbtt_fontinfo *font, int glyph, float scale_x, float scale_y, int *ix0, int *iy0, int *ix1, int *iy1) +{ + stbtt_GetGlyphBitmapBoxSubpixel(font, glyph, scale_x, scale_y,0.0f,0.0f, ix0, iy0, ix1, iy1); +} + +void stbtt_GetCodepointBitmapBoxSubpixel(const stbtt_fontinfo *font, int codepoint, float scale_x, float scale_y, float shift_x, float shift_y, int *ix0, int *iy0, int *ix1, int *iy1) +{ + stbtt_GetGlyphBitmapBoxSubpixel(font, stbtt_FindGlyphIndex(font,codepoint), scale_x, scale_y,shift_x,shift_y, ix0,iy0,ix1,iy1); +} + +void stbtt_GetCodepointBitmapBox(const stbtt_fontinfo *font, int codepoint, float scale_x, float scale_y, int *ix0, int *iy0, int *ix1, int *iy1) +{ + stbtt_GetCodepointBitmapBoxSubpixel(font, codepoint, scale_x, scale_y,0.0f,0.0f, ix0,iy0,ix1,iy1); +} + +typedef struct stbtt__edge { + float x0,y0, x1,y1; + int invert; +} stbtt__edge; + +typedef struct stbtt__active_edge +{ + int x,dx; + float ey; + struct stbtt__active_edge *next; + int valid; +} stbtt__active_edge; + +#define FIXSHIFT 10 +#define FIX (1 << FIXSHIFT) +#define FIXMASK (FIX-1) + +static stbtt__active_edge *new_active(stbtt__edge *e, int off_x, float start_point, void *userdata) +{ + stbtt__active_edge *z = (stbtt__active_edge *) STBTT_malloc(sizeof(*z), userdata); // @TODO: make a pool of these!!! + float dxdy = (e->x1 - e->x0) / (e->y1 - e->y0); + STBTT_assert(e->y0 <= start_point); + if (!z) return z; + // round dx down to avoid going too far + if (dxdy < 0) + z->dx = -STBTT_ifloor(FIX * -dxdy); + else + z->dx = STBTT_ifloor(FIX * dxdy); + z->x = STBTT_ifloor(FIX * (e->x0 + dxdy * (start_point - e->y0))); + z->x -= off_x * FIX; + z->ey = e->y1; + z->next = 0; + z->valid = e->invert ? 1 : -1; + return z; +} + +// note: this routine clips fills that extend off the edges... ideally this +// wouldn't happen, but it could happen if the truetype glyph bounding boxes +// are wrong, or if the user supplies a too-small bitmap +static void stbtt__fill_active_edges(unsigned char *scanline, int len, stbtt__active_edge *e, int max_weight) +{ + // non-zero winding fill + int x0=0, w=0; + + while (e) { + if (w == 0) { + // if we're currently at zero, we need to record the edge start point + x0 = e->x; w += e->valid; + } else { + int x1 = e->x; w += e->valid; + // if we went to zero, we need to draw + if (w == 0) { + int i = x0 >> FIXSHIFT; + int j = x1 >> FIXSHIFT; + + if (i < len && j >= 0) { + if (i == j) { + // x0,x1 are the same pixel, so compute combined coverage + scanline[i] = scanline[i] + (stbtt_uint8) ((x1 - x0) * max_weight >> FIXSHIFT); + } else { + if (i >= 0) // add antialiasing for x0 + scanline[i] = scanline[i] + (stbtt_uint8) (((FIX - (x0 & FIXMASK)) * max_weight) >> FIXSHIFT); + else + i = -1; // clip + + if (j < len) // add antialiasing for x1 + scanline[j] = scanline[j] + (stbtt_uint8) (((x1 & FIXMASK) * max_weight) >> FIXSHIFT); + else + j = len; // clip + + for (++i; i < j; ++i) // fill pixels between x0 and x1 + scanline[i] = scanline[i] + (stbtt_uint8) max_weight; + } + } + } + } + + e = e->next; + } +} + +static void stbtt__rasterize_sorted_edges(stbtt__bitmap *result, stbtt__edge *e, int n, int vsubsample, int off_x, int off_y, void *userdata) +{ + stbtt__active_edge *active = NULL; + int y,j=0; + int max_weight = (255 / vsubsample); // weight per vertical scanline + int s; // vertical subsample index + unsigned char scanline_data[512], *scanline; + + if (result->w > 512) + scanline = (unsigned char *) STBTT_malloc(result->w, userdata); + else + scanline = scanline_data; + + y = off_y * vsubsample; + e[n].y0 = (off_y + result->h) * (float) vsubsample + 1; + + while (j < result->h) { + STBTT_memset(scanline, 0, result->w); + for (s=0; s < vsubsample; ++s) { + // find center of pixel for this scanline + float scan_y = y + 0.5f; + stbtt__active_edge **step = &active; + + // update all active edges; + // remove all active edges that terminate before the center of this scanline + while (*step) { + stbtt__active_edge * z = *step; + if (z->ey <= scan_y) { + *step = z->next; // delete from list + STBTT_assert(z->valid); + z->valid = 0; + STBTT_free(z, userdata); + } else { + z->x += z->dx; // advance to position for current scanline + step = &((*step)->next); // advance through list + } + } + + // resort the list if needed + for(;;) { + int changed=0; + step = &active; + while (*step && (*step)->next) { + if ((*step)->x > (*step)->next->x) { + stbtt__active_edge *t = *step; + stbtt__active_edge *q = t->next; + + t->next = q->next; + q->next = t; + *step = q; + changed = 1; + } + step = &(*step)->next; + } + if (!changed) break; + } + + // insert all edges that start before the center of this scanline -- omit ones that also end on this scanline + while (e->y0 <= scan_y) { + if (e->y1 > scan_y) { + stbtt__active_edge *z = new_active(e, off_x, scan_y, userdata); + // find insertion point + if (active == NULL) + active = z; + else if (z->x < active->x) { + // insert at front + z->next = active; + active = z; + } else { + // find thing to insert AFTER + stbtt__active_edge *p = active; + while (p->next && p->next->x < z->x) + p = p->next; + // at this point, p->next->x is NOT < z->x + z->next = p->next; + p->next = z; + } + } + ++e; + } + + // now process all active edges in XOR fashion + if (active) + stbtt__fill_active_edges(scanline, result->w, active, max_weight); + + ++y; + } + STBTT_memcpy(result->pixels + j * result->stride, scanline, result->w); + ++j; + } + + while (active) { + stbtt__active_edge *z = active; + active = active->next; + STBTT_free(z, userdata); + } + + if (scanline != scanline_data) + STBTT_free(scanline, userdata); +} + +static int stbtt__edge_compare(const void *p, const void *q) +{ + stbtt__edge *a = (stbtt__edge *) p; + stbtt__edge *b = (stbtt__edge *) q; + + if (a->y0 < b->y0) return -1; + if (a->y0 > b->y0) return 1; + return 0; +} + +typedef struct +{ + float x,y; +} stbtt__point; + +static void stbtt__rasterize(stbtt__bitmap *result, stbtt__point *pts, int *wcount, int windings, float scale_x, float scale_y, float shift_x, float shift_y, int off_x, int off_y, int invert, void *userdata) +{ + float y_scale_inv = invert ? -scale_y : scale_y; + stbtt__edge *e; + int n,i,j,k,m; + int vsubsample = result->h < 8 ? 15 : 5; + // vsubsample should divide 255 evenly; otherwise we won't reach full opacity + + // now we have to blow out the windings into explicit edge lists + n = 0; + for (i=0; i < windings; ++i) + n += wcount[i]; + + e = (stbtt__edge *) STBTT_malloc(sizeof(*e) * (n+1), userdata); // add an extra one as a sentinel + if (e == 0) return; + n = 0; + + m=0; + for (i=0; i < windings; ++i) { + stbtt__point *p = pts + m; + m += wcount[i]; + j = wcount[i]-1; + for (k=0; k < wcount[i]; j=k++) { + int a=k,b=j; + // skip the edge if horizontal + if (p[j].y == p[k].y) + continue; + // add edge from j to k to the list + e[n].invert = 0; + if (invert ? p[j].y > p[k].y : p[j].y < p[k].y) { + e[n].invert = 1; + a=j,b=k; + } + e[n].x0 = p[a].x * scale_x + shift_x; + e[n].y0 = (p[a].y * y_scale_inv + shift_y) * vsubsample; + e[n].x1 = p[b].x * scale_x + shift_x; + e[n].y1 = (p[b].y * y_scale_inv + shift_y) * vsubsample; + ++n; + } + } + + // now sort the edges by their highest point (should snap to integer, and then by x) + STBTT_sort(e, n, sizeof(e[0]), stbtt__edge_compare); + + // now, traverse the scanlines and find the intersections on each scanline, use xor winding rule + stbtt__rasterize_sorted_edges(result, e, n, vsubsample, off_x, off_y, userdata); + + STBTT_free(e, userdata); +} + +static void stbtt__add_point(stbtt__point *points, int n, float x, float y) +{ + if (!points) return; // during first pass, it's unallocated + points[n].x = x; + points[n].y = y; +} + +// tesselate until threshhold p is happy... @TODO warped to compensate for non-linear stretching +static int stbtt__tesselate_curve(stbtt__point *points, int *num_points, float x0, float y0, float x1, float y1, float x2, float y2, float objspace_flatness_squared, int n) +{ + // midpoint + float mx = (x0 + 2*x1 + x2)/4; + float my = (y0 + 2*y1 + y2)/4; + // versus directly drawn line + float dx = (x0+x2)/2 - mx; + float dy = (y0+y2)/2 - my; + if (n > 16) // 65536 segments on one curve better be enough! + return 1; + if (dx*dx+dy*dy > objspace_flatness_squared) { // half-pixel error allowed... need to be smaller if AA + stbtt__tesselate_curve(points, num_points, x0,y0, (x0+x1)/2.0f,(y0+y1)/2.0f, mx,my, objspace_flatness_squared,n+1); + stbtt__tesselate_curve(points, num_points, mx,my, (x1+x2)/2.0f,(y1+y2)/2.0f, x2,y2, objspace_flatness_squared,n+1); + } else { + stbtt__add_point(points, *num_points,x2,y2); + *num_points = *num_points+1; + } + return 1; +} + +// returns number of contours +stbtt__point *stbtt_FlattenCurves(stbtt_vertex *vertices, int num_verts, float objspace_flatness, int **contour_lengths, int *num_contours, void *userdata) +{ + stbtt__point *points=0; + int num_points=0; + + float objspace_flatness_squared = objspace_flatness * objspace_flatness; + int i,n=0,start=0, pass; + + // count how many "moves" there are to get the contour count + for (i=0; i < num_verts; ++i) + if (vertices[i].type == STBTT_vmove) + ++n; + + *num_contours = n; + if (n == 0) return 0; + + *contour_lengths = (int *) STBTT_malloc(sizeof(**contour_lengths) * n, userdata); + + if (*contour_lengths == 0) { + *num_contours = 0; + return 0; + } + + // make two passes through the points so we don't need to realloc + for (pass=0; pass < 2; ++pass) { + float x=0,y=0; + if (pass == 1) { + points = (stbtt__point *) STBTT_malloc(num_points * sizeof(points[0]), userdata); + if (points == NULL) goto error; + } + num_points = 0; + n= -1; + for (i=0; i < num_verts; ++i) { + switch (vertices[i].type) { + case STBTT_vmove: + // start the next contour + if (n >= 0) + (*contour_lengths)[n] = num_points - start; + ++n; + start = num_points; + + x = vertices[i].x, y = vertices[i].y; + stbtt__add_point(points, num_points++, x,y); + break; + case STBTT_vline: + x = vertices[i].x, y = vertices[i].y; + stbtt__add_point(points, num_points++, x, y); + break; + case STBTT_vcurve: + stbtt__tesselate_curve(points, &num_points, x,y, + vertices[i].cx, vertices[i].cy, + vertices[i].x, vertices[i].y, + objspace_flatness_squared, 0); + x = vertices[i].x, y = vertices[i].y; + break; + } + } + (*contour_lengths)[n] = num_points - start; + } + + return points; +error: + STBTT_free(points, userdata); + STBTT_free(*contour_lengths, userdata); + *contour_lengths = 0; + *num_contours = 0; + return NULL; +} + +void stbtt_Rasterize(stbtt__bitmap *result, float flatness_in_pixels, stbtt_vertex *vertices, int num_verts, float scale_x, float scale_y, float shift_x, float shift_y, int x_off, int y_off, int invert, void *userdata) +{ + float scale = scale_x > scale_y ? scale_y : scale_x; + int winding_count, *winding_lengths; + stbtt__point *windings = stbtt_FlattenCurves(vertices, num_verts, flatness_in_pixels / scale, &winding_lengths, &winding_count, userdata); + if (windings) { + stbtt__rasterize(result, windings, winding_lengths, winding_count, scale_x, scale_y, shift_x, shift_y, x_off, y_off, invert, userdata); + STBTT_free(winding_lengths, userdata); + STBTT_free(windings, userdata); + } +} + +void stbtt_FreeBitmap(unsigned char *bitmap, void *userdata) +{ + STBTT_free(bitmap, userdata); +} + +unsigned char *stbtt_GetGlyphBitmapSubpixel(const stbtt_fontinfo *info, float scale_x, float scale_y, float shift_x, float shift_y, int glyph, int *width, int *height, int *xoff, int *yoff) +{ + int ix0,iy0,ix1,iy1; + stbtt__bitmap gbm; + stbtt_vertex *vertices; + int num_verts = stbtt_GetGlyphShape(info, glyph, &vertices); + + if (scale_x == 0) scale_x = scale_y; + if (scale_y == 0) { + if (scale_x == 0) return NULL; + scale_y = scale_x; + } + + stbtt_GetGlyphBitmapBoxSubpixel(info, glyph, scale_x, scale_y, shift_x, shift_y, &ix0,&iy0,&ix1,&iy1); + + // now we get the size + gbm.w = (ix1 - ix0); + gbm.h = (iy1 - iy0); + gbm.pixels = NULL; // in case we error + + if (width ) *width = gbm.w; + if (height) *height = gbm.h; + if (xoff ) *xoff = ix0; + if (yoff ) *yoff = iy0; + + if (gbm.w && gbm.h) { + gbm.pixels = (unsigned char *) STBTT_malloc(gbm.w * gbm.h, info->userdata); + if (gbm.pixels) { + gbm.stride = gbm.w; + + stbtt_Rasterize(&gbm, 0.35f, vertices, num_verts, scale_x, scale_y, shift_x, shift_y, ix0, iy0, 1, info->userdata); + } + } + STBTT_free(vertices, info->userdata); + return gbm.pixels; +} + +unsigned char *stbtt_GetGlyphBitmap(const stbtt_fontinfo *info, float scale_x, float scale_y, int glyph, int *width, int *height, int *xoff, int *yoff) +{ + return stbtt_GetGlyphBitmapSubpixel(info, scale_x, scale_y, 0.0f, 0.0f, glyph, width, height, xoff, yoff); +} + +void stbtt_MakeGlyphBitmapSubpixel(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, float shift_x, float shift_y, int glyph) +{ + int ix0,iy0; + stbtt_vertex *vertices; + int num_verts = stbtt_GetGlyphShape(info, glyph, &vertices); + stbtt__bitmap gbm; + + stbtt_GetGlyphBitmapBoxSubpixel(info, glyph, scale_x, scale_y, shift_x, shift_y, &ix0,&iy0,0,0); + gbm.pixels = output; + gbm.w = out_w; + gbm.h = out_h; + gbm.stride = out_stride; + + if (gbm.w && gbm.h) + stbtt_Rasterize(&gbm, 0.35f, vertices, num_verts, scale_x, scale_y, shift_x, shift_y, ix0,iy0, 1, info->userdata); + + STBTT_free(vertices, info->userdata); +} + +void stbtt_MakeGlyphBitmap(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, int glyph) +{ + stbtt_MakeGlyphBitmapSubpixel(info, output, out_w, out_h, out_stride, scale_x, scale_y, 0.0f,0.0f, glyph); +} + +unsigned char *stbtt_GetCodepointBitmapSubpixel(const stbtt_fontinfo *info, float scale_x, float scale_y, float shift_x, float shift_y, int codepoint, int *width, int *height, int *xoff, int *yoff) +{ + return stbtt_GetGlyphBitmapSubpixel(info, scale_x, scale_y,shift_x,shift_y, stbtt_FindGlyphIndex(info,codepoint), width,height,xoff,yoff); +} + +void stbtt_MakeCodepointBitmapSubpixel(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, float shift_x, float shift_y, int codepoint) +{ + stbtt_MakeGlyphBitmapSubpixel(info, output, out_w, out_h, out_stride, scale_x, scale_y, shift_x, shift_y, stbtt_FindGlyphIndex(info,codepoint)); +} + +unsigned char *stbtt_GetCodepointBitmap(const stbtt_fontinfo *info, float scale_x, float scale_y, int codepoint, int *width, int *height, int *xoff, int *yoff) +{ + return stbtt_GetCodepointBitmapSubpixel(info, scale_x, scale_y, 0.0f,0.0f, codepoint, width,height,xoff,yoff); +} + +void stbtt_MakeCodepointBitmap(const stbtt_fontinfo *info, unsigned char *output, int out_w, int out_h, int out_stride, float scale_x, float scale_y, int codepoint) +{ + stbtt_MakeCodepointBitmapSubpixel(info, output, out_w, out_h, out_stride, scale_x, scale_y, 0.0f,0.0f, codepoint); +} + +////////////////////////////////////////////////////////////////////////////// +// +// bitmap baking +// +// This is SUPER-CRAPPY packing to keep source code small + +extern int stbtt_BakeFontBitmap(const unsigned char *data, int offset, // font location (use offset=0 for plain .ttf) + float pixel_height, // height of font in pixels + unsigned char *pixels, int pw, int ph, // bitmap to be filled in + int first_char, int num_chars, // characters to bake + stbtt_bakedchar *chardata) +{ + float scale; + int x,y,bottom_y, i; + stbtt_fontinfo f; + stbtt_InitFont(&f, data, offset); + STBTT_memset(pixels, 0, pw*ph); // background of 0 around pixels + x=y=1; + bottom_y = 1; + + scale = stbtt_ScaleForPixelHeight(&f, pixel_height); + + for (i=0; i < num_chars; ++i) { + int advance, lsb, x0,y0,x1,y1,gw,gh; + int g = stbtt_FindGlyphIndex(&f, first_char + i); + stbtt_GetGlyphHMetrics(&f, g, &advance, &lsb); + stbtt_GetGlyphBitmapBox(&f, g, scale,scale, &x0,&y0,&x1,&y1); + gw = x1-x0; + gh = y1-y0; + if (x + gw + 1 >= pw) + y = bottom_y, x = 1; // advance to next row + if (y + gh + 1 >= ph) // check if it fits vertically AFTER potentially moving to next row + return -i; + STBTT_assert(x+gw < pw); + STBTT_assert(y+gh < ph); + stbtt_MakeGlyphBitmap(&f, pixels+x+y*pw, gw,gh,pw, scale,scale, g); + chardata[i].x0 = (stbtt_int16) x; + chardata[i].y0 = (stbtt_int16) y; + chardata[i].x1 = (stbtt_int16) (x + gw); + chardata[i].y1 = (stbtt_int16) (y + gh); + chardata[i].xadvance = scale * advance; + chardata[i].xoff = (float) x0; + chardata[i].yoff = (float) y0; + x = x + gw + 2; + if (y+gh+2 > bottom_y) + bottom_y = y+gh+2; + } + return bottom_y; +} + +void stbtt_GetBakedQuad(stbtt_bakedchar *chardata, int pw, int ph, int char_index, float *xpos, float *ypos, stbtt_aligned_quad *q, int opengl_fillrule) +{ + float d3d_bias = opengl_fillrule ? 0 : -0.5f; + float ipw = 1.0f / pw, iph = 1.0f / ph; + stbtt_bakedchar *b = chardata + char_index; + int round_x = STBTT_ifloor((*xpos + b->xoff) + 0.5); + int round_y = STBTT_ifloor((*ypos + b->yoff) + 0.5); + + q->x0 = round_x + d3d_bias; + q->y0 = round_y + d3d_bias; + q->x1 = round_x + b->x1 - b->x0 + d3d_bias; + q->y1 = round_y + b->y1 - b->y0 + d3d_bias; + + q->s0 = b->x0 * ipw; + q->t0 = b->y0 * iph; + q->s1 = b->x1 * ipw; + q->t1 = b->y1 * iph; + + *xpos += b->xadvance; +} + +////////////////////////////////////////////////////////////////////////////// +// +// font name matching -- recommended not to use this +// + +// check if a utf8 string contains a prefix which is the utf16 string; if so return length of matching utf8 string +static stbtt_int32 stbtt__CompareUTF8toUTF16_bigendian_prefix(const stbtt_uint8 *s1, stbtt_int32 len1, const stbtt_uint8 *s2, stbtt_int32 len2) +{ + stbtt_int32 i=0; + + // convert utf16 to utf8 and compare the results while converting + while (len2) { + stbtt_uint16 ch = s2[0]*256 + s2[1]; + if (ch < 0x80) { + if (i >= len1) return -1; + if (s1[i++] != ch) return -1; + } else if (ch < 0x800) { + if (i+1 >= len1) return -1; + if (s1[i++] != 0xc0 + (ch >> 6)) return -1; + if (s1[i++] != 0x80 + (ch & 0x3f)) return -1; + } else if (ch >= 0xd800 && ch < 0xdc00) { + stbtt_uint32 c; + stbtt_uint16 ch2 = s2[2]*256 + s2[3]; + if (i+3 >= len1) return -1; + c = ((ch - 0xd800) << 10) + (ch2 - 0xdc00) + 0x10000; + if (s1[i++] != 0xf0 + (c >> 18)) return -1; + if (s1[i++] != 0x80 + ((c >> 12) & 0x3f)) return -1; + if (s1[i++] != 0x80 + ((c >> 6) & 0x3f)) return -1; + if (s1[i++] != 0x80 + ((c ) & 0x3f)) return -1; + s2 += 2; // plus another 2 below + len2 -= 2; + } else if (ch >= 0xdc00 && ch < 0xe000) { + return -1; + } else { + if (i+2 >= len1) return -1; + if (s1[i++] != 0xe0 + (ch >> 12)) return -1; + if (s1[i++] != 0x80 + ((ch >> 6) & 0x3f)) return -1; + if (s1[i++] != 0x80 + ((ch ) & 0x3f)) return -1; + } + s2 += 2; + len2 -= 2; + } + return i; +} + +int stbtt_CompareUTF8toUTF16_bigendian(const char *s1, int len1, const char *s2, int len2) +{ + return len1 == stbtt__CompareUTF8toUTF16_bigendian_prefix((const stbtt_uint8*) s1, len1, (const stbtt_uint8*) s2, len2); +} + +// returns results in whatever encoding you request... but note that 2-byte encodings +// will be BIG-ENDIAN... use stbtt_CompareUTF8toUTF16_bigendian() to compare +const char *stbtt_GetFontNameString(const stbtt_fontinfo *font, int *length, int platformID, int encodingID, int languageID, int nameID) +{ + stbtt_int32 i,count,stringOffset; + stbtt_uint8 *fc = font->data; + stbtt_uint32 offset = font->fontstart; + stbtt_uint32 nm = stbtt__find_table(fc, offset, "name"); + if (!nm) return NULL; + + count = ttUSHORT(fc+nm+2); + stringOffset = nm + ttUSHORT(fc+nm+4); + for (i=0; i < count; ++i) { + stbtt_uint32 loc = nm + 6 + 12 * i; + if (platformID == ttUSHORT(fc+loc+0) && encodingID == ttUSHORT(fc+loc+2) + && languageID == ttUSHORT(fc+loc+4) && nameID == ttUSHORT(fc+loc+6)) { + *length = ttUSHORT(fc+loc+8); + return (const char *) (fc+stringOffset+ttUSHORT(fc+loc+10)); + } + } + return NULL; +} + +static int stbtt__matchpair(stbtt_uint8 *fc, stbtt_uint32 nm, stbtt_uint8 *name, stbtt_int32 nlen, stbtt_int32 target_id, stbtt_int32 next_id) +{ + stbtt_int32 i; + stbtt_int32 count = ttUSHORT(fc+nm+2); + stbtt_int32 stringOffset = nm + ttUSHORT(fc+nm+4); + + for (i=0; i < count; ++i) { + stbtt_uint32 loc = nm + 6 + 12 * i; + stbtt_int32 id = ttUSHORT(fc+loc+6); + if (id == target_id) { + // find the encoding + stbtt_int32 platform = ttUSHORT(fc+loc+0), encoding = ttUSHORT(fc+loc+2), language = ttUSHORT(fc+loc+4); + + // is this a Unicode encoding? + if (platform == 0 || (platform == 3 && encoding == 1) || (platform == 3 && encoding == 10)) { + stbtt_int32 slen = ttUSHORT(fc+loc+8); + stbtt_int32 off = ttUSHORT(fc+loc+10); + + // check if there's a prefix match + stbtt_int32 matchlen = stbtt__CompareUTF8toUTF16_bigendian_prefix(name, nlen, fc+stringOffset+off,slen); + if (matchlen >= 0) { + // check for target_id+1 immediately following, with same encoding & language + if (i+1 < count && ttUSHORT(fc+loc+12+6) == next_id && ttUSHORT(fc+loc+12) == platform && ttUSHORT(fc+loc+12+2) == encoding && ttUSHORT(fc+loc+12+4) == language) { + slen = ttUSHORT(fc+loc+12+8); + off = ttUSHORT(fc+loc+12+10); + if (slen == 0) { + if (matchlen == nlen) + return 1; + } else if (matchlen < nlen && name[matchlen] == ' ') { + ++matchlen; + if (stbtt_CompareUTF8toUTF16_bigendian((char*) (name+matchlen), nlen-matchlen, (char*)(fc+stringOffset+off),slen)) + return 1; + } + } else { + // if nothing immediately following + if (matchlen == nlen) + return 1; + } + } + } + + // @TODO handle other encodings + } + } + return 0; +} + +static int stbtt__matches(stbtt_uint8 *fc, stbtt_uint32 offset, stbtt_uint8 *name, stbtt_int32 flags) +{ + stbtt_int32 nlen = (stbtt_int32) STBTT_strlen((char *) name); + stbtt_uint32 nm,hd; + if (!stbtt__isfont(fc+offset)) return 0; + + // check italics/bold/underline flags in macStyle... + if (flags) { + hd = stbtt__find_table(fc, offset, "head"); + if ((ttUSHORT(fc+hd+44) & 7) != (flags & 7)) return 0; + } + + nm = stbtt__find_table(fc, offset, "name"); + if (!nm) return 0; + + if (flags) { + // if we checked the macStyle flags, then just check the family and ignore the subfamily + if (stbtt__matchpair(fc, nm, name, nlen, 16, -1)) return 1; + if (stbtt__matchpair(fc, nm, name, nlen, 1, -1)) return 1; + if (stbtt__matchpair(fc, nm, name, nlen, 3, -1)) return 1; + } else { + if (stbtt__matchpair(fc, nm, name, nlen, 16, 17)) return 1; + if (stbtt__matchpair(fc, nm, name, nlen, 1, 2)) return 1; + if (stbtt__matchpair(fc, nm, name, nlen, 3, -1)) return 1; + } + + return 0; +} + +int stbtt_FindMatchingFont(const unsigned char *font_collection, const char *name_utf8, stbtt_int32 flags) +{ + stbtt_int32 i; + for (i=0;;++i) { + stbtt_int32 off = stbtt_GetFontOffsetForIndex(font_collection, i); + if (off < 0) return off; + if (stbtt__matches((stbtt_uint8 *) font_collection, off, (stbtt_uint8*) name_utf8, flags)) + return off; + } +} + +#endif // STB_TRUETYPE_IMPLEMENTATION diff --git a/src/vm/vm.h b/src/vm/vm.h new file mode 100644 index 0000000..af5388f --- /dev/null +++ b/src/vm/vm.h @@ -0,0 +1,14 @@ +#pragma once + +#ifdef __cplusplus +extern "C" { +#endif + +#include "vm_stdinc.h" +#include "vm_log.h" +#include "vm_window.h" +#include "vm_input.h" + +#ifdef __cplusplus +} +#endif diff --git a/src/vm/vm_input.c b/src/vm/vm_input.c new file mode 100644 index 0000000..0c8da88 --- /dev/null +++ b/src/vm/vm_input.c @@ -0,0 +1,123 @@ +#include "vm_input.h" +#include "vm_log.h" +#include + +const int MAX_KEYS = SDL_NUM_SCANCODES; +const int MAX_MOUSE_BUTTONS = 3; + +struct { + bool keys[MAX_KEYS]; + bool locked_keys[MAX_KEYS]; + bool mouse_buttons[MAX_MOUSE_BUTTONS]; + bool locked_mouse_buttons[MAX_MOUSE_BUTTONS]; + uint mouse_x; + uint mouse_y; + int mouse_delta_x; + int mouse_delta_y; +} input_state; + +bool input_init() { + LOG_INFO("input_init\n"); + + memset(input_state.keys, 0, sizeof(bool) * MAX_KEYS); + memset(input_state.locked_keys, 0, sizeof(bool) * MAX_KEYS); + memset(input_state.mouse_buttons, 0, sizeof(bool) * MAX_MOUSE_BUTTONS); + memset(input_state.locked_mouse_buttons, 0, sizeof(bool) * MAX_MOUSE_BUTTONS); + input_state.mouse_x = 0; + input_state.mouse_y = 0; + input_state.mouse_delta_x = 0; + input_state.mouse_delta_y = 0; + + return true; +} + +void input_destroy() { + LOG_INFO("input_destroy\n"); +} + +void input_on_key_event(SDL_Event *event) { + if (event->type == SDL_KEYUP) { + int key = event->key.keysym.scancode; + input_state.keys[key] = false; + input_state.locked_keys[key] = false; + + } else if (event->type == SDL_KEYDOWN) { + int key = event->key.keysym.scancode; + input_state.keys[key] = !(input_state.locked_keys[key]); + } +} + +void input_on_mouse_event(SDL_Event *event) { + int button; + + switch (event->type) { + case SDL_MOUSEMOTION: + input_state.mouse_x = event->motion.x; + input_state.mouse_y = event->motion.y; + input_state.mouse_delta_x = event->motion.xrel; + input_state.mouse_delta_y = event->motion.yrel; + break; + + case SDL_MOUSEBUTTONUP: + button = event->button.button; + input_state.mouse_buttons[button] = false; + input_state.locked_mouse_buttons[button] = false; + break; + + case SDL_MOUSEBUTTONDOWN: + button = event->button.button; + input_state.mouse_buttons[button] = !(input_state.locked_mouse_buttons[button]); + break; + } +} + +bool input_is_key_down(KEYS key) { + SDL_Scancode scancode = SDL_GetScancodeFromKey(key); + return input_state.keys[scancode] && !input_state.locked_keys[scancode]; +} + +bool input_is_key_pressed(KEYS key) { + SDL_Scancode scancode = SDL_GetScancodeFromKey(key); + if (input_state.keys[scancode] && !input_state.locked_keys[scancode]) { + input_state.locked_keys[scancode] = true; + return true; + } else + return false; +} + +void input_lock_key(KEYS key) { + SDL_Scancode scancode = SDL_GetScancodeFromKey(key); + input_state.locked_keys[scancode] = true; +} + +bool input_is_mouse_button_down(MOUSE_BUTTONS button) { + return input_state.mouse_buttons[button] && !input_state.locked_mouse_buttons[button]; +} + +bool input_is_mouse_button_pressed(MOUSE_BUTTONS button) { + if (input_state.mouse_buttons[button] && !input_state.locked_mouse_buttons[button]) { + input_state.locked_mouse_buttons[button] = true; + return true; + } else + return false; +} + +void input_lock_mouse_button(MOUSE_BUTTONS button) { + input_state.locked_mouse_buttons[button] = true; +} + +int input_get_mouse_delta_x() { + return input_state.mouse_delta_x; +} + +int input_get_mouse_delta_y() { + return input_state.mouse_delta_y; +} + +uint input_get_mouse_x() { + return input_state.mouse_x; +} + +uint input_get_mouse_y() { + return input_state.mouse_y; +} diff --git a/src/vm/vm_input.h b/src/vm/vm_input.h new file mode 100644 index 0000000..672ad8b --- /dev/null +++ b/src/vm/vm_input.h @@ -0,0 +1,30 @@ +#pragma once + +#ifdef __cplusplus +extern "C" { +#endif + +#include "vm_stdinc.h" +#include "vm_input_codes.h" +#include "SDL.h" + +bool input_init(); +void input_destroy(); + +void input_on_key_event(SDL_Event *event); +void input_on_mouse_event(SDL_Event *event); + +bool input_is_key_down(KEYS key); +bool input_is_key_pressed(KEYS key); +void input_lock_key(KEYS key); +bool input_is_mouse_button_down(MOUSE_BUTTONS button); +bool input_is_mouse_button_pressed(MOUSE_BUTTONS button); +void input_lock_mouse_button(MOUSE_BUTTONS button); +int input_get_mouse_delta_x(); +int input_get_mouse_delta_y(); +uint input_get_mouse_x(); +uint input_get_mouse_y(); + +#ifdef __cplusplus +} +#endif diff --git a/src/vm/vm_input_codes.h b/src/vm/vm_input_codes.h new file mode 100644 index 0000000..5bd8259 --- /dev/null +++ b/src/vm/vm_input_codes.h @@ -0,0 +1,107 @@ +#pragma once + +#ifdef __cplusplus +extern "C" { +#endif + +#include "SDL.h" + +typedef enum { + MOUSE_LEFT = 1, + MOUSE_MIDDLE = 2, + MOUSE_RIGHT = 3 +} MOUSE_BUTTONS; + +// the idea behind this enum is that on other platforms (that may or may not use SDL) +// we could just remap to the platform's API, but use the same key constants + +typedef enum { + KSYM_UNKNOWN = 0, + KSYM_ESCAPE = SDLK_ESCAPE, + KSYM_BACKSPACE = SDLK_BACKSPACE, + KSYM_END = SDLK_END, + KSYM_SOFT_LEFT = 0, + KSYM_SOFT_RIGHT = 0, + KSYM_HOME = SDLK_HOME, + KSYM_BACK = 0, + KSYM_CALL = 0, + KSYM_ENDCALL = 0, + KSYM_0 = SDLK_0, + KSYM_1 = SDLK_1, + KSYM_2 = SDLK_2, + KSYM_3 = SDLK_3, + KSYM_4 = SDLK_4, + KSYM_5 = SDLK_5, + KSYM_6 = SDLK_6, + KSYM_7 = SDLK_7, + KSYM_8 = SDLK_8, + KSYM_9 = SDLK_9, + KSYM_ASTERISK = SDLK_ASTERISK, + KSYM_HASH = SDLK_HASH, + KSYM_UP = SDLK_UP, + KSYM_DOWN = SDLK_DOWN, + KSYM_LEFT = SDLK_LEFT, + KSYM_RIGHT = SDLK_RIGHT, + KSYM_DPAD_CENTER = 0, + KSYM_VOLUME_UP = 0, + KSYM_VOLUME_DOWN = 0, + KSYM_CAMERA = 0, + KSYM_CLEAR = 0, + KSYM_A = SDLK_a, + KSYM_B = SDLK_b, + KSYM_C = SDLK_c, + KSYM_D = SDLK_d, + KSYM_E = SDLK_e, + KSYM_F = SDLK_f, + KSYM_G = SDLK_g, + KSYM_H = SDLK_h, + KSYM_I = SDLK_i, + KSYM_J = SDLK_j, + KSYM_K = SDLK_k, + KSYM_L = SDLK_l, + KSYM_M = SDLK_m, + KSYM_N = SDLK_n, + KSYM_O = SDLK_o, + KSYM_P = SDLK_p, + KSYM_Q = SDLK_q, + KSYM_R = SDLK_r, + KSYM_S = SDLK_s, + KSYM_T = SDLK_t, + KSYM_U = SDLK_u, + KSYM_V = SDLK_v, + KSYM_W = SDLK_w, + KSYM_X = SDLK_x, + KSYM_Y = SDLK_y, + KSYM_Z = SDLK_z, + KSYM_COMMA = SDLK_COMMA, + KSYM_PERIOD = SDLK_PERIOD, + KSYM_LALT = SDLK_LALT, + KSYM_RALT = SDLK_RALT, + KSYM_RCTRL = SDLK_RCTRL, + KSYM_LCTRL = SDLK_LCTRL, + KSYM_LSHIFT = SDLK_LSHIFT, + KSYM_RSHIFT = SDLK_RSHIFT, + KSYM_TAB = SDLK_TAB, + KSYM_SPACE = SDLK_SPACE, + KSYM_SYM = 0, + KSYM_RETURN = SDLK_RETURN, + KSYM_DELETE = SDLK_DELETE, + KSYM_MINUS = SDLK_MINUS, + KSYM_EQUALS = SDLK_EQUALS, + KSYM_LEFTBRACKET = SDLK_LEFTBRACKET, + KSYM_RIGHTBRACKET = SDLK_RIGHTBRACKET, + KSYM_BACKSLASH = SDLK_BACKSLASH, + KSYM_SEMICOLON = SDLK_SEMICOLON, + KSYM_QUOTE = SDLK_QUOTE, + KSYM_SLASH = SDLK_SLASH, + KSYM_AT = SDLK_AT, + KSYM_PLUS = SDLK_PLUS, + KSYM_MENU = 0, + KSYM_SEARCH = 0, + KSYM_PAGEUP = SDLK_PAGEUP, + KSYM_PAGEDOWN = SDLK_PAGEDOWN +} KEYS; + +#ifdef __cplusplus +} +#endif diff --git a/src/vm/vm_log.c b/src/vm/vm_log.c new file mode 100644 index 0000000..d145bc1 --- /dev/null +++ b/src/vm/vm_log.c @@ -0,0 +1,73 @@ +#include "vm_log.h" + +#include +#include +#include +#include + +#ifdef _WIN32 +#define WIN32_LEAN_AND_MEAN +#define WIN32_EXTRA_LEAN +#include +#endif + +FILE *fp_log = NULL; +const char *log_filename = "run.log"; + +const size_t LOG_BUFFER_SIZE = 8096; +char _temp_buffer[LOG_BUFFER_SIZE]; +char _buffer[LOG_BUFFER_SIZE]; + +void log_init() { + fp_log = fopen(log_filename, "w"); + if (!fp_log) + { + snprintf(_buffer, LOG_BUFFER_SIZE, "Log file \"%s\" could not be opened. Further LOG_Write() calls will only appear in STDOUT.\n", log_filename); + printf("%s", _buffer); +#ifdef _MSC_VER + OutputDebugStringA(_buffer); +#endif + } +} + +void log_end() { + if (fp_log) + fclose(fp_log); +} + +void log_write(const char *tag, const char *format, ...) { + static char timestamp[15]; + + int hour; + int minute; + int second; + unsigned int milliseconds; + + struct timeb seconds; + ftime(&seconds); + struct tm *currentDate = localtime(&seconds.time); + + hour = currentDate->tm_hour; + minute = currentDate->tm_min; + second = currentDate->tm_sec; + milliseconds = seconds.millitm; + + // Prepend the date/time + snprintf(timestamp, 15, "[%02d:%02d:%02d,%03d]", hour, minute, second, milliseconds); + + // now the user-provided format string and arguments... + va_list args; + va_start(args, format); + vsnprintf(_temp_buffer, LOG_BUFFER_SIZE, format, args); + va_end(args); + + // write it all out + snprintf(_buffer, LOG_BUFFER_SIZE, "%s [%s] %s", timestamp, tag, _temp_buffer); + + if (fp_log) + fprintf(fp_log, "%s", _buffer); + printf("%s", _buffer); +#ifdef _MSC_VER + OutputDebugStringA(_buffer); +#endif +} diff --git a/src/vm/vm_log.h b/src/vm/vm_log.h new file mode 100644 index 0000000..283dc77 --- /dev/null +++ b/src/vm/vm_log.h @@ -0,0 +1,24 @@ +#pragma once + +#ifdef __cplusplus +extern "C" { +#endif + +#include "vm_stdinc.h" + +void log_init(); +void log_end(); +void log_write(const char *tag, const char *format, ...); + +#define LOG_INFO(...) (log_write("INFO", __VA_ARGS__)) +#define LOG_WARN(...) (log_write("WARN", __VA_ARGS__)) +#define LOG_ERROR(...) (log_write("ERROR", __VA_ARGS__)) +#ifdef DEBUG + #define LOG_DEBUG(...) (log_write("DEBUG", __VA_ARGS__)) +#else + #define LOG_DEBUG(...) +#endif + +#ifdef __cplusplus +} +#endif diff --git a/src/vm/vm_stdinc.h b/src/vm/vm_stdinc.h new file mode 100644 index 0000000..4f64623 --- /dev/null +++ b/src/vm/vm_stdinc.h @@ -0,0 +1,18 @@ +#pragma once + +#ifdef __cplusplus +extern "C" { +#endif + +#include +#include +#include + +typedef unsigned long ulong; +typedef unsigned int uint; +typedef unsigned short ushort; +typedef unsigned char uchar; + +#ifdef __cplusplus +} +#endif diff --git a/src/vm/vm_window.c b/src/vm/vm_window.c new file mode 100644 index 0000000..e7b4b47 --- /dev/null +++ b/src/vm/vm_window.c @@ -0,0 +1,190 @@ +#include "vm_window.h" +#include "vm_log.h" +#include "vm_input.h" +#include "fbgfx/surface.h" + +static SDL_Texture*create_texture(SDL_Renderer *renderer, int width, int height) { + return SDL_CreateTexture(renderer, SDL_PIXELFORMAT_BGR24, SDL_TEXTUREACCESS_STREAMING, width, height); +} + +static bool resize_texture(SDL_Renderer *renderer, SDL_Texture **texture, int width, int height) { + if (*texture) + SDL_DestroyTexture(*texture); + *texture = create_texture(renderer, width, height); + if (*texture == NULL) { + LOG_ERROR("Failed to resize texture to %dx%d\n", width, height); + return false; + } + + return true; +} + +static bool resize_window(WINDOW *window, int width, int height) { + if (window->fixed_viewport) + return true; + + LOG_INFO("Resizing window to %dx%d\n", width, height); + window->width = width; + window->height = height; + bool result = resize_texture(window->renderer, &window->texture, width, height); + if (result) + surface_resize(window->surface, width, height); + + return result; +} + +WINDOW * window_init(const char *title, int width, int height, int viewport_width, int viewport_height) { + LOG_INFO("window_init(%s, %d, %d, %d, %d)\n", title, width, height, viewport_width, viewport_height); + + WINDOW *window = malloc(sizeof(WINDOW)); + window->surface = NULL; + + LOG_INFO("Initializing SDL.\n"); + if (SDL_Init(SDL_INIT_EVERYTHING) != 0) { + LOG_ERROR("SDL_Init error: %s\n", SDL_GetError()); + window_destroy(window); + return NULL; + } + + window->initialized = true; + + window->width = width; + window->height = height; + + LOG_INFO("Creating a %dx%d window.\n", window->width, window->height); + window->window = SDL_CreateWindow(title, + SDL_WINDOWPOS_UNDEFINED, SDL_WINDOWPOS_UNDEFINED, + window->width, window->height, + SDL_WINDOW_RESIZABLE | SDL_WINDOW_SHOWN); + if (window->window == NULL) { + LOG_ERROR("SDL_CreateWindow error: %s\n", SDL_GetError()); + window_destroy(window); + return NULL; + } + + LOG_INFO("Creating an SDL renderer.\n"); + window->renderer = SDL_CreateRenderer(window->window, -1, 0); + if (window->renderer == NULL) { + LOG_ERROR("SDL_CreateRenderer error: %s\n", SDL_GetError()); + window_destroy(window); + return NULL; + } + + if (viewport_width != 0 && viewport_height != 0) { + LOG_INFO("Using fixed viewport of %dx%d\n", viewport_width, viewport_height); + window->fixed_viewport = true; + window->viewport_width = viewport_width; + window->viewport_height = viewport_height; + + SDL_SetHint(SDL_HINT_RENDER_SCALE_QUALITY, "nearest"); + SDL_RenderSetLogicalSize(window->renderer, window->viewport_width, window->viewport_height); + } else { + LOG_INFO("Using automatically sized viewport (always match window dimensions)\n"); + window->fixed_viewport = false; + window->viewport_width = width; + window->viewport_height = height; + } + + LOG_INFO("Creating a %dx%d texture for the 2D framebuffer.\n", window->viewport_width, window->viewport_height); + window->texture = create_texture(window->renderer, window->viewport_width, window->viewport_height); + if (window->texture == NULL) { + LOG_ERROR("SDL_CreateTexture error: %s\n", SDL_GetError()); + window_destroy(window); + return NULL; + } + + window->surface = surface_create(window->viewport_width, window->viewport_height, SURFACE_FORMAT_RGB, SURFACE_FLAGS_NONE); + + return window; +} + +bool window_do_events(WINDOW *window) { + bool quit = false; + SDL_Event e; + + while (SDL_PollEvent(&e)) { + switch (e.type) { + case SDL_QUIT: + LOG_INFO("Event: SDL_QUIT\n"); + quit = true; + break; + + case SDL_WINDOWEVENT: + if (e.window.event == SDL_WINDOWEVENT_RESIZED) { + LOG_INFO("Event: SDL_WINDOWEVENT_RESIZED\n"); + if (!resize_window(window, e.window.data1, e.window.data2)) + quit = true; + } + break; + + case SDL_KEYDOWN: + case SDL_KEYUP: + input_on_key_event(&e); + break; + + case SDL_MOUSEMOTION: + case SDL_MOUSEBUTTONDOWN: + case SDL_MOUSEBUTTONUP: + input_on_mouse_event(&e); + break; + } + } + + return !quit; +} + +void window_render(WINDOW *window) { + // HACK: there seems to be either some weird behaviour i'm just not understanding or a bug somewhere in + // either SDL or my own code that sets up these window / surface dimensions... + // basically, the "pass a NULL SDL_Rect to specify the entire surface region" does not seem to + // work as expected. if I use NULL's for all those arguments below, then I get a full-width + // but half-ish height (yes, not exactly half... seems random...) texture being displayed. + // eeerrrrggghh! wtf!? + + SDL_Rect src_region; + src_region.x = 0; + src_region.y = 0; + src_region.w = window->surface->width; + src_region.h = window->surface->height; + + SDL_Rect dest_region; + dest_region.x = 0; + dest_region.y = 0; + dest_region.w = window->width; + dest_region.h = window->height; + + SDL_UpdateTexture(window->texture, NULL, (void*)window->surface->pixels, window->surface->width * window->surface->bytes_per_pixel); + + SDL_RenderClear(window->renderer); +// SDL_RenderCopy(window->renderer, window->texture, &src_region, &dest_region); + SDL_RenderCopy(window->renderer, window->texture, NULL, NULL); + SDL_RenderPresent(window->renderer); +} + +void window_destroy(WINDOW *window) { + LOG_INFO("window_destroy\n"); + + surface_destroy(window->surface); + + if (window->texture) { + LOG_INFO("Destroying 2D framebuffer texture.\n"); + SDL_DestroyTexture(window->texture); + } + + if (window->renderer) { + LOG_INFO("Destroying SDL renderer.\n"); + SDL_DestroyRenderer(window->renderer); + } + + if (window->window) { + LOG_INFO("Destroying window.\n"); + SDL_DestroyWindow(window->window); + } + + if (window->initialized) { + LOG_INFO("SDL_Quit\n"); + SDL_Quit(); + } + + free(window); +} diff --git a/src/vm/vm_window.h b/src/vm/vm_window.h new file mode 100644 index 0000000..9c31f49 --- /dev/null +++ b/src/vm/vm_window.h @@ -0,0 +1,30 @@ +#pragma once + +#ifdef __cplusplus +extern "C" { +#endif + +#include "vm_stdinc.h" +#include "SDL.h" + +typedef struct { + bool initialized; + SDL_Window *window; + int width; + int height; + SDL_Renderer *renderer; + SDL_Texture *texture; + bool fixed_viewport; + int viewport_width; + int viewport_height; + struct SURFACE *surface; +} WINDOW; + +WINDOW * window_init(const char *title, int width, int height, int viewport_width, int viewport_height); +bool window_do_events(WINDOW *window); +void window_render(WINDOW *window); +void window_destroy(WINDOW *window); + +#ifdef __cplusplus +} +#endif