From 4740c9a85c6f4b6e4b84587e948990821491e521 Mon Sep 17 00:00:00 2001 From: benkyd Date: Wed, 11 May 2022 13:44:34 +0000 Subject: [PATCH] WOrking version --- LICENSE | 21 + legacy/resources/shaders/simple.frag | 27 + legacy/resources/shaders/simple.vert | 26 + legacy/resources/textures/bedrock.png | Bin 0 -> 225 bytes legacy/resources/textures/cobblestone.png | Bin 0 -> 568 bytes legacy/resources/textures/dirt.png | Bin 0 -> 266 bytes legacy/resources/textures/grass_side.png | Bin 0 -> 660 bytes legacy/resources/textures/grass_top.png | Bin 0 -> 766 bytes legacy/resources/textures/stone.png | Bin 0 -> 223 bytes legacy/src/config.hpp | 16 + legacy/src/game.cpp | 184 + legacy/src/game.hpp | 55 + legacy/src/main.cpp | 13 + legacy/src/physics/collider.cpp | 35 + legacy/src/physics/collider.hpp | 39 + legacy/src/renderer/renderer.cpp | 16 + legacy/src/renderer/renderer.hpp | 18 + legacy/src/util/filereader.cpp | 15 + legacy/src/util/filereader.hpp | 14 + legacy/src/world/block.cpp | 52 + legacy/src/world/block.hpp | 86 + legacy/src/world/chunk/chunk.cpp | 295 + legacy/src/world/chunk/chunk.hpp | 70 + legacy/src/world/chunk/face.hpp | 127 + legacy/src/world/chunk/voxel.cpp | 132 + legacy/src/world/chunk/voxel.hpp | 33 + legacy/src/world/entity.cpp | 59 + legacy/src/world/entity.hpp | 51 + legacy/src/world/generator/chunkgenerator.cpp | 14 + legacy/src/world/generator/chunkgenerator.hpp | 6 + legacy/src/world/generator/chunkmanager.hpp | 28 + legacy/src/world/world.cpp | 191 + legacy/src/world/world.hpp | 80 + src/Rendering/camera.cpp | 209 + src/Rendering/camera.hpp | 55 + src/Rendering/face.hpp | 128 + src/Rendering/frustrum.cpp | 3 + src/Rendering/frustrum.hpp | 29 + src/Rendering/mesh.cpp | 6 + src/Rendering/mesh.hpp | 27 + src/Rendering/renderable.cpp | 66 + src/Rendering/renderable.hpp | 45 + src/Rendering/rendermaster.cpp | 11 + src/Rendering/rendermaster.hpp | 41 + src/Rendering/shader.cpp | 121 + src/Rendering/shader.hpp | 30 + src/Rendering/texture.cpp | 60 + src/Rendering/texture.hpp | 14 + src/Rendering/voxelmesh.hpp | 23 + src/ThirdParty/fastnoise.cpp | 2250 +++++ src/ThirdParty/fastnoise.hpp | 311 + src/ThirdParty/glad.c | 1843 ++++ src/ThirdParty/stb_image.hpp | 7568 +++++++++++++++++ src/ThirdParty/stb_image_write.hpp | 7568 +++++++++++++++++ src/display.cpp | 120 + src/display.hpp | 42 + src/settings.hpp | 14 + src/threadpool.hpp | 3 + src/utilities.hpp | 10 + 59 files changed, 22300 insertions(+) create mode 100644 LICENSE create mode 100644 legacy/resources/shaders/simple.frag create mode 100644 legacy/resources/shaders/simple.vert create mode 100644 legacy/resources/textures/bedrock.png create mode 100644 legacy/resources/textures/cobblestone.png create mode 100644 legacy/resources/textures/dirt.png create mode 100644 legacy/resources/textures/grass_side.png create mode 100644 legacy/resources/textures/grass_top.png create mode 100644 legacy/resources/textures/stone.png create mode 100644 legacy/src/config.hpp create mode 100644 legacy/src/game.cpp create mode 100644 legacy/src/game.hpp create mode 100644 legacy/src/main.cpp create mode 100644 legacy/src/physics/collider.cpp create mode 100644 legacy/src/physics/collider.hpp create mode 100644 legacy/src/renderer/renderer.cpp create mode 100644 legacy/src/renderer/renderer.hpp create mode 100644 legacy/src/util/filereader.cpp create mode 100644 legacy/src/util/filereader.hpp create mode 100644 legacy/src/world/block.cpp create mode 100644 legacy/src/world/block.hpp create mode 100644 legacy/src/world/chunk/chunk.cpp create mode 100644 legacy/src/world/chunk/chunk.hpp create mode 100644 legacy/src/world/chunk/face.hpp create mode 100644 legacy/src/world/chunk/voxel.cpp create mode 100644 legacy/src/world/chunk/voxel.hpp create mode 100644 legacy/src/world/entity.cpp create mode 100644 legacy/src/world/entity.hpp create mode 100644 legacy/src/world/generator/chunkgenerator.cpp create mode 100644 legacy/src/world/generator/chunkgenerator.hpp create mode 100644 legacy/src/world/generator/chunkmanager.hpp create mode 100644 legacy/src/world/world.cpp create mode 100644 legacy/src/world/world.hpp create mode 100644 src/Rendering/camera.cpp create mode 100644 src/Rendering/camera.hpp create mode 100644 src/Rendering/face.hpp create mode 100644 src/Rendering/frustrum.cpp create mode 100644 src/Rendering/frustrum.hpp create mode 100644 src/Rendering/mesh.cpp create mode 100644 src/Rendering/mesh.hpp create mode 100644 src/Rendering/renderable.cpp create mode 100644 src/Rendering/renderable.hpp create mode 100644 src/Rendering/rendermaster.cpp create mode 100644 src/Rendering/rendermaster.hpp create mode 100644 src/Rendering/shader.cpp create mode 100644 src/Rendering/shader.hpp create mode 100644 src/Rendering/texture.cpp create mode 100644 src/Rendering/texture.hpp create mode 100644 src/Rendering/voxelmesh.hpp create mode 100644 src/ThirdParty/fastnoise.cpp create mode 100644 src/ThirdParty/fastnoise.hpp create mode 100644 src/ThirdParty/glad.c create mode 100644 src/ThirdParty/stb_image.hpp create mode 100644 src/ThirdParty/stb_image_write.hpp create mode 100644 src/display.cpp create mode 100644 src/display.hpp create mode 100644 src/settings.hpp create mode 100644 src/threadpool.hpp create mode 100644 src/utilities.hpp diff --git a/LICENSE b/LICENSE new file mode 100644 index 0000000..0ec9b90 --- /dev/null +++ b/LICENSE @@ -0,0 +1,21 @@ +MIT License + +Copyright (c) 2021 Benjamin Kyd + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/legacy/resources/shaders/simple.frag b/legacy/resources/shaders/simple.frag new file mode 100644 index 0000000..949c2ec --- /dev/null +++ b/legacy/resources/shaders/simple.frag @@ -0,0 +1,27 @@ +#version 450 + +vec3 SkyColour = vec3(186.0f / 255.0f, 214.0f / 255.0f, 254.0f / 255.0f); + +in vec3 TexCoord; +in float Distance; + +out vec4 outColour; + +uniform sampler2DArray tex; + +void main() { + + outColour = texture(tex, TexCoord); + //outColour = vec4(.9, .9, .9, 1); + + if (outColour.w == .0) + discard; + + float fogMax = 60000; + + vec3 colour = mix(outColour.xyz, SkyColour, min(1.0f, Distance / fogMax)); + + // Retain fragment transparency + outColour = vec4(colour, outColour.w); + +} diff --git a/legacy/resources/shaders/simple.vert b/legacy/resources/shaders/simple.vert new file mode 100644 index 0000000..bad13e0 --- /dev/null +++ b/legacy/resources/shaders/simple.vert @@ -0,0 +1,26 @@ +#version 450 + +layout (location = 0) in vec3 position; +layout (location = 1) in vec3 texcoord; + +out vec3 TexCoord; +out float Distance; + +uniform mat4 model; +uniform mat4 view; +uniform mat4 proj; + +void main() { + + TexCoord = texcoord; + + gl_Position = proj * view * model * vec4(position, 1.0); + + // Makes no sense but it works + Distance = ( + gl_Position.x * gl_Position.x + + gl_Position.y * gl_Position.y + + gl_Position.z * gl_Position.z + ); + +} diff --git a/legacy/resources/textures/bedrock.png b/legacy/resources/textures/bedrock.png new file mode 100644 index 0000000000000000000000000000000000000000..1643c9958fa499aa6a57a36efbde0c479e00fe6a GIT binary patch literal 225 zcmV<703QE|P)F#%PL-74aEz1~lKyIiDiVXa7M##mtc# z%Zg3STy?z)`(m#A!{X~9<&Lv_W@UB83}8|6RqkoxpZOmj5{fI&u##0A$Ed6X1#0Hi bBT~ROrseuXuHE=G00000NkvXXu0mjfLQ-Os literal 0 HcmV?d00001 diff --git a/legacy/resources/textures/cobblestone.png b/legacy/resources/textures/cobblestone.png new file mode 100644 index 0000000000000000000000000000000000000000..da3498c5c8d0f279a5e8f62e254241debe421344 GIT binary patch literal 568 zcmV-80>}M{P)N)G@VZQ44>2Kr1SYq5@WSmNv+pwt_HxbB@zh^G?`3fx7!)B*=%%2 zNRY{7NWI^0s#dF%&*$m&dNFCw=aT`F===TBe!u5H=?GlmpDL9~eDVAJwA<~d)oL*x zU|w2As zy3CzSCV5U_9{+;DpbiFK3bVqAFTeNzyLZLurvsovFYFH`|sW%Z4VIQzvE|-fJ z3hN5V?)N)alLR<-IvqAv0oCv@u7w`V=NDEI+d2EruJ8Nt`?O29*Y((K z=DVK`tGLh7MTtz%ds&v@Ga&DZ$;xQ1!Z)iihj}Hm*4l8eo1~Im3;+ie&}VWdpbDk0 z+*Vd{tJH?~A!2frcub(;r0KIVImhR2X8sP4PM&T#KQlJngXEsPNu_c;# zj2A0tbfmyB_f1S^g6x^QT2YWaZ|9A0rZL;l$|NdG{#j@PkULH z+3fY7k(rU<-=BXB-#+|e00DMx7KYDne=u+furu(9aWee+_J`s7$6pM8|NdiOVqygI zKfU?EAg;{EzyyR0Kfe41E9MpDVBnMBWOzB@5d%9nD_H!+tq%3Zzewm+w}3}cZM%OlR>Uv<6>d>GXDjGpd>fgAdo9X6nGiF zEPToE;n`P)-{1c-yq)rdficon^S`jZGsD}9d%@0^D3)h<(fg1A2DoCy7``oj#UPp| z%kX;AV+I8YL52q}-h-tds37$`9??re?uf8fGB zxRd#ed>KBS+YIJ|6o42Yd1N-oGMKBGWrcYfK#D*LK&F0sc9VghhlAnM*B@ZN{{8=- z;nLmb47}VN4DY{u2aAK`K!$+yF~JxhTR>ib`5opB5FZpKAU?7jC=5V+7MN2&0x*BT z*f75%Yee@76HF55tN$RAVQ~)&6HuB!jvWRx0P_mi z3&;uQ-kV?GL;BhEp3evn`jpnYGF^DuTM`Ky(w`6a;l0ff3PN_uY0G2zD(|B2pJ2 zUX+E$tenTqX>(3@a?Hc@i|38};N$22zxR9I=kWfj>%%o|Fc`46w}(t71Fq82QY4c} zn9XL`Y&L9dZDDhBQ)*bt_e3HAm&*n9wC}pk<#MPq*MLtJ%xx^8q@)BpJ3DeIaMUl) zrOj1rY;2&ctPH!myC^IygnISqo_07K*xug8m+27{6%|2iEdX3B;sv|GEN5e(pTdqT zFST|+-Fe-jEf$L$k}|vUawHPL{{FrIaGOjf*`%M(=TT5lATrd;!Mi#X3JJ~ca}zk# z;1e0aU=VJ%8x<85csJNDEZXXu1?+2ZRKVC&Js-HGNqKQ`5$%m9F&_Sfs;VlHVbvyt zGf@Gd!PK42W?{8jQCwUsTit&X(BA3>(G*Jmk?yHf3Pz(*irIy92~Lw|;vv*q4q+u1 zhar`Rdfaf zgKx9rIO91XCRJBgOXrb=A3~i{YK%5v%cY+6-_&bsYq;n?Cni$6qsS@)#CJ-eVM-|h zTV7s9Yu#~7Ed540orZex@wPVSjYuFX$di$2T=2C^N6UUGd_Es~Klj1!_hWW;7UkvT zGBb>rzNv>j9h%XNE>5A*aRjm5B-Yp0MJuWEN%yIrk{BZw0XR3RXR=GPYb+KML=M&e z*=Gtgw7I!ix>A3O^O(dpJToS2RQK2u&De&fh7U%Vu~GAl&a0T8pNGfeK{y=7>P|vZ zibkVos&)UH1O5x7GoOfvDVN*M3xkE^vH;Rxn#aX+GGJPa{BT7r)T6hZy0WrT4v8EH w1f+oT?ss`5iO#0ect137FjS<@Cj1Hi0=0vO8h?Sj)c^nh07*qoM6N<$f;^FEhyVZp literal 0 HcmV?d00001 diff --git a/legacy/resources/textures/stone.png b/legacy/resources/textures/stone.png new file mode 100644 index 0000000000000000000000000000000000000000..87e19ff46b57f2bb86553459b2ebb2805ebbc55a GIT binary patch literal 223 zcmV<503iQ~P)`Q z@@yAK5Xrp8Ww0?B)}a6jDz5Sm3r@C%T7;e8YL1N+@fol{wcN}(? + +#include "renderer/renderer.hpp" +#include "renderer/texture.hpp" +#include "renderer/shader.hpp" +#include "renderer/camera.hpp" + +#include "world/chunk/chunk.hpp" +#include "world/entity.hpp" +#include "world/world.hpp" +#include "world/block.hpp" + +#include "common.hpp" +#include "config.hpp" + + +Game::Game() { + +} + +void Game::Setup(int w, int h) { + + m_logger = std::make_shared(); + + *m_logger << "----------------" << LOGGER_ENDL; + *m_logger << "Minecraft 1.14.2" << LOGGER_ENDL; + *m_logger << "----------------" << LOGGER_ENDL; + *m_logger << LOGGER_ENDL; + +#ifdef __DEBUG + *m_logger << LOGGER_DEBUG << "Debug mode enabled" << LOGGER_ENDL; +#endif + + *m_logger << LOGGER_INFO << "Initializing display" << LOGGER_ENDL; + SDL_Init(SDL_INIT_VIDEO | SDL_INIT_AUDIO); + + SDL_GL_SetAttribute(SDL_GL_RED_SIZE, 8); + SDL_GL_SetAttribute(SDL_GL_GREEN_SIZE, 8); + SDL_GL_SetAttribute(SDL_GL_BLUE_SIZE, 8); + SDL_GL_SetAttribute(SDL_GL_ALPHA_SIZE, 8); + SDL_GL_SetAttribute(SDL_GL_BUFFER_SIZE, 32); + SDL_GL_SetAttribute(SDL_GL_DOUBLEBUFFER, 1); + + SDL_GL_SetAttribute(SDL_GL_MULTISAMPLEBUFFERS, 1); + SDL_GL_SetAttribute(SDL_GL_MULTISAMPLESAMPLES, 4); + + SDL_GL_SetAttribute(SDL_GL_CONTEXT_MAJOR_VERSION, 4); + SDL_GL_SetAttribute(SDL_GL_CONTEXT_MINOR_VERSION, 5); + + // Create GL window + *m_logger << LOGGER_INFO << "Creating window" << LOGGER_ENDL; + m_window = SDL_CreateWindow("Minecraft 1.14.2", + SDL_WINDOWPOS_CENTERED, + SDL_WINDOWPOS_CENTERED, w, h, + SDL_WINDOW_OPENGL | SDL_WINDOW_RESIZABLE); + + // Create GL context + *m_logger << LOGGER_INFO << "Creating OpenGL context" << LOGGER_ENDL; + m_glContext = SDL_GL_CreateContext(m_window); + + SDL_SetRelativeMouseMode(SDL_TRUE); + + // Set VSYNC swap interval + SDL_GL_SetSwapInterval(1); + + *m_logger << LOGGER_INFO << "Display set up" << LOGGER_ENDL; + + // Load OpenGL + gladLoadGLLoader(SDL_GL_GetProcAddress); + glEnable(GL_MULTISAMPLE); + // glEnable(GL_CULL_FACE); + glCullFace(GL_BACK); + glEnable(GL_DEPTH_TEST); + // glPolygonMode(GL_FRONT_AND_BACK, GL_LINE); + + *m_logger << LOGGER_INFO << "Loaded OpenGL" << LOGGER_ENDL; + *m_logger << LOGGER_ENDL; + IsDisplayOpen = true; + + std::shared_ptr playercamera = std::make_shared(w, h); + m_player = std::make_shared(glm::vec3(0), glm::vec3(0), playercamera); + + std::shared_ptr BlockDictionary = CBlockDictionary::GetInstance(); + + BlockDictionary->Build(); + + m_world = std::make_shared(); + + Texture texture; + m_world->SetTextureMap(texture.LoadTextures(BlockDictionary->Textures)); + + m_world->LoadWorld(); + +} + +void Game::Input(SDL_Event* e) { + + Uint8* state = (Uint8*)SDL_GetKeyboardState(NULL); + + while (SDL_PollEvent(e)) { + + + switch (e->type) { + + case SDL_KEYDOWN: + { + + if (e->key.keysym.sym == SDLK_ESCAPE) { + + IsMouseActive = !IsMouseActive; + + if (IsMouseActive) + SDL_SetRelativeMouseMode(SDL_TRUE); + else + SDL_SetRelativeMouseMode(SDL_FALSE); + + } + + break; + + } + + case SDL_WINDOWEVENT: + { + + if (e->window.event == SDL_WINDOWEVENT_RESIZED) { + + m_player->CameraUpdateProjection(e->window.data1, e->window.data2); + glViewport(0, 0, e->window.data1, e->window.data2); + + } + + break; + + } + + case SDL_QUIT: + { + + IsDisplayOpen = false; + + break; + + } + + } + + if (IsMouseActive) m_player->HandleMouseSDL(*e); + + } + + m_player->MoveSDL(state); + +} + +void Game::Run() { + + SDL_Event e; + + const float clear[] = { 186.0f / 255.0f, 214.0f / 255.0f, 254.0f / 255.0f }; + + m_renderer = std::make_unique(); + + while (IsDisplayOpen) { + + Input(&e); + + glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT); + glClearBufferfv(GL_COLOR, 0, clear); + + m_world->Update(m_player); + m_renderer->Render(m_world, m_player); + + SDL_GL_SwapWindow(m_window); + + } + +} diff --git a/legacy/src/game.hpp b/legacy/src/game.hpp new file mode 100644 index 0000000..daee357 --- /dev/null +++ b/legacy/src/game.hpp @@ -0,0 +1,55 @@ +#ifndef MINECRAFT_GAME_H_ +#define MINECRAFT_GAME_H_ + +#ifdef NDEBUG +#define __DEBUG +#endif + +#define __DEBUG +// #define __IMGUI + +#include +#include +#include + +#if _WIN32 +#include +#else +#include +#endif + +class Logger; + +class Renderer; +class Camera; + +class Player; +class World; + +class Game { +public: + Game(); + + void Setup(int w, int h); + + void Input(SDL_Event* e); + + void Run(); + + bool IsDisplayOpen = false; + bool IsMouseActive = true; + +private: + SDL_Window* m_window = nullptr; + SDL_GLContext m_glContext = nullptr; + + std::shared_ptr m_logger; + + std::shared_ptr m_renderer; + std::shared_ptr m_world; + + std::shared_ptr m_player; + +}; + +#endif diff --git a/legacy/src/main.cpp b/legacy/src/main.cpp new file mode 100644 index 0000000..6fc4fe4 --- /dev/null +++ b/legacy/src/main.cpp @@ -0,0 +1,13 @@ +#include + +#include "game.hpp" + +int main(int argc, char** argv) { + + Game game; + game.Setup(1080, 720); + game.Run(); + + return 0; + +} diff --git a/legacy/src/physics/collider.cpp b/legacy/src/physics/collider.cpp new file mode 100644 index 0000000..07fda40 --- /dev/null +++ b/legacy/src/physics/collider.cpp @@ -0,0 +1,35 @@ +#include "collider.hpp" + +EntityCollider::EntityCollider() { + +} + +glm::vec3 EntityCollider::TerrainCollide(std::vector terrain) { + + + +} + +bool EntityCollider::m_aabb(ColliderBox a, ColliderBox b) { + + return { + (a.Min.x <= b.Min.x + b.Max.x && a.Min.x + a.Max.x >= b.Min.x) && + (a.Min.y <= b.Min.y + b.Max.y && a.Min.y + a.Max.y >= b.Min.y) && + (a.Min.z <= b.Min.z + b.Max.z && a.Min.z + a.Max.z >= b.Min.z) + }; + +} + +float EntityCollider::m_xDepth(ColliderBox a, ColliderBox b) { + +} + + +float EntityCollider::m_yDepth(ColliderBox a, ColliderBox b) { + +} + + +float EntityCollider::m_zDepth(ColliderBox a, ColliderBox b) { + +} diff --git a/legacy/src/physics/collider.hpp b/legacy/src/physics/collider.hpp new file mode 100644 index 0000000..49e24f6 --- /dev/null +++ b/legacy/src/physics/collider.hpp @@ -0,0 +1,39 @@ +#ifndef MINECRAFT_PHYSICS_COLLIDER_H_ +#define MINECRAFT_PHYSICS_COLLIDER_H_ + +#include "../common.hpp" + +class ColliderBox { +public: + glm::vec3 Min; + glm::vec3 Max; +}; + +class Collider : public ColliderBox { +public: + +}; + +// TODO: Trees +class EntityCollider { +public: + + EntityCollider(); + + // Surrounding blocks indexed XYZ + // Returns point of collision + glm::vec3 TerrainCollide(std::vector surroundingBlocks); + + ColliderBox Bounds; + +private: + + bool m_aabb(ColliderBox a, ColliderBox b); + + float m_xDepth(ColliderBox a, ColliderBox b); + float m_yDepth(ColliderBox a, ColliderBox b); + float m_zDepth(ColliderBox a, ColliderBox b); + +}; + +#endif diff --git a/legacy/src/renderer/renderer.cpp b/legacy/src/renderer/renderer.cpp new file mode 100644 index 0000000..f927606 --- /dev/null +++ b/legacy/src/renderer/renderer.cpp @@ -0,0 +1,16 @@ +#include "renderer.hpp" + +#include "../world/chunk/chunk.hpp" +#include "../world/world.hpp" +#include "shader.hpp" + +Renderer::Renderer() { + +} + +// Perform the render passes +void Renderer::Render(std::shared_ptr world, std::shared_ptr entity) { + + world->Render(entity); + +} diff --git a/legacy/src/renderer/renderer.hpp b/legacy/src/renderer/renderer.hpp new file mode 100644 index 0000000..3edbec3 --- /dev/null +++ b/legacy/src/renderer/renderer.hpp @@ -0,0 +1,18 @@ +#ifndef MINECRAFT_RENDERER_RENDERER_H_ +#define MINECRAFT_RENDERER_RENDERER_H_ + +#include "../common.hpp" + +class Entity; +class World; + +// Does GL render passes then returns to the game loop +class Renderer { +public: + Renderer(); + + void Render(std::shared_ptr world, std::shared_ptr entity); + +}; + +#endif diff --git a/legacy/src/util/filereader.cpp b/legacy/src/util/filereader.cpp new file mode 100644 index 0000000..8228074 --- /dev/null +++ b/legacy/src/util/filereader.cpp @@ -0,0 +1,15 @@ +#include "filereader.hpp" + +#include + +FileReader::FileReader() { + +} + +std::string FileReader::LoadTextFromFile(std::string path) { + std::ifstream t(path); + std::string text((std::istreambuf_iterator(t)), + std::istreambuf_iterator()); + return text; +} + diff --git a/legacy/src/util/filereader.hpp b/legacy/src/util/filereader.hpp new file mode 100644 index 0000000..dc1cf24 --- /dev/null +++ b/legacy/src/util/filereader.hpp @@ -0,0 +1,14 @@ +#ifndef MINECRAFT_UTIL_FILEREADER_H_ +#define MINECRAFT_UTIL_FILEREADER_H_ + +#include + +class FileReader { +public: + FileReader(); + + std::string LoadTextFromFile(std::string path); + +}; + +#endif diff --git a/legacy/src/world/block.cpp b/legacy/src/world/block.cpp new file mode 100644 index 0000000..39b400d --- /dev/null +++ b/legacy/src/world/block.cpp @@ -0,0 +1,52 @@ +#include "block.hpp" + +#include "../config.hpp" + +#include + +std::shared_ptr CBlockDictionary::Instance; + +std::shared_ptr CBlockDictionary::GetInstance() { + + if (!CBlockDictionary::Instance) { + + CBlockDictionary::Instance = std::make_shared(); + + } + + return CBlockDictionary::Instance; + +} + + +void CBlockDictionary::Build() { + + // Order matters ! + RegisterTexture("stone.png"); + RegisterTexture("dirt.png"); + RegisterTexture("grass_side.png"); + RegisterTexture("grass_top.png"); + RegisterTexture("cobblestone.png"); + RegisterTexture("bedrock.png"); + + // Texture winding order - top, bottom, left, right, front, back + RegisterBlock(EBlockType::Air, { }); + RegisterBlock(EBlockType::Stone, { EFaceTexture::Stone, EFaceTexture::Stone, EFaceTexture::Stone, EFaceTexture::Stone, EFaceTexture::Stone, EFaceTexture::Stone }); + RegisterBlock(EBlockType::Dirt, { EFaceTexture::Dirt, EFaceTexture::Dirt, EFaceTexture::Dirt, EFaceTexture::Dirt, EFaceTexture::Dirt, EFaceTexture::Dirt }); + RegisterBlock(EBlockType::Grass, { EFaceTexture::Grass, EFaceTexture::Dirt, EFaceTexture::GrassSide, EFaceTexture::GrassSide, EFaceTexture::GrassSide, EFaceTexture::GrassSide }); + RegisterBlock(EBlockType::Cobblestone, { EFaceTexture::Cobblestone, EFaceTexture::Cobblestone, EFaceTexture::Cobblestone, EFaceTexture::Cobblestone, EFaceTexture::Cobblestone, EFaceTexture::Cobblestone }); + RegisterBlock(EBlockType::Bedrock, { EFaceTexture::Bedrock, EFaceTexture::Bedrock, EFaceTexture::Bedrock, EFaceTexture::Bedrock, EFaceTexture::Bedrock, EFaceTexture::Bedrock }); + +} + +void CBlockDictionary::RegisterTexture(std::string texture) { + + Textures.push_back(texture); + +} + +void CBlockDictionary::RegisterBlock(EBlockType::Block block, std::vector faceTextures) { + + BlockEntries[block] = std::make_shared((uint8_t)block, faceTextures); + +} diff --git a/legacy/src/world/block.hpp b/legacy/src/world/block.hpp new file mode 100644 index 0000000..ea7686e --- /dev/null +++ b/legacy/src/world/block.hpp @@ -0,0 +1,86 @@ +#ifndef MINECRAFT_WORLD_BLOCK_H_ +#define MINECRAFT_WORLD_BLOCK_H_ + +#include "../common.hpp" + + +namespace EBlockType { + + enum Block : uint8_t { + + Air = 0, + Stone, + Grass, + Dirt, + Cobblestone, + Bedrock + + }; + +} + +namespace EFaceTexture { + + enum Texture : uint16_t { + + Stone, + Dirt, + GrassSide, + Grass, + Cobblestone, + Bedrock + + }; + +} + +// Texture winding order - top, bottom, left, right, front, back +class CBlockEntry { +public: + + CBlockEntry(uint8_t id, std::vector faceTextures) + : ID(id), FaceTextures(faceTextures) { } + + uint8_t ID; + std::vector FaceTextures; + +}; + +// TODO: Make design of the class data oriented +// ie, import all the data used in the build from +// files and that +class CBlockDictionary { +public: + + static std::shared_ptr GetInstance(); + + static std::shared_ptr Instance; + +public: + + void Build(); + + // The index of the texutres path in this array is equal to + // that textures ID, to be referenced in the block entry + std::vector Textures; + + // Only supports up to 255 blocs, 0 being air + // word stores vectors of chunks which are 16x16x256 + // vectors of uint8_t which reference the block dictionary + std::map> BlockEntries; + + // Expects textures to be inserted in order, 0-... + void RegisterTexture(std::string texture); + + void RegisterBlock(EBlockType::Block block, std::vector faceTextures); + +}; + + +// static std::vector> TextureIdsAndPaths { +// {0, "dirt.png"}, +// {1, "grass_side.png"}, +// {2, "grass_top.png"} +// }; + +#endif diff --git a/legacy/src/world/chunk/chunk.cpp b/legacy/src/world/chunk/chunk.cpp new file mode 100644 index 0000000..d46c902 --- /dev/null +++ b/legacy/src/world/chunk/chunk.cpp @@ -0,0 +1,295 @@ +#include "chunk.hpp" +#include "voxel.hpp" + +#include "../../renderer/shader.hpp" +#include "../../renderer/camera.hpp" + +#include "../block.hpp" + +#include "../../util/fastnoise.hpp" + +#include + +static std::default_random_engine generator; + +Chunk::Chunk() { + +} + +Chunk::Chunk(int x, int z) { + + X = x, Z = z; + + Load(); + +} + +Chunk::Chunk(int x, int z, std::vector voxels) { + + X = x, Z = z; + + Voxels = voxels; + + Load(); + +} + +Chunk::Chunk(int x, int z, std::shared_ptr terrainGenerator) { + + X = x, Z = z; + int y; + + for (x = 0; x < CHUNK_WIDTH; x++) + for (y = 0; y < CHUNK_HEIGHT; y++) + for (z = 0; z < CHUNK_DEPTH; z++) { + + if (y == 0) { + Voxels.push_back((uint8_t)EBlockType::Bedrock); + continue; + } + + if (y == 1 && (float)rand() / (float)RAND_MAX > 0.5f) { + Voxels.push_back((uint8_t)EBlockType::Bedrock); + continue; + } + + if (pow((y / (float)CHUNK_HEIGHT), 1.1024f) + terrainGenerator->GetNoise(x + (Z * CHUNK_WIDTH), y, z + (X * CHUNK_DEPTH)) * 0.40f < 0.5f) { + + Voxels.push_back((uint8_t)EBlockType::Grass); + continue; + + } + + Voxels.push_back((uint8_t)EBlockType::Air); + + } + + for (x = 0; x < CHUNK_WIDTH; x++) + for (y = 0; y < CHUNK_HEIGHT; y++) + for (z = 0; z < CHUNK_DEPTH; z++) { + + if (BlockAt(x, y, z) == EBlockType::Bedrock) + continue; + + // No need for bounds checking as a closed loop + if (BlockAt(x, y + 1, z) == EBlockType::Grass) + Voxels[x + CHUNK_WIDTH * (y + CHUNK_HEIGHT * z)] = EBlockType::Dirt; + + } + + // Add stone 3 layers below dirt + for (x = 0; x < CHUNK_WIDTH; x++) + for (y = 0; y < CHUNK_HEIGHT; y++) + for (z = 0; z < CHUNK_DEPTH; z++) { + + if (BlockAt(x, y, z) == EBlockType::Bedrock) + continue; + + if (BlockAt(x, y + 1, z) == EBlockType::Dirt) + if (BlockAt(x, y + 2, z) == EBlockType::Dirt) + // if (BlockAt(x, y + 3, z) == EBlockType::Dirt) + Voxels[x + CHUNK_WIDTH * (y + CHUNK_HEIGHT * z)] = EBlockType::Stone; + + } + + // Add the rest of the stone + for (x = 0; x < CHUNK_WIDTH; x++) + for (y = 0; y < CHUNK_HEIGHT; y++) + for (z = 0; z < CHUNK_DEPTH; z++) { + + if (BlockAt(x, y, z) == EBlockType::Bedrock) + continue; + + if (BlockAt(x, y + 1, z) == EBlockType::Stone) + Voxels[x + CHUNK_WIDTH * (y + CHUNK_HEIGHT * z)] = EBlockType::Stone; + + } + + Load(); + +} + +void Chunk::Load() { + + if (Loaded) + return; + + m_model = glm::translate(glm::mat4(1.0f), { X * CHUNK_WIDTH, 0, Z * CHUNK_DEPTH }); + + if (!Voxels.empty()) { + m_mesh(); + Loaded = true; + return; + } + + // Generate a superflat chunk if nothing is there + // [x + WIDTH * (y + HEIGHT * z)] + for (int x = 0; x < CHUNK_WIDTH; x++) + for (int y = 0; y < CHUNK_HEIGHT; y++) + for (int z = 0; z < CHUNK_DEPTH; z++) { + + if (y > 32) { + Voxels.push_back((uint8_t)EBlockType::Air); + continue; + } + + if (y == 0) + Voxels.push_back((uint8_t)EBlockType::Bedrock); + else if (y < 28) + Voxels.push_back((uint8_t)EBlockType::Stone); + else if (y < 32) + Voxels.push_back((uint8_t)EBlockType::Dirt); + else + Voxels.push_back((uint8_t)EBlockType::Grass); + + } + + m_mesh(); + Loaded = true; + +} + +void Chunk::Unload() { + + m_vertices.clear(); + m_uvs.clear(); + + glBindVertexArray(m_vao); + + glDeleteBuffers(1, &m_vbo); + glDeleteVertexArrays(1, &m_vao); + + Loaded = false; + +} + +void Chunk::UploadMesh() { + + if (!MeshReady || !Loaded) + return; + + glGenVertexArrays(1, &m_vao); + glBindVertexArray(m_vao); + + glGenBuffers(1, &m_vbo); + glBindBuffer(GL_ARRAY_BUFFER, m_vbo); + + std::vector data; + data.insert(data.end(), m_vertices.begin(), m_vertices.end()); + data.insert(data.end(), m_uvs.begin(), m_uvs.end()); + + m_numVerts = m_vertices.size(); + + glBufferData(GL_ARRAY_BUFFER, data.size() * sizeof(glm::vec3), &data[0], GL_STATIC_DRAW); + + glEnableVertexAttribArray(0); + glVertexAttribPointer(0, 3, GL_FLOAT, GL_FALSE, 0, (const void*)0); + + glEnableVertexAttribArray(1); + glVertexAttribPointer(1, 3, GL_FLOAT, GL_FALSE, 0, (const void*)(m_vertices.size() * sizeof(glm::vec3))); + + m_vertices.clear(); + m_uvs.clear(); + + data.clear(); + + glBindVertexArray(0); + + MeshReady = !MeshReady; + +} + +void Chunk::Render(std::shared_ptr camera, std::shared_ptr shader) { + + if (!MeshReady || !Loaded) + return; + + shader->Use(); + glBindVertexArray(m_vao); + + GLint uniTrans = glGetUniformLocation(shader->Program, "model"); + glUniformMatrix4fv(uniTrans, 1, GL_FALSE, glm::value_ptr(m_model)); + + GLint uniView = glGetUniformLocation(shader->Program, "view"); + glUniformMatrix4fv(uniView, 1, GL_FALSE, glm::value_ptr(camera->GetViewMatrix())); + + GLint uniProj = glGetUniformLocation(shader->Program, "proj"); + glUniformMatrix4fv(uniProj, 1, GL_FALSE, glm::value_ptr(camera->GetProjectionMatrix())); + + glDrawArrays(GL_TRIANGLES, 0, m_numVerts); + +} + +void Chunk::Update(std::vector voxels) { + + Voxels = voxels; + m_mesh(); + +} + +uint8_t Chunk::BlockAt(int x, int y, int z) { + + if (x > CHUNK_WIDTH - 1) return 0; + if (y > CHUNK_HEIGHT - 1) return 0; + if (z > CHUNK_DEPTH - 1) return 0; + + if (x < 0) return 0; + if (y < 0) return 0; + if (z < 0) return 0; + + return Voxels[x + CHUNK_WIDTH * (y + CHUNK_HEIGHT * z)]; + +} + +void Chunk::m_mesh() { + + // TODO: Use greedy meshing for MAXIMUM performance + for (int x = 0; x < CHUNK_WIDTH; x++) + for (int y = 0; y < CHUNK_HEIGHT; y++) + for (int z = 0; z < CHUNK_DEPTH; z++) { + + std::vector tempVerts; + std::vector tempUVs; + + uint8_t block = BlockAt(x, y, z); + + if (block == EBlockType::Air) continue; + + Voxel tmp({x, y, z}, block); + + if (BlockAt(x + 1, y, z) == EBlockType::Air) + tmp.AddFace(EFaceType::Right); + + if (BlockAt(x - 1, y, z) == EBlockType::Air) + tmp.AddFace(EFaceType::Left); + + if (BlockAt(x, y + 1, z) == EBlockType::Air) + tmp.AddFace(EFaceType::Top); + + if (BlockAt(x, y - 1, z) == EBlockType::Air) + tmp.AddFace(EFaceType::Bottom); + + if (BlockAt(x, y, z + 1) == EBlockType::Air) + tmp.AddFace(EFaceType::Front); + + if (BlockAt(x, y, z - 1) == EBlockType::Air) + tmp.AddFace(EFaceType::Back); + + tmp.GetMesh(tempVerts, tempUVs); + + m_vertices.insert(m_vertices.end(), tempVerts.begin(), tempVerts.end()); + m_uvs.insert(m_uvs.end(), tempUVs.begin(), tempUVs.end()); + + tmp.Clear(); + + } + + MeshReady = true; + +} + +Chunk::~Chunk() { + + Unload(); + +} diff --git a/legacy/src/world/chunk/chunk.hpp b/legacy/src/world/chunk/chunk.hpp new file mode 100644 index 0000000..0422df7 --- /dev/null +++ b/legacy/src/world/chunk/chunk.hpp @@ -0,0 +1,70 @@ +#ifndef MINECRAFT_RENDERER_CHUNK_H_ +#define MINECRAFT_RENDERER_CHUNK_H_ + +#include "../../common.hpp" + +#define CHUNK_HEIGHT 128 +#define CHUNK_WIDTH 16 +#define CHUNK_DEPTH 16 + +class FastNoise; + +class Camera; +class Shader; + +class Voxel; + +class Chunk { +public: + + Chunk(); + Chunk(int x, int z); + Chunk(int x, int z, std::vector voxels); + Chunk(int x, int z, std::shared_ptr terrainGenerator); + + void Load(); + void Unload(); + + void UploadMesh(); + bool MeshReady = false; + + void Render(std::shared_ptr camera, std::shared_ptr shader); + + void Update(std::vector voxels); + + uint8_t BlockAt(int x, int y, int z); + + // Indexed sequentially [x + WIDTH * (y + HEIGHT * z)] = voxelID + // the voxel id is used to index the block dictionary to get properties + // to generate a mesh and send it to the GPU + std::vector Voxels; + + // To only be changed by the class its self + bool Loaded = false; + // To only be changed by render components + bool ShouldRender = false; + + // Chunk World pos + int X,Z; + + ~Chunk(); + +private: + + void m_mesh(); + + GLuint m_vao = 0; + GLuint m_vbo = 0; + + // Must be translated by a multiple of 16 in the x or z, nothing in y + glm::mat4 m_model; + + std::vector m_vertices; + int m_numVerts = 0; + + std::vector m_uvs; + + +}; + +#endif diff --git a/legacy/src/world/chunk/face.hpp b/legacy/src/world/chunk/face.hpp new file mode 100644 index 0000000..20dc525 --- /dev/null +++ b/legacy/src/world/chunk/face.hpp @@ -0,0 +1,127 @@ +#ifndef MINECRAFT_RENDERER_FACE_H_ +#define MINECRAFT_RENDERER_FACE_H_ + +#include "../../common.hpp" + +namespace EFaceType { + + enum Face : uint8_t { + Top, + Bottom, + Left, + Right, + Front, + Back, + }; + +} + +static std::vector CubeTopFace = { + { -0.5f, 0.5f, -0.5f }, + { 0.5f, 0.5f, -0.5f }, + { 0.5f, 0.5f, 0.5f }, + { 0.5f, 0.5f, 0.5f }, + { -0.5f, 0.5f, 0.5f }, + { -0.5f, 0.5f, -0.5f } +}; + +static std::vector CubeTopFaceUVs = { + { 0.0f, 0.0f }, + { 1.0f, 0.0f }, + { 1.0f, 1.0f }, + { 1.0f, 1.0f }, + { 0.0f, 1.0f }, + { 0.0f, 0.0f } +}; + +static std::vector CubeBottomFace = { + { -0.5f, -0.5f, -0.5f }, + { 0.5f, -0.5f, -0.5f }, + { 0.5f, -0.5f, 0.5f }, + { 0.5f, -0.5f, 0.5f }, + { -0.5f, -0.5f, 0.5f }, + { -0.5f, -0.5f, -0.5f } +}; + +static std::vector CubeBottomFaceUVs = { + { 0.0f, 0.0f }, + { 1.0f, 0.0f }, + { 1.0f, 1.0f }, + { 1.0f, 1.0f }, + { 0.0f, 1.0f }, + { 0.0f, 0.0f } +}; + +static std::vector CubeLeftFace = { + { -0.5f, 0.5f, 0.5f }, + { -0.5f, 0.5f, -0.5f }, + { -0.5f, -0.5f, -0.5f }, + { -0.5f, -0.5f, -0.5f }, + { -0.5f, -0.5f, 0.5f }, + { -0.5f, 0.5f, 0.5f } +}; + +static std::vector CubeLeftFaceUVs = { + { 0.0f, 0.0f }, + { 1.0f, 0.0f }, + { 1.0f, 1.0f }, + { 1.0f, 1.0f }, + { 0.0f, 1.0f }, + { 0.0f, 0.0f } +}; + +static std::vector CubeRightFace = { + { 0.5f, 0.5f, 0.5f }, + { 0.5f, 0.5f, -0.5f }, + { 0.5f, -0.5f, -0.5f }, + { 0.5f, -0.5f, -0.5f }, + { 0.5f, -0.5f, 0.5f }, + { 0.5f, 0.5f, 0.5f }, +}; + +static std::vector CubeRightFaceUVs = { + { 0.0f, 0.0f }, + { 1.0f, 0.0f }, + { 1.0f, 1.0f }, + { 1.0f, 1.0f }, + { 0.0f, 1.0f }, + { 0.0f, 0.0f } +}; + +static std::vector CubeFrontFace = { + { -0.5f, -0.5f, 0.5f }, + { 0.5f, -0.5f, 0.5f }, + { 0.5f, 0.5f, 0.5f }, + { 0.5f, 0.5f, 0.5f }, + { -0.5f, 0.5f, 0.5f }, + { -0.5f, -0.5f, 0.5f } +}; + +static std::vector CubeFrontFaceUVs = { + { 1.0f, 1.0f }, + { 0.0f, 1.0f }, + { 0.0f, 0.0f }, + { 0.0f, 0.0f }, + { 1.0f, 0.0f }, + { 1.0f, 1.0f } +}; + +static std::vector CubeBackFace = { + { -0.5f, -0.5f, -0.5f }, + { 0.5f, -0.5f, -0.5f }, + { 0.5f, 0.5f, -0.5f }, + { 0.5f, 0.5f, -0.5f }, + { -0.5f, 0.5f, -0.5f }, + { -0.5f, -0.5f, -0.5f } +}; + +static std::vector CubeBackFaceUVs = { + { 1.0f, 1.0f }, + { 0.0f, 1.0f }, + { 0.0f, 0.0f }, + { 0.0f, 0.0f }, + { 1.0f, 0.0f }, + { 1.0f, 1.0f } +}; + +#endif diff --git a/legacy/src/world/chunk/voxel.cpp b/legacy/src/world/chunk/voxel.cpp new file mode 100644 index 0000000..b0ff73f --- /dev/null +++ b/legacy/src/world/chunk/voxel.cpp @@ -0,0 +1,132 @@ +#include "voxel.hpp" + +#include +#include + +#include "../../renderer/shader.hpp" +#include "../../renderer/camera.hpp" + +#include "face.hpp" + +#include "../block.hpp" + +Voxel::Voxel(glm::vec3 coordsInChunk, uint8_t block) { + + // Texture winding order - top, bottom, left, right, front, back + + Block = block; + m_coordsInChunk = coordsInChunk; + +} + +void Voxel::AddFace(EFaceType::Face face) { + + std::vector verts; + std::vector uvs; + + switch (face) { + + case EFaceType::Top: + { + + verts = CubeTopFace; + uvs = CubeTopFaceUVs; + + break; + } + + case EFaceType::Bottom: + { + + verts = CubeBottomFace; + uvs = CubeBottomFaceUVs; + + break; + } + + case EFaceType::Left: + { + + verts = CubeLeftFace; + uvs = CubeLeftFaceUVs; + + break; + } + + case EFaceType::Right: + { + + verts = CubeRightFace; + uvs = CubeRightFaceUVs; + + break; + } + + case EFaceType::Front: + { + + verts = CubeFrontFace; + uvs = CubeFrontFaceUVs; + + break; + } + + case EFaceType::Back: + { + + verts = CubeBackFace; + uvs = CubeBackFaceUVs; + + break; + } + + } + + + verts = m_translateIntoChunk(verts, m_coordsInChunk); + m_vertices.insert(m_vertices.end(), verts.begin(), verts.end()); + + std::shared_ptr block = CBlockDictionary::GetInstance()->BlockEntries[Block]; + + uint16_t tex = block->FaceTextures[(uint16_t)face]; + + std::vector uvws = { + { uvs[0].x, uvs[0].y, (float)tex }, + { uvs[1].x, uvs[1].y, (float)tex }, + { uvs[2].x, uvs[2].y, (float)tex }, + { uvs[3].x, uvs[3].y, (float)tex }, + { uvs[4].x, uvs[4].y, (float)tex }, + { uvs[5].x, uvs[5].y, (float)tex }, + }; + + m_uvs.insert(m_uvs.end(), uvws.begin(), uvws.end()); + +} + +void Voxel::GetMesh(std::vector& verts, std::vector& uvs) { + + verts = m_vertices; + uvs = m_uvs; + +} + +void Voxel::Clear() { + + m_vertices.clear(); + m_uvs.clear(); + +} + +std::vector Voxel::m_translateIntoChunk(std::vector verts, glm::vec3 trans) { + + for (int i = 0; i < verts.size(); i++) { + + verts[i].x += trans.x; + verts[i].y += trans.y; + verts[i].z += trans.z; + + } + + return verts; + +} diff --git a/legacy/src/world/chunk/voxel.hpp b/legacy/src/world/chunk/voxel.hpp new file mode 100644 index 0000000..1e1b7b9 --- /dev/null +++ b/legacy/src/world/chunk/voxel.hpp @@ -0,0 +1,33 @@ +#ifndef MINECRAFT_RENDERER_VOXEL_H_ +#define MINECRAFT_RENDERER_VOXEL_H_ + +#include "../../common.hpp" + +#include "face.hpp" + +class Camera; +class Shader; + +class Voxel { +public: + Voxel(glm::vec3 coordsInChunk, uint8_t block); + + void AddFace(EFaceType::Face face); + void GetMesh(std::vector& verts, std::vector& uvs); + + void Clear(); + + uint8_t Block; + +private: + + glm::vec3 m_coordsInChunk; + + std::vector m_translateIntoChunk(std::vector verts, glm::vec3 trans); + + std::vector m_vertices; + std::vector m_uvs; + +}; + +#endif diff --git a/legacy/src/world/entity.cpp b/legacy/src/world/entity.cpp new file mode 100644 index 0000000..9e14d4f --- /dev/null +++ b/legacy/src/world/entity.cpp @@ -0,0 +1,59 @@ +#include "entity.hpp" + +#include "../renderer/camera.hpp" + +Entity::Entity(glm::vec3 postion, glm::vec3 direction, std::shared_ptr camera) + : Position(Position) + , Direction(direction) + , EntityCamera(camera) + { + + if (EntityCamera) { + EntityCamera->UpdateView(); + } +} + +Player::Player(glm::vec3 position, glm::vec3 direction, std::shared_ptr camera) + : Entity(position, direction, camera) { + + Position = { 0, 64, 0 }; + EntityCamera->Position = { Position.x, Position.y + EyePosition, Position.z }; + EntityCamera->UpdateView(); + +} + +void Player::MoveSDL(Uint8* state) { + + EntityCamera->MoveCamera(state); + Position = EntityCamera->Position; + Position.y -= EyePosition; + +} + +void Player::HandleMouseSDL(SDL_Event e) { + + EntityCamera->HandleMouse(e); + Direction = EntityCamera->LookDirection; + +} + +void Player::UpdatePosition(glm::vec3 position) { + + Position = position; + EntityCamera->UpdatePosition({ Position.x, Position.y + EyePosition, Position.z }); + +} + + +void Player::UpdateDirection(glm::vec3 direction) { + + Direction = direction; + EntityCamera->UpdateLookDirection(direction); + +} + +void Player::CameraUpdateProjection(int xres, int yres) { + + EntityCamera->UpdateProjection(xres, yres); + +} diff --git a/legacy/src/world/entity.hpp b/legacy/src/world/entity.hpp new file mode 100644 index 0000000..ade36b9 --- /dev/null +++ b/legacy/src/world/entity.hpp @@ -0,0 +1,51 @@ +#ifndef MINECRAFT_WORLD_ENTITY_H_ +#define MINECRAFT_WORLD_ENTITY_H_ + +#include "../common.hpp" + +class Camera; + +class Collider; + +class Entity { +public: + + Entity(glm::vec3 position, glm::vec3 direction = { 0.0f, 0.0f, 0.0f }, std::shared_ptr camera = std::make_shared()); + + // World position, 1.7 units below the + // camera position. + glm::vec3 Position; + // Look direction of the camera + glm::vec3 Direction; + // Velocity in direction + // of movement + glm::vec3 Velocity; + + // Can be null + std::shared_ptr EntityCamera; + + // Collider + // std::unique_ptr EntityCollider; + + // Mesh (or reference to) + +}; + +class Player : public Entity { +public: + + Player(glm::vec3 position, glm::vec3 direction, std::shared_ptr camera); + + float EyePosition = 1.7f; + + void MoveSDL(Uint8* state); + void HandleMouseSDL(SDL_Event e); + + void UpdatePosition(glm::vec3 position); + void UpdateDirection(glm::vec3 direction); + + void CameraUpdateProjection(int xres, int yres); + +}; + +#endif diff --git a/legacy/src/world/generator/chunkgenerator.cpp b/legacy/src/world/generator/chunkgenerator.cpp new file mode 100644 index 0000000..435dc81 --- /dev/null +++ b/legacy/src/world/generator/chunkgenerator.cpp @@ -0,0 +1,14 @@ +#include "../../util/fastnoise.hpp" + + +void dp() { + + FastNoise noise; + noise.SetSeed(121212); + + noise.SetNoiseType(FastNoise::SimplexFractal); + + noise.SetFractalOctaves(3); + +} + diff --git a/legacy/src/world/generator/chunkgenerator.hpp b/legacy/src/world/generator/chunkgenerator.hpp new file mode 100644 index 0000000..1425a26 --- /dev/null +++ b/legacy/src/world/generator/chunkgenerator.hpp @@ -0,0 +1,6 @@ +#ifndef MINECRAFT_WORLD_GENERATOR_CHUNKGENERATOR_H_ +#define MINECRAFT_WORLD_GENERATOR_CHUNKGENERATOR_H_ + + + +#endif diff --git a/legacy/src/world/generator/chunkmanager.hpp b/legacy/src/world/generator/chunkmanager.hpp new file mode 100644 index 0000000..8420177 --- /dev/null +++ b/legacy/src/world/generator/chunkmanager.hpp @@ -0,0 +1,28 @@ +#ifndef MINECRAFT_WORLD_GENERATOR_CUNKMANAGER_H_ +#define MINECRAFT_WORLD_GENERATOR_CUNKMANAGER_H_ + +#include "../../common.hpp" + + +class Frustrum; + +class ChunkManager { +public: + + // Instatntiated + ChunkManager(); + + void Update(); + + void Play(); + void Pause(); + + void LoadChunksAroundWorldPoint(glm::vec3 worldPoint); + + + + void CullFrustrumFromRenderQueue(); + +}; + +#endif diff --git a/legacy/src/world/world.cpp b/legacy/src/world/world.cpp new file mode 100644 index 0000000..f7e3fbb --- /dev/null +++ b/legacy/src/world/world.cpp @@ -0,0 +1,191 @@ +#include "world.hpp" + +#include +#include + +#include "chunk/chunk.hpp" + +#include "../renderer/shader.hpp" + +#include "../util/fastnoise.hpp" + +#include "../config.hpp" +#include "entity.hpp" + +World::World() { + +} + +void World::LoadWorld() { + + m_shaders["Basic"] = std::make_shared(); + m_shaders["Basic"]->Load(GameConfig.ResourceBase + "shaders/simple"); + m_shaders["Basic"]->Link(); + + srand(time(NULL)); + + m_noiseGenerator = std::make_shared(); + m_noiseGenerator->SetSeed(rand()); + + m_noiseGenerator->SetNoiseType(FastNoise::ValueFractal); + + m_noiseGenerator->SetFractalOctaves(5); + + // Generate a 54x54 chunk world + for (int x = -4; x < 50; x++) + for (int y = -50; y < 4; y++) { + + m_chunkLoaderQueue.push({ x, y }); + + } + + // Spawn generator threads + for (int i = 0; i < 6; i++) { + + m_generatorThreads.push_back(std::thread([&]() { + + m_loadChunks(); + + })); + + } + + m_generatorRunning = true; + +} + +void World::SetTextureMap(GLuint map) { + + m_textureMapID = map; + +} + +glm::vec3 World::GetChunkCoords(glm::vec3 worldCoords) { + + return { worldCoords.x / static_cast(CHUNK_WIDTH), + worldCoords.y / static_cast(CHUNK_HEIGHT), + worldCoords.z / static_cast(CHUNK_DEPTH) }; + +} + +glm::vec2 World::GetChunk(glm::vec3 worldCoords) { + + return { static_cast(worldCoords.x / CHUNK_WIDTH), static_cast(worldCoords.z / CHUNK_DEPTH) }; + +} + +std::vector> World::GetRenderableChunks() { + + std::vector> chunks; + + for (auto& chunk : m_chunks) { + + // Should the chunk be rendererd ? + if (chunk.second->ShouldRender) { + + m_chunkLoderMutex.lock(); + + if (chunk.second->MeshReady) + chunk.second->UploadMesh(); + + m_chunkLoderMutex.unlock(); + + // If not, add it + chunks.push_back(chunk.second); + + } + + } + + return chunks; + +} + +void World::Update(std::shared_ptr player) { + + // glm::vec2 inChunk = GetChunk(player->Position); + + // if (m_chunks.find(inChunk) == m_chunks.end()) { + + // m_chunkLoderMutex.lock(); + + // m_chunkLoaderQueue.push(inChunk); + + // m_chunkLoderMutex.unlock(); + + // } + + // std::cout << "Position: " << player->Position.x << ":" << player->Position.y << ":" << player->Position.z << std::endl; + // std::cout << "Chunk: " << inChunk.x << ":" << inChunk.y << std::endl << std::endl; + +} + +void World::Render(std::shared_ptr player) { + + glBindTexture(GL_TEXTURE_2D_ARRAY, m_textureMapID); + + std::vector> chunks = GetRenderableChunks(); + + + for (int i = 0; i < chunks.size(); i++) { + + chunks[i]->Render(player->EntityCamera, m_shaders["Basic"]); + + } + +} + +World::~World() { + + m_generatorRunning = false; + + for (int i = 0; i < m_generatorThreads.size(); i++) { + + m_generatorThreads[i].join(); + + } + + for (auto& chunk : m_chunks) { + + chunk.second->Unload(); + + } + +} + +void World::m_loadChunks() { + + while (m_generatorRunning) { + + m_chunkLoderMutex.lock(); + + glm::vec2 coords = m_chunkLoaderQueue.front(); + m_chunkLoaderQueue.pop(); + + m_chunkLoderMutex.unlock(); + + + std::shared_ptr loadingChunk = std::make_shared(coords.x, coords.y, m_noiseGenerator); + loadingChunk->ShouldRender = true; + std::cout << "Loaded chunk " << coords.x << ":" << coords.y << std::endl; + + + m_chunkLoderMutex.lock(); + + m_chunks[coords] = loadingChunk; + + m_chunkLoderMutex.unlock(); + + + while (m_chunkLoaderQueue.empty()) { + + if (!m_generatorRunning) break; + + static std::chrono::milliseconds dura(1); + std::this_thread::sleep_for(dura); + + } + + } + +} diff --git a/legacy/src/world/world.hpp b/legacy/src/world/world.hpp new file mode 100644 index 0000000..771dedc --- /dev/null +++ b/legacy/src/world/world.hpp @@ -0,0 +1,80 @@ +#ifndef MINECRAFT_WORLD_WORLD_H_ +#define MINECRAFT_WORLD_WORLD_H_ + +#include "../common.hpp" + +#include "../renderer/camera.hpp" + +#include "generator/chunkmanager.hpp" +#include "chunk/chunk.hpp" + +#include +#include +#include +#include + +class FastNoise; + +class Shader; +class Entity; + +class World { +public: + + // Default constructor + World(); + + // Preps the render threads and loads all of the shaders + void LoadWorld(); + + void SetTextureMap(GLuint map); + + // Takes world coordinates and gets a chunks coordinates + glm::vec3 GetChunkCoords(glm::vec3 wordCoords); + + // Takes world coordinates and gets the chunk those coorinates + // fall in + glm::vec2 GetChunk(glm::vec3 worldCoords); + + std::vector> GetRenderableChunks(); + + void Update(std::shared_ptr player); + void Render(std::shared_ptr player); + + ~World(); + +private: + + // GL stuff + + // Main texture map id + GLuint m_textureMapID; + + // Shaders indexed by name + std::map> m_shaders; + + + // Threads used for chunk generation + std::vector m_generatorThreads; + bool m_generatorRunning = false; + + + // Chuks + // Indexed by chunk coorinates + std::unordered_map> m_chunks; + + std::mutex m_chunkUpdaterMutex; + std::queue m_chunkUpdatesQueue; + + std::mutex m_chunkLoderMutex; + std::queue m_chunkLoaderQueue; + + // Generator + std::shared_ptr m_noiseGenerator; + + + void m_loadChunks(); + +}; + +#endif diff --git a/src/Rendering/camera.cpp b/src/Rendering/camera.cpp new file mode 100644 index 0000000..ed0f372 --- /dev/null +++ b/src/Rendering/camera.cpp @@ -0,0 +1,209 @@ +#include "camera.hpp" + +Camera::Camera() +{ + + projMatrix = glm::perspective( glm::radians( 45.0f ), 1.0f, 0.1f, 1000.0f ); + + Roll = 0.0f; + Pitch = 0.0f; + Yaw = 0.0f; + + Position = {}; + LookDirection = {}; + + viewMatrix = {}; + + UpdateView(); + +} + +Camera::Camera( int w, int h ) +{ + + projMatrix = glm::perspective( glm::radians( 45.0f ), (float) w / float( h ), 0.1f, 1000.0f ); + + Roll = 0.0f; + Pitch = 0.0f; + Yaw = 0.0f; + + Position = {}; + LookDirection = {}; + + viewMatrix = {}; + + UpdateView(); + +} + +void Camera::UpdateView() +{ + + // roll can be removed + glm::mat4 matRoll = glm::mat4( 1.0f ); //identity matrix; + glm::mat4 matPitch = glm::mat4( 1.0f );//identity matrix + glm::mat4 matYaw = glm::mat4( 1.0f ); //identity matrix + + // roll, pitch and yaw + matRoll = glm::rotate( matRoll, Roll, glm::vec3( 0.0f, 0.0f, 1.0f ) ); + matPitch = glm::rotate( matPitch, Pitch, glm::vec3( 1.0f, 0.0f, 0.0f ) ); + matYaw = glm::rotate( matYaw, Yaw, glm::vec3( 0.0f, 1.0f, 0.0f ) ); + + glm::mat4 rotate = matRoll * matPitch * matYaw; + + glm::mat4 translate = glm::mat4( 1.0f ); + translate = glm::translate( translate, -Position ); + + viewMatrix = rotate * translate; + + // Work out Look Vector + glm::mat4 inverseView = glm::inverse( viewMatrix ); + + LookDirection.x = inverseView[2][0]; + LookDirection.y = inverseView[2][1]; + LookDirection.z = inverseView[2][2]; + +} + +glm::mat4 Camera::GetViewMatrix() +{ + + return viewMatrix; + +} + +glm::mat4 Camera::GetProjectionMatrix() +{ + + return projMatrix; + +} + +void Camera::UpdateProjection( int width, int height ) +{ + + projMatrix = glm::perspective( glm::radians( 45.0f ), (float) width / (float) height, 0.1f, 1000.0f ); + +} + +void Camera::HandleMouse( SDL_Event e ) +{ + + if ( e.type != SDL_MOUSEMOTION ) + return; + + + float mouseDX = e.motion.xrel; + float mouseDY = e.motion.yrel; + + glm::vec2 mouseDelta { mouseDX, mouseDY }; + + MouseMoved( mouseDelta ); + +} + +void Camera::MoveCamera( Uint8* state ) +{ + + float dx = 0; + float dz = 0; + float dy = 0; + + // Rotate by camera direction + glm::mat2 rotate { + cos( Yaw ), -sin( Yaw ), + sin( Yaw ), cos( Yaw ) + }; + + glm::vec2 f( 0.0, 1.0 ); + f = f * rotate; + + if ( state[SDL_SCANCODE_W] ) + { + dz -= f.y; + dx -= f.x; + } + if ( state[SDL_SCANCODE_S] ) + { + dz += f.y; + dx += f.x; + } + if ( state[SDL_SCANCODE_A] ) + { + dz += f.x; + dx += -f.y; + } + if ( state[SDL_SCANCODE_D] ) + { + dz -= f.x; + dx -= -f.y; + } + if ( state[SDL_SCANCODE_SPACE] ) + { + dy += 1; + } + if ( state[SDL_SCANCODE_LSHIFT] ) + { + dy -= 1; + } + + // get current view matrix + glm::mat4 mat = GetViewMatrix(); + glm::vec3 forward( mat[0][2], mat[1][2], mat[2][2] ); + glm::vec3 strafe( mat[0][0], mat[1][0], mat[2][0] ); + + // forward vector must be negative to look forward. + // read :http://in2gpu.com/2015/05/17/view-matrix/ + Position.x += dx * CameraSpeed; + Position.z += dz * CameraSpeed; + Position.y += dy * CameraSpeed; + + // update the view matrix + UpdateView(); + +} + +void Camera::MouseMoved( glm::vec2 mouseDelta ) +{ + + // note that yaw and pitch must be converted to radians. + // this is done in UpdateView() by glm::rotate + Yaw += MouseSensitivity * (mouseDelta.x / 100); + Pitch += MouseSensitivity * (mouseDelta.y / 100); + Pitch = glm::clamp( Pitch, -M_PI / 2, M_PI / 2 ); + + UpdateView(); + +} + +void Camera::UpdatePosition( glm::vec3 position ) +{ + + Position = position; + + UpdateView(); + +} + +void Camera::UpdateEulerLookDirection( float roll, float pitch, float yaw ) +{ + + Roll = roll; Pitch = pitch; Yaw = yaw; + LookDirection.x = cos( Yaw ) * cos( Pitch ); + LookDirection.y = sin( Yaw ) * cos( Pitch ); + LookDirection.z = sin( Pitch ); + + UpdateView(); + +} + +void Camera::UpdateLookDirection( glm::vec3 lookDirection ) +{ + + LookDirection = lookDirection; + Pitch = asin( -lookDirection.y ); + Yaw = atan2( lookDirection.x, lookDirection.z ); + + UpdateView(); + +} diff --git a/src/Rendering/camera.hpp b/src/Rendering/camera.hpp new file mode 100644 index 0000000..f14d0eb --- /dev/null +++ b/src/Rendering/camera.hpp @@ -0,0 +1,55 @@ +#ifndef MINECRAFT_RENDERING_CAMERA_H_ +#define MINECRAFT_RENDERING_CAMERA_H_ + +#include +#include +#define GLM_ENABLE_EXPERIMENTAL +#include +#include + +#if _WIN32 +#include +#else +#include +#endif + +class Camera { +public: + Camera(); + Camera(int w, int h); + + void UpdateView(); + + glm::mat4 GetViewMatrix(); + glm::mat4 GetProjectionMatrix(); + glm::mat4 GetFrustrumMatrix(); + + void UpdateProjection(int width, int height); + + // Keyboard + void MoveCamera(Uint8* state); + // Mouse + void HandleMouse(SDL_Event e); + // Mouse Delta + void MouseMoved(glm::vec2 mouseDelta); + + // Updatable by + float MouseSensitivity = 0.1f; + float CameraSpeed = 2.0f; + + void UpdatePosition(glm::vec3 position); + void UpdateEulerLookDirection(float roll, float pitch, float yaw); + void UpdateLookDirection(glm::vec3 lookDirection); + + glm::vec3 Position = {}; + float Roll, Pitch, Yaw; + glm::vec3 LookDirection = {}; + +private: + + glm::mat4 viewMatrix = {}; + glm::mat4 projMatrix = {}; + +}; + +#endif diff --git a/src/Rendering/face.hpp b/src/Rendering/face.hpp new file mode 100644 index 0000000..885ca4e --- /dev/null +++ b/src/Rendering/face.hpp @@ -0,0 +1,128 @@ +#ifndef MINECRAFT_RENDERING_FACE_H_ +#define MINECRAFT_RENDERING_FACE_H_ + +#include +#include + +namespace EFaceType { + + enum Face : uint8_t { + Top, + Bottom, + Left, + Right, + Front, + Back, + }; + +} + +static std::vector CubeTopFace = { + { -0.5f, 0.5f, -0.5f }, + { 0.5f, 0.5f, -0.5f }, + { 0.5f, 0.5f, 0.5f }, + { 0.5f, 0.5f, 0.5f }, + { -0.5f, 0.5f, 0.5f }, + { -0.5f, 0.5f, -0.5f } +}; + +static std::vector CubeTopFaceUVs = { + { 0.0f, 0.0f }, + { 1.0f, 0.0f }, + { 1.0f, 1.0f }, + { 1.0f, 1.0f }, + { 0.0f, 1.0f }, + { 0.0f, 0.0f } +}; + +static std::vector CubeBottomFace = { + { -0.5f, -0.5f, -0.5f }, + { 0.5f, -0.5f, -0.5f }, + { 0.5f, -0.5f, 0.5f }, + { 0.5f, -0.5f, 0.5f }, + { -0.5f, -0.5f, 0.5f }, + { -0.5f, -0.5f, -0.5f } +}; + +static std::vector CubeBottomFaceUVs = { + { 0.0f, 0.0f }, + { 1.0f, 0.0f }, + { 1.0f, 1.0f }, + { 1.0f, 1.0f }, + { 0.0f, 1.0f }, + { 0.0f, 0.0f } +}; + +static std::vector CubeLeftFace = { + { -0.5f, 0.5f, 0.5f }, + { -0.5f, 0.5f, -0.5f }, + { -0.5f, -0.5f, -0.5f }, + { -0.5f, -0.5f, -0.5f }, + { -0.5f, -0.5f, 0.5f }, + { -0.5f, 0.5f, 0.5f } +}; + +static std::vector CubeLeftFaceUVs = { + { 0.0f, 0.0f }, + { 1.0f, 0.0f }, + { 1.0f, 1.0f }, + { 1.0f, 1.0f }, + { 0.0f, 1.0f }, + { 0.0f, 0.0f } +}; + +static std::vector CubeRightFace = { + { 0.5f, 0.5f, 0.5f }, + { 0.5f, 0.5f, -0.5f }, + { 0.5f, -0.5f, -0.5f }, + { 0.5f, -0.5f, -0.5f }, + { 0.5f, -0.5f, 0.5f }, + { 0.5f, 0.5f, 0.5f }, +}; + +static std::vector CubeRightFaceUVs = { + { 0.0f, 0.0f }, + { 1.0f, 0.0f }, + { 1.0f, 1.0f }, + { 1.0f, 1.0f }, + { 0.0f, 1.0f }, + { 0.0f, 0.0f } +}; + +static std::vector CubeFrontFace = { + { -0.5f, -0.5f, 0.5f }, + { 0.5f, -0.5f, 0.5f }, + { 0.5f, 0.5f, 0.5f }, + { 0.5f, 0.5f, 0.5f }, + { -0.5f, 0.5f, 0.5f }, + { -0.5f, -0.5f, 0.5f } +}; + +static std::vector CubeFrontFaceUVs = { + { 1.0f, 1.0f }, + { 0.0f, 1.0f }, + { 0.0f, 0.0f }, + { 0.0f, 0.0f }, + { 1.0f, 0.0f }, + { 1.0f, 1.0f } +}; + +static std::vector CubeBackFace = { + { -0.5f, -0.5f, -0.5f }, + { 0.5f, -0.5f, -0.5f }, + { 0.5f, 0.5f, -0.5f }, + { 0.5f, 0.5f, -0.5f }, + { -0.5f, 0.5f, -0.5f }, + { -0.5f, -0.5f, -0.5f } +}; + +static std::vector CubeBackFaceUVs = { + { 1.0f, 1.0f }, + { 0.0f, 1.0f }, + { 0.0f, 0.0f }, + { 0.0f, 0.0f }, + { 1.0f, 0.0f }, + { 1.0f, 1.0f } +}; + +#endif diff --git a/src/Rendering/frustrum.cpp b/src/Rendering/frustrum.cpp new file mode 100644 index 0000000..43a2d1a --- /dev/null +++ b/src/Rendering/frustrum.cpp @@ -0,0 +1,3 @@ +#include "frustrum.hpp" + + diff --git a/src/Rendering/frustrum.hpp b/src/Rendering/frustrum.hpp new file mode 100644 index 0000000..a29f37b --- /dev/null +++ b/src/Rendering/frustrum.hpp @@ -0,0 +1,29 @@ +#ifndef MINECRAFT_RENDERING_FRUSTRUM_H_ +#define MINECRAFT_RENDERING_FRUSTRUM_H_ + +namespace EFrustrumPlanes +{ + enum Planes + { + Right, + Left, + Top, + Bottom, + Far, + Near + }; +}; + +class FrustrumPlane +{ +public: + +}; + +class Frustrum +{ +public: + +}; + +#endif diff --git a/src/Rendering/mesh.cpp b/src/Rendering/mesh.cpp new file mode 100644 index 0000000..6056583 --- /dev/null +++ b/src/Rendering/mesh.cpp @@ -0,0 +1,6 @@ +#include "mesh.hpp" + +Mesh::Mesh() +{ + +} diff --git a/src/Rendering/mesh.hpp b/src/Rendering/mesh.hpp new file mode 100644 index 0000000..433cdf2 --- /dev/null +++ b/src/Rendering/mesh.hpp @@ -0,0 +1,27 @@ +#ifndef MINECRAFT_RENDERING_MESH_H_ +#define MINECRAFT_RENDERING_MESH_H_ + +#include "../common.hpp" + +class Vertex +{ +public: + glm::vec3 Position; + glm::vec3 SurfaceNormal; + +}; + +class Mesh +{ +public: + + Mesh(); + +private: + + GLuint mVAO; + GLuint mVBO; + +}; + +#endif diff --git a/src/Rendering/renderable.cpp b/src/Rendering/renderable.cpp new file mode 100644 index 0000000..68a7d31 --- /dev/null +++ b/src/Rendering/renderable.cpp @@ -0,0 +1,66 @@ +#include "renderable.hpp" + +#include + +Renderable::Renderable() +{ + +} + +void Renderable::Init() +{ + +} + +void Renderable::AddMesh( Mesh* mesh ) +{ + if (mesh == nullptr) return; + mMeshs.push_back( mesh ); +} + +void Renderable::RemoveMesh( Mesh* mesh ) +{ + // Renderable does not include mesh + if (std::find( mMeshs.begin(), mMeshs.end(), mesh ) == mMeshs.end()) + return; + + std::remove( mMeshs.begin(), mMeshs.end(), mesh ); + +} + +void Renderable::SetActiveMesh( Mesh* mesh ) +{ + // Renderable does not include mesh + if (std::find( mMeshs.begin(), mMeshs.end(), mesh ) == mMeshs.end()) + return; + + mActiveMesh = mesh; +} + +Mesh* Renderable::GetActiveMesh() +{ + return mActiveMesh; +} + +void Renderable::UpdateBuffer() +{ + +} + +void Renderable::Load() +{ + +} + +void Renderable::Unload() +{ + +} + +Renderable::~Renderable() +{ + Unload(); +} + + + diff --git a/src/Rendering/renderable.hpp b/src/Rendering/renderable.hpp new file mode 100644 index 0000000..7441d8c --- /dev/null +++ b/src/Rendering/renderable.hpp @@ -0,0 +1,45 @@ +#ifndef MINECRAFT_RENDERING_RENDERABLE_H_ +#define MINECRAFT_RENDERING_RENDERABLE_H_ + +#include "../common.hpp" + +class Mesh; + +// Basically a model but thats effort +// perhaps sub-class? +class Renderable +{ +public: + Renderable(); + + void Init(); + + // DOES NOT OWN MESH + void AddMesh( Mesh* ); + void RemoveMesh( Mesh* ); + void SetActiveMesh( Mesh* ); + Mesh* GetActiveMesh(); + + void UpdateBuffer(); + + // GPU Load methods + void Load(); + void Unload(); + + ~Renderable(); + +private: + + std::vector mBuff; + + std::vector mMeshs; + Mesh* mActiveMesh; + + // Meshes have uniform uniforms + GLuint mUBO; + + glm::mat4 mModelMatrix; + +}; + +#endif diff --git a/src/Rendering/rendermaster.cpp b/src/Rendering/rendermaster.cpp new file mode 100644 index 0000000..84719ea --- /dev/null +++ b/src/Rendering/rendermaster.cpp @@ -0,0 +1,11 @@ +#include "rendermaster.hpp" + +RenderMaster::RenderMaster() + : mWorldRenderer(), + mMeshRenderer() +{ + + + +} + diff --git a/src/Rendering/rendermaster.hpp b/src/Rendering/rendermaster.hpp new file mode 100644 index 0000000..9965d23 --- /dev/null +++ b/src/Rendering/rendermaster.hpp @@ -0,0 +1,41 @@ +#ifndef MINECRAFT_RENDERING_RENDERMASTER_H_ +#define MINECRAFT_RENDERING_RENDERMASTER_H_ + +/** + * Renderer Structure + * + * Mesh -> Renderable + * Mesh -> VoxelMesh + * Renderable -> Model + * Renderable -> World (static(?)) + * Renderable -> Entity (dynamic) + * Renderable -> Particle (dynamic) + * + * Kinda just winging it ngl +*/ + +class WorldRenderer +{ + +}; + + +class MeshRenderer +{ + +}; + + +class RenderMaster +{ +public: + RenderMaster(); + + + WorldRenderer mWorldRenderer; + MeshRenderer mMeshRenderer; + + +}; + +#endif diff --git a/src/Rendering/shader.cpp b/src/Rendering/shader.cpp new file mode 100644 index 0000000..a851d0e --- /dev/null +++ b/src/Rendering/shader.cpp @@ -0,0 +1,121 @@ +#include "shader.hpp" + +#include "../utilities.hpp" + +Shader::Shader() +{ + + Program = 0; + mFrag = 0; + mVert = 0; + + mLogger = std::make_shared(); + +} + + +void Shader::Load( std::string path ) +{ + + std::string vertexLocation = path + ".vert"; + Load( vertexLocation, GL_VERTEX_SHADER ); + *mLogger << LOGGER_INFO << "Vertex shader at '" << vertexLocation << "' loaded..." << LOGGER_ENDL; + + + std::string fragmentLocation = path + ".frag"; + Load( fragmentLocation, GL_FRAGMENT_SHADER ); + *mLogger << LOGGER_INFO << "Fragment shader at '" << fragmentLocation << "' loaded..." << LOGGER_ENDL; + +} + + +void Shader::Load( std::string path, GLenum type ) +{ + + GLuint activeShader = 0; + + if ( type == GL_VERTEX_SHADER ) + mVert = activeShader = glCreateShader( type ); + + if ( type == GL_FRAGMENT_SHADER ) + mFrag = activeShader = glCreateShader( type ); + + std::string loadedShaderSource = LoadTextFromFile( path ); + const char* shaderSource = loadedShaderSource.c_str(); + int shaderSourceLength = loadedShaderSource.length(); + + glShaderSource( activeShader, 1, &shaderSource, &shaderSourceLength ); + +} + +void Shader::Link() +{ + + if ( mVert == 0 || mFrag == 0 ) + { + *mLogger << LOGGER_ERROR << "Failed to link programs: Both programs not present" << LOGGER_ENDL; + return; + } + + glCompileShader( mVert ); + if ( mCheckShader( mVert ) ) + { + *mLogger << LOGGER_INFO << "Vertex shader '" << mVert << "' compiled..." << LOGGER_ENDL; + } + + glCompileShader( mFrag ); + if ( mCheckShader( mFrag ) ) + { + *mLogger << LOGGER_INFO << "Fragment shader '" << mFrag << "' compiled..." << LOGGER_ENDL; + } + + Program = glCreateProgram(); + + glAttachShader( Program, mVert ); + glAttachShader( Program, mFrag ); + + glLinkProgram( Program ); + + glDeleteShader( mVert ); + glDeleteShader( mFrag ); + + *mLogger << LOGGER_INFO << "Program '" << Program << "' loaded..." << LOGGER_ENDL; + +} + +void Shader::Use() +{ + + glUseProgram( Program ); + +} + + +bool Shader::mCheckShader( GLuint uid ) +{ + + GLint status = GL_TRUE; + + glGetShaderiv( uid, GL_COMPILE_STATUS, &status ); + + if ( status == GL_FALSE ) + { + char buf[512]; + glGetShaderInfoLog( uid, 512, NULL, buf ); + *mLogger << LOGGER_ERROR << buf << LOGGER_ENDL; + delete buf; + return false; + } + + return true; +} + + +Shader::~Shader() +{ + + glDeleteProgram( Program ); + glDeleteShader( mVert ); + glDeleteShader( mFrag ); + +} diff --git a/src/Rendering/shader.hpp b/src/Rendering/shader.hpp new file mode 100644 index 0000000..7f7e709 --- /dev/null +++ b/src/Rendering/shader.hpp @@ -0,0 +1,30 @@ +#ifndef MINECRAFT_RENDERING_SHADER_H_ +#define MINECRAFT_RENDERING_SHADER_H_ + +#include + +#include + +class Shader { +public: + Shader(); + + void Load(std::string path); + void Load(std::string path, GLenum type); + + GLuint Program; + void Link(); + + void Use(); + + ~Shader(); +private: + std::shared_ptr mLogger; + + bool mCheckShader(GLuint uid); + + GLuint mVert; + GLuint mFrag; +}; + +#endif diff --git a/src/Rendering/texture.cpp b/src/Rendering/texture.cpp new file mode 100644 index 0000000..7b91f10 --- /dev/null +++ b/src/Rendering/texture.cpp @@ -0,0 +1,60 @@ +#include "texture.hpp" + +#include + +#include "../settings.hpp" + +#define STB_IMAGE_IMPLEMENTATION +#include "../ThirdParty/stb_image.hpp" + +GLuint Texture::LoadTextures(std::vector textures) { + + Logger logger; + + std::string basePath = ResourceBase + "textures/"; + + int x = 16; + int y = 16; + + GLsizei layers = textures.size(); + GLubyte* texels = (GLubyte*)malloc(x * y * 4 * layers * sizeof(GLubyte)); + + for (int i = 0; i < layers; i++) { + + std::string path = basePath + textures[i]; + + int xR = 0; + int yR = 0; + int cR = 0; + + unsigned char* texture = stbi_load(path.c_str(), &xR, &yR, &cR, STBI_rgb_alpha); + + memcpy(texels + (i * x * y * 4), texture, x * y * 4); + + stbi_image_free(texture); + logger << LOGGER_INFO << "Texture at '" << path << "' Loaded..." << LOGGER_ENDL; + + } + + GLuint textureArray = 0; + + glGenTextures(1, &textureArray); + glBindTexture(GL_TEXTURE_2D_ARRAY, textureArray); + + glTexImage3D(GL_TEXTURE_2D_ARRAY, 0, GL_RGBA8, x, y, layers, 0, GL_RGBA, GL_UNSIGNED_BYTE, NULL); + + glTexSubImage3D(GL_TEXTURE_2D_ARRAY, 0, 0, 0, 0, x, y, layers, GL_RGBA, GL_UNSIGNED_BYTE, texels); + + glTexParameteri(GL_TEXTURE_2D_ARRAY, GL_TEXTURE_MIN_FILTER, GL_NEAREST); + glTexParameteri(GL_TEXTURE_2D_ARRAY, GL_TEXTURE_MAG_FILTER, GL_NEAREST); + + glTexParameteri(GL_TEXTURE_2D_ARRAY, GL_TEXTURE_WRAP_S, GL_CLAMP_TO_EDGE); + glTexParameteri(GL_TEXTURE_2D_ARRAY, GL_TEXTURE_WRAP_T, GL_CLAMP_TO_EDGE); + + glBindTexture(GL_TEXTURE_2D_ARRAY, 0); + + free(texels); + + return textureArray; + +} diff --git a/src/Rendering/texture.hpp b/src/Rendering/texture.hpp new file mode 100644 index 0000000..cffbcb1 --- /dev/null +++ b/src/Rendering/texture.hpp @@ -0,0 +1,14 @@ +#ifndef MINECRAFT_RENDERING_TEXTURE_H_ +#define MINECRAFT_RENDERING_TEXTURE_H_ + +#include +#include + +#include + +class Texture { +public: + GLuint LoadTextures(std::vector textures); +}; + +#endif diff --git a/src/Rendering/voxelmesh.hpp b/src/Rendering/voxelmesh.hpp new file mode 100644 index 0000000..bb94ecf --- /dev/null +++ b/src/Rendering/voxelmesh.hpp @@ -0,0 +1,23 @@ +#ifndef MINECRAFT_RENDERER_VOXELMESH_H_ +#define MINECRAFT_RENDERER_VOXELMESH_H_ + +#include "mesh.hpp" + +class VoxelMesh : public Mesh +{ +public: + + VoxelMesh(); + + + + int Width; + int Height; + int Depth; + + // Size is w*h*d + std::vector Voxels; + +}; + +#endif diff --git a/src/ThirdParty/fastnoise.cpp b/src/ThirdParty/fastnoise.cpp new file mode 100644 index 0000000..4fdba6b --- /dev/null +++ b/src/ThirdParty/fastnoise.cpp @@ -0,0 +1,2250 @@ +// FastNoise.cpp +// +// MIT License +// +// Copyright(c) 2017 Jordan Peck +// +// Permission is hereby granted, free of charge, to any person obtaining a copy +// of this software and associated documentation files(the "Software"), to deal +// in the Software without restriction, including without limitation the rights +// to use, copy, modify, merge, publish, distribute, sublicense, and / or sell +// copies of the Software, and to permit persons to whom the Software is +// furnished to do so, subject to the following conditions : +// +// The above copyright notice and this permission notice shall be included in all +// copies or substantial portions of the Software. +// +// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.IN NO EVENT SHALL THE +// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +// SOFTWARE. +// +// The developer's email is jorzixdan.me2@gzixmail.com (for great email, take +// off every 'zix'.) +// + +#include "fastnoise.hpp" + +#include +#include + +#include +#include + +const FN_DECIMAL GRAD_X[] = +{ + 1, -1, 1, -1, + 1, -1, 1, -1, + 0, 0, 0, 0 +}; +const FN_DECIMAL GRAD_Y[] = +{ + 1, 1, -1, -1, + 0, 0, 0, 0, + 1, -1, 1, -1 +}; +const FN_DECIMAL GRAD_Z[] = +{ + 0, 0, 0, 0, + 1, 1, -1, -1, + 1, 1, -1, -1 +}; + +const FN_DECIMAL GRAD_4D[] = +{ + 0,1,1,1,0,1,1,-1,0,1,-1,1,0,1,-1,-1, + 0,-1,1,1,0,-1,1,-1,0,-1,-1,1,0,-1,-1,-1, + 1,0,1,1,1,0,1,-1,1,0,-1,1,1,0,-1,-1, + -1,0,1,1,-1,0,1,-1,-1,0,-1,1,-1,0,-1,-1, + 1,1,0,1,1,1,0,-1,1,-1,0,1,1,-1,0,-1, + -1,1,0,1,-1,1,0,-1,-1,-1,0,1,-1,-1,0,-1, + 1,1,1,0,1,1,-1,0,1,-1,1,0,1,-1,-1,0, + -1,1,1,0,-1,1,-1,0,-1,-1,1,0,-1,-1,-1,0 +}; + +const FN_DECIMAL VAL_LUT[] = +{ + FN_DECIMAL(0.3490196078), FN_DECIMAL(0.4352941176), FN_DECIMAL(-0.4509803922), FN_DECIMAL(0.6392156863), FN_DECIMAL(0.5843137255), FN_DECIMAL(-0.1215686275), FN_DECIMAL(0.7176470588), FN_DECIMAL(-0.1058823529), FN_DECIMAL(0.3960784314), FN_DECIMAL(0.0431372549), FN_DECIMAL(-0.03529411765), FN_DECIMAL(0.3176470588), FN_DECIMAL(0.7254901961), FN_DECIMAL(0.137254902), FN_DECIMAL(0.8588235294), FN_DECIMAL(-0.8196078431), + FN_DECIMAL(-0.7960784314), FN_DECIMAL(-0.3333333333), FN_DECIMAL(-0.6705882353), FN_DECIMAL(-0.3882352941), FN_DECIMAL(0.262745098), FN_DECIMAL(0.3254901961), FN_DECIMAL(-0.6470588235), FN_DECIMAL(-0.9215686275), FN_DECIMAL(-0.5294117647), FN_DECIMAL(0.5294117647), FN_DECIMAL(-0.4666666667), FN_DECIMAL(0.8117647059), FN_DECIMAL(0.3803921569), FN_DECIMAL(0.662745098), FN_DECIMAL(0.03529411765), FN_DECIMAL(-0.6156862745), + FN_DECIMAL(-0.01960784314), FN_DECIMAL(-0.3568627451), FN_DECIMAL(-0.09019607843), FN_DECIMAL(0.7490196078), FN_DECIMAL(0.8352941176), FN_DECIMAL(-0.4039215686), FN_DECIMAL(-0.7490196078), FN_DECIMAL(0.9529411765), FN_DECIMAL(-0.0431372549), FN_DECIMAL(-0.9294117647), FN_DECIMAL(-0.6549019608), FN_DECIMAL(0.9215686275), FN_DECIMAL(-0.06666666667), FN_DECIMAL(-0.4431372549), FN_DECIMAL(0.4117647059), FN_DECIMAL(-0.4196078431), + FN_DECIMAL(-0.7176470588), FN_DECIMAL(-0.8117647059), FN_DECIMAL(-0.2549019608), FN_DECIMAL(0.4901960784), FN_DECIMAL(0.9137254902), FN_DECIMAL(0.7882352941), FN_DECIMAL(-1.0), FN_DECIMAL(-0.4745098039), FN_DECIMAL(0.7960784314), FN_DECIMAL(0.8509803922), FN_DECIMAL(-0.6784313725), FN_DECIMAL(0.4588235294), FN_DECIMAL(1.0), FN_DECIMAL(-0.1843137255), FN_DECIMAL(0.4509803922), FN_DECIMAL(0.1450980392), + FN_DECIMAL(-0.231372549), FN_DECIMAL(-0.968627451), FN_DECIMAL(-0.8588235294), FN_DECIMAL(0.4274509804), FN_DECIMAL(0.003921568627), FN_DECIMAL(-0.003921568627), FN_DECIMAL(0.2156862745), FN_DECIMAL(0.5058823529), FN_DECIMAL(0.7647058824), FN_DECIMAL(0.2078431373), FN_DECIMAL(-0.5921568627), FN_DECIMAL(0.5764705882), FN_DECIMAL(-0.1921568627), FN_DECIMAL(-0.937254902), FN_DECIMAL(0.08235294118), FN_DECIMAL(-0.08235294118), + FN_DECIMAL(0.9058823529), FN_DECIMAL(0.8274509804), FN_DECIMAL(0.02745098039), FN_DECIMAL(-0.168627451), FN_DECIMAL(-0.7803921569), FN_DECIMAL(0.1137254902), FN_DECIMAL(-0.9450980392), FN_DECIMAL(0.2), FN_DECIMAL(0.01960784314), FN_DECIMAL(0.5607843137), FN_DECIMAL(0.2705882353), FN_DECIMAL(0.4431372549), FN_DECIMAL(-0.9607843137), FN_DECIMAL(0.6156862745), FN_DECIMAL(0.9294117647), FN_DECIMAL(-0.07450980392), + FN_DECIMAL(0.3098039216), FN_DECIMAL(0.9921568627), FN_DECIMAL(-0.9137254902), FN_DECIMAL(-0.2941176471), FN_DECIMAL(-0.3411764706), FN_DECIMAL(-0.6235294118), FN_DECIMAL(-0.7647058824), FN_DECIMAL(-0.8901960784), FN_DECIMAL(0.05882352941), FN_DECIMAL(0.2392156863), FN_DECIMAL(0.7333333333), FN_DECIMAL(0.6549019608), FN_DECIMAL(0.2470588235), FN_DECIMAL(0.231372549), FN_DECIMAL(-0.3960784314), FN_DECIMAL(-0.05098039216), + FN_DECIMAL(-0.2235294118), FN_DECIMAL(-0.3725490196), FN_DECIMAL(0.6235294118), FN_DECIMAL(0.7019607843), FN_DECIMAL(-0.8274509804), FN_DECIMAL(0.4196078431), FN_DECIMAL(0.07450980392), FN_DECIMAL(0.8666666667), FN_DECIMAL(-0.537254902), FN_DECIMAL(-0.5058823529), FN_DECIMAL(-0.8039215686), FN_DECIMAL(0.09019607843), FN_DECIMAL(-0.4823529412), FN_DECIMAL(0.6705882353), FN_DECIMAL(-0.7882352941), FN_DECIMAL(0.09803921569), + FN_DECIMAL(-0.6078431373), FN_DECIMAL(0.8039215686), FN_DECIMAL(-0.6), FN_DECIMAL(-0.3254901961), FN_DECIMAL(-0.4117647059), FN_DECIMAL(-0.01176470588), FN_DECIMAL(0.4823529412), FN_DECIMAL(0.168627451), FN_DECIMAL(0.8745098039), FN_DECIMAL(-0.3647058824), FN_DECIMAL(-0.1607843137), FN_DECIMAL(0.568627451), FN_DECIMAL(-0.9921568627), FN_DECIMAL(0.9450980392), FN_DECIMAL(0.5137254902), FN_DECIMAL(0.01176470588), + FN_DECIMAL(-0.1450980392), FN_DECIMAL(-0.5529411765), FN_DECIMAL(-0.5764705882), FN_DECIMAL(-0.1137254902), FN_DECIMAL(0.5215686275), FN_DECIMAL(0.1607843137), FN_DECIMAL(0.3725490196), FN_DECIMAL(-0.2), FN_DECIMAL(-0.7254901961), FN_DECIMAL(0.631372549), FN_DECIMAL(0.7098039216), FN_DECIMAL(-0.568627451), FN_DECIMAL(0.1294117647), FN_DECIMAL(-0.3098039216), FN_DECIMAL(0.7411764706), FN_DECIMAL(-0.8509803922), + FN_DECIMAL(0.2549019608), FN_DECIMAL(-0.6392156863), FN_DECIMAL(-0.5607843137), FN_DECIMAL(-0.3176470588), FN_DECIMAL(0.937254902), FN_DECIMAL(0.9843137255), FN_DECIMAL(0.5921568627), FN_DECIMAL(0.6941176471), FN_DECIMAL(0.2862745098), FN_DECIMAL(-0.5215686275), FN_DECIMAL(0.1764705882), FN_DECIMAL(0.537254902), FN_DECIMAL(-0.4901960784), FN_DECIMAL(-0.4588235294), FN_DECIMAL(-0.2078431373), FN_DECIMAL(-0.2156862745), + FN_DECIMAL(0.7725490196), FN_DECIMAL(0.3647058824), FN_DECIMAL(-0.2392156863), FN_DECIMAL(0.2784313725), FN_DECIMAL(-0.8823529412), FN_DECIMAL(0.8980392157), FN_DECIMAL(0.1215686275), FN_DECIMAL(0.1058823529), FN_DECIMAL(-0.8745098039), FN_DECIMAL(-0.9843137255), FN_DECIMAL(-0.7019607843), FN_DECIMAL(0.9607843137), FN_DECIMAL(0.2941176471), FN_DECIMAL(0.3411764706), FN_DECIMAL(0.1529411765), FN_DECIMAL(0.06666666667), + FN_DECIMAL(-0.9764705882), FN_DECIMAL(0.3019607843), FN_DECIMAL(0.6470588235), FN_DECIMAL(-0.5843137255), FN_DECIMAL(0.05098039216), FN_DECIMAL(-0.5137254902), FN_DECIMAL(-0.137254902), FN_DECIMAL(0.3882352941), FN_DECIMAL(-0.262745098), FN_DECIMAL(-0.3019607843), FN_DECIMAL(-0.1764705882), FN_DECIMAL(-0.7568627451), FN_DECIMAL(0.1843137255), FN_DECIMAL(-0.5450980392), FN_DECIMAL(-0.4980392157), FN_DECIMAL(-0.2784313725), + FN_DECIMAL(-0.9529411765), FN_DECIMAL(-0.09803921569), FN_DECIMAL(0.8901960784), FN_DECIMAL(-0.2862745098), FN_DECIMAL(-0.3803921569), FN_DECIMAL(0.5529411765), FN_DECIMAL(0.7803921569), FN_DECIMAL(-0.8352941176), FN_DECIMAL(0.6862745098), FN_DECIMAL(0.7568627451), FN_DECIMAL(0.4980392157), FN_DECIMAL(-0.6862745098), FN_DECIMAL(-0.8980392157), FN_DECIMAL(-0.7725490196), FN_DECIMAL(-0.7098039216), FN_DECIMAL(-0.2470588235), + FN_DECIMAL(-0.9058823529), FN_DECIMAL(0.9764705882), FN_DECIMAL(0.1921568627), FN_DECIMAL(0.8431372549), FN_DECIMAL(-0.05882352941), FN_DECIMAL(0.3568627451), FN_DECIMAL(0.6078431373), FN_DECIMAL(0.5450980392), FN_DECIMAL(0.4039215686), FN_DECIMAL(-0.7333333333), FN_DECIMAL(-0.4274509804), FN_DECIMAL(0.6), FN_DECIMAL(0.6784313725), FN_DECIMAL(-0.631372549), FN_DECIMAL(-0.02745098039), FN_DECIMAL(-0.1294117647), + FN_DECIMAL(0.3333333333), FN_DECIMAL(-0.8431372549), FN_DECIMAL(0.2235294118), FN_DECIMAL(-0.3490196078), FN_DECIMAL(-0.6941176471), FN_DECIMAL(0.8823529412), FN_DECIMAL(0.4745098039), FN_DECIMAL(0.4666666667), FN_DECIMAL(-0.7411764706), FN_DECIMAL(-0.2705882353), FN_DECIMAL(0.968627451), FN_DECIMAL(0.8196078431), FN_DECIMAL(-0.662745098), FN_DECIMAL(-0.4352941176), FN_DECIMAL(-0.8666666667), FN_DECIMAL(-0.1529411765), +}; + +const FN_DECIMAL CELL_2D_X[] = +{ + FN_DECIMAL(-0.6440658039), FN_DECIMAL(-0.08028078721), FN_DECIMAL(0.9983546168), FN_DECIMAL(0.9869492062), FN_DECIMAL(0.9284746418), FN_DECIMAL(0.6051097552), FN_DECIMAL(-0.794167404), FN_DECIMAL(-0.3488667991), FN_DECIMAL(-0.943136526), FN_DECIMAL(-0.9968171318), FN_DECIMAL(0.8740961579), FN_DECIMAL(0.1421139764), FN_DECIMAL(0.4282553608), FN_DECIMAL(-0.9986665833), FN_DECIMAL(0.9996760121), FN_DECIMAL(-0.06248383632), + FN_DECIMAL(0.7120139305), FN_DECIMAL(0.8917660409), FN_DECIMAL(0.1094842955), FN_DECIMAL(-0.8730880804), FN_DECIMAL(0.2594811489), FN_DECIMAL(-0.6690063346), FN_DECIMAL(-0.9996834972), FN_DECIMAL(-0.8803608671), FN_DECIMAL(-0.8166554937), FN_DECIMAL(0.8955599676), FN_DECIMAL(-0.9398321388), FN_DECIMAL(0.07615451399), FN_DECIMAL(-0.7147270565), FN_DECIMAL(0.8707354457), FN_DECIMAL(-0.9580008579), FN_DECIMAL(0.4905965632), + FN_DECIMAL(0.786775944), FN_DECIMAL(0.1079711577), FN_DECIMAL(0.2686638979), FN_DECIMAL(0.6113487322), FN_DECIMAL(-0.530770584), FN_DECIMAL(-0.7837268286), FN_DECIMAL(-0.8558691039), FN_DECIMAL(-0.5726093896), FN_DECIMAL(-0.9830740914), FN_DECIMAL(0.7087766359), FN_DECIMAL(0.6807027153), FN_DECIMAL(-0.08864708788), FN_DECIMAL(0.6704485923), FN_DECIMAL(-0.1350735482), FN_DECIMAL(-0.9381333003), FN_DECIMAL(0.9756655376), + FN_DECIMAL(0.4231433671), FN_DECIMAL(-0.4959787385), FN_DECIMAL(0.1005554325), FN_DECIMAL(-0.7645857281), FN_DECIMAL(-0.5859053796), FN_DECIMAL(-0.9751154306), FN_DECIMAL(-0.6972258572), FN_DECIMAL(0.7907012002), FN_DECIMAL(-0.9109899213), FN_DECIMAL(-0.9584307894), FN_DECIMAL(-0.8269529333), FN_DECIMAL(0.2608264719), FN_DECIMAL(-0.7773760119), FN_DECIMAL(0.7606456974), FN_DECIMAL(-0.8961083758), FN_DECIMAL(-0.9838134719), + FN_DECIMAL(0.7338893576), FN_DECIMAL(0.2161226729), FN_DECIMAL(0.673509891), FN_DECIMAL(-0.5512056873), FN_DECIMAL(0.6899744332), FN_DECIMAL(0.868004831), FN_DECIMAL(0.5897430311), FN_DECIMAL(-0.8950444221), FN_DECIMAL(-0.3595752773), FN_DECIMAL(0.8209486981), FN_DECIMAL(-0.2912360132), FN_DECIMAL(-0.9965011374), FN_DECIMAL(0.9766994634), FN_DECIMAL(0.738790822), FN_DECIMAL(-0.4730947722), FN_DECIMAL(0.8946479441), + FN_DECIMAL(-0.6943628971), FN_DECIMAL(-0.6620468182), FN_DECIMAL(-0.0887255502), FN_DECIMAL(-0.7512250855), FN_DECIMAL(-0.5322986898), FN_DECIMAL(0.5226295385), FN_DECIMAL(0.2296318375), FN_DECIMAL(0.7915307344), FN_DECIMAL(-0.2756485999), FN_DECIMAL(-0.6900234522), FN_DECIMAL(0.07090588086), FN_DECIMAL(0.5981278485), FN_DECIMAL(0.3033429312), FN_DECIMAL(-0.7253142797), FN_DECIMAL(-0.9855874307), FN_DECIMAL(-0.1761843396), + FN_DECIMAL(-0.6438468325), FN_DECIMAL(-0.9956136595), FN_DECIMAL(0.8541580762), FN_DECIMAL(-0.9999807666), FN_DECIMAL(-0.02152416253), FN_DECIMAL(-0.8705983095), FN_DECIMAL(-0.1197138014), FN_DECIMAL(-0.992107781), FN_DECIMAL(-0.9091181546), FN_DECIMAL(0.788610536), FN_DECIMAL(-0.994636402), FN_DECIMAL(0.4211256853), FN_DECIMAL(0.3110430857), FN_DECIMAL(-0.4031127839), FN_DECIMAL(0.7610684239), FN_DECIMAL(0.7685674467), + FN_DECIMAL(0.152271555), FN_DECIMAL(-0.9364648723), FN_DECIMAL(0.1681333739), FN_DECIMAL(-0.3567427907), FN_DECIMAL(-0.418445483), FN_DECIMAL(-0.98774778), FN_DECIMAL(0.8705250765), FN_DECIMAL(-0.8911701067), FN_DECIMAL(-0.7315350966), FN_DECIMAL(0.6030885658), FN_DECIMAL(-0.4149130821), FN_DECIMAL(0.7585339481), FN_DECIMAL(0.6963196535), FN_DECIMAL(0.8332685012), FN_DECIMAL(-0.8086815232), FN_DECIMAL(0.7518116724), + FN_DECIMAL(-0.3490535894), FN_DECIMAL(0.6972110903), FN_DECIMAL(-0.8795676928), FN_DECIMAL(-0.6442331882), FN_DECIMAL(0.6610236811), FN_DECIMAL(-0.9853565782), FN_DECIMAL(-0.590338458), FN_DECIMAL(0.09843602117), FN_DECIMAL(0.5646534882), FN_DECIMAL(-0.6023259233), FN_DECIMAL(-0.3539248861), FN_DECIMAL(0.5132728656), FN_DECIMAL(0.9380385118), FN_DECIMAL(-0.7599270056), FN_DECIMAL(-0.7425936564), FN_DECIMAL(-0.6679610562), + FN_DECIMAL(-0.3018497816), FN_DECIMAL(0.814478266), FN_DECIMAL(0.03777430269), FN_DECIMAL(-0.7514235086), FN_DECIMAL(0.9662556939), FN_DECIMAL(-0.4720194901), FN_DECIMAL(-0.435054126), FN_DECIMAL(0.7091901235), FN_DECIMAL(0.929379209), FN_DECIMAL(0.9997434357), FN_DECIMAL(0.8306320299), FN_DECIMAL(-0.9434019629), FN_DECIMAL(-0.133133759), FN_DECIMAL(0.5048413216), FN_DECIMAL(0.3711995273), FN_DECIMAL(0.98552091), + FN_DECIMAL(0.7401857005), FN_DECIMAL(-0.9999981398), FN_DECIMAL(-0.2144033253), FN_DECIMAL(0.4808624681), FN_DECIMAL(-0.413835885), FN_DECIMAL(0.644229305), FN_DECIMAL(0.9626648696), FN_DECIMAL(0.1833665934), FN_DECIMAL(0.5794129), FN_DECIMAL(0.01404446873), FN_DECIMAL(0.4388494993), FN_DECIMAL(0.5213612322), FN_DECIMAL(-0.5281609948), FN_DECIMAL(-0.9745306846), FN_DECIMAL(-0.9904373013), FN_DECIMAL(0.9100232252), + FN_DECIMAL(-0.9914057719), FN_DECIMAL(0.7892627765), FN_DECIMAL(0.3364421659), FN_DECIMAL(-0.9416099764), FN_DECIMAL(0.7802732656), FN_DECIMAL(0.886302871), FN_DECIMAL(0.6524471291), FN_DECIMAL(0.5762186726), FN_DECIMAL(-0.08987644664), FN_DECIMAL(-0.2177026782), FN_DECIMAL(-0.9720345052), FN_DECIMAL(-0.05722538858), FN_DECIMAL(0.8105983127), FN_DECIMAL(0.3410261032), FN_DECIMAL(0.6452309645), FN_DECIMAL(-0.7810612152), + FN_DECIMAL(0.9989395718), FN_DECIMAL(-0.808247815), FN_DECIMAL(0.6370177929), FN_DECIMAL(0.5844658772), FN_DECIMAL(0.2054070861), FN_DECIMAL(0.055960522), FN_DECIMAL(-0.995827561), FN_DECIMAL(0.893409165), FN_DECIMAL(-0.931516824), FN_DECIMAL(0.328969469), FN_DECIMAL(-0.3193837488), FN_DECIMAL(0.7314755657), FN_DECIMAL(-0.7913517714), FN_DECIMAL(-0.2204109786), FN_DECIMAL(0.9955900414), FN_DECIMAL(-0.7112353139), + FN_DECIMAL(-0.7935008741), FN_DECIMAL(-0.9961918204), FN_DECIMAL(-0.9714163995), FN_DECIMAL(-0.9566188669), FN_DECIMAL(0.2748495632), FN_DECIMAL(-0.4681743221), FN_DECIMAL(-0.9614449642), FN_DECIMAL(0.585194072), FN_DECIMAL(0.4532946061), FN_DECIMAL(-0.9916113176), FN_DECIMAL(0.942479587), FN_DECIMAL(-0.9813704753), FN_DECIMAL(-0.6538429571), FN_DECIMAL(0.2923335053), FN_DECIMAL(-0.2246660704), FN_DECIMAL(-0.1800781949), + FN_DECIMAL(-0.9581216256), FN_DECIMAL(0.552215082), FN_DECIMAL(-0.9296791922), FN_DECIMAL(0.643183699), FN_DECIMAL(0.9997325981), FN_DECIMAL(-0.4606920354), FN_DECIMAL(-0.2148721265), FN_DECIMAL(0.3482070809), FN_DECIMAL(0.3075517813), FN_DECIMAL(0.6274756393), FN_DECIMAL(0.8910881765), FN_DECIMAL(-0.6397771309), FN_DECIMAL(-0.4479080125), FN_DECIMAL(-0.5247665011), FN_DECIMAL(-0.8386507094), FN_DECIMAL(0.3901291416), + FN_DECIMAL(0.1458336921), FN_DECIMAL(0.01624613149), FN_DECIMAL(-0.8273199879), FN_DECIMAL(0.5611100679), FN_DECIMAL(-0.8380219841), FN_DECIMAL(-0.9856122234), FN_DECIMAL(-0.861398618), FN_DECIMAL(0.6398413916), FN_DECIMAL(0.2694510795), FN_DECIMAL(0.4327334514), FN_DECIMAL(-0.9960265354), FN_DECIMAL(-0.939570655), FN_DECIMAL(-0.8846996446), FN_DECIMAL(0.7642113189), FN_DECIMAL(-0.7002080528), FN_DECIMAL(0.664508256), +}; +const FN_DECIMAL CELL_2D_Y[] = +{ + FN_DECIMAL(0.7649700911), FN_DECIMAL(0.9967722885), FN_DECIMAL(0.05734160033), FN_DECIMAL(-0.1610318741), FN_DECIMAL(0.371395799), FN_DECIMAL(-0.7961420628), FN_DECIMAL(0.6076990492), FN_DECIMAL(-0.9371723195), FN_DECIMAL(0.3324056156), FN_DECIMAL(0.07972205329), FN_DECIMAL(-0.4857529277), FN_DECIMAL(-0.9898503007), FN_DECIMAL(0.9036577593), FN_DECIMAL(0.05162417479), FN_DECIMAL(-0.02545330525), FN_DECIMAL(-0.998045976), + FN_DECIMAL(-0.7021653386), FN_DECIMAL(-0.4524967717), FN_DECIMAL(-0.9939885256), FN_DECIMAL(-0.4875625128), FN_DECIMAL(-0.9657481729), FN_DECIMAL(-0.7432567015), FN_DECIMAL(0.02515761212), FN_DECIMAL(0.4743044842), FN_DECIMAL(0.5771254669), FN_DECIMAL(0.4449408324), FN_DECIMAL(0.3416365773), FN_DECIMAL(0.9970960285), FN_DECIMAL(0.6994034849), FN_DECIMAL(0.4917517499), FN_DECIMAL(0.286765333), FN_DECIMAL(0.8713868327), + FN_DECIMAL(0.6172387009), FN_DECIMAL(0.9941540269), FN_DECIMAL(0.9632339851), FN_DECIMAL(-0.7913613129), FN_DECIMAL(0.847515538), FN_DECIMAL(0.6211056739), FN_DECIMAL(0.5171924952), FN_DECIMAL(-0.8198283277), FN_DECIMAL(-0.1832084353), FN_DECIMAL(0.7054329737), FN_DECIMAL(0.7325597678), FN_DECIMAL(0.9960630973), FN_DECIMAL(0.7419559859), FN_DECIMAL(0.9908355749), FN_DECIMAL(-0.346274329), FN_DECIMAL(0.2192641299), + FN_DECIMAL(-0.9060627411), FN_DECIMAL(-0.8683346653), FN_DECIMAL(0.9949314574), FN_DECIMAL(-0.6445220433), FN_DECIMAL(-0.8103794704), FN_DECIMAL(-0.2216977607), FN_DECIMAL(0.7168515217), FN_DECIMAL(0.612202264), FN_DECIMAL(-0.412428616), FN_DECIMAL(0.285325116), FN_DECIMAL(0.56227115), FN_DECIMAL(-0.9653857009), FN_DECIMAL(-0.6290361962), FN_DECIMAL(0.6491672535), FN_DECIMAL(0.443835306), FN_DECIMAL(-0.1791955706), + FN_DECIMAL(-0.6792690269), FN_DECIMAL(-0.9763662173), FN_DECIMAL(0.7391782104), FN_DECIMAL(0.8343693968), FN_DECIMAL(0.7238337389), FN_DECIMAL(0.4965557504), FN_DECIMAL(0.8075909592), FN_DECIMAL(-0.4459769977), FN_DECIMAL(-0.9331160806), FN_DECIMAL(-0.5710019572), FN_DECIMAL(0.9566512346), FN_DECIMAL(-0.08357920318), FN_DECIMAL(0.2146116448), FN_DECIMAL(-0.6739348049), FN_DECIMAL(0.8810115417), FN_DECIMAL(0.4467718167), + FN_DECIMAL(-0.7196250184), FN_DECIMAL(-0.749462481), FN_DECIMAL(0.9960561112), FN_DECIMAL(0.6600461127), FN_DECIMAL(-0.8465566164), FN_DECIMAL(-0.8525598897), FN_DECIMAL(-0.9732775654), FN_DECIMAL(0.6111293616), FN_DECIMAL(-0.9612584717), FN_DECIMAL(-0.7237870097), FN_DECIMAL(-0.9974830104), FN_DECIMAL(-0.8014006968), FN_DECIMAL(0.9528814544), FN_DECIMAL(-0.6884178931), FN_DECIMAL(-0.1691668301), FN_DECIMAL(0.9843571905), + FN_DECIMAL(0.7651544003), FN_DECIMAL(-0.09355982605), FN_DECIMAL(-0.5200134429), FN_DECIMAL(-0.006202125807), FN_DECIMAL(-0.9997683284), FN_DECIMAL(0.4919944954), FN_DECIMAL(-0.9928084436), FN_DECIMAL(-0.1253880012), FN_DECIMAL(-0.4165383308), FN_DECIMAL(-0.6148930171), FN_DECIMAL(-0.1034332049), FN_DECIMAL(-0.9070022917), FN_DECIMAL(-0.9503958117), FN_DECIMAL(0.9151503065), FN_DECIMAL(-0.6486716073), FN_DECIMAL(0.6397687707), + FN_DECIMAL(-0.9883386937), FN_DECIMAL(0.3507613761), FN_DECIMAL(0.9857642561), FN_DECIMAL(-0.9342026446), FN_DECIMAL(-0.9082419159), FN_DECIMAL(0.1560587169), FN_DECIMAL(0.4921240607), FN_DECIMAL(-0.453669308), FN_DECIMAL(0.6818037859), FN_DECIMAL(0.7976742329), FN_DECIMAL(0.9098610522), FN_DECIMAL(0.651633524), FN_DECIMAL(0.7177318024), FN_DECIMAL(-0.5528685241), FN_DECIMAL(0.5882467118), FN_DECIMAL(0.6593778956), + FN_DECIMAL(0.9371027648), FN_DECIMAL(-0.7168658839), FN_DECIMAL(-0.4757737632), FN_DECIMAL(0.7648291307), FN_DECIMAL(0.7503650398), FN_DECIMAL(0.1705063456), FN_DECIMAL(-0.8071558121), FN_DECIMAL(-0.9951433815), FN_DECIMAL(-0.8253280792), FN_DECIMAL(-0.7982502628), FN_DECIMAL(0.9352738503), FN_DECIMAL(0.8582254747), FN_DECIMAL(-0.3465310238), FN_DECIMAL(0.65000842), FN_DECIMAL(-0.6697422351), FN_DECIMAL(0.7441962291), + FN_DECIMAL(-0.9533555), FN_DECIMAL(0.5801940659), FN_DECIMAL(-0.9992862963), FN_DECIMAL(-0.659820211), FN_DECIMAL(0.2575848092), FN_DECIMAL(0.881588113), FN_DECIMAL(-0.9004043022), FN_DECIMAL(-0.7050172826), FN_DECIMAL(0.369126382), FN_DECIMAL(-0.02265088836), FN_DECIMAL(0.5568217228), FN_DECIMAL(-0.3316515286), FN_DECIMAL(0.991098079), FN_DECIMAL(-0.863212164), FN_DECIMAL(-0.9285531277), FN_DECIMAL(0.1695539323), + FN_DECIMAL(-0.672402505), FN_DECIMAL(-0.001928841934), FN_DECIMAL(0.9767452145), FN_DECIMAL(-0.8767960349), FN_DECIMAL(0.9103515037), FN_DECIMAL(-0.7648324016), FN_DECIMAL(0.2706960452), FN_DECIMAL(-0.9830446035), FN_DECIMAL(0.8150341657), FN_DECIMAL(-0.9999013716), FN_DECIMAL(-0.8985605806), FN_DECIMAL(0.8533360801), FN_DECIMAL(0.8491442537), FN_DECIMAL(-0.2242541966), FN_DECIMAL(-0.1379635899), FN_DECIMAL(-0.4145572694), + FN_DECIMAL(0.1308227633), FN_DECIMAL(0.6140555916), FN_DECIMAL(0.9417041303), FN_DECIMAL(-0.336705587), FN_DECIMAL(-0.6254387508), FN_DECIMAL(0.4631060578), FN_DECIMAL(-0.7578342456), FN_DECIMAL(-0.8172955655), FN_DECIMAL(-0.9959529228), FN_DECIMAL(-0.9760151351), FN_DECIMAL(0.2348380732), FN_DECIMAL(-0.9983612848), FN_DECIMAL(0.5856025746), FN_DECIMAL(-0.9400538266), FN_DECIMAL(-0.7639875669), FN_DECIMAL(0.6244544645), + FN_DECIMAL(0.04604054566), FN_DECIMAL(0.5888424828), FN_DECIMAL(0.7708490978), FN_DECIMAL(-0.8114182882), FN_DECIMAL(0.9786766212), FN_DECIMAL(-0.9984329822), FN_DECIMAL(0.09125496582), FN_DECIMAL(-0.4492438803), FN_DECIMAL(-0.3636982357), FN_DECIMAL(0.9443405575), FN_DECIMAL(-0.9476254645), FN_DECIMAL(-0.6818676535), FN_DECIMAL(-0.6113610831), FN_DECIMAL(0.9754070948), FN_DECIMAL(-0.0938108173), FN_DECIMAL(-0.7029540015), + FN_DECIMAL(-0.6085691109), FN_DECIMAL(-0.08718862881), FN_DECIMAL(-0.237381926), FN_DECIMAL(0.2913423132), FN_DECIMAL(0.9614872426), FN_DECIMAL(0.8836361266), FN_DECIMAL(-0.2749974196), FN_DECIMAL(-0.8108932717), FN_DECIMAL(-0.8913607575), FN_DECIMAL(0.129255541), FN_DECIMAL(-0.3342637104), FN_DECIMAL(-0.1921249337), FN_DECIMAL(-0.7566302845), FN_DECIMAL(-0.9563164339), FN_DECIMAL(-0.9744358146), FN_DECIMAL(0.9836522982), + FN_DECIMAL(-0.2863615732), FN_DECIMAL(0.8337016872), FN_DECIMAL(0.3683701937), FN_DECIMAL(0.7657119102), FN_DECIMAL(-0.02312427772), FN_DECIMAL(0.8875600535), FN_DECIMAL(0.976642191), FN_DECIMAL(0.9374176384), FN_DECIMAL(0.9515313457), FN_DECIMAL(-0.7786361937), FN_DECIMAL(-0.4538302125), FN_DECIMAL(-0.7685604874), FN_DECIMAL(-0.8940796454), FN_DECIMAL(-0.8512462154), FN_DECIMAL(0.5446696133), FN_DECIMAL(0.9207601495), + FN_DECIMAL(-0.9893091197), FN_DECIMAL(-0.9998680229), FN_DECIMAL(0.5617309299), FN_DECIMAL(-0.8277411985), FN_DECIMAL(0.545636467), FN_DECIMAL(0.1690223212), FN_DECIMAL(-0.5079295433), FN_DECIMAL(0.7685069899), FN_DECIMAL(-0.9630140787), FN_DECIMAL(0.9015219132), FN_DECIMAL(0.08905695279), FN_DECIMAL(-0.3423550559), FN_DECIMAL(-0.4661614943), FN_DECIMAL(-0.6449659371), FN_DECIMAL(0.7139388509), FN_DECIMAL(0.7472809229), +}; +const FN_DECIMAL CELL_3D_X[] = +{ + FN_DECIMAL(0.3752498686), FN_DECIMAL(0.687188096), FN_DECIMAL(0.2248135212), FN_DECIMAL(0.6692006647), FN_DECIMAL(-0.4376476931), FN_DECIMAL(0.6139972552), FN_DECIMAL(0.9494563929), FN_DECIMAL(0.8065108882), FN_DECIMAL(-0.2218812853), FN_DECIMAL(0.8484661167), FN_DECIMAL(0.5551817596), FN_DECIMAL(0.2133903499), FN_DECIMAL(0.5195126593), FN_DECIMAL(-0.6440141975), FN_DECIMAL(-0.5192897331), FN_DECIMAL(-0.3697654077), + FN_DECIMAL(-0.07927779647), FN_DECIMAL(0.4187757321), FN_DECIMAL(-0.750078731), FN_DECIMAL(0.6579554632), FN_DECIMAL(-0.6859803838), FN_DECIMAL(-0.6878407087), FN_DECIMAL(0.9490848347), FN_DECIMAL(0.5795829433), FN_DECIMAL(-0.5325976529), FN_DECIMAL(-0.1363699466), FN_DECIMAL(0.417665879), FN_DECIMAL(-0.9108236468), FN_DECIMAL(0.4438605427), FN_DECIMAL(0.819294887), FN_DECIMAL(-0.4033873915), FN_DECIMAL(-0.2817317705), + FN_DECIMAL(0.3969665622), FN_DECIMAL(0.5323450134), FN_DECIMAL(-0.6833017297), FN_DECIMAL(0.3881436661), FN_DECIMAL(-0.7119144767), FN_DECIMAL(-0.2306979838), FN_DECIMAL(-0.9398873022), FN_DECIMAL(0.1701906676), FN_DECIMAL(-0.4261839496), FN_DECIMAL(-0.003712295499), FN_DECIMAL(-0.734675004), FN_DECIMAL(-0.3195046015), FN_DECIMAL(0.7345307424), FN_DECIMAL(0.9766246496), FN_DECIMAL(-0.02003735175), FN_DECIMAL(-0.4824156342), + FN_DECIMAL(0.4245892007), FN_DECIMAL(0.9072427669), FN_DECIMAL(0.593346808), FN_DECIMAL(-0.8911762541), FN_DECIMAL(-0.7657571834), FN_DECIMAL(-0.5268198896), FN_DECIMAL(-0.8801903279), FN_DECIMAL(-0.6296409617), FN_DECIMAL(-0.09492481344), FN_DECIMAL(-0.4920470525), FN_DECIMAL(0.7307666154), FN_DECIMAL(-0.2514540636), FN_DECIMAL(-0.3356210347), FN_DECIMAL(-0.3522787894), FN_DECIMAL(0.87847885), FN_DECIMAL(-0.7424096346), + FN_DECIMAL(0.5757585274), FN_DECIMAL(0.4519299338), FN_DECIMAL(0.6420368628), FN_DECIMAL(-0.1128478447), FN_DECIMAL(0.499874883), FN_DECIMAL(0.5291681739), FN_DECIMAL(-0.5098837195), FN_DECIMAL(0.5639583502), FN_DECIMAL(-0.8456386526), FN_DECIMAL(-0.9657134875), FN_DECIMAL(-0.576437342), FN_DECIMAL(-0.5666013014), FN_DECIMAL(0.5667702405), FN_DECIMAL(-0.481316582), FN_DECIMAL(0.7313389916), FN_DECIMAL(-0.3805628566), + FN_DECIMAL(-0.6512675909), FN_DECIMAL(-0.2787156951), FN_DECIMAL(0.8648059114), FN_DECIMAL(-0.9730216276), FN_DECIMAL(-0.8335820906), FN_DECIMAL(0.2673159641), FN_DECIMAL(0.231150148), FN_DECIMAL(0.01286214638), FN_DECIMAL(0.6774953261), FN_DECIMAL(0.6542885718), FN_DECIMAL(-0.02545450161), FN_DECIMAL(0.2101238586), FN_DECIMAL(-0.5572105885), FN_DECIMAL(0.813705672), FN_DECIMAL(-0.7546026951), FN_DECIMAL(-0.2502500006), + FN_DECIMAL(-0.9979289381), FN_DECIMAL(0.7024037039), FN_DECIMAL(0.08990874624), FN_DECIMAL(0.8170812432), FN_DECIMAL(0.4226980265), FN_DECIMAL(-0.2442153475), FN_DECIMAL(-0.9183326731), FN_DECIMAL(0.6068222411), FN_DECIMAL(0.818676691), FN_DECIMAL(-0.7236735282), FN_DECIMAL(-0.5383903295), FN_DECIMAL(-0.6269337242), FN_DECIMAL(-0.0939331121), FN_DECIMAL(0.9203878539), FN_DECIMAL(-0.7256396824), FN_DECIMAL(0.6292431149), + FN_DECIMAL(0.4234156978), FN_DECIMAL(0.006685688024), FN_DECIMAL(-0.2598694113), FN_DECIMAL(0.6408036421), FN_DECIMAL(0.05899871622), FN_DECIMAL(0.7090281418), FN_DECIMAL(-0.5905222072), FN_DECIMAL(0.3128214264), FN_DECIMAL(-0.691925826), FN_DECIMAL(0.3634019349), FN_DECIMAL(-0.6772511147), FN_DECIMAL(-0.3204583896), FN_DECIMAL(-0.3906740409), FN_DECIMAL(-0.3342190395), FN_DECIMAL(-0.517779592), FN_DECIMAL(-0.6817711267), + FN_DECIMAL(0.6422383105), FN_DECIMAL(0.4388482478), FN_DECIMAL(0.2968562611), FN_DECIMAL(-0.2019778353), FN_DECIMAL(0.6014865048), FN_DECIMAL(0.9519280722), FN_DECIMAL(0.3398889569), FN_DECIMAL(0.8179709354), FN_DECIMAL(0.2365522154), FN_DECIMAL(0.3262175096), FN_DECIMAL(-0.8060715954), FN_DECIMAL(-0.2068642503), FN_DECIMAL(0.6208057279), FN_DECIMAL(-0.5274282502), FN_DECIMAL(-0.3722334928), FN_DECIMAL(-0.8923412971), + FN_DECIMAL(0.5341834201), FN_DECIMAL(-0.3663701513), FN_DECIMAL(-0.6114600319), FN_DECIMAL(0.5026307556), FN_DECIMAL(0.8396151729), FN_DECIMAL(0.9245042467), FN_DECIMAL(-0.7994843957), FN_DECIMAL(-0.5357200589), FN_DECIMAL(-0.6283359739), FN_DECIMAL(-0.61351886), FN_DECIMAL(-0.875632008), FN_DECIMAL(-0.5278879423), FN_DECIMAL(0.9087491985), FN_DECIMAL(-0.03500215466), FN_DECIMAL(-0.261365798), FN_DECIMAL(-0.579523541), + FN_DECIMAL(-0.3765052689), FN_DECIMAL(-0.74398252), FN_DECIMAL(0.4257318052), FN_DECIMAL(-0.1214508921), FN_DECIMAL(0.8561809753), FN_DECIMAL(0.6802835104), FN_DECIMAL(-0.5452131039), FN_DECIMAL(-0.1997156478), FN_DECIMAL(0.4562348357), FN_DECIMAL(-0.811704301), FN_DECIMAL(0.67793962), FN_DECIMAL(-0.9237819106), FN_DECIMAL(0.6973511259), FN_DECIMAL(-0.5189506), FN_DECIMAL(0.5517320032), FN_DECIMAL(-0.396710831), + FN_DECIMAL(0.5493762815), FN_DECIMAL(-0.2507853002), FN_DECIMAL(0.4788634005), FN_DECIMAL(0.387333516), FN_DECIMAL(-0.2176515694), FN_DECIMAL(0.6749832419), FN_DECIMAL(0.2148283022), FN_DECIMAL(-0.7521815872), FN_DECIMAL(0.4697000159), FN_DECIMAL(0.7890593699), FN_DECIMAL(-0.7606162952), FN_DECIMAL(0.01083397843), FN_DECIMAL(0.5254091908), FN_DECIMAL(-0.6748025877), FN_DECIMAL(0.751091524), FN_DECIMAL(0.05259056135), + FN_DECIMAL(0.01889481232), FN_DECIMAL(-0.6037423727), FN_DECIMAL(-0.6542965129), FN_DECIMAL(0.08873301081), FN_DECIMAL(-0.6191345671), FN_DECIMAL(0.4331858488), FN_DECIMAL(-0.3858351946), FN_DECIMAL(-0.1429059747), FN_DECIMAL(0.4118221036), FN_DECIMAL(-0.6247153214), FN_DECIMAL(-0.611423014), FN_DECIMAL(0.5542939606), FN_DECIMAL(-0.9432768808), FN_DECIMAL(-0.4567870451), FN_DECIMAL(-0.7349133547), FN_DECIMAL(0.399304489), + FN_DECIMAL(-0.7474927672), FN_DECIMAL(0.02589419753), FN_DECIMAL(0.783915821), FN_DECIMAL(0.6138668752), FN_DECIMAL(0.4276376047), FN_DECIMAL(-0.4347886353), FN_DECIMAL(0.02947841302), FN_DECIMAL(-0.833742746), FN_DECIMAL(0.3817221742), FN_DECIMAL(-0.8743368359), FN_DECIMAL(-0.3823443796), FN_DECIMAL(-0.6829243811), FN_DECIMAL(-0.3681903049), FN_DECIMAL(-0.367626833), FN_DECIMAL(-0.434583373), FN_DECIMAL(0.235891995), + FN_DECIMAL(-0.6874880269), FN_DECIMAL(-0.5115661773), FN_DECIMAL(-0.5534962601), FN_DECIMAL(0.5632777056), FN_DECIMAL(0.686191532), FN_DECIMAL(-0.05095871588), FN_DECIMAL(-0.06865785057), FN_DECIMAL(-0.5975288531), FN_DECIMAL(-0.6429790056), FN_DECIMAL(-0.3729361548), FN_DECIMAL(0.2237917666), FN_DECIMAL(0.6046773225), FN_DECIMAL(-0.5041542295), FN_DECIMAL(-0.03972191174), FN_DECIMAL(0.7028828406), FN_DECIMAL(-0.5560856498), + FN_DECIMAL(0.5898328456), FN_DECIMAL(-0.9308076766), FN_DECIMAL(0.4617069864), FN_DECIMAL(0.3190983137), FN_DECIMAL(0.9116567753), FN_DECIMAL(-0.45029554), FN_DECIMAL(0.3346334459), FN_DECIMAL(0.8525005645), FN_DECIMAL(0.2528483381), FN_DECIMAL(-0.8306630147), FN_DECIMAL(-0.6880390622), FN_DECIMAL(0.7448684026), FN_DECIMAL(-0.1963355843), FN_DECIMAL(-0.5900257974), FN_DECIMAL(0.9097057294), FN_DECIMAL(-0.2509196808), +}; +const FN_DECIMAL CELL_3D_Y[] = +{ + FN_DECIMAL(-0.6760585049), FN_DECIMAL(-0.09136176499), FN_DECIMAL(0.1681325679), FN_DECIMAL(-0.6688468686), FN_DECIMAL(-0.4822753902), FN_DECIMAL(-0.7891068824), FN_DECIMAL(-0.1877509944), FN_DECIMAL(0.548470914), FN_DECIMAL(-0.463339443), FN_DECIMAL(-0.4050542082), FN_DECIMAL(0.3218158513), FN_DECIMAL(0.2546493823), FN_DECIMAL(-0.3753271935), FN_DECIMAL(0.4745384887), FN_DECIMAL(0.481254652), FN_DECIMAL(-0.8934416489), + FN_DECIMAL(-0.6737085076), FN_DECIMAL(0.7469917228), FN_DECIMAL(0.3826230411), FN_DECIMAL(0.6751013678), FN_DECIMAL(-0.7248119515), FN_DECIMAL(-0.3224276742), FN_DECIMAL(-0.02076190936), FN_DECIMAL(-0.6404268166), FN_DECIMAL(-0.5292028444), FN_DECIMAL(0.7151414636), FN_DECIMAL(-0.6144655059), FN_DECIMAL(-0.369912124), FN_DECIMAL(0.6942067212), FN_DECIMAL(-0.4481558248), FN_DECIMAL(-0.6366894559), FN_DECIMAL(0.5956568471), + FN_DECIMAL(0.564274539), FN_DECIMAL(0.7145584688), FN_DECIMAL(0.6871918316), FN_DECIMAL(0.5657918509), FN_DECIMAL(-0.6275978114), FN_DECIMAL(0.4146983062), FN_DECIMAL(0.2638993789), FN_DECIMAL(-0.792633138), FN_DECIMAL(0.5706133514), FN_DECIMAL(0.8606546462), FN_DECIMAL(0.6490900316), FN_DECIMAL(-0.8242699196), FN_DECIMAL(0.6765819124), FN_DECIMAL(0.1959534069), FN_DECIMAL(-0.8426769757), FN_DECIMAL(-0.5917672797), + FN_DECIMAL(0.7517364266), FN_DECIMAL(0.03252559226), FN_DECIMAL(0.0883617105), FN_DECIMAL(0.4475064813), FN_DECIMAL(-0.1418643552), FN_DECIMAL(0.7343428473), FN_DECIMAL(0.3870192548), FN_DECIMAL(-0.7716703522), FN_DECIMAL(0.4839898327), FN_DECIMAL(0.7437439055), FN_DECIMAL(-0.5989573348), FN_DECIMAL(-0.8357068955), FN_DECIMAL(0.6086049038), FN_DECIMAL(0.9194627258), FN_DECIMAL(0.4718297238), FN_DECIMAL(-0.2650335884), + FN_DECIMAL(-0.6470352599), FN_DECIMAL(-0.5555181303), FN_DECIMAL(0.1222351235), FN_DECIMAL(0.7802044684), FN_DECIMAL(-0.8636947022), FN_DECIMAL(-0.2341352163), FN_DECIMAL(0.683030874), FN_DECIMAL(-0.5005858287), FN_DECIMAL(0.2334616211), FN_DECIMAL(0.2576877608), FN_DECIMAL(0.6666816727), FN_DECIMAL(-0.7663996863), FN_DECIMAL(0.794201982), FN_DECIMAL(0.6189308788), FN_DECIMAL(0.6071033261), FN_DECIMAL(-0.4206058253), + FN_DECIMAL(-0.3957336915), FN_DECIMAL(-0.8170257484), FN_DECIMAL(-0.1043240417), FN_DECIMAL(0.0002167596213), FN_DECIMAL(0.1816339018), FN_DECIMAL(-0.6838094939), FN_DECIMAL(-0.2495341969), FN_DECIMAL(-0.7116756954), FN_DECIMAL(-0.03361673621), FN_DECIMAL(-0.3350836431), FN_DECIMAL(0.2137186039), FN_DECIMAL(0.2557996786), FN_DECIMAL(0.7490117093), FN_DECIMAL(0.4942936549), FN_DECIMAL(-0.352686853), FN_DECIMAL(-0.3952445435), + FN_DECIMAL(-0.0459964767), FN_DECIMAL(-0.7115787471), FN_DECIMAL(0.08022899756), FN_DECIMAL(0.5362268157), FN_DECIMAL(-0.8258613686), FN_DECIMAL(0.1114171723), FN_DECIMAL(0.3882823051), FN_DECIMAL(-0.7915404457), FN_DECIMAL(0.3250957662), FN_DECIMAL(0.6401346464), FN_DECIMAL(-0.2662724517), FN_DECIMAL(-0.6727907114), FN_DECIMAL(-0.994730818), FN_DECIMAL(-0.3596358977), FN_DECIMAL(0.2344610069), FN_DECIMAL(-0.6645215546), + FN_DECIMAL(-0.7107590611), FN_DECIMAL(-0.4646617327), FN_DECIMAL(0.6717191355), FN_DECIMAL(0.5101893498), FN_DECIMAL(0.1185768238), FN_DECIMAL(0.236005093), FN_DECIMAL(-0.7811024061), FN_DECIMAL(0.5089325193), FN_DECIMAL(0.6073187658), FN_DECIMAL(-0.7930732557), FN_DECIMAL(-0.6822767155), FN_DECIMAL(0.3201532885), FN_DECIMAL(0.7545302807), FN_DECIMAL(0.1072664448), FN_DECIMAL(0.6784033173), FN_DECIMAL(-0.6595924967), + FN_DECIMAL(0.7276509498), FN_DECIMAL(0.5586689436), FN_DECIMAL(-0.6498636788), FN_DECIMAL(0.6789333174), FN_DECIMAL(0.7105966551), FN_DECIMAL(-0.2872214155), FN_DECIMAL(0.496746217), FN_DECIMAL(-0.3880337977), FN_DECIMAL(0.7324070604), FN_DECIMAL(-0.9326634749), FN_DECIMAL(-0.5867839255), FN_DECIMAL(0.8003043651), FN_DECIMAL(-0.1631882481), FN_DECIMAL(-0.6796374681), FN_DECIMAL(-0.8066678503), FN_DECIMAL(0.4238177418), + FN_DECIMAL(0.7715863549), FN_DECIMAL(0.5455367347), FN_DECIMAL(-0.03205115397), FN_DECIMAL(-0.6005545066), FN_DECIMAL(-0.5423640002), FN_DECIMAL(0.3569205906), FN_DECIMAL(-0.582071752), FN_DECIMAL(0.6407354361), FN_DECIMAL(0.7777142984), FN_DECIMAL(-0.09956428618), FN_DECIMAL(0.1100002681), FN_DECIMAL(0.8136349123), FN_DECIMAL(0.2923431904), FN_DECIMAL(0.9735794425), FN_DECIMAL(0.8324974864), FN_DECIMAL(-0.6179617717), + FN_DECIMAL(-0.9248386523), FN_DECIMAL(-0.6448780771), FN_DECIMAL(-0.5274402761), FN_DECIMAL(-0.7862170565), FN_DECIMAL(0.2682099744), FN_DECIMAL(-0.5848777694), FN_DECIMAL(-0.6364561467), FN_DECIMAL(-0.7167402514), FN_DECIMAL(-0.8677012494), FN_DECIMAL(0.4205286707), FN_DECIMAL(-0.7007832749), FN_DECIMAL(0.243272451), FN_DECIMAL(-0.1899846085), FN_DECIMAL(-0.6146124977), FN_DECIMAL(-0.8093357692), FN_DECIMAL(-0.03545096987), + FN_DECIMAL(-0.7191590868), FN_DECIMAL(0.7478645848), FN_DECIMAL(0.3623517328), FN_DECIMAL(0.8436992512), FN_DECIMAL(-0.2445711729), FN_DECIMAL(0.6897356637), FN_DECIMAL(-0.1708070787), FN_DECIMAL(0.4639272368), FN_DECIMAL(-0.7917186656), FN_DECIMAL(0.02980025428), FN_DECIMAL(0.6334156172), FN_DECIMAL(-0.9815544807), FN_DECIMAL(-0.2307217304), FN_DECIMAL(0.1080823318), FN_DECIMAL(0.5167601798), FN_DECIMAL(-0.845120016), + FN_DECIMAL(0.441572562), FN_DECIMAL(0.5876789172), FN_DECIMAL(-0.6365908737), FN_DECIMAL(0.68350166), FN_DECIMAL(0.5849723959), FN_DECIMAL(0.1164114357), FN_DECIMAL(-0.7379813884), FN_DECIMAL(-0.9613237178), FN_DECIMAL(-0.9071943084), FN_DECIMAL(-0.7682111105), FN_DECIMAL(0.639074459), FN_DECIMAL(-0.619358298), FN_DECIMAL(0.2807257131), FN_DECIMAL(-0.01800868791), FN_DECIMAL(0.3776607289), FN_DECIMAL(0.7207567823), + FN_DECIMAL(0.5536661486), FN_DECIMAL(-0.9974053117), FN_DECIMAL(-0.02047200006), FN_DECIMAL(-0.6739453804), FN_DECIMAL(-0.5607471297), FN_DECIMAL(0.8815553192), FN_DECIMAL(0.8275977415), FN_DECIMAL(0.3928902456), FN_DECIMAL(0.550991396), FN_DECIMAL(0.4247623676), FN_DECIMAL(-0.3436948871), FN_DECIMAL(-0.3653537677), FN_DECIMAL(0.3181702902), FN_DECIMAL(-0.6067173171), FN_DECIMAL(-0.8984128477), FN_DECIMAL(0.4220839766), + FN_DECIMAL(0.7238407199), FN_DECIMAL(-0.7766913695), FN_DECIMAL(0.6460037842), FN_DECIMAL(0.2544775664), FN_DECIMAL(0.6488840578), FN_DECIMAL(0.805016833), FN_DECIMAL(-0.9183807036), FN_DECIMAL(0.4144046357), FN_DECIMAL(0.270587208), FN_DECIMAL(-0.8813684494), FN_DECIMAL(0.6985971877), FN_DECIMAL(-0.7795603017), FN_DECIMAL(-0.8624480731), FN_DECIMAL(0.5532697017), FN_DECIMAL(0.711179521), FN_DECIMAL(-0.7798160574), + FN_DECIMAL(0.5225859041), FN_DECIMAL(0.1261859368), FN_DECIMAL(0.3398033582), FN_DECIMAL(-0.7472173667), FN_DECIMAL(-0.4032647119), FN_DECIMAL(-0.4246578154), FN_DECIMAL(0.8481212377), FN_DECIMAL(-0.2144838537), FN_DECIMAL(0.3431714491), FN_DECIMAL(0.5310188231), FN_DECIMAL(0.6682978632), FN_DECIMAL(0.3110433206), FN_DECIMAL(0.9263293599), FN_DECIMAL(-0.6155600569), FN_DECIMAL(0.07169784399), FN_DECIMAL(0.8985888773), +}; +const FN_DECIMAL CELL_3D_Z[] = +{ + FN_DECIMAL(-0.6341391283), FN_DECIMAL(-0.7207118346), FN_DECIMAL(0.9597866014), FN_DECIMAL(0.3237504235), FN_DECIMAL(-0.7588642466), FN_DECIMAL(-0.01782410481), FN_DECIMAL(0.2515593809), FN_DECIMAL(0.2207257205), FN_DECIMAL(-0.8579541106), FN_DECIMAL(0.3406410681), FN_DECIMAL(0.7669470462), FN_DECIMAL(-0.9431957648), FN_DECIMAL(0.7676171537), FN_DECIMAL(-0.6000491115), FN_DECIMAL(-0.7062096948), FN_DECIMAL(0.2550207115), + FN_DECIMAL(0.7347325213), FN_DECIMAL(0.5163625202), FN_DECIMAL(-0.5394270162), FN_DECIMAL(0.3336656285), FN_DECIMAL(-0.0638635111), FN_DECIMAL(-0.6503195787), FN_DECIMAL(0.3143356798), FN_DECIMAL(-0.5039217245), FN_DECIMAL(0.6605180464), FN_DECIMAL(-0.6855479011), FN_DECIMAL(-0.6693185756), FN_DECIMAL(0.1832083647), FN_DECIMAL(-0.5666258437), FN_DECIMAL(0.3576482138), FN_DECIMAL(-0.6571949095), FN_DECIMAL(-0.7522101635), + FN_DECIMAL(-0.7238865886), FN_DECIMAL(0.4538887323), FN_DECIMAL(0.2467106257), FN_DECIMAL(0.7274778869), FN_DECIMAL(0.3151170655), FN_DECIMAL(-0.8802293764), FN_DECIMAL(-0.2167232729), FN_DECIMAL(0.5854637865), FN_DECIMAL(0.7019741052), FN_DECIMAL(0.5091756071), FN_DECIMAL(0.1973189533), FN_DECIMAL(0.46743546), FN_DECIMAL(0.05197599597), FN_DECIMAL(0.088354718), FN_DECIMAL(0.5380464843), FN_DECIMAL(-0.6458224544), + FN_DECIMAL(-0.5045952393), FN_DECIMAL(0.419347884), FN_DECIMAL(0.8000823542), FN_DECIMAL(-0.07445020656), FN_DECIMAL(-0.6272881641), FN_DECIMAL(-0.428020311), FN_DECIMAL(-0.2747382083), FN_DECIMAL(-0.08987283726), FN_DECIMAL(0.8699098354), FN_DECIMAL(0.4524761885), FN_DECIMAL(-0.3274603257), FN_DECIMAL(0.4882262167), FN_DECIMAL(-0.7189983256), FN_DECIMAL(0.1746079907), FN_DECIMAL(0.0751772698), FN_DECIMAL(-0.6152927202), + FN_DECIMAL(0.4998474673), FN_DECIMAL(-0.6979677227), FN_DECIMAL(0.7568667263), FN_DECIMAL(-0.6152612058), FN_DECIMAL(0.06447140991), FN_DECIMAL(-0.8155744872), FN_DECIMAL(-0.5229602449), FN_DECIMAL(0.6567836838), FN_DECIMAL(-0.4799905631), FN_DECIMAL(0.03153534591), FN_DECIMAL(0.4724992466), FN_DECIMAL(-0.3026458097), FN_DECIMAL(-0.2191225827), FN_DECIMAL(-0.620692287), FN_DECIMAL(0.3107552588), FN_DECIMAL(0.8235670294), + FN_DECIMAL(0.6474915988), FN_DECIMAL(-0.5047637941), FN_DECIMAL(0.4911488878), FN_DECIMAL(-0.2307138167), FN_DECIMAL(-0.5216800015), FN_DECIMAL(0.6789305939), FN_DECIMAL(0.9403734863), FN_DECIMAL(0.702390397), FN_DECIMAL(0.7347584625), FN_DECIMAL(0.6779567958), FN_DECIMAL(0.9765635805), FN_DECIMAL(-0.9436177661), FN_DECIMAL(-0.358465925), FN_DECIMAL(-0.3058706624), FN_DECIMAL(0.5533414464), FN_DECIMAL(-0.8838306897), + FN_DECIMAL(0.04496841812), FN_DECIMAL(0.01687374963), FN_DECIMAL(-0.9927133148), FN_DECIMAL(-0.211752318), FN_DECIMAL(0.3732015249), FN_DECIMAL(0.9632990593), FN_DECIMAL(-0.07682417004), FN_DECIMAL(-0.07232213047), FN_DECIMAL(0.4733721775), FN_DECIMAL(0.2579229713), FN_DECIMAL(0.7995216286), FN_DECIMAL(0.3928189967), FN_DECIMAL(0.04107517667), FN_DECIMAL(0.1534542912), FN_DECIMAL(0.6468965045), FN_DECIMAL(0.4030684878), + FN_DECIMAL(-0.5617300988), FN_DECIMAL(-0.885463029), FN_DECIMAL(0.693729985), FN_DECIMAL(-0.5736527866), FN_DECIMAL(-0.9911905409), FN_DECIMAL(-0.66451538), FN_DECIMAL(0.2028855685), FN_DECIMAL(0.8019541421), FN_DECIMAL(-0.3903877149), FN_DECIMAL(-0.4888495114), FN_DECIMAL(-0.2753714057), FN_DECIMAL(-0.8915202143), FN_DECIMAL(0.5273119089), FN_DECIMAL(0.9363714773), FN_DECIMAL(-0.5212228249), FN_DECIMAL(-0.31642672), + FN_DECIMAL(0.2409440761), FN_DECIMAL(-0.703776404), FN_DECIMAL(-0.6996810411), FN_DECIMAL(-0.7058714505), FN_DECIMAL(-0.3650566783), FN_DECIMAL(0.1064744278), FN_DECIMAL(0.7985729102), FN_DECIMAL(0.424680257), FN_DECIMAL(-0.6384535592), FN_DECIMAL(0.1540161646), FN_DECIMAL(-0.07702731943), FN_DECIMAL(-0.5627789132), FN_DECIMAL(-0.7667919169), FN_DECIMAL(-0.509815999), FN_DECIMAL(0.4590525092), FN_DECIMAL(0.1552595611), + FN_DECIMAL(0.345402042), FN_DECIMAL(0.7537656024), FN_DECIMAL(0.7906259247), FN_DECIMAL(-0.6218493452), FN_DECIMAL(0.02979350071), FN_DECIMAL(-0.1337893489), FN_DECIMAL(-0.1483818606), FN_DECIMAL(0.549965562), FN_DECIMAL(0.01882482408), FN_DECIMAL(-0.7833783002), FN_DECIMAL(0.4702855809), FN_DECIMAL(0.2435827372), FN_DECIMAL(0.2978428332), FN_DECIMAL(0.2256499906), FN_DECIMAL(0.4885036897), FN_DECIMAL(0.5312962584), + FN_DECIMAL(0.05401156992), FN_DECIMAL(0.1749922158), FN_DECIMAL(-0.7352273018), FN_DECIMAL(0.6058980284), FN_DECIMAL(0.4416079111), FN_DECIMAL(0.4417378638), FN_DECIMAL(0.5455879807), FN_DECIMAL(-0.6681295324), FN_DECIMAL(0.1973431441), FN_DECIMAL(0.4053292055), FN_DECIMAL(0.2220375492), FN_DECIMAL(0.2957118467), FN_DECIMAL(0.6910913512), FN_DECIMAL(0.5940890106), FN_DECIMAL(-0.2014135283), FN_DECIMAL(-0.9172588213), + FN_DECIMAL(-0.4254361401), FN_DECIMAL(-0.6146586825), FN_DECIMAL(-0.7996193253), FN_DECIMAL(-0.3716777111), FN_DECIMAL(-0.9448876842), FN_DECIMAL(-0.2620349924), FN_DECIMAL(0.9615995749), FN_DECIMAL(-0.4679683524), FN_DECIMAL(0.3905937144), FN_DECIMAL(0.613593722), FN_DECIMAL(0.1422937358), FN_DECIMAL(0.1908754211), FN_DECIMAL(0.8189704912), FN_DECIMAL(-0.7300408736), FN_DECIMAL(-0.4108776451), FN_DECIMAL(-0.5319834504), + FN_DECIMAL(-0.8970265651), FN_DECIMAL(-0.5386359045), FN_DECIMAL(0.4082255906), FN_DECIMAL(0.7245356676), FN_DECIMAL(0.5239080873), FN_DECIMAL(-0.8937552226), FN_DECIMAL(-0.553637673), FN_DECIMAL(0.2354455182), FN_DECIMAL(-0.0860293075), FN_DECIMAL(-0.1399373318), FN_DECIMAL(-0.4666323327), FN_DECIMAL(0.5560157407), FN_DECIMAL(0.1772619533), FN_DECIMAL(-0.8893937725), FN_DECIMAL(-0.5632714576), FN_DECIMAL(-0.5666264959), + FN_DECIMAL(-0.3670263736), FN_DECIMAL(-0.06717242579), FN_DECIMAL(0.6205295181), FN_DECIMAL(-0.4110536264), FN_DECIMAL(0.7090054553), FN_DECIMAL(0.183899597), FN_DECIMAL(-0.5605470555), FN_DECIMAL(0.3879565548), FN_DECIMAL(0.7420893903), FN_DECIMAL(-0.2347595118), FN_DECIMAL(-0.8577217497), FN_DECIMAL(0.6325590203), FN_DECIMAL(-0.8736152276), FN_DECIMAL(0.7048011129), FN_DECIMAL(-0.06317948268), FN_DECIMAL(0.8753285574), + FN_DECIMAL(-0.05843650473), FN_DECIMAL(-0.3674922622), FN_DECIMAL(-0.5256624401), FN_DECIMAL(0.7861039337), FN_DECIMAL(0.3287714416), FN_DECIMAL(0.5910593099), FN_DECIMAL(-0.3896960134), FN_DECIMAL(0.6864605361), FN_DECIMAL(0.7164918431), FN_DECIMAL(-0.290014277), FN_DECIMAL(-0.6796169617), FN_DECIMAL(0.1632515592), FN_DECIMAL(0.04485347486), FN_DECIMAL(0.8320545697), FN_DECIMAL(0.01339408056), FN_DECIMAL(-0.2874989857), + FN_DECIMAL(0.615630723), FN_DECIMAL(0.3430367014), FN_DECIMAL(0.8193658136), FN_DECIMAL(-0.5829600957), FN_DECIMAL(0.07911697781), FN_DECIMAL(0.7854296063), FN_DECIMAL(-0.4107442306), FN_DECIMAL(0.4766964066), FN_DECIMAL(-0.9045999527), FN_DECIMAL(-0.1673856787), FN_DECIMAL(0.2828077348), FN_DECIMAL(-0.5902737632), FN_DECIMAL(-0.321506229), FN_DECIMAL(-0.5224513133), FN_DECIMAL(-0.4090169985), FN_DECIMAL(-0.3599685311), +}; + +static int FastFloor(FN_DECIMAL f) { return (f >= 0 ? (int)f : (int)f - 1); } +static int FastRound(FN_DECIMAL f) { return (f >= 0) ? (int)(f + FN_DECIMAL(0.5)) : (int)(f - FN_DECIMAL(0.5)); } +static int FastAbs(int i) { return abs(i); } +static FN_DECIMAL FastAbs(FN_DECIMAL f) { return fabs(f); } +static FN_DECIMAL Lerp(FN_DECIMAL a, FN_DECIMAL b, FN_DECIMAL t) { return a + t * (b - a); } +static FN_DECIMAL InterpHermiteFunc(FN_DECIMAL t) { return t*t*(3 - 2 * t); } +static FN_DECIMAL InterpQuinticFunc(FN_DECIMAL t) { return t*t*t*(t*(t * 6 - 15) + 10); } +static FN_DECIMAL CubicLerp(FN_DECIMAL a, FN_DECIMAL b, FN_DECIMAL c, FN_DECIMAL d, FN_DECIMAL t) +{ + FN_DECIMAL p = (d - c) - (a - b); + return t * t * t * p + t * t * ((a - b) - p) + t * (c - a) + b; +} + +void FastNoise::SetSeed(int seed) +{ + m_seed = seed; + + std::mt19937_64 gen(seed); + + for (int i = 0; i < 256; i++) + m_perm[i] = i; + + for (int j = 0; j < 256; j++) + { + int rng = (int)(gen() % (256 - j)); + int k = rng + j; + int l = m_perm[j]; + m_perm[j] = m_perm[j + 256] = m_perm[k]; + m_perm[k] = l; + m_perm12[j] = m_perm12[j + 256] = m_perm[j] % 12; + } +} + +void FastNoise::CalculateFractalBounding() +{ + FN_DECIMAL amp = m_gain; + FN_DECIMAL ampFractal = 1.0f; + for (int i = 1; i < m_octaves; i++) + { + ampFractal += amp; + amp *= m_gain; + } + m_fractalBounding = 1.0f / ampFractal; +} + +void FastNoise::SetCellularDistance2Indices(int cellularDistanceIndex0, int cellularDistanceIndex1) +{ + m_cellularDistanceIndex0 = std::min(cellularDistanceIndex0, cellularDistanceIndex1); + m_cellularDistanceIndex1 = std::max(cellularDistanceIndex0, cellularDistanceIndex1); + + m_cellularDistanceIndex0 = std::min(std::max(m_cellularDistanceIndex0, 0), FN_CELLULAR_INDEX_MAX); + m_cellularDistanceIndex1 = std::min(std::max(m_cellularDistanceIndex1, 0), FN_CELLULAR_INDEX_MAX); +} + +void FastNoise::GetCellularDistance2Indices(int& cellularDistanceIndex0, int& cellularDistanceIndex1) const +{ + cellularDistanceIndex0 = m_cellularDistanceIndex0; + cellularDistanceIndex1 = m_cellularDistanceIndex1; +} + +unsigned char FastNoise::Index2D_12(unsigned char offset, int x, int y) const +{ + return m_perm12[(x & 0xff) + m_perm[(y & 0xff) + offset]]; +} +unsigned char FastNoise::Index3D_12(unsigned char offset, int x, int y, int z) const +{ + return m_perm12[(x & 0xff) + m_perm[(y & 0xff) + m_perm[(z & 0xff) + offset]]]; +} +unsigned char FastNoise::Index4D_32(unsigned char offset, int x, int y, int z, int w) const +{ + return m_perm[(x & 0xff) + m_perm[(y & 0xff) + m_perm[(z & 0xff) + m_perm[(w & 0xff) + offset]]]] & 31; +} +unsigned char FastNoise::Index2D_256(unsigned char offset, int x, int y) const +{ + return m_perm[(x & 0xff) + m_perm[(y & 0xff) + offset]]; +} +unsigned char FastNoise::Index3D_256(unsigned char offset, int x, int y, int z) const +{ + return m_perm[(x & 0xff) + m_perm[(y & 0xff) + m_perm[(z & 0xff) + offset]]]; +} +unsigned char FastNoise::Index4D_256(unsigned char offset, int x, int y, int z, int w) const +{ + return m_perm[(x & 0xff) + m_perm[(y & 0xff) + m_perm[(z & 0xff) + m_perm[(w & 0xff) + offset]]]]; +} + +// Hashing +#define X_PRIME 1619 +#define Y_PRIME 31337 +#define Z_PRIME 6971 +#define W_PRIME 1013 + +static FN_DECIMAL ValCoord2D(int seed, int x, int y) +{ + int n = seed; + n ^= X_PRIME * x; + n ^= Y_PRIME * y; + + return (n * n * n * 60493) / FN_DECIMAL(2147483648); +} +static FN_DECIMAL ValCoord3D(int seed, int x, int y, int z) +{ + int n = seed; + n ^= X_PRIME * x; + n ^= Y_PRIME * y; + n ^= Z_PRIME * z; + + return (n * n * n * 60493) / FN_DECIMAL(2147483648); +} +static FN_DECIMAL ValCoord4D(int seed, int x, int y, int z, int w) +{ + int n = seed; + n ^= X_PRIME * x; + n ^= Y_PRIME * y; + n ^= Z_PRIME * z; + n ^= W_PRIME * w; + + return (n * n * n * 60493) / FN_DECIMAL(2147483648); +} + +FN_DECIMAL FastNoise::ValCoord2DFast(unsigned char offset, int x, int y) const +{ + return VAL_LUT[Index2D_256(offset, x, y)]; +} +FN_DECIMAL FastNoise::ValCoord3DFast(unsigned char offset, int x, int y, int z) const +{ + return VAL_LUT[Index3D_256(offset, x, y, z)]; +} + +FN_DECIMAL FastNoise::GradCoord2D(unsigned char offset, int x, int y, FN_DECIMAL xd, FN_DECIMAL yd) const +{ + unsigned char lutPos = Index2D_12(offset, x, y); + + return xd*GRAD_X[lutPos] + yd*GRAD_Y[lutPos]; +} +FN_DECIMAL FastNoise::GradCoord3D(unsigned char offset, int x, int y, int z, FN_DECIMAL xd, FN_DECIMAL yd, FN_DECIMAL zd) const +{ + unsigned char lutPos = Index3D_12(offset, x, y, z); + + return xd*GRAD_X[lutPos] + yd*GRAD_Y[lutPos] + zd*GRAD_Z[lutPos]; +} +FN_DECIMAL FastNoise::GradCoord4D(unsigned char offset, int x, int y, int z, int w, FN_DECIMAL xd, FN_DECIMAL yd, FN_DECIMAL zd, FN_DECIMAL wd) const +{ + unsigned char lutPos = Index4D_32(offset, x, y, z, w) << 2; + + return xd*GRAD_4D[lutPos] + yd*GRAD_4D[lutPos + 1] + zd*GRAD_4D[lutPos + 2] + wd*GRAD_4D[lutPos + 3]; +} + +FN_DECIMAL FastNoise::GetNoise(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + x *= m_frequency; + y *= m_frequency; + z *= m_frequency; + + switch (m_noiseType) + { + case Value: + return SingleValue(0, x, y, z); + case ValueFractal: + switch (m_fractalType) + { + case FBM: + return SingleValueFractalFBM(x, y, z); + case Billow: + return SingleValueFractalBillow(x, y, z); + case RigidMulti: + return SingleValueFractalRigidMulti(x, y, z); + default: + return 0; + } + case Perlin: + return SinglePerlin(0, x, y, z); + case PerlinFractal: + switch (m_fractalType) + { + case FBM: + return SinglePerlinFractalFBM(x, y, z); + case Billow: + return SinglePerlinFractalBillow(x, y, z); + case RigidMulti: + return SinglePerlinFractalRigidMulti(x, y, z); + default: + return 0; + } + case Simplex: + return SingleSimplex(0, x, y, z); + case SimplexFractal: + switch (m_fractalType) + { + case FBM: + return SingleSimplexFractalFBM(x, y, z); + case Billow: + return SingleSimplexFractalBillow(x, y, z); + case RigidMulti: + return SingleSimplexFractalRigidMulti(x, y, z); + default: + return 0; + } + case Cellular: + switch (m_cellularReturnType) + { + case CellValue: + case NoiseLookup: + case Distance: + return SingleCellular(x, y, z); + default: + return SingleCellular2Edge(x, y, z); + } + case WhiteNoise: + return GetWhiteNoise(x, y, z); + case Cubic: + return SingleCubic(0, x, y, z); + case CubicFractal: + switch (m_fractalType) + { + case FBM: + return SingleCubicFractalFBM(x, y, z); + case Billow: + return SingleCubicFractalBillow(x, y, z); + case RigidMulti: + return SingleCubicFractalRigidMulti(x, y, z); + } + default: + return 0; + } +} + +FN_DECIMAL FastNoise::GetNoise(FN_DECIMAL x, FN_DECIMAL y) const +{ + x *= m_frequency; + y *= m_frequency; + + switch (m_noiseType) + { + case Value: + return SingleValue(0, x, y); + case ValueFractal: + switch (m_fractalType) + { + case FBM: + return SingleValueFractalFBM(x, y); + case Billow: + return SingleValueFractalBillow(x, y); + case RigidMulti: + return SingleValueFractalRigidMulti(x, y); + } + case Perlin: + return SinglePerlin(0, x, y); + case PerlinFractal: + switch (m_fractalType) + { + case FBM: + return SinglePerlinFractalFBM(x, y); + case Billow: + return SinglePerlinFractalBillow(x, y); + case RigidMulti: + return SinglePerlinFractalRigidMulti(x, y); + } + case Simplex: + return SingleSimplex(0, x, y); + case SimplexFractal: + switch (m_fractalType) + { + case FBM: + return SingleSimplexFractalFBM(x, y); + case Billow: + return SingleSimplexFractalBillow(x, y); + case RigidMulti: + return SingleSimplexFractalRigidMulti(x, y); + } + case Cellular: + switch (m_cellularReturnType) + { + case CellValue: + case NoiseLookup: + case Distance: + return SingleCellular(x, y); + default: + return SingleCellular2Edge(x, y); + } + case WhiteNoise: + return GetWhiteNoise(x, y); + case Cubic: + return SingleCubic(0, x, y); + case CubicFractal: + switch (m_fractalType) + { + case FBM: + return SingleCubicFractalFBM(x, y); + case Billow: + return SingleCubicFractalBillow(x, y); + case RigidMulti: + return SingleCubicFractalRigidMulti(x, y); + } + } + return 0; +} + +// White Noise +FN_DECIMAL FastNoise::GetWhiteNoise(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z, FN_DECIMAL w) const +{ + return ValCoord4D(m_seed, + *reinterpret_cast(&x) ^ (*reinterpret_cast(&x) >> 16), + *reinterpret_cast(&y) ^ (*reinterpret_cast(&y) >> 16), + *reinterpret_cast(&z) ^ (*reinterpret_cast(&z) >> 16), + *reinterpret_cast(&w) ^ (*reinterpret_cast(&w) >> 16)); +} + +FN_DECIMAL FastNoise::GetWhiteNoise(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + return ValCoord3D(m_seed, + *reinterpret_cast(&x) ^ (*reinterpret_cast(&x) >> 16), + *reinterpret_cast(&y) ^ (*reinterpret_cast(&y) >> 16), + *reinterpret_cast(&z) ^ (*reinterpret_cast(&z) >> 16)); +} + +FN_DECIMAL FastNoise::GetWhiteNoise(FN_DECIMAL x, FN_DECIMAL y) const +{ + return ValCoord2D(m_seed, + *reinterpret_cast(&x) ^ (*reinterpret_cast(&x) >> 16), + *reinterpret_cast(&y) ^ (*reinterpret_cast(&y) >> 16)); +} + +FN_DECIMAL FastNoise::GetWhiteNoiseInt(int x, int y, int z, int w) const +{ + return ValCoord4D(m_seed, x, y, z, w); +} + +FN_DECIMAL FastNoise::GetWhiteNoiseInt(int x, int y, int z) const +{ + return ValCoord3D(m_seed, x, y, z); +} + +FN_DECIMAL FastNoise::GetWhiteNoiseInt(int x, int y) const +{ + return ValCoord2D(m_seed, x, y); +} + +// Value Noise +FN_DECIMAL FastNoise::GetValueFractal(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + x *= m_frequency; + y *= m_frequency; + z *= m_frequency; + + switch (m_fractalType) + { + case FBM: + return SingleValueFractalFBM(x, y, z); + case Billow: + return SingleValueFractalBillow(x, y, z); + case RigidMulti: + return SingleValueFractalRigidMulti(x, y, z); + default: + return 0; + } +} + +FN_DECIMAL FastNoise::SingleValueFractalFBM(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + FN_DECIMAL sum = SingleValue(m_perm[0], x, y, z); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + z *= m_lacunarity; + + amp *= m_gain; + sum += SingleValue(m_perm[i], x, y, z) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SingleValueFractalBillow(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + FN_DECIMAL sum = FastAbs(SingleValue(m_perm[0], x, y, z)) * 2 - 1; + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + z *= m_lacunarity; + + amp *= m_gain; + sum += (FastAbs(SingleValue(m_perm[i], x, y, z)) * 2 - 1) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SingleValueFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + FN_DECIMAL sum = 1 - FastAbs(SingleValue(m_perm[0], x, y, z)); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + z *= m_lacunarity; + + amp *= m_gain; + sum -= (1 - FastAbs(SingleValue(m_perm[i], x, y, z))) * amp; + } + + return sum; +} + +FN_DECIMAL FastNoise::GetValue(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + return SingleValue(0, x * m_frequency, y * m_frequency, z * m_frequency); +} + +FN_DECIMAL FastNoise::SingleValue(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + int x0 = FastFloor(x); + int y0 = FastFloor(y); + int z0 = FastFloor(z); + int x1 = x0 + 1; + int y1 = y0 + 1; + int z1 = z0 + 1; + + FN_DECIMAL xs, ys, zs; + switch (m_interp) + { + case Linear: + xs = x - (FN_DECIMAL)x0; + ys = y - (FN_DECIMAL)y0; + zs = z - (FN_DECIMAL)z0; + break; + case Hermite: + xs = InterpHermiteFunc(x - (FN_DECIMAL)x0); + ys = InterpHermiteFunc(y - (FN_DECIMAL)y0); + zs = InterpHermiteFunc(z - (FN_DECIMAL)z0); + break; + case Quintic: + xs = InterpQuinticFunc(x - (FN_DECIMAL)x0); + ys = InterpQuinticFunc(y - (FN_DECIMAL)y0); + zs = InterpQuinticFunc(z - (FN_DECIMAL)z0); + break; + } + + FN_DECIMAL xf00 = Lerp(ValCoord3DFast(offset, x0, y0, z0), ValCoord3DFast(offset, x1, y0, z0), xs); + FN_DECIMAL xf10 = Lerp(ValCoord3DFast(offset, x0, y1, z0), ValCoord3DFast(offset, x1, y1, z0), xs); + FN_DECIMAL xf01 = Lerp(ValCoord3DFast(offset, x0, y0, z1), ValCoord3DFast(offset, x1, y0, z1), xs); + FN_DECIMAL xf11 = Lerp(ValCoord3DFast(offset, x0, y1, z1), ValCoord3DFast(offset, x1, y1, z1), xs); + + FN_DECIMAL yf0 = Lerp(xf00, xf10, ys); + FN_DECIMAL yf1 = Lerp(xf01, xf11, ys); + + return Lerp(yf0, yf1, zs); +} + +FN_DECIMAL FastNoise::GetValueFractal(FN_DECIMAL x, FN_DECIMAL y) const +{ + x *= m_frequency; + y *= m_frequency; + + switch (m_fractalType) + { + case FBM: + return SingleValueFractalFBM(x, y); + case Billow: + return SingleValueFractalBillow(x, y); + case RigidMulti: + return SingleValueFractalRigidMulti(x, y); + default: + return 0; + } +} + +FN_DECIMAL FastNoise::SingleValueFractalFBM(FN_DECIMAL x, FN_DECIMAL y) const +{ + FN_DECIMAL sum = SingleValue(m_perm[0], x, y); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + + amp *= m_gain; + sum += SingleValue(m_perm[i], x, y) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SingleValueFractalBillow(FN_DECIMAL x, FN_DECIMAL y) const +{ + FN_DECIMAL sum = FastAbs(SingleValue(m_perm[0], x, y)) * 2 - 1; + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + amp *= m_gain; + sum += (FastAbs(SingleValue(m_perm[i], x, y)) * 2 - 1) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SingleValueFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y) const +{ + FN_DECIMAL sum = 1 - FastAbs(SingleValue(m_perm[0], x, y)); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + + amp *= m_gain; + sum -= (1 - FastAbs(SingleValue(m_perm[i], x, y))) * amp; + } + + return sum; +} + +FN_DECIMAL FastNoise::GetValue(FN_DECIMAL x, FN_DECIMAL y) const +{ + return SingleValue(0, x * m_frequency, y * m_frequency); +} + +FN_DECIMAL FastNoise::SingleValue(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y) const +{ + int x0 = FastFloor(x); + int y0 = FastFloor(y); + int x1 = x0 + 1; + int y1 = y0 + 1; + + FN_DECIMAL xs, ys; + switch (m_interp) + { + case Linear: + xs = x - (FN_DECIMAL)x0; + ys = y - (FN_DECIMAL)y0; + break; + case Hermite: + xs = InterpHermiteFunc(x - (FN_DECIMAL)x0); + ys = InterpHermiteFunc(y - (FN_DECIMAL)y0); + break; + case Quintic: + xs = InterpQuinticFunc(x - (FN_DECIMAL)x0); + ys = InterpQuinticFunc(y - (FN_DECIMAL)y0); + break; + } + + FN_DECIMAL xf0 = Lerp(ValCoord2DFast(offset, x0, y0), ValCoord2DFast(offset, x1, y0), xs); + FN_DECIMAL xf1 = Lerp(ValCoord2DFast(offset, x0, y1), ValCoord2DFast(offset, x1, y1), xs); + + return Lerp(xf0, xf1, ys); +} + +// Perlin Noise +FN_DECIMAL FastNoise::GetPerlinFractal(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + x *= m_frequency; + y *= m_frequency; + z *= m_frequency; + + switch (m_fractalType) + { + case FBM: + return SinglePerlinFractalFBM(x, y, z); + case Billow: + return SinglePerlinFractalBillow(x, y, z); + case RigidMulti: + return SinglePerlinFractalRigidMulti(x, y, z); + default: + return 0; + } +} + +FN_DECIMAL FastNoise::SinglePerlinFractalFBM(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + FN_DECIMAL sum = SinglePerlin(m_perm[0], x, y, z); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + z *= m_lacunarity; + + amp *= m_gain; + sum += SinglePerlin(m_perm[i], x, y, z) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SinglePerlinFractalBillow(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + FN_DECIMAL sum = FastAbs(SinglePerlin(m_perm[0], x, y, z)) * 2 - 1; + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + z *= m_lacunarity; + + amp *= m_gain; + sum += (FastAbs(SinglePerlin(m_perm[i], x, y, z)) * 2 - 1) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SinglePerlinFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + FN_DECIMAL sum = 1 - FastAbs(SinglePerlin(m_perm[0], x, y, z)); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + z *= m_lacunarity; + + amp *= m_gain; + sum -= (1 - FastAbs(SinglePerlin(m_perm[i], x, y, z))) * amp; + } + + return sum; +} + +FN_DECIMAL FastNoise::GetPerlin(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + return SinglePerlin(0, x * m_frequency, y * m_frequency, z * m_frequency); +} + +FN_DECIMAL FastNoise::SinglePerlin(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + int x0 = FastFloor(x); + int y0 = FastFloor(y); + int z0 = FastFloor(z); + int x1 = x0 + 1; + int y1 = y0 + 1; + int z1 = z0 + 1; + + FN_DECIMAL xs, ys, zs; + switch (m_interp) + { + case Linear: + xs = x - (FN_DECIMAL)x0; + ys = y - (FN_DECIMAL)y0; + zs = z - (FN_DECIMAL)z0; + break; + case Hermite: + xs = InterpHermiteFunc(x - (FN_DECIMAL)x0); + ys = InterpHermiteFunc(y - (FN_DECIMAL)y0); + zs = InterpHermiteFunc(z - (FN_DECIMAL)z0); + break; + case Quintic: + xs = InterpQuinticFunc(x - (FN_DECIMAL)x0); + ys = InterpQuinticFunc(y - (FN_DECIMAL)y0); + zs = InterpQuinticFunc(z - (FN_DECIMAL)z0); + break; + } + + FN_DECIMAL xd0 = x - (FN_DECIMAL)x0; + FN_DECIMAL yd0 = y - (FN_DECIMAL)y0; + FN_DECIMAL zd0 = z - (FN_DECIMAL)z0; + FN_DECIMAL xd1 = xd0 - 1; + FN_DECIMAL yd1 = yd0 - 1; + FN_DECIMAL zd1 = zd0 - 1; + + FN_DECIMAL xf00 = Lerp(GradCoord3D(offset, x0, y0, z0, xd0, yd0, zd0), GradCoord3D(offset, x1, y0, z0, xd1, yd0, zd0), xs); + FN_DECIMAL xf10 = Lerp(GradCoord3D(offset, x0, y1, z0, xd0, yd1, zd0), GradCoord3D(offset, x1, y1, z0, xd1, yd1, zd0), xs); + FN_DECIMAL xf01 = Lerp(GradCoord3D(offset, x0, y0, z1, xd0, yd0, zd1), GradCoord3D(offset, x1, y0, z1, xd1, yd0, zd1), xs); + FN_DECIMAL xf11 = Lerp(GradCoord3D(offset, x0, y1, z1, xd0, yd1, zd1), GradCoord3D(offset, x1, y1, z1, xd1, yd1, zd1), xs); + + FN_DECIMAL yf0 = Lerp(xf00, xf10, ys); + FN_DECIMAL yf1 = Lerp(xf01, xf11, ys); + + return Lerp(yf0, yf1, zs); +} + +FN_DECIMAL FastNoise::GetPerlinFractal(FN_DECIMAL x, FN_DECIMAL y) const +{ + x *= m_frequency; + y *= m_frequency; + + switch (m_fractalType) + { + case FBM: + return SinglePerlinFractalFBM(x, y); + case Billow: + return SinglePerlinFractalBillow(x, y); + case RigidMulti: + return SinglePerlinFractalRigidMulti(x, y); + default: + return 0; + } +} + +FN_DECIMAL FastNoise::SinglePerlinFractalFBM(FN_DECIMAL x, FN_DECIMAL y) const +{ + FN_DECIMAL sum = SinglePerlin(m_perm[0], x, y); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + + amp *= m_gain; + sum += SinglePerlin(m_perm[i], x, y) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SinglePerlinFractalBillow(FN_DECIMAL x, FN_DECIMAL y) const +{ + FN_DECIMAL sum = FastAbs(SinglePerlin(m_perm[0], x, y)) * 2 - 1; + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + + amp *= m_gain; + sum += (FastAbs(SinglePerlin(m_perm[i], x, y)) * 2 - 1) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SinglePerlinFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y) const +{ + FN_DECIMAL sum = 1 - FastAbs(SinglePerlin(m_perm[0], x, y)); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + + amp *= m_gain; + sum -= (1 - FastAbs(SinglePerlin(m_perm[i], x, y))) * amp; + } + + return sum; +} + +FN_DECIMAL FastNoise::GetPerlin(FN_DECIMAL x, FN_DECIMAL y) const +{ + return SinglePerlin(0, x * m_frequency, y * m_frequency); +} + +FN_DECIMAL FastNoise::SinglePerlin(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y) const +{ + int x0 = FastFloor(x); + int y0 = FastFloor(y); + int x1 = x0 + 1; + int y1 = y0 + 1; + + FN_DECIMAL xs, ys; + switch (m_interp) + { + case Linear: + xs = x - (FN_DECIMAL)x0; + ys = y - (FN_DECIMAL)y0; + break; + case Hermite: + xs = InterpHermiteFunc(x - (FN_DECIMAL)x0); + ys = InterpHermiteFunc(y - (FN_DECIMAL)y0); + break; + case Quintic: + xs = InterpQuinticFunc(x - (FN_DECIMAL)x0); + ys = InterpQuinticFunc(y - (FN_DECIMAL)y0); + break; + } + + FN_DECIMAL xd0 = x - (FN_DECIMAL)x0; + FN_DECIMAL yd0 = y - (FN_DECIMAL)y0; + FN_DECIMAL xd1 = xd0 - 1; + FN_DECIMAL yd1 = yd0 - 1; + + FN_DECIMAL xf0 = Lerp(GradCoord2D(offset, x0, y0, xd0, yd0), GradCoord2D(offset, x1, y0, xd1, yd0), xs); + FN_DECIMAL xf1 = Lerp(GradCoord2D(offset, x0, y1, xd0, yd1), GradCoord2D(offset, x1, y1, xd1, yd1), xs); + + return Lerp(xf0, xf1, ys); +} + +// Simplex Noise + +FN_DECIMAL FastNoise::GetSimplexFractal(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + x *= m_frequency; + y *= m_frequency; + z *= m_frequency; + + switch (m_fractalType) + { + case FBM: + return SingleSimplexFractalFBM(x, y, z); + case Billow: + return SingleSimplexFractalBillow(x, y, z); + case RigidMulti: + return SingleSimplexFractalRigidMulti(x, y, z); + default: + return 0; + } +} + +FN_DECIMAL FastNoise::SingleSimplexFractalFBM(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + FN_DECIMAL sum = SingleSimplex(m_perm[0], x, y, z); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + z *= m_lacunarity; + + amp *= m_gain; + sum += SingleSimplex(m_perm[i], x, y, z) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SingleSimplexFractalBillow(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + FN_DECIMAL sum = FastAbs(SingleSimplex(m_perm[0], x, y, z)) * 2 - 1; + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + z *= m_lacunarity; + + amp *= m_gain; + sum += (FastAbs(SingleSimplex(m_perm[i], x, y, z)) * 2 - 1) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SingleSimplexFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + FN_DECIMAL sum = 1 - FastAbs(SingleSimplex(m_perm[0], x, y, z)); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + z *= m_lacunarity; + + amp *= m_gain; + sum -= (1 - FastAbs(SingleSimplex(m_perm[i], x, y, z))) * amp; + } + + return sum; +} + +FN_DECIMAL FastNoise::GetSimplex(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + return SingleSimplex(0, x * m_frequency, y * m_frequency, z * m_frequency); +} + +static const FN_DECIMAL F3 = 1 / FN_DECIMAL(3); +static const FN_DECIMAL G3 = 1 / FN_DECIMAL(6); + +FN_DECIMAL FastNoise::SingleSimplex(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + FN_DECIMAL t = (x + y + z) * F3; + int i = FastFloor(x + t); + int j = FastFloor(y + t); + int k = FastFloor(z + t); + + t = (i + j + k) * G3; + FN_DECIMAL X0 = i - t; + FN_DECIMAL Y0 = j - t; + FN_DECIMAL Z0 = k - t; + + FN_DECIMAL x0 = x - X0; + FN_DECIMAL y0 = y - Y0; + FN_DECIMAL z0 = z - Z0; + + int i1, j1, k1; + int i2, j2, k2; + + if (x0 >= y0) + { + if (y0 >= z0) + { + i1 = 1; j1 = 0; k1 = 0; i2 = 1; j2 = 1; k2 = 0; + } + else if (x0 >= z0) + { + i1 = 1; j1 = 0; k1 = 0; i2 = 1; j2 = 0; k2 = 1; + } + else // x0 < z0 + { + i1 = 0; j1 = 0; k1 = 1; i2 = 1; j2 = 0; k2 = 1; + } + } + else // x0 < y0 + { + if (y0 < z0) + { + i1 = 0; j1 = 0; k1 = 1; i2 = 0; j2 = 1; k2 = 1; + } + else if (x0 < z0) + { + i1 = 0; j1 = 1; k1 = 0; i2 = 0; j2 = 1; k2 = 1; + } + else // x0 >= z0 + { + i1 = 0; j1 = 1; k1 = 0; i2 = 1; j2 = 1; k2 = 0; + } + } + + FN_DECIMAL x1 = x0 - i1 + G3; + FN_DECIMAL y1 = y0 - j1 + G3; + FN_DECIMAL z1 = z0 - k1 + G3; + FN_DECIMAL x2 = x0 - i2 + 2*G3; + FN_DECIMAL y2 = y0 - j2 + 2*G3; + FN_DECIMAL z2 = z0 - k2 + 2*G3; + FN_DECIMAL x3 = x0 - 1 + 3*G3; + FN_DECIMAL y3 = y0 - 1 + 3*G3; + FN_DECIMAL z3 = z0 - 1 + 3*G3; + + FN_DECIMAL n0, n1, n2, n3; + + t = FN_DECIMAL(0.6) - x0*x0 - y0*y0 - z0*z0; + if (t < 0) n0 = 0; + else + { + t *= t; + n0 = t*t*GradCoord3D(offset, i, j, k, x0, y0, z0); + } + + t = FN_DECIMAL(0.6) - x1*x1 - y1*y1 - z1*z1; + if (t < 0) n1 = 0; + else + { + t *= t; + n1 = t*t*GradCoord3D(offset, i + i1, j + j1, k + k1, x1, y1, z1); + } + + t = FN_DECIMAL(0.6) - x2*x2 - y2*y2 - z2*z2; + if (t < 0) n2 = 0; + else + { + t *= t; + n2 = t*t*GradCoord3D(offset, i + i2, j + j2, k + k2, x2, y2, z2); + } + + t = FN_DECIMAL(0.6) - x3*x3 - y3*y3 - z3*z3; + if (t < 0) n3 = 0; + else + { + t *= t; + n3 = t*t*GradCoord3D(offset, i + 1, j + 1, k + 1, x3, y3, z3); + } + + return 32 * (n0 + n1 + n2 + n3); +} + +FN_DECIMAL FastNoise::GetSimplexFractal(FN_DECIMAL x, FN_DECIMAL y) const +{ + x *= m_frequency; + y *= m_frequency; + + switch (m_fractalType) + { + case FBM: + return SingleSimplexFractalFBM(x, y); + case Billow: + return SingleSimplexFractalBillow(x, y); + case RigidMulti: + return SingleSimplexFractalRigidMulti(x, y); + default: + return 0; + } +} + +FN_DECIMAL FastNoise::SingleSimplexFractalFBM(FN_DECIMAL x, FN_DECIMAL y) const +{ + FN_DECIMAL sum = SingleSimplex(m_perm[0], x, y); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + + amp *= m_gain; + sum += SingleSimplex(m_perm[i], x, y) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SingleSimplexFractalBillow(FN_DECIMAL x, FN_DECIMAL y) const +{ + FN_DECIMAL sum = FastAbs(SingleSimplex(m_perm[0], x, y)) * 2 - 1; + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + + amp *= m_gain; + sum += (FastAbs(SingleSimplex(m_perm[i], x, y)) * 2 - 1) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SingleSimplexFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y) const +{ + FN_DECIMAL sum = 1 - FastAbs(SingleSimplex(m_perm[0], x, y)); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + + amp *= m_gain; + sum -= (1 - FastAbs(SingleSimplex(m_perm[i], x, y))) * amp; + } + + return sum; +} + +FN_DECIMAL FastNoise::SingleSimplexFractalBlend(FN_DECIMAL x, FN_DECIMAL y) const +{ + FN_DECIMAL sum = SingleSimplex(m_perm[0], x, y); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + + amp *= m_gain; + sum *= SingleSimplex(m_perm[i], x, y) * amp + 1; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::GetSimplex(FN_DECIMAL x, FN_DECIMAL y) const +{ + return SingleSimplex(0, x * m_frequency, y * m_frequency); +} + +//static const FN_DECIMAL F2 = 1 / FN_DECIMAL(2); +//static const FN_DECIMAL G2 = 1 / FN_DECIMAL(4); + +static const FN_DECIMAL SQRT3 = FN_DECIMAL(1.7320508075688772935274463415059); +static const FN_DECIMAL F2 = FN_DECIMAL(0.5) * (SQRT3 - FN_DECIMAL(1.0)); +static const FN_DECIMAL G2 = (FN_DECIMAL(3.0) - SQRT3) / FN_DECIMAL(6.0); + +FN_DECIMAL FastNoise::SingleSimplex(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y) const +{ + FN_DECIMAL t = (x + y) * F2; + int i = FastFloor(x + t); + int j = FastFloor(y + t); + + t = (i + j) * G2; + FN_DECIMAL X0 = i - t; + FN_DECIMAL Y0 = j - t; + + FN_DECIMAL x0 = x - X0; + FN_DECIMAL y0 = y - Y0; + + int i1, j1; + if (x0 > y0) + { + i1 = 1; j1 = 0; + } + else + { + i1 = 0; j1 = 1; + } + + FN_DECIMAL x1 = x0 - (FN_DECIMAL)i1 + G2; + FN_DECIMAL y1 = y0 - (FN_DECIMAL)j1 + G2; + FN_DECIMAL x2 = x0 - 1 + 2*G2; + FN_DECIMAL y2 = y0 - 1 + 2*G2; + + FN_DECIMAL n0, n1, n2; + + t = FN_DECIMAL(0.5) - x0*x0 - y0*y0; + if (t < 0) n0 = 0; + else + { + t *= t; + n0 = t * t * GradCoord2D(offset, i, j, x0, y0); + } + + t = FN_DECIMAL(0.5) - x1*x1 - y1*y1; + if (t < 0) n1 = 0; + else + { + t *= t; + n1 = t*t*GradCoord2D(offset, i + i1, j + j1, x1, y1); + } + + t = FN_DECIMAL(0.5) - x2*x2 - y2*y2; + if (t < 0) n2 = 0; + else + { + t *= t; + n2 = t*t*GradCoord2D(offset, i + 1, j + 1, x2, y2); + } + + return 70 * (n0 + n1 + n2); +} + +FN_DECIMAL FastNoise::GetSimplex(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z, FN_DECIMAL w) const +{ + return SingleSimplex(0, x * m_frequency, y * m_frequency, z * m_frequency, w * m_frequency); +} + +static const FN_DECIMAL F4 = (sqrt(FN_DECIMAL(5)) - 1) / 4; +static const FN_DECIMAL G4 = (5 - sqrt(FN_DECIMAL(5))) / 20; + +FN_DECIMAL FastNoise::SingleSimplex(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z, FN_DECIMAL w) const +{ + FN_DECIMAL n0, n1, n2, n3, n4; + FN_DECIMAL t = (x + y + z + w) * F4; + int i = FastFloor(x + t); + int j = FastFloor(y + t); + int k = FastFloor(z + t); + int l = FastFloor(w + t); + t = (i + j + k + l) * G4; + FN_DECIMAL X0 = i - t; + FN_DECIMAL Y0 = j - t; + FN_DECIMAL Z0 = k - t; + FN_DECIMAL W0 = l - t; + FN_DECIMAL x0 = x - X0; + FN_DECIMAL y0 = y - Y0; + FN_DECIMAL z0 = z - Z0; + FN_DECIMAL w0 = w - W0; + + int rankx = 0; + int ranky = 0; + int rankz = 0; + int rankw = 0; + + if (x0 > y0) rankx++; else ranky++; + if (x0 > z0) rankx++; else rankz++; + if (x0 > w0) rankx++; else rankw++; + if (y0 > z0) ranky++; else rankz++; + if (y0 > w0) ranky++; else rankw++; + if (z0 > w0) rankz++; else rankw++; + + int i1 = rankx >= 3 ? 1 : 0; + int j1 = ranky >= 3 ? 1 : 0; + int k1 = rankz >= 3 ? 1 : 0; + int l1 = rankw >= 3 ? 1 : 0; + + int i2 = rankx >= 2 ? 1 : 0; + int j2 = ranky >= 2 ? 1 : 0; + int k2 = rankz >= 2 ? 1 : 0; + int l2 = rankw >= 2 ? 1 : 0; + + int i3 = rankx >= 1 ? 1 : 0; + int j3 = ranky >= 1 ? 1 : 0; + int k3 = rankz >= 1 ? 1 : 0; + int l3 = rankw >= 1 ? 1 : 0; + + FN_DECIMAL x1 = x0 - i1 + G4; + FN_DECIMAL y1 = y0 - j1 + G4; + FN_DECIMAL z1 = z0 - k1 + G4; + FN_DECIMAL w1 = w0 - l1 + G4; + FN_DECIMAL x2 = x0 - i2 + 2*G4; + FN_DECIMAL y2 = y0 - j2 + 2*G4; + FN_DECIMAL z2 = z0 - k2 + 2*G4; + FN_DECIMAL w2 = w0 - l2 + 2*G4; + FN_DECIMAL x3 = x0 - i3 + 3*G4; + FN_DECIMAL y3 = y0 - j3 + 3*G4; + FN_DECIMAL z3 = z0 - k3 + 3*G4; + FN_DECIMAL w3 = w0 - l3 + 3*G4; + FN_DECIMAL x4 = x0 - 1 + 4*G4; + FN_DECIMAL y4 = y0 - 1 + 4*G4; + FN_DECIMAL z4 = z0 - 1 + 4*G4; + FN_DECIMAL w4 = w0 - 1 + 4*G4; + + t = FN_DECIMAL(0.6) - x0*x0 - y0*y0 - z0*z0 - w0*w0; + if (t < 0) n0 = 0; + else { + t *= t; + n0 = t * t * GradCoord4D(offset, i, j, k, l, x0, y0, z0, w0); + } + t = FN_DECIMAL(0.6) - x1*x1 - y1*y1 - z1*z1 - w1*w1; + if (t < 0) n1 = 0; + else { + t *= t; + n1 = t * t * GradCoord4D(offset, i + i1, j + j1, k + k1, l + l1, x1, y1, z1, w1); + } + t = FN_DECIMAL(0.6) - x2*x2 - y2*y2 - z2*z2 - w2*w2; + if (t < 0) n2 = 0; + else { + t *= t; + n2 = t * t * GradCoord4D(offset, i + i2, j + j2, k + k2, l + l2, x2, y2, z2, w2); + } + t = FN_DECIMAL(0.6) - x3*x3 - y3*y3 - z3*z3 - w3*w3; + if (t < 0) n3 = 0; + else { + t *= t; + n3 = t * t * GradCoord4D(offset, i + i3, j + j3, k + k3, l + l3, x3, y3, z3, w3); + } + t = FN_DECIMAL(0.6) - x4*x4 - y4*y4 - z4*z4 - w4*w4; + if (t < 0) n4 = 0; + else { + t *= t; + n4 = t * t * GradCoord4D(offset, i + 1, j + 1, k + 1, l + 1, x4, y4, z4, w4); + } + + return 27 * (n0 + n1 + n2 + n3 + n4); +} + +// Cubic Noise +FN_DECIMAL FastNoise::GetCubicFractal(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + x *= m_frequency; + y *= m_frequency; + z *= m_frequency; + + switch (m_fractalType) + { + case FBM: + return SingleCubicFractalFBM(x, y, z); + case Billow: + return SingleCubicFractalBillow(x, y, z); + case RigidMulti: + return SingleCubicFractalRigidMulti(x, y, z); + default: + return 0; + } +} + +FN_DECIMAL FastNoise::SingleCubicFractalFBM(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + FN_DECIMAL sum = SingleCubic(m_perm[0], x, y, z); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + z *= m_lacunarity; + + amp *= m_gain; + sum += SingleCubic(m_perm[i], x, y, z) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SingleCubicFractalBillow(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + FN_DECIMAL sum = FastAbs(SingleCubic(m_perm[0], x, y, z)) * 2 - 1; + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + z *= m_lacunarity; + + amp *= m_gain; + sum += (FastAbs(SingleCubic(m_perm[i], x, y, z)) * 2 - 1) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SingleCubicFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + FN_DECIMAL sum = 1 - FastAbs(SingleCubic(m_perm[0], x, y, z)); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + z *= m_lacunarity; + + amp *= m_gain; + sum -= (1 - FastAbs(SingleCubic(m_perm[i], x, y, z))) * amp; + } + + return sum; +} + +FN_DECIMAL FastNoise::GetCubic(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + return SingleCubic(0, x * m_frequency, y * m_frequency, z * m_frequency); +} + +const FN_DECIMAL CUBIC_3D_BOUNDING = 1 / (FN_DECIMAL(1.5) * FN_DECIMAL(1.5) * FN_DECIMAL(1.5)); + +FN_DECIMAL FastNoise::SingleCubic(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + int x1 = FastFloor(x); + int y1 = FastFloor(y); + int z1 = FastFloor(z); + + int x0 = x1 - 1; + int y0 = y1 - 1; + int z0 = z1 - 1; + int x2 = x1 + 1; + int y2 = y1 + 1; + int z2 = z1 + 1; + int x3 = x1 + 2; + int y3 = y1 + 2; + int z3 = z1 + 2; + + FN_DECIMAL xs = x - (FN_DECIMAL)x1; + FN_DECIMAL ys = y - (FN_DECIMAL)y1; + FN_DECIMAL zs = z - (FN_DECIMAL)z1; + + return CubicLerp( + CubicLerp( + CubicLerp(ValCoord3DFast(offset, x0, y0, z0), ValCoord3DFast(offset, x1, y0, z0), ValCoord3DFast(offset, x2, y0, z0), ValCoord3DFast(offset, x3, y0, z0), xs), + CubicLerp(ValCoord3DFast(offset, x0, y1, z0), ValCoord3DFast(offset, x1, y1, z0), ValCoord3DFast(offset, x2, y1, z0), ValCoord3DFast(offset, x3, y1, z0), xs), + CubicLerp(ValCoord3DFast(offset, x0, y2, z0), ValCoord3DFast(offset, x1, y2, z0), ValCoord3DFast(offset, x2, y2, z0), ValCoord3DFast(offset, x3, y2, z0), xs), + CubicLerp(ValCoord3DFast(offset, x0, y3, z0), ValCoord3DFast(offset, x1, y3, z0), ValCoord3DFast(offset, x2, y3, z0), ValCoord3DFast(offset, x3, y3, z0), xs), + ys), + CubicLerp( + CubicLerp(ValCoord3DFast(offset, x0, y0, z1), ValCoord3DFast(offset, x1, y0, z1), ValCoord3DFast(offset, x2, y0, z1), ValCoord3DFast(offset, x3, y0, z1), xs), + CubicLerp(ValCoord3DFast(offset, x0, y1, z1), ValCoord3DFast(offset, x1, y1, z1), ValCoord3DFast(offset, x2, y1, z1), ValCoord3DFast(offset, x3, y1, z1), xs), + CubicLerp(ValCoord3DFast(offset, x0, y2, z1), ValCoord3DFast(offset, x1, y2, z1), ValCoord3DFast(offset, x2, y2, z1), ValCoord3DFast(offset, x3, y2, z1), xs), + CubicLerp(ValCoord3DFast(offset, x0, y3, z1), ValCoord3DFast(offset, x1, y3, z1), ValCoord3DFast(offset, x2, y3, z1), ValCoord3DFast(offset, x3, y3, z1), xs), + ys), + CubicLerp( + CubicLerp(ValCoord3DFast(offset, x0, y0, z2), ValCoord3DFast(offset, x1, y0, z2), ValCoord3DFast(offset, x2, y0, z2), ValCoord3DFast(offset, x3, y0, z2), xs), + CubicLerp(ValCoord3DFast(offset, x0, y1, z2), ValCoord3DFast(offset, x1, y1, z2), ValCoord3DFast(offset, x2, y1, z2), ValCoord3DFast(offset, x3, y1, z2), xs), + CubicLerp(ValCoord3DFast(offset, x0, y2, z2), ValCoord3DFast(offset, x1, y2, z2), ValCoord3DFast(offset, x2, y2, z2), ValCoord3DFast(offset, x3, y2, z2), xs), + CubicLerp(ValCoord3DFast(offset, x0, y3, z2), ValCoord3DFast(offset, x1, y3, z2), ValCoord3DFast(offset, x2, y3, z2), ValCoord3DFast(offset, x3, y3, z2), xs), + ys), + CubicLerp( + CubicLerp(ValCoord3DFast(offset, x0, y0, z3), ValCoord3DFast(offset, x1, y0, z3), ValCoord3DFast(offset, x2, y0, z3), ValCoord3DFast(offset, x3, y0, z3), xs), + CubicLerp(ValCoord3DFast(offset, x0, y1, z3), ValCoord3DFast(offset, x1, y1, z3), ValCoord3DFast(offset, x2, y1, z3), ValCoord3DFast(offset, x3, y1, z3), xs), + CubicLerp(ValCoord3DFast(offset, x0, y2, z3), ValCoord3DFast(offset, x1, y2, z3), ValCoord3DFast(offset, x2, y2, z3), ValCoord3DFast(offset, x3, y2, z3), xs), + CubicLerp(ValCoord3DFast(offset, x0, y3, z3), ValCoord3DFast(offset, x1, y3, z3), ValCoord3DFast(offset, x2, y3, z3), ValCoord3DFast(offset, x3, y3, z3), xs), + ys), + zs) * CUBIC_3D_BOUNDING; +} + + +FN_DECIMAL FastNoise::GetCubicFractal(FN_DECIMAL x, FN_DECIMAL y) const +{ + x *= m_frequency; + y *= m_frequency; + + switch (m_fractalType) + { + case FBM: + return SingleCubicFractalFBM(x, y); + case Billow: + return SingleCubicFractalBillow(x, y); + case RigidMulti: + return SingleCubicFractalRigidMulti(x, y); + default: + return 0; + } +} + +FN_DECIMAL FastNoise::SingleCubicFractalFBM(FN_DECIMAL x, FN_DECIMAL y) const +{ + FN_DECIMAL sum = SingleCubic(m_perm[0], x, y); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + + amp *= m_gain; + sum += SingleCubic(m_perm[i], x, y) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SingleCubicFractalBillow(FN_DECIMAL x, FN_DECIMAL y) const +{ + FN_DECIMAL sum = FastAbs(SingleCubic(m_perm[0], x, y)) * 2 - 1; + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + + amp *= m_gain; + sum += (FastAbs(SingleCubic(m_perm[i], x, y)) * 2 - 1) * amp; + } + + return sum * m_fractalBounding; +} + +FN_DECIMAL FastNoise::SingleCubicFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y) const +{ + FN_DECIMAL sum = 1 - FastAbs(SingleCubic(m_perm[0], x, y)); + FN_DECIMAL amp = 1; + int i = 0; + + while (++i < m_octaves) + { + x *= m_lacunarity; + y *= m_lacunarity; + + amp *= m_gain; + sum -= (1 - FastAbs(SingleCubic(m_perm[i], x, y))) * amp; + } + + return sum; +} + +FN_DECIMAL FastNoise::GetCubic(FN_DECIMAL x, FN_DECIMAL y) const +{ + x *= m_frequency; + y *= m_frequency; + + return SingleCubic(0, x, y); +} + +const FN_DECIMAL CUBIC_2D_BOUNDING = 1 / (FN_DECIMAL(1.5) * FN_DECIMAL(1.5)); + +FN_DECIMAL FastNoise::SingleCubic(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y) const +{ + int x1 = FastFloor(x); + int y1 = FastFloor(y); + + int x0 = x1 - 1; + int y0 = y1 - 1; + int x2 = x1 + 1; + int y2 = y1 + 1; + int x3 = x1 + 2; + int y3 = y1 + 2; + + FN_DECIMAL xs = x - (FN_DECIMAL)x1; + FN_DECIMAL ys = y - (FN_DECIMAL)y1; + + return CubicLerp( + CubicLerp(ValCoord2DFast(offset, x0, y0), ValCoord2DFast(offset, x1, y0), ValCoord2DFast(offset, x2, y0), ValCoord2DFast(offset, x3, y0), xs), + CubicLerp(ValCoord2DFast(offset, x0, y1), ValCoord2DFast(offset, x1, y1), ValCoord2DFast(offset, x2, y1), ValCoord2DFast(offset, x3, y1), xs), + CubicLerp(ValCoord2DFast(offset, x0, y2), ValCoord2DFast(offset, x1, y2), ValCoord2DFast(offset, x2, y2), ValCoord2DFast(offset, x3, y2), xs), + CubicLerp(ValCoord2DFast(offset, x0, y3), ValCoord2DFast(offset, x1, y3), ValCoord2DFast(offset, x2, y3), ValCoord2DFast(offset, x3, y3), xs), + ys) * CUBIC_2D_BOUNDING; +} + +// Cellular Noise +FN_DECIMAL FastNoise::GetCellular(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + x *= m_frequency; + y *= m_frequency; + z *= m_frequency; + + switch (m_cellularReturnType) + { + case CellValue: + case NoiseLookup: + case Distance: + return SingleCellular(x, y, z); + default: + return SingleCellular2Edge(x, y, z); + } +} + +FN_DECIMAL FastNoise::SingleCellular(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + int xr = FastRound(x); + int yr = FastRound(y); + int zr = FastRound(z); + + FN_DECIMAL distance = 999999; + int xc, yc, zc; + + switch (m_cellularDistanceFunction) + { + case Euclidean: + for (int xi = xr - 1; xi <= xr + 1; xi++) + { + for (int yi = yr - 1; yi <= yr + 1; yi++) + { + for (int zi = zr - 1; zi <= zr + 1; zi++) + { + unsigned char lutPos = Index3D_256(0, xi, yi, zi); + + FN_DECIMAL vecX = xi - x + CELL_3D_X[lutPos] * m_cellularJitter; + FN_DECIMAL vecY = yi - y + CELL_3D_Y[lutPos] * m_cellularJitter; + FN_DECIMAL vecZ = zi - z + CELL_3D_Z[lutPos] * m_cellularJitter; + + FN_DECIMAL newDistance = vecX * vecX + vecY * vecY + vecZ * vecZ; + + if (newDistance < distance) + { + distance = newDistance; + xc = xi; + yc = yi; + zc = zi; + } + } + } + } + break; + case Manhattan: + for (int xi = xr - 1; xi <= xr + 1; xi++) + { + for (int yi = yr - 1; yi <= yr + 1; yi++) + { + for (int zi = zr - 1; zi <= zr + 1; zi++) + { + unsigned char lutPos = Index3D_256(0, xi, yi, zi); + + FN_DECIMAL vecX = xi - x + CELL_3D_X[lutPos] * m_cellularJitter; + FN_DECIMAL vecY = yi - y + CELL_3D_Y[lutPos] * m_cellularJitter; + FN_DECIMAL vecZ = zi - z + CELL_3D_Z[lutPos] * m_cellularJitter; + + FN_DECIMAL newDistance = FastAbs(vecX) + FastAbs(vecY) + FastAbs(vecZ); + + if (newDistance < distance) + { + distance = newDistance; + xc = xi; + yc = yi; + zc = zi; + } + } + } + } + break; + case Natural: + for (int xi = xr - 1; xi <= xr + 1; xi++) + { + for (int yi = yr - 1; yi <= yr + 1; yi++) + { + for (int zi = zr - 1; zi <= zr + 1; zi++) + { + unsigned char lutPos = Index3D_256(0, xi, yi, zi); + + FN_DECIMAL vecX = xi - x + CELL_3D_X[lutPos] * m_cellularJitter; + FN_DECIMAL vecY = yi - y + CELL_3D_Y[lutPos] * m_cellularJitter; + FN_DECIMAL vecZ = zi - z + CELL_3D_Z[lutPos] * m_cellularJitter; + + FN_DECIMAL newDistance = (FastAbs(vecX) + FastAbs(vecY) + FastAbs(vecZ)) + (vecX * vecX + vecY * vecY + vecZ * vecZ); + + if (newDistance < distance) + { + distance = newDistance; + xc = xi; + yc = yi; + zc = zi; + } + } + } + } + break; + default: + break; + } + + unsigned char lutPos; + switch (m_cellularReturnType) + { + case CellValue: + return ValCoord3D(m_seed, xc, yc, zc); + + case NoiseLookup: + assert(m_cellularNoiseLookup); + + lutPos = Index3D_256(0, xc, yc, zc); + return m_cellularNoiseLookup->GetNoise(xc + CELL_3D_X[lutPos] * m_cellularJitter, yc + CELL_3D_Y[lutPos] * m_cellularJitter, zc + CELL_3D_Z[lutPos] * m_cellularJitter); + + case Distance: + return distance; + default: + return 0; + } +} + +FN_DECIMAL FastNoise::SingleCellular2Edge(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const +{ + int xr = FastRound(x); + int yr = FastRound(y); + int zr = FastRound(z); + + FN_DECIMAL distance[FN_CELLULAR_INDEX_MAX+1] = { 999999,999999,999999,999999 }; + + switch (m_cellularDistanceFunction) + { + case Euclidean: + for (int xi = xr - 1; xi <= xr + 1; xi++) + { + for (int yi = yr - 1; yi <= yr + 1; yi++) + { + for (int zi = zr - 1; zi <= zr + 1; zi++) + { + unsigned char lutPos = Index3D_256(0, xi, yi, zi); + + FN_DECIMAL vecX = xi - x + CELL_3D_X[lutPos] * m_cellularJitter; + FN_DECIMAL vecY = yi - y + CELL_3D_Y[lutPos] * m_cellularJitter; + FN_DECIMAL vecZ = zi - z + CELL_3D_Z[lutPos] * m_cellularJitter; + + FN_DECIMAL newDistance = vecX * vecX + vecY * vecY + vecZ * vecZ; + + for (int i = m_cellularDistanceIndex1; i > 0; i--) + distance[i] = fmax(fmin(distance[i], newDistance), distance[i - 1]); + distance[0] = fmin(distance[0], newDistance); + } + } + } + break; + case Manhattan: + for (int xi = xr - 1; xi <= xr + 1; xi++) + { + for (int yi = yr - 1; yi <= yr + 1; yi++) + { + for (int zi = zr - 1; zi <= zr + 1; zi++) + { + unsigned char lutPos = Index3D_256(0, xi, yi, zi); + + FN_DECIMAL vecX = xi - x + CELL_3D_X[lutPos] * m_cellularJitter; + FN_DECIMAL vecY = yi - y + CELL_3D_Y[lutPos] * m_cellularJitter; + FN_DECIMAL vecZ = zi - z + CELL_3D_Z[lutPos] * m_cellularJitter; + + FN_DECIMAL newDistance = FastAbs(vecX) + FastAbs(vecY) + FastAbs(vecZ); + + for (int i = m_cellularDistanceIndex1; i > 0; i--) + distance[i] = fmax(fmin(distance[i], newDistance), distance[i - 1]); + distance[0] = fmin(distance[0], newDistance); + } + } + } + break; + case Natural: + for (int xi = xr - 1; xi <= xr + 1; xi++) + { + for (int yi = yr - 1; yi <= yr + 1; yi++) + { + for (int zi = zr - 1; zi <= zr + 1; zi++) + { + unsigned char lutPos = Index3D_256(0, xi, yi, zi); + + FN_DECIMAL vecX = xi - x + CELL_3D_X[lutPos] * m_cellularJitter; + FN_DECIMAL vecY = yi - y + CELL_3D_Y[lutPos] * m_cellularJitter; + FN_DECIMAL vecZ = zi - z + CELL_3D_Z[lutPos] * m_cellularJitter; + + FN_DECIMAL newDistance = (FastAbs(vecX) + FastAbs(vecY) + FastAbs(vecZ)) + (vecX * vecX + vecY * vecY + vecZ * vecZ); + + for (int i = m_cellularDistanceIndex1; i > 0; i--) + distance[i] = fmax(fmin(distance[i], newDistance), distance[i - 1]); + distance[0] = fmin(distance[0], newDistance); + } + } + } + break; + default: + break; + } + + switch (m_cellularReturnType) + { + case Distance2: + return distance[m_cellularDistanceIndex1]; + case Distance2Add: + return distance[m_cellularDistanceIndex1] + distance[m_cellularDistanceIndex0]; + case Distance2Sub: + return distance[m_cellularDistanceIndex1] - distance[m_cellularDistanceIndex0]; + case Distance2Mul: + return distance[m_cellularDistanceIndex1] * distance[m_cellularDistanceIndex0]; + case Distance2Div: + return distance[m_cellularDistanceIndex0] / distance[m_cellularDistanceIndex1]; + default: + return 0; + } +} + +FN_DECIMAL FastNoise::GetCellular(FN_DECIMAL x, FN_DECIMAL y) const +{ + x *= m_frequency; + y *= m_frequency; + + switch (m_cellularReturnType) + { + case CellValue: + case NoiseLookup: + case Distance: + return SingleCellular(x, y); + default: + return SingleCellular2Edge(x, y); + } +} + +FN_DECIMAL FastNoise::SingleCellular(FN_DECIMAL x, FN_DECIMAL y) const +{ + int xr = FastRound(x); + int yr = FastRound(y); + + FN_DECIMAL distance = 999999; + int xc, yc; + + switch (m_cellularDistanceFunction) + { + default: + case Euclidean: + for (int xi = xr - 1; xi <= xr + 1; xi++) + { + for (int yi = yr - 1; yi <= yr + 1; yi++) + { + unsigned char lutPos = Index2D_256(0, xi, yi); + + FN_DECIMAL vecX = xi - x + CELL_2D_X[lutPos] * m_cellularJitter; + FN_DECIMAL vecY = yi - y + CELL_2D_Y[lutPos] * m_cellularJitter; + + FN_DECIMAL newDistance = vecX * vecX + vecY * vecY; + + if (newDistance < distance) + { + distance = newDistance; + xc = xi; + yc = yi; + } + } + } + break; + case Manhattan: + for (int xi = xr - 1; xi <= xr + 1; xi++) + { + for (int yi = yr - 1; yi <= yr + 1; yi++) + { + unsigned char lutPos = Index2D_256(0, xi, yi); + + FN_DECIMAL vecX = xi - x + CELL_2D_X[lutPos] * m_cellularJitter; + FN_DECIMAL vecY = yi - y + CELL_2D_Y[lutPos] * m_cellularJitter; + + FN_DECIMAL newDistance = (FastAbs(vecX) + FastAbs(vecY)); + + if (newDistance < distance) + { + distance = newDistance; + xc = xi; + yc = yi; + } + } + } + break; + case Natural: + for (int xi = xr - 1; xi <= xr + 1; xi++) + { + for (int yi = yr - 1; yi <= yr + 1; yi++) + { + unsigned char lutPos = Index2D_256(0, xi, yi); + + FN_DECIMAL vecX = xi - x + CELL_2D_X[lutPos] * m_cellularJitter; + FN_DECIMAL vecY = yi - y + CELL_2D_Y[lutPos] * m_cellularJitter; + + FN_DECIMAL newDistance = (FastAbs(vecX) + FastAbs(vecY)) + (vecX * vecX + vecY * vecY); + + if (newDistance < distance) + { + distance = newDistance; + xc = xi; + yc = yi; + } + } + } + break; + } + + unsigned char lutPos; + switch (m_cellularReturnType) + { + case CellValue: + return ValCoord2D(m_seed, xc, yc); + + case NoiseLookup: + assert(m_cellularNoiseLookup); + + lutPos = Index2D_256(0, xc, yc); + return m_cellularNoiseLookup->GetNoise(xc + CELL_2D_X[lutPos] * m_cellularJitter, yc + CELL_2D_Y[lutPos] * m_cellularJitter); + + case Distance: + return distance; + default: + return 0; + } +} + +FN_DECIMAL FastNoise::SingleCellular2Edge(FN_DECIMAL x, FN_DECIMAL y) const +{ + int xr = FastRound(x); + int yr = FastRound(y); + + FN_DECIMAL distance[FN_CELLULAR_INDEX_MAX + 1] = { 999999,999999,999999,999999 }; + + switch (m_cellularDistanceFunction) + { + default: + case Euclidean: + for (int xi = xr - 1; xi <= xr + 1; xi++) + { + for (int yi = yr - 1; yi <= yr + 1; yi++) + { + unsigned char lutPos = Index2D_256(0, xi, yi); + + FN_DECIMAL vecX = xi - x + CELL_2D_X[lutPos] * m_cellularJitter; + FN_DECIMAL vecY = yi - y + CELL_2D_Y[lutPos] * m_cellularJitter; + + FN_DECIMAL newDistance = vecX * vecX + vecY * vecY; + + for (int i = m_cellularDistanceIndex1; i > 0; i--) + distance[i] = fmax(fmin(distance[i], newDistance), distance[i - 1]); + distance[0] = fmin(distance[0], newDistance); + } + } + break; + case Manhattan: + for (int xi = xr - 1; xi <= xr + 1; xi++) + { + for (int yi = yr - 1; yi <= yr + 1; yi++) + { + unsigned char lutPos = Index2D_256(0, xi, yi); + + FN_DECIMAL vecX = xi - x + CELL_2D_X[lutPos] * m_cellularJitter; + FN_DECIMAL vecY = yi - y + CELL_2D_Y[lutPos] * m_cellularJitter; + + FN_DECIMAL newDistance = FastAbs(vecX) + FastAbs(vecY); + + for (int i = m_cellularDistanceIndex1; i > 0; i--) + distance[i] = fmax(fmin(distance[i], newDistance), distance[i - 1]); + distance[0] = fmin(distance[0], newDistance); + } + } + break; + case Natural: + for (int xi = xr - 1; xi <= xr + 1; xi++) + { + for (int yi = yr - 1; yi <= yr + 1; yi++) + { + unsigned char lutPos = Index2D_256(0, xi, yi); + + FN_DECIMAL vecX = xi - x + CELL_2D_X[lutPos] * m_cellularJitter; + FN_DECIMAL vecY = yi - y + CELL_2D_Y[lutPos] * m_cellularJitter; + + FN_DECIMAL newDistance = (FastAbs(vecX) + FastAbs(vecY)) + (vecX * vecX + vecY * vecY); + + for (int i = m_cellularDistanceIndex1; i > 0; i--) + distance[i] = fmax(fmin(distance[i], newDistance), distance[i - 1]); + distance[0] = fmin(distance[0], newDistance); + } + } + break; + } + + switch (m_cellularReturnType) + { + case Distance2: + return distance[m_cellularDistanceIndex1]; + case Distance2Add: + return distance[m_cellularDistanceIndex1] + distance[m_cellularDistanceIndex0]; + case Distance2Sub: + return distance[m_cellularDistanceIndex1] - distance[m_cellularDistanceIndex0]; + case Distance2Mul: + return distance[m_cellularDistanceIndex1] * distance[m_cellularDistanceIndex0]; + case Distance2Div: + return distance[m_cellularDistanceIndex0] / distance[m_cellularDistanceIndex1]; + default: + return 0; + } +} + +void FastNoise::GradientPerturb(FN_DECIMAL& x, FN_DECIMAL& y, FN_DECIMAL& z) const +{ + SingleGradientPerturb(0, m_gradientPerturbAmp, m_frequency, x, y, z); +} + +void FastNoise::GradientPerturbFractal(FN_DECIMAL& x, FN_DECIMAL& y, FN_DECIMAL& z) const +{ + FN_DECIMAL amp = m_gradientPerturbAmp * m_fractalBounding; + FN_DECIMAL freq = m_frequency; + int i = 0; + + SingleGradientPerturb(m_perm[0], amp, m_frequency, x, y, z); + + while (++i < m_octaves) + { + freq *= m_lacunarity; + amp *= m_gain; + SingleGradientPerturb(m_perm[i], amp, freq, x, y, z); + } +} + +void FastNoise::SingleGradientPerturb(unsigned char offset, FN_DECIMAL warpAmp, FN_DECIMAL frequency, FN_DECIMAL& x, FN_DECIMAL& y, FN_DECIMAL& z) const +{ + FN_DECIMAL xf = x * frequency; + FN_DECIMAL yf = y * frequency; + FN_DECIMAL zf = z * frequency; + + int x0 = FastFloor(xf); + int y0 = FastFloor(yf); + int z0 = FastFloor(zf); + int x1 = x0 + 1; + int y1 = y0 + 1; + int z1 = z0 + 1; + + FN_DECIMAL xs, ys, zs; + switch (m_interp) + { + default: + case Linear: + xs = xf - (FN_DECIMAL)x0; + ys = yf - (FN_DECIMAL)y0; + zs = zf - (FN_DECIMAL)z0; + break; + case Hermite: + xs = InterpHermiteFunc(xf - (FN_DECIMAL)x0); + ys = InterpHermiteFunc(yf - (FN_DECIMAL)y0); + zs = InterpHermiteFunc(zf - (FN_DECIMAL)z0); + break; + case Quintic: + xs = InterpQuinticFunc(xf - (FN_DECIMAL)x0); + ys = InterpQuinticFunc(yf - (FN_DECIMAL)y0); + zs = InterpQuinticFunc(zf - (FN_DECIMAL)z0); + break; + } + + int lutPos0 = Index3D_256(offset, x0, y0, z0); + int lutPos1 = Index3D_256(offset, x1, y0, z0); + + FN_DECIMAL lx0x = Lerp(CELL_3D_X[lutPos0], CELL_3D_X[lutPos1], xs); + FN_DECIMAL ly0x = Lerp(CELL_3D_Y[lutPos0], CELL_3D_Y[lutPos1], xs); + FN_DECIMAL lz0x = Lerp(CELL_3D_Z[lutPos0], CELL_3D_Z[lutPos1], xs); + + lutPos0 = Index3D_256(offset, x0, y1, z0); + lutPos1 = Index3D_256(offset, x1, y1, z0); + + FN_DECIMAL lx1x = Lerp(CELL_3D_X[lutPos0], CELL_3D_X[lutPos1], xs); + FN_DECIMAL ly1x = Lerp(CELL_3D_Y[lutPos0], CELL_3D_Y[lutPos1], xs); + FN_DECIMAL lz1x = Lerp(CELL_3D_Z[lutPos0], CELL_3D_Z[lutPos1], xs); + + FN_DECIMAL lx0y = Lerp(lx0x, lx1x, ys); + FN_DECIMAL ly0y = Lerp(ly0x, ly1x, ys); + FN_DECIMAL lz0y = Lerp(lz0x, lz1x, ys); + + lutPos0 = Index3D_256(offset, x0, y0, z1); + lutPos1 = Index3D_256(offset, x1, y0, z1); + + lx0x = Lerp(CELL_3D_X[lutPos0], CELL_3D_X[lutPos1], xs); + ly0x = Lerp(CELL_3D_Y[lutPos0], CELL_3D_Y[lutPos1], xs); + lz0x = Lerp(CELL_3D_Z[lutPos0], CELL_3D_Z[lutPos1], xs); + + lutPos0 = Index3D_256(offset, x0, y1, z1); + lutPos1 = Index3D_256(offset, x1, y1, z1); + + lx1x = Lerp(CELL_3D_X[lutPos0], CELL_3D_X[lutPos1], xs); + ly1x = Lerp(CELL_3D_Y[lutPos0], CELL_3D_Y[lutPos1], xs); + lz1x = Lerp(CELL_3D_Z[lutPos0], CELL_3D_Z[lutPos1], xs); + + x += Lerp(lx0y, Lerp(lx0x, lx1x, ys), zs) * warpAmp; + y += Lerp(ly0y, Lerp(ly0x, ly1x, ys), zs) * warpAmp; + z += Lerp(lz0y, Lerp(lz0x, lz1x, ys), zs) * warpAmp; +} + +void FastNoise::GradientPerturb(FN_DECIMAL& x, FN_DECIMAL& y) const +{ + SingleGradientPerturb(0, m_gradientPerturbAmp, m_frequency, x, y); +} + +void FastNoise::GradientPerturbFractal(FN_DECIMAL& x, FN_DECIMAL& y) const +{ + FN_DECIMAL amp = m_gradientPerturbAmp * m_fractalBounding; + FN_DECIMAL freq = m_frequency; + int i = 0; + + SingleGradientPerturb(m_perm[0], amp, m_frequency, x, y); + + while (++i < m_octaves) + { + freq *= m_lacunarity; + amp *= m_gain; + SingleGradientPerturb(m_perm[i], amp, freq, x, y); + } +} + +void FastNoise::SingleGradientPerturb(unsigned char offset, FN_DECIMAL warpAmp, FN_DECIMAL frequency, FN_DECIMAL& x, FN_DECIMAL& y) const +{ + FN_DECIMAL xf = x * frequency; + FN_DECIMAL yf = y * frequency; + + int x0 = FastFloor(xf); + int y0 = FastFloor(yf); + int x1 = x0 + 1; + int y1 = y0 + 1; + + FN_DECIMAL xs, ys; + switch (m_interp) + { + default: + case Linear: + xs = xf - (FN_DECIMAL)x0; + ys = yf - (FN_DECIMAL)y0; + break; + case Hermite: + xs = InterpHermiteFunc(xf - (FN_DECIMAL)x0); + ys = InterpHermiteFunc(yf - (FN_DECIMAL)y0); + break; + case Quintic: + xs = InterpQuinticFunc(xf - (FN_DECIMAL)x0); + ys = InterpQuinticFunc(yf - (FN_DECIMAL)y0); + break; + } + + int lutPos0 = Index2D_256(offset, x0, y0); + int lutPos1 = Index2D_256(offset, x1, y0); + + FN_DECIMAL lx0x = Lerp(CELL_2D_X[lutPos0], CELL_2D_X[lutPos1], xs); + FN_DECIMAL ly0x = Lerp(CELL_2D_Y[lutPos0], CELL_2D_Y[lutPos1], xs); + + lutPos0 = Index2D_256(offset, x0, y1); + lutPos1 = Index2D_256(offset, x1, y1); + + FN_DECIMAL lx1x = Lerp(CELL_2D_X[lutPos0], CELL_2D_X[lutPos1], xs); + FN_DECIMAL ly1x = Lerp(CELL_2D_Y[lutPos0], CELL_2D_Y[lutPos1], xs); + + x += Lerp(lx0x, lx1x, ys) * warpAmp; + y += Lerp(ly0x, ly1x, ys) * warpAmp; +} diff --git a/src/ThirdParty/fastnoise.hpp b/src/ThirdParty/fastnoise.hpp new file mode 100644 index 0000000..f6c8d21 --- /dev/null +++ b/src/ThirdParty/fastnoise.hpp @@ -0,0 +1,311 @@ +// FastNoise.h +// +// MIT License +// +// Copyright(c) 2017 Jordan Peck +// +// Permission is hereby granted, free of charge, to any person obtaining a copy +// of this software and associated documentation files(the "Software"), to deal +// in the Software without restriction, including without limitation the rights +// to use, copy, modify, merge, publish, distribute, sublicense, and / or sell +// copies of the Software, and to permit persons to whom the Software is +// furnished to do so, subject to the following conditions : +// +// The above copyright notice and this permission notice shall be included in all +// copies or substantial portions of the Software. +// +// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.IN NO EVENT SHALL THE +// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +// SOFTWARE. +// +// The developer's email is jorzixdan.me2@gzixmail.com (for great email, take +// off every 'zix'.) +// + +// VERSION: 0.4.1 + +#ifndef FASTNOISE_H +#define FASTNOISE_H + +// Uncomment the line below to use doubles throughout FastNoise instead of floats +//#define FN_USE_DOUBLES + +#define FN_CELLULAR_INDEX_MAX 3 + +#ifdef FN_USE_DOUBLES +typedef double FN_DECIMAL; +#else +typedef float FN_DECIMAL; +#endif + +class FastNoise +{ +public: + explicit FastNoise(int seed = 1337) { SetSeed(seed); CalculateFractalBounding(); } + + enum NoiseType { Value, ValueFractal, Perlin, PerlinFractal, Simplex, SimplexFractal, Cellular, WhiteNoise, Cubic, CubicFractal }; + enum Interp { Linear, Hermite, Quintic }; + enum FractalType { FBM, Billow, RigidMulti }; + enum CellularDistanceFunction { Euclidean, Manhattan, Natural }; + enum CellularReturnType { CellValue, NoiseLookup, Distance, Distance2, Distance2Add, Distance2Sub, Distance2Mul, Distance2Div }; + + // Sets seed used for all noise types + // Default: 1337 + void SetSeed(int seed); + + // Returns seed used for all noise types + int GetSeed() const { return m_seed; } + + // Sets frequency for all noise types + // Default: 0.01 + void SetFrequency(FN_DECIMAL frequency) { m_frequency = frequency; } + + // Returns frequency used for all noise types + FN_DECIMAL GetFrequency() const { return m_frequency; } + + // Changes the interpolation method used to smooth between noise values + // Possible interpolation methods (lowest to highest quality) : + // - Linear + // - Hermite + // - Quintic + // Used in Value, Perlin Noise and Position Warping + // Default: Quintic + void SetInterp(Interp interp) { m_interp = interp; } + + // Returns interpolation method used for supported noise types + Interp GetInterp() const { return m_interp; } + + // Sets noise return type of GetNoise(...) + // Default: Simplex + void SetNoiseType(NoiseType noiseType) { m_noiseType = noiseType; } + + // Returns the noise type used by GetNoise + NoiseType GetNoiseType() const { return m_noiseType; } + + // Sets octave count for all fractal noise types + // Default: 3 + void SetFractalOctaves(int octaves) { m_octaves = octaves; CalculateFractalBounding(); } + + // Returns octave count for all fractal noise types + int GetFractalOctaves() const { return m_octaves; } + + // Sets octave lacunarity for all fractal noise types + // Default: 2.0 + void SetFractalLacunarity(FN_DECIMAL lacunarity) { m_lacunarity = lacunarity; } + + // Returns octave lacunarity for all fractal noise types + FN_DECIMAL GetFractalLacunarity() const { return m_lacunarity; } + + // Sets octave gain for all fractal noise types + // Default: 0.5 + void SetFractalGain(FN_DECIMAL gain) { m_gain = gain; CalculateFractalBounding(); } + + // Returns octave gain for all fractal noise types + FN_DECIMAL GetFractalGain() const { return m_gain; } + + // Sets method for combining octaves in all fractal noise types + // Default: FBM + void SetFractalType(FractalType fractalType) { m_fractalType = fractalType; } + + // Returns method for combining octaves in all fractal noise types + FractalType GetFractalType() const { return m_fractalType; } + + + // Sets distance function used in cellular noise calculations + // Default: Euclidean + void SetCellularDistanceFunction(CellularDistanceFunction cellularDistanceFunction) { m_cellularDistanceFunction = cellularDistanceFunction; } + + // Returns the distance function used in cellular noise calculations + CellularDistanceFunction GetCellularDistanceFunction() const { return m_cellularDistanceFunction; } + + // Sets return type from cellular noise calculations + // Note: NoiseLookup requires another FastNoise object be set with SetCellularNoiseLookup() to function + // Default: CellValue + void SetCellularReturnType(CellularReturnType cellularReturnType) { m_cellularReturnType = cellularReturnType; } + + // Returns the return type from cellular noise calculations + CellularReturnType GetCellularReturnType() const { return m_cellularReturnType; } + + // Noise used to calculate a cell value if cellular return type is NoiseLookup + // The lookup value is acquired through GetNoise() so ensure you SetNoiseType() on the noise lookup, value, Perlin or simplex is recommended + void SetCellularNoiseLookup(FastNoise* noise) { m_cellularNoiseLookup = noise; } + + // Returns the noise used to calculate a cell value if the cellular return type is NoiseLookup + FastNoise* GetCellularNoiseLookup() const { return m_cellularNoiseLookup; } + + // Sets the 2 distance indices used for distance2 return types + // Default: 0, 1 + // Note: index0 should be lower than index1 + // Both indices must be >= 0, index1 must be < 4 + void SetCellularDistance2Indices(int cellularDistanceIndex0, int cellularDistanceIndex1); + + // Returns the 2 distance indices used for distance2 return types + void GetCellularDistance2Indices(int& cellularDistanceIndex0, int& cellularDistanceIndex1) const; + + // Sets the maximum distance a cellular point can move from its grid position + // Setting this high will make artifacts more common + // Default: 0.45 + void SetCellularJitter(FN_DECIMAL cellularJitter) { m_cellularJitter = cellularJitter; } + + // Returns the maximum distance a cellular point can move from its grid position + FN_DECIMAL GetCellularJitter() const { return m_cellularJitter; } + + // Sets the maximum warp distance from original location when using GradientPerturb{Fractal}(...) + // Default: 1.0 + void SetGradientPerturbAmp(FN_DECIMAL gradientPerturbAmp) { m_gradientPerturbAmp = gradientPerturbAmp; } + + // Returns the maximum warp distance from original location when using GradientPerturb{Fractal}(...) + FN_DECIMAL GetGradientPerturbAmp() const { return m_gradientPerturbAmp; } + + //2D + FN_DECIMAL GetValue(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL GetValueFractal(FN_DECIMAL x, FN_DECIMAL y) const; + + FN_DECIMAL GetPerlin(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL GetPerlinFractal(FN_DECIMAL x, FN_DECIMAL y) const; + + FN_DECIMAL GetSimplex(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL GetSimplexFractal(FN_DECIMAL x, FN_DECIMAL y) const; + + FN_DECIMAL GetCellular(FN_DECIMAL x, FN_DECIMAL y) const; + + FN_DECIMAL GetWhiteNoise(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL GetWhiteNoiseInt(int x, int y) const; + + FN_DECIMAL GetCubic(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL GetCubicFractal(FN_DECIMAL x, FN_DECIMAL y) const; + + FN_DECIMAL GetNoise(FN_DECIMAL x, FN_DECIMAL y) const; + + void GradientPerturb(FN_DECIMAL& x, FN_DECIMAL& y) const; + void GradientPerturbFractal(FN_DECIMAL& x, FN_DECIMAL& y) const; + + //3D + FN_DECIMAL GetValue(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL GetValueFractal(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + + FN_DECIMAL GetPerlin(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL GetPerlinFractal(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + + FN_DECIMAL GetSimplex(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL GetSimplexFractal(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + + FN_DECIMAL GetCellular(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + + FN_DECIMAL GetWhiteNoise(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL GetWhiteNoiseInt(int x, int y, int z) const; + + FN_DECIMAL GetCubic(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL GetCubicFractal(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + + FN_DECIMAL GetNoise(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + + void GradientPerturb(FN_DECIMAL& x, FN_DECIMAL& y, FN_DECIMAL& z) const; + void GradientPerturbFractal(FN_DECIMAL& x, FN_DECIMAL& y, FN_DECIMAL& z) const; + + //4D + FN_DECIMAL GetSimplex(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z, FN_DECIMAL w) const; + + FN_DECIMAL GetWhiteNoise(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z, FN_DECIMAL w) const; + FN_DECIMAL GetWhiteNoiseInt(int x, int y, int z, int w) const; + +private: + unsigned char m_perm[512]; + unsigned char m_perm12[512]; + + int m_seed = 1337; + FN_DECIMAL m_frequency = FN_DECIMAL(0.01); + Interp m_interp = Quintic; + NoiseType m_noiseType = Simplex; + + int m_octaves = 3; + FN_DECIMAL m_lacunarity = FN_DECIMAL(2); + FN_DECIMAL m_gain = FN_DECIMAL(0.5); + FractalType m_fractalType = FBM; + FN_DECIMAL m_fractalBounding; + + CellularDistanceFunction m_cellularDistanceFunction = Euclidean; + CellularReturnType m_cellularReturnType = CellValue; + FastNoise* m_cellularNoiseLookup = nullptr; + int m_cellularDistanceIndex0 = 0; + int m_cellularDistanceIndex1 = 1; + FN_DECIMAL m_cellularJitter = FN_DECIMAL(0.45); + + FN_DECIMAL m_gradientPerturbAmp = FN_DECIMAL(1); + + void CalculateFractalBounding(); + + //2D + FN_DECIMAL SingleValueFractalFBM(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL SingleValueFractalBillow(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL SingleValueFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL SingleValue(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y) const; + + FN_DECIMAL SinglePerlinFractalFBM(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL SinglePerlinFractalBillow(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL SinglePerlinFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL SinglePerlin(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y) const; + + FN_DECIMAL SingleSimplexFractalFBM(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL SingleSimplexFractalBillow(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL SingleSimplexFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL SingleSimplexFractalBlend(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL SingleSimplex(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y) const; + + FN_DECIMAL SingleCubicFractalFBM(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL SingleCubicFractalBillow(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL SingleCubicFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL SingleCubic(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y) const; + + FN_DECIMAL SingleCellular(FN_DECIMAL x, FN_DECIMAL y) const; + FN_DECIMAL SingleCellular2Edge(FN_DECIMAL x, FN_DECIMAL y) const; + + void SingleGradientPerturb(unsigned char offset, FN_DECIMAL warpAmp, FN_DECIMAL frequency, FN_DECIMAL& x, FN_DECIMAL& y) const; + + //3D + FN_DECIMAL SingleValueFractalFBM(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL SingleValueFractalBillow(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL SingleValueFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL SingleValue(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + + FN_DECIMAL SinglePerlinFractalFBM(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL SinglePerlinFractalBillow(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL SinglePerlinFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL SinglePerlin(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + + FN_DECIMAL SingleSimplexFractalFBM(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL SingleSimplexFractalBillow(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL SingleSimplexFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL SingleSimplex(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + + FN_DECIMAL SingleCubicFractalFBM(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL SingleCubicFractalBillow(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL SingleCubicFractalRigidMulti(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL SingleCubic(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + + FN_DECIMAL SingleCellular(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + FN_DECIMAL SingleCellular2Edge(FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z) const; + + void SingleGradientPerturb(unsigned char offset, FN_DECIMAL warpAmp, FN_DECIMAL frequency, FN_DECIMAL& x, FN_DECIMAL& y, FN_DECIMAL& z) const; + + //4D + FN_DECIMAL SingleSimplex(unsigned char offset, FN_DECIMAL x, FN_DECIMAL y, FN_DECIMAL z, FN_DECIMAL w) const; + + inline unsigned char Index2D_12(unsigned char offset, int x, int y) const; + inline unsigned char Index3D_12(unsigned char offset, int x, int y, int z) const; + inline unsigned char Index4D_32(unsigned char offset, int x, int y, int z, int w) const; + inline unsigned char Index2D_256(unsigned char offset, int x, int y) const; + inline unsigned char Index3D_256(unsigned char offset, int x, int y, int z) const; + inline unsigned char Index4D_256(unsigned char offset, int x, int y, int z, int w) const; + + inline FN_DECIMAL ValCoord2DFast(unsigned char offset, int x, int y) const; + inline FN_DECIMAL ValCoord3DFast(unsigned char offset, int x, int y, int z) const; + inline FN_DECIMAL GradCoord2D(unsigned char offset, int x, int y, FN_DECIMAL xd, FN_DECIMAL yd) const; + inline FN_DECIMAL GradCoord3D(unsigned char offset, int x, int y, int z, FN_DECIMAL xd, FN_DECIMAL yd, FN_DECIMAL zd) const; + inline FN_DECIMAL GradCoord4D(unsigned char offset, int x, int y, int z, int w, FN_DECIMAL xd, FN_DECIMAL yd, FN_DECIMAL zd, FN_DECIMAL wd) const; +}; +#endif diff --git a/src/ThirdParty/glad.c b/src/ThirdParty/glad.c new file mode 100644 index 0000000..5e04eda --- /dev/null +++ b/src/ThirdParty/glad.c @@ -0,0 +1,1843 @@ +/* + + OpenGL loader generated by glad 0.1.28 on Tue Jan 15 14:49:19 2019. + + Language/Generator: C/C++ + Specification: gl + APIs: gl=3.3 + Profile: compatibility + Extensions: + + Loader: True + Local files: False + Omit khrplatform: False + Reproducible: False + + Commandline: + --profile="compatibility" --api="gl=3.3" --generator="c" --spec="gl" --extensions="" + Online: + https://glad.dav1d.de/#profile=compatibility&language=c&specification=gl&loader=on&api=gl%3D3.3 +*/ + +#include +#include +#include +#include + +static void* get_proc(const char* namez); + +#if defined(_WIN32) || defined(__CYGWIN__) +#include +static HMODULE libGL; + +typedef void* (APIENTRYP PFNWGLGETPROCADDRESSPROC_PRIVATE)(const char*); +static PFNWGLGETPROCADDRESSPROC_PRIVATE gladGetProcAddressPtr; + +#ifdef _MSC_VER +#ifdef __has_include +#if __has_include() +#define HAVE_WINAPIFAMILY 1 +#endif +#elif _MSC_VER >= 1700 && !_USING_V110_SDK71_ +#define HAVE_WINAPIFAMILY 1 +#endif +#endif + +#ifdef HAVE_WINAPIFAMILY +#include +#if !WINAPI_FAMILY_PARTITION(WINAPI_PARTITION_DESKTOP) && WINAPI_FAMILY_PARTITION(WINAPI_PARTITION_APP) +#define IS_UWP 1 +#endif +#endif + +static +int open_gl(void) { +#ifndef IS_UWP + libGL = LoadLibraryW(L"opengl32.dll"); + if (libGL != NULL) { + void (*tmp)(void); + tmp = (void(*)(void)) GetProcAddress(libGL, "wglGetProcAddress"); + gladGetProcAddressPtr = (PFNWGLGETPROCADDRESSPROC_PRIVATE)tmp; + return gladGetProcAddressPtr != NULL; + } +#endif + + return 0; +} + +static +void close_gl(void) { + if (libGL != NULL) { + FreeLibrary((HMODULE)libGL); + libGL = NULL; + } +} +#else +#include +static void* libGL; + +#if !defined(__APPLE__) && !defined(__HAIKU__) +typedef void* (APIENTRYP PFNGLXGETPROCADDRESSPROC_PRIVATE)(const char*); +static PFNGLXGETPROCADDRESSPROC_PRIVATE gladGetProcAddressPtr; +#endif + +static +int open_gl(void) { +#ifdef __APPLE__ + static const char* NAMES[] = { + "../Frameworks/OpenGL.framework/OpenGL", + "/Library/Frameworks/OpenGL.framework/OpenGL", + "/System/Library/Frameworks/OpenGL.framework/OpenGL", + "/System/Library/Frameworks/OpenGL.framework/Versions/Current/OpenGL" + }; +#else + static const char* NAMES[] = { "libGL.so.1", "libGL.so" }; +#endif + + unsigned int index = 0; + for (index = 0; index < (sizeof(NAMES) / sizeof(NAMES[0])); index++) { + libGL = dlopen(NAMES[index], RTLD_NOW | RTLD_GLOBAL); + + if (libGL != NULL) { +#if defined(__APPLE__) || defined(__HAIKU__) + return 1; +#else + gladGetProcAddressPtr = (PFNGLXGETPROCADDRESSPROC_PRIVATE)dlsym(libGL, + "glXGetProcAddressARB"); + return gladGetProcAddressPtr != NULL; +#endif + } + } + + return 0; +} + +static +void close_gl(void) { + if (libGL != NULL) { + dlclose(libGL); + libGL = NULL; + } +} +#endif + +static +void* get_proc(const char* namez) { + void* result = NULL; + if (libGL == NULL) return NULL; + +#if !defined(__APPLE__) && !defined(__HAIKU__) + if (gladGetProcAddressPtr != NULL) { + result = gladGetProcAddressPtr(namez); + } +#endif + if (result == NULL) { +#if defined(_WIN32) || defined(__CYGWIN__) + result = (void*)GetProcAddress((HMODULE)libGL, namez); +#else + result = dlsym(libGL, namez); +#endif + } + + return result; +} + +int gladLoadGL(void) { + int status = 0; + + if (open_gl()) { + status = gladLoadGLLoader(&get_proc); + close_gl(); + } + + return status; +} + +struct gladGLversionStruct GLVersion = { 0, 0 }; + +#if defined(GL_ES_VERSION_3_0) || defined(GL_VERSION_3_0) +#define _GLAD_IS_SOME_NEW_VERSION 1 +#endif + +static int max_loaded_major; +static int max_loaded_minor; + +static const char* exts = NULL; +static int num_exts_i = 0; +static char** exts_i = NULL; + +static int get_exts(void) { +#ifdef _GLAD_IS_SOME_NEW_VERSION + if (max_loaded_major < 3) { +#endif + exts = (const char*)glGetString(GL_EXTENSIONS); +#ifdef _GLAD_IS_SOME_NEW_VERSION + } + else { + unsigned int index; + + num_exts_i = 0; + glGetIntegerv(GL_NUM_EXTENSIONS, &num_exts_i); + if (num_exts_i > 0) { + char** tmp_exts_i = (char**)realloc((void*)exts_i, (size_t)num_exts_i * (sizeof * exts_i)); + if (tmp_exts_i == NULL) { + return 0; + } + exts_i = tmp_exts_i; + } + + if (exts_i == NULL) { + return 0; + } + + for (index = 0; index < (unsigned)num_exts_i; index++) { + const char* gl_str_tmp = (const char*)glGetStringi(GL_EXTENSIONS, index); + size_t len = strlen(gl_str_tmp); + + char* local_str = (char*)malloc((len + 1) * sizeof(char)); + if (local_str != NULL) { + memcpy(local_str, gl_str_tmp, (len + 1) * sizeof(char)); + } + exts_i[index] = local_str; + } + } +#endif + return 1; +} + +static void free_exts(void) { + if (exts_i != NULL) { + int index; + for (index = 0; index < num_exts_i; index++) { + free((char*)exts_i[index]); + } + free((void*)exts_i); + exts_i = NULL; + } +} + +static int has_ext(const char* ext) { +#ifdef _GLAD_IS_SOME_NEW_VERSION + if (max_loaded_major < 3) { +#endif + const char* extensions; + const char* loc; + const char* terminator; + extensions = exts; + if (extensions == NULL || ext == NULL) { + return 0; + } + + while (1) { + loc = strstr(extensions, ext); + if (loc == NULL) { + return 0; + } + + terminator = loc + strlen(ext); + if ((loc == extensions || *(loc - 1) == ' ') && + (*terminator == ' ' || *terminator == '\0')) { + return 1; + } + extensions = terminator; + } +#ifdef _GLAD_IS_SOME_NEW_VERSION + } + else { + int index; + if (exts_i == NULL) return 0; + for (index = 0; index < num_exts_i; index++) { + const char* e = exts_i[index]; + + if (exts_i[index] != NULL && strcmp(e, ext) == 0) { + return 1; + } + } + } +#endif + + return 0; +} +int GLAD_GL_VERSION_1_0 = 0; +int GLAD_GL_VERSION_1_1 = 0; +int GLAD_GL_VERSION_1_2 = 0; +int GLAD_GL_VERSION_1_3 = 0; +int GLAD_GL_VERSION_1_4 = 0; +int GLAD_GL_VERSION_1_5 = 0; +int GLAD_GL_VERSION_2_0 = 0; +int GLAD_GL_VERSION_2_1 = 0; +int GLAD_GL_VERSION_3_0 = 0; +int GLAD_GL_VERSION_3_1 = 0; +int GLAD_GL_VERSION_3_2 = 0; +int GLAD_GL_VERSION_3_3 = 0; +PFNGLACCUMPROC glad_glAccum = NULL; +PFNGLACTIVETEXTUREPROC glad_glActiveTexture = NULL; +PFNGLALPHAFUNCPROC glad_glAlphaFunc = NULL; +PFNGLARETEXTURESRESIDENTPROC glad_glAreTexturesResident = NULL; +PFNGLARRAYELEMENTPROC glad_glArrayElement = NULL; +PFNGLATTACHSHADERPROC glad_glAttachShader = NULL; +PFNGLBEGINPROC glad_glBegin = NULL; +PFNGLBEGINCONDITIONALRENDERPROC glad_glBeginConditionalRender = NULL; +PFNGLBEGINQUERYPROC glad_glBeginQuery = NULL; +PFNGLBEGINTRANSFORMFEEDBACKPROC glad_glBeginTransformFeedback = NULL; +PFNGLBINDATTRIBLOCATIONPROC glad_glBindAttribLocation = NULL; +PFNGLBINDBUFFERPROC glad_glBindBuffer = NULL; +PFNGLBINDBUFFERBASEPROC glad_glBindBufferBase = NULL; +PFNGLBINDBUFFERRANGEPROC glad_glBindBufferRange = NULL; +PFNGLBINDFRAGDATALOCATIONPROC glad_glBindFragDataLocation = NULL; +PFNGLBINDFRAGDATALOCATIONINDEXEDPROC glad_glBindFragDataLocationIndexed = NULL; +PFNGLBINDFRAMEBUFFERPROC glad_glBindFramebuffer = NULL; +PFNGLBINDRENDERBUFFERPROC glad_glBindRenderbuffer = NULL; +PFNGLBINDSAMPLERPROC glad_glBindSampler = NULL; +PFNGLBINDTEXTUREPROC glad_glBindTexture = NULL; +PFNGLBINDVERTEXARRAYPROC glad_glBindVertexArray = NULL; +PFNGLBITMAPPROC glad_glBitmap = NULL; +PFNGLBLENDCOLORPROC glad_glBlendColor = NULL; +PFNGLBLENDEQUATIONPROC glad_glBlendEquation = NULL; +PFNGLBLENDEQUATIONSEPARATEPROC glad_glBlendEquationSeparate = NULL; +PFNGLBLENDFUNCPROC glad_glBlendFunc = NULL; +PFNGLBLENDFUNCSEPARATEPROC glad_glBlendFuncSeparate = NULL; +PFNGLBLITFRAMEBUFFERPROC glad_glBlitFramebuffer = NULL; +PFNGLBUFFERDATAPROC glad_glBufferData = NULL; +PFNGLBUFFERSUBDATAPROC glad_glBufferSubData = NULL; +PFNGLCALLLISTPROC glad_glCallList = NULL; +PFNGLCALLLISTSPROC glad_glCallLists = NULL; +PFNGLCHECKFRAMEBUFFERSTATUSPROC glad_glCheckFramebufferStatus = NULL; +PFNGLCLAMPCOLORPROC glad_glClampColor = NULL; +PFNGLCLEARPROC glad_glClear = NULL; +PFNGLCLEARACCUMPROC glad_glClearAccum = NULL; +PFNGLCLEARBUFFERFIPROC glad_glClearBufferfi = NULL; +PFNGLCLEARBUFFERFVPROC glad_glClearBufferfv = NULL; +PFNGLCLEARBUFFERIVPROC glad_glClearBufferiv = NULL; +PFNGLCLEARBUFFERUIVPROC glad_glClearBufferuiv = NULL; +PFNGLCLEARCOLORPROC glad_glClearColor = NULL; +PFNGLCLEARDEPTHPROC glad_glClearDepth = NULL; +PFNGLCLEARINDEXPROC glad_glClearIndex = NULL; +PFNGLCLEARSTENCILPROC glad_glClearStencil = NULL; +PFNGLCLIENTACTIVETEXTUREPROC glad_glClientActiveTexture = NULL; +PFNGLCLIENTWAITSYNCPROC glad_glClientWaitSync = NULL; +PFNGLCLIPPLANEPROC glad_glClipPlane = NULL; +PFNGLCOLOR3BPROC glad_glColor3b = NULL; +PFNGLCOLOR3BVPROC glad_glColor3bv = NULL; +PFNGLCOLOR3DPROC glad_glColor3d = NULL; +PFNGLCOLOR3DVPROC glad_glColor3dv = NULL; +PFNGLCOLOR3FPROC glad_glColor3f = NULL; +PFNGLCOLOR3FVPROC glad_glColor3fv = NULL; +PFNGLCOLOR3IPROC glad_glColor3i = NULL; +PFNGLCOLOR3IVPROC glad_glColor3iv = NULL; +PFNGLCOLOR3SPROC glad_glColor3s = NULL; +PFNGLCOLOR3SVPROC glad_glColor3sv = NULL; +PFNGLCOLOR3UBPROC glad_glColor3ub = NULL; +PFNGLCOLOR3UBVPROC glad_glColor3ubv = NULL; +PFNGLCOLOR3UIPROC glad_glColor3ui = NULL; +PFNGLCOLOR3UIVPROC glad_glColor3uiv = NULL; +PFNGLCOLOR3USPROC glad_glColor3us = NULL; +PFNGLCOLOR3USVPROC glad_glColor3usv = NULL; +PFNGLCOLOR4BPROC glad_glColor4b = NULL; +PFNGLCOLOR4BVPROC glad_glColor4bv = NULL; +PFNGLCOLOR4DPROC glad_glColor4d = NULL; +PFNGLCOLOR4DVPROC glad_glColor4dv = NULL; +PFNGLCOLOR4FPROC glad_glColor4f = NULL; +PFNGLCOLOR4FVPROC glad_glColor4fv = NULL; +PFNGLCOLOR4IPROC glad_glColor4i = NULL; +PFNGLCOLOR4IVPROC glad_glColor4iv = NULL; +PFNGLCOLOR4SPROC glad_glColor4s = NULL; +PFNGLCOLOR4SVPROC glad_glColor4sv = NULL; +PFNGLCOLOR4UBPROC glad_glColor4ub = NULL; +PFNGLCOLOR4UBVPROC glad_glColor4ubv = NULL; +PFNGLCOLOR4UIPROC glad_glColor4ui = NULL; +PFNGLCOLOR4UIVPROC glad_glColor4uiv = NULL; +PFNGLCOLOR4USPROC glad_glColor4us = NULL; +PFNGLCOLOR4USVPROC glad_glColor4usv = NULL; +PFNGLCOLORMASKPROC glad_glColorMask = NULL; +PFNGLCOLORMASKIPROC glad_glColorMaski = NULL; +PFNGLCOLORMATERIALPROC glad_glColorMaterial = NULL; +PFNGLCOLORP3UIPROC glad_glColorP3ui = NULL; +PFNGLCOLORP3UIVPROC glad_glColorP3uiv = NULL; +PFNGLCOLORP4UIPROC glad_glColorP4ui = NULL; +PFNGLCOLORP4UIVPROC glad_glColorP4uiv = NULL; +PFNGLCOLORPOINTERPROC glad_glColorPointer = NULL; +PFNGLCOMPILESHADERPROC glad_glCompileShader = NULL; +PFNGLCOMPRESSEDTEXIMAGE1DPROC glad_glCompressedTexImage1D = NULL; +PFNGLCOMPRESSEDTEXIMAGE2DPROC glad_glCompressedTexImage2D = NULL; +PFNGLCOMPRESSEDTEXIMAGE3DPROC glad_glCompressedTexImage3D = NULL; +PFNGLCOMPRESSEDTEXSUBIMAGE1DPROC glad_glCompressedTexSubImage1D = NULL; +PFNGLCOMPRESSEDTEXSUBIMAGE2DPROC glad_glCompressedTexSubImage2D = NULL; +PFNGLCOMPRESSEDTEXSUBIMAGE3DPROC glad_glCompressedTexSubImage3D = NULL; +PFNGLCOPYBUFFERSUBDATAPROC glad_glCopyBufferSubData = NULL; +PFNGLCOPYPIXELSPROC glad_glCopyPixels = NULL; +PFNGLCOPYTEXIMAGE1DPROC glad_glCopyTexImage1D = NULL; +PFNGLCOPYTEXIMAGE2DPROC glad_glCopyTexImage2D = NULL; +PFNGLCOPYTEXSUBIMAGE1DPROC glad_glCopyTexSubImage1D = NULL; +PFNGLCOPYTEXSUBIMAGE2DPROC glad_glCopyTexSubImage2D = NULL; +PFNGLCOPYTEXSUBIMAGE3DPROC glad_glCopyTexSubImage3D = NULL; +PFNGLCREATEPROGRAMPROC glad_glCreateProgram = NULL; +PFNGLCREATESHADERPROC glad_glCreateShader = NULL; +PFNGLCULLFACEPROC glad_glCullFace = NULL; +PFNGLDELETEBUFFERSPROC glad_glDeleteBuffers = NULL; +PFNGLDELETEFRAMEBUFFERSPROC glad_glDeleteFramebuffers = NULL; +PFNGLDELETELISTSPROC glad_glDeleteLists = NULL; +PFNGLDELETEPROGRAMPROC glad_glDeleteProgram = NULL; +PFNGLDELETEQUERIESPROC glad_glDeleteQueries = NULL; +PFNGLDELETERENDERBUFFERSPROC glad_glDeleteRenderbuffers = NULL; +PFNGLDELETESAMPLERSPROC glad_glDeleteSamplers = NULL; +PFNGLDELETESHADERPROC glad_glDeleteShader = NULL; +PFNGLDELETESYNCPROC glad_glDeleteSync = NULL; +PFNGLDELETETEXTURESPROC glad_glDeleteTextures = NULL; +PFNGLDELETEVERTEXARRAYSPROC glad_glDeleteVertexArrays = NULL; +PFNGLDEPTHFUNCPROC glad_glDepthFunc = NULL; +PFNGLDEPTHMASKPROC glad_glDepthMask = NULL; +PFNGLDEPTHRANGEPROC glad_glDepthRange = NULL; +PFNGLDETACHSHADERPROC glad_glDetachShader = NULL; +PFNGLDISABLEPROC glad_glDisable = NULL; +PFNGLDISABLECLIENTSTATEPROC glad_glDisableClientState = NULL; +PFNGLDISABLEVERTEXATTRIBARRAYPROC glad_glDisableVertexAttribArray = NULL; +PFNGLDISABLEIPROC glad_glDisablei = NULL; +PFNGLDRAWARRAYSPROC glad_glDrawArrays = NULL; +PFNGLDRAWARRAYSINSTANCEDPROC glad_glDrawArraysInstanced = NULL; +PFNGLDRAWBUFFERPROC glad_glDrawBuffer = NULL; +PFNGLDRAWBUFFERSPROC glad_glDrawBuffers = NULL; +PFNGLDRAWELEMENTSPROC glad_glDrawElements = NULL; +PFNGLDRAWELEMENTSBASEVERTEXPROC glad_glDrawElementsBaseVertex = NULL; +PFNGLDRAWELEMENTSINSTANCEDPROC glad_glDrawElementsInstanced = NULL; +PFNGLDRAWELEMENTSINSTANCEDBASEVERTEXPROC glad_glDrawElementsInstancedBaseVertex = NULL; +PFNGLDRAWPIXELSPROC glad_glDrawPixels = NULL; +PFNGLDRAWRANGEELEMENTSPROC glad_glDrawRangeElements = NULL; +PFNGLDRAWRANGEELEMENTSBASEVERTEXPROC glad_glDrawRangeElementsBaseVertex = NULL; +PFNGLEDGEFLAGPROC glad_glEdgeFlag = NULL; +PFNGLEDGEFLAGPOINTERPROC glad_glEdgeFlagPointer = NULL; +PFNGLEDGEFLAGVPROC glad_glEdgeFlagv = NULL; +PFNGLENABLEPROC glad_glEnable = NULL; +PFNGLENABLECLIENTSTATEPROC glad_glEnableClientState = NULL; +PFNGLENABLEVERTEXATTRIBARRAYPROC glad_glEnableVertexAttribArray = NULL; +PFNGLENABLEIPROC glad_glEnablei = NULL; +PFNGLENDPROC glad_glEnd = NULL; +PFNGLENDCONDITIONALRENDERPROC glad_glEndConditionalRender = NULL; +PFNGLENDLISTPROC glad_glEndList = NULL; +PFNGLENDQUERYPROC glad_glEndQuery = NULL; +PFNGLENDTRANSFORMFEEDBACKPROC glad_glEndTransformFeedback = NULL; +PFNGLEVALCOORD1DPROC glad_glEvalCoord1d = NULL; +PFNGLEVALCOORD1DVPROC glad_glEvalCoord1dv = NULL; +PFNGLEVALCOORD1FPROC glad_glEvalCoord1f = NULL; +PFNGLEVALCOORD1FVPROC glad_glEvalCoord1fv = NULL; +PFNGLEVALCOORD2DPROC glad_glEvalCoord2d = NULL; +PFNGLEVALCOORD2DVPROC glad_glEvalCoord2dv = NULL; +PFNGLEVALCOORD2FPROC glad_glEvalCoord2f = NULL; +PFNGLEVALCOORD2FVPROC glad_glEvalCoord2fv = NULL; +PFNGLEVALMESH1PROC glad_glEvalMesh1 = NULL; +PFNGLEVALMESH2PROC glad_glEvalMesh2 = NULL; +PFNGLEVALPOINT1PROC glad_glEvalPoint1 = NULL; +PFNGLEVALPOINT2PROC glad_glEvalPoint2 = NULL; +PFNGLFEEDBACKBUFFERPROC glad_glFeedbackBuffer = NULL; +PFNGLFENCESYNCPROC glad_glFenceSync = NULL; +PFNGLFINISHPROC glad_glFinish = NULL; +PFNGLFLUSHPROC glad_glFlush = NULL; +PFNGLFLUSHMAPPEDBUFFERRANGEPROC glad_glFlushMappedBufferRange = NULL; +PFNGLFOGCOORDPOINTERPROC glad_glFogCoordPointer = NULL; +PFNGLFOGCOORDDPROC glad_glFogCoordd = NULL; +PFNGLFOGCOORDDVPROC glad_glFogCoorddv = NULL; +PFNGLFOGCOORDFPROC glad_glFogCoordf = NULL; +PFNGLFOGCOORDFVPROC glad_glFogCoordfv = NULL; +PFNGLFOGFPROC glad_glFogf = NULL; +PFNGLFOGFVPROC glad_glFogfv = NULL; +PFNGLFOGIPROC glad_glFogi = NULL; +PFNGLFOGIVPROC glad_glFogiv = NULL; +PFNGLFRAMEBUFFERRENDERBUFFERPROC glad_glFramebufferRenderbuffer = NULL; +PFNGLFRAMEBUFFERTEXTUREPROC glad_glFramebufferTexture = NULL; +PFNGLFRAMEBUFFERTEXTURE1DPROC glad_glFramebufferTexture1D = NULL; +PFNGLFRAMEBUFFERTEXTURE2DPROC glad_glFramebufferTexture2D = NULL; +PFNGLFRAMEBUFFERTEXTURE3DPROC glad_glFramebufferTexture3D = NULL; +PFNGLFRAMEBUFFERTEXTURELAYERPROC glad_glFramebufferTextureLayer = NULL; +PFNGLFRONTFACEPROC glad_glFrontFace = NULL; +PFNGLFRUSTUMPROC glad_glFrustum = NULL; +PFNGLGENBUFFERSPROC glad_glGenBuffers = NULL; +PFNGLGENFRAMEBUFFERSPROC glad_glGenFramebuffers = NULL; +PFNGLGENLISTSPROC glad_glGenLists = NULL; +PFNGLGENQUERIESPROC glad_glGenQueries = NULL; +PFNGLGENRENDERBUFFERSPROC glad_glGenRenderbuffers = NULL; +PFNGLGENSAMPLERSPROC glad_glGenSamplers = NULL; +PFNGLGENTEXTURESPROC glad_glGenTextures = NULL; +PFNGLGENVERTEXARRAYSPROC glad_glGenVertexArrays = NULL; +PFNGLGENERATEMIPMAPPROC glad_glGenerateMipmap = NULL; +PFNGLGETACTIVEATTRIBPROC glad_glGetActiveAttrib = NULL; +PFNGLGETACTIVEUNIFORMPROC glad_glGetActiveUniform = NULL; +PFNGLGETACTIVEUNIFORMBLOCKNAMEPROC glad_glGetActiveUniformBlockName = NULL; +PFNGLGETACTIVEUNIFORMBLOCKIVPROC glad_glGetActiveUniformBlockiv = NULL; +PFNGLGETACTIVEUNIFORMNAMEPROC glad_glGetActiveUniformName = NULL; +PFNGLGETACTIVEUNIFORMSIVPROC glad_glGetActiveUniformsiv = NULL; +PFNGLGETATTACHEDSHADERSPROC glad_glGetAttachedShaders = NULL; +PFNGLGETATTRIBLOCATIONPROC glad_glGetAttribLocation = NULL; +PFNGLGETBOOLEANI_VPROC glad_glGetBooleani_v = NULL; +PFNGLGETBOOLEANVPROC glad_glGetBooleanv = NULL; +PFNGLGETBUFFERPARAMETERI64VPROC glad_glGetBufferParameteri64v = NULL; +PFNGLGETBUFFERPARAMETERIVPROC glad_glGetBufferParameteriv = NULL; +PFNGLGETBUFFERPOINTERVPROC glad_glGetBufferPointerv = NULL; +PFNGLGETBUFFERSUBDATAPROC glad_glGetBufferSubData = NULL; +PFNGLGETCLIPPLANEPROC glad_glGetClipPlane = NULL; +PFNGLGETCOMPRESSEDTEXIMAGEPROC glad_glGetCompressedTexImage = NULL; +PFNGLGETDOUBLEVPROC glad_glGetDoublev = NULL; +PFNGLGETERRORPROC glad_glGetError = NULL; +PFNGLGETFLOATVPROC glad_glGetFloatv = NULL; +PFNGLGETFRAGDATAINDEXPROC glad_glGetFragDataIndex = NULL; +PFNGLGETFRAGDATALOCATIONPROC glad_glGetFragDataLocation = NULL; +PFNGLGETFRAMEBUFFERATTACHMENTPARAMETERIVPROC glad_glGetFramebufferAttachmentParameteriv = NULL; +PFNGLGETINTEGER64I_VPROC glad_glGetInteger64i_v = NULL; +PFNGLGETINTEGER64VPROC glad_glGetInteger64v = NULL; +PFNGLGETINTEGERI_VPROC glad_glGetIntegeri_v = NULL; +PFNGLGETINTEGERVPROC glad_glGetIntegerv = NULL; +PFNGLGETLIGHTFVPROC glad_glGetLightfv = NULL; +PFNGLGETLIGHTIVPROC glad_glGetLightiv = NULL; +PFNGLGETMAPDVPROC glad_glGetMapdv = NULL; +PFNGLGETMAPFVPROC glad_glGetMapfv = NULL; +PFNGLGETMAPIVPROC glad_glGetMapiv = NULL; +PFNGLGETMATERIALFVPROC glad_glGetMaterialfv = NULL; +PFNGLGETMATERIALIVPROC glad_glGetMaterialiv = NULL; +PFNGLGETMULTISAMPLEFVPROC glad_glGetMultisamplefv = NULL; +PFNGLGETPIXELMAPFVPROC glad_glGetPixelMapfv = NULL; +PFNGLGETPIXELMAPUIVPROC glad_glGetPixelMapuiv = NULL; +PFNGLGETPIXELMAPUSVPROC glad_glGetPixelMapusv = NULL; +PFNGLGETPOINTERVPROC glad_glGetPointerv = NULL; +PFNGLGETPOLYGONSTIPPLEPROC glad_glGetPolygonStipple = NULL; +PFNGLGETPROGRAMINFOLOGPROC glad_glGetProgramInfoLog = NULL; +PFNGLGETPROGRAMIVPROC glad_glGetProgramiv = NULL; +PFNGLGETQUERYOBJECTI64VPROC glad_glGetQueryObjecti64v = NULL; +PFNGLGETQUERYOBJECTIVPROC glad_glGetQueryObjectiv = NULL; +PFNGLGETQUERYOBJECTUI64VPROC glad_glGetQueryObjectui64v = NULL; +PFNGLGETQUERYOBJECTUIVPROC glad_glGetQueryObjectuiv = NULL; +PFNGLGETQUERYIVPROC glad_glGetQueryiv = NULL; +PFNGLGETRENDERBUFFERPARAMETERIVPROC glad_glGetRenderbufferParameteriv = NULL; +PFNGLGETSAMPLERPARAMETERIIVPROC glad_glGetSamplerParameterIiv = NULL; +PFNGLGETSAMPLERPARAMETERIUIVPROC glad_glGetSamplerParameterIuiv = NULL; +PFNGLGETSAMPLERPARAMETERFVPROC glad_glGetSamplerParameterfv = NULL; +PFNGLGETSAMPLERPARAMETERIVPROC glad_glGetSamplerParameteriv = NULL; +PFNGLGETSHADERINFOLOGPROC glad_glGetShaderInfoLog = NULL; +PFNGLGETSHADERSOURCEPROC glad_glGetShaderSource = NULL; +PFNGLGETSHADERIVPROC glad_glGetShaderiv = NULL; +PFNGLGETSTRINGPROC glad_glGetString = NULL; +PFNGLGETSTRINGIPROC glad_glGetStringi = NULL; +PFNGLGETSYNCIVPROC glad_glGetSynciv = NULL; +PFNGLGETTEXENVFVPROC glad_glGetTexEnvfv = NULL; +PFNGLGETTEXENVIVPROC glad_glGetTexEnviv = NULL; +PFNGLGETTEXGENDVPROC glad_glGetTexGendv = NULL; +PFNGLGETTEXGENFVPROC glad_glGetTexGenfv = NULL; +PFNGLGETTEXGENIVPROC glad_glGetTexGeniv = NULL; +PFNGLGETTEXIMAGEPROC glad_glGetTexImage = NULL; +PFNGLGETTEXLEVELPARAMETERFVPROC glad_glGetTexLevelParameterfv = NULL; +PFNGLGETTEXLEVELPARAMETERIVPROC glad_glGetTexLevelParameteriv = NULL; +PFNGLGETTEXPARAMETERIIVPROC glad_glGetTexParameterIiv = NULL; +PFNGLGETTEXPARAMETERIUIVPROC glad_glGetTexParameterIuiv = NULL; +PFNGLGETTEXPARAMETERFVPROC glad_glGetTexParameterfv = NULL; +PFNGLGETTEXPARAMETERIVPROC glad_glGetTexParameteriv = NULL; +PFNGLGETTRANSFORMFEEDBACKVARYINGPROC glad_glGetTransformFeedbackVarying = NULL; +PFNGLGETUNIFORMBLOCKINDEXPROC glad_glGetUniformBlockIndex = NULL; +PFNGLGETUNIFORMINDICESPROC glad_glGetUniformIndices = NULL; +PFNGLGETUNIFORMLOCATIONPROC glad_glGetUniformLocation = NULL; +PFNGLGETUNIFORMFVPROC glad_glGetUniformfv = NULL; +PFNGLGETUNIFORMIVPROC glad_glGetUniformiv = NULL; +PFNGLGETUNIFORMUIVPROC glad_glGetUniformuiv = NULL; +PFNGLGETVERTEXATTRIBIIVPROC glad_glGetVertexAttribIiv = NULL; +PFNGLGETVERTEXATTRIBIUIVPROC glad_glGetVertexAttribIuiv = NULL; +PFNGLGETVERTEXATTRIBPOINTERVPROC glad_glGetVertexAttribPointerv = NULL; +PFNGLGETVERTEXATTRIBDVPROC glad_glGetVertexAttribdv = NULL; +PFNGLGETVERTEXATTRIBFVPROC glad_glGetVertexAttribfv = NULL; +PFNGLGETVERTEXATTRIBIVPROC glad_glGetVertexAttribiv = NULL; +PFNGLHINTPROC glad_glHint = NULL; +PFNGLINDEXMASKPROC glad_glIndexMask = NULL; +PFNGLINDEXPOINTERPROC glad_glIndexPointer = NULL; +PFNGLINDEXDPROC glad_glIndexd = NULL; +PFNGLINDEXDVPROC glad_glIndexdv = NULL; +PFNGLINDEXFPROC glad_glIndexf = NULL; +PFNGLINDEXFVPROC glad_glIndexfv = NULL; +PFNGLINDEXIPROC glad_glIndexi = NULL; +PFNGLINDEXIVPROC glad_glIndexiv = NULL; +PFNGLINDEXSPROC glad_glIndexs = NULL; +PFNGLINDEXSVPROC glad_glIndexsv = NULL; +PFNGLINDEXUBPROC glad_glIndexub = NULL; +PFNGLINDEXUBVPROC glad_glIndexubv = NULL; +PFNGLINITNAMESPROC glad_glInitNames = NULL; +PFNGLINTERLEAVEDARRAYSPROC glad_glInterleavedArrays = NULL; +PFNGLISBUFFERPROC glad_glIsBuffer = NULL; +PFNGLISENABLEDPROC glad_glIsEnabled = NULL; +PFNGLISENABLEDIPROC glad_glIsEnabledi = NULL; +PFNGLISFRAMEBUFFERPROC glad_glIsFramebuffer = NULL; +PFNGLISLISTPROC glad_glIsList = NULL; +PFNGLISPROGRAMPROC glad_glIsProgram = NULL; +PFNGLISQUERYPROC glad_glIsQuery = NULL; +PFNGLISRENDERBUFFERPROC glad_glIsRenderbuffer = NULL; +PFNGLISSAMPLERPROC glad_glIsSampler = NULL; +PFNGLISSHADERPROC glad_glIsShader = NULL; +PFNGLISSYNCPROC glad_glIsSync = NULL; +PFNGLISTEXTUREPROC glad_glIsTexture = NULL; +PFNGLISVERTEXARRAYPROC glad_glIsVertexArray = NULL; +PFNGLLIGHTMODELFPROC glad_glLightModelf = NULL; +PFNGLLIGHTMODELFVPROC glad_glLightModelfv = NULL; +PFNGLLIGHTMODELIPROC glad_glLightModeli = NULL; +PFNGLLIGHTMODELIVPROC glad_glLightModeliv = NULL; +PFNGLLIGHTFPROC glad_glLightf = NULL; +PFNGLLIGHTFVPROC glad_glLightfv = NULL; +PFNGLLIGHTIPROC glad_glLighti = NULL; +PFNGLLIGHTIVPROC glad_glLightiv = NULL; +PFNGLLINESTIPPLEPROC glad_glLineStipple = NULL; +PFNGLLINEWIDTHPROC glad_glLineWidth = NULL; +PFNGLLINKPROGRAMPROC glad_glLinkProgram = NULL; +PFNGLLISTBASEPROC glad_glListBase = NULL; +PFNGLLOADIDENTITYPROC glad_glLoadIdentity = NULL; +PFNGLLOADMATRIXDPROC glad_glLoadMatrixd = NULL; +PFNGLLOADMATRIXFPROC glad_glLoadMatrixf = NULL; +PFNGLLOADNAMEPROC glad_glLoadName = NULL; +PFNGLLOADTRANSPOSEMATRIXDPROC glad_glLoadTransposeMatrixd = NULL; +PFNGLLOADTRANSPOSEMATRIXFPROC glad_glLoadTransposeMatrixf = NULL; +PFNGLLOGICOPPROC glad_glLogicOp = NULL; +PFNGLMAP1DPROC glad_glMap1d = NULL; +PFNGLMAP1FPROC glad_glMap1f = NULL; +PFNGLMAP2DPROC glad_glMap2d = NULL; +PFNGLMAP2FPROC glad_glMap2f = NULL; +PFNGLMAPBUFFERPROC glad_glMapBuffer = NULL; +PFNGLMAPBUFFERRANGEPROC glad_glMapBufferRange = NULL; +PFNGLMAPGRID1DPROC glad_glMapGrid1d = NULL; +PFNGLMAPGRID1FPROC glad_glMapGrid1f = NULL; +PFNGLMAPGRID2DPROC glad_glMapGrid2d = NULL; +PFNGLMAPGRID2FPROC glad_glMapGrid2f = NULL; +PFNGLMATERIALFPROC glad_glMaterialf = NULL; +PFNGLMATERIALFVPROC glad_glMaterialfv = NULL; +PFNGLMATERIALIPROC glad_glMateriali = NULL; +PFNGLMATERIALIVPROC glad_glMaterialiv = NULL; +PFNGLMATRIXMODEPROC glad_glMatrixMode = NULL; +PFNGLMULTMATRIXDPROC glad_glMultMatrixd = NULL; +PFNGLMULTMATRIXFPROC glad_glMultMatrixf = NULL; +PFNGLMULTTRANSPOSEMATRIXDPROC glad_glMultTransposeMatrixd = NULL; +PFNGLMULTTRANSPOSEMATRIXFPROC glad_glMultTransposeMatrixf = NULL; +PFNGLMULTIDRAWARRAYSPROC glad_glMultiDrawArrays = NULL; +PFNGLMULTIDRAWELEMENTSPROC glad_glMultiDrawElements = NULL; +PFNGLMULTIDRAWELEMENTSBASEVERTEXPROC glad_glMultiDrawElementsBaseVertex = NULL; +PFNGLMULTITEXCOORD1DPROC glad_glMultiTexCoord1d = NULL; +PFNGLMULTITEXCOORD1DVPROC glad_glMultiTexCoord1dv = NULL; +PFNGLMULTITEXCOORD1FPROC glad_glMultiTexCoord1f = NULL; +PFNGLMULTITEXCOORD1FVPROC glad_glMultiTexCoord1fv = NULL; +PFNGLMULTITEXCOORD1IPROC glad_glMultiTexCoord1i = NULL; +PFNGLMULTITEXCOORD1IVPROC glad_glMultiTexCoord1iv = NULL; +PFNGLMULTITEXCOORD1SPROC glad_glMultiTexCoord1s = NULL; +PFNGLMULTITEXCOORD1SVPROC glad_glMultiTexCoord1sv = NULL; +PFNGLMULTITEXCOORD2DPROC glad_glMultiTexCoord2d = NULL; +PFNGLMULTITEXCOORD2DVPROC glad_glMultiTexCoord2dv = NULL; +PFNGLMULTITEXCOORD2FPROC glad_glMultiTexCoord2f = NULL; +PFNGLMULTITEXCOORD2FVPROC glad_glMultiTexCoord2fv = NULL; +PFNGLMULTITEXCOORD2IPROC glad_glMultiTexCoord2i = NULL; +PFNGLMULTITEXCOORD2IVPROC glad_glMultiTexCoord2iv = NULL; +PFNGLMULTITEXCOORD2SPROC glad_glMultiTexCoord2s = NULL; +PFNGLMULTITEXCOORD2SVPROC glad_glMultiTexCoord2sv = NULL; +PFNGLMULTITEXCOORD3DPROC glad_glMultiTexCoord3d = NULL; +PFNGLMULTITEXCOORD3DVPROC glad_glMultiTexCoord3dv = NULL; +PFNGLMULTITEXCOORD3FPROC glad_glMultiTexCoord3f = NULL; +PFNGLMULTITEXCOORD3FVPROC glad_glMultiTexCoord3fv = NULL; +PFNGLMULTITEXCOORD3IPROC glad_glMultiTexCoord3i = NULL; +PFNGLMULTITEXCOORD3IVPROC glad_glMultiTexCoord3iv = NULL; +PFNGLMULTITEXCOORD3SPROC glad_glMultiTexCoord3s = NULL; +PFNGLMULTITEXCOORD3SVPROC glad_glMultiTexCoord3sv = NULL; +PFNGLMULTITEXCOORD4DPROC glad_glMultiTexCoord4d = NULL; +PFNGLMULTITEXCOORD4DVPROC glad_glMultiTexCoord4dv = NULL; +PFNGLMULTITEXCOORD4FPROC glad_glMultiTexCoord4f = NULL; +PFNGLMULTITEXCOORD4FVPROC glad_glMultiTexCoord4fv = NULL; +PFNGLMULTITEXCOORD4IPROC glad_glMultiTexCoord4i = NULL; +PFNGLMULTITEXCOORD4IVPROC glad_glMultiTexCoord4iv = NULL; +PFNGLMULTITEXCOORD4SPROC glad_glMultiTexCoord4s = NULL; +PFNGLMULTITEXCOORD4SVPROC glad_glMultiTexCoord4sv = NULL; +PFNGLMULTITEXCOORDP1UIPROC glad_glMultiTexCoordP1ui = NULL; +PFNGLMULTITEXCOORDP1UIVPROC glad_glMultiTexCoordP1uiv = NULL; +PFNGLMULTITEXCOORDP2UIPROC glad_glMultiTexCoordP2ui = NULL; +PFNGLMULTITEXCOORDP2UIVPROC glad_glMultiTexCoordP2uiv = NULL; +PFNGLMULTITEXCOORDP3UIPROC glad_glMultiTexCoordP3ui = NULL; +PFNGLMULTITEXCOORDP3UIVPROC glad_glMultiTexCoordP3uiv = NULL; +PFNGLMULTITEXCOORDP4UIPROC glad_glMultiTexCoordP4ui = NULL; +PFNGLMULTITEXCOORDP4UIVPROC glad_glMultiTexCoordP4uiv = NULL; +PFNGLNEWLISTPROC glad_glNewList = NULL; +PFNGLNORMAL3BPROC glad_glNormal3b = NULL; +PFNGLNORMAL3BVPROC glad_glNormal3bv = NULL; +PFNGLNORMAL3DPROC glad_glNormal3d = NULL; +PFNGLNORMAL3DVPROC glad_glNormal3dv = NULL; +PFNGLNORMAL3FPROC glad_glNormal3f = NULL; +PFNGLNORMAL3FVPROC glad_glNormal3fv = NULL; +PFNGLNORMAL3IPROC glad_glNormal3i = NULL; +PFNGLNORMAL3IVPROC glad_glNormal3iv = NULL; +PFNGLNORMAL3SPROC glad_glNormal3s = NULL; +PFNGLNORMAL3SVPROC glad_glNormal3sv = NULL; +PFNGLNORMALP3UIPROC glad_glNormalP3ui = NULL; +PFNGLNORMALP3UIVPROC glad_glNormalP3uiv = NULL; +PFNGLNORMALPOINTERPROC glad_glNormalPointer = NULL; +PFNGLORTHOPROC glad_glOrtho = NULL; +PFNGLPASSTHROUGHPROC glad_glPassThrough = NULL; +PFNGLPIXELMAPFVPROC glad_glPixelMapfv = NULL; +PFNGLPIXELMAPUIVPROC glad_glPixelMapuiv = NULL; +PFNGLPIXELMAPUSVPROC glad_glPixelMapusv = NULL; +PFNGLPIXELSTOREFPROC glad_glPixelStoref = NULL; +PFNGLPIXELSTOREIPROC glad_glPixelStorei = NULL; +PFNGLPIXELTRANSFERFPROC glad_glPixelTransferf = NULL; +PFNGLPIXELTRANSFERIPROC glad_glPixelTransferi = NULL; +PFNGLPIXELZOOMPROC glad_glPixelZoom = NULL; +PFNGLPOINTPARAMETERFPROC glad_glPointParameterf = NULL; +PFNGLPOINTPARAMETERFVPROC glad_glPointParameterfv = NULL; +PFNGLPOINTPARAMETERIPROC glad_glPointParameteri = NULL; +PFNGLPOINTPARAMETERIVPROC glad_glPointParameteriv = NULL; +PFNGLPOINTSIZEPROC glad_glPointSize = NULL; +PFNGLPOLYGONMODEPROC glad_glPolygonMode = NULL; +PFNGLPOLYGONOFFSETPROC glad_glPolygonOffset = NULL; +PFNGLPOLYGONSTIPPLEPROC glad_glPolygonStipple = NULL; +PFNGLPOPATTRIBPROC glad_glPopAttrib = NULL; +PFNGLPOPCLIENTATTRIBPROC glad_glPopClientAttrib = NULL; +PFNGLPOPMATRIXPROC glad_glPopMatrix = NULL; +PFNGLPOPNAMEPROC glad_glPopName = NULL; +PFNGLPRIMITIVERESTARTINDEXPROC glad_glPrimitiveRestartIndex = NULL; +PFNGLPRIORITIZETEXTURESPROC glad_glPrioritizeTextures = NULL; +PFNGLPROVOKINGVERTEXPROC glad_glProvokingVertex = NULL; +PFNGLPUSHATTRIBPROC glad_glPushAttrib = NULL; +PFNGLPUSHCLIENTATTRIBPROC glad_glPushClientAttrib = NULL; +PFNGLPUSHMATRIXPROC glad_glPushMatrix = NULL; +PFNGLPUSHNAMEPROC glad_glPushName = NULL; +PFNGLQUERYCOUNTERPROC glad_glQueryCounter = NULL; +PFNGLRASTERPOS2DPROC glad_glRasterPos2d = NULL; +PFNGLRASTERPOS2DVPROC glad_glRasterPos2dv = NULL; +PFNGLRASTERPOS2FPROC glad_glRasterPos2f = NULL; +PFNGLRASTERPOS2FVPROC glad_glRasterPos2fv = NULL; +PFNGLRASTERPOS2IPROC glad_glRasterPos2i = NULL; +PFNGLRASTERPOS2IVPROC glad_glRasterPos2iv = NULL; +PFNGLRASTERPOS2SPROC glad_glRasterPos2s = NULL; +PFNGLRASTERPOS2SVPROC glad_glRasterPos2sv = NULL; +PFNGLRASTERPOS3DPROC glad_glRasterPos3d = NULL; +PFNGLRASTERPOS3DVPROC glad_glRasterPos3dv = NULL; +PFNGLRASTERPOS3FPROC glad_glRasterPos3f = NULL; +PFNGLRASTERPOS3FVPROC glad_glRasterPos3fv = NULL; +PFNGLRASTERPOS3IPROC glad_glRasterPos3i = NULL; +PFNGLRASTERPOS3IVPROC glad_glRasterPos3iv = NULL; +PFNGLRASTERPOS3SPROC glad_glRasterPos3s = NULL; +PFNGLRASTERPOS3SVPROC glad_glRasterPos3sv = NULL; +PFNGLRASTERPOS4DPROC glad_glRasterPos4d = NULL; +PFNGLRASTERPOS4DVPROC glad_glRasterPos4dv = NULL; +PFNGLRASTERPOS4FPROC glad_glRasterPos4f = NULL; +PFNGLRASTERPOS4FVPROC glad_glRasterPos4fv = NULL; +PFNGLRASTERPOS4IPROC glad_glRasterPos4i = NULL; +PFNGLRASTERPOS4IVPROC glad_glRasterPos4iv = NULL; +PFNGLRASTERPOS4SPROC glad_glRasterPos4s = NULL; +PFNGLRASTERPOS4SVPROC glad_glRasterPos4sv = NULL; +PFNGLREADBUFFERPROC glad_glReadBuffer = NULL; +PFNGLREADPIXELSPROC glad_glReadPixels = NULL; +PFNGLRECTDPROC glad_glRectd = NULL; +PFNGLRECTDVPROC glad_glRectdv = NULL; +PFNGLRECTFPROC glad_glRectf = NULL; +PFNGLRECTFVPROC glad_glRectfv = NULL; +PFNGLRECTIPROC glad_glRecti = NULL; +PFNGLRECTIVPROC glad_glRectiv = NULL; +PFNGLRECTSPROC glad_glRects = NULL; +PFNGLRECTSVPROC glad_glRectsv = NULL; +PFNGLRENDERMODEPROC glad_glRenderMode = NULL; +PFNGLRENDERBUFFERSTORAGEPROC glad_glRenderbufferStorage = NULL; +PFNGLRENDERBUFFERSTORAGEMULTISAMPLEPROC glad_glRenderbufferStorageMultisample = NULL; +PFNGLROTATEDPROC glad_glRotated = NULL; +PFNGLROTATEFPROC glad_glRotatef = NULL; +PFNGLSAMPLECOVERAGEPROC glad_glSampleCoverage = NULL; +PFNGLSAMPLEMASKIPROC glad_glSampleMaski = NULL; +PFNGLSAMPLERPARAMETERIIVPROC glad_glSamplerParameterIiv = NULL; +PFNGLSAMPLERPARAMETERIUIVPROC glad_glSamplerParameterIuiv = NULL; +PFNGLSAMPLERPARAMETERFPROC glad_glSamplerParameterf = NULL; +PFNGLSAMPLERPARAMETERFVPROC glad_glSamplerParameterfv = NULL; +PFNGLSAMPLERPARAMETERIPROC glad_glSamplerParameteri = NULL; +PFNGLSAMPLERPARAMETERIVPROC glad_glSamplerParameteriv = NULL; +PFNGLSCALEDPROC glad_glScaled = NULL; +PFNGLSCALEFPROC glad_glScalef = NULL; +PFNGLSCISSORPROC glad_glScissor = NULL; +PFNGLSECONDARYCOLOR3BPROC glad_glSecondaryColor3b = NULL; +PFNGLSECONDARYCOLOR3BVPROC glad_glSecondaryColor3bv = NULL; +PFNGLSECONDARYCOLOR3DPROC glad_glSecondaryColor3d = NULL; +PFNGLSECONDARYCOLOR3DVPROC glad_glSecondaryColor3dv = NULL; +PFNGLSECONDARYCOLOR3FPROC glad_glSecondaryColor3f = NULL; +PFNGLSECONDARYCOLOR3FVPROC glad_glSecondaryColor3fv = NULL; +PFNGLSECONDARYCOLOR3IPROC glad_glSecondaryColor3i = NULL; +PFNGLSECONDARYCOLOR3IVPROC glad_glSecondaryColor3iv = NULL; +PFNGLSECONDARYCOLOR3SPROC glad_glSecondaryColor3s = NULL; +PFNGLSECONDARYCOLOR3SVPROC glad_glSecondaryColor3sv = NULL; +PFNGLSECONDARYCOLOR3UBPROC glad_glSecondaryColor3ub = NULL; +PFNGLSECONDARYCOLOR3UBVPROC glad_glSecondaryColor3ubv = NULL; +PFNGLSECONDARYCOLOR3UIPROC glad_glSecondaryColor3ui = NULL; +PFNGLSECONDARYCOLOR3UIVPROC glad_glSecondaryColor3uiv = NULL; +PFNGLSECONDARYCOLOR3USPROC glad_glSecondaryColor3us = NULL; +PFNGLSECONDARYCOLOR3USVPROC glad_glSecondaryColor3usv = NULL; +PFNGLSECONDARYCOLORP3UIPROC glad_glSecondaryColorP3ui = NULL; +PFNGLSECONDARYCOLORP3UIVPROC glad_glSecondaryColorP3uiv = NULL; +PFNGLSECONDARYCOLORPOINTERPROC glad_glSecondaryColorPointer = NULL; +PFNGLSELECTBUFFERPROC glad_glSelectBuffer = NULL; +PFNGLSHADEMODELPROC glad_glShadeModel = NULL; +PFNGLSHADERSOURCEPROC glad_glShaderSource = NULL; +PFNGLSTENCILFUNCPROC glad_glStencilFunc = NULL; +PFNGLSTENCILFUNCSEPARATEPROC glad_glStencilFuncSeparate = NULL; +PFNGLSTENCILMASKPROC glad_glStencilMask = NULL; +PFNGLSTENCILMASKSEPARATEPROC glad_glStencilMaskSeparate = NULL; +PFNGLSTENCILOPPROC glad_glStencilOp = NULL; +PFNGLSTENCILOPSEPARATEPROC glad_glStencilOpSeparate = NULL; +PFNGLTEXBUFFERPROC glad_glTexBuffer = NULL; +PFNGLTEXCOORD1DPROC glad_glTexCoord1d = NULL; +PFNGLTEXCOORD1DVPROC glad_glTexCoord1dv = NULL; +PFNGLTEXCOORD1FPROC glad_glTexCoord1f = NULL; +PFNGLTEXCOORD1FVPROC glad_glTexCoord1fv = NULL; +PFNGLTEXCOORD1IPROC glad_glTexCoord1i = NULL; +PFNGLTEXCOORD1IVPROC glad_glTexCoord1iv = NULL; +PFNGLTEXCOORD1SPROC glad_glTexCoord1s = NULL; +PFNGLTEXCOORD1SVPROC glad_glTexCoord1sv = NULL; +PFNGLTEXCOORD2DPROC glad_glTexCoord2d = NULL; +PFNGLTEXCOORD2DVPROC glad_glTexCoord2dv = NULL; +PFNGLTEXCOORD2FPROC glad_glTexCoord2f = NULL; +PFNGLTEXCOORD2FVPROC glad_glTexCoord2fv = NULL; +PFNGLTEXCOORD2IPROC glad_glTexCoord2i = NULL; +PFNGLTEXCOORD2IVPROC glad_glTexCoord2iv = NULL; +PFNGLTEXCOORD2SPROC glad_glTexCoord2s = NULL; +PFNGLTEXCOORD2SVPROC glad_glTexCoord2sv = NULL; +PFNGLTEXCOORD3DPROC glad_glTexCoord3d = NULL; +PFNGLTEXCOORD3DVPROC glad_glTexCoord3dv = NULL; +PFNGLTEXCOORD3FPROC glad_glTexCoord3f = NULL; +PFNGLTEXCOORD3FVPROC glad_glTexCoord3fv = NULL; +PFNGLTEXCOORD3IPROC glad_glTexCoord3i = NULL; +PFNGLTEXCOORD3IVPROC glad_glTexCoord3iv = NULL; +PFNGLTEXCOORD3SPROC glad_glTexCoord3s = NULL; +PFNGLTEXCOORD3SVPROC glad_glTexCoord3sv = NULL; +PFNGLTEXCOORD4DPROC glad_glTexCoord4d = NULL; +PFNGLTEXCOORD4DVPROC glad_glTexCoord4dv = NULL; +PFNGLTEXCOORD4FPROC glad_glTexCoord4f = NULL; +PFNGLTEXCOORD4FVPROC glad_glTexCoord4fv = NULL; +PFNGLTEXCOORD4IPROC glad_glTexCoord4i = NULL; +PFNGLTEXCOORD4IVPROC glad_glTexCoord4iv = NULL; +PFNGLTEXCOORD4SPROC glad_glTexCoord4s = NULL; +PFNGLTEXCOORD4SVPROC glad_glTexCoord4sv = NULL; +PFNGLTEXCOORDP1UIPROC glad_glTexCoordP1ui = NULL; +PFNGLTEXCOORDP1UIVPROC glad_glTexCoordP1uiv = NULL; +PFNGLTEXCOORDP2UIPROC glad_glTexCoordP2ui = NULL; +PFNGLTEXCOORDP2UIVPROC glad_glTexCoordP2uiv = NULL; +PFNGLTEXCOORDP3UIPROC glad_glTexCoordP3ui = NULL; +PFNGLTEXCOORDP3UIVPROC glad_glTexCoordP3uiv = NULL; +PFNGLTEXCOORDP4UIPROC glad_glTexCoordP4ui = NULL; +PFNGLTEXCOORDP4UIVPROC glad_glTexCoordP4uiv = NULL; +PFNGLTEXCOORDPOINTERPROC glad_glTexCoordPointer = NULL; +PFNGLTEXENVFPROC glad_glTexEnvf = NULL; +PFNGLTEXENVFVPROC glad_glTexEnvfv = NULL; +PFNGLTEXENVIPROC glad_glTexEnvi = NULL; +PFNGLTEXENVIVPROC glad_glTexEnviv = NULL; +PFNGLTEXGENDPROC glad_glTexGend = NULL; +PFNGLTEXGENDVPROC glad_glTexGendv = NULL; +PFNGLTEXGENFPROC glad_glTexGenf = NULL; +PFNGLTEXGENFVPROC glad_glTexGenfv = NULL; +PFNGLTEXGENIPROC glad_glTexGeni = NULL; +PFNGLTEXGENIVPROC glad_glTexGeniv = NULL; +PFNGLTEXIMAGE1DPROC glad_glTexImage1D = NULL; +PFNGLTEXIMAGE2DPROC glad_glTexImage2D = NULL; +PFNGLTEXIMAGE2DMULTISAMPLEPROC glad_glTexImage2DMultisample = NULL; +PFNGLTEXIMAGE3DPROC glad_glTexImage3D = NULL; +PFNGLTEXIMAGE3DMULTISAMPLEPROC glad_glTexImage3DMultisample = NULL; +PFNGLTEXPARAMETERIIVPROC glad_glTexParameterIiv = NULL; +PFNGLTEXPARAMETERIUIVPROC glad_glTexParameterIuiv = NULL; +PFNGLTEXPARAMETERFPROC glad_glTexParameterf = NULL; +PFNGLTEXPARAMETERFVPROC glad_glTexParameterfv = NULL; +PFNGLTEXPARAMETERIPROC glad_glTexParameteri = NULL; +PFNGLTEXPARAMETERIVPROC glad_glTexParameteriv = NULL; +PFNGLTEXSUBIMAGE1DPROC glad_glTexSubImage1D = NULL; +PFNGLTEXSUBIMAGE2DPROC glad_glTexSubImage2D = NULL; +PFNGLTEXSUBIMAGE3DPROC glad_glTexSubImage3D = NULL; +PFNGLTRANSFORMFEEDBACKVARYINGSPROC glad_glTransformFeedbackVaryings = NULL; +PFNGLTRANSLATEDPROC glad_glTranslated = NULL; +PFNGLTRANSLATEFPROC glad_glTranslatef = NULL; +PFNGLUNIFORM1FPROC glad_glUniform1f = NULL; +PFNGLUNIFORM1FVPROC glad_glUniform1fv = NULL; +PFNGLUNIFORM1IPROC glad_glUniform1i = NULL; +PFNGLUNIFORM1IVPROC glad_glUniform1iv = NULL; +PFNGLUNIFORM1UIPROC glad_glUniform1ui = NULL; +PFNGLUNIFORM1UIVPROC glad_glUniform1uiv = NULL; +PFNGLUNIFORM2FPROC glad_glUniform2f = NULL; +PFNGLUNIFORM2FVPROC glad_glUniform2fv = NULL; +PFNGLUNIFORM2IPROC glad_glUniform2i = NULL; +PFNGLUNIFORM2IVPROC glad_glUniform2iv = NULL; +PFNGLUNIFORM2UIPROC glad_glUniform2ui = NULL; +PFNGLUNIFORM2UIVPROC glad_glUniform2uiv = NULL; +PFNGLUNIFORM3FPROC glad_glUniform3f = NULL; +PFNGLUNIFORM3FVPROC glad_glUniform3fv = NULL; +PFNGLUNIFORM3IPROC glad_glUniform3i = NULL; +PFNGLUNIFORM3IVPROC glad_glUniform3iv = NULL; +PFNGLUNIFORM3UIPROC glad_glUniform3ui = NULL; +PFNGLUNIFORM3UIVPROC glad_glUniform3uiv = NULL; +PFNGLUNIFORM4FPROC glad_glUniform4f = NULL; +PFNGLUNIFORM4FVPROC glad_glUniform4fv = NULL; +PFNGLUNIFORM4IPROC glad_glUniform4i = NULL; +PFNGLUNIFORM4IVPROC glad_glUniform4iv = NULL; +PFNGLUNIFORM4UIPROC glad_glUniform4ui = NULL; +PFNGLUNIFORM4UIVPROC glad_glUniform4uiv = NULL; +PFNGLUNIFORMBLOCKBINDINGPROC glad_glUniformBlockBinding = NULL; +PFNGLUNIFORMMATRIX2FVPROC glad_glUniformMatrix2fv = NULL; +PFNGLUNIFORMMATRIX2X3FVPROC glad_glUniformMatrix2x3fv = NULL; +PFNGLUNIFORMMATRIX2X4FVPROC glad_glUniformMatrix2x4fv = NULL; +PFNGLUNIFORMMATRIX3FVPROC glad_glUniformMatrix3fv = NULL; +PFNGLUNIFORMMATRIX3X2FVPROC glad_glUniformMatrix3x2fv = NULL; +PFNGLUNIFORMMATRIX3X4FVPROC glad_glUniformMatrix3x4fv = NULL; +PFNGLUNIFORMMATRIX4FVPROC glad_glUniformMatrix4fv = NULL; +PFNGLUNIFORMMATRIX4X2FVPROC glad_glUniformMatrix4x2fv = NULL; +PFNGLUNIFORMMATRIX4X3FVPROC glad_glUniformMatrix4x3fv = NULL; +PFNGLUNMAPBUFFERPROC glad_glUnmapBuffer = NULL; +PFNGLUSEPROGRAMPROC glad_glUseProgram = NULL; +PFNGLVALIDATEPROGRAMPROC glad_glValidateProgram = NULL; +PFNGLVERTEX2DPROC glad_glVertex2d = NULL; +PFNGLVERTEX2DVPROC glad_glVertex2dv = NULL; +PFNGLVERTEX2FPROC glad_glVertex2f = NULL; +PFNGLVERTEX2FVPROC glad_glVertex2fv = NULL; +PFNGLVERTEX2IPROC glad_glVertex2i = NULL; +PFNGLVERTEX2IVPROC glad_glVertex2iv = NULL; +PFNGLVERTEX2SPROC glad_glVertex2s = NULL; +PFNGLVERTEX2SVPROC glad_glVertex2sv = NULL; +PFNGLVERTEX3DPROC glad_glVertex3d = NULL; +PFNGLVERTEX3DVPROC glad_glVertex3dv = NULL; +PFNGLVERTEX3FPROC glad_glVertex3f = NULL; +PFNGLVERTEX3FVPROC glad_glVertex3fv = NULL; +PFNGLVERTEX3IPROC glad_glVertex3i = NULL; +PFNGLVERTEX3IVPROC glad_glVertex3iv = NULL; +PFNGLVERTEX3SPROC glad_glVertex3s = NULL; +PFNGLVERTEX3SVPROC glad_glVertex3sv = NULL; +PFNGLVERTEX4DPROC glad_glVertex4d = NULL; +PFNGLVERTEX4DVPROC glad_glVertex4dv = NULL; +PFNGLVERTEX4FPROC glad_glVertex4f = NULL; +PFNGLVERTEX4FVPROC glad_glVertex4fv = NULL; +PFNGLVERTEX4IPROC glad_glVertex4i = NULL; +PFNGLVERTEX4IVPROC glad_glVertex4iv = NULL; +PFNGLVERTEX4SPROC glad_glVertex4s = NULL; +PFNGLVERTEX4SVPROC glad_glVertex4sv = NULL; +PFNGLVERTEXATTRIB1DPROC glad_glVertexAttrib1d = NULL; +PFNGLVERTEXATTRIB1DVPROC glad_glVertexAttrib1dv = NULL; +PFNGLVERTEXATTRIB1FPROC glad_glVertexAttrib1f = NULL; +PFNGLVERTEXATTRIB1FVPROC glad_glVertexAttrib1fv = NULL; +PFNGLVERTEXATTRIB1SPROC glad_glVertexAttrib1s = NULL; +PFNGLVERTEXATTRIB1SVPROC glad_glVertexAttrib1sv = NULL; +PFNGLVERTEXATTRIB2DPROC glad_glVertexAttrib2d = NULL; +PFNGLVERTEXATTRIB2DVPROC glad_glVertexAttrib2dv = NULL; +PFNGLVERTEXATTRIB2FPROC glad_glVertexAttrib2f = NULL; +PFNGLVERTEXATTRIB2FVPROC glad_glVertexAttrib2fv = NULL; +PFNGLVERTEXATTRIB2SPROC glad_glVertexAttrib2s = NULL; +PFNGLVERTEXATTRIB2SVPROC glad_glVertexAttrib2sv = NULL; +PFNGLVERTEXATTRIB3DPROC glad_glVertexAttrib3d = NULL; +PFNGLVERTEXATTRIB3DVPROC glad_glVertexAttrib3dv = NULL; +PFNGLVERTEXATTRIB3FPROC glad_glVertexAttrib3f = NULL; +PFNGLVERTEXATTRIB3FVPROC glad_glVertexAttrib3fv = NULL; +PFNGLVERTEXATTRIB3SPROC glad_glVertexAttrib3s = NULL; +PFNGLVERTEXATTRIB3SVPROC glad_glVertexAttrib3sv = NULL; +PFNGLVERTEXATTRIB4NBVPROC glad_glVertexAttrib4Nbv = NULL; +PFNGLVERTEXATTRIB4NIVPROC glad_glVertexAttrib4Niv = NULL; +PFNGLVERTEXATTRIB4NSVPROC glad_glVertexAttrib4Nsv = NULL; +PFNGLVERTEXATTRIB4NUBPROC glad_glVertexAttrib4Nub = NULL; +PFNGLVERTEXATTRIB4NUBVPROC glad_glVertexAttrib4Nubv = NULL; +PFNGLVERTEXATTRIB4NUIVPROC glad_glVertexAttrib4Nuiv = NULL; +PFNGLVERTEXATTRIB4NUSVPROC glad_glVertexAttrib4Nusv = NULL; +PFNGLVERTEXATTRIB4BVPROC glad_glVertexAttrib4bv = NULL; +PFNGLVERTEXATTRIB4DPROC glad_glVertexAttrib4d = NULL; +PFNGLVERTEXATTRIB4DVPROC glad_glVertexAttrib4dv = NULL; +PFNGLVERTEXATTRIB4FPROC glad_glVertexAttrib4f = NULL; +PFNGLVERTEXATTRIB4FVPROC glad_glVertexAttrib4fv = NULL; +PFNGLVERTEXATTRIB4IVPROC glad_glVertexAttrib4iv = NULL; +PFNGLVERTEXATTRIB4SPROC glad_glVertexAttrib4s = NULL; +PFNGLVERTEXATTRIB4SVPROC glad_glVertexAttrib4sv = NULL; +PFNGLVERTEXATTRIB4UBVPROC glad_glVertexAttrib4ubv = NULL; +PFNGLVERTEXATTRIB4UIVPROC glad_glVertexAttrib4uiv = NULL; +PFNGLVERTEXATTRIB4USVPROC glad_glVertexAttrib4usv = NULL; +PFNGLVERTEXATTRIBDIVISORPROC glad_glVertexAttribDivisor = NULL; +PFNGLVERTEXATTRIBI1IPROC glad_glVertexAttribI1i = NULL; +PFNGLVERTEXATTRIBI1IVPROC glad_glVertexAttribI1iv = NULL; +PFNGLVERTEXATTRIBI1UIPROC glad_glVertexAttribI1ui = NULL; +PFNGLVERTEXATTRIBI1UIVPROC glad_glVertexAttribI1uiv = NULL; +PFNGLVERTEXATTRIBI2IPROC glad_glVertexAttribI2i = NULL; +PFNGLVERTEXATTRIBI2IVPROC glad_glVertexAttribI2iv = NULL; +PFNGLVERTEXATTRIBI2UIPROC glad_glVertexAttribI2ui = NULL; +PFNGLVERTEXATTRIBI2UIVPROC glad_glVertexAttribI2uiv = NULL; +PFNGLVERTEXATTRIBI3IPROC glad_glVertexAttribI3i = NULL; +PFNGLVERTEXATTRIBI3IVPROC glad_glVertexAttribI3iv = NULL; +PFNGLVERTEXATTRIBI3UIPROC glad_glVertexAttribI3ui = NULL; +PFNGLVERTEXATTRIBI3UIVPROC glad_glVertexAttribI3uiv = NULL; +PFNGLVERTEXATTRIBI4BVPROC glad_glVertexAttribI4bv = NULL; +PFNGLVERTEXATTRIBI4IPROC glad_glVertexAttribI4i = NULL; +PFNGLVERTEXATTRIBI4IVPROC glad_glVertexAttribI4iv = NULL; +PFNGLVERTEXATTRIBI4SVPROC glad_glVertexAttribI4sv = NULL; +PFNGLVERTEXATTRIBI4UBVPROC glad_glVertexAttribI4ubv = NULL; +PFNGLVERTEXATTRIBI4UIPROC glad_glVertexAttribI4ui = NULL; +PFNGLVERTEXATTRIBI4UIVPROC glad_glVertexAttribI4uiv = NULL; +PFNGLVERTEXATTRIBI4USVPROC glad_glVertexAttribI4usv = NULL; +PFNGLVERTEXATTRIBIPOINTERPROC glad_glVertexAttribIPointer = NULL; +PFNGLVERTEXATTRIBP1UIPROC glad_glVertexAttribP1ui = NULL; +PFNGLVERTEXATTRIBP1UIVPROC glad_glVertexAttribP1uiv = NULL; +PFNGLVERTEXATTRIBP2UIPROC glad_glVertexAttribP2ui = NULL; +PFNGLVERTEXATTRIBP2UIVPROC glad_glVertexAttribP2uiv = NULL; +PFNGLVERTEXATTRIBP3UIPROC glad_glVertexAttribP3ui = NULL; +PFNGLVERTEXATTRIBP3UIVPROC glad_glVertexAttribP3uiv = NULL; +PFNGLVERTEXATTRIBP4UIPROC glad_glVertexAttribP4ui = NULL; +PFNGLVERTEXATTRIBP4UIVPROC glad_glVertexAttribP4uiv = NULL; +PFNGLVERTEXATTRIBPOINTERPROC glad_glVertexAttribPointer = NULL; +PFNGLVERTEXP2UIPROC glad_glVertexP2ui = NULL; +PFNGLVERTEXP2UIVPROC glad_glVertexP2uiv = NULL; +PFNGLVERTEXP3UIPROC glad_glVertexP3ui = NULL; +PFNGLVERTEXP3UIVPROC glad_glVertexP3uiv = NULL; +PFNGLVERTEXP4UIPROC glad_glVertexP4ui = NULL; +PFNGLVERTEXP4UIVPROC glad_glVertexP4uiv = NULL; +PFNGLVERTEXPOINTERPROC glad_glVertexPointer = NULL; +PFNGLVIEWPORTPROC glad_glViewport = NULL; +PFNGLWAITSYNCPROC glad_glWaitSync = NULL; +PFNGLWINDOWPOS2DPROC glad_glWindowPos2d = NULL; +PFNGLWINDOWPOS2DVPROC glad_glWindowPos2dv = NULL; +PFNGLWINDOWPOS2FPROC glad_glWindowPos2f = NULL; +PFNGLWINDOWPOS2FVPROC glad_glWindowPos2fv = NULL; +PFNGLWINDOWPOS2IPROC glad_glWindowPos2i = NULL; +PFNGLWINDOWPOS2IVPROC glad_glWindowPos2iv = NULL; +PFNGLWINDOWPOS2SPROC glad_glWindowPos2s = NULL; +PFNGLWINDOWPOS2SVPROC glad_glWindowPos2sv = NULL; +PFNGLWINDOWPOS3DPROC glad_glWindowPos3d = NULL; +PFNGLWINDOWPOS3DVPROC glad_glWindowPos3dv = NULL; +PFNGLWINDOWPOS3FPROC glad_glWindowPos3f = NULL; +PFNGLWINDOWPOS3FVPROC glad_glWindowPos3fv = NULL; +PFNGLWINDOWPOS3IPROC glad_glWindowPos3i = NULL; +PFNGLWINDOWPOS3IVPROC glad_glWindowPos3iv = NULL; +PFNGLWINDOWPOS3SPROC glad_glWindowPos3s = NULL; +PFNGLWINDOWPOS3SVPROC glad_glWindowPos3sv = NULL; +static void load_GL_VERSION_1_0(GLADloadproc load) { + if (!GLAD_GL_VERSION_1_0) return; + glad_glCullFace = (PFNGLCULLFACEPROC)load("glCullFace"); + glad_glFrontFace = (PFNGLFRONTFACEPROC)load("glFrontFace"); + glad_glHint = (PFNGLHINTPROC)load("glHint"); + glad_glLineWidth = (PFNGLLINEWIDTHPROC)load("glLineWidth"); + glad_glPointSize = (PFNGLPOINTSIZEPROC)load("glPointSize"); + glad_glPolygonMode = (PFNGLPOLYGONMODEPROC)load("glPolygonMode"); + glad_glScissor = (PFNGLSCISSORPROC)load("glScissor"); + glad_glTexParameterf = (PFNGLTEXPARAMETERFPROC)load("glTexParameterf"); + glad_glTexParameterfv = (PFNGLTEXPARAMETERFVPROC)load("glTexParameterfv"); + glad_glTexParameteri = (PFNGLTEXPARAMETERIPROC)load("glTexParameteri"); + glad_glTexParameteriv = (PFNGLTEXPARAMETERIVPROC)load("glTexParameteriv"); + glad_glTexImage1D = (PFNGLTEXIMAGE1DPROC)load("glTexImage1D"); + glad_glTexImage2D = (PFNGLTEXIMAGE2DPROC)load("glTexImage2D"); + glad_glDrawBuffer = (PFNGLDRAWBUFFERPROC)load("glDrawBuffer"); + glad_glClear = (PFNGLCLEARPROC)load("glClear"); + glad_glClearColor = (PFNGLCLEARCOLORPROC)load("glClearColor"); + glad_glClearStencil = (PFNGLCLEARSTENCILPROC)load("glClearStencil"); + glad_glClearDepth = (PFNGLCLEARDEPTHPROC)load("glClearDepth"); + glad_glStencilMask = (PFNGLSTENCILMASKPROC)load("glStencilMask"); + glad_glColorMask = (PFNGLCOLORMASKPROC)load("glColorMask"); + glad_glDepthMask = (PFNGLDEPTHMASKPROC)load("glDepthMask"); + glad_glDisable = (PFNGLDISABLEPROC)load("glDisable"); + glad_glEnable = (PFNGLENABLEPROC)load("glEnable"); + glad_glFinish = (PFNGLFINISHPROC)load("glFinish"); + glad_glFlush = (PFNGLFLUSHPROC)load("glFlush"); + glad_glBlendFunc = (PFNGLBLENDFUNCPROC)load("glBlendFunc"); + glad_glLogicOp = (PFNGLLOGICOPPROC)load("glLogicOp"); + glad_glStencilFunc = (PFNGLSTENCILFUNCPROC)load("glStencilFunc"); + glad_glStencilOp = (PFNGLSTENCILOPPROC)load("glStencilOp"); + glad_glDepthFunc = (PFNGLDEPTHFUNCPROC)load("glDepthFunc"); + glad_glPixelStoref = (PFNGLPIXELSTOREFPROC)load("glPixelStoref"); + glad_glPixelStorei = (PFNGLPIXELSTOREIPROC)load("glPixelStorei"); + glad_glReadBuffer = (PFNGLREADBUFFERPROC)load("glReadBuffer"); + glad_glReadPixels = (PFNGLREADPIXELSPROC)load("glReadPixels"); + glad_glGetBooleanv = (PFNGLGETBOOLEANVPROC)load("glGetBooleanv"); + glad_glGetDoublev = (PFNGLGETDOUBLEVPROC)load("glGetDoublev"); + glad_glGetError = (PFNGLGETERRORPROC)load("glGetError"); + glad_glGetFloatv = (PFNGLGETFLOATVPROC)load("glGetFloatv"); + glad_glGetIntegerv = (PFNGLGETINTEGERVPROC)load("glGetIntegerv"); + glad_glGetString = (PFNGLGETSTRINGPROC)load("glGetString"); + glad_glGetTexImage = (PFNGLGETTEXIMAGEPROC)load("glGetTexImage"); + glad_glGetTexParameterfv = (PFNGLGETTEXPARAMETERFVPROC)load("glGetTexParameterfv"); + glad_glGetTexParameteriv = (PFNGLGETTEXPARAMETERIVPROC)load("glGetTexParameteriv"); + glad_glGetTexLevelParameterfv = (PFNGLGETTEXLEVELPARAMETERFVPROC)load("glGetTexLevelParameterfv"); + glad_glGetTexLevelParameteriv = (PFNGLGETTEXLEVELPARAMETERIVPROC)load("glGetTexLevelParameteriv"); + glad_glIsEnabled = (PFNGLISENABLEDPROC)load("glIsEnabled"); + glad_glDepthRange = (PFNGLDEPTHRANGEPROC)load("glDepthRange"); + glad_glViewport = (PFNGLVIEWPORTPROC)load("glViewport"); + glad_glNewList = (PFNGLNEWLISTPROC)load("glNewList"); + glad_glEndList = (PFNGLENDLISTPROC)load("glEndList"); + glad_glCallList = (PFNGLCALLLISTPROC)load("glCallList"); + glad_glCallLists = (PFNGLCALLLISTSPROC)load("glCallLists"); + glad_glDeleteLists = (PFNGLDELETELISTSPROC)load("glDeleteLists"); + glad_glGenLists = (PFNGLGENLISTSPROC)load("glGenLists"); + glad_glListBase = (PFNGLLISTBASEPROC)load("glListBase"); + glad_glBegin = (PFNGLBEGINPROC)load("glBegin"); + glad_glBitmap = (PFNGLBITMAPPROC)load("glBitmap"); + glad_glColor3b = (PFNGLCOLOR3BPROC)load("glColor3b"); + glad_glColor3bv = (PFNGLCOLOR3BVPROC)load("glColor3bv"); + glad_glColor3d = (PFNGLCOLOR3DPROC)load("glColor3d"); + glad_glColor3dv = (PFNGLCOLOR3DVPROC)load("glColor3dv"); + glad_glColor3f = (PFNGLCOLOR3FPROC)load("glColor3f"); + glad_glColor3fv = (PFNGLCOLOR3FVPROC)load("glColor3fv"); + glad_glColor3i = (PFNGLCOLOR3IPROC)load("glColor3i"); + glad_glColor3iv = (PFNGLCOLOR3IVPROC)load("glColor3iv"); + glad_glColor3s = (PFNGLCOLOR3SPROC)load("glColor3s"); + glad_glColor3sv = (PFNGLCOLOR3SVPROC)load("glColor3sv"); + glad_glColor3ub = (PFNGLCOLOR3UBPROC)load("glColor3ub"); + glad_glColor3ubv = (PFNGLCOLOR3UBVPROC)load("glColor3ubv"); + glad_glColor3ui = (PFNGLCOLOR3UIPROC)load("glColor3ui"); + glad_glColor3uiv = (PFNGLCOLOR3UIVPROC)load("glColor3uiv"); + glad_glColor3us = (PFNGLCOLOR3USPROC)load("glColor3us"); + glad_glColor3usv = (PFNGLCOLOR3USVPROC)load("glColor3usv"); + glad_glColor4b = (PFNGLCOLOR4BPROC)load("glColor4b"); + glad_glColor4bv = (PFNGLCOLOR4BVPROC)load("glColor4bv"); + glad_glColor4d = (PFNGLCOLOR4DPROC)load("glColor4d"); + glad_glColor4dv = (PFNGLCOLOR4DVPROC)load("glColor4dv"); + glad_glColor4f = (PFNGLCOLOR4FPROC)load("glColor4f"); + glad_glColor4fv = (PFNGLCOLOR4FVPROC)load("glColor4fv"); + glad_glColor4i = (PFNGLCOLOR4IPROC)load("glColor4i"); + glad_glColor4iv = (PFNGLCOLOR4IVPROC)load("glColor4iv"); + glad_glColor4s = (PFNGLCOLOR4SPROC)load("glColor4s"); + glad_glColor4sv = (PFNGLCOLOR4SVPROC)load("glColor4sv"); + glad_glColor4ub = (PFNGLCOLOR4UBPROC)load("glColor4ub"); + glad_glColor4ubv = (PFNGLCOLOR4UBVPROC)load("glColor4ubv"); + glad_glColor4ui = (PFNGLCOLOR4UIPROC)load("glColor4ui"); + glad_glColor4uiv = (PFNGLCOLOR4UIVPROC)load("glColor4uiv"); + glad_glColor4us = (PFNGLCOLOR4USPROC)load("glColor4us"); + glad_glColor4usv = (PFNGLCOLOR4USVPROC)load("glColor4usv"); + glad_glEdgeFlag = (PFNGLEDGEFLAGPROC)load("glEdgeFlag"); + glad_glEdgeFlagv = (PFNGLEDGEFLAGVPROC)load("glEdgeFlagv"); + glad_glEnd = (PFNGLENDPROC)load("glEnd"); + glad_glIndexd = (PFNGLINDEXDPROC)load("glIndexd"); + glad_glIndexdv = (PFNGLINDEXDVPROC)load("glIndexdv"); + glad_glIndexf = (PFNGLINDEXFPROC)load("glIndexf"); + glad_glIndexfv = (PFNGLINDEXFVPROC)load("glIndexfv"); + glad_glIndexi = (PFNGLINDEXIPROC)load("glIndexi"); + glad_glIndexiv = (PFNGLINDEXIVPROC)load("glIndexiv"); + glad_glIndexs = (PFNGLINDEXSPROC)load("glIndexs"); + glad_glIndexsv = (PFNGLINDEXSVPROC)load("glIndexsv"); + glad_glNormal3b = (PFNGLNORMAL3BPROC)load("glNormal3b"); + glad_glNormal3bv = (PFNGLNORMAL3BVPROC)load("glNormal3bv"); + glad_glNormal3d = (PFNGLNORMAL3DPROC)load("glNormal3d"); + glad_glNormal3dv = (PFNGLNORMAL3DVPROC)load("glNormal3dv"); + glad_glNormal3f = (PFNGLNORMAL3FPROC)load("glNormal3f"); + glad_glNormal3fv = (PFNGLNORMAL3FVPROC)load("glNormal3fv"); + glad_glNormal3i = (PFNGLNORMAL3IPROC)load("glNormal3i"); + glad_glNormal3iv = (PFNGLNORMAL3IVPROC)load("glNormal3iv"); + glad_glNormal3s = (PFNGLNORMAL3SPROC)load("glNormal3s"); + glad_glNormal3sv = (PFNGLNORMAL3SVPROC)load("glNormal3sv"); + glad_glRasterPos2d = (PFNGLRASTERPOS2DPROC)load("glRasterPos2d"); + glad_glRasterPos2dv = (PFNGLRASTERPOS2DVPROC)load("glRasterPos2dv"); + glad_glRasterPos2f = (PFNGLRASTERPOS2FPROC)load("glRasterPos2f"); + glad_glRasterPos2fv = (PFNGLRASTERPOS2FVPROC)load("glRasterPos2fv"); + glad_glRasterPos2i = (PFNGLRASTERPOS2IPROC)load("glRasterPos2i"); + glad_glRasterPos2iv = (PFNGLRASTERPOS2IVPROC)load("glRasterPos2iv"); + glad_glRasterPos2s = (PFNGLRASTERPOS2SPROC)load("glRasterPos2s"); + glad_glRasterPos2sv = (PFNGLRASTERPOS2SVPROC)load("glRasterPos2sv"); + glad_glRasterPos3d = (PFNGLRASTERPOS3DPROC)load("glRasterPos3d"); + glad_glRasterPos3dv = (PFNGLRASTERPOS3DVPROC)load("glRasterPos3dv"); + glad_glRasterPos3f = (PFNGLRASTERPOS3FPROC)load("glRasterPos3f"); + glad_glRasterPos3fv = (PFNGLRASTERPOS3FVPROC)load("glRasterPos3fv"); + glad_glRasterPos3i = (PFNGLRASTERPOS3IPROC)load("glRasterPos3i"); + glad_glRasterPos3iv = (PFNGLRASTERPOS3IVPROC)load("glRasterPos3iv"); + glad_glRasterPos3s = (PFNGLRASTERPOS3SPROC)load("glRasterPos3s"); + glad_glRasterPos3sv = (PFNGLRASTERPOS3SVPROC)load("glRasterPos3sv"); + glad_glRasterPos4d = (PFNGLRASTERPOS4DPROC)load("glRasterPos4d"); + glad_glRasterPos4dv = (PFNGLRASTERPOS4DVPROC)load("glRasterPos4dv"); + glad_glRasterPos4f = (PFNGLRASTERPOS4FPROC)load("glRasterPos4f"); + glad_glRasterPos4fv = (PFNGLRASTERPOS4FVPROC)load("glRasterPos4fv"); + glad_glRasterPos4i = (PFNGLRASTERPOS4IPROC)load("glRasterPos4i"); + glad_glRasterPos4iv = (PFNGLRASTERPOS4IVPROC)load("glRasterPos4iv"); + glad_glRasterPos4s = (PFNGLRASTERPOS4SPROC)load("glRasterPos4s"); + glad_glRasterPos4sv = (PFNGLRASTERPOS4SVPROC)load("glRasterPos4sv"); + glad_glRectd = (PFNGLRECTDPROC)load("glRectd"); + glad_glRectdv = (PFNGLRECTDVPROC)load("glRectdv"); + glad_glRectf = (PFNGLRECTFPROC)load("glRectf"); + glad_glRectfv = (PFNGLRECTFVPROC)load("glRectfv"); + glad_glRecti = (PFNGLRECTIPROC)load("glRecti"); + glad_glRectiv = (PFNGLRECTIVPROC)load("glRectiv"); + glad_glRects = (PFNGLRECTSPROC)load("glRects"); + glad_glRectsv = (PFNGLRECTSVPROC)load("glRectsv"); + glad_glTexCoord1d = (PFNGLTEXCOORD1DPROC)load("glTexCoord1d"); + glad_glTexCoord1dv = (PFNGLTEXCOORD1DVPROC)load("glTexCoord1dv"); + glad_glTexCoord1f = (PFNGLTEXCOORD1FPROC)load("glTexCoord1f"); + glad_glTexCoord1fv = (PFNGLTEXCOORD1FVPROC)load("glTexCoord1fv"); + glad_glTexCoord1i = (PFNGLTEXCOORD1IPROC)load("glTexCoord1i"); + glad_glTexCoord1iv = (PFNGLTEXCOORD1IVPROC)load("glTexCoord1iv"); + glad_glTexCoord1s = (PFNGLTEXCOORD1SPROC)load("glTexCoord1s"); + glad_glTexCoord1sv = (PFNGLTEXCOORD1SVPROC)load("glTexCoord1sv"); + glad_glTexCoord2d = (PFNGLTEXCOORD2DPROC)load("glTexCoord2d"); + glad_glTexCoord2dv = (PFNGLTEXCOORD2DVPROC)load("glTexCoord2dv"); + glad_glTexCoord2f = (PFNGLTEXCOORD2FPROC)load("glTexCoord2f"); + glad_glTexCoord2fv = (PFNGLTEXCOORD2FVPROC)load("glTexCoord2fv"); + glad_glTexCoord2i = (PFNGLTEXCOORD2IPROC)load("glTexCoord2i"); + glad_glTexCoord2iv = (PFNGLTEXCOORD2IVPROC)load("glTexCoord2iv"); + glad_glTexCoord2s = (PFNGLTEXCOORD2SPROC)load("glTexCoord2s"); + glad_glTexCoord2sv = (PFNGLTEXCOORD2SVPROC)load("glTexCoord2sv"); + glad_glTexCoord3d = (PFNGLTEXCOORD3DPROC)load("glTexCoord3d"); + glad_glTexCoord3dv = (PFNGLTEXCOORD3DVPROC)load("glTexCoord3dv"); + glad_glTexCoord3f = (PFNGLTEXCOORD3FPROC)load("glTexCoord3f"); + glad_glTexCoord3fv = (PFNGLTEXCOORD3FVPROC)load("glTexCoord3fv"); + glad_glTexCoord3i = (PFNGLTEXCOORD3IPROC)load("glTexCoord3i"); + glad_glTexCoord3iv = (PFNGLTEXCOORD3IVPROC)load("glTexCoord3iv"); + glad_glTexCoord3s = (PFNGLTEXCOORD3SPROC)load("glTexCoord3s"); + glad_glTexCoord3sv = (PFNGLTEXCOORD3SVPROC)load("glTexCoord3sv"); + glad_glTexCoord4d = (PFNGLTEXCOORD4DPROC)load("glTexCoord4d"); + glad_glTexCoord4dv = (PFNGLTEXCOORD4DVPROC)load("glTexCoord4dv"); + glad_glTexCoord4f = (PFNGLTEXCOORD4FPROC)load("glTexCoord4f"); + glad_glTexCoord4fv = (PFNGLTEXCOORD4FVPROC)load("glTexCoord4fv"); + glad_glTexCoord4i = (PFNGLTEXCOORD4IPROC)load("glTexCoord4i"); + glad_glTexCoord4iv = (PFNGLTEXCOORD4IVPROC)load("glTexCoord4iv"); + glad_glTexCoord4s = (PFNGLTEXCOORD4SPROC)load("glTexCoord4s"); + glad_glTexCoord4sv = (PFNGLTEXCOORD4SVPROC)load("glTexCoord4sv"); + glad_glVertex2d = (PFNGLVERTEX2DPROC)load("glVertex2d"); + glad_glVertex2dv = (PFNGLVERTEX2DVPROC)load("glVertex2dv"); + glad_glVertex2f = (PFNGLVERTEX2FPROC)load("glVertex2f"); + glad_glVertex2fv = (PFNGLVERTEX2FVPROC)load("glVertex2fv"); + glad_glVertex2i = (PFNGLVERTEX2IPROC)load("glVertex2i"); + glad_glVertex2iv = (PFNGLVERTEX2IVPROC)load("glVertex2iv"); + glad_glVertex2s = (PFNGLVERTEX2SPROC)load("glVertex2s"); + glad_glVertex2sv = (PFNGLVERTEX2SVPROC)load("glVertex2sv"); + glad_glVertex3d = (PFNGLVERTEX3DPROC)load("glVertex3d"); + glad_glVertex3dv = (PFNGLVERTEX3DVPROC)load("glVertex3dv"); + glad_glVertex3f = (PFNGLVERTEX3FPROC)load("glVertex3f"); + glad_glVertex3fv = (PFNGLVERTEX3FVPROC)load("glVertex3fv"); + glad_glVertex3i = (PFNGLVERTEX3IPROC)load("glVertex3i"); + glad_glVertex3iv = (PFNGLVERTEX3IVPROC)load("glVertex3iv"); + glad_glVertex3s = (PFNGLVERTEX3SPROC)load("glVertex3s"); + glad_glVertex3sv = (PFNGLVERTEX3SVPROC)load("glVertex3sv"); + glad_glVertex4d = (PFNGLVERTEX4DPROC)load("glVertex4d"); + glad_glVertex4dv = (PFNGLVERTEX4DVPROC)load("glVertex4dv"); + glad_glVertex4f = (PFNGLVERTEX4FPROC)load("glVertex4f"); + glad_glVertex4fv = (PFNGLVERTEX4FVPROC)load("glVertex4fv"); + glad_glVertex4i = (PFNGLVERTEX4IPROC)load("glVertex4i"); + glad_glVertex4iv = (PFNGLVERTEX4IVPROC)load("glVertex4iv"); + glad_glVertex4s = (PFNGLVERTEX4SPROC)load("glVertex4s"); + glad_glVertex4sv = (PFNGLVERTEX4SVPROC)load("glVertex4sv"); + glad_glClipPlane = (PFNGLCLIPPLANEPROC)load("glClipPlane"); + glad_glColorMaterial = (PFNGLCOLORMATERIALPROC)load("glColorMaterial"); + glad_glFogf = (PFNGLFOGFPROC)load("glFogf"); + glad_glFogfv = (PFNGLFOGFVPROC)load("glFogfv"); + glad_glFogi = (PFNGLFOGIPROC)load("glFogi"); + glad_glFogiv = (PFNGLFOGIVPROC)load("glFogiv"); + glad_glLightf = (PFNGLLIGHTFPROC)load("glLightf"); + glad_glLightfv = (PFNGLLIGHTFVPROC)load("glLightfv"); + glad_glLighti = (PFNGLLIGHTIPROC)load("glLighti"); + glad_glLightiv = (PFNGLLIGHTIVPROC)load("glLightiv"); + glad_glLightModelf = (PFNGLLIGHTMODELFPROC)load("glLightModelf"); + glad_glLightModelfv = (PFNGLLIGHTMODELFVPROC)load("glLightModelfv"); + glad_glLightModeli = (PFNGLLIGHTMODELIPROC)load("glLightModeli"); + glad_glLightModeliv = (PFNGLLIGHTMODELIVPROC)load("glLightModeliv"); + glad_glLineStipple = (PFNGLLINESTIPPLEPROC)load("glLineStipple"); + glad_glMaterialf = (PFNGLMATERIALFPROC)load("glMaterialf"); + glad_glMaterialfv = (PFNGLMATERIALFVPROC)load("glMaterialfv"); + glad_glMateriali = (PFNGLMATERIALIPROC)load("glMateriali"); + glad_glMaterialiv = (PFNGLMATERIALIVPROC)load("glMaterialiv"); + glad_glPolygonStipple = (PFNGLPOLYGONSTIPPLEPROC)load("glPolygonStipple"); + glad_glShadeModel = (PFNGLSHADEMODELPROC)load("glShadeModel"); + glad_glTexEnvf = (PFNGLTEXENVFPROC)load("glTexEnvf"); + glad_glTexEnvfv = (PFNGLTEXENVFVPROC)load("glTexEnvfv"); + glad_glTexEnvi = (PFNGLTEXENVIPROC)load("glTexEnvi"); + glad_glTexEnviv = (PFNGLTEXENVIVPROC)load("glTexEnviv"); + glad_glTexGend = (PFNGLTEXGENDPROC)load("glTexGend"); + glad_glTexGendv = (PFNGLTEXGENDVPROC)load("glTexGendv"); + glad_glTexGenf = (PFNGLTEXGENFPROC)load("glTexGenf"); + glad_glTexGenfv = (PFNGLTEXGENFVPROC)load("glTexGenfv"); + glad_glTexGeni = (PFNGLTEXGENIPROC)load("glTexGeni"); + glad_glTexGeniv = (PFNGLTEXGENIVPROC)load("glTexGeniv"); + glad_glFeedbackBuffer = (PFNGLFEEDBACKBUFFERPROC)load("glFeedbackBuffer"); + glad_glSelectBuffer = (PFNGLSELECTBUFFERPROC)load("glSelectBuffer"); + glad_glRenderMode = (PFNGLRENDERMODEPROC)load("glRenderMode"); + glad_glInitNames = (PFNGLINITNAMESPROC)load("glInitNames"); + glad_glLoadName = (PFNGLLOADNAMEPROC)load("glLoadName"); + glad_glPassThrough = (PFNGLPASSTHROUGHPROC)load("glPassThrough"); + glad_glPopName = (PFNGLPOPNAMEPROC)load("glPopName"); + glad_glPushName = (PFNGLPUSHNAMEPROC)load("glPushName"); + glad_glClearAccum = (PFNGLCLEARACCUMPROC)load("glClearAccum"); + glad_glClearIndex = (PFNGLCLEARINDEXPROC)load("glClearIndex"); + glad_glIndexMask = (PFNGLINDEXMASKPROC)load("glIndexMask"); + glad_glAccum = (PFNGLACCUMPROC)load("glAccum"); + glad_glPopAttrib = (PFNGLPOPATTRIBPROC)load("glPopAttrib"); + glad_glPushAttrib = (PFNGLPUSHATTRIBPROC)load("glPushAttrib"); + glad_glMap1d = (PFNGLMAP1DPROC)load("glMap1d"); + glad_glMap1f = (PFNGLMAP1FPROC)load("glMap1f"); + glad_glMap2d = (PFNGLMAP2DPROC)load("glMap2d"); + glad_glMap2f = (PFNGLMAP2FPROC)load("glMap2f"); + glad_glMapGrid1d = (PFNGLMAPGRID1DPROC)load("glMapGrid1d"); + glad_glMapGrid1f = (PFNGLMAPGRID1FPROC)load("glMapGrid1f"); + glad_glMapGrid2d = (PFNGLMAPGRID2DPROC)load("glMapGrid2d"); + glad_glMapGrid2f = (PFNGLMAPGRID2FPROC)load("glMapGrid2f"); + glad_glEvalCoord1d = (PFNGLEVALCOORD1DPROC)load("glEvalCoord1d"); + glad_glEvalCoord1dv = (PFNGLEVALCOORD1DVPROC)load("glEvalCoord1dv"); + glad_glEvalCoord1f = (PFNGLEVALCOORD1FPROC)load("glEvalCoord1f"); + glad_glEvalCoord1fv = (PFNGLEVALCOORD1FVPROC)load("glEvalCoord1fv"); + glad_glEvalCoord2d = (PFNGLEVALCOORD2DPROC)load("glEvalCoord2d"); + glad_glEvalCoord2dv = (PFNGLEVALCOORD2DVPROC)load("glEvalCoord2dv"); + glad_glEvalCoord2f = (PFNGLEVALCOORD2FPROC)load("glEvalCoord2f"); + glad_glEvalCoord2fv = (PFNGLEVALCOORD2FVPROC)load("glEvalCoord2fv"); + glad_glEvalMesh1 = (PFNGLEVALMESH1PROC)load("glEvalMesh1"); + glad_glEvalPoint1 = (PFNGLEVALPOINT1PROC)load("glEvalPoint1"); + glad_glEvalMesh2 = (PFNGLEVALMESH2PROC)load("glEvalMesh2"); + glad_glEvalPoint2 = (PFNGLEVALPOINT2PROC)load("glEvalPoint2"); + glad_glAlphaFunc = (PFNGLALPHAFUNCPROC)load("glAlphaFunc"); + glad_glPixelZoom = (PFNGLPIXELZOOMPROC)load("glPixelZoom"); + glad_glPixelTransferf = (PFNGLPIXELTRANSFERFPROC)load("glPixelTransferf"); + glad_glPixelTransferi = (PFNGLPIXELTRANSFERIPROC)load("glPixelTransferi"); + glad_glPixelMapfv = (PFNGLPIXELMAPFVPROC)load("glPixelMapfv"); + glad_glPixelMapuiv = (PFNGLPIXELMAPUIVPROC)load("glPixelMapuiv"); + glad_glPixelMapusv = (PFNGLPIXELMAPUSVPROC)load("glPixelMapusv"); + glad_glCopyPixels = (PFNGLCOPYPIXELSPROC)load("glCopyPixels"); + glad_glDrawPixels = (PFNGLDRAWPIXELSPROC)load("glDrawPixels"); + glad_glGetClipPlane = (PFNGLGETCLIPPLANEPROC)load("glGetClipPlane"); + glad_glGetLightfv = (PFNGLGETLIGHTFVPROC)load("glGetLightfv"); + glad_glGetLightiv = (PFNGLGETLIGHTIVPROC)load("glGetLightiv"); + glad_glGetMapdv = (PFNGLGETMAPDVPROC)load("glGetMapdv"); + glad_glGetMapfv = (PFNGLGETMAPFVPROC)load("glGetMapfv"); + glad_glGetMapiv = (PFNGLGETMAPIVPROC)load("glGetMapiv"); + glad_glGetMaterialfv = (PFNGLGETMATERIALFVPROC)load("glGetMaterialfv"); + glad_glGetMaterialiv = (PFNGLGETMATERIALIVPROC)load("glGetMaterialiv"); + glad_glGetPixelMapfv = (PFNGLGETPIXELMAPFVPROC)load("glGetPixelMapfv"); + glad_glGetPixelMapuiv = (PFNGLGETPIXELMAPUIVPROC)load("glGetPixelMapuiv"); + glad_glGetPixelMapusv = (PFNGLGETPIXELMAPUSVPROC)load("glGetPixelMapusv"); + glad_glGetPolygonStipple = (PFNGLGETPOLYGONSTIPPLEPROC)load("glGetPolygonStipple"); + glad_glGetTexEnvfv = (PFNGLGETTEXENVFVPROC)load("glGetTexEnvfv"); + glad_glGetTexEnviv = (PFNGLGETTEXENVIVPROC)load("glGetTexEnviv"); + glad_glGetTexGendv = (PFNGLGETTEXGENDVPROC)load("glGetTexGendv"); + glad_glGetTexGenfv = (PFNGLGETTEXGENFVPROC)load("glGetTexGenfv"); + glad_glGetTexGeniv = (PFNGLGETTEXGENIVPROC)load("glGetTexGeniv"); + glad_glIsList = (PFNGLISLISTPROC)load("glIsList"); + glad_glFrustum = (PFNGLFRUSTUMPROC)load("glFrustum"); + glad_glLoadIdentity = (PFNGLLOADIDENTITYPROC)load("glLoadIdentity"); + glad_glLoadMatrixf = (PFNGLLOADMATRIXFPROC)load("glLoadMatrixf"); + glad_glLoadMatrixd = (PFNGLLOADMATRIXDPROC)load("glLoadMatrixd"); + glad_glMatrixMode = (PFNGLMATRIXMODEPROC)load("glMatrixMode"); + glad_glMultMatrixf = (PFNGLMULTMATRIXFPROC)load("glMultMatrixf"); + glad_glMultMatrixd = (PFNGLMULTMATRIXDPROC)load("glMultMatrixd"); + glad_glOrtho = (PFNGLORTHOPROC)load("glOrtho"); + glad_glPopMatrix = (PFNGLPOPMATRIXPROC)load("glPopMatrix"); + glad_glPushMatrix = (PFNGLPUSHMATRIXPROC)load("glPushMatrix"); + glad_glRotated = (PFNGLROTATEDPROC)load("glRotated"); + glad_glRotatef = (PFNGLROTATEFPROC)load("glRotatef"); + glad_glScaled = (PFNGLSCALEDPROC)load("glScaled"); + glad_glScalef = (PFNGLSCALEFPROC)load("glScalef"); + glad_glTranslated = (PFNGLTRANSLATEDPROC)load("glTranslated"); + glad_glTranslatef = (PFNGLTRANSLATEFPROC)load("glTranslatef"); +} +static void load_GL_VERSION_1_1(GLADloadproc load) { + if (!GLAD_GL_VERSION_1_1) return; + glad_glDrawArrays = (PFNGLDRAWARRAYSPROC)load("glDrawArrays"); + glad_glDrawElements = (PFNGLDRAWELEMENTSPROC)load("glDrawElements"); + glad_glGetPointerv = (PFNGLGETPOINTERVPROC)load("glGetPointerv"); + glad_glPolygonOffset = (PFNGLPOLYGONOFFSETPROC)load("glPolygonOffset"); + glad_glCopyTexImage1D = (PFNGLCOPYTEXIMAGE1DPROC)load("glCopyTexImage1D"); + glad_glCopyTexImage2D = (PFNGLCOPYTEXIMAGE2DPROC)load("glCopyTexImage2D"); + glad_glCopyTexSubImage1D = (PFNGLCOPYTEXSUBIMAGE1DPROC)load("glCopyTexSubImage1D"); + glad_glCopyTexSubImage2D = (PFNGLCOPYTEXSUBIMAGE2DPROC)load("glCopyTexSubImage2D"); + glad_glTexSubImage1D = (PFNGLTEXSUBIMAGE1DPROC)load("glTexSubImage1D"); + glad_glTexSubImage2D = (PFNGLTEXSUBIMAGE2DPROC)load("glTexSubImage2D"); + glad_glBindTexture = (PFNGLBINDTEXTUREPROC)load("glBindTexture"); + glad_glDeleteTextures = (PFNGLDELETETEXTURESPROC)load("glDeleteTextures"); + glad_glGenTextures = (PFNGLGENTEXTURESPROC)load("glGenTextures"); + glad_glIsTexture = (PFNGLISTEXTUREPROC)load("glIsTexture"); + glad_glArrayElement = (PFNGLARRAYELEMENTPROC)load("glArrayElement"); + glad_glColorPointer = (PFNGLCOLORPOINTERPROC)load("glColorPointer"); + glad_glDisableClientState = (PFNGLDISABLECLIENTSTATEPROC)load("glDisableClientState"); + glad_glEdgeFlagPointer = (PFNGLEDGEFLAGPOINTERPROC)load("glEdgeFlagPointer"); + glad_glEnableClientState = (PFNGLENABLECLIENTSTATEPROC)load("glEnableClientState"); + glad_glIndexPointer = (PFNGLINDEXPOINTERPROC)load("glIndexPointer"); + glad_glInterleavedArrays = (PFNGLINTERLEAVEDARRAYSPROC)load("glInterleavedArrays"); + glad_glNormalPointer = (PFNGLNORMALPOINTERPROC)load("glNormalPointer"); + glad_glTexCoordPointer = (PFNGLTEXCOORDPOINTERPROC)load("glTexCoordPointer"); + glad_glVertexPointer = (PFNGLVERTEXPOINTERPROC)load("glVertexPointer"); + glad_glAreTexturesResident = (PFNGLARETEXTURESRESIDENTPROC)load("glAreTexturesResident"); + glad_glPrioritizeTextures = (PFNGLPRIORITIZETEXTURESPROC)load("glPrioritizeTextures"); + glad_glIndexub = (PFNGLINDEXUBPROC)load("glIndexub"); + glad_glIndexubv = (PFNGLINDEXUBVPROC)load("glIndexubv"); + glad_glPopClientAttrib = (PFNGLPOPCLIENTATTRIBPROC)load("glPopClientAttrib"); + glad_glPushClientAttrib = (PFNGLPUSHCLIENTATTRIBPROC)load("glPushClientAttrib"); +} +static void load_GL_VERSION_1_2(GLADloadproc load) { + if (!GLAD_GL_VERSION_1_2) return; + glad_glDrawRangeElements = (PFNGLDRAWRANGEELEMENTSPROC)load("glDrawRangeElements"); + glad_glTexImage3D = (PFNGLTEXIMAGE3DPROC)load("glTexImage3D"); + glad_glTexSubImage3D = (PFNGLTEXSUBIMAGE3DPROC)load("glTexSubImage3D"); + glad_glCopyTexSubImage3D = (PFNGLCOPYTEXSUBIMAGE3DPROC)load("glCopyTexSubImage3D"); +} +static void load_GL_VERSION_1_3(GLADloadproc load) { + if (!GLAD_GL_VERSION_1_3) return; + glad_glActiveTexture = (PFNGLACTIVETEXTUREPROC)load("glActiveTexture"); + glad_glSampleCoverage = (PFNGLSAMPLECOVERAGEPROC)load("glSampleCoverage"); + glad_glCompressedTexImage3D = (PFNGLCOMPRESSEDTEXIMAGE3DPROC)load("glCompressedTexImage3D"); + glad_glCompressedTexImage2D = (PFNGLCOMPRESSEDTEXIMAGE2DPROC)load("glCompressedTexImage2D"); + glad_glCompressedTexImage1D = (PFNGLCOMPRESSEDTEXIMAGE1DPROC)load("glCompressedTexImage1D"); + glad_glCompressedTexSubImage3D = (PFNGLCOMPRESSEDTEXSUBIMAGE3DPROC)load("glCompressedTexSubImage3D"); + glad_glCompressedTexSubImage2D = (PFNGLCOMPRESSEDTEXSUBIMAGE2DPROC)load("glCompressedTexSubImage2D"); + glad_glCompressedTexSubImage1D = (PFNGLCOMPRESSEDTEXSUBIMAGE1DPROC)load("glCompressedTexSubImage1D"); + glad_glGetCompressedTexImage = (PFNGLGETCOMPRESSEDTEXIMAGEPROC)load("glGetCompressedTexImage"); + glad_glClientActiveTexture = (PFNGLCLIENTACTIVETEXTUREPROC)load("glClientActiveTexture"); + glad_glMultiTexCoord1d = (PFNGLMULTITEXCOORD1DPROC)load("glMultiTexCoord1d"); + glad_glMultiTexCoord1dv = (PFNGLMULTITEXCOORD1DVPROC)load("glMultiTexCoord1dv"); + glad_glMultiTexCoord1f = (PFNGLMULTITEXCOORD1FPROC)load("glMultiTexCoord1f"); + glad_glMultiTexCoord1fv = (PFNGLMULTITEXCOORD1FVPROC)load("glMultiTexCoord1fv"); + glad_glMultiTexCoord1i = (PFNGLMULTITEXCOORD1IPROC)load("glMultiTexCoord1i"); + glad_glMultiTexCoord1iv = (PFNGLMULTITEXCOORD1IVPROC)load("glMultiTexCoord1iv"); + glad_glMultiTexCoord1s = (PFNGLMULTITEXCOORD1SPROC)load("glMultiTexCoord1s"); + glad_glMultiTexCoord1sv = (PFNGLMULTITEXCOORD1SVPROC)load("glMultiTexCoord1sv"); + glad_glMultiTexCoord2d = (PFNGLMULTITEXCOORD2DPROC)load("glMultiTexCoord2d"); + glad_glMultiTexCoord2dv = (PFNGLMULTITEXCOORD2DVPROC)load("glMultiTexCoord2dv"); + glad_glMultiTexCoord2f = (PFNGLMULTITEXCOORD2FPROC)load("glMultiTexCoord2f"); + glad_glMultiTexCoord2fv = (PFNGLMULTITEXCOORD2FVPROC)load("glMultiTexCoord2fv"); + glad_glMultiTexCoord2i = (PFNGLMULTITEXCOORD2IPROC)load("glMultiTexCoord2i"); + glad_glMultiTexCoord2iv = (PFNGLMULTITEXCOORD2IVPROC)load("glMultiTexCoord2iv"); + glad_glMultiTexCoord2s = (PFNGLMULTITEXCOORD2SPROC)load("glMultiTexCoord2s"); + glad_glMultiTexCoord2sv = (PFNGLMULTITEXCOORD2SVPROC)load("glMultiTexCoord2sv"); + glad_glMultiTexCoord3d = (PFNGLMULTITEXCOORD3DPROC)load("glMultiTexCoord3d"); + glad_glMultiTexCoord3dv = (PFNGLMULTITEXCOORD3DVPROC)load("glMultiTexCoord3dv"); + glad_glMultiTexCoord3f = (PFNGLMULTITEXCOORD3FPROC)load("glMultiTexCoord3f"); + glad_glMultiTexCoord3fv = (PFNGLMULTITEXCOORD3FVPROC)load("glMultiTexCoord3fv"); + glad_glMultiTexCoord3i = (PFNGLMULTITEXCOORD3IPROC)load("glMultiTexCoord3i"); + glad_glMultiTexCoord3iv = (PFNGLMULTITEXCOORD3IVPROC)load("glMultiTexCoord3iv"); + glad_glMultiTexCoord3s = (PFNGLMULTITEXCOORD3SPROC)load("glMultiTexCoord3s"); + glad_glMultiTexCoord3sv = (PFNGLMULTITEXCOORD3SVPROC)load("glMultiTexCoord3sv"); + glad_glMultiTexCoord4d = (PFNGLMULTITEXCOORD4DPROC)load("glMultiTexCoord4d"); + glad_glMultiTexCoord4dv = (PFNGLMULTITEXCOORD4DVPROC)load("glMultiTexCoord4dv"); + glad_glMultiTexCoord4f = (PFNGLMULTITEXCOORD4FPROC)load("glMultiTexCoord4f"); + glad_glMultiTexCoord4fv = (PFNGLMULTITEXCOORD4FVPROC)load("glMultiTexCoord4fv"); + glad_glMultiTexCoord4i = (PFNGLMULTITEXCOORD4IPROC)load("glMultiTexCoord4i"); + glad_glMultiTexCoord4iv = (PFNGLMULTITEXCOORD4IVPROC)load("glMultiTexCoord4iv"); + glad_glMultiTexCoord4s = (PFNGLMULTITEXCOORD4SPROC)load("glMultiTexCoord4s"); + glad_glMultiTexCoord4sv = (PFNGLMULTITEXCOORD4SVPROC)load("glMultiTexCoord4sv"); + glad_glLoadTransposeMatrixf = (PFNGLLOADTRANSPOSEMATRIXFPROC)load("glLoadTransposeMatrixf"); + glad_glLoadTransposeMatrixd = (PFNGLLOADTRANSPOSEMATRIXDPROC)load("glLoadTransposeMatrixd"); + glad_glMultTransposeMatrixf = (PFNGLMULTTRANSPOSEMATRIXFPROC)load("glMultTransposeMatrixf"); + glad_glMultTransposeMatrixd = (PFNGLMULTTRANSPOSEMATRIXDPROC)load("glMultTransposeMatrixd"); +} +static void load_GL_VERSION_1_4(GLADloadproc load) { + if (!GLAD_GL_VERSION_1_4) return; + glad_glBlendFuncSeparate = (PFNGLBLENDFUNCSEPARATEPROC)load("glBlendFuncSeparate"); + glad_glMultiDrawArrays = (PFNGLMULTIDRAWARRAYSPROC)load("glMultiDrawArrays"); + glad_glMultiDrawElements = (PFNGLMULTIDRAWELEMENTSPROC)load("glMultiDrawElements"); + glad_glPointParameterf = (PFNGLPOINTPARAMETERFPROC)load("glPointParameterf"); + glad_glPointParameterfv = (PFNGLPOINTPARAMETERFVPROC)load("glPointParameterfv"); + glad_glPointParameteri = (PFNGLPOINTPARAMETERIPROC)load("glPointParameteri"); + glad_glPointParameteriv = (PFNGLPOINTPARAMETERIVPROC)load("glPointParameteriv"); + glad_glFogCoordf = (PFNGLFOGCOORDFPROC)load("glFogCoordf"); + glad_glFogCoordfv = (PFNGLFOGCOORDFVPROC)load("glFogCoordfv"); + glad_glFogCoordd = (PFNGLFOGCOORDDPROC)load("glFogCoordd"); + glad_glFogCoorddv = (PFNGLFOGCOORDDVPROC)load("glFogCoorddv"); + glad_glFogCoordPointer = (PFNGLFOGCOORDPOINTERPROC)load("glFogCoordPointer"); + glad_glSecondaryColor3b = (PFNGLSECONDARYCOLOR3BPROC)load("glSecondaryColor3b"); + glad_glSecondaryColor3bv = (PFNGLSECONDARYCOLOR3BVPROC)load("glSecondaryColor3bv"); + glad_glSecondaryColor3d = (PFNGLSECONDARYCOLOR3DPROC)load("glSecondaryColor3d"); + glad_glSecondaryColor3dv = (PFNGLSECONDARYCOLOR3DVPROC)load("glSecondaryColor3dv"); + glad_glSecondaryColor3f = (PFNGLSECONDARYCOLOR3FPROC)load("glSecondaryColor3f"); + glad_glSecondaryColor3fv = (PFNGLSECONDARYCOLOR3FVPROC)load("glSecondaryColor3fv"); + glad_glSecondaryColor3i = (PFNGLSECONDARYCOLOR3IPROC)load("glSecondaryColor3i"); + glad_glSecondaryColor3iv = (PFNGLSECONDARYCOLOR3IVPROC)load("glSecondaryColor3iv"); + glad_glSecondaryColor3s = (PFNGLSECONDARYCOLOR3SPROC)load("glSecondaryColor3s"); + glad_glSecondaryColor3sv = (PFNGLSECONDARYCOLOR3SVPROC)load("glSecondaryColor3sv"); + glad_glSecondaryColor3ub = (PFNGLSECONDARYCOLOR3UBPROC)load("glSecondaryColor3ub"); + glad_glSecondaryColor3ubv = (PFNGLSECONDARYCOLOR3UBVPROC)load("glSecondaryColor3ubv"); + glad_glSecondaryColor3ui = (PFNGLSECONDARYCOLOR3UIPROC)load("glSecondaryColor3ui"); + glad_glSecondaryColor3uiv = (PFNGLSECONDARYCOLOR3UIVPROC)load("glSecondaryColor3uiv"); + glad_glSecondaryColor3us = (PFNGLSECONDARYCOLOR3USPROC)load("glSecondaryColor3us"); + glad_glSecondaryColor3usv = (PFNGLSECONDARYCOLOR3USVPROC)load("glSecondaryColor3usv"); + glad_glSecondaryColorPointer = (PFNGLSECONDARYCOLORPOINTERPROC)load("glSecondaryColorPointer"); + glad_glWindowPos2d = (PFNGLWINDOWPOS2DPROC)load("glWindowPos2d"); + glad_glWindowPos2dv = (PFNGLWINDOWPOS2DVPROC)load("glWindowPos2dv"); + glad_glWindowPos2f = (PFNGLWINDOWPOS2FPROC)load("glWindowPos2f"); + glad_glWindowPos2fv = (PFNGLWINDOWPOS2FVPROC)load("glWindowPos2fv"); + glad_glWindowPos2i = (PFNGLWINDOWPOS2IPROC)load("glWindowPos2i"); + glad_glWindowPos2iv = (PFNGLWINDOWPOS2IVPROC)load("glWindowPos2iv"); + glad_glWindowPos2s = (PFNGLWINDOWPOS2SPROC)load("glWindowPos2s"); + glad_glWindowPos2sv = (PFNGLWINDOWPOS2SVPROC)load("glWindowPos2sv"); + glad_glWindowPos3d = (PFNGLWINDOWPOS3DPROC)load("glWindowPos3d"); + glad_glWindowPos3dv = (PFNGLWINDOWPOS3DVPROC)load("glWindowPos3dv"); + glad_glWindowPos3f = (PFNGLWINDOWPOS3FPROC)load("glWindowPos3f"); + glad_glWindowPos3fv = (PFNGLWINDOWPOS3FVPROC)load("glWindowPos3fv"); + glad_glWindowPos3i = (PFNGLWINDOWPOS3IPROC)load("glWindowPos3i"); + glad_glWindowPos3iv = (PFNGLWINDOWPOS3IVPROC)load("glWindowPos3iv"); + glad_glWindowPos3s = (PFNGLWINDOWPOS3SPROC)load("glWindowPos3s"); + glad_glWindowPos3sv = (PFNGLWINDOWPOS3SVPROC)load("glWindowPos3sv"); + glad_glBlendColor = (PFNGLBLENDCOLORPROC)load("glBlendColor"); + glad_glBlendEquation = (PFNGLBLENDEQUATIONPROC)load("glBlendEquation"); +} +static void load_GL_VERSION_1_5(GLADloadproc load) { + if (!GLAD_GL_VERSION_1_5) return; + glad_glGenQueries = (PFNGLGENQUERIESPROC)load("glGenQueries"); + glad_glDeleteQueries = (PFNGLDELETEQUERIESPROC)load("glDeleteQueries"); + glad_glIsQuery = (PFNGLISQUERYPROC)load("glIsQuery"); + glad_glBeginQuery = (PFNGLBEGINQUERYPROC)load("glBeginQuery"); + glad_glEndQuery = (PFNGLENDQUERYPROC)load("glEndQuery"); + glad_glGetQueryiv = (PFNGLGETQUERYIVPROC)load("glGetQueryiv"); + glad_glGetQueryObjectiv = (PFNGLGETQUERYOBJECTIVPROC)load("glGetQueryObjectiv"); + glad_glGetQueryObjectuiv = (PFNGLGETQUERYOBJECTUIVPROC)load("glGetQueryObjectuiv"); + glad_glBindBuffer = (PFNGLBINDBUFFERPROC)load("glBindBuffer"); + glad_glDeleteBuffers = (PFNGLDELETEBUFFERSPROC)load("glDeleteBuffers"); + glad_glGenBuffers = (PFNGLGENBUFFERSPROC)load("glGenBuffers"); + glad_glIsBuffer = (PFNGLISBUFFERPROC)load("glIsBuffer"); + glad_glBufferData = (PFNGLBUFFERDATAPROC)load("glBufferData"); + glad_glBufferSubData = (PFNGLBUFFERSUBDATAPROC)load("glBufferSubData"); + glad_glGetBufferSubData = (PFNGLGETBUFFERSUBDATAPROC)load("glGetBufferSubData"); + glad_glMapBuffer = (PFNGLMAPBUFFERPROC)load("glMapBuffer"); + glad_glUnmapBuffer = (PFNGLUNMAPBUFFERPROC)load("glUnmapBuffer"); + glad_glGetBufferParameteriv = (PFNGLGETBUFFERPARAMETERIVPROC)load("glGetBufferParameteriv"); + glad_glGetBufferPointerv = (PFNGLGETBUFFERPOINTERVPROC)load("glGetBufferPointerv"); +} +static void load_GL_VERSION_2_0(GLADloadproc load) { + if (!GLAD_GL_VERSION_2_0) return; + glad_glBlendEquationSeparate = (PFNGLBLENDEQUATIONSEPARATEPROC)load("glBlendEquationSeparate"); + glad_glDrawBuffers = (PFNGLDRAWBUFFERSPROC)load("glDrawBuffers"); + glad_glStencilOpSeparate = (PFNGLSTENCILOPSEPARATEPROC)load("glStencilOpSeparate"); + glad_glStencilFuncSeparate = (PFNGLSTENCILFUNCSEPARATEPROC)load("glStencilFuncSeparate"); + glad_glStencilMaskSeparate = (PFNGLSTENCILMASKSEPARATEPROC)load("glStencilMaskSeparate"); + glad_glAttachShader = (PFNGLATTACHSHADERPROC)load("glAttachShader"); + glad_glBindAttribLocation = (PFNGLBINDATTRIBLOCATIONPROC)load("glBindAttribLocation"); + glad_glCompileShader = (PFNGLCOMPILESHADERPROC)load("glCompileShader"); + glad_glCreateProgram = (PFNGLCREATEPROGRAMPROC)load("glCreateProgram"); + glad_glCreateShader = (PFNGLCREATESHADERPROC)load("glCreateShader"); + glad_glDeleteProgram = (PFNGLDELETEPROGRAMPROC)load("glDeleteProgram"); + glad_glDeleteShader = (PFNGLDELETESHADERPROC)load("glDeleteShader"); + glad_glDetachShader = (PFNGLDETACHSHADERPROC)load("glDetachShader"); + glad_glDisableVertexAttribArray = (PFNGLDISABLEVERTEXATTRIBARRAYPROC)load("glDisableVertexAttribArray"); + glad_glEnableVertexAttribArray = (PFNGLENABLEVERTEXATTRIBARRAYPROC)load("glEnableVertexAttribArray"); + glad_glGetActiveAttrib = (PFNGLGETACTIVEATTRIBPROC)load("glGetActiveAttrib"); + glad_glGetActiveUniform = (PFNGLGETACTIVEUNIFORMPROC)load("glGetActiveUniform"); + glad_glGetAttachedShaders = (PFNGLGETATTACHEDSHADERSPROC)load("glGetAttachedShaders"); + glad_glGetAttribLocation = (PFNGLGETATTRIBLOCATIONPROC)load("glGetAttribLocation"); + glad_glGetProgramiv = (PFNGLGETPROGRAMIVPROC)load("glGetProgramiv"); + glad_glGetProgramInfoLog = (PFNGLGETPROGRAMINFOLOGPROC)load("glGetProgramInfoLog"); + glad_glGetShaderiv = (PFNGLGETSHADERIVPROC)load("glGetShaderiv"); + glad_glGetShaderInfoLog = (PFNGLGETSHADERINFOLOGPROC)load("glGetShaderInfoLog"); + glad_glGetShaderSource = (PFNGLGETSHADERSOURCEPROC)load("glGetShaderSource"); + glad_glGetUniformLocation = (PFNGLGETUNIFORMLOCATIONPROC)load("glGetUniformLocation"); + glad_glGetUniformfv = (PFNGLGETUNIFORMFVPROC)load("glGetUniformfv"); + glad_glGetUniformiv = (PFNGLGETUNIFORMIVPROC)load("glGetUniformiv"); + glad_glGetVertexAttribdv = (PFNGLGETVERTEXATTRIBDVPROC)load("glGetVertexAttribdv"); + glad_glGetVertexAttribfv = (PFNGLGETVERTEXATTRIBFVPROC)load("glGetVertexAttribfv"); + glad_glGetVertexAttribiv = (PFNGLGETVERTEXATTRIBIVPROC)load("glGetVertexAttribiv"); + glad_glGetVertexAttribPointerv = (PFNGLGETVERTEXATTRIBPOINTERVPROC)load("glGetVertexAttribPointerv"); + glad_glIsProgram = (PFNGLISPROGRAMPROC)load("glIsProgram"); + glad_glIsShader = (PFNGLISSHADERPROC)load("glIsShader"); + glad_glLinkProgram = (PFNGLLINKPROGRAMPROC)load("glLinkProgram"); + glad_glShaderSource = (PFNGLSHADERSOURCEPROC)load("glShaderSource"); + glad_glUseProgram = (PFNGLUSEPROGRAMPROC)load("glUseProgram"); + glad_glUniform1f = (PFNGLUNIFORM1FPROC)load("glUniform1f"); + glad_glUniform2f = (PFNGLUNIFORM2FPROC)load("glUniform2f"); + glad_glUniform3f = (PFNGLUNIFORM3FPROC)load("glUniform3f"); + glad_glUniform4f = (PFNGLUNIFORM4FPROC)load("glUniform4f"); + glad_glUniform1i = (PFNGLUNIFORM1IPROC)load("glUniform1i"); + glad_glUniform2i = (PFNGLUNIFORM2IPROC)load("glUniform2i"); + glad_glUniform3i = (PFNGLUNIFORM3IPROC)load("glUniform3i"); + glad_glUniform4i = (PFNGLUNIFORM4IPROC)load("glUniform4i"); + glad_glUniform1fv = (PFNGLUNIFORM1FVPROC)load("glUniform1fv"); + glad_glUniform2fv = (PFNGLUNIFORM2FVPROC)load("glUniform2fv"); + glad_glUniform3fv = (PFNGLUNIFORM3FVPROC)load("glUniform3fv"); + glad_glUniform4fv = (PFNGLUNIFORM4FVPROC)load("glUniform4fv"); + glad_glUniform1iv = (PFNGLUNIFORM1IVPROC)load("glUniform1iv"); + glad_glUniform2iv = (PFNGLUNIFORM2IVPROC)load("glUniform2iv"); + glad_glUniform3iv = (PFNGLUNIFORM3IVPROC)load("glUniform3iv"); + glad_glUniform4iv = (PFNGLUNIFORM4IVPROC)load("glUniform4iv"); + glad_glUniformMatrix2fv = (PFNGLUNIFORMMATRIX2FVPROC)load("glUniformMatrix2fv"); + glad_glUniformMatrix3fv = (PFNGLUNIFORMMATRIX3FVPROC)load("glUniformMatrix3fv"); + glad_glUniformMatrix4fv = (PFNGLUNIFORMMATRIX4FVPROC)load("glUniformMatrix4fv"); + glad_glValidateProgram = (PFNGLVALIDATEPROGRAMPROC)load("glValidateProgram"); + glad_glVertexAttrib1d = (PFNGLVERTEXATTRIB1DPROC)load("glVertexAttrib1d"); + glad_glVertexAttrib1dv = (PFNGLVERTEXATTRIB1DVPROC)load("glVertexAttrib1dv"); + glad_glVertexAttrib1f = (PFNGLVERTEXATTRIB1FPROC)load("glVertexAttrib1f"); + glad_glVertexAttrib1fv = (PFNGLVERTEXATTRIB1FVPROC)load("glVertexAttrib1fv"); + glad_glVertexAttrib1s = (PFNGLVERTEXATTRIB1SPROC)load("glVertexAttrib1s"); + glad_glVertexAttrib1sv = (PFNGLVERTEXATTRIB1SVPROC)load("glVertexAttrib1sv"); + glad_glVertexAttrib2d = (PFNGLVERTEXATTRIB2DPROC)load("glVertexAttrib2d"); + glad_glVertexAttrib2dv = (PFNGLVERTEXATTRIB2DVPROC)load("glVertexAttrib2dv"); + glad_glVertexAttrib2f = (PFNGLVERTEXATTRIB2FPROC)load("glVertexAttrib2f"); + glad_glVertexAttrib2fv = (PFNGLVERTEXATTRIB2FVPROC)load("glVertexAttrib2fv"); + glad_glVertexAttrib2s = (PFNGLVERTEXATTRIB2SPROC)load("glVertexAttrib2s"); + glad_glVertexAttrib2sv = (PFNGLVERTEXATTRIB2SVPROC)load("glVertexAttrib2sv"); + glad_glVertexAttrib3d = (PFNGLVERTEXATTRIB3DPROC)load("glVertexAttrib3d"); + glad_glVertexAttrib3dv = (PFNGLVERTEXATTRIB3DVPROC)load("glVertexAttrib3dv"); + glad_glVertexAttrib3f = (PFNGLVERTEXATTRIB3FPROC)load("glVertexAttrib3f"); + glad_glVertexAttrib3fv = (PFNGLVERTEXATTRIB3FVPROC)load("glVertexAttrib3fv"); + glad_glVertexAttrib3s = (PFNGLVERTEXATTRIB3SPROC)load("glVertexAttrib3s"); + glad_glVertexAttrib3sv = (PFNGLVERTEXATTRIB3SVPROC)load("glVertexAttrib3sv"); + glad_glVertexAttrib4Nbv = (PFNGLVERTEXATTRIB4NBVPROC)load("glVertexAttrib4Nbv"); + glad_glVertexAttrib4Niv = (PFNGLVERTEXATTRIB4NIVPROC)load("glVertexAttrib4Niv"); + glad_glVertexAttrib4Nsv = (PFNGLVERTEXATTRIB4NSVPROC)load("glVertexAttrib4Nsv"); + glad_glVertexAttrib4Nub = (PFNGLVERTEXATTRIB4NUBPROC)load("glVertexAttrib4Nub"); + glad_glVertexAttrib4Nubv = (PFNGLVERTEXATTRIB4NUBVPROC)load("glVertexAttrib4Nubv"); + glad_glVertexAttrib4Nuiv = (PFNGLVERTEXATTRIB4NUIVPROC)load("glVertexAttrib4Nuiv"); + glad_glVertexAttrib4Nusv = (PFNGLVERTEXATTRIB4NUSVPROC)load("glVertexAttrib4Nusv"); + glad_glVertexAttrib4bv = (PFNGLVERTEXATTRIB4BVPROC)load("glVertexAttrib4bv"); + glad_glVertexAttrib4d = (PFNGLVERTEXATTRIB4DPROC)load("glVertexAttrib4d"); + glad_glVertexAttrib4dv = (PFNGLVERTEXATTRIB4DVPROC)load("glVertexAttrib4dv"); + glad_glVertexAttrib4f = (PFNGLVERTEXATTRIB4FPROC)load("glVertexAttrib4f"); + glad_glVertexAttrib4fv = (PFNGLVERTEXATTRIB4FVPROC)load("glVertexAttrib4fv"); + glad_glVertexAttrib4iv = (PFNGLVERTEXATTRIB4IVPROC)load("glVertexAttrib4iv"); + glad_glVertexAttrib4s = (PFNGLVERTEXATTRIB4SPROC)load("glVertexAttrib4s"); + glad_glVertexAttrib4sv = (PFNGLVERTEXATTRIB4SVPROC)load("glVertexAttrib4sv"); + glad_glVertexAttrib4ubv = (PFNGLVERTEXATTRIB4UBVPROC)load("glVertexAttrib4ubv"); + glad_glVertexAttrib4uiv = (PFNGLVERTEXATTRIB4UIVPROC)load("glVertexAttrib4uiv"); + glad_glVertexAttrib4usv = (PFNGLVERTEXATTRIB4USVPROC)load("glVertexAttrib4usv"); + glad_glVertexAttribPointer = (PFNGLVERTEXATTRIBPOINTERPROC)load("glVertexAttribPointer"); +} +static void load_GL_VERSION_2_1(GLADloadproc load) { + if (!GLAD_GL_VERSION_2_1) return; + glad_glUniformMatrix2x3fv = (PFNGLUNIFORMMATRIX2X3FVPROC)load("glUniformMatrix2x3fv"); + glad_glUniformMatrix3x2fv = (PFNGLUNIFORMMATRIX3X2FVPROC)load("glUniformMatrix3x2fv"); + glad_glUniformMatrix2x4fv = (PFNGLUNIFORMMATRIX2X4FVPROC)load("glUniformMatrix2x4fv"); + glad_glUniformMatrix4x2fv = (PFNGLUNIFORMMATRIX4X2FVPROC)load("glUniformMatrix4x2fv"); + glad_glUniformMatrix3x4fv = (PFNGLUNIFORMMATRIX3X4FVPROC)load("glUniformMatrix3x4fv"); + glad_glUniformMatrix4x3fv = (PFNGLUNIFORMMATRIX4X3FVPROC)load("glUniformMatrix4x3fv"); +} +static void load_GL_VERSION_3_0(GLADloadproc load) { + if (!GLAD_GL_VERSION_3_0) return; + glad_glColorMaski = (PFNGLCOLORMASKIPROC)load("glColorMaski"); + glad_glGetBooleani_v = (PFNGLGETBOOLEANI_VPROC)load("glGetBooleani_v"); + glad_glGetIntegeri_v = (PFNGLGETINTEGERI_VPROC)load("glGetIntegeri_v"); + glad_glEnablei = (PFNGLENABLEIPROC)load("glEnablei"); + glad_glDisablei = (PFNGLDISABLEIPROC)load("glDisablei"); + glad_glIsEnabledi = (PFNGLISENABLEDIPROC)load("glIsEnabledi"); + glad_glBeginTransformFeedback = (PFNGLBEGINTRANSFORMFEEDBACKPROC)load("glBeginTransformFeedback"); + glad_glEndTransformFeedback = (PFNGLENDTRANSFORMFEEDBACKPROC)load("glEndTransformFeedback"); + glad_glBindBufferRange = (PFNGLBINDBUFFERRANGEPROC)load("glBindBufferRange"); + glad_glBindBufferBase = (PFNGLBINDBUFFERBASEPROC)load("glBindBufferBase"); + glad_glTransformFeedbackVaryings = (PFNGLTRANSFORMFEEDBACKVARYINGSPROC)load("glTransformFeedbackVaryings"); + glad_glGetTransformFeedbackVarying = (PFNGLGETTRANSFORMFEEDBACKVARYINGPROC)load("glGetTransformFeedbackVarying"); + glad_glClampColor = (PFNGLCLAMPCOLORPROC)load("glClampColor"); + glad_glBeginConditionalRender = (PFNGLBEGINCONDITIONALRENDERPROC)load("glBeginConditionalRender"); + glad_glEndConditionalRender = (PFNGLENDCONDITIONALRENDERPROC)load("glEndConditionalRender"); + glad_glVertexAttribIPointer = (PFNGLVERTEXATTRIBIPOINTERPROC)load("glVertexAttribIPointer"); + glad_glGetVertexAttribIiv = (PFNGLGETVERTEXATTRIBIIVPROC)load("glGetVertexAttribIiv"); + glad_glGetVertexAttribIuiv = (PFNGLGETVERTEXATTRIBIUIVPROC)load("glGetVertexAttribIuiv"); + glad_glVertexAttribI1i = (PFNGLVERTEXATTRIBI1IPROC)load("glVertexAttribI1i"); + glad_glVertexAttribI2i = (PFNGLVERTEXATTRIBI2IPROC)load("glVertexAttribI2i"); + glad_glVertexAttribI3i = (PFNGLVERTEXATTRIBI3IPROC)load("glVertexAttribI3i"); + glad_glVertexAttribI4i = (PFNGLVERTEXATTRIBI4IPROC)load("glVertexAttribI4i"); + glad_glVertexAttribI1ui = (PFNGLVERTEXATTRIBI1UIPROC)load("glVertexAttribI1ui"); + glad_glVertexAttribI2ui = (PFNGLVERTEXATTRIBI2UIPROC)load("glVertexAttribI2ui"); + glad_glVertexAttribI3ui = (PFNGLVERTEXATTRIBI3UIPROC)load("glVertexAttribI3ui"); + glad_glVertexAttribI4ui = (PFNGLVERTEXATTRIBI4UIPROC)load("glVertexAttribI4ui"); + glad_glVertexAttribI1iv = (PFNGLVERTEXATTRIBI1IVPROC)load("glVertexAttribI1iv"); + glad_glVertexAttribI2iv = (PFNGLVERTEXATTRIBI2IVPROC)load("glVertexAttribI2iv"); + glad_glVertexAttribI3iv = (PFNGLVERTEXATTRIBI3IVPROC)load("glVertexAttribI3iv"); + glad_glVertexAttribI4iv = (PFNGLVERTEXATTRIBI4IVPROC)load("glVertexAttribI4iv"); + glad_glVertexAttribI1uiv = (PFNGLVERTEXATTRIBI1UIVPROC)load("glVertexAttribI1uiv"); + glad_glVertexAttribI2uiv = (PFNGLVERTEXATTRIBI2UIVPROC)load("glVertexAttribI2uiv"); + glad_glVertexAttribI3uiv = (PFNGLVERTEXATTRIBI3UIVPROC)load("glVertexAttribI3uiv"); + glad_glVertexAttribI4uiv = (PFNGLVERTEXATTRIBI4UIVPROC)load("glVertexAttribI4uiv"); + glad_glVertexAttribI4bv = (PFNGLVERTEXATTRIBI4BVPROC)load("glVertexAttribI4bv"); + glad_glVertexAttribI4sv = (PFNGLVERTEXATTRIBI4SVPROC)load("glVertexAttribI4sv"); + glad_glVertexAttribI4ubv = (PFNGLVERTEXATTRIBI4UBVPROC)load("glVertexAttribI4ubv"); + glad_glVertexAttribI4usv = (PFNGLVERTEXATTRIBI4USVPROC)load("glVertexAttribI4usv"); + glad_glGetUniformuiv = (PFNGLGETUNIFORMUIVPROC)load("glGetUniformuiv"); + glad_glBindFragDataLocation = (PFNGLBINDFRAGDATALOCATIONPROC)load("glBindFragDataLocation"); + glad_glGetFragDataLocation = (PFNGLGETFRAGDATALOCATIONPROC)load("glGetFragDataLocation"); + glad_glUniform1ui = (PFNGLUNIFORM1UIPROC)load("glUniform1ui"); + glad_glUniform2ui = (PFNGLUNIFORM2UIPROC)load("glUniform2ui"); + glad_glUniform3ui = (PFNGLUNIFORM3UIPROC)load("glUniform3ui"); + glad_glUniform4ui = (PFNGLUNIFORM4UIPROC)load("glUniform4ui"); + glad_glUniform1uiv = (PFNGLUNIFORM1UIVPROC)load("glUniform1uiv"); + glad_glUniform2uiv = (PFNGLUNIFORM2UIVPROC)load("glUniform2uiv"); + glad_glUniform3uiv = (PFNGLUNIFORM3UIVPROC)load("glUniform3uiv"); + glad_glUniform4uiv = (PFNGLUNIFORM4UIVPROC)load("glUniform4uiv"); + glad_glTexParameterIiv = (PFNGLTEXPARAMETERIIVPROC)load("glTexParameterIiv"); + glad_glTexParameterIuiv = (PFNGLTEXPARAMETERIUIVPROC)load("glTexParameterIuiv"); + glad_glGetTexParameterIiv = (PFNGLGETTEXPARAMETERIIVPROC)load("glGetTexParameterIiv"); + glad_glGetTexParameterIuiv = (PFNGLGETTEXPARAMETERIUIVPROC)load("glGetTexParameterIuiv"); + glad_glClearBufferiv = (PFNGLCLEARBUFFERIVPROC)load("glClearBufferiv"); + glad_glClearBufferuiv = (PFNGLCLEARBUFFERUIVPROC)load("glClearBufferuiv"); + glad_glClearBufferfv = (PFNGLCLEARBUFFERFVPROC)load("glClearBufferfv"); + glad_glClearBufferfi = (PFNGLCLEARBUFFERFIPROC)load("glClearBufferfi"); + glad_glGetStringi = (PFNGLGETSTRINGIPROC)load("glGetStringi"); + glad_glIsRenderbuffer = (PFNGLISRENDERBUFFERPROC)load("glIsRenderbuffer"); + glad_glBindRenderbuffer = (PFNGLBINDRENDERBUFFERPROC)load("glBindRenderbuffer"); + glad_glDeleteRenderbuffers = (PFNGLDELETERENDERBUFFERSPROC)load("glDeleteRenderbuffers"); + glad_glGenRenderbuffers = (PFNGLGENRENDERBUFFERSPROC)load("glGenRenderbuffers"); + glad_glRenderbufferStorage = (PFNGLRENDERBUFFERSTORAGEPROC)load("glRenderbufferStorage"); + glad_glGetRenderbufferParameteriv = (PFNGLGETRENDERBUFFERPARAMETERIVPROC)load("glGetRenderbufferParameteriv"); + glad_glIsFramebuffer = (PFNGLISFRAMEBUFFERPROC)load("glIsFramebuffer"); + glad_glBindFramebuffer = (PFNGLBINDFRAMEBUFFERPROC)load("glBindFramebuffer"); + glad_glDeleteFramebuffers = (PFNGLDELETEFRAMEBUFFERSPROC)load("glDeleteFramebuffers"); + glad_glGenFramebuffers = (PFNGLGENFRAMEBUFFERSPROC)load("glGenFramebuffers"); + glad_glCheckFramebufferStatus = (PFNGLCHECKFRAMEBUFFERSTATUSPROC)load("glCheckFramebufferStatus"); + glad_glFramebufferTexture1D = (PFNGLFRAMEBUFFERTEXTURE1DPROC)load("glFramebufferTexture1D"); + glad_glFramebufferTexture2D = (PFNGLFRAMEBUFFERTEXTURE2DPROC)load("glFramebufferTexture2D"); + glad_glFramebufferTexture3D = (PFNGLFRAMEBUFFERTEXTURE3DPROC)load("glFramebufferTexture3D"); + glad_glFramebufferRenderbuffer = (PFNGLFRAMEBUFFERRENDERBUFFERPROC)load("glFramebufferRenderbuffer"); + glad_glGetFramebufferAttachmentParameteriv = (PFNGLGETFRAMEBUFFERATTACHMENTPARAMETERIVPROC)load("glGetFramebufferAttachmentParameteriv"); + glad_glGenerateMipmap = (PFNGLGENERATEMIPMAPPROC)load("glGenerateMipmap"); + glad_glBlitFramebuffer = (PFNGLBLITFRAMEBUFFERPROC)load("glBlitFramebuffer"); + glad_glRenderbufferStorageMultisample = (PFNGLRENDERBUFFERSTORAGEMULTISAMPLEPROC)load("glRenderbufferStorageMultisample"); + glad_glFramebufferTextureLayer = (PFNGLFRAMEBUFFERTEXTURELAYERPROC)load("glFramebufferTextureLayer"); + glad_glMapBufferRange = (PFNGLMAPBUFFERRANGEPROC)load("glMapBufferRange"); + glad_glFlushMappedBufferRange = (PFNGLFLUSHMAPPEDBUFFERRANGEPROC)load("glFlushMappedBufferRange"); + glad_glBindVertexArray = (PFNGLBINDVERTEXARRAYPROC)load("glBindVertexArray"); + glad_glDeleteVertexArrays = (PFNGLDELETEVERTEXARRAYSPROC)load("glDeleteVertexArrays"); + glad_glGenVertexArrays = (PFNGLGENVERTEXARRAYSPROC)load("glGenVertexArrays"); + glad_glIsVertexArray = (PFNGLISVERTEXARRAYPROC)load("glIsVertexArray"); +} +static void load_GL_VERSION_3_1(GLADloadproc load) { + if (!GLAD_GL_VERSION_3_1) return; + glad_glDrawArraysInstanced = (PFNGLDRAWARRAYSINSTANCEDPROC)load("glDrawArraysInstanced"); + glad_glDrawElementsInstanced = (PFNGLDRAWELEMENTSINSTANCEDPROC)load("glDrawElementsInstanced"); + glad_glTexBuffer = (PFNGLTEXBUFFERPROC)load("glTexBuffer"); + glad_glPrimitiveRestartIndex = (PFNGLPRIMITIVERESTARTINDEXPROC)load("glPrimitiveRestartIndex"); + glad_glCopyBufferSubData = (PFNGLCOPYBUFFERSUBDATAPROC)load("glCopyBufferSubData"); + glad_glGetUniformIndices = (PFNGLGETUNIFORMINDICESPROC)load("glGetUniformIndices"); + glad_glGetActiveUniformsiv = (PFNGLGETACTIVEUNIFORMSIVPROC)load("glGetActiveUniformsiv"); + glad_glGetActiveUniformName = (PFNGLGETACTIVEUNIFORMNAMEPROC)load("glGetActiveUniformName"); + glad_glGetUniformBlockIndex = (PFNGLGETUNIFORMBLOCKINDEXPROC)load("glGetUniformBlockIndex"); + glad_glGetActiveUniformBlockiv = (PFNGLGETACTIVEUNIFORMBLOCKIVPROC)load("glGetActiveUniformBlockiv"); + glad_glGetActiveUniformBlockName = (PFNGLGETACTIVEUNIFORMBLOCKNAMEPROC)load("glGetActiveUniformBlockName"); + glad_glUniformBlockBinding = (PFNGLUNIFORMBLOCKBINDINGPROC)load("glUniformBlockBinding"); + glad_glBindBufferRange = (PFNGLBINDBUFFERRANGEPROC)load("glBindBufferRange"); + glad_glBindBufferBase = (PFNGLBINDBUFFERBASEPROC)load("glBindBufferBase"); + glad_glGetIntegeri_v = (PFNGLGETINTEGERI_VPROC)load("glGetIntegeri_v"); +} +static void load_GL_VERSION_3_2(GLADloadproc load) { + if (!GLAD_GL_VERSION_3_2) return; + glad_glDrawElementsBaseVertex = (PFNGLDRAWELEMENTSBASEVERTEXPROC)load("glDrawElementsBaseVertex"); + glad_glDrawRangeElementsBaseVertex = (PFNGLDRAWRANGEELEMENTSBASEVERTEXPROC)load("glDrawRangeElementsBaseVertex"); + glad_glDrawElementsInstancedBaseVertex = (PFNGLDRAWELEMENTSINSTANCEDBASEVERTEXPROC)load("glDrawElementsInstancedBaseVertex"); + glad_glMultiDrawElementsBaseVertex = (PFNGLMULTIDRAWELEMENTSBASEVERTEXPROC)load("glMultiDrawElementsBaseVertex"); + glad_glProvokingVertex = (PFNGLPROVOKINGVERTEXPROC)load("glProvokingVertex"); + glad_glFenceSync = (PFNGLFENCESYNCPROC)load("glFenceSync"); + glad_glIsSync = (PFNGLISSYNCPROC)load("glIsSync"); + glad_glDeleteSync = (PFNGLDELETESYNCPROC)load("glDeleteSync"); + glad_glClientWaitSync = (PFNGLCLIENTWAITSYNCPROC)load("glClientWaitSync"); + glad_glWaitSync = (PFNGLWAITSYNCPROC)load("glWaitSync"); + glad_glGetInteger64v = (PFNGLGETINTEGER64VPROC)load("glGetInteger64v"); + glad_glGetSynciv = (PFNGLGETSYNCIVPROC)load("glGetSynciv"); + glad_glGetInteger64i_v = (PFNGLGETINTEGER64I_VPROC)load("glGetInteger64i_v"); + glad_glGetBufferParameteri64v = (PFNGLGETBUFFERPARAMETERI64VPROC)load("glGetBufferParameteri64v"); + glad_glFramebufferTexture = (PFNGLFRAMEBUFFERTEXTUREPROC)load("glFramebufferTexture"); + glad_glTexImage2DMultisample = (PFNGLTEXIMAGE2DMULTISAMPLEPROC)load("glTexImage2DMultisample"); + glad_glTexImage3DMultisample = (PFNGLTEXIMAGE3DMULTISAMPLEPROC)load("glTexImage3DMultisample"); + glad_glGetMultisamplefv = (PFNGLGETMULTISAMPLEFVPROC)load("glGetMultisamplefv"); + glad_glSampleMaski = (PFNGLSAMPLEMASKIPROC)load("glSampleMaski"); +} +static void load_GL_VERSION_3_3(GLADloadproc load) { + if (!GLAD_GL_VERSION_3_3) return; + glad_glBindFragDataLocationIndexed = (PFNGLBINDFRAGDATALOCATIONINDEXEDPROC)load("glBindFragDataLocationIndexed"); + glad_glGetFragDataIndex = (PFNGLGETFRAGDATAINDEXPROC)load("glGetFragDataIndex"); + glad_glGenSamplers = (PFNGLGENSAMPLERSPROC)load("glGenSamplers"); + glad_glDeleteSamplers = (PFNGLDELETESAMPLERSPROC)load("glDeleteSamplers"); + glad_glIsSampler = (PFNGLISSAMPLERPROC)load("glIsSampler"); + glad_glBindSampler = (PFNGLBINDSAMPLERPROC)load("glBindSampler"); + glad_glSamplerParameteri = (PFNGLSAMPLERPARAMETERIPROC)load("glSamplerParameteri"); + glad_glSamplerParameteriv = (PFNGLSAMPLERPARAMETERIVPROC)load("glSamplerParameteriv"); + glad_glSamplerParameterf = (PFNGLSAMPLERPARAMETERFPROC)load("glSamplerParameterf"); + glad_glSamplerParameterfv = (PFNGLSAMPLERPARAMETERFVPROC)load("glSamplerParameterfv"); + glad_glSamplerParameterIiv = (PFNGLSAMPLERPARAMETERIIVPROC)load("glSamplerParameterIiv"); + glad_glSamplerParameterIuiv = (PFNGLSAMPLERPARAMETERIUIVPROC)load("glSamplerParameterIuiv"); + glad_glGetSamplerParameteriv = (PFNGLGETSAMPLERPARAMETERIVPROC)load("glGetSamplerParameteriv"); + glad_glGetSamplerParameterIiv = (PFNGLGETSAMPLERPARAMETERIIVPROC)load("glGetSamplerParameterIiv"); + glad_glGetSamplerParameterfv = (PFNGLGETSAMPLERPARAMETERFVPROC)load("glGetSamplerParameterfv"); + glad_glGetSamplerParameterIuiv = (PFNGLGETSAMPLERPARAMETERIUIVPROC)load("glGetSamplerParameterIuiv"); + glad_glQueryCounter = (PFNGLQUERYCOUNTERPROC)load("glQueryCounter"); + glad_glGetQueryObjecti64v = (PFNGLGETQUERYOBJECTI64VPROC)load("glGetQueryObjecti64v"); + glad_glGetQueryObjectui64v = (PFNGLGETQUERYOBJECTUI64VPROC)load("glGetQueryObjectui64v"); + glad_glVertexAttribDivisor = (PFNGLVERTEXATTRIBDIVISORPROC)load("glVertexAttribDivisor"); + glad_glVertexAttribP1ui = (PFNGLVERTEXATTRIBP1UIPROC)load("glVertexAttribP1ui"); + glad_glVertexAttribP1uiv = (PFNGLVERTEXATTRIBP1UIVPROC)load("glVertexAttribP1uiv"); + glad_glVertexAttribP2ui = (PFNGLVERTEXATTRIBP2UIPROC)load("glVertexAttribP2ui"); + glad_glVertexAttribP2uiv = (PFNGLVERTEXATTRIBP2UIVPROC)load("glVertexAttribP2uiv"); + glad_glVertexAttribP3ui = (PFNGLVERTEXATTRIBP3UIPROC)load("glVertexAttribP3ui"); + glad_glVertexAttribP3uiv = (PFNGLVERTEXATTRIBP3UIVPROC)load("glVertexAttribP3uiv"); + glad_glVertexAttribP4ui = (PFNGLVERTEXATTRIBP4UIPROC)load("glVertexAttribP4ui"); + glad_glVertexAttribP4uiv = (PFNGLVERTEXATTRIBP4UIVPROC)load("glVertexAttribP4uiv"); + glad_glVertexP2ui = (PFNGLVERTEXP2UIPROC)load("glVertexP2ui"); + glad_glVertexP2uiv = (PFNGLVERTEXP2UIVPROC)load("glVertexP2uiv"); + glad_glVertexP3ui = (PFNGLVERTEXP3UIPROC)load("glVertexP3ui"); + glad_glVertexP3uiv = (PFNGLVERTEXP3UIVPROC)load("glVertexP3uiv"); + glad_glVertexP4ui = (PFNGLVERTEXP4UIPROC)load("glVertexP4ui"); + glad_glVertexP4uiv = (PFNGLVERTEXP4UIVPROC)load("glVertexP4uiv"); + glad_glTexCoordP1ui = (PFNGLTEXCOORDP1UIPROC)load("glTexCoordP1ui"); + glad_glTexCoordP1uiv = (PFNGLTEXCOORDP1UIVPROC)load("glTexCoordP1uiv"); + glad_glTexCoordP2ui = (PFNGLTEXCOORDP2UIPROC)load("glTexCoordP2ui"); + glad_glTexCoordP2uiv = (PFNGLTEXCOORDP2UIVPROC)load("glTexCoordP2uiv"); + glad_glTexCoordP3ui = (PFNGLTEXCOORDP3UIPROC)load("glTexCoordP3ui"); + glad_glTexCoordP3uiv = (PFNGLTEXCOORDP3UIVPROC)load("glTexCoordP3uiv"); + glad_glTexCoordP4ui = (PFNGLTEXCOORDP4UIPROC)load("glTexCoordP4ui"); + glad_glTexCoordP4uiv = (PFNGLTEXCOORDP4UIVPROC)load("glTexCoordP4uiv"); + glad_glMultiTexCoordP1ui = (PFNGLMULTITEXCOORDP1UIPROC)load("glMultiTexCoordP1ui"); + glad_glMultiTexCoordP1uiv = (PFNGLMULTITEXCOORDP1UIVPROC)load("glMultiTexCoordP1uiv"); + glad_glMultiTexCoordP2ui = (PFNGLMULTITEXCOORDP2UIPROC)load("glMultiTexCoordP2ui"); + glad_glMultiTexCoordP2uiv = (PFNGLMULTITEXCOORDP2UIVPROC)load("glMultiTexCoordP2uiv"); + glad_glMultiTexCoordP3ui = (PFNGLMULTITEXCOORDP3UIPROC)load("glMultiTexCoordP3ui"); + glad_glMultiTexCoordP3uiv = (PFNGLMULTITEXCOORDP3UIVPROC)load("glMultiTexCoordP3uiv"); + glad_glMultiTexCoordP4ui = (PFNGLMULTITEXCOORDP4UIPROC)load("glMultiTexCoordP4ui"); + glad_glMultiTexCoordP4uiv = (PFNGLMULTITEXCOORDP4UIVPROC)load("glMultiTexCoordP4uiv"); + glad_glNormalP3ui = (PFNGLNORMALP3UIPROC)load("glNormalP3ui"); + glad_glNormalP3uiv = (PFNGLNORMALP3UIVPROC)load("glNormalP3uiv"); + glad_glColorP3ui = (PFNGLCOLORP3UIPROC)load("glColorP3ui"); + glad_glColorP3uiv = (PFNGLCOLORP3UIVPROC)load("glColorP3uiv"); + glad_glColorP4ui = (PFNGLCOLORP4UIPROC)load("glColorP4ui"); + glad_glColorP4uiv = (PFNGLCOLORP4UIVPROC)load("glColorP4uiv"); + glad_glSecondaryColorP3ui = (PFNGLSECONDARYCOLORP3UIPROC)load("glSecondaryColorP3ui"); + glad_glSecondaryColorP3uiv = (PFNGLSECONDARYCOLORP3UIVPROC)load("glSecondaryColorP3uiv"); +} +static int find_extensionsGL(void) { + if (!get_exts()) return 0; + (void)& has_ext; + free_exts(); + return 1; +} + +static void find_coreGL(void) { + + /* Thank you @elmindreda + * https://github.com/elmindreda/greg/blob/master/templates/greg.c.in#L176 + * https://github.com/glfw/glfw/blob/master/src/context.c#L36 + */ + int i, major, minor; + + const char* version; + const char* prefixes[] = { + "OpenGL ES-CM ", + "OpenGL ES-CL ", + "OpenGL ES ", + NULL + }; + + version = (const char*)glGetString(GL_VERSION); + if (!version) return; + + for (i = 0; prefixes[i]; i++) { + const size_t length = strlen(prefixes[i]); + if (strncmp(version, prefixes[i], length) == 0) { + version += length; + break; + } + } + + /* PR #18 */ +#ifdef _MSC_VER + sscanf_s(version, "%d.%d", &major, &minor); +#else + sscanf(version, "%d.%d", &major, &minor); +#endif + + GLVersion.major = major; GLVersion.minor = minor; + max_loaded_major = major; max_loaded_minor = minor; + GLAD_GL_VERSION_1_0 = (major == 1 && minor >= 0) || major > 1; + GLAD_GL_VERSION_1_1 = (major == 1 && minor >= 1) || major > 1; + GLAD_GL_VERSION_1_2 = (major == 1 && minor >= 2) || major > 1; + GLAD_GL_VERSION_1_3 = (major == 1 && minor >= 3) || major > 1; + GLAD_GL_VERSION_1_4 = (major == 1 && minor >= 4) || major > 1; + GLAD_GL_VERSION_1_5 = (major == 1 && minor >= 5) || major > 1; + GLAD_GL_VERSION_2_0 = (major == 2 && minor >= 0) || major > 2; + GLAD_GL_VERSION_2_1 = (major == 2 && minor >= 1) || major > 2; + GLAD_GL_VERSION_3_0 = (major == 3 && minor >= 0) || major > 3; + GLAD_GL_VERSION_3_1 = (major == 3 && minor >= 1) || major > 3; + GLAD_GL_VERSION_3_2 = (major == 3 && minor >= 2) || major > 3; + GLAD_GL_VERSION_3_3 = (major == 3 && minor >= 3) || major > 3; + if (GLVersion.major > 3 || (GLVersion.major >= 3 && GLVersion.minor >= 3)) { + max_loaded_major = 3; + max_loaded_minor = 3; + } +} + +int gladLoadGLLoader(GLADloadproc load) { + GLVersion.major = 0; GLVersion.minor = 0; + glGetString = (PFNGLGETSTRINGPROC)load("glGetString"); + if (glGetString == NULL) return 0; + if (glGetString(GL_VERSION) == NULL) return 0; + find_coreGL(); + load_GL_VERSION_1_0(load); + load_GL_VERSION_1_1(load); + load_GL_VERSION_1_2(load); + load_GL_VERSION_1_3(load); + load_GL_VERSION_1_4(load); + load_GL_VERSION_1_5(load); + load_GL_VERSION_2_0(load); + load_GL_VERSION_2_1(load); + load_GL_VERSION_3_0(load); + load_GL_VERSION_3_1(load); + load_GL_VERSION_3_2(load); + load_GL_VERSION_3_3(load); + + if (!find_extensionsGL()) return 0; + return GLVersion.major != 0 || GLVersion.minor != 0; +} + diff --git a/src/ThirdParty/stb_image.hpp b/src/ThirdParty/stb_image.hpp new file mode 100644 index 0000000..764761a --- /dev/null +++ b/src/ThirdParty/stb_image.hpp @@ -0,0 +1,7568 @@ +/* stb_image - v2.19 - public domain image loader - http://nothings.org/stb + 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. + + // i.e. it should look like this: + #include ... + #include ... + #include ... + #define STB_IMAGE_IMPLEMENTATION + #include "stb_image.h" + + You can #define STBI_ASSERT(x) before the #include to avoid using assert.h. + And #define STBI_MALLOC, STBI_REALLOC, and STBI_FREE to avoid using malloc,realloc,free + + + QUICK NOTES: + Primarily of interest to game developers and other people who can + avoid problematic images and only need the trivial interface + + JPEG baseline & progressive (12 bpc/arithmetic not supported, same as stock IJG lib) + PNG 1/2/4/8/16-bit-per-channel + + TGA (not sure what subset, if a subset) + BMP non-1bpp, non-RLE + PSD (composited view only, no extra channels, 8/16 bit-per-channel) + + GIF (*comp always reports as 4-channel) + HDR (radiance rgbE format) + PIC (Softimage PIC) + PNM (PPM and PGM binary only) + + Animated GIF still needs a proper API, but here's one way to do it: + http://gist.github.com/urraka/685d9a6340b26b830d49 + + - decode from memory or through FILE (define STBI_NO_STDIO to remove code) + - decode from arbitrary I/O callbacks + - SIMD acceleration on x86/x64 (SSE2) and ARM (NEON) + + Full documentation under "DOCUMENTATION" below. + + +LICENSE + + See end of file for license information. + +RECENT REVISION HISTORY: + + 2.19 (2018-02-11) fix warning + 2.18 (2018-01-30) fix warnings + 2.17 (2018-01-29) bugfix, 1-bit BMP, 16-bitness query, fix warnings + 2.16 (2017-07-23) all functions have 16-bit variants; optimizations; bugfixes + 2.15 (2017-03-18) fix png-1,2,4; all Imagenet JPGs; no runtime SSE detection on GCC + 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs + 2.13 (2016-12-04) experimental 16-bit API, only for PNG so far; fixes + 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes + 2.11 (2016-04-02) 16-bit PNGS; enable SSE2 in non-gcc x64 + RGB-format JPEG; remove white matting in PSD; + allocate large structures on the stack; + correct channel count for PNG & BMP + 2.10 (2016-01-22) avoid warning introduced in 2.09 + 2.09 (2016-01-16) 16-bit TGA; comments in PNM files; STBI_REALLOC_SIZED + + See end of file for full revision history. + + + ============================ Contributors ========================= + + Image formats Extensions, features + Sean Barrett (jpeg, png, bmp) Jetro Lauha (stbi_info) + Nicolas Schulz (hdr, psd) Martin "SpartanJ" Golini (stbi_info) + Jonathan Dummer (tga) James "moose2000" Brown (iPhone PNG) + Jean-Marc Lienher (gif) Ben "Disch" Wenger (io callbacks) + Tom Seddon (pic) Omar Cornut (1/2/4-bit PNG) + Thatcher Ulrich (psd) Nicolas Guillemot (vertical flip) + Ken Miller (pgm, ppm) Richard Mitton (16-bit PSD) + github:urraka (animated gif) Junggon Kim (PNM comments) + Christopher Forseth (animated gif) Daniel Gibson (16-bit TGA) + socks-the-fox (16-bit PNG) + Jeremy Sawicki (handle all ImageNet JPGs) + Optimizations & bugfixes Mikhail Morozov (1-bit BMP) + Fabian "ryg" Giesen Anael Seghezzi (is-16-bit query) + Arseny Kapoulkine + John-Mark Allen + + Bug & warning fixes + Marc LeBlanc David Woo Guillaume George Martins Mozeiko + Christpher Lloyd Jerry Jansson Joseph Thomson Phil Jordan + Dave Moore Roy Eltham Hayaki Saito Nathan Reed + Won Chun Luke Graham Johan Duparc Nick Verigakis + the Horde3D community Thomas Ruf Ronny Chevalier github:rlyeh + Janez Zemva John Bartholomew Michal Cichon github:romigrou + Jonathan Blow Ken Hamada Tero Hanninen github:svdijk + Laurent Gomila Cort Stratton Sergio Gonzalez github:snagar + Aruelien Pocheville Thibault Reuille Cass Everitt github:Zelex + Ryamond Barbiero Paul Du Bois Engin Manap github:grim210 + Aldo Culquicondor Philipp Wiesemann Dale Weiler github:sammyhw + Oriol Ferrer Mesia Josh Tobin Matthew Gregan github:phprus + Julian Raschke Gregory Mullen Baldur Karlsson github:poppolopoppo + Christian Floisand Kevin Schmidt github:darealshinji + Blazej Dariusz Roszkowski github:Michaelangel007 +*/ + +#ifndef STBI_INCLUDE_STB_IMAGE_H +#define STBI_INCLUDE_STB_IMAGE_H + +// DOCUMENTATION +// +// Limitations: +// - no 12-bit-per-channel JPEG +// - no JPEGs with arithmetic coding +// - GIF always returns *comp=4 +// +// Basic usage (see HDR discussion below for HDR usage): +// 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 *channels_in_file -- outputs # of image components in image file +// int desired_channels -- 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, or NULL on an allocation failure or if the image is +// corrupt or invalid. 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 'desired_channels' if desired_channels is non-zero, or +// *channels_in_file otherwise. If desired_channels is non-zero, +// *channels_in_file has the number of components that _would_ have been +// output otherwise. E.g. if you set desired_channels to 4, you will always +// get RGBA output, but you can check *channels_in_file to see if it's trivially +// opaque because e.g. there were only 3 channels in the source image. +// +// 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, *channels_in_file 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. +// +// =========================================================================== +// +// Philosophy +// +// stb libraries are designed with the following priorities: +// +// 1. easy to use +// 2. easy to maintain +// 3. good performance +// +// Sometimes I let "good performance" creep up in priority over "easy to maintain", +// and for best performance I may provide less-easy-to-use APIs that give higher +// performance, in addition to the easy to use ones. Nevertheless, it's important +// to keep in mind that from the standpoint of you, a client of this library, +// all you care about is #1 and #3, and stb libraries DO NOT emphasize #3 above all. +// +// Some secondary priorities arise directly from the first two, some of which +// make more explicit reasons why performance can't be emphasized. +// +// - Portable ("ease of use") +// - Small source code footprint ("easy to maintain") +// - No dependencies ("ease of use") +// +// =========================================================================== +// +// 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). +// +// =========================================================================== +// +// SIMD support +// +// The JPEG decoder will try to automatically use SIMD kernels on x86 when +// supported by the compiler. For ARM Neon support, you must explicitly +// request it. +// +// (The old do-it-yourself SIMD API is no longer supported in the current +// code.) +// +// On x86, SSE2 will automatically be used when available based on a run-time +// test; if not, the generic C versions are used as a fall-back. On ARM targets, +// the typical path is to have separate builds for NEON and non-NEON devices +// (at least this is true for iOS and Android). Therefore, the NEON support is +// toggled by a build flag: define STBI_NEON to get NEON loops. +// +// If for some reason you do not want to use any of SIMD code, or if +// you have issues compiling it, you can disable it entirely by +// defining STBI_NO_SIMD. +// +// =========================================================================== +// +// 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); +// +// =========================================================================== +// +// iPhone PNG support: +// +// By default we convert iphone-formatted PNGs back to RGB, even though +// they are internally encoded differently. 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 (which +// is BGR stored in RGB). +// +// 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). +// +// =========================================================================== +// +// ADDITIONAL CONFIGURATION +// +// - You can suppress implementation of any of the decoders to reduce +// your code footprint by #defining one or more of the following +// symbols before creating the implementation. +// +// STBI_NO_JPEG +// STBI_NO_PNG +// STBI_NO_BMP +// STBI_NO_PSD +// STBI_NO_TGA +// STBI_NO_GIF +// STBI_NO_HDR +// STBI_NO_PIC +// STBI_NO_PNM (.ppm and .pgm) +// +// - You can request *only* certain decoders and suppress all other ones +// (this will be more forward-compatible, as addition of new decoders +// doesn't require you to disable them explicitly): +// +// STBI_ONLY_JPEG +// STBI_ONLY_PNG +// STBI_ONLY_BMP +// STBI_ONLY_PSD +// STBI_ONLY_TGA +// STBI_ONLY_GIF +// STBI_ONLY_HDR +// STBI_ONLY_PIC +// STBI_ONLY_PNM (.ppm and .pgm) +// +// - If you use STBI_NO_PNG (or _ONLY_ without PNG), and you still +// want the zlib decoder to be available, #define STBI_SUPPORT_ZLIB +// + + +#ifndef STBI_NO_STDIO +#include +#endif // STBI_NO_STDIO + +#define STBI_VERSION 1 + +enum +{ + STBI_default = 0, // only used for desired_channels + + STBI_grey = 1, + STBI_grey_alpha = 2, + STBI_rgb = 3, + STBI_rgb_alpha = 4 +}; + +typedef unsigned char stbi_uc; +typedef unsigned short stbi_us; + +#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 + // + + 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; + + //////////////////////////////////// + // + // 8-bits-per-channel interface + // + + STBIDEF stbi_uc* stbi_load_from_memory(stbi_uc const* buffer, int len, int* x, int* y, int* channels_in_file, int desired_channels); + STBIDEF stbi_uc* stbi_load_from_callbacks(stbi_io_callbacks const* clbk, void* user, int* x, int* y, int* channels_in_file, int desired_channels); +#ifndef STBI_NO_GIF + STBIDEF stbi_uc* stbi_load_gif_from_memory(stbi_uc const* buffer, int len, int** delays, int* x, int* y, int* z, int* comp, int req_comp); +#endif + + +#ifndef STBI_NO_STDIO + STBIDEF stbi_uc* stbi_load(char const* filename, int* x, int* y, int* channels_in_file, int desired_channels); + STBIDEF stbi_uc* stbi_load_from_file(FILE* f, int* x, int* y, int* channels_in_file, int desired_channels); + // for stbi_load_from_file, file pointer is left pointing immediately after image +#endif + +//////////////////////////////////// +// +// 16-bits-per-channel interface +// + + STBIDEF stbi_us* stbi_load_16_from_memory(stbi_uc const* buffer, int len, int* x, int* y, int* channels_in_file, int desired_channels); + STBIDEF stbi_us* stbi_load_16_from_callbacks(stbi_io_callbacks const* clbk, void* user, int* x, int* y, int* channels_in_file, int desired_channels); + +#ifndef STBI_NO_STDIO + STBIDEF stbi_us* stbi_load_16(char const* filename, int* x, int* y, int* channels_in_file, int desired_channels); + STBIDEF stbi_us* stbi_load_from_file_16(FILE* f, int* x, int* y, int* channels_in_file, int desired_channels); +#endif + + //////////////////////////////////// + // + // float-per-channel interface + // +#ifndef STBI_NO_LINEAR + STBIDEF float* stbi_loadf_from_memory(stbi_uc const* buffer, int len, int* x, int* y, int* channels_in_file, int desired_channels); + STBIDEF float* stbi_loadf_from_callbacks(stbi_io_callbacks const* clbk, void* user, int* x, int* y, int* channels_in_file, int desired_channels); + +#ifndef STBI_NO_STDIO + STBIDEF float* stbi_loadf(char const* filename, int* x, int* y, int* channels_in_file, int desired_channels); + STBIDEF float* stbi_loadf_from_file(FILE* f, int* x, int* y, int* channels_in_file, int desired_channels); +#endif +#endif + +#ifndef STBI_NO_HDR + STBIDEF void stbi_hdr_to_ldr_gamma(float gamma); + STBIDEF void stbi_hdr_to_ldr_scale(float scale); +#endif // STBI_NO_HDR + +#ifndef STBI_NO_LINEAR + STBIDEF void stbi_ldr_to_hdr_gamma(float gamma); + STBIDEF void stbi_ldr_to_hdr_scale(float scale); +#endif // STBI_NO_LINEAR + + // stbi_is_hdr is always defined, but always returns false if STBI_NO_HDR + 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); + STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const* buffer, int len); + STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const* clbk, void* user); + +#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); + STBIDEF int stbi_is_16_bit(char const* filename); + STBIDEF int stbi_is_16_bit_from_file(FILE* f); +#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); + + // flip the image vertically, so the first pixel in the output array is the bottom left + STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip); + + // 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); + + +#ifdef __cplusplus +} +#endif + +// +// +//// end header file ///////////////////////////////////////////////////// +#endif // STBI_INCLUDE_STB_IMAGE_H + +#ifdef STB_IMAGE_IMPLEMENTATION + +#if defined(STBI_ONLY_JPEG) || defined(STBI_ONLY_PNG) || defined(STBI_ONLY_BMP) \ + || defined(STBI_ONLY_TGA) || defined(STBI_ONLY_GIF) || defined(STBI_ONLY_PSD) \ + || defined(STBI_ONLY_HDR) || defined(STBI_ONLY_PIC) || defined(STBI_ONLY_PNM) \ + || defined(STBI_ONLY_ZLIB) +#ifndef STBI_ONLY_JPEG +#define STBI_NO_JPEG +#endif +#ifndef STBI_ONLY_PNG +#define STBI_NO_PNG +#endif +#ifndef STBI_ONLY_BMP +#define STBI_NO_BMP +#endif +#ifndef STBI_ONLY_PSD +#define STBI_NO_PSD +#endif +#ifndef STBI_ONLY_TGA +#define STBI_NO_TGA +#endif +#ifndef STBI_ONLY_GIF +#define STBI_NO_GIF +#endif +#ifndef STBI_ONLY_HDR +#define STBI_NO_HDR +#endif +#ifndef STBI_ONLY_PIC +#define STBI_NO_PIC +#endif +#ifndef STBI_ONLY_PNM +#define STBI_NO_PNM +#endif +#endif + +#if defined(STBI_NO_PNG) && !defined(STBI_SUPPORT_ZLIB) && !defined(STBI_NO_ZLIB) +#define STBI_NO_ZLIB +#endif + + +#include +#include // ptrdiff_t on osx +#include +#include +#include + +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +#include // ldexp, pow +#endif + +#ifndef STBI_NO_STDIO +#include +#endif + +#ifndef STBI_ASSERT +#include +#define STBI_ASSERT(x) assert(x) +#endif + + +#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 + +#if defined(STBI_MALLOC) && defined(STBI_FREE) && (defined(STBI_REALLOC) || defined(STBI_REALLOC_SIZED)) +// ok +#elif !defined(STBI_MALLOC) && !defined(STBI_FREE) && !defined(STBI_REALLOC) && !defined(STBI_REALLOC_SIZED) +// ok +#else +#error "Must define all or none of STBI_MALLOC, STBI_FREE, and STBI_REALLOC (or STBI_REALLOC_SIZED)." +#endif + +#ifndef STBI_MALLOC +#define STBI_MALLOC(sz) malloc(sz) +#define STBI_REALLOC(p,newsz) realloc(p,newsz) +#define STBI_FREE(p) free(p) +#endif + +#ifndef STBI_REALLOC_SIZED +#define STBI_REALLOC_SIZED(p,oldsz,newsz) STBI_REALLOC(p,newsz) +#endif + +// x86/x64 detection +#if defined(__x86_64__) || defined(_M_X64) +#define STBI__X64_TARGET +#elif defined(__i386) || defined(_M_IX86) +#define STBI__X86_TARGET +#endif + +#if defined(__GNUC__) && defined(STBI__X86_TARGET) && !defined(__SSE2__) && !defined(STBI_NO_SIMD) +// gcc doesn't support sse2 intrinsics unless you compile with -msse2, +// which in turn means it gets to use SSE2 everywhere. This is unfortunate, +// but previous attempts to provide the SSE2 functions with runtime +// detection caused numerous issues. The way architecture extensions are +// exposed in GCC/Clang is, sadly, not really suited for one-file libs. +// New behavior: if compiled with -msse2, we use SSE2 without any +// detection; if not, we don't use it at all. +#define STBI_NO_SIMD +#endif + +#if defined(__MINGW32__) && defined(STBI__X86_TARGET) && !defined(STBI_MINGW_ENABLE_SSE2) && !defined(STBI_NO_SIMD) +// Note that __MINGW32__ doesn't actually mean 32-bit, so we have to avoid STBI__X64_TARGET +// +// 32-bit MinGW wants ESP to be 16-byte aligned, but this is not in the +// Windows ABI and VC++ as well as Windows DLLs don't maintain that invariant. +// As a result, enabling SSE2 on 32-bit MinGW is dangerous when not +// simultaneously enabling "-mstackrealign". +// +// See https://github.com/nothings/stb/issues/81 for more information. +// +// So default to no SSE2 on 32-bit MinGW. If you've read this far and added +// -mstackrealign to your build settings, feel free to #define STBI_MINGW_ENABLE_SSE2. +#define STBI_NO_SIMD +#endif + +#if !defined(STBI_NO_SIMD) && (defined(STBI__X86_TARGET) || defined(STBI__X64_TARGET)) +#define STBI_SSE2 +#include + +#ifdef _MSC_VER + +#if _MSC_VER >= 1400 // not VC6 +#include // __cpuid +static int stbi__cpuid3(void) +{ + int info[4]; + __cpuid(info, 1); + return info[3]; +} +#else +static int stbi__cpuid3(void) +{ + int res; + __asm { + mov eax, 1 + cpuid + mov res, edx + } + return res; +} +#endif + +#define STBI_SIMD_ALIGN(type, name) __declspec(align(16)) type name + +static int stbi__sse2_available(void) +{ + int info3 = stbi__cpuid3(); + return ((info3 >> 26) & 1) != 0; +} +#else // assume GCC-style if not VC++ +#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) + +static int stbi__sse2_available(void) +{ + // If we're even attempting to compile this on GCC/Clang, that means + // -msse2 is on, which means the compiler is allowed to use SSE2 + // instructions at will, and so are we. + return 1; +} +#endif +#endif + +// ARM NEON +#if defined(STBI_NO_SIMD) && defined(STBI_NEON) +#undef STBI_NEON +#endif + +#ifdef STBI_NEON +#include +// assume GCC or Clang on ARM targets +#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) +#endif + +#ifndef STBI_SIMD_ALIGN +#define STBI_SIMD_ALIGN(type, name) type name +#endif + +/////////////////////////////////////////////// +// +// 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, * img_buffer_original_end; +} 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 = s->img_buffer_original_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); + s->img_buffer_original_end = s->img_buffer_end; +} + +#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; + s->img_buffer_end = s->img_buffer_original_end; +} + +enum +{ + STBI_ORDER_RGB, + STBI_ORDER_BGR +}; + +typedef struct +{ + int bits_per_channel; + int num_channels; + int channel_order; +} stbi__result_info; + +#ifndef STBI_NO_JPEG +static int stbi__jpeg_test(stbi__context* s); +static void* stbi__jpeg_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri); +static int stbi__jpeg_info(stbi__context* s, int* x, int* y, int* comp); +#endif + +#ifndef STBI_NO_PNG +static int stbi__png_test(stbi__context* s); +static void* stbi__png_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri); +static int stbi__png_info(stbi__context* s, int* x, int* y, int* comp); +static int stbi__png_is16(stbi__context* s); +#endif + +#ifndef STBI_NO_BMP +static int stbi__bmp_test(stbi__context* s); +static void* stbi__bmp_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri); +static int stbi__bmp_info(stbi__context* s, int* x, int* y, int* comp); +#endif + +#ifndef STBI_NO_TGA +static int stbi__tga_test(stbi__context* s); +static void* stbi__tga_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri); +static int stbi__tga_info(stbi__context* s, int* x, int* y, int* comp); +#endif + +#ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context* s); +static void* stbi__psd_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri, int bpc); +static int stbi__psd_info(stbi__context* s, int* x, int* y, int* comp); +static int stbi__psd_is16(stbi__context* s); +#endif + +#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, stbi__result_info* ri); +static int stbi__hdr_info(stbi__context* s, int* x, int* y, int* comp); +#endif + +#ifndef STBI_NO_PIC +static int stbi__pic_test(stbi__context* s); +static void* stbi__pic_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri); +static int stbi__pic_info(stbi__context* s, int* x, int* y, int* comp); +#endif + +#ifndef STBI_NO_GIF +static int stbi__gif_test(stbi__context* s); +static void* stbi__gif_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri); +static void* stbi__load_gif_main(stbi__context* s, int** delays, int* x, int* y, int* z, int* comp, int req_comp); +static int stbi__gif_info(stbi__context* s, int* x, int* y, int* comp); +#endif + +#ifndef STBI_NO_PNM +static int stbi__pnm_test(stbi__context* s); +static void* stbi__pnm_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri); +static int stbi__pnm_info(stbi__context* s, int* x, int* y, int* comp); +#endif + +// 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 STBI_MALLOC(size); +} + +// stb_image uses ints pervasively, including for offset calculations. +// therefore the largest decoded image size we can support with the +// current code, even on 64-bit targets, is INT_MAX. this is not a +// significant limitation for the intended use case. +// +// we do, however, need to make sure our size calculations don't +// overflow. hence a few helper functions for size calculations that +// multiply integers together, making sure that they're non-negative +// and no overflow occurs. + +// return 1 if the sum is valid, 0 on overflow. +// negative terms are considered invalid. +static int stbi__addsizes_valid(int a, int b) +{ + if (b < 0) return 0; + // now 0 <= b <= INT_MAX, hence also + // 0 <= INT_MAX - b <= INTMAX. + // And "a + b <= INT_MAX" (which might overflow) is the + // same as a <= INT_MAX - b (no overflow) + return a <= INT_MAX - b; +} + +// returns 1 if the product is valid, 0 on overflow. +// negative factors are considered invalid. +static int stbi__mul2sizes_valid(int a, int b) +{ + if (a < 0 || b < 0) return 0; + if (b == 0) return 1; // mul-by-0 is always safe + // portable way to check for no overflows in a*b + return a <= INT_MAX / b; +} + +// returns 1 if "a*b + add" has no negative terms/factors and doesn't overflow +static int stbi__mad2sizes_valid(int a, int b, int add) +{ + return stbi__mul2sizes_valid(a, b) && stbi__addsizes_valid(a * b, add); +} + +// returns 1 if "a*b*c + add" has no negative terms/factors and doesn't overflow +static int stbi__mad3sizes_valid(int a, int b, int c, int add) +{ + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a * b, c) && + stbi__addsizes_valid(a * b * c, add); +} + +// returns 1 if "a*b*c*d + add" has no negative terms/factors and doesn't overflow +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +static int stbi__mad4sizes_valid(int a, int b, int c, int d, int add) +{ + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a * b, c) && + stbi__mul2sizes_valid(a * b * c, d) && stbi__addsizes_valid(a * b * c * d, add); +} +#endif + +// mallocs with size overflow checking +static void* stbi__malloc_mad2(int a, int b, int add) +{ + if (!stbi__mad2sizes_valid(a, b, add)) return NULL; + return stbi__malloc(a * b + add); +} + +static void* stbi__malloc_mad3(int a, int b, int c, int add) +{ + if (!stbi__mad3sizes_valid(a, b, c, add)) return NULL; + return stbi__malloc(a * b * c + add); +} + +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +static void* stbi__malloc_mad4(int a, int b, int c, int d, int add) +{ + if (!stbi__mad4sizes_valid(a, b, c, d, add)) return NULL; + return stbi__malloc(a * b * c * d + add); +} +#endif + +// 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 *)(size_t) (stbi__err(x,y)?NULL:NULL)) +#define stbi__errpuc(x,y) ((unsigned char *)(size_t) (stbi__err(x,y)?NULL:NULL)) + +STBIDEF void stbi_image_free(void* retval_from_stbi_load) +{ + STBI_FREE(retval_from_stbi_load); +} + +#ifndef STBI_NO_LINEAR +static float* stbi__ldr_to_hdr(stbi_uc* data, int x, int y, int comp); +#endif + +#ifndef STBI_NO_HDR +static stbi_uc* stbi__hdr_to_ldr(float* data, int x, int y, int comp); +#endif + +static int stbi__vertically_flip_on_load = 0; + +STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip) +{ + stbi__vertically_flip_on_load = flag_true_if_should_flip; +} + +static void* stbi__load_main(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri, int bpc) +{ + memset(ri, 0, sizeof(*ri)); // make sure it's initialized if we add new fields + ri->bits_per_channel = 8; // default is 8 so most paths don't have to be changed + ri->channel_order = STBI_ORDER_RGB; // all current input & output are this, but this is here so we can add BGR order + ri->num_channels = 0; + +#ifndef STBI_NO_JPEG + if (stbi__jpeg_test(s)) return stbi__jpeg_load(s, x, y, comp, req_comp, ri); +#endif +#ifndef STBI_NO_PNG + if (stbi__png_test(s)) return stbi__png_load(s, x, y, comp, req_comp, ri); +#endif +#ifndef STBI_NO_BMP + if (stbi__bmp_test(s)) return stbi__bmp_load(s, x, y, comp, req_comp, ri); +#endif +#ifndef STBI_NO_GIF + if (stbi__gif_test(s)) return stbi__gif_load(s, x, y, comp, req_comp, ri); +#endif +#ifndef STBI_NO_PSD + if (stbi__psd_test(s)) return stbi__psd_load(s, x, y, comp, req_comp, ri, bpc); +#endif +#ifndef STBI_NO_PIC + if (stbi__pic_test(s)) return stbi__pic_load(s, x, y, comp, req_comp, ri); +#endif +#ifndef STBI_NO_PNM + if (stbi__pnm_test(s)) return stbi__pnm_load(s, x, y, comp, req_comp, ri); +#endif + +#ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) { + float* hdr = stbi__hdr_load(s, x, y, comp, req_comp, ri); + return stbi__hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp); + } +#endif + +#ifndef STBI_NO_TGA + // test tga last because it's a crappy test! + if (stbi__tga_test(s)) + return stbi__tga_load(s, x, y, comp, req_comp, ri); +#endif + + return stbi__errpuc("unknown image type", "Image not of any known type, or corrupt"); +} + +static stbi_uc* stbi__convert_16_to_8(stbi__uint16* orig, int w, int h, int channels) +{ + int i; + int img_len = w * h * channels; + stbi_uc* reduced; + + reduced = (stbi_uc*)stbi__malloc(img_len); + if (reduced == NULL) return stbi__errpuc("outofmem", "Out of memory"); + + for (i = 0; i < img_len; ++i) + reduced[i] = (stbi_uc)((orig[i] >> 8) & 0xFF); // top half of each byte is sufficient approx of 16->8 bit scaling + + STBI_FREE(orig); + return reduced; +} + +static stbi__uint16* stbi__convert_8_to_16(stbi_uc* orig, int w, int h, int channels) +{ + int i; + int img_len = w * h * channels; + stbi__uint16* enlarged; + + enlarged = (stbi__uint16*)stbi__malloc(img_len * 2); + if (enlarged == NULL) return (stbi__uint16*)stbi__errpuc("outofmem", "Out of memory"); + + for (i = 0; i < img_len; ++i) + enlarged[i] = (stbi__uint16)((orig[i] << 8) + orig[i]); // replicate to high and low byte, maps 0->0, 255->0xffff + + STBI_FREE(orig); + return enlarged; +} + +static void stbi__vertical_flip(void* image, int w, int h, int bytes_per_pixel) +{ + int row; + size_t bytes_per_row = (size_t)w * bytes_per_pixel; + stbi_uc temp[2048]; + stbi_uc* bytes = (stbi_uc*)image; + + for (row = 0; row < (h >> 1); row++) { + stbi_uc* row0 = bytes + row * bytes_per_row; + stbi_uc* row1 = bytes + (h - row - 1) * bytes_per_row; + // swap row0 with row1 + size_t bytes_left = bytes_per_row; + while (bytes_left) { + size_t bytes_copy = (bytes_left < sizeof(temp)) ? bytes_left : sizeof(temp); + memcpy(temp, row0, bytes_copy); + memcpy(row0, row1, bytes_copy); + memcpy(row1, temp, bytes_copy); + row0 += bytes_copy; + row1 += bytes_copy; + bytes_left -= bytes_copy; + } + } +} + +static void stbi__vertical_flip_slices(void* image, int w, int h, int z, int bytes_per_pixel) +{ + int slice; + int slice_size = w * h * bytes_per_pixel; + + stbi_uc* bytes = (stbi_uc*)image; + for (slice = 0; slice < z; ++slice) { + stbi__vertical_flip(bytes, w, h, bytes_per_pixel); + bytes += slice_size; + } +} + +static unsigned char* stbi__load_and_postprocess_8bit(stbi__context* s, int* x, int* y, int* comp, int req_comp) +{ + stbi__result_info ri; + void* result = stbi__load_main(s, x, y, comp, req_comp, &ri, 8); + + if (result == NULL) + return NULL; + + if (ri.bits_per_channel != 8) { + STBI_ASSERT(ri.bits_per_channel == 16); + result = stbi__convert_16_to_8((stbi__uint16*)result, *x, *y, req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 8; + } + + // @TODO: move stbi__convert_format to here + + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi_uc)); + } + + return (unsigned char*)result; +} + +static stbi__uint16* stbi__load_and_postprocess_16bit(stbi__context* s, int* x, int* y, int* comp, int req_comp) +{ + stbi__result_info ri; + void* result = stbi__load_main(s, x, y, comp, req_comp, &ri, 16); + + if (result == NULL) + return NULL; + + if (ri.bits_per_channel != 16) { + STBI_ASSERT(ri.bits_per_channel == 8); + result = stbi__convert_8_to_16((stbi_uc*)result, *x, *y, req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 16; + } + + // @TODO: move stbi__convert_format16 to here + // @TODO: special case RGB-to-Y (and RGBA-to-YA) for 8-bit-to-16-bit case to keep more precision + + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi__uint16)); + } + + return (stbi__uint16*)result; +} + +#if !defined(STBI_NO_HDR) || !defined(STBI_NO_LINEAR) +static void stbi__float_postprocess(float* result, int* x, int* y, int* comp, int req_comp) +{ + if (stbi__vertically_flip_on_load && result != NULL) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(float)); + } +} +#endif + +#ifndef STBI_NO_STDIO + +static FILE* stbi__fopen(char const* filename, char const* mode) +{ + FILE* f; +#if defined(_MSC_VER) && _MSC_VER >= 1400 + if (0 != fopen_s(&f, filename, mode)) + f = 0; +#else + f = fopen(filename, mode); +#endif + return f; +} + + +STBIDEF stbi_uc* 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 stbi_uc* 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_and_postprocess_8bit(&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; +} + +STBIDEF stbi__uint16* stbi_load_from_file_16(FILE* f, int* x, int* y, int* comp, int req_comp) +{ + stbi__uint16* result; + stbi__context s; + stbi__start_file(&s, f); + result = stbi__load_and_postprocess_16bit(&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; +} + +STBIDEF stbi_us* stbi_load_16(char const* filename, int* x, int* y, int* comp, int req_comp) +{ + FILE* f = stbi__fopen(filename, "rb"); + stbi__uint16* result; + if (!f) return (stbi_us*)stbi__errpuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file_16(f, x, y, comp, req_comp); + fclose(f); + return result; +} + + +#endif //!STBI_NO_STDIO + +STBIDEF stbi_us* stbi_load_16_from_memory(stbi_uc const* buffer, int len, int* x, int* y, int* channels_in_file, int desired_channels) +{ + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__load_and_postprocess_16bit(&s, x, y, channels_in_file, desired_channels); +} + +STBIDEF stbi_us* stbi_load_16_from_callbacks(stbi_io_callbacks const* clbk, void* user, int* x, int* y, int* channels_in_file, int desired_channels) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks*)clbk, user); + return stbi__load_and_postprocess_16bit(&s, x, y, channels_in_file, desired_channels); +} + +STBIDEF stbi_uc* 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_and_postprocess_8bit(&s, x, y, comp, req_comp); +} + +STBIDEF stbi_uc* 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_and_postprocess_8bit(&s, x, y, comp, req_comp); +} + +#ifndef STBI_NO_GIF +STBIDEF stbi_uc* stbi_load_gif_from_memory(stbi_uc const* buffer, int len, int** delays, int* x, int* y, int* z, int* comp, int req_comp) +{ + unsigned char* result; + stbi__context s; + stbi__start_mem(&s, buffer, len); + + result = (unsigned char*)stbi__load_gif_main(&s, delays, x, y, z, comp, req_comp); + if (stbi__vertically_flip_on_load) { + stbi__vertical_flip_slices(result, *x, *y, *z, *comp); + } + + return result; +} +#endif + +#ifndef STBI_NO_LINEAR +static 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)) { + stbi__result_info ri; + float* hdr_data = stbi__hdr_load(s, x, y, comp, req_comp, &ri); + if (hdr_data) + stbi__float_postprocess(hdr_data, x, y, comp, req_comp); + return hdr_data; + } +#endif + data = stbi__load_and_postprocess_8bit(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"); +} + +STBIDEF 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); +} + +STBIDEF 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 +STBIDEF 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; +} + +STBIDEF 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_LINEAR + +// these is-hdr-or-not is defined independent of whether STBI_NO_LINEAR is +// defined, for API simplicity; if STBI_NO_LINEAR is defined, it always +// reports false! + +STBIDEF 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 + long pos = ftell(f); + int res; + stbi__context s; + stbi__start_file(&s, f); + res = stbi__hdr_test(&s); + fseek(f, pos, SEEK_SET); + return res; +#else + STBI_NOTUSED(f); + 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 + STBI_NOTUSED(clbk); + STBI_NOTUSED(user); + return 0; +#endif +} + +#ifndef STBI_NO_LINEAR +static float stbi__l2h_gamma = 2.2f, stbi__l2h_scale = 1.0f; + +STBIDEF void stbi_ldr_to_hdr_gamma(float gamma) { stbi__l2h_gamma = gamma; } +STBIDEF void stbi_ldr_to_hdr_scale(float scale) { stbi__l2h_scale = scale; } +#endif + +static float stbi__h2l_gamma_i = 1.0f / 2.2f, stbi__h2l_scale_i = 1.0f; + +STBIDEF void stbi_hdr_to_ldr_gamma(float gamma) { stbi__h2l_gamma_i = 1 / gamma; } +STBIDEF void stbi_hdr_to_ldr_scale(float scale) { stbi__h2l_scale_i = 1 / scale; } + + +////////////////////////////////////////////////////////////////////////////// +// +// Common code used by all image loaders +// + +enum +{ + STBI__SCAN_load = 0, + STBI__SCAN_type, + STBI__SCAN_header +}; + +static void stbi__refill_buffer(stbi__context* s) +{ + int n = (s->io.read)(s->io_user_data, (char*)s->buffer_start, s->buflen); + if (n == 0) { + // 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 (n < 0) { + s->img_buffer = s->img_buffer_end; + return; + } + if (s->io.read) { + int blen = (int)(s->img_buffer_end - s->img_buffer); + if (blen < n) { + s->img_buffer = s->img_buffer_end; + (s->io.skip)(s->io_user_data, n - blen); + return; + } + } + s->img_buffer += n; +} + +static int stbi__getn(stbi__context* s, 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); +} + +#if defined(STBI_NO_BMP) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) +// nothing +#else +static int stbi__get16le(stbi__context* s) +{ + int z = stbi__get8(s); + return z + (stbi__get8(s) << 8); +} +#endif + +#ifndef STBI_NO_BMP +static stbi__uint32 stbi__get32le(stbi__context* s) +{ + stbi__uint32 z = stbi__get16le(s); + return z + (stbi__get16le(s) << 16); +} +#endif + +#define STBI__BYTECAST(x) ((stbi_uc) ((x) & 255)) // truncate int to byte without warnings + + +////////////////////////////////////////////////////////////////////////////// +// +// 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_mad3(req_comp, x, y, 0); + if (good == NULL) { + STBI_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 STBI__COMBO(a,b) ((a)*8+(b)) +#define STBI__CASE(a,b) case STBI__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 (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1, 2) { dest[0] = src[0], dest[1] = 255; } break; + STBI__CASE(1, 3) { dest[0] = dest[1] = dest[2] = src[0]; } break; + STBI__CASE(1, 4) { dest[0] = dest[1] = dest[2] = src[0], dest[3] = 255; } break; + STBI__CASE(2, 1) { dest[0] = src[0]; } break; + STBI__CASE(2, 3) { dest[0] = dest[1] = dest[2] = src[0]; } break; + STBI__CASE(2, 4) { dest[0] = dest[1] = dest[2] = src[0], dest[3] = src[1]; } break; + STBI__CASE(3, 4) { dest[0] = src[0], dest[1] = src[1], dest[2] = src[2], dest[3] = 255; } break; + STBI__CASE(3, 1) { dest[0] = stbi__compute_y(src[0], src[1], src[2]); } break; + STBI__CASE(3, 2) { dest[0] = stbi__compute_y(src[0], src[1], src[2]), dest[1] = 255; } break; + STBI__CASE(4, 1) { dest[0] = stbi__compute_y(src[0], src[1], src[2]); } break; + STBI__CASE(4, 2) { dest[0] = stbi__compute_y(src[0], src[1], src[2]), dest[1] = src[3]; } break; + STBI__CASE(4, 3) { dest[0] = src[0], dest[1] = src[1], dest[2] = src[2]; } break; + default: STBI_ASSERT(0); + } +#undef STBI__CASE + } + + STBI_FREE(data); + return good; +} + +static stbi__uint16 stbi__compute_y_16(int r, int g, int b) +{ + return (stbi__uint16)(((r * 77) + (g * 150) + (29 * b)) >> 8); +} + +static stbi__uint16* stbi__convert_format16(stbi__uint16* data, int img_n, int req_comp, unsigned int x, unsigned int y) +{ + int i, j; + stbi__uint16* good; + + if (req_comp == img_n) return data; + STBI_ASSERT(req_comp >= 1 && req_comp <= 4); + + good = (stbi__uint16*)stbi__malloc(req_comp * x * y * 2); + if (good == NULL) { + STBI_FREE(data); + return (stbi__uint16*)stbi__errpuc("outofmem", "Out of memory"); + } + + for (j = 0; j < (int)y; ++j) { + stbi__uint16* src = data + j * x * img_n; + stbi__uint16* dest = good + j * x * req_comp; + +#define STBI__COMBO(a,b) ((a)*8+(b)) +#define STBI__CASE(a,b) case STBI__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 (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1, 2) { dest[0] = src[0], dest[1] = 0xffff; } break; + STBI__CASE(1, 3) { dest[0] = dest[1] = dest[2] = src[0]; } break; + STBI__CASE(1, 4) { dest[0] = dest[1] = dest[2] = src[0], dest[3] = 0xffff; } break; + STBI__CASE(2, 1) { dest[0] = src[0]; } break; + STBI__CASE(2, 3) { dest[0] = dest[1] = dest[2] = src[0]; } break; + STBI__CASE(2, 4) { dest[0] = dest[1] = dest[2] = src[0], dest[3] = src[1]; } break; + STBI__CASE(3, 4) { dest[0] = src[0], dest[1] = src[1], dest[2] = src[2], dest[3] = 0xffff; } break; + STBI__CASE(3, 1) { dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); } break; + STBI__CASE(3, 2) { dest[0] = stbi__compute_y_16(src[0], src[1], src[2]), dest[1] = 0xffff; } break; + STBI__CASE(4, 1) { dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); } break; + STBI__CASE(4, 2) { dest[0] = stbi__compute_y_16(src[0], src[1], src[2]), dest[1] = src[3]; } break; + STBI__CASE(4, 3) { dest[0] = src[0], dest[1] = src[1], dest[2] = src[2]; } break; + default: STBI_ASSERT(0); + } +#undef STBI__CASE + } + + STBI_FREE(data); + return good; +} + +#ifndef STBI_NO_LINEAR +static float* stbi__ldr_to_hdr(stbi_uc* data, int x, int y, int comp) +{ + int i, k, n; + float* output; + if (!data) return NULL; + output = (float*)stbi__malloc_mad4(x, y, comp, sizeof(float), 0); + if (output == NULL) { STBI_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; + } + STBI_FREE(data); + return output; +} +#endif + +#ifndef STBI_NO_HDR +#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; + if (!data) return NULL; + output = (stbi_uc*)stbi__malloc_mad3(x, y, comp, 0); + if (output == NULL) { STBI_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); + } + } + STBI_FREE(data); + return output; +} +#endif + +////////////////////////////////////////////////////////////////////////////// +// +// "baseline" JPEG/JFIF decoder +// +// simple implementation +// - 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 +// - some SIMD kernels for common paths on targets with SSE2/NEON +// - uses a lot of intermediate memory, could cache poorly + +#ifndef STBI_NO_JPEG + +// 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 +{ + stbi__context* s; + stbi__huffman huff_dc[4]; + stbi__huffman huff_ac[4]; + stbi__uint16 dequant[4][64]; + stbi__int16 fast_ac[4][1 << FAST_BITS]; + + // 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, * raw_coeff; + stbi_uc* linebuf; + short* coeff; // progressive only + int coeff_w, coeff_h; // number of 8x8 coefficient blocks + } 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 progressive; + int spec_start; + int spec_end; + int succ_high; + int succ_low; + int eob_run; + int jfif; + int app14_color_transform; // Adobe APP14 tag + int rgb; + + int scan_n, order[4]; + int restart_interval, todo; + + // kernels + void (*idct_block_kernel)(stbi_uc* out, int out_stride, short data[64]); + void (*YCbCr_to_RGB_kernel)(stbi_uc* out, const stbi_uc* y, const stbi_uc* pcb, const stbi_uc* pcr, int count, int step); + stbi_uc* (*resample_row_hv_2_kernel)(stbi_uc* out, stbi_uc* in_near, stbi_uc* in_far, int w, int hs); +} stbi__jpeg; + +static int stbi__build_huffman(stbi__huffman* h, int* count) +{ + int i, j, k = 0; + unsigned int 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 >= (1u << 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; +} + +// build a table that decodes both magnitude and value of small ACs in +// one go. +static void stbi__build_fast_ac(stbi__int16* fast_ac, stbi__huffman* h) +{ + int i; + for (i = 0; i < (1 << FAST_BITS); ++i) { + stbi_uc fast = h->fast[i]; + fast_ac[i] = 0; + if (fast < 255) { + int rs = h->values[fast]; + int run = (rs >> 4) & 15; + int magbits = rs & 15; + int len = h->size[fast]; + + if (magbits && len + magbits <= FAST_BITS) { + // magnitude code followed by receive_extend code + int k = ((i << len) & ((1 << FAST_BITS) - 1)) >> (FAST_BITS - magbits); + int m = 1 << (magbits - 1); + if (k < m) k += (~0U << magbits) + 1; + // if the result is small enough, we can fit it in fast_ac table + if (k >= -128 && k <= 127) + fast_ac[i] = (stbi__int16)((k * 256) + (run * 16) + (len + magbits)); + } + } + } +} + +static void stbi__grow_buffer_unsafe(stbi__jpeg* j) +{ + do { + unsigned int b = j->nomore ? 0 : stbi__get8(j->s); + if (b == 0xff) { + int c = stbi__get8(j->s); + while (c == 0xff) c = stbi__get8(j->s); // consume fill bytes + 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 const 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]; +} + +// bias[n] = (-1<code_bits < n) stbi__grow_buffer_unsafe(j); + + sgn = (stbi__int32)j->code_buffer >> 31; // sign bit is always in MSB + k = stbi_lrot(j->code_buffer, n); + STBI_ASSERT(n >= 0 && n < (int)(sizeof(stbi__bmask) / sizeof(*stbi__bmask))); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k + (stbi__jbias[n] & ~sgn); +} + +// get some unsigned bits +stbi_inline static int stbi__jpeg_get_bits(stbi__jpeg* j, int n) +{ + unsigned int k; + if (j->code_bits < n) stbi__grow_buffer_unsafe(j); + k = stbi_lrot(j->code_buffer, n); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k; +} + +stbi_inline static int stbi__jpeg_get_bit(stbi__jpeg* j) +{ + unsigned int k; + if (j->code_bits < 1) stbi__grow_buffer_unsafe(j); + k = j->code_buffer; + j->code_buffer <<= 1; + --j->code_bits; + return k & 0x80000000; +} + +// 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 const 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, stbi__int16* fac, int b, stbi__uint16* dequant) +{ + int diff, dc, k; + int t; + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + t = stbi__jpeg_huff_decode(j, hdc); + if (t < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + + // 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 * dequant[0]); + + // decode AC components, see JPEG spec + k = 1; + do { + unsigned int zig; + int c, r, s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS) - 1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + j->code_buffer <<= s; + j->code_bits -= s; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)((r >> 8) * dequant[zig]); + } + else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (rs != 0xf0) break; // end block + k += 16; + } + else { + k += r; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)(stbi__extend_receive(j, s) * dequant[zig]); + } + } + } while (k < 64); + return 1; +} + +static int stbi__jpeg_decode_block_prog_dc(stbi__jpeg* j, short data[64], stbi__huffman* hdc, int b) +{ + int diff, dc; + int t; + if (j->spec_end != 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + + if (j->succ_high == 0) { + // first scan for DC coefficient, must be first + memset(data, 0, 64 * sizeof(data[0])); // 0 all the ac values now + t = stbi__jpeg_huff_decode(j, hdc); + diff = t ? stbi__extend_receive(j, t) : 0; + + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + data[0] = (short)(dc << j->succ_low); + } + else { + // refinement scan for DC coefficient + if (stbi__jpeg_get_bit(j)) + data[0] += (short)(1 << j->succ_low); + } + return 1; +} + +// @OPTIMIZE: store non-zigzagged during the decode passes, +// and only de-zigzag when dequantizing +static int stbi__jpeg_decode_block_prog_ac(stbi__jpeg* j, short data[64], stbi__huffman* hac, stbi__int16* fac) +{ + int k; + if (j->spec_start == 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + + if (j->succ_high == 0) { + int shift = j->succ_low; + + if (j->eob_run) { + --j->eob_run; + return 1; + } + + k = j->spec_start; + do { + unsigned int zig; + int c, r, s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS) - 1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + j->code_buffer <<= s; + j->code_bits -= s; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)((r >> 8) << shift); + } + else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r); + if (r) + j->eob_run += stbi__jpeg_get_bits(j, r); + --j->eob_run; + break; + } + k += 16; + } + else { + k += r; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)(stbi__extend_receive(j, s) << shift); + } + } + } while (k <= j->spec_end); + } + else { + // refinement scan for these AC coefficients + + short bit = (short)(1 << j->succ_low); + + if (j->eob_run) { + --j->eob_run; + for (k = j->spec_start; k <= j->spec_end; ++k) { + short* p = &data[stbi__jpeg_dezigzag[k]]; + if (*p != 0) + if (stbi__jpeg_get_bit(j)) + if ((*p & bit) == 0) { + if (*p > 0) + * p += bit; + else + *p -= bit; + } + } + } + else { + k = j->spec_start; + do { + int r, s; + int rs = stbi__jpeg_huff_decode(j, hac); // @OPTIMIZE see if we can use the fast path here, advance-by-r is so slow, eh + if (rs < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r) - 1; + if (r) + j->eob_run += stbi__jpeg_get_bits(j, r); + r = 64; // force end of block + } + else { + // r=15 s=0 should write 16 0s, so we just do + // a run of 15 0s and then write s (which is 0), + // so we don't have to do anything special here + } + } + else { + if (s != 1) return stbi__err("bad huffman code", "Corrupt JPEG"); + // sign bit + if (stbi__jpeg_get_bit(j)) + s = bit; + else + s = -bit; + } + + // advance by r + while (k <= j->spec_end) { + short* p = &data[stbi__jpeg_dezigzag[k++]]; + if (*p != 0) { + if (stbi__jpeg_get_bit(j)) + if ((*p & bit) == 0) { + if (*p > 0) + * p += bit; + else + *p -= bit; + } + } + else { + if (r == 0) { + *p = (short)s; + break; + } + --r; + } + } + } while (k <= j->spec_end); + } + } + 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) * 4096) + +// 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; + +static void stbi__idct_block(stbi_uc* out, int out_stride, short data[64]) +{ + int i, val[64], * v = val; + stbi_uc* o; + short* d = data; + + // columns + for (i = 0; i < 8; ++i, ++d, ++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] * 4; + v[0] = v[8] = v[16] = v[24] = v[32] = v[40] = v[48] = v[56] = dcterm; + } + else { + STBI__IDCT_1D(d[0], d[8], d[16], d[24], d[32], d[40], d[48], d[56]) + // 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_SSE2 +// sse2 integer IDCT. not the fastest possible implementation but it +// produces bit-identical results to the generic C version so it's +// fully "transparent". +static void stbi__idct_simd(stbi_uc* out, int out_stride, short data[64]) +{ + // This is constructed to match our regular (generic) integer IDCT exactly. + __m128i row0, row1, row2, row3, row4, row5, row6, row7; + __m128i tmp; + + // dot product constant: even elems=x, odd elems=y +#define dct_const(x,y) _mm_setr_epi16((x),(y),(x),(y),(x),(y),(x),(y)) + +// out(0) = c0[even]*x + c0[odd]*y (c0, x, y 16-bit, out 32-bit) +// out(1) = c1[even]*x + c1[odd]*y +#define dct_rot(out0,out1, x,y,c0,c1) \ + __m128i c0##lo = _mm_unpacklo_epi16((x),(y)); \ + __m128i c0##hi = _mm_unpackhi_epi16((x),(y)); \ + __m128i out0##_l = _mm_madd_epi16(c0##lo, c0); \ + __m128i out0##_h = _mm_madd_epi16(c0##hi, c0); \ + __m128i out1##_l = _mm_madd_epi16(c0##lo, c1); \ + __m128i out1##_h = _mm_madd_epi16(c0##hi, c1) + + // out = in << 12 (in 16-bit, out 32-bit) +#define dct_widen(out, in) \ + __m128i out##_l = _mm_srai_epi32(_mm_unpacklo_epi16(_mm_setzero_si128(), (in)), 4); \ + __m128i out##_h = _mm_srai_epi32(_mm_unpackhi_epi16(_mm_setzero_si128(), (in)), 4) + + // wide add +#define dct_wadd(out, a, b) \ + __m128i out##_l = _mm_add_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_add_epi32(a##_h, b##_h) + + // wide sub +#define dct_wsub(out, a, b) \ + __m128i out##_l = _mm_sub_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_sub_epi32(a##_h, b##_h) + + // butterfly a/b, add bias, then shift by "s" and pack +#define dct_bfly32o(out0, out1, a,b,bias,s) \ + { \ + __m128i abiased_l = _mm_add_epi32(a##_l, bias); \ + __m128i abiased_h = _mm_add_epi32(a##_h, bias); \ + dct_wadd(sum, abiased, b); \ + dct_wsub(dif, abiased, b); \ + out0 = _mm_packs_epi32(_mm_srai_epi32(sum_l, s), _mm_srai_epi32(sum_h, s)); \ + out1 = _mm_packs_epi32(_mm_srai_epi32(dif_l, s), _mm_srai_epi32(dif_h, s)); \ + } + + // 8-bit interleave step (for transposes) +#define dct_interleave8(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi8(a, b); \ + b = _mm_unpackhi_epi8(tmp, b) + + // 16-bit interleave step (for transposes) +#define dct_interleave16(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi16(a, b); \ + b = _mm_unpackhi_epi16(tmp, b) + +#define dct_pass(bias,shift) \ + { \ + /* even part */ \ + dct_rot(t2e,t3e, row2,row6, rot0_0,rot0_1); \ + __m128i sum04 = _mm_add_epi16(row0, row4); \ + __m128i dif04 = _mm_sub_epi16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + dct_rot(y0o,y2o, row7,row3, rot2_0,rot2_1); \ + dct_rot(y1o,y3o, row5,row1, rot3_0,rot3_1); \ + __m128i sum17 = _mm_add_epi16(row1, row7); \ + __m128i sum35 = _mm_add_epi16(row3, row5); \ + dct_rot(y4o,y5o, sum17,sum35, rot1_0,rot1_1); \ + dct_wadd(x4, y0o, y4o); \ + dct_wadd(x5, y1o, y5o); \ + dct_wadd(x6, y2o, y5o); \ + dct_wadd(x7, y3o, y4o); \ + dct_bfly32o(row0,row7, x0,x7,bias,shift); \ + dct_bfly32o(row1,row6, x1,x6,bias,shift); \ + dct_bfly32o(row2,row5, x2,x5,bias,shift); \ + dct_bfly32o(row3,row4, x3,x4,bias,shift); \ + } + + __m128i rot0_0 = dct_const(stbi__f2f(0.5411961f), stbi__f2f(0.5411961f) + stbi__f2f(-1.847759065f)); + __m128i rot0_1 = dct_const(stbi__f2f(0.5411961f) + stbi__f2f(0.765366865f), stbi__f2f(0.5411961f)); + __m128i rot1_0 = dct_const(stbi__f2f(1.175875602f) + stbi__f2f(-0.899976223f), stbi__f2f(1.175875602f)); + __m128i rot1_1 = dct_const(stbi__f2f(1.175875602f), stbi__f2f(1.175875602f) + stbi__f2f(-2.562915447f)); + __m128i rot2_0 = dct_const(stbi__f2f(-1.961570560f) + stbi__f2f(0.298631336f), stbi__f2f(-1.961570560f)); + __m128i rot2_1 = dct_const(stbi__f2f(-1.961570560f), stbi__f2f(-1.961570560f) + stbi__f2f(3.072711026f)); + __m128i rot3_0 = dct_const(stbi__f2f(-0.390180644f) + stbi__f2f(2.053119869f), stbi__f2f(-0.390180644f)); + __m128i rot3_1 = dct_const(stbi__f2f(-0.390180644f), stbi__f2f(-0.390180644f) + stbi__f2f(1.501321110f)); + + // rounding biases in column/row passes, see stbi__idct_block for explanation. + __m128i bias_0 = _mm_set1_epi32(512); + __m128i bias_1 = _mm_set1_epi32(65536 + (128 << 17)); + + // load + row0 = _mm_load_si128((const __m128i*) (data + 0 * 8)); + row1 = _mm_load_si128((const __m128i*) (data + 1 * 8)); + row2 = _mm_load_si128((const __m128i*) (data + 2 * 8)); + row3 = _mm_load_si128((const __m128i*) (data + 3 * 8)); + row4 = _mm_load_si128((const __m128i*) (data + 4 * 8)); + row5 = _mm_load_si128((const __m128i*) (data + 5 * 8)); + row6 = _mm_load_si128((const __m128i*) (data + 6 * 8)); + row7 = _mm_load_si128((const __m128i*) (data + 7 * 8)); + + // column pass + dct_pass(bias_0, 10); + + { + // 16bit 8x8 transpose pass 1 + dct_interleave16(row0, row4); + dct_interleave16(row1, row5); + dct_interleave16(row2, row6); + dct_interleave16(row3, row7); + + // transpose pass 2 + dct_interleave16(row0, row2); + dct_interleave16(row1, row3); + dct_interleave16(row4, row6); + dct_interleave16(row5, row7); + + // transpose pass 3 + dct_interleave16(row0, row1); + dct_interleave16(row2, row3); + dct_interleave16(row4, row5); + dct_interleave16(row6, row7); + } + + // row pass + dct_pass(bias_1, 17); + + { + // pack + __m128i p0 = _mm_packus_epi16(row0, row1); // a0a1a2a3...a7b0b1b2b3...b7 + __m128i p1 = _mm_packus_epi16(row2, row3); + __m128i p2 = _mm_packus_epi16(row4, row5); + __m128i p3 = _mm_packus_epi16(row6, row7); + + // 8bit 8x8 transpose pass 1 + dct_interleave8(p0, p2); // a0e0a1e1... + dct_interleave8(p1, p3); // c0g0c1g1... + + // transpose pass 2 + dct_interleave8(p0, p1); // a0c0e0g0... + dct_interleave8(p2, p3); // b0d0f0h0... + + // transpose pass 3 + dct_interleave8(p0, p2); // a0b0c0d0... + dct_interleave8(p1, p3); // a4b4c4d4... + + // store + _mm_storel_epi64((__m128i*) out, p0); out += out_stride; + _mm_storel_epi64((__m128i*) out, _mm_shuffle_epi32(p0, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i*) out, p2); out += out_stride; + _mm_storel_epi64((__m128i*) out, _mm_shuffle_epi32(p2, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i*) out, p1); out += out_stride; + _mm_storel_epi64((__m128i*) out, _mm_shuffle_epi32(p1, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i*) out, p3); out += out_stride; + _mm_storel_epi64((__m128i*) out, _mm_shuffle_epi32(p3, 0x4e)); + } + +#undef dct_const +#undef dct_rot +#undef dct_widen +#undef dct_wadd +#undef dct_wsub +#undef dct_bfly32o +#undef dct_interleave8 +#undef dct_interleave16 +#undef dct_pass +} + +#endif // STBI_SSE2 + +#ifdef STBI_NEON + +// NEON integer IDCT. should produce bit-identical +// results to the generic C version. +static void stbi__idct_simd(stbi_uc* out, int out_stride, short data[64]) +{ + int16x8_t row0, row1, row2, row3, row4, row5, row6, row7; + + int16x4_t rot0_0 = vdup_n_s16(stbi__f2f(0.5411961f)); + int16x4_t rot0_1 = vdup_n_s16(stbi__f2f(-1.847759065f)); + int16x4_t rot0_2 = vdup_n_s16(stbi__f2f(0.765366865f)); + int16x4_t rot1_0 = vdup_n_s16(stbi__f2f(1.175875602f)); + int16x4_t rot1_1 = vdup_n_s16(stbi__f2f(-0.899976223f)); + int16x4_t rot1_2 = vdup_n_s16(stbi__f2f(-2.562915447f)); + int16x4_t rot2_0 = vdup_n_s16(stbi__f2f(-1.961570560f)); + int16x4_t rot2_1 = vdup_n_s16(stbi__f2f(-0.390180644f)); + int16x4_t rot3_0 = vdup_n_s16(stbi__f2f(0.298631336f)); + int16x4_t rot3_1 = vdup_n_s16(stbi__f2f(2.053119869f)); + int16x4_t rot3_2 = vdup_n_s16(stbi__f2f(3.072711026f)); + int16x4_t rot3_3 = vdup_n_s16(stbi__f2f(1.501321110f)); + +#define dct_long_mul(out, inq, coeff) \ + int32x4_t out##_l = vmull_s16(vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmull_s16(vget_high_s16(inq), coeff) + +#define dct_long_mac(out, acc, inq, coeff) \ + int32x4_t out##_l = vmlal_s16(acc##_l, vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmlal_s16(acc##_h, vget_high_s16(inq), coeff) + +#define dct_widen(out, inq) \ + int32x4_t out##_l = vshll_n_s16(vget_low_s16(inq), 12); \ + int32x4_t out##_h = vshll_n_s16(vget_high_s16(inq), 12) + + // wide add +#define dct_wadd(out, a, b) \ + int32x4_t out##_l = vaddq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vaddq_s32(a##_h, b##_h) + +// wide sub +#define dct_wsub(out, a, b) \ + int32x4_t out##_l = vsubq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vsubq_s32(a##_h, b##_h) + +// butterfly a/b, then shift using "shiftop" by "s" and pack +#define dct_bfly32o(out0,out1, a,b,shiftop,s) \ + { \ + dct_wadd(sum, a, b); \ + dct_wsub(dif, a, b); \ + out0 = vcombine_s16(shiftop(sum_l, s), shiftop(sum_h, s)); \ + out1 = vcombine_s16(shiftop(dif_l, s), shiftop(dif_h, s)); \ + } + +#define dct_pass(shiftop, shift) \ + { \ + /* even part */ \ + int16x8_t sum26 = vaddq_s16(row2, row6); \ + dct_long_mul(p1e, sum26, rot0_0); \ + dct_long_mac(t2e, p1e, row6, rot0_1); \ + dct_long_mac(t3e, p1e, row2, rot0_2); \ + int16x8_t sum04 = vaddq_s16(row0, row4); \ + int16x8_t dif04 = vsubq_s16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + int16x8_t sum15 = vaddq_s16(row1, row5); \ + int16x8_t sum17 = vaddq_s16(row1, row7); \ + int16x8_t sum35 = vaddq_s16(row3, row5); \ + int16x8_t sum37 = vaddq_s16(row3, row7); \ + int16x8_t sumodd = vaddq_s16(sum17, sum35); \ + dct_long_mul(p5o, sumodd, rot1_0); \ + dct_long_mac(p1o, p5o, sum17, rot1_1); \ + dct_long_mac(p2o, p5o, sum35, rot1_2); \ + dct_long_mul(p3o, sum37, rot2_0); \ + dct_long_mul(p4o, sum15, rot2_1); \ + dct_wadd(sump13o, p1o, p3o); \ + dct_wadd(sump24o, p2o, p4o); \ + dct_wadd(sump23o, p2o, p3o); \ + dct_wadd(sump14o, p1o, p4o); \ + dct_long_mac(x4, sump13o, row7, rot3_0); \ + dct_long_mac(x5, sump24o, row5, rot3_1); \ + dct_long_mac(x6, sump23o, row3, rot3_2); \ + dct_long_mac(x7, sump14o, row1, rot3_3); \ + dct_bfly32o(row0,row7, x0,x7,shiftop,shift); \ + dct_bfly32o(row1,row6, x1,x6,shiftop,shift); \ + dct_bfly32o(row2,row5, x2,x5,shiftop,shift); \ + dct_bfly32o(row3,row4, x3,x4,shiftop,shift); \ + } + + // load + row0 = vld1q_s16(data + 0 * 8); + row1 = vld1q_s16(data + 1 * 8); + row2 = vld1q_s16(data + 2 * 8); + row3 = vld1q_s16(data + 3 * 8); + row4 = vld1q_s16(data + 4 * 8); + row5 = vld1q_s16(data + 5 * 8); + row6 = vld1q_s16(data + 6 * 8); + row7 = vld1q_s16(data + 7 * 8); + + // add DC bias + row0 = vaddq_s16(row0, vsetq_lane_s16(1024, vdupq_n_s16(0), 0)); + + // column pass + dct_pass(vrshrn_n_s32, 10); + + // 16bit 8x8 transpose + { + // these three map to a single VTRN.16, VTRN.32, and VSWP, respectively. + // whether compilers actually get this is another story, sadly. +#define dct_trn16(x, y) { int16x8x2_t t = vtrnq_s16(x, y); x = t.val[0]; y = t.val[1]; } +#define dct_trn32(x, y) { int32x4x2_t t = vtrnq_s32(vreinterpretq_s32_s16(x), vreinterpretq_s32_s16(y)); x = vreinterpretq_s16_s32(t.val[0]); y = vreinterpretq_s16_s32(t.val[1]); } +#define dct_trn64(x, y) { int16x8_t x0 = x; int16x8_t y0 = y; x = vcombine_s16(vget_low_s16(x0), vget_low_s16(y0)); y = vcombine_s16(vget_high_s16(x0), vget_high_s16(y0)); } + + // pass 1 + dct_trn16(row0, row1); // a0b0a2b2a4b4a6b6 + dct_trn16(row2, row3); + dct_trn16(row4, row5); + dct_trn16(row6, row7); + + // pass 2 + dct_trn32(row0, row2); // a0b0c0d0a4b4c4d4 + dct_trn32(row1, row3); + dct_trn32(row4, row6); + dct_trn32(row5, row7); + + // pass 3 + dct_trn64(row0, row4); // a0b0c0d0e0f0g0h0 + dct_trn64(row1, row5); + dct_trn64(row2, row6); + dct_trn64(row3, row7); + +#undef dct_trn16 +#undef dct_trn32 +#undef dct_trn64 + } + + // row pass + // vrshrn_n_s32 only supports shifts up to 16, we need + // 17. so do a non-rounding shift of 16 first then follow + // up with a rounding shift by 1. + dct_pass(vshrn_n_s32, 16); + + { + // pack and round + uint8x8_t p0 = vqrshrun_n_s16(row0, 1); + uint8x8_t p1 = vqrshrun_n_s16(row1, 1); + uint8x8_t p2 = vqrshrun_n_s16(row2, 1); + uint8x8_t p3 = vqrshrun_n_s16(row3, 1); + uint8x8_t p4 = vqrshrun_n_s16(row4, 1); + uint8x8_t p5 = vqrshrun_n_s16(row5, 1); + uint8x8_t p6 = vqrshrun_n_s16(row6, 1); + uint8x8_t p7 = vqrshrun_n_s16(row7, 1); + + // again, these can translate into one instruction, but often don't. +#define dct_trn8_8(x, y) { uint8x8x2_t t = vtrn_u8(x, y); x = t.val[0]; y = t.val[1]; } +#define dct_trn8_16(x, y) { uint16x4x2_t t = vtrn_u16(vreinterpret_u16_u8(x), vreinterpret_u16_u8(y)); x = vreinterpret_u8_u16(t.val[0]); y = vreinterpret_u8_u16(t.val[1]); } +#define dct_trn8_32(x, y) { uint32x2x2_t t = vtrn_u32(vreinterpret_u32_u8(x), vreinterpret_u32_u8(y)); x = vreinterpret_u8_u32(t.val[0]); y = vreinterpret_u8_u32(t.val[1]); } + + // sadly can't use interleaved stores here since we only write + // 8 bytes to each scan line! + + // 8x8 8-bit transpose pass 1 + dct_trn8_8(p0, p1); + dct_trn8_8(p2, p3); + dct_trn8_8(p4, p5); + dct_trn8_8(p6, p7); + + // pass 2 + dct_trn8_16(p0, p2); + dct_trn8_16(p1, p3); + dct_trn8_16(p4, p6); + dct_trn8_16(p5, p7); + + // pass 3 + dct_trn8_32(p0, p4); + dct_trn8_32(p1, p5); + dct_trn8_32(p2, p6); + dct_trn8_32(p3, p7); + + // store + vst1_u8(out, p0); out += out_stride; + vst1_u8(out, p1); out += out_stride; + vst1_u8(out, p2); out += out_stride; + vst1_u8(out, p3); out += out_stride; + vst1_u8(out, p4); out += out_stride; + vst1_u8(out, p5); out += out_stride; + vst1_u8(out, p6); out += out_stride; + vst1_u8(out, p7); + +#undef dct_trn8_8 +#undef dct_trn8_16 +#undef dct_trn8_32 + } + +#undef dct_long_mul +#undef dct_long_mac +#undef dct_widen +#undef dct_wadd +#undef dct_wsub +#undef dct_bfly32o +#undef dct_pass +} + +#endif // STBI_NEON + +#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); // consume repeated 0xff fill bytes + 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 = j->img_comp[3].dc_pred = 0; + j->marker = STBI__MARKER_none; + j->todo = j->restart_interval ? j->restart_interval : 0x7fffffff; + j->eob_run = 0; + // 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->progressive) { + if (z->scan_n == 1) { + int i, j; + STBI_SIMD_ALIGN(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) { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc + z->img_comp[n].hd, z->huff_ac + ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; + z->idct_block_kernel(z->img_comp[n].data + z->img_comp[n].w2 * j * 8 + i * 8, z->img_comp[n].w2, data); + // 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); + } + } + } + return 1; + } + else { // interleaved + int i, j, k, x, y; + STBI_SIMD_ALIGN(short, data[64]); + for (j = 0; j < z->img_mcu_y; ++j) { + for (i = 0; i < z->img_mcu_x; ++i) { + // 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; + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc + z->img_comp[n].hd, z->huff_ac + ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; + z->idct_block_kernel(z->img_comp[n].data + z->img_comp[n].w2 * y2 + x2, z->img_comp[n].w2, data); + } + } + } + // 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 (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } + else { + if (z->scan_n == 1) { + int i, j; + int n = z->order[0]; + // 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) { + short* data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + if (z->spec_start == 0) { + if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } + else { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block_prog_ac(z, data, &z->huff_ac[ha], z->fast_ac[ha])) + return 0; + } + // 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 (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + else { // interleaved + int i, j, k, x, y; + 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); + int y2 = (j * z->img_comp[n].v + y); + short* data = z->img_comp[n].coeff + 64 * (x2 + y2 * z->img_comp[n].coeff_w); + if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } + } + } + // 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 (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } +} + +static void stbi__jpeg_dequantize(short* data, stbi__uint16* dequant) +{ + int i; + for (i = 0; i < 64; ++i) + data[i] *= dequant[i]; +} + +static void stbi__jpeg_finish(stbi__jpeg* z) +{ + if (z->progressive) { + // dequantize and idct the data + int i, j, n; + for (n = 0; n < z->s->img_n; ++n) { + int w = (z->img_comp[n].x + 7) >> 3; + int h = (z->img_comp[n].y + 7) >> 3; + for (j = 0; j < h; ++j) { + for (i = 0; i < w; ++i) { + short* data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + stbi__jpeg_dequantize(data, z->dequant[z->img_comp[n].tq]); + z->idct_block_kernel(z->img_comp[n].data + z->img_comp[n].w2 * j * 8 + i * 8, z->img_comp[n].w2, data); + } + } + } + } +} + +static int stbi__process_marker(stbi__jpeg* z, int m) +{ + int L; + switch (m) { + case STBI__MARKER_none: // no marker found + return stbi__err("expected marker", "Corrupt JPEG"); + + 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, sixteen = (p != 0); + int t = q & 15, i; + if (p != 0 && p != 1) 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__uint16)(sixteen ? stbi__get16be(z->s) : stbi__get8(z->s)); + L -= (sixteen ? 129 : 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); + if (tc != 0) + stbi__build_fast_ac(z->fast_ac[th], z->huff_ac + th); + L -= n; + } + return L == 0; + } + + // check for comment block or APP blocks + if ((m >= 0xE0 && m <= 0xEF) || m == 0xFE) { + L = stbi__get16be(z->s); + if (L < 2) { + if (m == 0xFE) + return stbi__err("bad COM len", "Corrupt JPEG"); + else + return stbi__err("bad APP len", "Corrupt JPEG"); + } + L -= 2; + + if (m == 0xE0 && L >= 5) { // JFIF APP0 segment + static const unsigned char tag[5] = { 'J','F','I','F','\0' }; + int ok = 1; + int i; + for (i = 0; i < 5; ++i) + if (stbi__get8(z->s) != tag[i]) + ok = 0; + L -= 5; + if (ok) + z->jfif = 1; + } + else if (m == 0xEE && L >= 12) { // Adobe APP14 segment + static const unsigned char tag[6] = { 'A','d','o','b','e','\0' }; + int ok = 1; + int i; + for (i = 0; i < 6; ++i) + if (stbi__get8(z->s) != tag[i]) + ok = 0; + L -= 6; + if (ok) { + stbi__get8(z->s); // version + stbi__get16be(z->s); // flags0 + stbi__get16be(z->s); // flags1 + z->app14_color_transform = stbi__get8(z->s); // color transform + L -= 6; + } + } + + stbi__skip(z->s, L); + return 1; + } + + return stbi__err("unknown marker", "Corrupt JPEG"); +} + +// after we see 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 SOS component count", "Corrupt JPEG"); + if (Ls != 6 + 2 * z->scan_n) return stbi__err("bad SOS len", "Corrupt JPEG"); + for (i = 0; i < z->scan_n; ++i) { + int id = stbi__get8(z->s), which; + int q = stbi__get8(z->s); + for (which = 0; which < z->s->img_n; ++which) + if (z->img_comp[which].id == id) + break; + if (which == z->s->img_n) return 0; // no match + z->img_comp[which].hd = q >> 4; if (z->img_comp[which].hd > 3) return stbi__err("bad DC huff", "Corrupt JPEG"); + z->img_comp[which].ha = q & 15; if (z->img_comp[which].ha > 3) return stbi__err("bad AC huff", "Corrupt JPEG"); + z->order[i] = which; + } + + { + int aa; + z->spec_start = stbi__get8(z->s); + z->spec_end = stbi__get8(z->s); // should be 63, but might be 0 + aa = stbi__get8(z->s); + z->succ_high = (aa >> 4); + z->succ_low = (aa & 15); + if (z->progressive) { + if (z->spec_start > 63 || z->spec_end > 63 || z->spec_start > z->spec_end || z->succ_high > 13 || z->succ_low > 13) + return stbi__err("bad SOS", "Corrupt JPEG"); + } + else { + if (z->spec_start != 0) return stbi__err("bad SOS", "Corrupt JPEG"); + if (z->succ_high != 0 || z->succ_low != 0) return stbi__err("bad SOS", "Corrupt JPEG"); + z->spec_end = 63; + } + } + + return 1; +} + +static int stbi__free_jpeg_components(stbi__jpeg* z, int ncomp, int why) +{ + int i; + for (i = 0; i < ncomp; ++i) { + if (z->img_comp[i].raw_data) { + STBI_FREE(z->img_comp[i].raw_data); + z->img_comp[i].raw_data = NULL; + z->img_comp[i].data = NULL; + } + if (z->img_comp[i].raw_coeff) { + STBI_FREE(z->img_comp[i].raw_coeff); + z->img_comp[i].raw_coeff = 0; + z->img_comp[i].coeff = 0; + } + if (z->img_comp[i].linebuf) { + STBI_FREE(z->img_comp[i].linebuf); + z->img_comp[i].linebuf = NULL; + } + } + return why; +} + +static int stbi__process_frame_header(stbi__jpeg* z, int scan) +{ + stbi__context* s = z->s; + int Lf, p, i, q, h_max = 1, v_max = 1, c; + Lf = stbi__get16be(s); if (Lf < 11) return stbi__err("bad SOF len", "Corrupt JPEG"); // 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 && c != 4) return stbi__err("bad component count", "Corrupt JPEG"); + s->img_n = c; + for (i = 0; i < c; ++i) { + z->img_comp[i].data = NULL; + z->img_comp[i].linebuf = NULL; + } + + if (Lf != 8 + 3 * s->img_n) return stbi__err("bad SOF len", "Corrupt JPEG"); + + z->rgb = 0; + for (i = 0; i < s->img_n; ++i) { + static const unsigned char rgb[3] = { 'R', 'G', 'B' }; + z->img_comp[i].id = stbi__get8(s); + if (s->img_n == 3 && z->img_comp[i].id == rgb[i]) + ++z->rgb; + q = stbi__get8(s); + z->img_comp[i].h = (q >> 4); if (!z->img_comp[i].h || z->img_comp[i].h > 4) return stbi__err("bad H", "Corrupt JPEG"); + z->img_comp[i].v = q & 15; if (!z->img_comp[i].v || z->img_comp[i].v > 4) return stbi__err("bad V", "Corrupt JPEG"); + z->img_comp[i].tq = stbi__get8(s); if (z->img_comp[i].tq > 3) return stbi__err("bad TQ", "Corrupt JPEG"); + } + + if (scan != STBI__SCAN_load) return 1; + + if (!stbi__mad3sizes_valid(s->img_x, s->img_y, s->img_n, 0)) 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; + // these sizes can't be more than 17 bits + 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 + // + // img_mcu_x, img_mcu_y: <=17 bits; comp[i].h and .v are <=4 (checked earlier) + // so these muls can't overflow with 32-bit ints (which we require) + 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].coeff = 0; + z->img_comp[i].raw_coeff = 0; + z->img_comp[i].linebuf = NULL; + z->img_comp[i].raw_data = stbi__malloc_mad2(z->img_comp[i].w2, z->img_comp[i].h2, 15); + if (z->img_comp[i].raw_data == NULL) + return stbi__free_jpeg_components(z, i + 1, stbi__err("outofmem", "Out of memory")); + // align blocks for idct using mmx/sse + z->img_comp[i].data = (stbi_uc*)(((size_t)z->img_comp[i].raw_data + 15) & ~15); + if (z->progressive) { + // w2, h2 are multiples of 8 (see above) + z->img_comp[i].coeff_w = z->img_comp[i].w2 / 8; + z->img_comp[i].coeff_h = z->img_comp[i].h2 / 8; + z->img_comp[i].raw_coeff = stbi__malloc_mad3(z->img_comp[i].w2, z->img_comp[i].h2, sizeof(short), 15); + if (z->img_comp[i].raw_coeff == NULL) + return stbi__free_jpeg_components(z, i + 1, stbi__err("outofmem", "Out of memory")); + z->img_comp[i].coeff = (short*)(((size_t)z->img_comp[i].raw_coeff + 15) & ~15); + } + } + + return 1; +} + +// use comparisons since in some cases we handle more than one case (e.g. 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 || (x) == 0xc2) +#define stbi__SOS(x) ((x) == 0xda) + +#define stbi__SOF_progressive(x) ((x) == 0xc2) + +static int stbi__decode_jpeg_header(stbi__jpeg* z, int scan) +{ + int m; + z->jfif = 0; + z->app14_color_transform = -1; // valid values are 0,1,2 + z->marker = STBI__MARKER_none; // initialize cached marker to empty + m = stbi__get_marker(z); + if (!stbi__SOI(m)) return stbi__err("no SOI", "Corrupt JPEG"); + if (scan == STBI__SCAN_type) return 1; + m = stbi__get_marker(z); + while (!stbi__SOF(m)) { + if (!stbi__process_marker(z, m)) return 0; + m = stbi__get_marker(z); + while (m == STBI__MARKER_none) { + // some files have extra padding after their blocks, so ok, we'll scan + if (stbi__at_eof(z->s)) return stbi__err("no SOF", "Corrupt JPEG"); + m = stbi__get_marker(z); + } + } + z->progressive = stbi__SOF_progressive(m); + if (!stbi__process_frame_header(z, scan)) return 0; + return 1; +} + +// decode image to YCbCr format +static int stbi__decode_jpeg_image(stbi__jpeg* j) +{ + int m; + for (m = 0; m < 4; m++) { + j->img_comp[m].raw_data = NULL; + j->img_comp[m].raw_coeff = NULL; + } + j->restart_interval = 0; + if (!stbi__decode_jpeg_header(j, STBI__SCAN_load)) return 0; + m = stbi__get_marker(j); + while (!stbi__EOI(m)) { + if (stbi__SOS(m)) { + if (!stbi__process_scan_header(j)) return 0; + if (!stbi__parse_entropy_coded_data(j)) return 0; + if (j->marker == STBI__MARKER_none) { + // 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; + } + } + // if we reach eof without hitting a marker, stbi__get_marker() below will fail and we'll eventually return 0 + } + } + else if (stbi__DNL(m)) { + int Ld = stbi__get16be(j->s); + stbi__uint32 NL = stbi__get16be(j->s); + if (Ld != 4) return stbi__err("bad DNL len", "Corrupt JPEG"); + if (NL != j->s->img_y) return stbi__err("bad DNL height", "Corrupt JPEG"); + } + else { + if (!stbi__process_marker(j, m)) return 0; + } + m = stbi__get_marker(j); + } + if (j->progressive) + stbi__jpeg_finish(j); + return 1; +} + +// 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; +} + +#if defined(STBI_SSE2) || defined(STBI_NEON) +static stbi_uc* stbi__resample_row_hv_2_simd(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 = 0, t0, t1; + + if (w == 1) { + out[0] = out[1] = stbi__div4(3 * in_near[0] + in_far[0] + 2); + return out; + } + + t1 = 3 * in_near[0] + in_far[0]; + // process groups of 8 pixels for as long as we can. + // note we can't handle the last pixel in a row in this loop + // because we need to handle the filter boundary conditions. + for (; i < ((w - 1) & ~7); i += 8) { +#if defined(STBI_SSE2) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + __m128i zero = _mm_setzero_si128(); + __m128i farb = _mm_loadl_epi64((__m128i*) (in_far + i)); + __m128i nearb = _mm_loadl_epi64((__m128i*) (in_near + i)); + __m128i farw = _mm_unpacklo_epi8(farb, zero); + __m128i nearw = _mm_unpacklo_epi8(nearb, zero); + __m128i diff = _mm_sub_epi16(farw, nearw); + __m128i nears = _mm_slli_epi16(nearw, 2); + __m128i curr = _mm_add_epi16(nears, diff); // current row + + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + __m128i prv0 = _mm_slli_si128(curr, 2); + __m128i nxt0 = _mm_srli_si128(curr, 2); + __m128i prev = _mm_insert_epi16(prv0, t1, 0); + __m128i next = _mm_insert_epi16(nxt0, 3 * in_near[i + 8] + in_far[i + 8], 7); + + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + __m128i bias = _mm_set1_epi16(8); + __m128i curs = _mm_slli_epi16(curr, 2); + __m128i prvd = _mm_sub_epi16(prev, curr); + __m128i nxtd = _mm_sub_epi16(next, curr); + __m128i curb = _mm_add_epi16(curs, bias); + __m128i even = _mm_add_epi16(prvd, curb); + __m128i odd = _mm_add_epi16(nxtd, curb); + + // interleave even and odd pixels, then undo scaling. + __m128i int0 = _mm_unpacklo_epi16(even, odd); + __m128i int1 = _mm_unpackhi_epi16(even, odd); + __m128i de0 = _mm_srli_epi16(int0, 4); + __m128i de1 = _mm_srli_epi16(int1, 4); + + // pack and write output + __m128i outv = _mm_packus_epi16(de0, de1); + _mm_storeu_si128((__m128i*) (out + i * 2), outv); +#elif defined(STBI_NEON) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + uint8x8_t farb = vld1_u8(in_far + i); + uint8x8_t nearb = vld1_u8(in_near + i); + int16x8_t diff = vreinterpretq_s16_u16(vsubl_u8(farb, nearb)); + int16x8_t nears = vreinterpretq_s16_u16(vshll_n_u8(nearb, 2)); + int16x8_t curr = vaddq_s16(nears, diff); // current row + + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + int16x8_t prv0 = vextq_s16(curr, curr, 7); + int16x8_t nxt0 = vextq_s16(curr, curr, 1); + int16x8_t prev = vsetq_lane_s16(t1, prv0, 0); + int16x8_t next = vsetq_lane_s16(3 * in_near[i + 8] + in_far[i + 8], nxt0, 7); + + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + int16x8_t curs = vshlq_n_s16(curr, 2); + int16x8_t prvd = vsubq_s16(prev, curr); + int16x8_t nxtd = vsubq_s16(next, curr); + int16x8_t even = vaddq_s16(curs, prvd); + int16x8_t odd = vaddq_s16(curs, nxtd); + + // undo scaling and round, then store with even/odd phases interleaved + uint8x8x2_t o; + o.val[0] = vqrshrun_n_s16(even, 4); + o.val[1] = vqrshrun_n_s16(odd, 4); + vst2_u8(out + i * 2, o); +#endif + + // "previous" value for next iter + t1 = 3 * in_near[i + 7] + in_far[i + 7]; + } + + t0 = t1; + t1 = 3 * in_near[i] + in_far[i]; + out[i * 2] = stbi__div16(3 * t1 + t0 + 8); + + for (++i; i < w; ++i) { + t0 = t1; + t1 = 3 * in_near[i] + in_far[i]; + out[i * 2 - 1] = stbi__div16(3 * t0 + t1 + 8); + out[i * 2] = stbi__div16(3 * t1 + t0 + 8); + } + out[w * 2 - 1] = stbi__div4(t1 + 2); + + STBI_NOTUSED(hs); + + return out; +} +#endif + +static 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; +} + +// this is a reduced-precision calculation of YCbCr-to-RGB introduced +// to make sure the code produces the same results in both SIMD and scalar +#define stbi__float2fixed(x) (((int) ((x) * 4096.0f + 0.5f)) << 8) +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] << 20) + (1 << 19); // rounding + int r, g, b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr * stbi__float2fixed(1.40200f); + g = y_fixed + (cr * -stbi__float2fixed(0.71414f)) + ((cb * -stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb * stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned)r > 255) { if (r < 0) r = 0; else r = 255; } + if ((unsigned)g > 255) { if (g < 0) g = 0; else g = 255; } + if ((unsigned)b > 255) { if (b < 0) b = 0; else b = 255; } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } +} + +#if defined(STBI_SSE2) || defined(STBI_NEON) +static void stbi__YCbCr_to_RGB_simd(stbi_uc* out, stbi_uc const* y, stbi_uc const* pcb, stbi_uc const* pcr, int count, int step) +{ + int i = 0; + +#ifdef STBI_SSE2 + // step == 3 is pretty ugly on the final interleave, and i'm not convinced + // it's useful in practice (you wouldn't use it for textures, for example). + // so just accelerate step == 4 case. + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + __m128i signflip = _mm_set1_epi8(-0x80); + __m128i cr_const0 = _mm_set1_epi16((short)(1.40200f * 4096.0f + 0.5f)); + __m128i cr_const1 = _mm_set1_epi16(-(short)(0.71414f * 4096.0f + 0.5f)); + __m128i cb_const0 = _mm_set1_epi16(-(short)(0.34414f * 4096.0f + 0.5f)); + __m128i cb_const1 = _mm_set1_epi16((short)(1.77200f * 4096.0f + 0.5f)); + __m128i y_bias = _mm_set1_epi8((char)(unsigned char)128); + __m128i xw = _mm_set1_epi16(255); // alpha channel + + for (; i + 7 < count; i += 8) { + // load + __m128i y_bytes = _mm_loadl_epi64((__m128i*) (y + i)); + __m128i cr_bytes = _mm_loadl_epi64((__m128i*) (pcr + i)); + __m128i cb_bytes = _mm_loadl_epi64((__m128i*) (pcb + i)); + __m128i cr_biased = _mm_xor_si128(cr_bytes, signflip); // -128 + __m128i cb_biased = _mm_xor_si128(cb_bytes, signflip); // -128 + + // unpack to short (and left-shift cr, cb by 8) + __m128i yw = _mm_unpacklo_epi8(y_bias, y_bytes); + __m128i crw = _mm_unpacklo_epi8(_mm_setzero_si128(), cr_biased); + __m128i cbw = _mm_unpacklo_epi8(_mm_setzero_si128(), cb_biased); + + // color transform + __m128i yws = _mm_srli_epi16(yw, 4); + __m128i cr0 = _mm_mulhi_epi16(cr_const0, crw); + __m128i cb0 = _mm_mulhi_epi16(cb_const0, cbw); + __m128i cb1 = _mm_mulhi_epi16(cbw, cb_const1); + __m128i cr1 = _mm_mulhi_epi16(crw, cr_const1); + __m128i rws = _mm_add_epi16(cr0, yws); + __m128i gwt = _mm_add_epi16(cb0, yws); + __m128i bws = _mm_add_epi16(yws, cb1); + __m128i gws = _mm_add_epi16(gwt, cr1); + + // descale + __m128i rw = _mm_srai_epi16(rws, 4); + __m128i bw = _mm_srai_epi16(bws, 4); + __m128i gw = _mm_srai_epi16(gws, 4); + + // back to byte, set up for transpose + __m128i brb = _mm_packus_epi16(rw, bw); + __m128i gxb = _mm_packus_epi16(gw, xw); + + // transpose to interleave channels + __m128i t0 = _mm_unpacklo_epi8(brb, gxb); + __m128i t1 = _mm_unpackhi_epi8(brb, gxb); + __m128i o0 = _mm_unpacklo_epi16(t0, t1); + __m128i o1 = _mm_unpackhi_epi16(t0, t1); + + // store + _mm_storeu_si128((__m128i*) (out + 0), o0); + _mm_storeu_si128((__m128i*) (out + 16), o1); + out += 32; + } + } +#endif + +#ifdef STBI_NEON + // in this version, step=3 support would be easy to add. but is there demand? + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + uint8x8_t signflip = vdup_n_u8(0x80); + int16x8_t cr_const0 = vdupq_n_s16((short)(1.40200f * 4096.0f + 0.5f)); + int16x8_t cr_const1 = vdupq_n_s16(-(short)(0.71414f * 4096.0f + 0.5f)); + int16x8_t cb_const0 = vdupq_n_s16(-(short)(0.34414f * 4096.0f + 0.5f)); + int16x8_t cb_const1 = vdupq_n_s16((short)(1.77200f * 4096.0f + 0.5f)); + + for (; i + 7 < count; i += 8) { + // load + uint8x8_t y_bytes = vld1_u8(y + i); + uint8x8_t cr_bytes = vld1_u8(pcr + i); + uint8x8_t cb_bytes = vld1_u8(pcb + i); + int8x8_t cr_biased = vreinterpret_s8_u8(vsub_u8(cr_bytes, signflip)); + int8x8_t cb_biased = vreinterpret_s8_u8(vsub_u8(cb_bytes, signflip)); + + // expand to s16 + int16x8_t yws = vreinterpretq_s16_u16(vshll_n_u8(y_bytes, 4)); + int16x8_t crw = vshll_n_s8(cr_biased, 7); + int16x8_t cbw = vshll_n_s8(cb_biased, 7); + + // color transform + int16x8_t cr0 = vqdmulhq_s16(crw, cr_const0); + int16x8_t cb0 = vqdmulhq_s16(cbw, cb_const0); + int16x8_t cr1 = vqdmulhq_s16(crw, cr_const1); + int16x8_t cb1 = vqdmulhq_s16(cbw, cb_const1); + int16x8_t rws = vaddq_s16(yws, cr0); + int16x8_t gws = vaddq_s16(vaddq_s16(yws, cb0), cr1); + int16x8_t bws = vaddq_s16(yws, cb1); + + // undo scaling, round, convert to byte + uint8x8x4_t o; + o.val[0] = vqrshrun_n_s16(rws, 4); + o.val[1] = vqrshrun_n_s16(gws, 4); + o.val[2] = vqrshrun_n_s16(bws, 4); + o.val[3] = vdup_n_u8(255); + + // store, interleaving r/g/b/a + vst4_u8(out, o); + out += 8 * 4; + } + } +#endif + + for (; i < count; ++i) { + int y_fixed = (y[i] << 20) + (1 << 19); // rounding + int r, g, b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr * stbi__float2fixed(1.40200f); + g = y_fixed + cr * -stbi__float2fixed(0.71414f) + ((cb * -stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb * stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned)r > 255) { if (r < 0) r = 0; else r = 255; } + if ((unsigned)g > 255) { if (g < 0) g = 0; else g = 255; } + if ((unsigned)b > 255) { if (b < 0) b = 0; else b = 255; } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } +} +#endif + +// set up the kernels +static void stbi__setup_jpeg(stbi__jpeg* j) +{ + j->idct_block_kernel = stbi__idct_block; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_row; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2; + +#ifdef STBI_SSE2 + if (stbi__sse2_available()) { + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; + } +#endif + +#ifdef STBI_NEON + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; +#endif +} + +// clean up the temporary component buffers +static void stbi__cleanup_jpeg(stbi__jpeg* j) +{ + stbi__free_jpeg_components(j, j->s->img_n, 0); +} + +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; + +// fast 0..255 * 0..255 => 0..255 rounded multiplication +static stbi_uc stbi__blinn_8x8(stbi_uc x, stbi_uc y) +{ + unsigned int t = x * y + 128; + return (stbi_uc)((t + (t >> 8)) >> 8); +} + +static stbi_uc* load_jpeg_image(stbi__jpeg* z, int* out_x, int* out_y, int* comp, int req_comp) +{ + int n, decode_n, is_rgb; + 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, but leave in YCbCr format + if (!stbi__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 >= 3 ? 3 : 1; + + is_rgb = z->s->img_n == 3 && (z->rgb == 3 || (z->app14_color_transform == 0 && !z->jfif)); + + if (z->s->img_n == 3 && n < 3 && !is_rgb) + 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 = z->resample_row_hv_2_kernel; + else r->resample = stbi__resample_row_generic; + } + + // can't error after this so, this is safe + output = (stbi_uc*)stbi__malloc_mad3(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) { + if (is_rgb) { + for (i = 0; i < z->s->img_x; ++i) { + out[0] = y[i]; + out[1] = coutput[1][i]; + out[2] = coutput[2][i]; + out[3] = 255; + out += n; + } + } + else { + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + } + } + else if (z->s->img_n == 4) { + if (z->app14_color_transform == 0) { // CMYK + for (i = 0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(coutput[0][i], m); + out[1] = stbi__blinn_8x8(coutput[1][i], m); + out[2] = stbi__blinn_8x8(coutput[2][i], m); + out[3] = 255; + out += n; + } + } + else if (z->app14_color_transform == 2) { // YCCK + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + for (i = 0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(255 - out[0], m); + out[1] = stbi__blinn_8x8(255 - out[1], m); + out[2] = stbi__blinn_8x8(255 - out[2], m); + out += n; + } + } + else { // YCbCr + alpha? Ignore the fourth channel for now + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + } + } + else + for (i = 0; i < z->s->img_x; ++i) { + out[0] = out[1] = out[2] = y[i]; + out[3] = 255; // not used if n==3 + out += n; + } + } + else { + if (is_rgb) { + if (n == 1) + for (i = 0; i < z->s->img_x; ++i) + * out++ = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + else { + for (i = 0; i < z->s->img_x; ++i, out += 2) { + out[0] = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + out[1] = 255; + } + } + } + else if (z->s->img_n == 4 && z->app14_color_transform == 0) { + for (i = 0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + stbi_uc r = stbi__blinn_8x8(coutput[0][i], m); + stbi_uc g = stbi__blinn_8x8(coutput[1][i], m); + stbi_uc b = stbi__blinn_8x8(coutput[2][i], m); + out[0] = stbi__compute_y(r, g, b); + out[1] = 255; + out += n; + } + } + else if (z->s->img_n == 4 && z->app14_color_transform == 2) { + for (i = 0; i < z->s->img_x; ++i) { + out[0] = stbi__blinn_8x8(255 - coutput[0][i], coutput[3][i]); + out[1] = 255; + 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 >= 3 ? 3 : 1; // report original components, not output + return output; + } +} + +static void* stbi__jpeg_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri) +{ + unsigned char* result; + stbi__jpeg* j = (stbi__jpeg*)stbi__malloc(sizeof(stbi__jpeg)); + STBI_NOTUSED(ri); + j->s = s; + stbi__setup_jpeg(j); + result = load_jpeg_image(j, x, y, comp, req_comp); + STBI_FREE(j); + return result; +} + +static int stbi__jpeg_test(stbi__context* s) +{ + int r; + stbi__jpeg* j = (stbi__jpeg*)stbi__malloc(sizeof(stbi__jpeg)); + j->s = s; + stbi__setup_jpeg(j); + r = stbi__decode_jpeg_header(j, STBI__SCAN_type); + stbi__rewind(s); + STBI_FREE(j); + return r; +} + +static int stbi__jpeg_info_raw(stbi__jpeg* j, int* x, int* y, int* comp) +{ + if (!stbi__decode_jpeg_header(j, STBI__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 >= 3 ? 3 : 1; + return 1; +} + +static int stbi__jpeg_info(stbi__context* s, int* x, int* y, int* comp) +{ + int result; + stbi__jpeg* j = (stbi__jpeg*)(stbi__malloc(sizeof(stbi__jpeg))); + j->s = s; + result = stbi__jpeg_info_raw(j, x, y, comp); + STBI_FREE(j); + return result; +} +#endif + +// 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 + +#ifndef STBI_NO_ZLIB + +// 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, const 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, 0, sizeof(z->fast)); + for (i = 0; i < num; ++i) + ++sizes[sizelist[i]]; + sizes[0] = 0; + for (i = 1; i < 16; ++i) + if (sizes[i] > (1 << i)) + return stbi__err("bad sizes", "Corrupt PNG"); + code = 0; + for (i = 1; i < 16; ++i) { + next_code[i] = code; + z->firstcode[i] = (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 PNG"); + 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]; + stbi__uint16 fastv = (stbi__uint16)((s << 9) | i); + z->size[c] = (stbi_uc)s; + z->value[c] = (stbi__uint16)i; + if (s <= STBI__ZFAST_BITS) { + int j = stbi__bit_reverse(next_code[s], s); + while (j < (1 << STBI__ZFAST_BITS)) { + z->fast[j] = fastv; + j += (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 |= (unsigned int)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; +} + +static int stbi__zhuffman_decode_slowpath(stbi__zbuf* a, stbi__zhuffman* z) +{ + int b, s, k; + // 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]; +} + +stbi_inline static int stbi__zhuffman_decode(stbi__zbuf* a, stbi__zhuffman* z) +{ + int b, s; + if (a->num_bits < 16) stbi__fill_bits(a); + b = z->fast[a->code_buffer & STBI__ZFAST_MASK]; + if (b) { + s = b >> 9; + a->code_buffer >>= s; + a->num_bits -= s; + return b & 511; + } + return stbi__zhuffman_decode_slowpath(a, z); +} + +static int stbi__zexpand(stbi__zbuf* z, char* zout, int n) // need to make room for n bytes +{ + char* q; + int cur, limit, old_limit; + z->zout = zout; + if (!z->z_expandable) return stbi__err("output buffer limit", "Corrupt PNG"); + cur = (int)(z->zout - z->zout_start); + limit = old_limit = (int)(z->zout_end - z->zout_start); + while (cur + n > limit) + limit *= 2; + q = (char*)STBI_REALLOC_SIZED(z->zout_start, old_limit, limit); + STBI_NOTUSED(old_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 const 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 const 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 const 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 const int stbi__zdist_extra[32] = +{ 0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13 }; + +static int stbi__parse_huffman_block(stbi__zbuf* a) +{ + char* zout = a->zout; + for (;;) { + int z = stbi__zhuffman_decode(a, &a->z_length); + if (z < 256) { + if (z < 0) return stbi__err("bad huffman code", "Corrupt PNG"); // error in huffman codes + if (zout >= a->zout_end) { + if (!stbi__zexpand(a, zout, 1)) return 0; + zout = a->zout; + } + *zout++ = (char)z; + } + else { + stbi_uc* p; + int len, dist; + if (z == 256) { + a->zout = zout; + return 1; + } + z -= 257; + len = stbi__zlength_base[z]; + if (stbi__zlength_extra[z]) len += stbi__zreceive(a, stbi__zlength_extra[z]); + z = stbi__zhuffman_decode(a, &a->z_distance); + if (z < 0) return stbi__err("bad huffman code", "Corrupt PNG"); + dist = stbi__zdist_base[z]; + if (stbi__zdist_extra[z]) dist += stbi__zreceive(a, stbi__zdist_extra[z]); + if (zout - a->zout_start < dist) return stbi__err("bad dist", "Corrupt PNG"); + if (zout + len > a->zout_end) { + if (!stbi__zexpand(a, zout, len)) return 0; + zout = a->zout; + } + p = (stbi_uc*)(zout - dist); + if (dist == 1) { // run of one byte; common in images. + stbi_uc v = *p; + if (len) { do *zout++ = v; while (--len); } + } + else { + if (len) { do *zout++ = *p++; while (--len); } + } + } + } +} + +static int stbi__compute_huffman_codes(stbi__zbuf* a) +{ + static const 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; + int ntot = hlit + hdist; + + 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 < ntot) { + int c = stbi__zhuffman_decode(a, &z_codelength); + if (c < 0 || c >= 19) return stbi__err("bad codelengths", "Corrupt PNG"); + if (c < 16) + lencodes[n++] = (stbi_uc)c; + else { + stbi_uc fill = 0; + if (c == 16) { + c = stbi__zreceive(a, 2) + 3; + if (n == 0) return stbi__err("bad codelengths", "Corrupt PNG"); + fill = lencodes[n - 1]; + } + else if (c == 17) + c = stbi__zreceive(a, 3) + 3; + else { + STBI_ASSERT(c == 18); + c = stbi__zreceive(a, 7) + 11; + } + if (ntot - n < c) return stbi__err("bad codelengths", "Corrupt PNG"); + memset(lencodes + n, fill, c); + n += c; + } + } + if (n != ntot) 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_uncompressed_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, a->zout, len)) return 0; + memcpy(a->zout, a->zbuffer, len); + a->zbuffer += len; + a->zout += len; + return 1; +} + +static int stbi__parse_zlib_header(stbi__zbuf* a) +{ + int cmf = stbi__zget8(a); + int cm = cmf & 15; + /* int cinfo = cmf >> 4; */ + int flg = stbi__zget8(a); + if ((cmf * 256 + flg) % 31 != 0) return stbi__err("bad zlib header", "Corrupt PNG"); // 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; +} + +static const stbi_uc stbi__zdefault_length[288] = +{ + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, 7,7,7,7,7,7,7,7,8,8,8,8,8,8,8,8 +}; +static const stbi_uc stbi__zdefault_distance[32] = +{ + 5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5 +}; +/* +Init algorithm: +{ + 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_uncompressed_block(a)) return 0; + } + else if (type == 3) { + return 0; + } + else { + if (type == 1) { + // use fixed code lengths + 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 { + STBI_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 { + STBI_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 { + STBI_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; +} +#endif + +// 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 + +#ifndef STBI_NO_PNG +typedef struct +{ + stbi__uint32 length; + stbi__uint32 type; +} stbi__pngchunk; + +static stbi__pngchunk stbi__get_chunk_header(stbi__context* s) +{ + stbi__pngchunk c; + c.length = stbi__get32be(s); + c.type = stbi__get32be(s); + return c; +} + +static int stbi__check_png_header(stbi__context* s) +{ + static const 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; + int depth; +} stbi__png; + + +enum { + STBI__F_none = 0, + STBI__F_sub = 1, + STBI__F_up = 2, + STBI__F_avg = 3, + STBI__F_paeth = 4, + // synthetic filters used for first scanline to avoid needing a dummy row of 0s + 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; +} + +static const stbi_uc stbi__depth_scale_table[9] = { 0, 0xff, 0x55, 0, 0x11, 0,0,0, 0x01 }; + +// 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, int depth, int color) +{ + int bytes = (depth == 16 ? 2 : 1); + stbi__context* s = a->s; + stbi__uint32 i, j, stride = x * out_n * bytes; + stbi__uint32 img_len, img_width_bytes; + int k; + int img_n = s->img_n; // copy it into a local for later + + int output_bytes = out_n * bytes; + int filter_bytes = img_n * bytes; + int width = x; + + STBI_ASSERT(out_n == s->img_n || out_n == s->img_n + 1); + a->out = (stbi_uc*)stbi__malloc_mad3(x, y, output_bytes, 0); // extra bytes to write off the end into + if (!a->out) return stbi__err("outofmem", "Out of memory"); + + if (!stbi__mad3sizes_valid(img_n, x, depth, 7)) return stbi__err("too large", "Corrupt PNG"); + img_width_bytes = (((img_n * x * depth) + 7) >> 3); + img_len = (img_width_bytes + 1) * y; + + // we used to check for exact match between raw_len and img_len on non-interlaced PNGs, + // but issue #276 reported a PNG in the wild that had extra data at the end (all zeros), + // so just check for raw_len < img_len always. + if (raw_len < img_len) return stbi__err("not enough pixels", "Corrupt PNG"); + + for (j = 0; j < y; ++j) { + stbi_uc* cur = a->out + stride * j; + stbi_uc* prior; + int filter = *raw++; + + if (filter > 4) + return stbi__err("invalid filter", "Corrupt PNG"); + + if (depth < 8) { + STBI_ASSERT(img_width_bytes <= x); + cur += x * out_n - img_width_bytes; // store output to the rightmost img_len bytes, so we can decode in place + filter_bytes = 1; + width = img_width_bytes; + } + prior = cur - stride; // bugfix: need to compute this after 'cur +=' computation above + + // if first row, use special filter that doesn't sample previous row + if (j == 0) filter = first_row_filter[filter]; + + // handle first byte explicitly + for (k = 0; k < filter_bytes; ++k) { + switch (filter) { + case STBI__F_none: cur[k] = raw[k]; break; + case STBI__F_sub: cur[k] = raw[k]; break; + case STBI__F_up: cur[k] = STBI__BYTECAST(raw[k] + prior[k]); break; + case STBI__F_avg: cur[k] = STBI__BYTECAST(raw[k] + (prior[k] >> 1)); break; + case STBI__F_paeth: cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(0, prior[k], 0)); break; + case STBI__F_avg_first: cur[k] = raw[k]; break; + case STBI__F_paeth_first: cur[k] = raw[k]; break; + } + } + + if (depth == 8) { + if (img_n != out_n) + cur[img_n] = 255; // first pixel + raw += img_n; + cur += out_n; + prior += out_n; + } + else if (depth == 16) { + if (img_n != out_n) { + cur[filter_bytes] = 255; // first pixel top byte + cur[filter_bytes + 1] = 255; // first pixel bottom byte + } + raw += filter_bytes; + cur += output_bytes; + prior += output_bytes; + } + else { + raw += 1; + cur += 1; + prior += 1; + } + + // this is a little gross, so that we don't switch per-pixel or per-component + if (depth < 8 || img_n == out_n) { + int nk = (width - 1) * filter_bytes; +#define STBI__CASE(f) \ + case f: \ + for (k=0; k < nk; ++k) + switch (filter) { + // "none" filter turns into a memcpy here; make that explicit. + case STBI__F_none: memcpy(cur, raw, nk); break; + STBI__CASE(STBI__F_sub) { cur[k] = STBI__BYTECAST(raw[k] + cur[k - filter_bytes]); } break; + STBI__CASE(STBI__F_up) { cur[k] = STBI__BYTECAST(raw[k] + prior[k]); } break; + STBI__CASE(STBI__F_avg) { cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k - filter_bytes]) >> 1)); } break; + STBI__CASE(STBI__F_paeth) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k - filter_bytes], prior[k], prior[k - filter_bytes])); } break; + STBI__CASE(STBI__F_avg_first) { cur[k] = STBI__BYTECAST(raw[k] + (cur[k - filter_bytes] >> 1)); } break; + STBI__CASE(STBI__F_paeth_first) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k - filter_bytes], 0, 0)); } break; + } +#undef STBI__CASE + raw += nk; + } + else { + STBI_ASSERT(img_n + 1 == out_n); +#define STBI__CASE(f) \ + case f: \ + for (i=x-1; i >= 1; --i, cur[filter_bytes]=255,raw+=filter_bytes,cur+=output_bytes,prior+=output_bytes) \ + for (k=0; k < filter_bytes; ++k) + switch (filter) { + STBI__CASE(STBI__F_none) { cur[k] = raw[k]; } break; + STBI__CASE(STBI__F_sub) { cur[k] = STBI__BYTECAST(raw[k] + cur[k - output_bytes]); } break; + STBI__CASE(STBI__F_up) { cur[k] = STBI__BYTECAST(raw[k] + prior[k]); } break; + STBI__CASE(STBI__F_avg) { cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k - output_bytes]) >> 1)); } break; + STBI__CASE(STBI__F_paeth) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k - output_bytes], prior[k], prior[k - output_bytes])); } break; + STBI__CASE(STBI__F_avg_first) { cur[k] = STBI__BYTECAST(raw[k] + (cur[k - output_bytes] >> 1)); } break; + STBI__CASE(STBI__F_paeth_first) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k - output_bytes], 0, 0)); } break; + } +#undef STBI__CASE + + // the loop above sets the high byte of the pixels' alpha, but for + // 16 bit png files we also need the low byte set. we'll do that here. + if (depth == 16) { + cur = a->out + stride * j; // start at the beginning of the row again + for (i = 0; i < x; ++i, cur += output_bytes) { + cur[filter_bytes + 1] = 255; + } + } + } + } + + // we make a separate pass to expand bits to pixels; for performance, + // this could run two scanlines behind the above code, so it won't + // intefere with filtering but will still be in the cache. + if (depth < 8) { + for (j = 0; j < y; ++j) { + stbi_uc* cur = a->out + stride * j; + stbi_uc* in = a->out + stride * j + x * out_n - img_width_bytes; + // unpack 1/2/4-bit into a 8-bit buffer. allows us to keep the common 8-bit path optimal at minimal cost for 1/2/4-bit + // png guarante byte alignment, if width is not multiple of 8/4/2 we'll decode dummy trailing data that will be skipped in the later loop + stbi_uc scale = (color == 0) ? stbi__depth_scale_table[depth] : 1; // scale grayscale values to 0..255 range + + // note that the final byte might overshoot and write more data than desired. + // we can allocate enough data that this never writes out of memory, but it + // could also overwrite the next scanline. can it overwrite non-empty data + // on the next scanline? yes, consider 1-pixel-wide scanlines with 1-bit-per-pixel. + // so we need to explicitly clamp the final ones + + if (depth == 4) { + for (k = x * img_n; k >= 2; k -= 2, ++in) { + *cur++ = scale * ((*in >> 4)); + *cur++ = scale * ((*in) & 0x0f); + } + if (k > 0)* cur++ = scale * ((*in >> 4)); + } + else if (depth == 2) { + for (k = x * img_n; k >= 4; k -= 4, ++in) { + *cur++ = scale * ((*in >> 6)); + *cur++ = scale * ((*in >> 4) & 0x03); + *cur++ = scale * ((*in >> 2) & 0x03); + *cur++ = scale * ((*in) & 0x03); + } + if (k > 0)* cur++ = scale * ((*in >> 6)); + if (k > 1)* cur++ = scale * ((*in >> 4) & 0x03); + if (k > 2)* cur++ = scale * ((*in >> 2) & 0x03); + } + else if (depth == 1) { + for (k = x * img_n; k >= 8; k -= 8, ++in) { + *cur++ = scale * ((*in >> 7)); + *cur++ = scale * ((*in >> 6) & 0x01); + *cur++ = scale * ((*in >> 5) & 0x01); + *cur++ = scale * ((*in >> 4) & 0x01); + *cur++ = scale * ((*in >> 3) & 0x01); + *cur++ = scale * ((*in >> 2) & 0x01); + *cur++ = scale * ((*in >> 1) & 0x01); + *cur++ = scale * ((*in) & 0x01); + } + if (k > 0)* cur++ = scale * ((*in >> 7)); + if (k > 1)* cur++ = scale * ((*in >> 6) & 0x01); + if (k > 2)* cur++ = scale * ((*in >> 5) & 0x01); + if (k > 3)* cur++ = scale * ((*in >> 4) & 0x01); + if (k > 4)* cur++ = scale * ((*in >> 3) & 0x01); + if (k > 5)* cur++ = scale * ((*in >> 2) & 0x01); + if (k > 6)* cur++ = scale * ((*in >> 1) & 0x01); + } + if (img_n != out_n) { + int q; + // insert alpha = 255 + cur = a->out + stride * j; + if (img_n == 1) { + for (q = x - 1; q >= 0; --q) { + cur[q * 2 + 1] = 255; + cur[q * 2 + 0] = cur[q]; + } + } + else { + STBI_ASSERT(img_n == 3); + for (q = x - 1; q >= 0; --q) { + cur[q * 4 + 3] = 255; + cur[q * 4 + 2] = cur[q * 3 + 2]; + cur[q * 4 + 1] = cur[q * 3 + 1]; + cur[q * 4 + 0] = cur[q * 3 + 0]; + } + } + } + } + } + else if (depth == 16) { + // force the image data from big-endian to platform-native. + // this is done in a separate pass due to the decoding relying + // on the data being untouched, but could probably be done + // per-line during decode if care is taken. + stbi_uc* cur = a->out; + stbi__uint16* cur16 = (stbi__uint16*)cur; + + for (i = 0; i < x * y * out_n; ++i, cur16++, cur += 2) { + *cur16 = (cur[0] << 8) | cur[1]; + } + } + + return 1; +} + +static int stbi__create_png_image(stbi__png* a, stbi_uc* image_data, stbi__uint32 image_data_len, int out_n, int depth, int color, int interlaced) +{ + int bytes = (depth == 16 ? 2 : 1); + int out_bytes = out_n * bytes; + stbi_uc* final; + int p; + if (!interlaced) + return stbi__create_png_image_raw(a, image_data, image_data_len, out_n, a->s->img_x, a->s->img_y, depth, color); + + // de-interlacing + final = (stbi_uc*)stbi__malloc_mad3(a->s->img_x, a->s->img_y, out_bytes, 0); + 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) { + stbi__uint32 img_len = ((((a->s->img_n * x * depth) + 7) >> 3) + 1) * y; + if (!stbi__create_png_image_raw(a, image_data, image_data_len, out_n, x, y, depth, color)) { + STBI_FREE(final); + return 0; + } + for (j = 0; j < y; ++j) { + for (i = 0; i < x; ++i) { + int out_y = j * yspc[p] + yorig[p]; + int out_x = i * xspc[p] + xorig[p]; + memcpy(final + out_y * a->s->img_x * out_bytes + out_x * out_bytes, + a->out + (j * x + i) * out_bytes, out_bytes); + } + } + STBI_FREE(a->out); + image_data += img_len; + image_data_len -= img_len; + } + } + a->out = final; + + return 1; +} + +static int stbi__compute_transparency(stbi__png* z, 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__compute_transparency16(stbi__png* z, stbi__uint16 tc[3], int out_n) +{ + stbi__context* s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi__uint16* p = (stbi__uint16*)z->out; + + // compute color-based transparency, assuming we've + // already got 65535 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 : 65535); + 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_mad2(pixel_count, pal_img_n, 0); + 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; + } + } + STBI_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) { + stbi_uc half = a / 2; + p[0] = (p[2] * 255 + half) / a; + p[1] = (p[1] * 255 + half) / a; + p[2] = (t * 255 + half) / 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; + } + } + } +} + +#define STBI__PNG_TYPE(a,b,c,d) (((unsigned) (a) << 24) + ((unsigned) (b) << 16) + ((unsigned) (c) << 8) + (unsigned) (d)) + +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__uint16 tc16[3]; + stbi__uint32 ioff = 0, idata_limit = 0, i, pal_len = 0; + int first = 1, k, interlace = 0, color = 0, is_iphone = 0; + stbi__context* s = z->s; + + z->expanded = NULL; + z->idata = NULL; + z->out = NULL; + + if (!stbi__check_png_header(s)) return 0; + + if (scan == STBI__SCAN_type) return 1; + + for (;;) { + stbi__pngchunk c = stbi__get_chunk_header(s); + switch (c.type) { + case STBI__PNG_TYPE('C', 'g', 'B', 'I'): + is_iphone = 1; + stbi__skip(s, c.length); + break; + case STBI__PNG_TYPE('I', 'H', 'D', 'R'): { + int comp, filter; + if (!first) return stbi__err("multiple IHDR", "Corrupt PNG"); + first = 0; + if (c.length != 13) return stbi__err("bad IHDR len", "Corrupt PNG"); + s->img_x = stbi__get32be(s); if (s->img_x > (1 << 24)) return stbi__err("too large", "Very large image (corrupt?)"); + s->img_y = stbi__get32be(s); if (s->img_y > (1 << 24)) return stbi__err("too large", "Very large image (corrupt?)"); + z->depth = stbi__get8(s); if (z->depth != 1 && z->depth != 2 && z->depth != 4 && z->depth != 8 && z->depth != 16) return stbi__err("1/2/4/8/16-bit only", "PNG not supported: 1/2/4/8/16-bit only"); + color = stbi__get8(s); if (color > 6) return stbi__err("bad ctype", "Corrupt PNG"); + if (color == 3 && z->depth == 16) return stbi__err("bad ctype", "Corrupt PNG"); + if (color == 3) pal_img_n = 3; else if (color & 1) return stbi__err("bad ctype", "Corrupt PNG"); + comp = stbi__get8(s); if (comp) return stbi__err("bad comp method", "Corrupt PNG"); + filter = stbi__get8(s); if (filter) return stbi__err("bad filter method", "Corrupt PNG"); + interlace = stbi__get8(s); if (interlace > 1) return stbi__err("bad interlace method", "Corrupt PNG"); + if (!s->img_x || !s->img_y) return stbi__err("0-pixel image", "Corrupt PNG"); + if (!pal_img_n) { + s->img_n = (color & 2 ? 3 : 1) + (color & 4 ? 1 : 0); + if ((1 << 30) / s->img_x / s->img_n < s->img_y) return stbi__err("too large", "Image too large to decode"); + if (scan == STBI__SCAN_header) return 1; + } + else { + // 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 STBI__PNG_TYPE('P', 'L', 'T', 'E'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (c.length > 256 * 3) return stbi__err("invalid PLTE", "Corrupt PNG"); + pal_len = c.length / 3; + if (pal_len * 3 != c.length) return stbi__err("invalid PLTE", "Corrupt PNG"); + for (i = 0; i < pal_len; ++i) { + palette[i * 4 + 0] = stbi__get8(s); + palette[i * 4 + 1] = stbi__get8(s); + palette[i * 4 + 2] = stbi__get8(s); + palette[i * 4 + 3] = 255; + } + break; + } + + case STBI__PNG_TYPE('t', 'R', 'N', 'S'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (z->idata) return stbi__err("tRNS after IDAT", "Corrupt PNG"); + if (pal_img_n) { + if (scan == STBI__SCAN_header) { s->img_n = 4; return 1; } + if (pal_len == 0) return stbi__err("tRNS before PLTE", "Corrupt PNG"); + if (c.length > pal_len) return stbi__err("bad tRNS len", "Corrupt PNG"); + pal_img_n = 4; + for (i = 0; i < c.length; ++i) + palette[i * 4 + 3] = stbi__get8(s); + } + else { + if (!(s->img_n & 1)) return stbi__err("tRNS with alpha", "Corrupt PNG"); + if (c.length != (stbi__uint32)s->img_n * 2) return stbi__err("bad tRNS len", "Corrupt PNG"); + has_trans = 1; + if (z->depth == 16) { + for (k = 0; k < s->img_n; ++k) tc16[k] = (stbi__uint16)stbi__get16be(s); // copy the values as-is + } + else { + for (k = 0; k < s->img_n; ++k) tc[k] = (stbi_uc)(stbi__get16be(s) & 255) * stbi__depth_scale_table[z->depth]; // non 8-bit images will be larger + } + } + break; + } + + case STBI__PNG_TYPE('I', 'D', 'A', 'T'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (pal_img_n && !pal_len) return stbi__err("no PLTE", "Corrupt PNG"); + if (scan == STBI__SCAN_header) { s->img_n = pal_img_n; return 1; } + if ((int)(ioff + c.length) < (int)ioff) return 0; + if (ioff + c.length > idata_limit) { + stbi__uint32 idata_limit_old = 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; + STBI_NOTUSED(idata_limit_old); + p = (stbi_uc*)STBI_REALLOC_SIZED(z->idata, idata_limit_old, idata_limit); if (p == NULL) return stbi__err("outofmem", "Out of memory"); + z->idata = p; + } + if (!stbi__getn(s, z->idata + ioff, c.length)) return stbi__err("outofdata", "Corrupt PNG"); + ioff += c.length; + break; + } + + case STBI__PNG_TYPE('I', 'E', 'N', 'D'): { + stbi__uint32 raw_len, bpl; + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (scan != STBI__SCAN_load) return 1; + if (z->idata == NULL) return stbi__err("no IDAT", "Corrupt PNG"); + // initial guess for decoded data size to avoid unnecessary reallocs + bpl = (s->img_x * z->depth + 7) / 8; // bytes per line, per component + raw_len = bpl * s->img_y * s->img_n /* pixels */ + s->img_y /* filter mode per row */; + z->expanded = (stbi_uc*)stbi_zlib_decode_malloc_guesssize_headerflag((char*)z->idata, ioff, raw_len, (int*)& raw_len, !is_iphone); + if (z->expanded == NULL) return 0; // zlib should set error + STBI_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, z->depth, color, interlace)) return 0; + if (has_trans) { + if (z->depth == 16) { + if (!stbi__compute_transparency16(z, tc16, s->img_out_n)) return 0; + } + else { + 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; + } + else if (has_trans) { + // non-paletted image with tRNS -> source image has (constant) alpha + ++s->img_n; + } + STBI_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 void* stbi__do_png(stbi__png* p, int* x, int* y, int* n, int req_comp, stbi__result_info* ri) +{ + void* result = NULL; + if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); + if (stbi__parse_png_file(p, STBI__SCAN_load, req_comp)) { + if (p->depth < 8) + ri->bits_per_channel = 8; + else + ri->bits_per_channel = p->depth; + result = p->out; + p->out = NULL; + if (req_comp && req_comp != p->s->img_out_n) { + if (ri->bits_per_channel == 8) + result = stbi__convert_format((unsigned char*)result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); + else + result = stbi__convert_format16((stbi__uint16*)result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); + p->s->img_out_n = req_comp; + if (result == NULL) return result; + } + *x = p->s->img_x; + *y = p->s->img_y; + if (n)* n = p->s->img_n; + } + STBI_FREE(p->out); p->out = NULL; + STBI_FREE(p->expanded); p->expanded = NULL; + STBI_FREE(p->idata); p->idata = NULL; + + return result; +} + +static void* stbi__png_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri) +{ + stbi__png p; + p.s = s; + return stbi__do_png(&p, x, y, comp, req_comp, ri); +} + +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, STBI__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); +} + +static int stbi__png_is16(stbi__context* s) +{ + stbi__png p; + p.s = s; + if (!stbi__png_info_raw(&p, NULL, NULL, NULL)) + return 0; + if (p.depth != 16) { + stbi__rewind(p.s); + return 0; + } + return 1; +} +#endif + +// Microsoft/Windows BMP image + +#ifndef STBI_NO_BMP +static int stbi__bmp_test_raw(stbi__context* s) +{ + int r; + int sz; + if (stbi__get8(s) != 'B') return 0; + if (stbi__get8(s) != 'M') return 0; + stbi__get32le(s); // 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; +} + +// extract an arbitrarily-aligned N-bit value (N=bits) +// from v, and then make it 8-bits long and fractionally +// extend it to full full range. +static int stbi__shiftsigned(int v, int shift, int bits) +{ + static unsigned int mul_table[9] = { + 0, + 0xff/*0b11111111*/, 0x55/*0b01010101*/, 0x49/*0b01001001*/, 0x11/*0b00010001*/, + 0x21/*0b00100001*/, 0x41/*0b01000001*/, 0x81/*0b10000001*/, 0x01/*0b00000001*/, + }; + static unsigned int shift_table[9] = { + 0, 0,0,1,0,2,4,6,0, + }; + if (shift < 0) + v <<= -shift; + else + v >>= shift; + STBI_ASSERT(v >= 0 && v < 256); + v >>= (8 - bits); + STBI_ASSERT(bits >= 0 && bits <= 8); + return (int)((unsigned)v * mul_table[bits]) >> shift_table[bits]; +} + +typedef struct +{ + int bpp, offset, hsz; + unsigned int mr, mg, mb, ma, all_a; +} stbi__bmp_data; + +static void* stbi__bmp_parse_header(stbi__context* s, stbi__bmp_data* info) +{ + int 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 + info->offset = stbi__get32le(s); + info->hsz = hsz = stbi__get32le(s); + info->mr = info->mg = info->mb = info->ma = 0; + + 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"); + info->bpp = stbi__get16le(s); + if (hsz != 12) { + int 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 (info->bpp == 16 || info->bpp == 32) { + if (compress == 0) { + if (info->bpp == 32) { + info->mr = 0xffu << 16; + info->mg = 0xffu << 8; + info->mb = 0xffu << 0; + info->ma = 0xffu << 24; + info->all_a = 0; // if all_a is 0 at end, then we loaded alpha channel but it was all 0 + } + else { + info->mr = 31u << 10; + info->mg = 31u << 5; + info->mb = 31u << 0; + } + } + else if (compress == 3) { + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + // not documented, but generated by photoshop and handled by mspaint + if (info->mr == info->mg && info->mg == info->mb) { + // ?!?!? + return stbi__errpuc("bad BMP", "bad BMP"); + } + } + else + return stbi__errpuc("bad BMP", "bad BMP"); + } + } + else { + int i; + if (hsz != 108 && hsz != 124) + return stbi__errpuc("bad BMP", "bad BMP"); + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + info->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 + } + } + } + return (void*)1; +} + + +static void* stbi__bmp_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri) +{ + stbi_uc* out; + unsigned int mr = 0, mg = 0, mb = 0, ma = 0, all_a; + stbi_uc pal[256][4]; + int psize = 0, i, j, width; + int flip_vertically, pad, target; + stbi__bmp_data info; + STBI_NOTUSED(ri); + + info.all_a = 255; + if (stbi__bmp_parse_header(s, &info) == NULL) + return NULL; // error code already set + + flip_vertically = ((int)s->img_y) > 0; + s->img_y = abs((int)s->img_y); + + mr = info.mr; + mg = info.mg; + mb = info.mb; + ma = info.ma; + all_a = info.all_a; + + if (info.hsz == 12) { + if (info.bpp < 24) + psize = (info.offset - 14 - 24) / 3; + } + else { + if (info.bpp < 16) + psize = (info.offset - 14 - info.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 + + // sanity-check size + if (!stbi__mad3sizes_valid(target, s->img_x, s->img_y, 0)) + return stbi__errpuc("too large", "Corrupt BMP"); + + out = (stbi_uc*)stbi__malloc_mad3(target, s->img_x, s->img_y, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + if (info.bpp < 16) { + int z = 0; + if (psize == 0 || psize > 256) { STBI_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 (info.hsz != 12) stbi__get8(s); + pal[i][3] = 255; + } + stbi__skip(s, info.offset - 14 - info.hsz - psize * (info.hsz == 12 ? 3 : 4)); + if (info.bpp == 1) width = (s->img_x + 7) >> 3; + else if (info.bpp == 4) width = (s->img_x + 1) >> 1; + else if (info.bpp == 8) width = s->img_x; + else { STBI_FREE(out); return stbi__errpuc("bad bpp", "Corrupt BMP"); } + pad = (-width) & 3; + if (info.bpp == 1) { + for (j = 0; j < (int)s->img_y; ++j) { + int bit_offset = 7, v = stbi__get8(s); + for (i = 0; i < (int)s->img_x; ++i) { + int color = (v >> bit_offset) & 0x1; + out[z++] = pal[color][0]; + out[z++] = pal[color][1]; + out[z++] = pal[color][2]; + if ((--bit_offset) < 0) { + bit_offset = 7; + v = stbi__get8(s); + } + } + stbi__skip(s, pad); + } + } + else { + 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 (info.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 = (info.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, info.offset - 14 - info.hsz); + if (info.bpp == 24) width = 3 * s->img_x; + else if (info.bpp == 16) width = 2 * s->img_x; + else /* bpp = 32 and pad = 0 */ width = 0; + pad = (-width) & 3; + if (info.bpp == 24) { + easy = 1; + } + else if (info.bpp == 32) { + if (mb == 0xff && mg == 0xff00 && mr == 0x00ff0000 && ma == 0xff000000) + easy = 2; + } + if (!easy) { + if (!mr || !mg || !mb) { STBI_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); + all_a |= a; + if (target == 4) out[z++] = a; + } + } + else { + int bpp = info.bpp; + for (i = 0; i < (int)s->img_x; ++i) { + stbi__uint32 v = (bpp == 16 ? (stbi__uint32)stbi__get16le(s) : stbi__get32le(s)); + unsigned 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); + all_a |= a; + if (target == 4) out[z++] = STBI__BYTECAST(a); + } + } + stbi__skip(s, pad); + } + } + + // if alpha channel is all 0s, replace with all 255s + if (target == 4 && all_a == 0) + for (i = 4 * s->img_x * s->img_y - 1; i >= 0; i -= 4) + out[i] = 255; + + 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; +} +#endif + +// Targa Truevision - TGA +// by Jonathan Dummer +#ifndef STBI_NO_TGA +// returns STBI_rgb or whatever, 0 on error +static int stbi__tga_get_comp(int bits_per_pixel, int is_grey, int* is_rgb16) +{ + // only RGB or RGBA (incl. 16bit) or grey allowed + if (is_rgb16)* is_rgb16 = 0; + switch (bits_per_pixel) { + case 8: return STBI_grey; + case 16: if (is_grey) return STBI_grey_alpha; + // fallthrough + case 15: if (is_rgb16) * is_rgb16 = 1; + return STBI_rgb; + case 24: // fallthrough + case 32: return bits_per_pixel / 8; + default: return 0; + } +} + +static int stbi__tga_info(stbi__context* s, int* x, int* y, int* comp) +{ + int tga_w, tga_h, tga_comp, tga_image_type, tga_bits_per_pixel, tga_colormap_bpp; + int sz, tga_colormap_type; + stbi__get8(s); // discard Offset + tga_colormap_type = stbi__get8(s); // colormap type + if (tga_colormap_type > 1) { + stbi__rewind(s); + return 0; // only RGB or indexed allowed + } + tga_image_type = stbi__get8(s); // image type + if (tga_colormap_type == 1) { // colormapped (paletted) image + if (tga_image_type != 1 && tga_image_type != 9) { + stbi__rewind(s); + return 0; + } + stbi__skip(s, 4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ((sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32)) { + stbi__rewind(s); + return 0; + } + stbi__skip(s, 4); // skip image x and y origin + tga_colormap_bpp = sz; + } + else { // "normal" image w/o colormap - only RGB or grey allowed, +/- RLE + if ((tga_image_type != 2) && (tga_image_type != 3) && (tga_image_type != 10) && (tga_image_type != 11)) { + stbi__rewind(s); + return 0; // only RGB or grey allowed, +/- RLE + } + stbi__skip(s, 9); // skip colormap specification and image x/y origin + tga_colormap_bpp = 0; + } + 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 + } + tga_bits_per_pixel = stbi__get8(s); // bits per pixel + stbi__get8(s); // ignore alpha bits + if (tga_colormap_bpp != 0) { + if ((tga_bits_per_pixel != 8) && (tga_bits_per_pixel != 16)) { + // when using a colormap, tga_bits_per_pixel is the size of the indexes + // I don't think anything but 8 or 16bit indexes makes sense + stbi__rewind(s); + return 0; + } + tga_comp = stbi__tga_get_comp(tga_colormap_bpp, 0, NULL); + } + else { + tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3) || (tga_image_type == 11), NULL); + } + if (!tga_comp) { + stbi__rewind(s); + return 0; + } + if (x)* x = tga_w; + if (y)* y = tga_h; + if (comp)* comp = tga_comp; + return 1; // seems to have passed everything +} + +static int stbi__tga_test(stbi__context* s) +{ + int res = 0; + int sz, tga_color_type; + stbi__get8(s); // discard Offset + tga_color_type = stbi__get8(s); // color type + if (tga_color_type > 1) goto errorEnd; // only RGB or indexed allowed + sz = stbi__get8(s); // image type + if (tga_color_type == 1) { // colormapped (paletted) image + if (sz != 1 && sz != 9) goto errorEnd; // colortype 1 demands image type 1 or 9 + stbi__skip(s, 4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ((sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32)) goto errorEnd; + stbi__skip(s, 4); // skip image x and y origin + } + else { // "normal" image w/o colormap + if ((sz != 2) && (sz != 3) && (sz != 10) && (sz != 11)) goto errorEnd; // only RGB or grey allowed, +/- RLE + stbi__skip(s, 9); // skip colormap specification and image x/y origin + } + if (stbi__get16le(s) < 1) goto errorEnd; // test width + if (stbi__get16le(s) < 1) goto errorEnd; // test height + sz = stbi__get8(s); // bits per pixel + if ((tga_color_type == 1) && (sz != 8) && (sz != 16)) goto errorEnd; // for colormapped images, bpp is size of an index + if ((sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32)) goto errorEnd; + + res = 1; // if we got this far, everything's good and we can return 1 instead of 0 + +errorEnd: + stbi__rewind(s); + return res; +} + +// read 16bit value and convert to 24bit RGB +static void stbi__tga_read_rgb16(stbi__context* s, stbi_uc* out) +{ + stbi__uint16 px = (stbi__uint16)stbi__get16le(s); + stbi__uint16 fiveBitMask = 31; + // we have 3 channels with 5bits each + int r = (px >> 10) & fiveBitMask; + int g = (px >> 5) & fiveBitMask; + int b = px & fiveBitMask; + // Note that this saves the data in RGB(A) order, so it doesn't need to be swapped later + out[0] = (stbi_uc)((r * 255) / 31); + out[1] = (stbi_uc)((g * 255) / 31); + out[2] = (stbi_uc)((b * 255) / 31); + + // some people claim that the most significant bit might be used for alpha + // (possibly if an alpha-bit is set in the "image descriptor byte") + // but that only made 16bit test images completely translucent.. + // so let's treat all 15 and 16bit TGAs as RGB with no alpha. +} + +static void* stbi__tga_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri) +{ + // 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_rgb16 = 0; + int tga_inverted = stbi__get8(s); + // int tga_alpha_bits = tga_inverted & 15; // the 4 lowest bits - unused (useless?) + // image data + unsigned char* tga_data; + unsigned char* tga_palette = NULL; + int i, j; + unsigned char raw_data[4] = { 0 }; + int RLE_count = 0; + int RLE_repeating = 0; + int read_next_pixel = 1; + STBI_NOTUSED(ri); + + // do a tiny bit of precessing + if (tga_image_type >= 8) + { + tga_image_type -= 8; + tga_is_RLE = 1; + } + tga_inverted = 1 - ((tga_inverted >> 5) & 1); + + // If I'm paletted, then I'll use the number of bits from the palette + if (tga_indexed) tga_comp = stbi__tga_get_comp(tga_palette_bits, 0, &tga_rgb16); + else tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3), &tga_rgb16); + + if (!tga_comp) // shouldn't really happen, stbi__tga_test() should have ensured basic consistency + return stbi__errpuc("bad format", "Can't find out TGA pixelformat"); + + // tga info + *x = tga_width; + *y = tga_height; + if (comp)* comp = tga_comp; + + if (!stbi__mad3sizes_valid(tga_width, tga_height, tga_comp, 0)) + return stbi__errpuc("too large", "Corrupt TGA"); + + tga_data = (unsigned char*)stbi__malloc_mad3(tga_width, tga_height, tga_comp, 0); + 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 && !tga_rgb16) { + for (i = 0; i < tga_height; ++i) { + int row = tga_inverted ? tga_height - i - 1 : i; + stbi_uc* tga_row = tga_data + row * 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_mad2(tga_palette_len, tga_comp, 0); + if (!tga_palette) { + STBI_FREE(tga_data); + return stbi__errpuc("outofmem", "Out of memory"); + } + if (tga_rgb16) { + stbi_uc* pal_entry = tga_palette; + STBI_ASSERT(tga_comp == STBI_rgb); + for (i = 0; i < tga_palette_len; ++i) { + stbi__tga_read_rgb16(s, pal_entry); + pal_entry += tga_comp; + } + } + else if (!stbi__getn(s, tga_palette, tga_palette_len * tga_comp)) { + STBI_FREE(tga_data); + STBI_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 index, then perform the lookup + int pal_idx = (tga_bits_per_pixel == 8) ? stbi__get8(s) : stbi__get16le(s); + if (pal_idx >= tga_palette_len) { + // invalid index + pal_idx = 0; + } + pal_idx *= tga_comp; + for (j = 0; j < tga_comp; ++j) { + raw_data[j] = tga_palette[pal_idx + j]; + } + } + else if (tga_rgb16) { + STBI_ASSERT(tga_comp == STBI_rgb); + stbi__tga_read_rgb16(s, raw_data); + } + else { + // read in the data raw + for (j = 0; j < tga_comp; ++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) + { + STBI_FREE(tga_palette); + } + } + + // swap RGB - if the source data was RGB16, it already is in the right order + if (tga_comp >= 3 && !tga_rgb16) + { + 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; +} +#endif + +// ************************************************************************************************* +// Photoshop PSD loader -- PD by Thatcher Ulrich, integration by Nicolas Schulz, tweaked by STB + +#ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context* s) +{ + int r = (stbi__get32be(s) == 0x38425053); + stbi__rewind(s); + return r; +} + +static int stbi__psd_decode_rle(stbi__context* s, stbi_uc* p, int pixelCount) +{ + int count, nleft, len; + + count = 0; + while ((nleft = pixelCount - count) > 0) { + len = stbi__get8(s); + if (len == 128) { + // No-op. + } + else if (len < 128) { + // Copy next len+1 bytes literally. + len++; + if (len > nleft) return 0; // corrupt data + 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 = 257 - len; + if (len > nleft) return 0; // corrupt data + val = stbi__get8(s); + count += len; + while (len) { + *p = val; + p += 4; + len--; + } + } + } + + return 1; +} + +static void* stbi__psd_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri, int bpc) +{ + int pixelCount; + int channelCount, compression; + int channel, i; + int bitdepth; + int w, h; + stbi_uc* out; + STBI_NOTUSED(ri); + + // 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. + bitdepth = stbi__get16be(s); + if (bitdepth != 8 && bitdepth != 16) + return stbi__errpuc("unsupported bit depth", "PSD bit depth is not 8 or 16 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"); + + // Check size + if (!stbi__mad3sizes_valid(4, w, h, 0)) + return stbi__errpuc("too large", "Corrupt PSD"); + + // Create the destination image. + + if (!compression && bitdepth == 16 && bpc == 16) { + out = (stbi_uc*)stbi__malloc_mad3(8, w, h, 0); + ri->bits_per_channel = 16; + } + else + 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 += 4) + * p = (channel == 3 ? 255 : 0); + } + else { + // Read the RLE data. + if (!stbi__psd_decode_rle(s, p, pixelCount)) { + STBI_FREE(out); + return stbi__errpuc("corrupt", "bad RLE data"); + } + } + } + + } + 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 (or 16-bit) value for each pixel in the image. + + // Read the data by channel. + for (channel = 0; channel < 4; channel++) { + if (channel >= channelCount) { + // Fill this channel with default data. + if (bitdepth == 16 && bpc == 16) { + stbi__uint16* q = ((stbi__uint16*)out) + channel; + stbi__uint16 val = channel == 3 ? 65535 : 0; + for (i = 0; i < pixelCount; i++, q += 4) + * q = val; + } + else { + stbi_uc* p = out + channel; + stbi_uc val = channel == 3 ? 255 : 0; + for (i = 0; i < pixelCount; i++, p += 4) + * p = val; + } + } + else { + if (ri->bits_per_channel == 16) { // output bpc + stbi__uint16* q = ((stbi__uint16*)out) + channel; + for (i = 0; i < pixelCount; i++, q += 4) + * q = (stbi__uint16)stbi__get16be(s); + } + else { + stbi_uc* p = out + channel; + if (bitdepth == 16) { // input bpc + for (i = 0; i < pixelCount; i++, p += 4) + * p = (stbi_uc)(stbi__get16be(s) >> 8); + } + else { + for (i = 0; i < pixelCount; i++, p += 4) + * p = stbi__get8(s); + } + } + } + } + } + + // remove weird white matte from PSD + if (channelCount >= 4) { + if (ri->bits_per_channel == 16) { + for (i = 0; i < w * h; ++i) { + stbi__uint16* pixel = (stbi__uint16*)out + 4 * i; + if (pixel[3] != 0 && pixel[3] != 65535) { + float a = pixel[3] / 65535.0f; + float ra = 1.0f / a; + float inv_a = 65535.0f * (1 - ra); + pixel[0] = (stbi__uint16)(pixel[0] * ra + inv_a); + pixel[1] = (stbi__uint16)(pixel[1] * ra + inv_a); + pixel[2] = (stbi__uint16)(pixel[2] * ra + inv_a); + } + } + } + else { + for (i = 0; i < w * h; ++i) { + unsigned char* pixel = out + 4 * i; + if (pixel[3] != 0 && pixel[3] != 255) { + float a = pixel[3] / 255.0f; + float ra = 1.0f / a; + float inv_a = 255.0f * (1 - ra); + pixel[0] = (unsigned char)(pixel[0] * ra + inv_a); + pixel[1] = (unsigned char)(pixel[1] * ra + inv_a); + pixel[2] = (unsigned char)(pixel[2] * ra + inv_a); + } + } + } + } + + // convert to desired output format + if (req_comp && req_comp != 4) { + if (ri->bits_per_channel == 16) + out = (stbi_uc*)stbi__convert_format16((stbi__uint16*)out, 4, req_comp, w, h); + else + 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 = 4; + *y = h; + *x = w; + + return out; +} +#endif + +// ************************************************************************************************* +// 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/ + +#ifndef STBI_NO_PIC +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; y < height; ++y) { + int packet_idx; + + for (packet_idx = 0; packet_idx < num_packets; ++packet_idx) { + stbi__pic_packet* packet = &packets[packet_idx]; + stbi_uc* dest = result + y * width * 4; + + switch (packet->type) { + default: + return stbi__errpuc("bad format", "packet has bad compression type"); + + case 0: {//uncompressed + int x; + + for (x = 0; x < width; ++x, dest += 4) + if (!stbi__readval(s, packet->channel, dest)) + return 0; + break; + } + + case 1://Pure RLE + { + int left = width, i; + + while (left > 0) { + stbi_uc count, value[4]; + + count = 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; i < count; ++i, dest += 4) + stbi__copyval(packet->channel, 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]; + + 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; i < count; ++i, dest += 4) + stbi__copyval(packet->channel, dest, value); + } + else { // Raw + ++count; + if (count > left) return stbi__errpuc("bad file", "scanline overrun"); + + for (i = 0; i < count; ++i, dest += 4) + if (!stbi__readval(s, packet->channel, dest)) + return 0; + } + left -= count; + } + break; + } + } + } + } + + return result; +} + +static void* stbi__pic_load(stbi__context* s, int* px, int* py, int* comp, int req_comp, stbi__result_info* ri) +{ + stbi_uc* result; + int i, x, y, internal_comp; + STBI_NOTUSED(ri); + + if (!comp) comp = &internal_comp; + + 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 (!stbi__mad3sizes_valid(x, y, 4, 0)) return stbi__errpuc("too large", "PIC 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_mad3(x, y, 4, 0); + memset(result, 0xff, x * y * 4); + + if (!stbi__pic_load_core(s, x, y, comp, result)) { + STBI_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; +} +#endif + +// ************************************************************************************************* +// GIF loader -- public domain by Jean-Marc Lienher -- simplified/shrunk by stb + +#ifndef STBI_NO_GIF +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) + stbi_uc* background; // The current "background" as far as a gif is concerned + stbi_uc* history; + int flags, bgindex, ratio, transparent, eflags; + stbi_uc pal[256][4]; + stbi_uc lpal[256][4]; + stbi__gif_lzw codes[8192]; + 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; + int delay; +} 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 == i ? 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 = (stbi__gif*)stbi__malloc(sizeof(stbi__gif)); + if (!stbi__gif_header(s, g, comp, 1)) { + STBI_FREE(g); + stbi__rewind(s); + return 0; + } + if (x)* x = g->w; + if (y)* y = g->h; + STBI_FREE(g); + return 1; +} + +static void stbi__out_gif_code(stbi__gif* g, stbi__uint16 code) +{ + stbi_uc* p, * c; + int idx; + + // 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; + + idx = g->cur_x + g->cur_y; + p = &g->out[idx]; + g->history[idx / 4] = 1; + + c = &g->color_table[g->codes[code].suffix * 4]; + if (c[3] > 128) { // don't render transparent pixels; + 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, init_code; + stbi__uint32 first; + stbi__int32 codesize, codemask, avail, oldcode, bits, valid_bits, clear; + stbi__gif_lzw* p; + + lzw_cs = stbi__get8(s); + if (lzw_cs > 12) return NULL; + clear = 1 << lzw_cs; + first = 1; + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + bits = 0; + valid_bits = 0; + for (init_code = 0; init_code < clear; init_code++) { + g->codes[init_code].prefix = -1; + g->codes[init_code].first = (stbi_uc)init_code; + g->codes[init_code].suffix = (stbi_uc)init_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 > 8192) { + 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"); + } + } + } +} + +// this function is designed to support animated gifs, although stb_image doesn't support it +// two back is the image from two frames ago, used for a very specific disposal format +static stbi_uc* stbi__gif_load_next(stbi__context* s, stbi__gif* g, int* comp, int req_comp, stbi_uc* two_back) +{ + int dispose; + int first_frame; + int pi; + int pcount; + + // on first frame, any non-written pixels get the background colour (non-transparent) + first_frame = 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); + g->background = (stbi_uc*)stbi__malloc(4 * g->w * g->h); + g->history = (stbi_uc*)stbi__malloc(g->w * g->h); + if (g->out == 0) return stbi__errpuc("outofmem", "Out of memory"); + + // image is treated as "tranparent" at the start - ie, nothing overwrites the current background; + // background colour is only used for pixels that are not rendered first frame, after that "background" + // color refers to teh color that was there the previous frame. + memset(g->out, 0x00, 4 * g->w * g->h); + memset(g->background, 0x00, 4 * g->w * g->h); // state of the background (starts transparent) + memset(g->history, 0x00, g->w * g->h); // pixels that were affected previous frame + first_frame = 1; + } + else { + // second frame - how do we dispoase of the previous one? + dispose = (g->eflags & 0x1C) >> 2; + pcount = g->w * g->h; + + if ((dispose == 3) && (two_back == 0)) { + dispose = 2; // if I don't have an image to revert back to, default to the old background + } + + if (dispose == 3) { // use previous graphic + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { + memcpy(&g->out[pi * 4], &two_back[pi * 4], 4); + } + } + } + else if (dispose == 2) { + // restore what was changed last frame to background before that frame; + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { + memcpy(&g->out[pi * 4], &g->background[pi * 4], 4); + } + } + } + else { + // This is a non-disposal case eithe way, so just + // leave the pixels as is, and they will become the new background + // 1: do not dispose + // 0: not specified. + } + + // background is what out is after the undoing of the previou frame; + memcpy(g->background, g->out, 4 * g->w * g->h); + } + + // clear my history; + memset(g->history, 0x00, g->w * g->h); // pixels that were affected previous frame + + for (;;) { + int tag = stbi__get8(s); + switch (tag) { + 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) { + 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 this was the first frame, + pcount = g->w * g->h; + if (first_frame && (g->bgindex > 0)) { + // if first frame, any pixel not drawn to gets the background color + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi] == 0) { + g->pal[g->bgindex][3] = 255; // just in case it was made transparent, undo that; It will be reset next frame if need be; + memcpy(&g->out[pi * 4], &g->pal[g->bgindex], 4); + } + } + } + + return o; + } + + case 0x21: // Comment Extension. + { + int len; + int ext = stbi__get8(s); + if (ext == 0xF9) { // Graphic Control Extension. + len = stbi__get8(s); + if (len == 4) { + g->eflags = stbi__get8(s); + g->delay = 10 * stbi__get16le(s); // delay - 1/100th of a second, saving as 1/1000ths. + + // unset old transparent + if (g->transparent >= 0) { + g->pal[g->transparent][3] = 255; + } + if (g->eflags & 0x01) { + g->transparent = stbi__get8(s); + if (g->transparent >= 0) { + g->pal[g->transparent][3] = 0; + } + } + else { + // don't need transparent + stbi__skip(s, 1); + g->transparent = -1; + } + } + 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 void* stbi__load_gif_main(stbi__context* s, int** delays, int* x, int* y, int* z, int* comp, int req_comp) +{ + if (stbi__gif_test(s)) { + int layers = 0; + stbi_uc* u = 0; + stbi_uc* out = 0; + stbi_uc* two_back = 0; + stbi__gif g; + int stride; + memset(&g, 0, sizeof(g)); + if (delays) { + *delays = 0; + } + + do { + u = stbi__gif_load_next(s, &g, comp, req_comp, two_back); + if (u == (stbi_uc*)s) u = 0; // end of animated gif marker + + if (u) { + *x = g.w; + *y = g.h; + ++layers; + stride = g.w * g.h * 4; + + if (out) { + out = (stbi_uc*)STBI_REALLOC(out, layers * stride); + if (delays) { + *delays = (int*)STBI_REALLOC(*delays, sizeof(int) * layers); + } + } + else { + out = (stbi_uc*)stbi__malloc(layers * stride); + if (delays) { + *delays = (int*)stbi__malloc(layers * sizeof(int)); + } + } + memcpy(out + ((layers - 1) * stride), u, stride); + if (layers >= 2) { + two_back = out - 2 * stride; + } + + if (delays) { + (*delays)[layers - 1U] = g.delay; + } + } + } while (u != 0); + + // free temp buffer; + STBI_FREE(g.out); + STBI_FREE(g.history); + STBI_FREE(g.background); + + // do the final conversion after loading everything; + if (req_comp && req_comp != 4) + out = stbi__convert_format(out, 4, req_comp, layers * g.w, g.h); + + *z = layers; + return out; + } + else { + return stbi__errpuc("not GIF", "Image was not as a gif type."); + } +} + +static void* stbi__gif_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri) +{ + stbi_uc* u = 0; + stbi__gif g; + memset(&g, 0, sizeof(g)); + + u = stbi__gif_load_next(s, &g, comp, req_comp, 0); + if (u == (stbi_uc*)s) u = 0; // end of animated gif marker + if (u) { + *x = g.w; + *y = g.h; + + // moved conversion to after successful load so that the same + // can be done for multiple frames. + if (req_comp && req_comp != 4) + u = stbi__convert_format(u, 4, req_comp, g.w, g.h); + } + + // free buffers needed for multiple frame loading; + STBI_FREE(g.history); + STBI_FREE(g.background); + + 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); +} +#endif + +// ************************************************************************************************* +// Radiance RGBE HDR loader +// originally by Nicolas Schulz +#ifndef STBI_NO_HDR +static int stbi__hdr_test_core(stbi__context* s, const char* signature) +{ + int i; + for (i = 0; signature[i]; ++i) + if (stbi__get8(s) != signature[i]) + return 0; + stbi__rewind(s); + return 1; +} + +static int stbi__hdr_test(stbi__context* s) +{ + int r = stbi__hdr_test_core(s, "#?RADIANCE\n"); + stbi__rewind(s); + if (!r) { + r = stbi__hdr_test_core(s, "#?RGBE\n"); + 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, stbi__result_info* ri) +{ + 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; + const char* headerToken; + STBI_NOTUSED(ri); + + // Check identifier + headerToken = stbi__hdr_gettoken(s, buffer); + if (strcmp(headerToken, "#?RADIANCE") != 0 && strcmp(headerToken, "#?RGBE") != 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; + + if (!stbi__mad4sizes_valid(width, height, req_comp, sizeof(float), 0)) + return stbi__errpf("too large", "HDR image is too large"); + + // Read data + hdr_data = (float*)stbi__malloc_mad4(width, height, req_comp, sizeof(float), 0); + if (!hdr_data) + return stbi__errpf("outofmem", "Out of memory"); + + // 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; + STBI_FREE(scanline); + goto main_decode_loop; // yes, this makes no sense + } + len <<= 8; + len |= stbi__get8(s); + if (len != width) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("invalid decoded scanline length", "corrupt HDR"); } + if (scanline == NULL) { + scanline = (stbi_uc*)stbi__malloc_mad2(width, 4, 0); + if (!scanline) { + STBI_FREE(hdr_data); + return stbi__errpf("outofmem", "Out of memory"); + } + } + + for (k = 0; k < 4; ++k) { + int nleft; + i = 0; + while ((nleft = width - i) > 0) { + count = stbi__get8(s); + if (count > 128) { + // Run + value = stbi__get8(s); + count -= 128; + if (count > nleft) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = value; + } + else { + // Dump + if (count > nleft) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } + 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); + } + if (scanline) + STBI_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; + int dummy; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + if (stbi__hdr_test(s) == 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 + +#ifndef STBI_NO_BMP +static int stbi__bmp_info(stbi__context* s, int* x, int* y, int* comp) +{ + void* p; + stbi__bmp_data info; + + info.all_a = 255; + p = stbi__bmp_parse_header(s, &info); + stbi__rewind(s); + if (p == NULL) + return 0; + if (x)* x = s->img_x; + if (y)* y = s->img_y; + if (comp)* comp = info.ma ? 4 : 3; + return 1; +} +#endif + +#ifndef STBI_NO_PSD +static int stbi__psd_info(stbi__context* s, int* x, int* y, int* comp) +{ + int channelCount, dummy, depth; + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + 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); + depth = stbi__get16be(s); + if (depth != 8 && depth != 16) { + stbi__rewind(s); + return 0; + } + if (stbi__get16be(s) != 3) { + stbi__rewind(s); + return 0; + } + *comp = 4; + return 1; +} + +static int stbi__psd_is16(stbi__context* s) +{ + int channelCount, depth; + 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; + } + (void)stbi__get32be(s); + (void)stbi__get32be(s); + depth = stbi__get16be(s); + if (depth != 16) { + stbi__rewind(s); + return 0; + } + return 1; +} +#endif + +#ifndef STBI_NO_PIC +static int stbi__pic_info(stbi__context* s, int* x, int* y, int* comp) +{ + int act_comp = 0, num_packets = 0, chained, dummy; + stbi__pic_packet packets[10]; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + if (!stbi__pic_is4(s, "\x53\x80\xF6\x34")) { + stbi__rewind(s); + return 0; + } + + stbi__skip(s, 88); + + *x = stbi__get16be(s); + *y = stbi__get16be(s); + if (stbi__at_eof(s)) { + stbi__rewind(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; +} +#endif + +// ************************************************************************************************* +// Portable Gray Map and Portable Pixel Map loader +// by Ken Miller +// +// PGM: http://netpbm.sourceforge.net/doc/pgm.html +// PPM: http://netpbm.sourceforge.net/doc/ppm.html +// +// Known limitations: +// Does not support comments in the header section +// Does not support ASCII image data (formats P2 and P3) +// Does not support 16-bit-per-channel + +#ifndef STBI_NO_PNM + +static int stbi__pnm_test(stbi__context* s) +{ + char p, t; + p = (char)stbi__get8(s); + t = (char)stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind(s); + return 0; + } + return 1; +} + +static void* stbi__pnm_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri) +{ + stbi_uc* out; + STBI_NOTUSED(ri); + + if (!stbi__pnm_info(s, (int*)& s->img_x, (int*)& s->img_y, (int*)& s->img_n)) + return 0; + + *x = s->img_x; + *y = s->img_y; + if (comp)* comp = s->img_n; + + if (!stbi__mad3sizes_valid(s->img_n, s->img_x, s->img_y, 0)) + return stbi__errpuc("too large", "PNM too large"); + + out = (stbi_uc*)stbi__malloc_mad3(s->img_n, s->img_x, s->img_y, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + stbi__getn(s, out, s->img_n * s->img_x * s->img_y); + + if (req_comp && req_comp != s->img_n) { + out = stbi__convert_format(out, s->img_n, req_comp, s->img_x, s->img_y); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + return out; +} + +static int stbi__pnm_isspace(char c) +{ + return c == ' ' || c == '\t' || c == '\n' || c == '\v' || c == '\f' || c == '\r'; +} + +static void stbi__pnm_skip_whitespace(stbi__context* s, char* c) +{ + for (;;) { + while (!stbi__at_eof(s) && stbi__pnm_isspace(*c)) + * c = (char)stbi__get8(s); + + if (stbi__at_eof(s) || *c != '#') + break; + + while (!stbi__at_eof(s) && *c != '\n' && *c != '\r') + * c = (char)stbi__get8(s); + } +} + +static int stbi__pnm_isdigit(char c) +{ + return c >= '0' && c <= '9'; +} + +static int stbi__pnm_getinteger(stbi__context* s, char* c) +{ + int value = 0; + + while (!stbi__at_eof(s) && stbi__pnm_isdigit(*c)) { + value = value * 10 + (*c - '0'); + *c = (char)stbi__get8(s); + } + + return value; +} + +static int stbi__pnm_info(stbi__context* s, int* x, int* y, int* comp) +{ + int maxv, dummy; + char c, p, t; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + stbi__rewind(s); + + // Get identifier + p = (char)stbi__get8(s); + t = (char)stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind(s); + return 0; + } + + *comp = (t == '6') ? 3 : 1; // '5' is 1-component .pgm; '6' is 3-component .ppm + + c = (char)stbi__get8(s); + stbi__pnm_skip_whitespace(s, &c); + + *x = stbi__pnm_getinteger(s, &c); // read width + stbi__pnm_skip_whitespace(s, &c); + + *y = stbi__pnm_getinteger(s, &c); // read height + stbi__pnm_skip_whitespace(s, &c); + + maxv = stbi__pnm_getinteger(s, &c); // read max value + + if (maxv > 255) + return stbi__err("max value > 255", "PPM image not 8-bit"); + else + return 1; +} +#endif + +static int stbi__info_main(stbi__context* s, int* x, int* y, int* comp) +{ +#ifndef STBI_NO_JPEG + if (stbi__jpeg_info(s, x, y, comp)) return 1; +#endif + +#ifndef STBI_NO_PNG + if (stbi__png_info(s, x, y, comp)) return 1; +#endif + +#ifndef STBI_NO_GIF + if (stbi__gif_info(s, x, y, comp)) return 1; +#endif + +#ifndef STBI_NO_BMP + if (stbi__bmp_info(s, x, y, comp)) return 1; +#endif + +#ifndef STBI_NO_PSD + if (stbi__psd_info(s, x, y, comp)) return 1; +#endif + +#ifndef STBI_NO_PIC + if (stbi__pic_info(s, x, y, comp)) return 1; +#endif + +#ifndef STBI_NO_PNM + if (stbi__pnm_info(s, x, y, comp)) return 1; +#endif + +#ifndef STBI_NO_HDR + if (stbi__hdr_info(s, x, y, comp)) return 1; +#endif + + // test tga last because it's a crappy test! +#ifndef STBI_NO_TGA + if (stbi__tga_info(s, x, y, comp)) + return 1; +#endif + return stbi__err("unknown image type", "Image not of any known type, or corrupt"); +} + +static int stbi__is_16_main(stbi__context* s) +{ +#ifndef STBI_NO_PNG + if (stbi__png_is16(s)) return 1; +#endif + +#ifndef STBI_NO_PSD + if (stbi__psd_is16(s)) return 1; +#endif + + return 0; +} + +#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; +} + +STBIDEF int stbi_is_16_bit(char const* filename) +{ + FILE* f = stbi__fopen(filename, "rb"); + int result; + if (!f) return stbi__err("can't fopen", "Unable to open file"); + result = stbi_is_16_bit_from_file(f); + fclose(f); + return result; +} + +STBIDEF int stbi_is_16_bit_from_file(FILE* f) +{ + int r; + stbi__context s; + long pos = ftell(f); + stbi__start_file(&s, f); + r = stbi__is_16_main(&s); + 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); +} + +STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const* buffer, int len) +{ + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__is_16_main(&s); +} + +STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const* c, void* user) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks*)c, user); + return stbi__is_16_main(&s); +} + +#endif // STB_IMAGE_IMPLEMENTATION + +/* + revision history: + 2.19 (2018-02-11) fix warning + 2.18 (2018-01-30) fix warnings + 2.17 (2018-01-29) change sbti__shiftsigned to avoid clang -O2 bug + 1-bit BMP + *_is_16_bit api + avoid warnings + 2.16 (2017-07-23) all functions have 16-bit variants; + STBI_NO_STDIO works again; + compilation fixes; + fix rounding in unpremultiply; + optimize vertical flip; + disable raw_len validation; + documentation fixes + 2.15 (2017-03-18) fix png-1,2,4 bug; now all Imagenet JPGs decode; + warning fixes; disable run-time SSE detection on gcc; + uniform handling of optional "return" values; + thread-safe initialization of zlib tables + 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs + 2.13 (2016-11-29) add 16-bit API, only supported for PNG right now + 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes + 2.11 (2016-04-02) allocate large structures on the stack + remove white matting for transparent PSD + fix reported channel count for PNG & BMP + re-enable SSE2 in non-gcc 64-bit + support RGB-formatted JPEG + read 16-bit PNGs (only as 8-bit) + 2.10 (2016-01-22) avoid warning introduced in 2.09 by STBI_REALLOC_SIZED + 2.09 (2016-01-16) allow comments in PNM files + 16-bit-per-pixel TGA (not bit-per-component) + info() for TGA could break due to .hdr handling + info() for BMP to shares code instead of sloppy parse + can use STBI_REALLOC_SIZED if allocator doesn't support realloc + code cleanup + 2.08 (2015-09-13) fix to 2.07 cleanup, reading RGB PSD as RGBA + 2.07 (2015-09-13) fix compiler warnings + partial animated GIF support + limited 16-bpc PSD support + #ifdef unused functions + bug with < 92 byte PIC,PNM,HDR,TGA + 2.06 (2015-04-19) fix bug where PSD returns wrong '*comp' value + 2.05 (2015-04-19) fix bug in progressive JPEG handling, fix warning + 2.04 (2015-04-15) try to re-enable SIMD on MinGW 64-bit + 2.03 (2015-04-12) extra corruption checking (mmozeiko) + stbi_set_flip_vertically_on_load (nguillemot) + fix NEON support; fix mingw support + 2.02 (2015-01-19) fix incorrect assert, fix warning + 2.01 (2015-01-17) fix various warnings; suppress SIMD on gcc 32-bit without -msse2 + 2.00b (2014-12-25) fix STBI_MALLOC in progressive JPEG + 2.00 (2014-12-25) optimize JPG, including x86 SSE2 & NEON SIMD (ryg) + progressive JPEG (stb) + PGM/PPM support (Ken Miller) + STBI_MALLOC,STBI_REALLOC,STBI_FREE + GIF bugfix -- seemingly never worked + STBI_NO_*, STBI_ONLY_* + 1.48 (2014-12-14) fix incorrectly-named assert() + 1.47 (2014-12-14) 1/2/4-bit PNG support, both direct and paletted (Omar Cornut & stb) + optimize PNG (ryg) + fix bug in interlaced PNG with user-specified channel count (stb) + 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 (2006-11-19) + first released version +*/ + + +/* +------------------------------------------------------------------------------ +This software is available under 2 licenses -- choose whichever you prefer. +------------------------------------------------------------------------------ +ALTERNATIVE A - MIT License +Copyright (c) 2017 Sean Barrett +Permission is hereby granted, free of charge, to any person obtaining a copy of +this software and associated documentation files (the "Software"), to deal in +the Software without restriction, including without limitation the rights to +use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies +of the Software, and to permit persons to whom the Software is furnished to do +so, subject to the following conditions: +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. +------------------------------------------------------------------------------ +ALTERNATIVE B - Public Domain (www.unlicense.org) +This is free and unencumbered software released into the public domain. +Anyone is free to copy, modify, publish, use, compile, sell, or distribute this +software, either in source code form or as a compiled binary, for any purpose, +commercial or non-commercial, and by any means. +In jurisdictions that recognize copyright laws, the author or authors of this +software dedicate any and all copyright interest in the software to the public +domain. We make this dedication for the benefit of the public at large and to +the detriment of our heirs and successors. We intend this dedication to be an +overt act of relinquishment in perpetuity of all present and future rights to +this software under copyright law. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN +ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION +WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. +------------------------------------------------------------------------------ +*/ diff --git a/src/ThirdParty/stb_image_write.hpp b/src/ThirdParty/stb_image_write.hpp new file mode 100644 index 0000000..764761a --- /dev/null +++ b/src/ThirdParty/stb_image_write.hpp @@ -0,0 +1,7568 @@ +/* stb_image - v2.19 - public domain image loader - http://nothings.org/stb + 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. + + // i.e. it should look like this: + #include ... + #include ... + #include ... + #define STB_IMAGE_IMPLEMENTATION + #include "stb_image.h" + + You can #define STBI_ASSERT(x) before the #include to avoid using assert.h. + And #define STBI_MALLOC, STBI_REALLOC, and STBI_FREE to avoid using malloc,realloc,free + + + QUICK NOTES: + Primarily of interest to game developers and other people who can + avoid problematic images and only need the trivial interface + + JPEG baseline & progressive (12 bpc/arithmetic not supported, same as stock IJG lib) + PNG 1/2/4/8/16-bit-per-channel + + TGA (not sure what subset, if a subset) + BMP non-1bpp, non-RLE + PSD (composited view only, no extra channels, 8/16 bit-per-channel) + + GIF (*comp always reports as 4-channel) + HDR (radiance rgbE format) + PIC (Softimage PIC) + PNM (PPM and PGM binary only) + + Animated GIF still needs a proper API, but here's one way to do it: + http://gist.github.com/urraka/685d9a6340b26b830d49 + + - decode from memory or through FILE (define STBI_NO_STDIO to remove code) + - decode from arbitrary I/O callbacks + - SIMD acceleration on x86/x64 (SSE2) and ARM (NEON) + + Full documentation under "DOCUMENTATION" below. + + +LICENSE + + See end of file for license information. + +RECENT REVISION HISTORY: + + 2.19 (2018-02-11) fix warning + 2.18 (2018-01-30) fix warnings + 2.17 (2018-01-29) bugfix, 1-bit BMP, 16-bitness query, fix warnings + 2.16 (2017-07-23) all functions have 16-bit variants; optimizations; bugfixes + 2.15 (2017-03-18) fix png-1,2,4; all Imagenet JPGs; no runtime SSE detection on GCC + 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs + 2.13 (2016-12-04) experimental 16-bit API, only for PNG so far; fixes + 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes + 2.11 (2016-04-02) 16-bit PNGS; enable SSE2 in non-gcc x64 + RGB-format JPEG; remove white matting in PSD; + allocate large structures on the stack; + correct channel count for PNG & BMP + 2.10 (2016-01-22) avoid warning introduced in 2.09 + 2.09 (2016-01-16) 16-bit TGA; comments in PNM files; STBI_REALLOC_SIZED + + See end of file for full revision history. + + + ============================ Contributors ========================= + + Image formats Extensions, features + Sean Barrett (jpeg, png, bmp) Jetro Lauha (stbi_info) + Nicolas Schulz (hdr, psd) Martin "SpartanJ" Golini (stbi_info) + Jonathan Dummer (tga) James "moose2000" Brown (iPhone PNG) + Jean-Marc Lienher (gif) Ben "Disch" Wenger (io callbacks) + Tom Seddon (pic) Omar Cornut (1/2/4-bit PNG) + Thatcher Ulrich (psd) Nicolas Guillemot (vertical flip) + Ken Miller (pgm, ppm) Richard Mitton (16-bit PSD) + github:urraka (animated gif) Junggon Kim (PNM comments) + Christopher Forseth (animated gif) Daniel Gibson (16-bit TGA) + socks-the-fox (16-bit PNG) + Jeremy Sawicki (handle all ImageNet JPGs) + Optimizations & bugfixes Mikhail Morozov (1-bit BMP) + Fabian "ryg" Giesen Anael Seghezzi (is-16-bit query) + Arseny Kapoulkine + John-Mark Allen + + Bug & warning fixes + Marc LeBlanc David Woo Guillaume George Martins Mozeiko + Christpher Lloyd Jerry Jansson Joseph Thomson Phil Jordan + Dave Moore Roy Eltham Hayaki Saito Nathan Reed + Won Chun Luke Graham Johan Duparc Nick Verigakis + the Horde3D community Thomas Ruf Ronny Chevalier github:rlyeh + Janez Zemva John Bartholomew Michal Cichon github:romigrou + Jonathan Blow Ken Hamada Tero Hanninen github:svdijk + Laurent Gomila Cort Stratton Sergio Gonzalez github:snagar + Aruelien Pocheville Thibault Reuille Cass Everitt github:Zelex + Ryamond Barbiero Paul Du Bois Engin Manap github:grim210 + Aldo Culquicondor Philipp Wiesemann Dale Weiler github:sammyhw + Oriol Ferrer Mesia Josh Tobin Matthew Gregan github:phprus + Julian Raschke Gregory Mullen Baldur Karlsson github:poppolopoppo + Christian Floisand Kevin Schmidt github:darealshinji + Blazej Dariusz Roszkowski github:Michaelangel007 +*/ + +#ifndef STBI_INCLUDE_STB_IMAGE_H +#define STBI_INCLUDE_STB_IMAGE_H + +// DOCUMENTATION +// +// Limitations: +// - no 12-bit-per-channel JPEG +// - no JPEGs with arithmetic coding +// - GIF always returns *comp=4 +// +// Basic usage (see HDR discussion below for HDR usage): +// 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 *channels_in_file -- outputs # of image components in image file +// int desired_channels -- 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, or NULL on an allocation failure or if the image is +// corrupt or invalid. 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 'desired_channels' if desired_channels is non-zero, or +// *channels_in_file otherwise. If desired_channels is non-zero, +// *channels_in_file has the number of components that _would_ have been +// output otherwise. E.g. if you set desired_channels to 4, you will always +// get RGBA output, but you can check *channels_in_file to see if it's trivially +// opaque because e.g. there were only 3 channels in the source image. +// +// 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, *channels_in_file 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. +// +// =========================================================================== +// +// Philosophy +// +// stb libraries are designed with the following priorities: +// +// 1. easy to use +// 2. easy to maintain +// 3. good performance +// +// Sometimes I let "good performance" creep up in priority over "easy to maintain", +// and for best performance I may provide less-easy-to-use APIs that give higher +// performance, in addition to the easy to use ones. Nevertheless, it's important +// to keep in mind that from the standpoint of you, a client of this library, +// all you care about is #1 and #3, and stb libraries DO NOT emphasize #3 above all. +// +// Some secondary priorities arise directly from the first two, some of which +// make more explicit reasons why performance can't be emphasized. +// +// - Portable ("ease of use") +// - Small source code footprint ("easy to maintain") +// - No dependencies ("ease of use") +// +// =========================================================================== +// +// 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). +// +// =========================================================================== +// +// SIMD support +// +// The JPEG decoder will try to automatically use SIMD kernels on x86 when +// supported by the compiler. For ARM Neon support, you must explicitly +// request it. +// +// (The old do-it-yourself SIMD API is no longer supported in the current +// code.) +// +// On x86, SSE2 will automatically be used when available based on a run-time +// test; if not, the generic C versions are used as a fall-back. On ARM targets, +// the typical path is to have separate builds for NEON and non-NEON devices +// (at least this is true for iOS and Android). Therefore, the NEON support is +// toggled by a build flag: define STBI_NEON to get NEON loops. +// +// If for some reason you do not want to use any of SIMD code, or if +// you have issues compiling it, you can disable it entirely by +// defining STBI_NO_SIMD. +// +// =========================================================================== +// +// 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); +// +// =========================================================================== +// +// iPhone PNG support: +// +// By default we convert iphone-formatted PNGs back to RGB, even though +// they are internally encoded differently. 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 (which +// is BGR stored in RGB). +// +// 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). +// +// =========================================================================== +// +// ADDITIONAL CONFIGURATION +// +// - You can suppress implementation of any of the decoders to reduce +// your code footprint by #defining one or more of the following +// symbols before creating the implementation. +// +// STBI_NO_JPEG +// STBI_NO_PNG +// STBI_NO_BMP +// STBI_NO_PSD +// STBI_NO_TGA +// STBI_NO_GIF +// STBI_NO_HDR +// STBI_NO_PIC +// STBI_NO_PNM (.ppm and .pgm) +// +// - You can request *only* certain decoders and suppress all other ones +// (this will be more forward-compatible, as addition of new decoders +// doesn't require you to disable them explicitly): +// +// STBI_ONLY_JPEG +// STBI_ONLY_PNG +// STBI_ONLY_BMP +// STBI_ONLY_PSD +// STBI_ONLY_TGA +// STBI_ONLY_GIF +// STBI_ONLY_HDR +// STBI_ONLY_PIC +// STBI_ONLY_PNM (.ppm and .pgm) +// +// - If you use STBI_NO_PNG (or _ONLY_ without PNG), and you still +// want the zlib decoder to be available, #define STBI_SUPPORT_ZLIB +// + + +#ifndef STBI_NO_STDIO +#include +#endif // STBI_NO_STDIO + +#define STBI_VERSION 1 + +enum +{ + STBI_default = 0, // only used for desired_channels + + STBI_grey = 1, + STBI_grey_alpha = 2, + STBI_rgb = 3, + STBI_rgb_alpha = 4 +}; + +typedef unsigned char stbi_uc; +typedef unsigned short stbi_us; + +#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 + // + + 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; + + //////////////////////////////////// + // + // 8-bits-per-channel interface + // + + STBIDEF stbi_uc* stbi_load_from_memory(stbi_uc const* buffer, int len, int* x, int* y, int* channels_in_file, int desired_channels); + STBIDEF stbi_uc* stbi_load_from_callbacks(stbi_io_callbacks const* clbk, void* user, int* x, int* y, int* channels_in_file, int desired_channels); +#ifndef STBI_NO_GIF + STBIDEF stbi_uc* stbi_load_gif_from_memory(stbi_uc const* buffer, int len, int** delays, int* x, int* y, int* z, int* comp, int req_comp); +#endif + + +#ifndef STBI_NO_STDIO + STBIDEF stbi_uc* stbi_load(char const* filename, int* x, int* y, int* channels_in_file, int desired_channels); + STBIDEF stbi_uc* stbi_load_from_file(FILE* f, int* x, int* y, int* channels_in_file, int desired_channels); + // for stbi_load_from_file, file pointer is left pointing immediately after image +#endif + +//////////////////////////////////// +// +// 16-bits-per-channel interface +// + + STBIDEF stbi_us* stbi_load_16_from_memory(stbi_uc const* buffer, int len, int* x, int* y, int* channels_in_file, int desired_channels); + STBIDEF stbi_us* stbi_load_16_from_callbacks(stbi_io_callbacks const* clbk, void* user, int* x, int* y, int* channels_in_file, int desired_channels); + +#ifndef STBI_NO_STDIO + STBIDEF stbi_us* stbi_load_16(char const* filename, int* x, int* y, int* channels_in_file, int desired_channels); + STBIDEF stbi_us* stbi_load_from_file_16(FILE* f, int* x, int* y, int* channels_in_file, int desired_channels); +#endif + + //////////////////////////////////// + // + // float-per-channel interface + // +#ifndef STBI_NO_LINEAR + STBIDEF float* stbi_loadf_from_memory(stbi_uc const* buffer, int len, int* x, int* y, int* channels_in_file, int desired_channels); + STBIDEF float* stbi_loadf_from_callbacks(stbi_io_callbacks const* clbk, void* user, int* x, int* y, int* channels_in_file, int desired_channels); + +#ifndef STBI_NO_STDIO + STBIDEF float* stbi_loadf(char const* filename, int* x, int* y, int* channels_in_file, int desired_channels); + STBIDEF float* stbi_loadf_from_file(FILE* f, int* x, int* y, int* channels_in_file, int desired_channels); +#endif +#endif + +#ifndef STBI_NO_HDR + STBIDEF void stbi_hdr_to_ldr_gamma(float gamma); + STBIDEF void stbi_hdr_to_ldr_scale(float scale); +#endif // STBI_NO_HDR + +#ifndef STBI_NO_LINEAR + STBIDEF void stbi_ldr_to_hdr_gamma(float gamma); + STBIDEF void stbi_ldr_to_hdr_scale(float scale); +#endif // STBI_NO_LINEAR + + // stbi_is_hdr is always defined, but always returns false if STBI_NO_HDR + 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); + STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const* buffer, int len); + STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const* clbk, void* user); + +#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); + STBIDEF int stbi_is_16_bit(char const* filename); + STBIDEF int stbi_is_16_bit_from_file(FILE* f); +#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); + + // flip the image vertically, so the first pixel in the output array is the bottom left + STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip); + + // 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); + + +#ifdef __cplusplus +} +#endif + +// +// +//// end header file ///////////////////////////////////////////////////// +#endif // STBI_INCLUDE_STB_IMAGE_H + +#ifdef STB_IMAGE_IMPLEMENTATION + +#if defined(STBI_ONLY_JPEG) || defined(STBI_ONLY_PNG) || defined(STBI_ONLY_BMP) \ + || defined(STBI_ONLY_TGA) || defined(STBI_ONLY_GIF) || defined(STBI_ONLY_PSD) \ + || defined(STBI_ONLY_HDR) || defined(STBI_ONLY_PIC) || defined(STBI_ONLY_PNM) \ + || defined(STBI_ONLY_ZLIB) +#ifndef STBI_ONLY_JPEG +#define STBI_NO_JPEG +#endif +#ifndef STBI_ONLY_PNG +#define STBI_NO_PNG +#endif +#ifndef STBI_ONLY_BMP +#define STBI_NO_BMP +#endif +#ifndef STBI_ONLY_PSD +#define STBI_NO_PSD +#endif +#ifndef STBI_ONLY_TGA +#define STBI_NO_TGA +#endif +#ifndef STBI_ONLY_GIF +#define STBI_NO_GIF +#endif +#ifndef STBI_ONLY_HDR +#define STBI_NO_HDR +#endif +#ifndef STBI_ONLY_PIC +#define STBI_NO_PIC +#endif +#ifndef STBI_ONLY_PNM +#define STBI_NO_PNM +#endif +#endif + +#if defined(STBI_NO_PNG) && !defined(STBI_SUPPORT_ZLIB) && !defined(STBI_NO_ZLIB) +#define STBI_NO_ZLIB +#endif + + +#include +#include // ptrdiff_t on osx +#include +#include +#include + +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +#include // ldexp, pow +#endif + +#ifndef STBI_NO_STDIO +#include +#endif + +#ifndef STBI_ASSERT +#include +#define STBI_ASSERT(x) assert(x) +#endif + + +#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 + +#if defined(STBI_MALLOC) && defined(STBI_FREE) && (defined(STBI_REALLOC) || defined(STBI_REALLOC_SIZED)) +// ok +#elif !defined(STBI_MALLOC) && !defined(STBI_FREE) && !defined(STBI_REALLOC) && !defined(STBI_REALLOC_SIZED) +// ok +#else +#error "Must define all or none of STBI_MALLOC, STBI_FREE, and STBI_REALLOC (or STBI_REALLOC_SIZED)." +#endif + +#ifndef STBI_MALLOC +#define STBI_MALLOC(sz) malloc(sz) +#define STBI_REALLOC(p,newsz) realloc(p,newsz) +#define STBI_FREE(p) free(p) +#endif + +#ifndef STBI_REALLOC_SIZED +#define STBI_REALLOC_SIZED(p,oldsz,newsz) STBI_REALLOC(p,newsz) +#endif + +// x86/x64 detection +#if defined(__x86_64__) || defined(_M_X64) +#define STBI__X64_TARGET +#elif defined(__i386) || defined(_M_IX86) +#define STBI__X86_TARGET +#endif + +#if defined(__GNUC__) && defined(STBI__X86_TARGET) && !defined(__SSE2__) && !defined(STBI_NO_SIMD) +// gcc doesn't support sse2 intrinsics unless you compile with -msse2, +// which in turn means it gets to use SSE2 everywhere. This is unfortunate, +// but previous attempts to provide the SSE2 functions with runtime +// detection caused numerous issues. The way architecture extensions are +// exposed in GCC/Clang is, sadly, not really suited for one-file libs. +// New behavior: if compiled with -msse2, we use SSE2 without any +// detection; if not, we don't use it at all. +#define STBI_NO_SIMD +#endif + +#if defined(__MINGW32__) && defined(STBI__X86_TARGET) && !defined(STBI_MINGW_ENABLE_SSE2) && !defined(STBI_NO_SIMD) +// Note that __MINGW32__ doesn't actually mean 32-bit, so we have to avoid STBI__X64_TARGET +// +// 32-bit MinGW wants ESP to be 16-byte aligned, but this is not in the +// Windows ABI and VC++ as well as Windows DLLs don't maintain that invariant. +// As a result, enabling SSE2 on 32-bit MinGW is dangerous when not +// simultaneously enabling "-mstackrealign". +// +// See https://github.com/nothings/stb/issues/81 for more information. +// +// So default to no SSE2 on 32-bit MinGW. If you've read this far and added +// -mstackrealign to your build settings, feel free to #define STBI_MINGW_ENABLE_SSE2. +#define STBI_NO_SIMD +#endif + +#if !defined(STBI_NO_SIMD) && (defined(STBI__X86_TARGET) || defined(STBI__X64_TARGET)) +#define STBI_SSE2 +#include + +#ifdef _MSC_VER + +#if _MSC_VER >= 1400 // not VC6 +#include // __cpuid +static int stbi__cpuid3(void) +{ + int info[4]; + __cpuid(info, 1); + return info[3]; +} +#else +static int stbi__cpuid3(void) +{ + int res; + __asm { + mov eax, 1 + cpuid + mov res, edx + } + return res; +} +#endif + +#define STBI_SIMD_ALIGN(type, name) __declspec(align(16)) type name + +static int stbi__sse2_available(void) +{ + int info3 = stbi__cpuid3(); + return ((info3 >> 26) & 1) != 0; +} +#else // assume GCC-style if not VC++ +#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) + +static int stbi__sse2_available(void) +{ + // If we're even attempting to compile this on GCC/Clang, that means + // -msse2 is on, which means the compiler is allowed to use SSE2 + // instructions at will, and so are we. + return 1; +} +#endif +#endif + +// ARM NEON +#if defined(STBI_NO_SIMD) && defined(STBI_NEON) +#undef STBI_NEON +#endif + +#ifdef STBI_NEON +#include +// assume GCC or Clang on ARM targets +#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16))) +#endif + +#ifndef STBI_SIMD_ALIGN +#define STBI_SIMD_ALIGN(type, name) type name +#endif + +/////////////////////////////////////////////// +// +// 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, * img_buffer_original_end; +} 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 = s->img_buffer_original_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); + s->img_buffer_original_end = s->img_buffer_end; +} + +#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; + s->img_buffer_end = s->img_buffer_original_end; +} + +enum +{ + STBI_ORDER_RGB, + STBI_ORDER_BGR +}; + +typedef struct +{ + int bits_per_channel; + int num_channels; + int channel_order; +} stbi__result_info; + +#ifndef STBI_NO_JPEG +static int stbi__jpeg_test(stbi__context* s); +static void* stbi__jpeg_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri); +static int stbi__jpeg_info(stbi__context* s, int* x, int* y, int* comp); +#endif + +#ifndef STBI_NO_PNG +static int stbi__png_test(stbi__context* s); +static void* stbi__png_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri); +static int stbi__png_info(stbi__context* s, int* x, int* y, int* comp); +static int stbi__png_is16(stbi__context* s); +#endif + +#ifndef STBI_NO_BMP +static int stbi__bmp_test(stbi__context* s); +static void* stbi__bmp_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri); +static int stbi__bmp_info(stbi__context* s, int* x, int* y, int* comp); +#endif + +#ifndef STBI_NO_TGA +static int stbi__tga_test(stbi__context* s); +static void* stbi__tga_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri); +static int stbi__tga_info(stbi__context* s, int* x, int* y, int* comp); +#endif + +#ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context* s); +static void* stbi__psd_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri, int bpc); +static int stbi__psd_info(stbi__context* s, int* x, int* y, int* comp); +static int stbi__psd_is16(stbi__context* s); +#endif + +#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, stbi__result_info* ri); +static int stbi__hdr_info(stbi__context* s, int* x, int* y, int* comp); +#endif + +#ifndef STBI_NO_PIC +static int stbi__pic_test(stbi__context* s); +static void* stbi__pic_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri); +static int stbi__pic_info(stbi__context* s, int* x, int* y, int* comp); +#endif + +#ifndef STBI_NO_GIF +static int stbi__gif_test(stbi__context* s); +static void* stbi__gif_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri); +static void* stbi__load_gif_main(stbi__context* s, int** delays, int* x, int* y, int* z, int* comp, int req_comp); +static int stbi__gif_info(stbi__context* s, int* x, int* y, int* comp); +#endif + +#ifndef STBI_NO_PNM +static int stbi__pnm_test(stbi__context* s); +static void* stbi__pnm_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri); +static int stbi__pnm_info(stbi__context* s, int* x, int* y, int* comp); +#endif + +// 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 STBI_MALLOC(size); +} + +// stb_image uses ints pervasively, including for offset calculations. +// therefore the largest decoded image size we can support with the +// current code, even on 64-bit targets, is INT_MAX. this is not a +// significant limitation for the intended use case. +// +// we do, however, need to make sure our size calculations don't +// overflow. hence a few helper functions for size calculations that +// multiply integers together, making sure that they're non-negative +// and no overflow occurs. + +// return 1 if the sum is valid, 0 on overflow. +// negative terms are considered invalid. +static int stbi__addsizes_valid(int a, int b) +{ + if (b < 0) return 0; + // now 0 <= b <= INT_MAX, hence also + // 0 <= INT_MAX - b <= INTMAX. + // And "a + b <= INT_MAX" (which might overflow) is the + // same as a <= INT_MAX - b (no overflow) + return a <= INT_MAX - b; +} + +// returns 1 if the product is valid, 0 on overflow. +// negative factors are considered invalid. +static int stbi__mul2sizes_valid(int a, int b) +{ + if (a < 0 || b < 0) return 0; + if (b == 0) return 1; // mul-by-0 is always safe + // portable way to check for no overflows in a*b + return a <= INT_MAX / b; +} + +// returns 1 if "a*b + add" has no negative terms/factors and doesn't overflow +static int stbi__mad2sizes_valid(int a, int b, int add) +{ + return stbi__mul2sizes_valid(a, b) && stbi__addsizes_valid(a * b, add); +} + +// returns 1 if "a*b*c + add" has no negative terms/factors and doesn't overflow +static int stbi__mad3sizes_valid(int a, int b, int c, int add) +{ + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a * b, c) && + stbi__addsizes_valid(a * b * c, add); +} + +// returns 1 if "a*b*c*d + add" has no negative terms/factors and doesn't overflow +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +static int stbi__mad4sizes_valid(int a, int b, int c, int d, int add) +{ + return stbi__mul2sizes_valid(a, b) && stbi__mul2sizes_valid(a * b, c) && + stbi__mul2sizes_valid(a * b * c, d) && stbi__addsizes_valid(a * b * c * d, add); +} +#endif + +// mallocs with size overflow checking +static void* stbi__malloc_mad2(int a, int b, int add) +{ + if (!stbi__mad2sizes_valid(a, b, add)) return NULL; + return stbi__malloc(a * b + add); +} + +static void* stbi__malloc_mad3(int a, int b, int c, int add) +{ + if (!stbi__mad3sizes_valid(a, b, c, add)) return NULL; + return stbi__malloc(a * b * c + add); +} + +#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR) +static void* stbi__malloc_mad4(int a, int b, int c, int d, int add) +{ + if (!stbi__mad4sizes_valid(a, b, c, d, add)) return NULL; + return stbi__malloc(a * b * c * d + add); +} +#endif + +// 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 *)(size_t) (stbi__err(x,y)?NULL:NULL)) +#define stbi__errpuc(x,y) ((unsigned char *)(size_t) (stbi__err(x,y)?NULL:NULL)) + +STBIDEF void stbi_image_free(void* retval_from_stbi_load) +{ + STBI_FREE(retval_from_stbi_load); +} + +#ifndef STBI_NO_LINEAR +static float* stbi__ldr_to_hdr(stbi_uc* data, int x, int y, int comp); +#endif + +#ifndef STBI_NO_HDR +static stbi_uc* stbi__hdr_to_ldr(float* data, int x, int y, int comp); +#endif + +static int stbi__vertically_flip_on_load = 0; + +STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip) +{ + stbi__vertically_flip_on_load = flag_true_if_should_flip; +} + +static void* stbi__load_main(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri, int bpc) +{ + memset(ri, 0, sizeof(*ri)); // make sure it's initialized if we add new fields + ri->bits_per_channel = 8; // default is 8 so most paths don't have to be changed + ri->channel_order = STBI_ORDER_RGB; // all current input & output are this, but this is here so we can add BGR order + ri->num_channels = 0; + +#ifndef STBI_NO_JPEG + if (stbi__jpeg_test(s)) return stbi__jpeg_load(s, x, y, comp, req_comp, ri); +#endif +#ifndef STBI_NO_PNG + if (stbi__png_test(s)) return stbi__png_load(s, x, y, comp, req_comp, ri); +#endif +#ifndef STBI_NO_BMP + if (stbi__bmp_test(s)) return stbi__bmp_load(s, x, y, comp, req_comp, ri); +#endif +#ifndef STBI_NO_GIF + if (stbi__gif_test(s)) return stbi__gif_load(s, x, y, comp, req_comp, ri); +#endif +#ifndef STBI_NO_PSD + if (stbi__psd_test(s)) return stbi__psd_load(s, x, y, comp, req_comp, ri, bpc); +#endif +#ifndef STBI_NO_PIC + if (stbi__pic_test(s)) return stbi__pic_load(s, x, y, comp, req_comp, ri); +#endif +#ifndef STBI_NO_PNM + if (stbi__pnm_test(s)) return stbi__pnm_load(s, x, y, comp, req_comp, ri); +#endif + +#ifndef STBI_NO_HDR + if (stbi__hdr_test(s)) { + float* hdr = stbi__hdr_load(s, x, y, comp, req_comp, ri); + return stbi__hdr_to_ldr(hdr, *x, *y, req_comp ? req_comp : *comp); + } +#endif + +#ifndef STBI_NO_TGA + // test tga last because it's a crappy test! + if (stbi__tga_test(s)) + return stbi__tga_load(s, x, y, comp, req_comp, ri); +#endif + + return stbi__errpuc("unknown image type", "Image not of any known type, or corrupt"); +} + +static stbi_uc* stbi__convert_16_to_8(stbi__uint16* orig, int w, int h, int channels) +{ + int i; + int img_len = w * h * channels; + stbi_uc* reduced; + + reduced = (stbi_uc*)stbi__malloc(img_len); + if (reduced == NULL) return stbi__errpuc("outofmem", "Out of memory"); + + for (i = 0; i < img_len; ++i) + reduced[i] = (stbi_uc)((orig[i] >> 8) & 0xFF); // top half of each byte is sufficient approx of 16->8 bit scaling + + STBI_FREE(orig); + return reduced; +} + +static stbi__uint16* stbi__convert_8_to_16(stbi_uc* orig, int w, int h, int channels) +{ + int i; + int img_len = w * h * channels; + stbi__uint16* enlarged; + + enlarged = (stbi__uint16*)stbi__malloc(img_len * 2); + if (enlarged == NULL) return (stbi__uint16*)stbi__errpuc("outofmem", "Out of memory"); + + for (i = 0; i < img_len; ++i) + enlarged[i] = (stbi__uint16)((orig[i] << 8) + orig[i]); // replicate to high and low byte, maps 0->0, 255->0xffff + + STBI_FREE(orig); + return enlarged; +} + +static void stbi__vertical_flip(void* image, int w, int h, int bytes_per_pixel) +{ + int row; + size_t bytes_per_row = (size_t)w * bytes_per_pixel; + stbi_uc temp[2048]; + stbi_uc* bytes = (stbi_uc*)image; + + for (row = 0; row < (h >> 1); row++) { + stbi_uc* row0 = bytes + row * bytes_per_row; + stbi_uc* row1 = bytes + (h - row - 1) * bytes_per_row; + // swap row0 with row1 + size_t bytes_left = bytes_per_row; + while (bytes_left) { + size_t bytes_copy = (bytes_left < sizeof(temp)) ? bytes_left : sizeof(temp); + memcpy(temp, row0, bytes_copy); + memcpy(row0, row1, bytes_copy); + memcpy(row1, temp, bytes_copy); + row0 += bytes_copy; + row1 += bytes_copy; + bytes_left -= bytes_copy; + } + } +} + +static void stbi__vertical_flip_slices(void* image, int w, int h, int z, int bytes_per_pixel) +{ + int slice; + int slice_size = w * h * bytes_per_pixel; + + stbi_uc* bytes = (stbi_uc*)image; + for (slice = 0; slice < z; ++slice) { + stbi__vertical_flip(bytes, w, h, bytes_per_pixel); + bytes += slice_size; + } +} + +static unsigned char* stbi__load_and_postprocess_8bit(stbi__context* s, int* x, int* y, int* comp, int req_comp) +{ + stbi__result_info ri; + void* result = stbi__load_main(s, x, y, comp, req_comp, &ri, 8); + + if (result == NULL) + return NULL; + + if (ri.bits_per_channel != 8) { + STBI_ASSERT(ri.bits_per_channel == 16); + result = stbi__convert_16_to_8((stbi__uint16*)result, *x, *y, req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 8; + } + + // @TODO: move stbi__convert_format to here + + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi_uc)); + } + + return (unsigned char*)result; +} + +static stbi__uint16* stbi__load_and_postprocess_16bit(stbi__context* s, int* x, int* y, int* comp, int req_comp) +{ + stbi__result_info ri; + void* result = stbi__load_main(s, x, y, comp, req_comp, &ri, 16); + + if (result == NULL) + return NULL; + + if (ri.bits_per_channel != 16) { + STBI_ASSERT(ri.bits_per_channel == 8); + result = stbi__convert_8_to_16((stbi_uc*)result, *x, *y, req_comp == 0 ? *comp : req_comp); + ri.bits_per_channel = 16; + } + + // @TODO: move stbi__convert_format16 to here + // @TODO: special case RGB-to-Y (and RGBA-to-YA) for 8-bit-to-16-bit case to keep more precision + + if (stbi__vertically_flip_on_load) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(stbi__uint16)); + } + + return (stbi__uint16*)result; +} + +#if !defined(STBI_NO_HDR) || !defined(STBI_NO_LINEAR) +static void stbi__float_postprocess(float* result, int* x, int* y, int* comp, int req_comp) +{ + if (stbi__vertically_flip_on_load && result != NULL) { + int channels = req_comp ? req_comp : *comp; + stbi__vertical_flip(result, *x, *y, channels * sizeof(float)); + } +} +#endif + +#ifndef STBI_NO_STDIO + +static FILE* stbi__fopen(char const* filename, char const* mode) +{ + FILE* f; +#if defined(_MSC_VER) && _MSC_VER >= 1400 + if (0 != fopen_s(&f, filename, mode)) + f = 0; +#else + f = fopen(filename, mode); +#endif + return f; +} + + +STBIDEF stbi_uc* 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 stbi_uc* 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_and_postprocess_8bit(&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; +} + +STBIDEF stbi__uint16* stbi_load_from_file_16(FILE* f, int* x, int* y, int* comp, int req_comp) +{ + stbi__uint16* result; + stbi__context s; + stbi__start_file(&s, f); + result = stbi__load_and_postprocess_16bit(&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; +} + +STBIDEF stbi_us* stbi_load_16(char const* filename, int* x, int* y, int* comp, int req_comp) +{ + FILE* f = stbi__fopen(filename, "rb"); + stbi__uint16* result; + if (!f) return (stbi_us*)stbi__errpuc("can't fopen", "Unable to open file"); + result = stbi_load_from_file_16(f, x, y, comp, req_comp); + fclose(f); + return result; +} + + +#endif //!STBI_NO_STDIO + +STBIDEF stbi_us* stbi_load_16_from_memory(stbi_uc const* buffer, int len, int* x, int* y, int* channels_in_file, int desired_channels) +{ + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__load_and_postprocess_16bit(&s, x, y, channels_in_file, desired_channels); +} + +STBIDEF stbi_us* stbi_load_16_from_callbacks(stbi_io_callbacks const* clbk, void* user, int* x, int* y, int* channels_in_file, int desired_channels) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks*)clbk, user); + return stbi__load_and_postprocess_16bit(&s, x, y, channels_in_file, desired_channels); +} + +STBIDEF stbi_uc* 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_and_postprocess_8bit(&s, x, y, comp, req_comp); +} + +STBIDEF stbi_uc* 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_and_postprocess_8bit(&s, x, y, comp, req_comp); +} + +#ifndef STBI_NO_GIF +STBIDEF stbi_uc* stbi_load_gif_from_memory(stbi_uc const* buffer, int len, int** delays, int* x, int* y, int* z, int* comp, int req_comp) +{ + unsigned char* result; + stbi__context s; + stbi__start_mem(&s, buffer, len); + + result = (unsigned char*)stbi__load_gif_main(&s, delays, x, y, z, comp, req_comp); + if (stbi__vertically_flip_on_load) { + stbi__vertical_flip_slices(result, *x, *y, *z, *comp); + } + + return result; +} +#endif + +#ifndef STBI_NO_LINEAR +static 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)) { + stbi__result_info ri; + float* hdr_data = stbi__hdr_load(s, x, y, comp, req_comp, &ri); + if (hdr_data) + stbi__float_postprocess(hdr_data, x, y, comp, req_comp); + return hdr_data; + } +#endif + data = stbi__load_and_postprocess_8bit(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"); +} + +STBIDEF 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); +} + +STBIDEF 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 +STBIDEF 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; +} + +STBIDEF 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_LINEAR + +// these is-hdr-or-not is defined independent of whether STBI_NO_LINEAR is +// defined, for API simplicity; if STBI_NO_LINEAR is defined, it always +// reports false! + +STBIDEF 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 + long pos = ftell(f); + int res; + stbi__context s; + stbi__start_file(&s, f); + res = stbi__hdr_test(&s); + fseek(f, pos, SEEK_SET); + return res; +#else + STBI_NOTUSED(f); + 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 + STBI_NOTUSED(clbk); + STBI_NOTUSED(user); + return 0; +#endif +} + +#ifndef STBI_NO_LINEAR +static float stbi__l2h_gamma = 2.2f, stbi__l2h_scale = 1.0f; + +STBIDEF void stbi_ldr_to_hdr_gamma(float gamma) { stbi__l2h_gamma = gamma; } +STBIDEF void stbi_ldr_to_hdr_scale(float scale) { stbi__l2h_scale = scale; } +#endif + +static float stbi__h2l_gamma_i = 1.0f / 2.2f, stbi__h2l_scale_i = 1.0f; + +STBIDEF void stbi_hdr_to_ldr_gamma(float gamma) { stbi__h2l_gamma_i = 1 / gamma; } +STBIDEF void stbi_hdr_to_ldr_scale(float scale) { stbi__h2l_scale_i = 1 / scale; } + + +////////////////////////////////////////////////////////////////////////////// +// +// Common code used by all image loaders +// + +enum +{ + STBI__SCAN_load = 0, + STBI__SCAN_type, + STBI__SCAN_header +}; + +static void stbi__refill_buffer(stbi__context* s) +{ + int n = (s->io.read)(s->io_user_data, (char*)s->buffer_start, s->buflen); + if (n == 0) { + // 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 (n < 0) { + s->img_buffer = s->img_buffer_end; + return; + } + if (s->io.read) { + int blen = (int)(s->img_buffer_end - s->img_buffer); + if (blen < n) { + s->img_buffer = s->img_buffer_end; + (s->io.skip)(s->io_user_data, n - blen); + return; + } + } + s->img_buffer += n; +} + +static int stbi__getn(stbi__context* s, 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); +} + +#if defined(STBI_NO_BMP) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) +// nothing +#else +static int stbi__get16le(stbi__context* s) +{ + int z = stbi__get8(s); + return z + (stbi__get8(s) << 8); +} +#endif + +#ifndef STBI_NO_BMP +static stbi__uint32 stbi__get32le(stbi__context* s) +{ + stbi__uint32 z = stbi__get16le(s); + return z + (stbi__get16le(s) << 16); +} +#endif + +#define STBI__BYTECAST(x) ((stbi_uc) ((x) & 255)) // truncate int to byte without warnings + + +////////////////////////////////////////////////////////////////////////////// +// +// 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_mad3(req_comp, x, y, 0); + if (good == NULL) { + STBI_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 STBI__COMBO(a,b) ((a)*8+(b)) +#define STBI__CASE(a,b) case STBI__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 (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1, 2) { dest[0] = src[0], dest[1] = 255; } break; + STBI__CASE(1, 3) { dest[0] = dest[1] = dest[2] = src[0]; } break; + STBI__CASE(1, 4) { dest[0] = dest[1] = dest[2] = src[0], dest[3] = 255; } break; + STBI__CASE(2, 1) { dest[0] = src[0]; } break; + STBI__CASE(2, 3) { dest[0] = dest[1] = dest[2] = src[0]; } break; + STBI__CASE(2, 4) { dest[0] = dest[1] = dest[2] = src[0], dest[3] = src[1]; } break; + STBI__CASE(3, 4) { dest[0] = src[0], dest[1] = src[1], dest[2] = src[2], dest[3] = 255; } break; + STBI__CASE(3, 1) { dest[0] = stbi__compute_y(src[0], src[1], src[2]); } break; + STBI__CASE(3, 2) { dest[0] = stbi__compute_y(src[0], src[1], src[2]), dest[1] = 255; } break; + STBI__CASE(4, 1) { dest[0] = stbi__compute_y(src[0], src[1], src[2]); } break; + STBI__CASE(4, 2) { dest[0] = stbi__compute_y(src[0], src[1], src[2]), dest[1] = src[3]; } break; + STBI__CASE(4, 3) { dest[0] = src[0], dest[1] = src[1], dest[2] = src[2]; } break; + default: STBI_ASSERT(0); + } +#undef STBI__CASE + } + + STBI_FREE(data); + return good; +} + +static stbi__uint16 stbi__compute_y_16(int r, int g, int b) +{ + return (stbi__uint16)(((r * 77) + (g * 150) + (29 * b)) >> 8); +} + +static stbi__uint16* stbi__convert_format16(stbi__uint16* data, int img_n, int req_comp, unsigned int x, unsigned int y) +{ + int i, j; + stbi__uint16* good; + + if (req_comp == img_n) return data; + STBI_ASSERT(req_comp >= 1 && req_comp <= 4); + + good = (stbi__uint16*)stbi__malloc(req_comp * x * y * 2); + if (good == NULL) { + STBI_FREE(data); + return (stbi__uint16*)stbi__errpuc("outofmem", "Out of memory"); + } + + for (j = 0; j < (int)y; ++j) { + stbi__uint16* src = data + j * x * img_n; + stbi__uint16* dest = good + j * x * req_comp; + +#define STBI__COMBO(a,b) ((a)*8+(b)) +#define STBI__CASE(a,b) case STBI__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 (STBI__COMBO(img_n, req_comp)) { + STBI__CASE(1, 2) { dest[0] = src[0], dest[1] = 0xffff; } break; + STBI__CASE(1, 3) { dest[0] = dest[1] = dest[2] = src[0]; } break; + STBI__CASE(1, 4) { dest[0] = dest[1] = dest[2] = src[0], dest[3] = 0xffff; } break; + STBI__CASE(2, 1) { dest[0] = src[0]; } break; + STBI__CASE(2, 3) { dest[0] = dest[1] = dest[2] = src[0]; } break; + STBI__CASE(2, 4) { dest[0] = dest[1] = dest[2] = src[0], dest[3] = src[1]; } break; + STBI__CASE(3, 4) { dest[0] = src[0], dest[1] = src[1], dest[2] = src[2], dest[3] = 0xffff; } break; + STBI__CASE(3, 1) { dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); } break; + STBI__CASE(3, 2) { dest[0] = stbi__compute_y_16(src[0], src[1], src[2]), dest[1] = 0xffff; } break; + STBI__CASE(4, 1) { dest[0] = stbi__compute_y_16(src[0], src[1], src[2]); } break; + STBI__CASE(4, 2) { dest[0] = stbi__compute_y_16(src[0], src[1], src[2]), dest[1] = src[3]; } break; + STBI__CASE(4, 3) { dest[0] = src[0], dest[1] = src[1], dest[2] = src[2]; } break; + default: STBI_ASSERT(0); + } +#undef STBI__CASE + } + + STBI_FREE(data); + return good; +} + +#ifndef STBI_NO_LINEAR +static float* stbi__ldr_to_hdr(stbi_uc* data, int x, int y, int comp) +{ + int i, k, n; + float* output; + if (!data) return NULL; + output = (float*)stbi__malloc_mad4(x, y, comp, sizeof(float), 0); + if (output == NULL) { STBI_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; + } + STBI_FREE(data); + return output; +} +#endif + +#ifndef STBI_NO_HDR +#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; + if (!data) return NULL; + output = (stbi_uc*)stbi__malloc_mad3(x, y, comp, 0); + if (output == NULL) { STBI_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); + } + } + STBI_FREE(data); + return output; +} +#endif + +////////////////////////////////////////////////////////////////////////////// +// +// "baseline" JPEG/JFIF decoder +// +// simple implementation +// - 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 +// - some SIMD kernels for common paths on targets with SSE2/NEON +// - uses a lot of intermediate memory, could cache poorly + +#ifndef STBI_NO_JPEG + +// 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 +{ + stbi__context* s; + stbi__huffman huff_dc[4]; + stbi__huffman huff_ac[4]; + stbi__uint16 dequant[4][64]; + stbi__int16 fast_ac[4][1 << FAST_BITS]; + + // 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, * raw_coeff; + stbi_uc* linebuf; + short* coeff; // progressive only + int coeff_w, coeff_h; // number of 8x8 coefficient blocks + } 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 progressive; + int spec_start; + int spec_end; + int succ_high; + int succ_low; + int eob_run; + int jfif; + int app14_color_transform; // Adobe APP14 tag + int rgb; + + int scan_n, order[4]; + int restart_interval, todo; + + // kernels + void (*idct_block_kernel)(stbi_uc* out, int out_stride, short data[64]); + void (*YCbCr_to_RGB_kernel)(stbi_uc* out, const stbi_uc* y, const stbi_uc* pcb, const stbi_uc* pcr, int count, int step); + stbi_uc* (*resample_row_hv_2_kernel)(stbi_uc* out, stbi_uc* in_near, stbi_uc* in_far, int w, int hs); +} stbi__jpeg; + +static int stbi__build_huffman(stbi__huffman* h, int* count) +{ + int i, j, k = 0; + unsigned int 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 >= (1u << 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; +} + +// build a table that decodes both magnitude and value of small ACs in +// one go. +static void stbi__build_fast_ac(stbi__int16* fast_ac, stbi__huffman* h) +{ + int i; + for (i = 0; i < (1 << FAST_BITS); ++i) { + stbi_uc fast = h->fast[i]; + fast_ac[i] = 0; + if (fast < 255) { + int rs = h->values[fast]; + int run = (rs >> 4) & 15; + int magbits = rs & 15; + int len = h->size[fast]; + + if (magbits && len + magbits <= FAST_BITS) { + // magnitude code followed by receive_extend code + int k = ((i << len) & ((1 << FAST_BITS) - 1)) >> (FAST_BITS - magbits); + int m = 1 << (magbits - 1); + if (k < m) k += (~0U << magbits) + 1; + // if the result is small enough, we can fit it in fast_ac table + if (k >= -128 && k <= 127) + fast_ac[i] = (stbi__int16)((k * 256) + (run * 16) + (len + magbits)); + } + } + } +} + +static void stbi__grow_buffer_unsafe(stbi__jpeg* j) +{ + do { + unsigned int b = j->nomore ? 0 : stbi__get8(j->s); + if (b == 0xff) { + int c = stbi__get8(j->s); + while (c == 0xff) c = stbi__get8(j->s); // consume fill bytes + 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 const 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]; +} + +// bias[n] = (-1<code_bits < n) stbi__grow_buffer_unsafe(j); + + sgn = (stbi__int32)j->code_buffer >> 31; // sign bit is always in MSB + k = stbi_lrot(j->code_buffer, n); + STBI_ASSERT(n >= 0 && n < (int)(sizeof(stbi__bmask) / sizeof(*stbi__bmask))); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k + (stbi__jbias[n] & ~sgn); +} + +// get some unsigned bits +stbi_inline static int stbi__jpeg_get_bits(stbi__jpeg* j, int n) +{ + unsigned int k; + if (j->code_bits < n) stbi__grow_buffer_unsafe(j); + k = stbi_lrot(j->code_buffer, n); + j->code_buffer = k & ~stbi__bmask[n]; + k &= stbi__bmask[n]; + j->code_bits -= n; + return k; +} + +stbi_inline static int stbi__jpeg_get_bit(stbi__jpeg* j) +{ + unsigned int k; + if (j->code_bits < 1) stbi__grow_buffer_unsafe(j); + k = j->code_buffer; + j->code_buffer <<= 1; + --j->code_bits; + return k & 0x80000000; +} + +// 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 const 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, stbi__int16* fac, int b, stbi__uint16* dequant) +{ + int diff, dc, k; + int t; + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + t = stbi__jpeg_huff_decode(j, hdc); + if (t < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + + // 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 * dequant[0]); + + // decode AC components, see JPEG spec + k = 1; + do { + unsigned int zig; + int c, r, s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS) - 1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + j->code_buffer <<= s; + j->code_bits -= s; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)((r >> 8) * dequant[zig]); + } + else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (rs != 0xf0) break; // end block + k += 16; + } + else { + k += r; + // decode into unzigzag'd location + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)(stbi__extend_receive(j, s) * dequant[zig]); + } + } + } while (k < 64); + return 1; +} + +static int stbi__jpeg_decode_block_prog_dc(stbi__jpeg* j, short data[64], stbi__huffman* hdc, int b) +{ + int diff, dc; + int t; + if (j->spec_end != 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + + if (j->succ_high == 0) { + // first scan for DC coefficient, must be first + memset(data, 0, 64 * sizeof(data[0])); // 0 all the ac values now + t = stbi__jpeg_huff_decode(j, hdc); + diff = t ? stbi__extend_receive(j, t) : 0; + + dc = j->img_comp[b].dc_pred + diff; + j->img_comp[b].dc_pred = dc; + data[0] = (short)(dc << j->succ_low); + } + else { + // refinement scan for DC coefficient + if (stbi__jpeg_get_bit(j)) + data[0] += (short)(1 << j->succ_low); + } + return 1; +} + +// @OPTIMIZE: store non-zigzagged during the decode passes, +// and only de-zigzag when dequantizing +static int stbi__jpeg_decode_block_prog_ac(stbi__jpeg* j, short data[64], stbi__huffman* hac, stbi__int16* fac) +{ + int k; + if (j->spec_start == 0) return stbi__err("can't merge dc and ac", "Corrupt JPEG"); + + if (j->succ_high == 0) { + int shift = j->succ_low; + + if (j->eob_run) { + --j->eob_run; + return 1; + } + + k = j->spec_start; + do { + unsigned int zig; + int c, r, s; + if (j->code_bits < 16) stbi__grow_buffer_unsafe(j); + c = (j->code_buffer >> (32 - FAST_BITS)) & ((1 << FAST_BITS) - 1); + r = fac[c]; + if (r) { // fast-AC path + k += (r >> 4) & 15; // run + s = r & 15; // combined length + j->code_buffer <<= s; + j->code_bits -= s; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)((r >> 8) << shift); + } + else { + int rs = stbi__jpeg_huff_decode(j, hac); + if (rs < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r); + if (r) + j->eob_run += stbi__jpeg_get_bits(j, r); + --j->eob_run; + break; + } + k += 16; + } + else { + k += r; + zig = stbi__jpeg_dezigzag[k++]; + data[zig] = (short)(stbi__extend_receive(j, s) << shift); + } + } + } while (k <= j->spec_end); + } + else { + // refinement scan for these AC coefficients + + short bit = (short)(1 << j->succ_low); + + if (j->eob_run) { + --j->eob_run; + for (k = j->spec_start; k <= j->spec_end; ++k) { + short* p = &data[stbi__jpeg_dezigzag[k]]; + if (*p != 0) + if (stbi__jpeg_get_bit(j)) + if ((*p & bit) == 0) { + if (*p > 0) + * p += bit; + else + *p -= bit; + } + } + } + else { + k = j->spec_start; + do { + int r, s; + int rs = stbi__jpeg_huff_decode(j, hac); // @OPTIMIZE see if we can use the fast path here, advance-by-r is so slow, eh + if (rs < 0) return stbi__err("bad huffman code", "Corrupt JPEG"); + s = rs & 15; + r = rs >> 4; + if (s == 0) { + if (r < 15) { + j->eob_run = (1 << r) - 1; + if (r) + j->eob_run += stbi__jpeg_get_bits(j, r); + r = 64; // force end of block + } + else { + // r=15 s=0 should write 16 0s, so we just do + // a run of 15 0s and then write s (which is 0), + // so we don't have to do anything special here + } + } + else { + if (s != 1) return stbi__err("bad huffman code", "Corrupt JPEG"); + // sign bit + if (stbi__jpeg_get_bit(j)) + s = bit; + else + s = -bit; + } + + // advance by r + while (k <= j->spec_end) { + short* p = &data[stbi__jpeg_dezigzag[k++]]; + if (*p != 0) { + if (stbi__jpeg_get_bit(j)) + if ((*p & bit) == 0) { + if (*p > 0) + * p += bit; + else + *p -= bit; + } + } + else { + if (r == 0) { + *p = (short)s; + break; + } + --r; + } + } + } while (k <= j->spec_end); + } + } + 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) * 4096) + +// 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; + +static void stbi__idct_block(stbi_uc* out, int out_stride, short data[64]) +{ + int i, val[64], * v = val; + stbi_uc* o; + short* d = data; + + // columns + for (i = 0; i < 8; ++i, ++d, ++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] * 4; + v[0] = v[8] = v[16] = v[24] = v[32] = v[40] = v[48] = v[56] = dcterm; + } + else { + STBI__IDCT_1D(d[0], d[8], d[16], d[24], d[32], d[40], d[48], d[56]) + // 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_SSE2 +// sse2 integer IDCT. not the fastest possible implementation but it +// produces bit-identical results to the generic C version so it's +// fully "transparent". +static void stbi__idct_simd(stbi_uc* out, int out_stride, short data[64]) +{ + // This is constructed to match our regular (generic) integer IDCT exactly. + __m128i row0, row1, row2, row3, row4, row5, row6, row7; + __m128i tmp; + + // dot product constant: even elems=x, odd elems=y +#define dct_const(x,y) _mm_setr_epi16((x),(y),(x),(y),(x),(y),(x),(y)) + +// out(0) = c0[even]*x + c0[odd]*y (c0, x, y 16-bit, out 32-bit) +// out(1) = c1[even]*x + c1[odd]*y +#define dct_rot(out0,out1, x,y,c0,c1) \ + __m128i c0##lo = _mm_unpacklo_epi16((x),(y)); \ + __m128i c0##hi = _mm_unpackhi_epi16((x),(y)); \ + __m128i out0##_l = _mm_madd_epi16(c0##lo, c0); \ + __m128i out0##_h = _mm_madd_epi16(c0##hi, c0); \ + __m128i out1##_l = _mm_madd_epi16(c0##lo, c1); \ + __m128i out1##_h = _mm_madd_epi16(c0##hi, c1) + + // out = in << 12 (in 16-bit, out 32-bit) +#define dct_widen(out, in) \ + __m128i out##_l = _mm_srai_epi32(_mm_unpacklo_epi16(_mm_setzero_si128(), (in)), 4); \ + __m128i out##_h = _mm_srai_epi32(_mm_unpackhi_epi16(_mm_setzero_si128(), (in)), 4) + + // wide add +#define dct_wadd(out, a, b) \ + __m128i out##_l = _mm_add_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_add_epi32(a##_h, b##_h) + + // wide sub +#define dct_wsub(out, a, b) \ + __m128i out##_l = _mm_sub_epi32(a##_l, b##_l); \ + __m128i out##_h = _mm_sub_epi32(a##_h, b##_h) + + // butterfly a/b, add bias, then shift by "s" and pack +#define dct_bfly32o(out0, out1, a,b,bias,s) \ + { \ + __m128i abiased_l = _mm_add_epi32(a##_l, bias); \ + __m128i abiased_h = _mm_add_epi32(a##_h, bias); \ + dct_wadd(sum, abiased, b); \ + dct_wsub(dif, abiased, b); \ + out0 = _mm_packs_epi32(_mm_srai_epi32(sum_l, s), _mm_srai_epi32(sum_h, s)); \ + out1 = _mm_packs_epi32(_mm_srai_epi32(dif_l, s), _mm_srai_epi32(dif_h, s)); \ + } + + // 8-bit interleave step (for transposes) +#define dct_interleave8(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi8(a, b); \ + b = _mm_unpackhi_epi8(tmp, b) + + // 16-bit interleave step (for transposes) +#define dct_interleave16(a, b) \ + tmp = a; \ + a = _mm_unpacklo_epi16(a, b); \ + b = _mm_unpackhi_epi16(tmp, b) + +#define dct_pass(bias,shift) \ + { \ + /* even part */ \ + dct_rot(t2e,t3e, row2,row6, rot0_0,rot0_1); \ + __m128i sum04 = _mm_add_epi16(row0, row4); \ + __m128i dif04 = _mm_sub_epi16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + dct_rot(y0o,y2o, row7,row3, rot2_0,rot2_1); \ + dct_rot(y1o,y3o, row5,row1, rot3_0,rot3_1); \ + __m128i sum17 = _mm_add_epi16(row1, row7); \ + __m128i sum35 = _mm_add_epi16(row3, row5); \ + dct_rot(y4o,y5o, sum17,sum35, rot1_0,rot1_1); \ + dct_wadd(x4, y0o, y4o); \ + dct_wadd(x5, y1o, y5o); \ + dct_wadd(x6, y2o, y5o); \ + dct_wadd(x7, y3o, y4o); \ + dct_bfly32o(row0,row7, x0,x7,bias,shift); \ + dct_bfly32o(row1,row6, x1,x6,bias,shift); \ + dct_bfly32o(row2,row5, x2,x5,bias,shift); \ + dct_bfly32o(row3,row4, x3,x4,bias,shift); \ + } + + __m128i rot0_0 = dct_const(stbi__f2f(0.5411961f), stbi__f2f(0.5411961f) + stbi__f2f(-1.847759065f)); + __m128i rot0_1 = dct_const(stbi__f2f(0.5411961f) + stbi__f2f(0.765366865f), stbi__f2f(0.5411961f)); + __m128i rot1_0 = dct_const(stbi__f2f(1.175875602f) + stbi__f2f(-0.899976223f), stbi__f2f(1.175875602f)); + __m128i rot1_1 = dct_const(stbi__f2f(1.175875602f), stbi__f2f(1.175875602f) + stbi__f2f(-2.562915447f)); + __m128i rot2_0 = dct_const(stbi__f2f(-1.961570560f) + stbi__f2f(0.298631336f), stbi__f2f(-1.961570560f)); + __m128i rot2_1 = dct_const(stbi__f2f(-1.961570560f), stbi__f2f(-1.961570560f) + stbi__f2f(3.072711026f)); + __m128i rot3_0 = dct_const(stbi__f2f(-0.390180644f) + stbi__f2f(2.053119869f), stbi__f2f(-0.390180644f)); + __m128i rot3_1 = dct_const(stbi__f2f(-0.390180644f), stbi__f2f(-0.390180644f) + stbi__f2f(1.501321110f)); + + // rounding biases in column/row passes, see stbi__idct_block for explanation. + __m128i bias_0 = _mm_set1_epi32(512); + __m128i bias_1 = _mm_set1_epi32(65536 + (128 << 17)); + + // load + row0 = _mm_load_si128((const __m128i*) (data + 0 * 8)); + row1 = _mm_load_si128((const __m128i*) (data + 1 * 8)); + row2 = _mm_load_si128((const __m128i*) (data + 2 * 8)); + row3 = _mm_load_si128((const __m128i*) (data + 3 * 8)); + row4 = _mm_load_si128((const __m128i*) (data + 4 * 8)); + row5 = _mm_load_si128((const __m128i*) (data + 5 * 8)); + row6 = _mm_load_si128((const __m128i*) (data + 6 * 8)); + row7 = _mm_load_si128((const __m128i*) (data + 7 * 8)); + + // column pass + dct_pass(bias_0, 10); + + { + // 16bit 8x8 transpose pass 1 + dct_interleave16(row0, row4); + dct_interleave16(row1, row5); + dct_interleave16(row2, row6); + dct_interleave16(row3, row7); + + // transpose pass 2 + dct_interleave16(row0, row2); + dct_interleave16(row1, row3); + dct_interleave16(row4, row6); + dct_interleave16(row5, row7); + + // transpose pass 3 + dct_interleave16(row0, row1); + dct_interleave16(row2, row3); + dct_interleave16(row4, row5); + dct_interleave16(row6, row7); + } + + // row pass + dct_pass(bias_1, 17); + + { + // pack + __m128i p0 = _mm_packus_epi16(row0, row1); // a0a1a2a3...a7b0b1b2b3...b7 + __m128i p1 = _mm_packus_epi16(row2, row3); + __m128i p2 = _mm_packus_epi16(row4, row5); + __m128i p3 = _mm_packus_epi16(row6, row7); + + // 8bit 8x8 transpose pass 1 + dct_interleave8(p0, p2); // a0e0a1e1... + dct_interleave8(p1, p3); // c0g0c1g1... + + // transpose pass 2 + dct_interleave8(p0, p1); // a0c0e0g0... + dct_interleave8(p2, p3); // b0d0f0h0... + + // transpose pass 3 + dct_interleave8(p0, p2); // a0b0c0d0... + dct_interleave8(p1, p3); // a4b4c4d4... + + // store + _mm_storel_epi64((__m128i*) out, p0); out += out_stride; + _mm_storel_epi64((__m128i*) out, _mm_shuffle_epi32(p0, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i*) out, p2); out += out_stride; + _mm_storel_epi64((__m128i*) out, _mm_shuffle_epi32(p2, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i*) out, p1); out += out_stride; + _mm_storel_epi64((__m128i*) out, _mm_shuffle_epi32(p1, 0x4e)); out += out_stride; + _mm_storel_epi64((__m128i*) out, p3); out += out_stride; + _mm_storel_epi64((__m128i*) out, _mm_shuffle_epi32(p3, 0x4e)); + } + +#undef dct_const +#undef dct_rot +#undef dct_widen +#undef dct_wadd +#undef dct_wsub +#undef dct_bfly32o +#undef dct_interleave8 +#undef dct_interleave16 +#undef dct_pass +} + +#endif // STBI_SSE2 + +#ifdef STBI_NEON + +// NEON integer IDCT. should produce bit-identical +// results to the generic C version. +static void stbi__idct_simd(stbi_uc* out, int out_stride, short data[64]) +{ + int16x8_t row0, row1, row2, row3, row4, row5, row6, row7; + + int16x4_t rot0_0 = vdup_n_s16(stbi__f2f(0.5411961f)); + int16x4_t rot0_1 = vdup_n_s16(stbi__f2f(-1.847759065f)); + int16x4_t rot0_2 = vdup_n_s16(stbi__f2f(0.765366865f)); + int16x4_t rot1_0 = vdup_n_s16(stbi__f2f(1.175875602f)); + int16x4_t rot1_1 = vdup_n_s16(stbi__f2f(-0.899976223f)); + int16x4_t rot1_2 = vdup_n_s16(stbi__f2f(-2.562915447f)); + int16x4_t rot2_0 = vdup_n_s16(stbi__f2f(-1.961570560f)); + int16x4_t rot2_1 = vdup_n_s16(stbi__f2f(-0.390180644f)); + int16x4_t rot3_0 = vdup_n_s16(stbi__f2f(0.298631336f)); + int16x4_t rot3_1 = vdup_n_s16(stbi__f2f(2.053119869f)); + int16x4_t rot3_2 = vdup_n_s16(stbi__f2f(3.072711026f)); + int16x4_t rot3_3 = vdup_n_s16(stbi__f2f(1.501321110f)); + +#define dct_long_mul(out, inq, coeff) \ + int32x4_t out##_l = vmull_s16(vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmull_s16(vget_high_s16(inq), coeff) + +#define dct_long_mac(out, acc, inq, coeff) \ + int32x4_t out##_l = vmlal_s16(acc##_l, vget_low_s16(inq), coeff); \ + int32x4_t out##_h = vmlal_s16(acc##_h, vget_high_s16(inq), coeff) + +#define dct_widen(out, inq) \ + int32x4_t out##_l = vshll_n_s16(vget_low_s16(inq), 12); \ + int32x4_t out##_h = vshll_n_s16(vget_high_s16(inq), 12) + + // wide add +#define dct_wadd(out, a, b) \ + int32x4_t out##_l = vaddq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vaddq_s32(a##_h, b##_h) + +// wide sub +#define dct_wsub(out, a, b) \ + int32x4_t out##_l = vsubq_s32(a##_l, b##_l); \ + int32x4_t out##_h = vsubq_s32(a##_h, b##_h) + +// butterfly a/b, then shift using "shiftop" by "s" and pack +#define dct_bfly32o(out0,out1, a,b,shiftop,s) \ + { \ + dct_wadd(sum, a, b); \ + dct_wsub(dif, a, b); \ + out0 = vcombine_s16(shiftop(sum_l, s), shiftop(sum_h, s)); \ + out1 = vcombine_s16(shiftop(dif_l, s), shiftop(dif_h, s)); \ + } + +#define dct_pass(shiftop, shift) \ + { \ + /* even part */ \ + int16x8_t sum26 = vaddq_s16(row2, row6); \ + dct_long_mul(p1e, sum26, rot0_0); \ + dct_long_mac(t2e, p1e, row6, rot0_1); \ + dct_long_mac(t3e, p1e, row2, rot0_2); \ + int16x8_t sum04 = vaddq_s16(row0, row4); \ + int16x8_t dif04 = vsubq_s16(row0, row4); \ + dct_widen(t0e, sum04); \ + dct_widen(t1e, dif04); \ + dct_wadd(x0, t0e, t3e); \ + dct_wsub(x3, t0e, t3e); \ + dct_wadd(x1, t1e, t2e); \ + dct_wsub(x2, t1e, t2e); \ + /* odd part */ \ + int16x8_t sum15 = vaddq_s16(row1, row5); \ + int16x8_t sum17 = vaddq_s16(row1, row7); \ + int16x8_t sum35 = vaddq_s16(row3, row5); \ + int16x8_t sum37 = vaddq_s16(row3, row7); \ + int16x8_t sumodd = vaddq_s16(sum17, sum35); \ + dct_long_mul(p5o, sumodd, rot1_0); \ + dct_long_mac(p1o, p5o, sum17, rot1_1); \ + dct_long_mac(p2o, p5o, sum35, rot1_2); \ + dct_long_mul(p3o, sum37, rot2_0); \ + dct_long_mul(p4o, sum15, rot2_1); \ + dct_wadd(sump13o, p1o, p3o); \ + dct_wadd(sump24o, p2o, p4o); \ + dct_wadd(sump23o, p2o, p3o); \ + dct_wadd(sump14o, p1o, p4o); \ + dct_long_mac(x4, sump13o, row7, rot3_0); \ + dct_long_mac(x5, sump24o, row5, rot3_1); \ + dct_long_mac(x6, sump23o, row3, rot3_2); \ + dct_long_mac(x7, sump14o, row1, rot3_3); \ + dct_bfly32o(row0,row7, x0,x7,shiftop,shift); \ + dct_bfly32o(row1,row6, x1,x6,shiftop,shift); \ + dct_bfly32o(row2,row5, x2,x5,shiftop,shift); \ + dct_bfly32o(row3,row4, x3,x4,shiftop,shift); \ + } + + // load + row0 = vld1q_s16(data + 0 * 8); + row1 = vld1q_s16(data + 1 * 8); + row2 = vld1q_s16(data + 2 * 8); + row3 = vld1q_s16(data + 3 * 8); + row4 = vld1q_s16(data + 4 * 8); + row5 = vld1q_s16(data + 5 * 8); + row6 = vld1q_s16(data + 6 * 8); + row7 = vld1q_s16(data + 7 * 8); + + // add DC bias + row0 = vaddq_s16(row0, vsetq_lane_s16(1024, vdupq_n_s16(0), 0)); + + // column pass + dct_pass(vrshrn_n_s32, 10); + + // 16bit 8x8 transpose + { + // these three map to a single VTRN.16, VTRN.32, and VSWP, respectively. + // whether compilers actually get this is another story, sadly. +#define dct_trn16(x, y) { int16x8x2_t t = vtrnq_s16(x, y); x = t.val[0]; y = t.val[1]; } +#define dct_trn32(x, y) { int32x4x2_t t = vtrnq_s32(vreinterpretq_s32_s16(x), vreinterpretq_s32_s16(y)); x = vreinterpretq_s16_s32(t.val[0]); y = vreinterpretq_s16_s32(t.val[1]); } +#define dct_trn64(x, y) { int16x8_t x0 = x; int16x8_t y0 = y; x = vcombine_s16(vget_low_s16(x0), vget_low_s16(y0)); y = vcombine_s16(vget_high_s16(x0), vget_high_s16(y0)); } + + // pass 1 + dct_trn16(row0, row1); // a0b0a2b2a4b4a6b6 + dct_trn16(row2, row3); + dct_trn16(row4, row5); + dct_trn16(row6, row7); + + // pass 2 + dct_trn32(row0, row2); // a0b0c0d0a4b4c4d4 + dct_trn32(row1, row3); + dct_trn32(row4, row6); + dct_trn32(row5, row7); + + // pass 3 + dct_trn64(row0, row4); // a0b0c0d0e0f0g0h0 + dct_trn64(row1, row5); + dct_trn64(row2, row6); + dct_trn64(row3, row7); + +#undef dct_trn16 +#undef dct_trn32 +#undef dct_trn64 + } + + // row pass + // vrshrn_n_s32 only supports shifts up to 16, we need + // 17. so do a non-rounding shift of 16 first then follow + // up with a rounding shift by 1. + dct_pass(vshrn_n_s32, 16); + + { + // pack and round + uint8x8_t p0 = vqrshrun_n_s16(row0, 1); + uint8x8_t p1 = vqrshrun_n_s16(row1, 1); + uint8x8_t p2 = vqrshrun_n_s16(row2, 1); + uint8x8_t p3 = vqrshrun_n_s16(row3, 1); + uint8x8_t p4 = vqrshrun_n_s16(row4, 1); + uint8x8_t p5 = vqrshrun_n_s16(row5, 1); + uint8x8_t p6 = vqrshrun_n_s16(row6, 1); + uint8x8_t p7 = vqrshrun_n_s16(row7, 1); + + // again, these can translate into one instruction, but often don't. +#define dct_trn8_8(x, y) { uint8x8x2_t t = vtrn_u8(x, y); x = t.val[0]; y = t.val[1]; } +#define dct_trn8_16(x, y) { uint16x4x2_t t = vtrn_u16(vreinterpret_u16_u8(x), vreinterpret_u16_u8(y)); x = vreinterpret_u8_u16(t.val[0]); y = vreinterpret_u8_u16(t.val[1]); } +#define dct_trn8_32(x, y) { uint32x2x2_t t = vtrn_u32(vreinterpret_u32_u8(x), vreinterpret_u32_u8(y)); x = vreinterpret_u8_u32(t.val[0]); y = vreinterpret_u8_u32(t.val[1]); } + + // sadly can't use interleaved stores here since we only write + // 8 bytes to each scan line! + + // 8x8 8-bit transpose pass 1 + dct_trn8_8(p0, p1); + dct_trn8_8(p2, p3); + dct_trn8_8(p4, p5); + dct_trn8_8(p6, p7); + + // pass 2 + dct_trn8_16(p0, p2); + dct_trn8_16(p1, p3); + dct_trn8_16(p4, p6); + dct_trn8_16(p5, p7); + + // pass 3 + dct_trn8_32(p0, p4); + dct_trn8_32(p1, p5); + dct_trn8_32(p2, p6); + dct_trn8_32(p3, p7); + + // store + vst1_u8(out, p0); out += out_stride; + vst1_u8(out, p1); out += out_stride; + vst1_u8(out, p2); out += out_stride; + vst1_u8(out, p3); out += out_stride; + vst1_u8(out, p4); out += out_stride; + vst1_u8(out, p5); out += out_stride; + vst1_u8(out, p6); out += out_stride; + vst1_u8(out, p7); + +#undef dct_trn8_8 +#undef dct_trn8_16 +#undef dct_trn8_32 + } + +#undef dct_long_mul +#undef dct_long_mac +#undef dct_widen +#undef dct_wadd +#undef dct_wsub +#undef dct_bfly32o +#undef dct_pass +} + +#endif // STBI_NEON + +#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); // consume repeated 0xff fill bytes + 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 = j->img_comp[3].dc_pred = 0; + j->marker = STBI__MARKER_none; + j->todo = j->restart_interval ? j->restart_interval : 0x7fffffff; + j->eob_run = 0; + // 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->progressive) { + if (z->scan_n == 1) { + int i, j; + STBI_SIMD_ALIGN(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) { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc + z->img_comp[n].hd, z->huff_ac + ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; + z->idct_block_kernel(z->img_comp[n].data + z->img_comp[n].w2 * j * 8 + i * 8, z->img_comp[n].w2, data); + // 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); + } + } + } + return 1; + } + else { // interleaved + int i, j, k, x, y; + STBI_SIMD_ALIGN(short, data[64]); + for (j = 0; j < z->img_mcu_y; ++j) { + for (i = 0; i < z->img_mcu_x; ++i) { + // 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; + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block(z, data, z->huff_dc + z->img_comp[n].hd, z->huff_ac + ha, z->fast_ac[ha], n, z->dequant[z->img_comp[n].tq])) return 0; + z->idct_block_kernel(z->img_comp[n].data + z->img_comp[n].w2 * y2 + x2, z->img_comp[n].w2, data); + } + } + } + // 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 (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } + else { + if (z->scan_n == 1) { + int i, j; + int n = z->order[0]; + // 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) { + short* data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + if (z->spec_start == 0) { + if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } + else { + int ha = z->img_comp[n].ha; + if (!stbi__jpeg_decode_block_prog_ac(z, data, &z->huff_ac[ha], z->fast_ac[ha])) + return 0; + } + // 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 (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + else { // interleaved + int i, j, k, x, y; + 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); + int y2 = (j * z->img_comp[n].v + y); + short* data = z->img_comp[n].coeff + 64 * (x2 + y2 * z->img_comp[n].coeff_w); + if (!stbi__jpeg_decode_block_prog_dc(z, data, &z->huff_dc[z->img_comp[n].hd], n)) + return 0; + } + } + } + // 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 (!STBI__RESTART(z->marker)) return 1; + stbi__jpeg_reset(z); + } + } + } + return 1; + } + } +} + +static void stbi__jpeg_dequantize(short* data, stbi__uint16* dequant) +{ + int i; + for (i = 0; i < 64; ++i) + data[i] *= dequant[i]; +} + +static void stbi__jpeg_finish(stbi__jpeg* z) +{ + if (z->progressive) { + // dequantize and idct the data + int i, j, n; + for (n = 0; n < z->s->img_n; ++n) { + int w = (z->img_comp[n].x + 7) >> 3; + int h = (z->img_comp[n].y + 7) >> 3; + for (j = 0; j < h; ++j) { + for (i = 0; i < w; ++i) { + short* data = z->img_comp[n].coeff + 64 * (i + j * z->img_comp[n].coeff_w); + stbi__jpeg_dequantize(data, z->dequant[z->img_comp[n].tq]); + z->idct_block_kernel(z->img_comp[n].data + z->img_comp[n].w2 * j * 8 + i * 8, z->img_comp[n].w2, data); + } + } + } + } +} + +static int stbi__process_marker(stbi__jpeg* z, int m) +{ + int L; + switch (m) { + case STBI__MARKER_none: // no marker found + return stbi__err("expected marker", "Corrupt JPEG"); + + 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, sixteen = (p != 0); + int t = q & 15, i; + if (p != 0 && p != 1) 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__uint16)(sixteen ? stbi__get16be(z->s) : stbi__get8(z->s)); + L -= (sixteen ? 129 : 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); + if (tc != 0) + stbi__build_fast_ac(z->fast_ac[th], z->huff_ac + th); + L -= n; + } + return L == 0; + } + + // check for comment block or APP blocks + if ((m >= 0xE0 && m <= 0xEF) || m == 0xFE) { + L = stbi__get16be(z->s); + if (L < 2) { + if (m == 0xFE) + return stbi__err("bad COM len", "Corrupt JPEG"); + else + return stbi__err("bad APP len", "Corrupt JPEG"); + } + L -= 2; + + if (m == 0xE0 && L >= 5) { // JFIF APP0 segment + static const unsigned char tag[5] = { 'J','F','I','F','\0' }; + int ok = 1; + int i; + for (i = 0; i < 5; ++i) + if (stbi__get8(z->s) != tag[i]) + ok = 0; + L -= 5; + if (ok) + z->jfif = 1; + } + else if (m == 0xEE && L >= 12) { // Adobe APP14 segment + static const unsigned char tag[6] = { 'A','d','o','b','e','\0' }; + int ok = 1; + int i; + for (i = 0; i < 6; ++i) + if (stbi__get8(z->s) != tag[i]) + ok = 0; + L -= 6; + if (ok) { + stbi__get8(z->s); // version + stbi__get16be(z->s); // flags0 + stbi__get16be(z->s); // flags1 + z->app14_color_transform = stbi__get8(z->s); // color transform + L -= 6; + } + } + + stbi__skip(z->s, L); + return 1; + } + + return stbi__err("unknown marker", "Corrupt JPEG"); +} + +// after we see 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 SOS component count", "Corrupt JPEG"); + if (Ls != 6 + 2 * z->scan_n) return stbi__err("bad SOS len", "Corrupt JPEG"); + for (i = 0; i < z->scan_n; ++i) { + int id = stbi__get8(z->s), which; + int q = stbi__get8(z->s); + for (which = 0; which < z->s->img_n; ++which) + if (z->img_comp[which].id == id) + break; + if (which == z->s->img_n) return 0; // no match + z->img_comp[which].hd = q >> 4; if (z->img_comp[which].hd > 3) return stbi__err("bad DC huff", "Corrupt JPEG"); + z->img_comp[which].ha = q & 15; if (z->img_comp[which].ha > 3) return stbi__err("bad AC huff", "Corrupt JPEG"); + z->order[i] = which; + } + + { + int aa; + z->spec_start = stbi__get8(z->s); + z->spec_end = stbi__get8(z->s); // should be 63, but might be 0 + aa = stbi__get8(z->s); + z->succ_high = (aa >> 4); + z->succ_low = (aa & 15); + if (z->progressive) { + if (z->spec_start > 63 || z->spec_end > 63 || z->spec_start > z->spec_end || z->succ_high > 13 || z->succ_low > 13) + return stbi__err("bad SOS", "Corrupt JPEG"); + } + else { + if (z->spec_start != 0) return stbi__err("bad SOS", "Corrupt JPEG"); + if (z->succ_high != 0 || z->succ_low != 0) return stbi__err("bad SOS", "Corrupt JPEG"); + z->spec_end = 63; + } + } + + return 1; +} + +static int stbi__free_jpeg_components(stbi__jpeg* z, int ncomp, int why) +{ + int i; + for (i = 0; i < ncomp; ++i) { + if (z->img_comp[i].raw_data) { + STBI_FREE(z->img_comp[i].raw_data); + z->img_comp[i].raw_data = NULL; + z->img_comp[i].data = NULL; + } + if (z->img_comp[i].raw_coeff) { + STBI_FREE(z->img_comp[i].raw_coeff); + z->img_comp[i].raw_coeff = 0; + z->img_comp[i].coeff = 0; + } + if (z->img_comp[i].linebuf) { + STBI_FREE(z->img_comp[i].linebuf); + z->img_comp[i].linebuf = NULL; + } + } + return why; +} + +static int stbi__process_frame_header(stbi__jpeg* z, int scan) +{ + stbi__context* s = z->s; + int Lf, p, i, q, h_max = 1, v_max = 1, c; + Lf = stbi__get16be(s); if (Lf < 11) return stbi__err("bad SOF len", "Corrupt JPEG"); // 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 && c != 4) return stbi__err("bad component count", "Corrupt JPEG"); + s->img_n = c; + for (i = 0; i < c; ++i) { + z->img_comp[i].data = NULL; + z->img_comp[i].linebuf = NULL; + } + + if (Lf != 8 + 3 * s->img_n) return stbi__err("bad SOF len", "Corrupt JPEG"); + + z->rgb = 0; + for (i = 0; i < s->img_n; ++i) { + static const unsigned char rgb[3] = { 'R', 'G', 'B' }; + z->img_comp[i].id = stbi__get8(s); + if (s->img_n == 3 && z->img_comp[i].id == rgb[i]) + ++z->rgb; + q = stbi__get8(s); + z->img_comp[i].h = (q >> 4); if (!z->img_comp[i].h || z->img_comp[i].h > 4) return stbi__err("bad H", "Corrupt JPEG"); + z->img_comp[i].v = q & 15; if (!z->img_comp[i].v || z->img_comp[i].v > 4) return stbi__err("bad V", "Corrupt JPEG"); + z->img_comp[i].tq = stbi__get8(s); if (z->img_comp[i].tq > 3) return stbi__err("bad TQ", "Corrupt JPEG"); + } + + if (scan != STBI__SCAN_load) return 1; + + if (!stbi__mad3sizes_valid(s->img_x, s->img_y, s->img_n, 0)) 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; + // these sizes can't be more than 17 bits + 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 + // + // img_mcu_x, img_mcu_y: <=17 bits; comp[i].h and .v are <=4 (checked earlier) + // so these muls can't overflow with 32-bit ints (which we require) + 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].coeff = 0; + z->img_comp[i].raw_coeff = 0; + z->img_comp[i].linebuf = NULL; + z->img_comp[i].raw_data = stbi__malloc_mad2(z->img_comp[i].w2, z->img_comp[i].h2, 15); + if (z->img_comp[i].raw_data == NULL) + return stbi__free_jpeg_components(z, i + 1, stbi__err("outofmem", "Out of memory")); + // align blocks for idct using mmx/sse + z->img_comp[i].data = (stbi_uc*)(((size_t)z->img_comp[i].raw_data + 15) & ~15); + if (z->progressive) { + // w2, h2 are multiples of 8 (see above) + z->img_comp[i].coeff_w = z->img_comp[i].w2 / 8; + z->img_comp[i].coeff_h = z->img_comp[i].h2 / 8; + z->img_comp[i].raw_coeff = stbi__malloc_mad3(z->img_comp[i].w2, z->img_comp[i].h2, sizeof(short), 15); + if (z->img_comp[i].raw_coeff == NULL) + return stbi__free_jpeg_components(z, i + 1, stbi__err("outofmem", "Out of memory")); + z->img_comp[i].coeff = (short*)(((size_t)z->img_comp[i].raw_coeff + 15) & ~15); + } + } + + return 1; +} + +// use comparisons since in some cases we handle more than one case (e.g. 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 || (x) == 0xc2) +#define stbi__SOS(x) ((x) == 0xda) + +#define stbi__SOF_progressive(x) ((x) == 0xc2) + +static int stbi__decode_jpeg_header(stbi__jpeg* z, int scan) +{ + int m; + z->jfif = 0; + z->app14_color_transform = -1; // valid values are 0,1,2 + z->marker = STBI__MARKER_none; // initialize cached marker to empty + m = stbi__get_marker(z); + if (!stbi__SOI(m)) return stbi__err("no SOI", "Corrupt JPEG"); + if (scan == STBI__SCAN_type) return 1; + m = stbi__get_marker(z); + while (!stbi__SOF(m)) { + if (!stbi__process_marker(z, m)) return 0; + m = stbi__get_marker(z); + while (m == STBI__MARKER_none) { + // some files have extra padding after their blocks, so ok, we'll scan + if (stbi__at_eof(z->s)) return stbi__err("no SOF", "Corrupt JPEG"); + m = stbi__get_marker(z); + } + } + z->progressive = stbi__SOF_progressive(m); + if (!stbi__process_frame_header(z, scan)) return 0; + return 1; +} + +// decode image to YCbCr format +static int stbi__decode_jpeg_image(stbi__jpeg* j) +{ + int m; + for (m = 0; m < 4; m++) { + j->img_comp[m].raw_data = NULL; + j->img_comp[m].raw_coeff = NULL; + } + j->restart_interval = 0; + if (!stbi__decode_jpeg_header(j, STBI__SCAN_load)) return 0; + m = stbi__get_marker(j); + while (!stbi__EOI(m)) { + if (stbi__SOS(m)) { + if (!stbi__process_scan_header(j)) return 0; + if (!stbi__parse_entropy_coded_data(j)) return 0; + if (j->marker == STBI__MARKER_none) { + // 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; + } + } + // if we reach eof without hitting a marker, stbi__get_marker() below will fail and we'll eventually return 0 + } + } + else if (stbi__DNL(m)) { + int Ld = stbi__get16be(j->s); + stbi__uint32 NL = stbi__get16be(j->s); + if (Ld != 4) return stbi__err("bad DNL len", "Corrupt JPEG"); + if (NL != j->s->img_y) return stbi__err("bad DNL height", "Corrupt JPEG"); + } + else { + if (!stbi__process_marker(j, m)) return 0; + } + m = stbi__get_marker(j); + } + if (j->progressive) + stbi__jpeg_finish(j); + return 1; +} + +// 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; +} + +#if defined(STBI_SSE2) || defined(STBI_NEON) +static stbi_uc* stbi__resample_row_hv_2_simd(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 = 0, t0, t1; + + if (w == 1) { + out[0] = out[1] = stbi__div4(3 * in_near[0] + in_far[0] + 2); + return out; + } + + t1 = 3 * in_near[0] + in_far[0]; + // process groups of 8 pixels for as long as we can. + // note we can't handle the last pixel in a row in this loop + // because we need to handle the filter boundary conditions. + for (; i < ((w - 1) & ~7); i += 8) { +#if defined(STBI_SSE2) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + __m128i zero = _mm_setzero_si128(); + __m128i farb = _mm_loadl_epi64((__m128i*) (in_far + i)); + __m128i nearb = _mm_loadl_epi64((__m128i*) (in_near + i)); + __m128i farw = _mm_unpacklo_epi8(farb, zero); + __m128i nearw = _mm_unpacklo_epi8(nearb, zero); + __m128i diff = _mm_sub_epi16(farw, nearw); + __m128i nears = _mm_slli_epi16(nearw, 2); + __m128i curr = _mm_add_epi16(nears, diff); // current row + + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + __m128i prv0 = _mm_slli_si128(curr, 2); + __m128i nxt0 = _mm_srli_si128(curr, 2); + __m128i prev = _mm_insert_epi16(prv0, t1, 0); + __m128i next = _mm_insert_epi16(nxt0, 3 * in_near[i + 8] + in_far[i + 8], 7); + + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + __m128i bias = _mm_set1_epi16(8); + __m128i curs = _mm_slli_epi16(curr, 2); + __m128i prvd = _mm_sub_epi16(prev, curr); + __m128i nxtd = _mm_sub_epi16(next, curr); + __m128i curb = _mm_add_epi16(curs, bias); + __m128i even = _mm_add_epi16(prvd, curb); + __m128i odd = _mm_add_epi16(nxtd, curb); + + // interleave even and odd pixels, then undo scaling. + __m128i int0 = _mm_unpacklo_epi16(even, odd); + __m128i int1 = _mm_unpackhi_epi16(even, odd); + __m128i de0 = _mm_srli_epi16(int0, 4); + __m128i de1 = _mm_srli_epi16(int1, 4); + + // pack and write output + __m128i outv = _mm_packus_epi16(de0, de1); + _mm_storeu_si128((__m128i*) (out + i * 2), outv); +#elif defined(STBI_NEON) + // load and perform the vertical filtering pass + // this uses 3*x + y = 4*x + (y - x) + uint8x8_t farb = vld1_u8(in_far + i); + uint8x8_t nearb = vld1_u8(in_near + i); + int16x8_t diff = vreinterpretq_s16_u16(vsubl_u8(farb, nearb)); + int16x8_t nears = vreinterpretq_s16_u16(vshll_n_u8(nearb, 2)); + int16x8_t curr = vaddq_s16(nears, diff); // current row + + // horizontal filter works the same based on shifted vers of current + // row. "prev" is current row shifted right by 1 pixel; we need to + // insert the previous pixel value (from t1). + // "next" is current row shifted left by 1 pixel, with first pixel + // of next block of 8 pixels added in. + int16x8_t prv0 = vextq_s16(curr, curr, 7); + int16x8_t nxt0 = vextq_s16(curr, curr, 1); + int16x8_t prev = vsetq_lane_s16(t1, prv0, 0); + int16x8_t next = vsetq_lane_s16(3 * in_near[i + 8] + in_far[i + 8], nxt0, 7); + + // horizontal filter, polyphase implementation since it's convenient: + // even pixels = 3*cur + prev = cur*4 + (prev - cur) + // odd pixels = 3*cur + next = cur*4 + (next - cur) + // note the shared term. + int16x8_t curs = vshlq_n_s16(curr, 2); + int16x8_t prvd = vsubq_s16(prev, curr); + int16x8_t nxtd = vsubq_s16(next, curr); + int16x8_t even = vaddq_s16(curs, prvd); + int16x8_t odd = vaddq_s16(curs, nxtd); + + // undo scaling and round, then store with even/odd phases interleaved + uint8x8x2_t o; + o.val[0] = vqrshrun_n_s16(even, 4); + o.val[1] = vqrshrun_n_s16(odd, 4); + vst2_u8(out + i * 2, o); +#endif + + // "previous" value for next iter + t1 = 3 * in_near[i + 7] + in_far[i + 7]; + } + + t0 = t1; + t1 = 3 * in_near[i] + in_far[i]; + out[i * 2] = stbi__div16(3 * t1 + t0 + 8); + + for (++i; i < w; ++i) { + t0 = t1; + t1 = 3 * in_near[i] + in_far[i]; + out[i * 2 - 1] = stbi__div16(3 * t0 + t1 + 8); + out[i * 2] = stbi__div16(3 * t1 + t0 + 8); + } + out[w * 2 - 1] = stbi__div4(t1 + 2); + + STBI_NOTUSED(hs); + + return out; +} +#endif + +static 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; +} + +// this is a reduced-precision calculation of YCbCr-to-RGB introduced +// to make sure the code produces the same results in both SIMD and scalar +#define stbi__float2fixed(x) (((int) ((x) * 4096.0f + 0.5f)) << 8) +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] << 20) + (1 << 19); // rounding + int r, g, b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr * stbi__float2fixed(1.40200f); + g = y_fixed + (cr * -stbi__float2fixed(0.71414f)) + ((cb * -stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb * stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned)r > 255) { if (r < 0) r = 0; else r = 255; } + if ((unsigned)g > 255) { if (g < 0) g = 0; else g = 255; } + if ((unsigned)b > 255) { if (b < 0) b = 0; else b = 255; } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } +} + +#if defined(STBI_SSE2) || defined(STBI_NEON) +static void stbi__YCbCr_to_RGB_simd(stbi_uc* out, stbi_uc const* y, stbi_uc const* pcb, stbi_uc const* pcr, int count, int step) +{ + int i = 0; + +#ifdef STBI_SSE2 + // step == 3 is pretty ugly on the final interleave, and i'm not convinced + // it's useful in practice (you wouldn't use it for textures, for example). + // so just accelerate step == 4 case. + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + __m128i signflip = _mm_set1_epi8(-0x80); + __m128i cr_const0 = _mm_set1_epi16((short)(1.40200f * 4096.0f + 0.5f)); + __m128i cr_const1 = _mm_set1_epi16(-(short)(0.71414f * 4096.0f + 0.5f)); + __m128i cb_const0 = _mm_set1_epi16(-(short)(0.34414f * 4096.0f + 0.5f)); + __m128i cb_const1 = _mm_set1_epi16((short)(1.77200f * 4096.0f + 0.5f)); + __m128i y_bias = _mm_set1_epi8((char)(unsigned char)128); + __m128i xw = _mm_set1_epi16(255); // alpha channel + + for (; i + 7 < count; i += 8) { + // load + __m128i y_bytes = _mm_loadl_epi64((__m128i*) (y + i)); + __m128i cr_bytes = _mm_loadl_epi64((__m128i*) (pcr + i)); + __m128i cb_bytes = _mm_loadl_epi64((__m128i*) (pcb + i)); + __m128i cr_biased = _mm_xor_si128(cr_bytes, signflip); // -128 + __m128i cb_biased = _mm_xor_si128(cb_bytes, signflip); // -128 + + // unpack to short (and left-shift cr, cb by 8) + __m128i yw = _mm_unpacklo_epi8(y_bias, y_bytes); + __m128i crw = _mm_unpacklo_epi8(_mm_setzero_si128(), cr_biased); + __m128i cbw = _mm_unpacklo_epi8(_mm_setzero_si128(), cb_biased); + + // color transform + __m128i yws = _mm_srli_epi16(yw, 4); + __m128i cr0 = _mm_mulhi_epi16(cr_const0, crw); + __m128i cb0 = _mm_mulhi_epi16(cb_const0, cbw); + __m128i cb1 = _mm_mulhi_epi16(cbw, cb_const1); + __m128i cr1 = _mm_mulhi_epi16(crw, cr_const1); + __m128i rws = _mm_add_epi16(cr0, yws); + __m128i gwt = _mm_add_epi16(cb0, yws); + __m128i bws = _mm_add_epi16(yws, cb1); + __m128i gws = _mm_add_epi16(gwt, cr1); + + // descale + __m128i rw = _mm_srai_epi16(rws, 4); + __m128i bw = _mm_srai_epi16(bws, 4); + __m128i gw = _mm_srai_epi16(gws, 4); + + // back to byte, set up for transpose + __m128i brb = _mm_packus_epi16(rw, bw); + __m128i gxb = _mm_packus_epi16(gw, xw); + + // transpose to interleave channels + __m128i t0 = _mm_unpacklo_epi8(brb, gxb); + __m128i t1 = _mm_unpackhi_epi8(brb, gxb); + __m128i o0 = _mm_unpacklo_epi16(t0, t1); + __m128i o1 = _mm_unpackhi_epi16(t0, t1); + + // store + _mm_storeu_si128((__m128i*) (out + 0), o0); + _mm_storeu_si128((__m128i*) (out + 16), o1); + out += 32; + } + } +#endif + +#ifdef STBI_NEON + // in this version, step=3 support would be easy to add. but is there demand? + if (step == 4) { + // this is a fairly straightforward implementation and not super-optimized. + uint8x8_t signflip = vdup_n_u8(0x80); + int16x8_t cr_const0 = vdupq_n_s16((short)(1.40200f * 4096.0f + 0.5f)); + int16x8_t cr_const1 = vdupq_n_s16(-(short)(0.71414f * 4096.0f + 0.5f)); + int16x8_t cb_const0 = vdupq_n_s16(-(short)(0.34414f * 4096.0f + 0.5f)); + int16x8_t cb_const1 = vdupq_n_s16((short)(1.77200f * 4096.0f + 0.5f)); + + for (; i + 7 < count; i += 8) { + // load + uint8x8_t y_bytes = vld1_u8(y + i); + uint8x8_t cr_bytes = vld1_u8(pcr + i); + uint8x8_t cb_bytes = vld1_u8(pcb + i); + int8x8_t cr_biased = vreinterpret_s8_u8(vsub_u8(cr_bytes, signflip)); + int8x8_t cb_biased = vreinterpret_s8_u8(vsub_u8(cb_bytes, signflip)); + + // expand to s16 + int16x8_t yws = vreinterpretq_s16_u16(vshll_n_u8(y_bytes, 4)); + int16x8_t crw = vshll_n_s8(cr_biased, 7); + int16x8_t cbw = vshll_n_s8(cb_biased, 7); + + // color transform + int16x8_t cr0 = vqdmulhq_s16(crw, cr_const0); + int16x8_t cb0 = vqdmulhq_s16(cbw, cb_const0); + int16x8_t cr1 = vqdmulhq_s16(crw, cr_const1); + int16x8_t cb1 = vqdmulhq_s16(cbw, cb_const1); + int16x8_t rws = vaddq_s16(yws, cr0); + int16x8_t gws = vaddq_s16(vaddq_s16(yws, cb0), cr1); + int16x8_t bws = vaddq_s16(yws, cb1); + + // undo scaling, round, convert to byte + uint8x8x4_t o; + o.val[0] = vqrshrun_n_s16(rws, 4); + o.val[1] = vqrshrun_n_s16(gws, 4); + o.val[2] = vqrshrun_n_s16(bws, 4); + o.val[3] = vdup_n_u8(255); + + // store, interleaving r/g/b/a + vst4_u8(out, o); + out += 8 * 4; + } + } +#endif + + for (; i < count; ++i) { + int y_fixed = (y[i] << 20) + (1 << 19); // rounding + int r, g, b; + int cr = pcr[i] - 128; + int cb = pcb[i] - 128; + r = y_fixed + cr * stbi__float2fixed(1.40200f); + g = y_fixed + cr * -stbi__float2fixed(0.71414f) + ((cb * -stbi__float2fixed(0.34414f)) & 0xffff0000); + b = y_fixed + cb * stbi__float2fixed(1.77200f); + r >>= 20; + g >>= 20; + b >>= 20; + if ((unsigned)r > 255) { if (r < 0) r = 0; else r = 255; } + if ((unsigned)g > 255) { if (g < 0) g = 0; else g = 255; } + if ((unsigned)b > 255) { if (b < 0) b = 0; else b = 255; } + out[0] = (stbi_uc)r; + out[1] = (stbi_uc)g; + out[2] = (stbi_uc)b; + out[3] = 255; + out += step; + } +} +#endif + +// set up the kernels +static void stbi__setup_jpeg(stbi__jpeg* j) +{ + j->idct_block_kernel = stbi__idct_block; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_row; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2; + +#ifdef STBI_SSE2 + if (stbi__sse2_available()) { + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; + } +#endif + +#ifdef STBI_NEON + j->idct_block_kernel = stbi__idct_simd; + j->YCbCr_to_RGB_kernel = stbi__YCbCr_to_RGB_simd; + j->resample_row_hv_2_kernel = stbi__resample_row_hv_2_simd; +#endif +} + +// clean up the temporary component buffers +static void stbi__cleanup_jpeg(stbi__jpeg* j) +{ + stbi__free_jpeg_components(j, j->s->img_n, 0); +} + +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; + +// fast 0..255 * 0..255 => 0..255 rounded multiplication +static stbi_uc stbi__blinn_8x8(stbi_uc x, stbi_uc y) +{ + unsigned int t = x * y + 128; + return (stbi_uc)((t + (t >> 8)) >> 8); +} + +static stbi_uc* load_jpeg_image(stbi__jpeg* z, int* out_x, int* out_y, int* comp, int req_comp) +{ + int n, decode_n, is_rgb; + 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, but leave in YCbCr format + if (!stbi__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 >= 3 ? 3 : 1; + + is_rgb = z->s->img_n == 3 && (z->rgb == 3 || (z->app14_color_transform == 0 && !z->jfif)); + + if (z->s->img_n == 3 && n < 3 && !is_rgb) + 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 = z->resample_row_hv_2_kernel; + else r->resample = stbi__resample_row_generic; + } + + // can't error after this so, this is safe + output = (stbi_uc*)stbi__malloc_mad3(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) { + if (is_rgb) { + for (i = 0; i < z->s->img_x; ++i) { + out[0] = y[i]; + out[1] = coutput[1][i]; + out[2] = coutput[2][i]; + out[3] = 255; + out += n; + } + } + else { + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + } + } + else if (z->s->img_n == 4) { + if (z->app14_color_transform == 0) { // CMYK + for (i = 0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(coutput[0][i], m); + out[1] = stbi__blinn_8x8(coutput[1][i], m); + out[2] = stbi__blinn_8x8(coutput[2][i], m); + out[3] = 255; + out += n; + } + } + else if (z->app14_color_transform == 2) { // YCCK + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + for (i = 0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + out[0] = stbi__blinn_8x8(255 - out[0], m); + out[1] = stbi__blinn_8x8(255 - out[1], m); + out[2] = stbi__blinn_8x8(255 - out[2], m); + out += n; + } + } + else { // YCbCr + alpha? Ignore the fourth channel for now + z->YCbCr_to_RGB_kernel(out, y, coutput[1], coutput[2], z->s->img_x, n); + } + } + else + for (i = 0; i < z->s->img_x; ++i) { + out[0] = out[1] = out[2] = y[i]; + out[3] = 255; // not used if n==3 + out += n; + } + } + else { + if (is_rgb) { + if (n == 1) + for (i = 0; i < z->s->img_x; ++i) + * out++ = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + else { + for (i = 0; i < z->s->img_x; ++i, out += 2) { + out[0] = stbi__compute_y(coutput[0][i], coutput[1][i], coutput[2][i]); + out[1] = 255; + } + } + } + else if (z->s->img_n == 4 && z->app14_color_transform == 0) { + for (i = 0; i < z->s->img_x; ++i) { + stbi_uc m = coutput[3][i]; + stbi_uc r = stbi__blinn_8x8(coutput[0][i], m); + stbi_uc g = stbi__blinn_8x8(coutput[1][i], m); + stbi_uc b = stbi__blinn_8x8(coutput[2][i], m); + out[0] = stbi__compute_y(r, g, b); + out[1] = 255; + out += n; + } + } + else if (z->s->img_n == 4 && z->app14_color_transform == 2) { + for (i = 0; i < z->s->img_x; ++i) { + out[0] = stbi__blinn_8x8(255 - coutput[0][i], coutput[3][i]); + out[1] = 255; + 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 >= 3 ? 3 : 1; // report original components, not output + return output; + } +} + +static void* stbi__jpeg_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri) +{ + unsigned char* result; + stbi__jpeg* j = (stbi__jpeg*)stbi__malloc(sizeof(stbi__jpeg)); + STBI_NOTUSED(ri); + j->s = s; + stbi__setup_jpeg(j); + result = load_jpeg_image(j, x, y, comp, req_comp); + STBI_FREE(j); + return result; +} + +static int stbi__jpeg_test(stbi__context* s) +{ + int r; + stbi__jpeg* j = (stbi__jpeg*)stbi__malloc(sizeof(stbi__jpeg)); + j->s = s; + stbi__setup_jpeg(j); + r = stbi__decode_jpeg_header(j, STBI__SCAN_type); + stbi__rewind(s); + STBI_FREE(j); + return r; +} + +static int stbi__jpeg_info_raw(stbi__jpeg* j, int* x, int* y, int* comp) +{ + if (!stbi__decode_jpeg_header(j, STBI__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 >= 3 ? 3 : 1; + return 1; +} + +static int stbi__jpeg_info(stbi__context* s, int* x, int* y, int* comp) +{ + int result; + stbi__jpeg* j = (stbi__jpeg*)(stbi__malloc(sizeof(stbi__jpeg))); + j->s = s; + result = stbi__jpeg_info_raw(j, x, y, comp); + STBI_FREE(j); + return result; +} +#endif + +// 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 + +#ifndef STBI_NO_ZLIB + +// 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, const 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, 0, sizeof(z->fast)); + for (i = 0; i < num; ++i) + ++sizes[sizelist[i]]; + sizes[0] = 0; + for (i = 1; i < 16; ++i) + if (sizes[i] > (1 << i)) + return stbi__err("bad sizes", "Corrupt PNG"); + code = 0; + for (i = 1; i < 16; ++i) { + next_code[i] = code; + z->firstcode[i] = (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 PNG"); + 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]; + stbi__uint16 fastv = (stbi__uint16)((s << 9) | i); + z->size[c] = (stbi_uc)s; + z->value[c] = (stbi__uint16)i; + if (s <= STBI__ZFAST_BITS) { + int j = stbi__bit_reverse(next_code[s], s); + while (j < (1 << STBI__ZFAST_BITS)) { + z->fast[j] = fastv; + j += (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 |= (unsigned int)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; +} + +static int stbi__zhuffman_decode_slowpath(stbi__zbuf* a, stbi__zhuffman* z) +{ + int b, s, k; + // 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]; +} + +stbi_inline static int stbi__zhuffman_decode(stbi__zbuf* a, stbi__zhuffman* z) +{ + int b, s; + if (a->num_bits < 16) stbi__fill_bits(a); + b = z->fast[a->code_buffer & STBI__ZFAST_MASK]; + if (b) { + s = b >> 9; + a->code_buffer >>= s; + a->num_bits -= s; + return b & 511; + } + return stbi__zhuffman_decode_slowpath(a, z); +} + +static int stbi__zexpand(stbi__zbuf* z, char* zout, int n) // need to make room for n bytes +{ + char* q; + int cur, limit, old_limit; + z->zout = zout; + if (!z->z_expandable) return stbi__err("output buffer limit", "Corrupt PNG"); + cur = (int)(z->zout - z->zout_start); + limit = old_limit = (int)(z->zout_end - z->zout_start); + while (cur + n > limit) + limit *= 2; + q = (char*)STBI_REALLOC_SIZED(z->zout_start, old_limit, limit); + STBI_NOTUSED(old_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 const 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 const 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 const 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 const int stbi__zdist_extra[32] = +{ 0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13 }; + +static int stbi__parse_huffman_block(stbi__zbuf* a) +{ + char* zout = a->zout; + for (;;) { + int z = stbi__zhuffman_decode(a, &a->z_length); + if (z < 256) { + if (z < 0) return stbi__err("bad huffman code", "Corrupt PNG"); // error in huffman codes + if (zout >= a->zout_end) { + if (!stbi__zexpand(a, zout, 1)) return 0; + zout = a->zout; + } + *zout++ = (char)z; + } + else { + stbi_uc* p; + int len, dist; + if (z == 256) { + a->zout = zout; + return 1; + } + z -= 257; + len = stbi__zlength_base[z]; + if (stbi__zlength_extra[z]) len += stbi__zreceive(a, stbi__zlength_extra[z]); + z = stbi__zhuffman_decode(a, &a->z_distance); + if (z < 0) return stbi__err("bad huffman code", "Corrupt PNG"); + dist = stbi__zdist_base[z]; + if (stbi__zdist_extra[z]) dist += stbi__zreceive(a, stbi__zdist_extra[z]); + if (zout - a->zout_start < dist) return stbi__err("bad dist", "Corrupt PNG"); + if (zout + len > a->zout_end) { + if (!stbi__zexpand(a, zout, len)) return 0; + zout = a->zout; + } + p = (stbi_uc*)(zout - dist); + if (dist == 1) { // run of one byte; common in images. + stbi_uc v = *p; + if (len) { do *zout++ = v; while (--len); } + } + else { + if (len) { do *zout++ = *p++; while (--len); } + } + } + } +} + +static int stbi__compute_huffman_codes(stbi__zbuf* a) +{ + static const 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; + int ntot = hlit + hdist; + + 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 < ntot) { + int c = stbi__zhuffman_decode(a, &z_codelength); + if (c < 0 || c >= 19) return stbi__err("bad codelengths", "Corrupt PNG"); + if (c < 16) + lencodes[n++] = (stbi_uc)c; + else { + stbi_uc fill = 0; + if (c == 16) { + c = stbi__zreceive(a, 2) + 3; + if (n == 0) return stbi__err("bad codelengths", "Corrupt PNG"); + fill = lencodes[n - 1]; + } + else if (c == 17) + c = stbi__zreceive(a, 3) + 3; + else { + STBI_ASSERT(c == 18); + c = stbi__zreceive(a, 7) + 11; + } + if (ntot - n < c) return stbi__err("bad codelengths", "Corrupt PNG"); + memset(lencodes + n, fill, c); + n += c; + } + } + if (n != ntot) 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_uncompressed_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, a->zout, len)) return 0; + memcpy(a->zout, a->zbuffer, len); + a->zbuffer += len; + a->zout += len; + return 1; +} + +static int stbi__parse_zlib_header(stbi__zbuf* a) +{ + int cmf = stbi__zget8(a); + int cm = cmf & 15; + /* int cinfo = cmf >> 4; */ + int flg = stbi__zget8(a); + if ((cmf * 256 + flg) % 31 != 0) return stbi__err("bad zlib header", "Corrupt PNG"); // 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; +} + +static const stbi_uc stbi__zdefault_length[288] = +{ + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, + 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, + 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, 7,7,7,7,7,7,7,7,8,8,8,8,8,8,8,8 +}; +static const stbi_uc stbi__zdefault_distance[32] = +{ + 5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5,5 +}; +/* +Init algorithm: +{ + 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_uncompressed_block(a)) return 0; + } + else if (type == 3) { + return 0; + } + else { + if (type == 1) { + // use fixed code lengths + 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 { + STBI_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 { + STBI_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 { + STBI_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; +} +#endif + +// 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 + +#ifndef STBI_NO_PNG +typedef struct +{ + stbi__uint32 length; + stbi__uint32 type; +} stbi__pngchunk; + +static stbi__pngchunk stbi__get_chunk_header(stbi__context* s) +{ + stbi__pngchunk c; + c.length = stbi__get32be(s); + c.type = stbi__get32be(s); + return c; +} + +static int stbi__check_png_header(stbi__context* s) +{ + static const 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; + int depth; +} stbi__png; + + +enum { + STBI__F_none = 0, + STBI__F_sub = 1, + STBI__F_up = 2, + STBI__F_avg = 3, + STBI__F_paeth = 4, + // synthetic filters used for first scanline to avoid needing a dummy row of 0s + 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; +} + +static const stbi_uc stbi__depth_scale_table[9] = { 0, 0xff, 0x55, 0, 0x11, 0,0,0, 0x01 }; + +// 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, int depth, int color) +{ + int bytes = (depth == 16 ? 2 : 1); + stbi__context* s = a->s; + stbi__uint32 i, j, stride = x * out_n * bytes; + stbi__uint32 img_len, img_width_bytes; + int k; + int img_n = s->img_n; // copy it into a local for later + + int output_bytes = out_n * bytes; + int filter_bytes = img_n * bytes; + int width = x; + + STBI_ASSERT(out_n == s->img_n || out_n == s->img_n + 1); + a->out = (stbi_uc*)stbi__malloc_mad3(x, y, output_bytes, 0); // extra bytes to write off the end into + if (!a->out) return stbi__err("outofmem", "Out of memory"); + + if (!stbi__mad3sizes_valid(img_n, x, depth, 7)) return stbi__err("too large", "Corrupt PNG"); + img_width_bytes = (((img_n * x * depth) + 7) >> 3); + img_len = (img_width_bytes + 1) * y; + + // we used to check for exact match between raw_len and img_len on non-interlaced PNGs, + // but issue #276 reported a PNG in the wild that had extra data at the end (all zeros), + // so just check for raw_len < img_len always. + if (raw_len < img_len) return stbi__err("not enough pixels", "Corrupt PNG"); + + for (j = 0; j < y; ++j) { + stbi_uc* cur = a->out + stride * j; + stbi_uc* prior; + int filter = *raw++; + + if (filter > 4) + return stbi__err("invalid filter", "Corrupt PNG"); + + if (depth < 8) { + STBI_ASSERT(img_width_bytes <= x); + cur += x * out_n - img_width_bytes; // store output to the rightmost img_len bytes, so we can decode in place + filter_bytes = 1; + width = img_width_bytes; + } + prior = cur - stride; // bugfix: need to compute this after 'cur +=' computation above + + // if first row, use special filter that doesn't sample previous row + if (j == 0) filter = first_row_filter[filter]; + + // handle first byte explicitly + for (k = 0; k < filter_bytes; ++k) { + switch (filter) { + case STBI__F_none: cur[k] = raw[k]; break; + case STBI__F_sub: cur[k] = raw[k]; break; + case STBI__F_up: cur[k] = STBI__BYTECAST(raw[k] + prior[k]); break; + case STBI__F_avg: cur[k] = STBI__BYTECAST(raw[k] + (prior[k] >> 1)); break; + case STBI__F_paeth: cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(0, prior[k], 0)); break; + case STBI__F_avg_first: cur[k] = raw[k]; break; + case STBI__F_paeth_first: cur[k] = raw[k]; break; + } + } + + if (depth == 8) { + if (img_n != out_n) + cur[img_n] = 255; // first pixel + raw += img_n; + cur += out_n; + prior += out_n; + } + else if (depth == 16) { + if (img_n != out_n) { + cur[filter_bytes] = 255; // first pixel top byte + cur[filter_bytes + 1] = 255; // first pixel bottom byte + } + raw += filter_bytes; + cur += output_bytes; + prior += output_bytes; + } + else { + raw += 1; + cur += 1; + prior += 1; + } + + // this is a little gross, so that we don't switch per-pixel or per-component + if (depth < 8 || img_n == out_n) { + int nk = (width - 1) * filter_bytes; +#define STBI__CASE(f) \ + case f: \ + for (k=0; k < nk; ++k) + switch (filter) { + // "none" filter turns into a memcpy here; make that explicit. + case STBI__F_none: memcpy(cur, raw, nk); break; + STBI__CASE(STBI__F_sub) { cur[k] = STBI__BYTECAST(raw[k] + cur[k - filter_bytes]); } break; + STBI__CASE(STBI__F_up) { cur[k] = STBI__BYTECAST(raw[k] + prior[k]); } break; + STBI__CASE(STBI__F_avg) { cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k - filter_bytes]) >> 1)); } break; + STBI__CASE(STBI__F_paeth) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k - filter_bytes], prior[k], prior[k - filter_bytes])); } break; + STBI__CASE(STBI__F_avg_first) { cur[k] = STBI__BYTECAST(raw[k] + (cur[k - filter_bytes] >> 1)); } break; + STBI__CASE(STBI__F_paeth_first) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k - filter_bytes], 0, 0)); } break; + } +#undef STBI__CASE + raw += nk; + } + else { + STBI_ASSERT(img_n + 1 == out_n); +#define STBI__CASE(f) \ + case f: \ + for (i=x-1; i >= 1; --i, cur[filter_bytes]=255,raw+=filter_bytes,cur+=output_bytes,prior+=output_bytes) \ + for (k=0; k < filter_bytes; ++k) + switch (filter) { + STBI__CASE(STBI__F_none) { cur[k] = raw[k]; } break; + STBI__CASE(STBI__F_sub) { cur[k] = STBI__BYTECAST(raw[k] + cur[k - output_bytes]); } break; + STBI__CASE(STBI__F_up) { cur[k] = STBI__BYTECAST(raw[k] + prior[k]); } break; + STBI__CASE(STBI__F_avg) { cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k - output_bytes]) >> 1)); } break; + STBI__CASE(STBI__F_paeth) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k - output_bytes], prior[k], prior[k - output_bytes])); } break; + STBI__CASE(STBI__F_avg_first) { cur[k] = STBI__BYTECAST(raw[k] + (cur[k - output_bytes] >> 1)); } break; + STBI__CASE(STBI__F_paeth_first) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k - output_bytes], 0, 0)); } break; + } +#undef STBI__CASE + + // the loop above sets the high byte of the pixels' alpha, but for + // 16 bit png files we also need the low byte set. we'll do that here. + if (depth == 16) { + cur = a->out + stride * j; // start at the beginning of the row again + for (i = 0; i < x; ++i, cur += output_bytes) { + cur[filter_bytes + 1] = 255; + } + } + } + } + + // we make a separate pass to expand bits to pixels; for performance, + // this could run two scanlines behind the above code, so it won't + // intefere with filtering but will still be in the cache. + if (depth < 8) { + for (j = 0; j < y; ++j) { + stbi_uc* cur = a->out + stride * j; + stbi_uc* in = a->out + stride * j + x * out_n - img_width_bytes; + // unpack 1/2/4-bit into a 8-bit buffer. allows us to keep the common 8-bit path optimal at minimal cost for 1/2/4-bit + // png guarante byte alignment, if width is not multiple of 8/4/2 we'll decode dummy trailing data that will be skipped in the later loop + stbi_uc scale = (color == 0) ? stbi__depth_scale_table[depth] : 1; // scale grayscale values to 0..255 range + + // note that the final byte might overshoot and write more data than desired. + // we can allocate enough data that this never writes out of memory, but it + // could also overwrite the next scanline. can it overwrite non-empty data + // on the next scanline? yes, consider 1-pixel-wide scanlines with 1-bit-per-pixel. + // so we need to explicitly clamp the final ones + + if (depth == 4) { + for (k = x * img_n; k >= 2; k -= 2, ++in) { + *cur++ = scale * ((*in >> 4)); + *cur++ = scale * ((*in) & 0x0f); + } + if (k > 0)* cur++ = scale * ((*in >> 4)); + } + else if (depth == 2) { + for (k = x * img_n; k >= 4; k -= 4, ++in) { + *cur++ = scale * ((*in >> 6)); + *cur++ = scale * ((*in >> 4) & 0x03); + *cur++ = scale * ((*in >> 2) & 0x03); + *cur++ = scale * ((*in) & 0x03); + } + if (k > 0)* cur++ = scale * ((*in >> 6)); + if (k > 1)* cur++ = scale * ((*in >> 4) & 0x03); + if (k > 2)* cur++ = scale * ((*in >> 2) & 0x03); + } + else if (depth == 1) { + for (k = x * img_n; k >= 8; k -= 8, ++in) { + *cur++ = scale * ((*in >> 7)); + *cur++ = scale * ((*in >> 6) & 0x01); + *cur++ = scale * ((*in >> 5) & 0x01); + *cur++ = scale * ((*in >> 4) & 0x01); + *cur++ = scale * ((*in >> 3) & 0x01); + *cur++ = scale * ((*in >> 2) & 0x01); + *cur++ = scale * ((*in >> 1) & 0x01); + *cur++ = scale * ((*in) & 0x01); + } + if (k > 0)* cur++ = scale * ((*in >> 7)); + if (k > 1)* cur++ = scale * ((*in >> 6) & 0x01); + if (k > 2)* cur++ = scale * ((*in >> 5) & 0x01); + if (k > 3)* cur++ = scale * ((*in >> 4) & 0x01); + if (k > 4)* cur++ = scale * ((*in >> 3) & 0x01); + if (k > 5)* cur++ = scale * ((*in >> 2) & 0x01); + if (k > 6)* cur++ = scale * ((*in >> 1) & 0x01); + } + if (img_n != out_n) { + int q; + // insert alpha = 255 + cur = a->out + stride * j; + if (img_n == 1) { + for (q = x - 1; q >= 0; --q) { + cur[q * 2 + 1] = 255; + cur[q * 2 + 0] = cur[q]; + } + } + else { + STBI_ASSERT(img_n == 3); + for (q = x - 1; q >= 0; --q) { + cur[q * 4 + 3] = 255; + cur[q * 4 + 2] = cur[q * 3 + 2]; + cur[q * 4 + 1] = cur[q * 3 + 1]; + cur[q * 4 + 0] = cur[q * 3 + 0]; + } + } + } + } + } + else if (depth == 16) { + // force the image data from big-endian to platform-native. + // this is done in a separate pass due to the decoding relying + // on the data being untouched, but could probably be done + // per-line during decode if care is taken. + stbi_uc* cur = a->out; + stbi__uint16* cur16 = (stbi__uint16*)cur; + + for (i = 0; i < x * y * out_n; ++i, cur16++, cur += 2) { + *cur16 = (cur[0] << 8) | cur[1]; + } + } + + return 1; +} + +static int stbi__create_png_image(stbi__png* a, stbi_uc* image_data, stbi__uint32 image_data_len, int out_n, int depth, int color, int interlaced) +{ + int bytes = (depth == 16 ? 2 : 1); + int out_bytes = out_n * bytes; + stbi_uc* final; + int p; + if (!interlaced) + return stbi__create_png_image_raw(a, image_data, image_data_len, out_n, a->s->img_x, a->s->img_y, depth, color); + + // de-interlacing + final = (stbi_uc*)stbi__malloc_mad3(a->s->img_x, a->s->img_y, out_bytes, 0); + 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) { + stbi__uint32 img_len = ((((a->s->img_n * x * depth) + 7) >> 3) + 1) * y; + if (!stbi__create_png_image_raw(a, image_data, image_data_len, out_n, x, y, depth, color)) { + STBI_FREE(final); + return 0; + } + for (j = 0; j < y; ++j) { + for (i = 0; i < x; ++i) { + int out_y = j * yspc[p] + yorig[p]; + int out_x = i * xspc[p] + xorig[p]; + memcpy(final + out_y * a->s->img_x * out_bytes + out_x * out_bytes, + a->out + (j * x + i) * out_bytes, out_bytes); + } + } + STBI_FREE(a->out); + image_data += img_len; + image_data_len -= img_len; + } + } + a->out = final; + + return 1; +} + +static int stbi__compute_transparency(stbi__png* z, 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__compute_transparency16(stbi__png* z, stbi__uint16 tc[3], int out_n) +{ + stbi__context* s = z->s; + stbi__uint32 i, pixel_count = s->img_x * s->img_y; + stbi__uint16* p = (stbi__uint16*)z->out; + + // compute color-based transparency, assuming we've + // already got 65535 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 : 65535); + 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_mad2(pixel_count, pal_img_n, 0); + 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; + } + } + STBI_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) { + stbi_uc half = a / 2; + p[0] = (p[2] * 255 + half) / a; + p[1] = (p[1] * 255 + half) / a; + p[2] = (t * 255 + half) / 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; + } + } + } +} + +#define STBI__PNG_TYPE(a,b,c,d) (((unsigned) (a) << 24) + ((unsigned) (b) << 16) + ((unsigned) (c) << 8) + (unsigned) (d)) + +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__uint16 tc16[3]; + stbi__uint32 ioff = 0, idata_limit = 0, i, pal_len = 0; + int first = 1, k, interlace = 0, color = 0, is_iphone = 0; + stbi__context* s = z->s; + + z->expanded = NULL; + z->idata = NULL; + z->out = NULL; + + if (!stbi__check_png_header(s)) return 0; + + if (scan == STBI__SCAN_type) return 1; + + for (;;) { + stbi__pngchunk c = stbi__get_chunk_header(s); + switch (c.type) { + case STBI__PNG_TYPE('C', 'g', 'B', 'I'): + is_iphone = 1; + stbi__skip(s, c.length); + break; + case STBI__PNG_TYPE('I', 'H', 'D', 'R'): { + int comp, filter; + if (!first) return stbi__err("multiple IHDR", "Corrupt PNG"); + first = 0; + if (c.length != 13) return stbi__err("bad IHDR len", "Corrupt PNG"); + s->img_x = stbi__get32be(s); if (s->img_x > (1 << 24)) return stbi__err("too large", "Very large image (corrupt?)"); + s->img_y = stbi__get32be(s); if (s->img_y > (1 << 24)) return stbi__err("too large", "Very large image (corrupt?)"); + z->depth = stbi__get8(s); if (z->depth != 1 && z->depth != 2 && z->depth != 4 && z->depth != 8 && z->depth != 16) return stbi__err("1/2/4/8/16-bit only", "PNG not supported: 1/2/4/8/16-bit only"); + color = stbi__get8(s); if (color > 6) return stbi__err("bad ctype", "Corrupt PNG"); + if (color == 3 && z->depth == 16) return stbi__err("bad ctype", "Corrupt PNG"); + if (color == 3) pal_img_n = 3; else if (color & 1) return stbi__err("bad ctype", "Corrupt PNG"); + comp = stbi__get8(s); if (comp) return stbi__err("bad comp method", "Corrupt PNG"); + filter = stbi__get8(s); if (filter) return stbi__err("bad filter method", "Corrupt PNG"); + interlace = stbi__get8(s); if (interlace > 1) return stbi__err("bad interlace method", "Corrupt PNG"); + if (!s->img_x || !s->img_y) return stbi__err("0-pixel image", "Corrupt PNG"); + if (!pal_img_n) { + s->img_n = (color & 2 ? 3 : 1) + (color & 4 ? 1 : 0); + if ((1 << 30) / s->img_x / s->img_n < s->img_y) return stbi__err("too large", "Image too large to decode"); + if (scan == STBI__SCAN_header) return 1; + } + else { + // 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 STBI__PNG_TYPE('P', 'L', 'T', 'E'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (c.length > 256 * 3) return stbi__err("invalid PLTE", "Corrupt PNG"); + pal_len = c.length / 3; + if (pal_len * 3 != c.length) return stbi__err("invalid PLTE", "Corrupt PNG"); + for (i = 0; i < pal_len; ++i) { + palette[i * 4 + 0] = stbi__get8(s); + palette[i * 4 + 1] = stbi__get8(s); + palette[i * 4 + 2] = stbi__get8(s); + palette[i * 4 + 3] = 255; + } + break; + } + + case STBI__PNG_TYPE('t', 'R', 'N', 'S'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (z->idata) return stbi__err("tRNS after IDAT", "Corrupt PNG"); + if (pal_img_n) { + if (scan == STBI__SCAN_header) { s->img_n = 4; return 1; } + if (pal_len == 0) return stbi__err("tRNS before PLTE", "Corrupt PNG"); + if (c.length > pal_len) return stbi__err("bad tRNS len", "Corrupt PNG"); + pal_img_n = 4; + for (i = 0; i < c.length; ++i) + palette[i * 4 + 3] = stbi__get8(s); + } + else { + if (!(s->img_n & 1)) return stbi__err("tRNS with alpha", "Corrupt PNG"); + if (c.length != (stbi__uint32)s->img_n * 2) return stbi__err("bad tRNS len", "Corrupt PNG"); + has_trans = 1; + if (z->depth == 16) { + for (k = 0; k < s->img_n; ++k) tc16[k] = (stbi__uint16)stbi__get16be(s); // copy the values as-is + } + else { + for (k = 0; k < s->img_n; ++k) tc[k] = (stbi_uc)(stbi__get16be(s) & 255) * stbi__depth_scale_table[z->depth]; // non 8-bit images will be larger + } + } + break; + } + + case STBI__PNG_TYPE('I', 'D', 'A', 'T'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (pal_img_n && !pal_len) return stbi__err("no PLTE", "Corrupt PNG"); + if (scan == STBI__SCAN_header) { s->img_n = pal_img_n; return 1; } + if ((int)(ioff + c.length) < (int)ioff) return 0; + if (ioff + c.length > idata_limit) { + stbi__uint32 idata_limit_old = 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; + STBI_NOTUSED(idata_limit_old); + p = (stbi_uc*)STBI_REALLOC_SIZED(z->idata, idata_limit_old, idata_limit); if (p == NULL) return stbi__err("outofmem", "Out of memory"); + z->idata = p; + } + if (!stbi__getn(s, z->idata + ioff, c.length)) return stbi__err("outofdata", "Corrupt PNG"); + ioff += c.length; + break; + } + + case STBI__PNG_TYPE('I', 'E', 'N', 'D'): { + stbi__uint32 raw_len, bpl; + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (scan != STBI__SCAN_load) return 1; + if (z->idata == NULL) return stbi__err("no IDAT", "Corrupt PNG"); + // initial guess for decoded data size to avoid unnecessary reallocs + bpl = (s->img_x * z->depth + 7) / 8; // bytes per line, per component + raw_len = bpl * s->img_y * s->img_n /* pixels */ + s->img_y /* filter mode per row */; + z->expanded = (stbi_uc*)stbi_zlib_decode_malloc_guesssize_headerflag((char*)z->idata, ioff, raw_len, (int*)& raw_len, !is_iphone); + if (z->expanded == NULL) return 0; // zlib should set error + STBI_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, z->depth, color, interlace)) return 0; + if (has_trans) { + if (z->depth == 16) { + if (!stbi__compute_transparency16(z, tc16, s->img_out_n)) return 0; + } + else { + 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; + } + else if (has_trans) { + // non-paletted image with tRNS -> source image has (constant) alpha + ++s->img_n; + } + STBI_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 void* stbi__do_png(stbi__png* p, int* x, int* y, int* n, int req_comp, stbi__result_info* ri) +{ + void* result = NULL; + if (req_comp < 0 || req_comp > 4) return stbi__errpuc("bad req_comp", "Internal error"); + if (stbi__parse_png_file(p, STBI__SCAN_load, req_comp)) { + if (p->depth < 8) + ri->bits_per_channel = 8; + else + ri->bits_per_channel = p->depth; + result = p->out; + p->out = NULL; + if (req_comp && req_comp != p->s->img_out_n) { + if (ri->bits_per_channel == 8) + result = stbi__convert_format((unsigned char*)result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); + else + result = stbi__convert_format16((stbi__uint16*)result, p->s->img_out_n, req_comp, p->s->img_x, p->s->img_y); + p->s->img_out_n = req_comp; + if (result == NULL) return result; + } + *x = p->s->img_x; + *y = p->s->img_y; + if (n)* n = p->s->img_n; + } + STBI_FREE(p->out); p->out = NULL; + STBI_FREE(p->expanded); p->expanded = NULL; + STBI_FREE(p->idata); p->idata = NULL; + + return result; +} + +static void* stbi__png_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri) +{ + stbi__png p; + p.s = s; + return stbi__do_png(&p, x, y, comp, req_comp, ri); +} + +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, STBI__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); +} + +static int stbi__png_is16(stbi__context* s) +{ + stbi__png p; + p.s = s; + if (!stbi__png_info_raw(&p, NULL, NULL, NULL)) + return 0; + if (p.depth != 16) { + stbi__rewind(p.s); + return 0; + } + return 1; +} +#endif + +// Microsoft/Windows BMP image + +#ifndef STBI_NO_BMP +static int stbi__bmp_test_raw(stbi__context* s) +{ + int r; + int sz; + if (stbi__get8(s) != 'B') return 0; + if (stbi__get8(s) != 'M') return 0; + stbi__get32le(s); // 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; +} + +// extract an arbitrarily-aligned N-bit value (N=bits) +// from v, and then make it 8-bits long and fractionally +// extend it to full full range. +static int stbi__shiftsigned(int v, int shift, int bits) +{ + static unsigned int mul_table[9] = { + 0, + 0xff/*0b11111111*/, 0x55/*0b01010101*/, 0x49/*0b01001001*/, 0x11/*0b00010001*/, + 0x21/*0b00100001*/, 0x41/*0b01000001*/, 0x81/*0b10000001*/, 0x01/*0b00000001*/, + }; + static unsigned int shift_table[9] = { + 0, 0,0,1,0,2,4,6,0, + }; + if (shift < 0) + v <<= -shift; + else + v >>= shift; + STBI_ASSERT(v >= 0 && v < 256); + v >>= (8 - bits); + STBI_ASSERT(bits >= 0 && bits <= 8); + return (int)((unsigned)v * mul_table[bits]) >> shift_table[bits]; +} + +typedef struct +{ + int bpp, offset, hsz; + unsigned int mr, mg, mb, ma, all_a; +} stbi__bmp_data; + +static void* stbi__bmp_parse_header(stbi__context* s, stbi__bmp_data* info) +{ + int 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 + info->offset = stbi__get32le(s); + info->hsz = hsz = stbi__get32le(s); + info->mr = info->mg = info->mb = info->ma = 0; + + 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"); + info->bpp = stbi__get16le(s); + if (hsz != 12) { + int 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 (info->bpp == 16 || info->bpp == 32) { + if (compress == 0) { + if (info->bpp == 32) { + info->mr = 0xffu << 16; + info->mg = 0xffu << 8; + info->mb = 0xffu << 0; + info->ma = 0xffu << 24; + info->all_a = 0; // if all_a is 0 at end, then we loaded alpha channel but it was all 0 + } + else { + info->mr = 31u << 10; + info->mg = 31u << 5; + info->mb = 31u << 0; + } + } + else if (compress == 3) { + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + // not documented, but generated by photoshop and handled by mspaint + if (info->mr == info->mg && info->mg == info->mb) { + // ?!?!? + return stbi__errpuc("bad BMP", "bad BMP"); + } + } + else + return stbi__errpuc("bad BMP", "bad BMP"); + } + } + else { + int i; + if (hsz != 108 && hsz != 124) + return stbi__errpuc("bad BMP", "bad BMP"); + info->mr = stbi__get32le(s); + info->mg = stbi__get32le(s); + info->mb = stbi__get32le(s); + info->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 + } + } + } + return (void*)1; +} + + +static void* stbi__bmp_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri) +{ + stbi_uc* out; + unsigned int mr = 0, mg = 0, mb = 0, ma = 0, all_a; + stbi_uc pal[256][4]; + int psize = 0, i, j, width; + int flip_vertically, pad, target; + stbi__bmp_data info; + STBI_NOTUSED(ri); + + info.all_a = 255; + if (stbi__bmp_parse_header(s, &info) == NULL) + return NULL; // error code already set + + flip_vertically = ((int)s->img_y) > 0; + s->img_y = abs((int)s->img_y); + + mr = info.mr; + mg = info.mg; + mb = info.mb; + ma = info.ma; + all_a = info.all_a; + + if (info.hsz == 12) { + if (info.bpp < 24) + psize = (info.offset - 14 - 24) / 3; + } + else { + if (info.bpp < 16) + psize = (info.offset - 14 - info.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 + + // sanity-check size + if (!stbi__mad3sizes_valid(target, s->img_x, s->img_y, 0)) + return stbi__errpuc("too large", "Corrupt BMP"); + + out = (stbi_uc*)stbi__malloc_mad3(target, s->img_x, s->img_y, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + if (info.bpp < 16) { + int z = 0; + if (psize == 0 || psize > 256) { STBI_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 (info.hsz != 12) stbi__get8(s); + pal[i][3] = 255; + } + stbi__skip(s, info.offset - 14 - info.hsz - psize * (info.hsz == 12 ? 3 : 4)); + if (info.bpp == 1) width = (s->img_x + 7) >> 3; + else if (info.bpp == 4) width = (s->img_x + 1) >> 1; + else if (info.bpp == 8) width = s->img_x; + else { STBI_FREE(out); return stbi__errpuc("bad bpp", "Corrupt BMP"); } + pad = (-width) & 3; + if (info.bpp == 1) { + for (j = 0; j < (int)s->img_y; ++j) { + int bit_offset = 7, v = stbi__get8(s); + for (i = 0; i < (int)s->img_x; ++i) { + int color = (v >> bit_offset) & 0x1; + out[z++] = pal[color][0]; + out[z++] = pal[color][1]; + out[z++] = pal[color][2]; + if ((--bit_offset) < 0) { + bit_offset = 7; + v = stbi__get8(s); + } + } + stbi__skip(s, pad); + } + } + else { + 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 (info.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 = (info.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, info.offset - 14 - info.hsz); + if (info.bpp == 24) width = 3 * s->img_x; + else if (info.bpp == 16) width = 2 * s->img_x; + else /* bpp = 32 and pad = 0 */ width = 0; + pad = (-width) & 3; + if (info.bpp == 24) { + easy = 1; + } + else if (info.bpp == 32) { + if (mb == 0xff && mg == 0xff00 && mr == 0x00ff0000 && ma == 0xff000000) + easy = 2; + } + if (!easy) { + if (!mr || !mg || !mb) { STBI_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); + all_a |= a; + if (target == 4) out[z++] = a; + } + } + else { + int bpp = info.bpp; + for (i = 0; i < (int)s->img_x; ++i) { + stbi__uint32 v = (bpp == 16 ? (stbi__uint32)stbi__get16le(s) : stbi__get32le(s)); + unsigned 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); + all_a |= a; + if (target == 4) out[z++] = STBI__BYTECAST(a); + } + } + stbi__skip(s, pad); + } + } + + // if alpha channel is all 0s, replace with all 255s + if (target == 4 && all_a == 0) + for (i = 4 * s->img_x * s->img_y - 1; i >= 0; i -= 4) + out[i] = 255; + + 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; +} +#endif + +// Targa Truevision - TGA +// by Jonathan Dummer +#ifndef STBI_NO_TGA +// returns STBI_rgb or whatever, 0 on error +static int stbi__tga_get_comp(int bits_per_pixel, int is_grey, int* is_rgb16) +{ + // only RGB or RGBA (incl. 16bit) or grey allowed + if (is_rgb16)* is_rgb16 = 0; + switch (bits_per_pixel) { + case 8: return STBI_grey; + case 16: if (is_grey) return STBI_grey_alpha; + // fallthrough + case 15: if (is_rgb16) * is_rgb16 = 1; + return STBI_rgb; + case 24: // fallthrough + case 32: return bits_per_pixel / 8; + default: return 0; + } +} + +static int stbi__tga_info(stbi__context* s, int* x, int* y, int* comp) +{ + int tga_w, tga_h, tga_comp, tga_image_type, tga_bits_per_pixel, tga_colormap_bpp; + int sz, tga_colormap_type; + stbi__get8(s); // discard Offset + tga_colormap_type = stbi__get8(s); // colormap type + if (tga_colormap_type > 1) { + stbi__rewind(s); + return 0; // only RGB or indexed allowed + } + tga_image_type = stbi__get8(s); // image type + if (tga_colormap_type == 1) { // colormapped (paletted) image + if (tga_image_type != 1 && tga_image_type != 9) { + stbi__rewind(s); + return 0; + } + stbi__skip(s, 4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ((sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32)) { + stbi__rewind(s); + return 0; + } + stbi__skip(s, 4); // skip image x and y origin + tga_colormap_bpp = sz; + } + else { // "normal" image w/o colormap - only RGB or grey allowed, +/- RLE + if ((tga_image_type != 2) && (tga_image_type != 3) && (tga_image_type != 10) && (tga_image_type != 11)) { + stbi__rewind(s); + return 0; // only RGB or grey allowed, +/- RLE + } + stbi__skip(s, 9); // skip colormap specification and image x/y origin + tga_colormap_bpp = 0; + } + 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 + } + tga_bits_per_pixel = stbi__get8(s); // bits per pixel + stbi__get8(s); // ignore alpha bits + if (tga_colormap_bpp != 0) { + if ((tga_bits_per_pixel != 8) && (tga_bits_per_pixel != 16)) { + // when using a colormap, tga_bits_per_pixel is the size of the indexes + // I don't think anything but 8 or 16bit indexes makes sense + stbi__rewind(s); + return 0; + } + tga_comp = stbi__tga_get_comp(tga_colormap_bpp, 0, NULL); + } + else { + tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3) || (tga_image_type == 11), NULL); + } + if (!tga_comp) { + stbi__rewind(s); + return 0; + } + if (x)* x = tga_w; + if (y)* y = tga_h; + if (comp)* comp = tga_comp; + return 1; // seems to have passed everything +} + +static int stbi__tga_test(stbi__context* s) +{ + int res = 0; + int sz, tga_color_type; + stbi__get8(s); // discard Offset + tga_color_type = stbi__get8(s); // color type + if (tga_color_type > 1) goto errorEnd; // only RGB or indexed allowed + sz = stbi__get8(s); // image type + if (tga_color_type == 1) { // colormapped (paletted) image + if (sz != 1 && sz != 9) goto errorEnd; // colortype 1 demands image type 1 or 9 + stbi__skip(s, 4); // skip index of first colormap entry and number of entries + sz = stbi__get8(s); // check bits per palette color entry + if ((sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32)) goto errorEnd; + stbi__skip(s, 4); // skip image x and y origin + } + else { // "normal" image w/o colormap + if ((sz != 2) && (sz != 3) && (sz != 10) && (sz != 11)) goto errorEnd; // only RGB or grey allowed, +/- RLE + stbi__skip(s, 9); // skip colormap specification and image x/y origin + } + if (stbi__get16le(s) < 1) goto errorEnd; // test width + if (stbi__get16le(s) < 1) goto errorEnd; // test height + sz = stbi__get8(s); // bits per pixel + if ((tga_color_type == 1) && (sz != 8) && (sz != 16)) goto errorEnd; // for colormapped images, bpp is size of an index + if ((sz != 8) && (sz != 15) && (sz != 16) && (sz != 24) && (sz != 32)) goto errorEnd; + + res = 1; // if we got this far, everything's good and we can return 1 instead of 0 + +errorEnd: + stbi__rewind(s); + return res; +} + +// read 16bit value and convert to 24bit RGB +static void stbi__tga_read_rgb16(stbi__context* s, stbi_uc* out) +{ + stbi__uint16 px = (stbi__uint16)stbi__get16le(s); + stbi__uint16 fiveBitMask = 31; + // we have 3 channels with 5bits each + int r = (px >> 10) & fiveBitMask; + int g = (px >> 5) & fiveBitMask; + int b = px & fiveBitMask; + // Note that this saves the data in RGB(A) order, so it doesn't need to be swapped later + out[0] = (stbi_uc)((r * 255) / 31); + out[1] = (stbi_uc)((g * 255) / 31); + out[2] = (stbi_uc)((b * 255) / 31); + + // some people claim that the most significant bit might be used for alpha + // (possibly if an alpha-bit is set in the "image descriptor byte") + // but that only made 16bit test images completely translucent.. + // so let's treat all 15 and 16bit TGAs as RGB with no alpha. +} + +static void* stbi__tga_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri) +{ + // 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_rgb16 = 0; + int tga_inverted = stbi__get8(s); + // int tga_alpha_bits = tga_inverted & 15; // the 4 lowest bits - unused (useless?) + // image data + unsigned char* tga_data; + unsigned char* tga_palette = NULL; + int i, j; + unsigned char raw_data[4] = { 0 }; + int RLE_count = 0; + int RLE_repeating = 0; + int read_next_pixel = 1; + STBI_NOTUSED(ri); + + // do a tiny bit of precessing + if (tga_image_type >= 8) + { + tga_image_type -= 8; + tga_is_RLE = 1; + } + tga_inverted = 1 - ((tga_inverted >> 5) & 1); + + // If I'm paletted, then I'll use the number of bits from the palette + if (tga_indexed) tga_comp = stbi__tga_get_comp(tga_palette_bits, 0, &tga_rgb16); + else tga_comp = stbi__tga_get_comp(tga_bits_per_pixel, (tga_image_type == 3), &tga_rgb16); + + if (!tga_comp) // shouldn't really happen, stbi__tga_test() should have ensured basic consistency + return stbi__errpuc("bad format", "Can't find out TGA pixelformat"); + + // tga info + *x = tga_width; + *y = tga_height; + if (comp)* comp = tga_comp; + + if (!stbi__mad3sizes_valid(tga_width, tga_height, tga_comp, 0)) + return stbi__errpuc("too large", "Corrupt TGA"); + + tga_data = (unsigned char*)stbi__malloc_mad3(tga_width, tga_height, tga_comp, 0); + 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 && !tga_rgb16) { + for (i = 0; i < tga_height; ++i) { + int row = tga_inverted ? tga_height - i - 1 : i; + stbi_uc* tga_row = tga_data + row * 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_mad2(tga_palette_len, tga_comp, 0); + if (!tga_palette) { + STBI_FREE(tga_data); + return stbi__errpuc("outofmem", "Out of memory"); + } + if (tga_rgb16) { + stbi_uc* pal_entry = tga_palette; + STBI_ASSERT(tga_comp == STBI_rgb); + for (i = 0; i < tga_palette_len; ++i) { + stbi__tga_read_rgb16(s, pal_entry); + pal_entry += tga_comp; + } + } + else if (!stbi__getn(s, tga_palette, tga_palette_len * tga_comp)) { + STBI_FREE(tga_data); + STBI_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 index, then perform the lookup + int pal_idx = (tga_bits_per_pixel == 8) ? stbi__get8(s) : stbi__get16le(s); + if (pal_idx >= tga_palette_len) { + // invalid index + pal_idx = 0; + } + pal_idx *= tga_comp; + for (j = 0; j < tga_comp; ++j) { + raw_data[j] = tga_palette[pal_idx + j]; + } + } + else if (tga_rgb16) { + STBI_ASSERT(tga_comp == STBI_rgb); + stbi__tga_read_rgb16(s, raw_data); + } + else { + // read in the data raw + for (j = 0; j < tga_comp; ++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) + { + STBI_FREE(tga_palette); + } + } + + // swap RGB - if the source data was RGB16, it already is in the right order + if (tga_comp >= 3 && !tga_rgb16) + { + 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; +} +#endif + +// ************************************************************************************************* +// Photoshop PSD loader -- PD by Thatcher Ulrich, integration by Nicolas Schulz, tweaked by STB + +#ifndef STBI_NO_PSD +static int stbi__psd_test(stbi__context* s) +{ + int r = (stbi__get32be(s) == 0x38425053); + stbi__rewind(s); + return r; +} + +static int stbi__psd_decode_rle(stbi__context* s, stbi_uc* p, int pixelCount) +{ + int count, nleft, len; + + count = 0; + while ((nleft = pixelCount - count) > 0) { + len = stbi__get8(s); + if (len == 128) { + // No-op. + } + else if (len < 128) { + // Copy next len+1 bytes literally. + len++; + if (len > nleft) return 0; // corrupt data + 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 = 257 - len; + if (len > nleft) return 0; // corrupt data + val = stbi__get8(s); + count += len; + while (len) { + *p = val; + p += 4; + len--; + } + } + } + + return 1; +} + +static void* stbi__psd_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri, int bpc) +{ + int pixelCount; + int channelCount, compression; + int channel, i; + int bitdepth; + int w, h; + stbi_uc* out; + STBI_NOTUSED(ri); + + // 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. + bitdepth = stbi__get16be(s); + if (bitdepth != 8 && bitdepth != 16) + return stbi__errpuc("unsupported bit depth", "PSD bit depth is not 8 or 16 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"); + + // Check size + if (!stbi__mad3sizes_valid(4, w, h, 0)) + return stbi__errpuc("too large", "Corrupt PSD"); + + // Create the destination image. + + if (!compression && bitdepth == 16 && bpc == 16) { + out = (stbi_uc*)stbi__malloc_mad3(8, w, h, 0); + ri->bits_per_channel = 16; + } + else + 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 += 4) + * p = (channel == 3 ? 255 : 0); + } + else { + // Read the RLE data. + if (!stbi__psd_decode_rle(s, p, pixelCount)) { + STBI_FREE(out); + return stbi__errpuc("corrupt", "bad RLE data"); + } + } + } + + } + 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 (or 16-bit) value for each pixel in the image. + + // Read the data by channel. + for (channel = 0; channel < 4; channel++) { + if (channel >= channelCount) { + // Fill this channel with default data. + if (bitdepth == 16 && bpc == 16) { + stbi__uint16* q = ((stbi__uint16*)out) + channel; + stbi__uint16 val = channel == 3 ? 65535 : 0; + for (i = 0; i < pixelCount; i++, q += 4) + * q = val; + } + else { + stbi_uc* p = out + channel; + stbi_uc val = channel == 3 ? 255 : 0; + for (i = 0; i < pixelCount; i++, p += 4) + * p = val; + } + } + else { + if (ri->bits_per_channel == 16) { // output bpc + stbi__uint16* q = ((stbi__uint16*)out) + channel; + for (i = 0; i < pixelCount; i++, q += 4) + * q = (stbi__uint16)stbi__get16be(s); + } + else { + stbi_uc* p = out + channel; + if (bitdepth == 16) { // input bpc + for (i = 0; i < pixelCount; i++, p += 4) + * p = (stbi_uc)(stbi__get16be(s) >> 8); + } + else { + for (i = 0; i < pixelCount; i++, p += 4) + * p = stbi__get8(s); + } + } + } + } + } + + // remove weird white matte from PSD + if (channelCount >= 4) { + if (ri->bits_per_channel == 16) { + for (i = 0; i < w * h; ++i) { + stbi__uint16* pixel = (stbi__uint16*)out + 4 * i; + if (pixel[3] != 0 && pixel[3] != 65535) { + float a = pixel[3] / 65535.0f; + float ra = 1.0f / a; + float inv_a = 65535.0f * (1 - ra); + pixel[0] = (stbi__uint16)(pixel[0] * ra + inv_a); + pixel[1] = (stbi__uint16)(pixel[1] * ra + inv_a); + pixel[2] = (stbi__uint16)(pixel[2] * ra + inv_a); + } + } + } + else { + for (i = 0; i < w * h; ++i) { + unsigned char* pixel = out + 4 * i; + if (pixel[3] != 0 && pixel[3] != 255) { + float a = pixel[3] / 255.0f; + float ra = 1.0f / a; + float inv_a = 255.0f * (1 - ra); + pixel[0] = (unsigned char)(pixel[0] * ra + inv_a); + pixel[1] = (unsigned char)(pixel[1] * ra + inv_a); + pixel[2] = (unsigned char)(pixel[2] * ra + inv_a); + } + } + } + } + + // convert to desired output format + if (req_comp && req_comp != 4) { + if (ri->bits_per_channel == 16) + out = (stbi_uc*)stbi__convert_format16((stbi__uint16*)out, 4, req_comp, w, h); + else + 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 = 4; + *y = h; + *x = w; + + return out; +} +#endif + +// ************************************************************************************************* +// 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/ + +#ifndef STBI_NO_PIC +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; y < height; ++y) { + int packet_idx; + + for (packet_idx = 0; packet_idx < num_packets; ++packet_idx) { + stbi__pic_packet* packet = &packets[packet_idx]; + stbi_uc* dest = result + y * width * 4; + + switch (packet->type) { + default: + return stbi__errpuc("bad format", "packet has bad compression type"); + + case 0: {//uncompressed + int x; + + for (x = 0; x < width; ++x, dest += 4) + if (!stbi__readval(s, packet->channel, dest)) + return 0; + break; + } + + case 1://Pure RLE + { + int left = width, i; + + while (left > 0) { + stbi_uc count, value[4]; + + count = 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; i < count; ++i, dest += 4) + stbi__copyval(packet->channel, 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]; + + 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; i < count; ++i, dest += 4) + stbi__copyval(packet->channel, dest, value); + } + else { // Raw + ++count; + if (count > left) return stbi__errpuc("bad file", "scanline overrun"); + + for (i = 0; i < count; ++i, dest += 4) + if (!stbi__readval(s, packet->channel, dest)) + return 0; + } + left -= count; + } + break; + } + } + } + } + + return result; +} + +static void* stbi__pic_load(stbi__context* s, int* px, int* py, int* comp, int req_comp, stbi__result_info* ri) +{ + stbi_uc* result; + int i, x, y, internal_comp; + STBI_NOTUSED(ri); + + if (!comp) comp = &internal_comp; + + 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 (!stbi__mad3sizes_valid(x, y, 4, 0)) return stbi__errpuc("too large", "PIC 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_mad3(x, y, 4, 0); + memset(result, 0xff, x * y * 4); + + if (!stbi__pic_load_core(s, x, y, comp, result)) { + STBI_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; +} +#endif + +// ************************************************************************************************* +// GIF loader -- public domain by Jean-Marc Lienher -- simplified/shrunk by stb + +#ifndef STBI_NO_GIF +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) + stbi_uc* background; // The current "background" as far as a gif is concerned + stbi_uc* history; + int flags, bgindex, ratio, transparent, eflags; + stbi_uc pal[256][4]; + stbi_uc lpal[256][4]; + stbi__gif_lzw codes[8192]; + 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; + int delay; +} 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 == i ? 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 = (stbi__gif*)stbi__malloc(sizeof(stbi__gif)); + if (!stbi__gif_header(s, g, comp, 1)) { + STBI_FREE(g); + stbi__rewind(s); + return 0; + } + if (x)* x = g->w; + if (y)* y = g->h; + STBI_FREE(g); + return 1; +} + +static void stbi__out_gif_code(stbi__gif* g, stbi__uint16 code) +{ + stbi_uc* p, * c; + int idx; + + // 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; + + idx = g->cur_x + g->cur_y; + p = &g->out[idx]; + g->history[idx / 4] = 1; + + c = &g->color_table[g->codes[code].suffix * 4]; + if (c[3] > 128) { // don't render transparent pixels; + 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, init_code; + stbi__uint32 first; + stbi__int32 codesize, codemask, avail, oldcode, bits, valid_bits, clear; + stbi__gif_lzw* p; + + lzw_cs = stbi__get8(s); + if (lzw_cs > 12) return NULL; + clear = 1 << lzw_cs; + first = 1; + codesize = lzw_cs + 1; + codemask = (1 << codesize) - 1; + bits = 0; + valid_bits = 0; + for (init_code = 0; init_code < clear; init_code++) { + g->codes[init_code].prefix = -1; + g->codes[init_code].first = (stbi_uc)init_code; + g->codes[init_code].suffix = (stbi_uc)init_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 > 8192) { + 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"); + } + } + } +} + +// this function is designed to support animated gifs, although stb_image doesn't support it +// two back is the image from two frames ago, used for a very specific disposal format +static stbi_uc* stbi__gif_load_next(stbi__context* s, stbi__gif* g, int* comp, int req_comp, stbi_uc* two_back) +{ + int dispose; + int first_frame; + int pi; + int pcount; + + // on first frame, any non-written pixels get the background colour (non-transparent) + first_frame = 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); + g->background = (stbi_uc*)stbi__malloc(4 * g->w * g->h); + g->history = (stbi_uc*)stbi__malloc(g->w * g->h); + if (g->out == 0) return stbi__errpuc("outofmem", "Out of memory"); + + // image is treated as "tranparent" at the start - ie, nothing overwrites the current background; + // background colour is only used for pixels that are not rendered first frame, after that "background" + // color refers to teh color that was there the previous frame. + memset(g->out, 0x00, 4 * g->w * g->h); + memset(g->background, 0x00, 4 * g->w * g->h); // state of the background (starts transparent) + memset(g->history, 0x00, g->w * g->h); // pixels that were affected previous frame + first_frame = 1; + } + else { + // second frame - how do we dispoase of the previous one? + dispose = (g->eflags & 0x1C) >> 2; + pcount = g->w * g->h; + + if ((dispose == 3) && (two_back == 0)) { + dispose = 2; // if I don't have an image to revert back to, default to the old background + } + + if (dispose == 3) { // use previous graphic + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { + memcpy(&g->out[pi * 4], &two_back[pi * 4], 4); + } + } + } + else if (dispose == 2) { + // restore what was changed last frame to background before that frame; + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi]) { + memcpy(&g->out[pi * 4], &g->background[pi * 4], 4); + } + } + } + else { + // This is a non-disposal case eithe way, so just + // leave the pixels as is, and they will become the new background + // 1: do not dispose + // 0: not specified. + } + + // background is what out is after the undoing of the previou frame; + memcpy(g->background, g->out, 4 * g->w * g->h); + } + + // clear my history; + memset(g->history, 0x00, g->w * g->h); // pixels that were affected previous frame + + for (;;) { + int tag = stbi__get8(s); + switch (tag) { + 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) { + 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 this was the first frame, + pcount = g->w * g->h; + if (first_frame && (g->bgindex > 0)) { + // if first frame, any pixel not drawn to gets the background color + for (pi = 0; pi < pcount; ++pi) { + if (g->history[pi] == 0) { + g->pal[g->bgindex][3] = 255; // just in case it was made transparent, undo that; It will be reset next frame if need be; + memcpy(&g->out[pi * 4], &g->pal[g->bgindex], 4); + } + } + } + + return o; + } + + case 0x21: // Comment Extension. + { + int len; + int ext = stbi__get8(s); + if (ext == 0xF9) { // Graphic Control Extension. + len = stbi__get8(s); + if (len == 4) { + g->eflags = stbi__get8(s); + g->delay = 10 * stbi__get16le(s); // delay - 1/100th of a second, saving as 1/1000ths. + + // unset old transparent + if (g->transparent >= 0) { + g->pal[g->transparent][3] = 255; + } + if (g->eflags & 0x01) { + g->transparent = stbi__get8(s); + if (g->transparent >= 0) { + g->pal[g->transparent][3] = 0; + } + } + else { + // don't need transparent + stbi__skip(s, 1); + g->transparent = -1; + } + } + 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 void* stbi__load_gif_main(stbi__context* s, int** delays, int* x, int* y, int* z, int* comp, int req_comp) +{ + if (stbi__gif_test(s)) { + int layers = 0; + stbi_uc* u = 0; + stbi_uc* out = 0; + stbi_uc* two_back = 0; + stbi__gif g; + int stride; + memset(&g, 0, sizeof(g)); + if (delays) { + *delays = 0; + } + + do { + u = stbi__gif_load_next(s, &g, comp, req_comp, two_back); + if (u == (stbi_uc*)s) u = 0; // end of animated gif marker + + if (u) { + *x = g.w; + *y = g.h; + ++layers; + stride = g.w * g.h * 4; + + if (out) { + out = (stbi_uc*)STBI_REALLOC(out, layers * stride); + if (delays) { + *delays = (int*)STBI_REALLOC(*delays, sizeof(int) * layers); + } + } + else { + out = (stbi_uc*)stbi__malloc(layers * stride); + if (delays) { + *delays = (int*)stbi__malloc(layers * sizeof(int)); + } + } + memcpy(out + ((layers - 1) * stride), u, stride); + if (layers >= 2) { + two_back = out - 2 * stride; + } + + if (delays) { + (*delays)[layers - 1U] = g.delay; + } + } + } while (u != 0); + + // free temp buffer; + STBI_FREE(g.out); + STBI_FREE(g.history); + STBI_FREE(g.background); + + // do the final conversion after loading everything; + if (req_comp && req_comp != 4) + out = stbi__convert_format(out, 4, req_comp, layers * g.w, g.h); + + *z = layers; + return out; + } + else { + return stbi__errpuc("not GIF", "Image was not as a gif type."); + } +} + +static void* stbi__gif_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri) +{ + stbi_uc* u = 0; + stbi__gif g; + memset(&g, 0, sizeof(g)); + + u = stbi__gif_load_next(s, &g, comp, req_comp, 0); + if (u == (stbi_uc*)s) u = 0; // end of animated gif marker + if (u) { + *x = g.w; + *y = g.h; + + // moved conversion to after successful load so that the same + // can be done for multiple frames. + if (req_comp && req_comp != 4) + u = stbi__convert_format(u, 4, req_comp, g.w, g.h); + } + + // free buffers needed for multiple frame loading; + STBI_FREE(g.history); + STBI_FREE(g.background); + + 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); +} +#endif + +// ************************************************************************************************* +// Radiance RGBE HDR loader +// originally by Nicolas Schulz +#ifndef STBI_NO_HDR +static int stbi__hdr_test_core(stbi__context* s, const char* signature) +{ + int i; + for (i = 0; signature[i]; ++i) + if (stbi__get8(s) != signature[i]) + return 0; + stbi__rewind(s); + return 1; +} + +static int stbi__hdr_test(stbi__context* s) +{ + int r = stbi__hdr_test_core(s, "#?RADIANCE\n"); + stbi__rewind(s); + if (!r) { + r = stbi__hdr_test_core(s, "#?RGBE\n"); + 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, stbi__result_info* ri) +{ + 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; + const char* headerToken; + STBI_NOTUSED(ri); + + // Check identifier + headerToken = stbi__hdr_gettoken(s, buffer); + if (strcmp(headerToken, "#?RADIANCE") != 0 && strcmp(headerToken, "#?RGBE") != 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; + + if (!stbi__mad4sizes_valid(width, height, req_comp, sizeof(float), 0)) + return stbi__errpf("too large", "HDR image is too large"); + + // Read data + hdr_data = (float*)stbi__malloc_mad4(width, height, req_comp, sizeof(float), 0); + if (!hdr_data) + return stbi__errpf("outofmem", "Out of memory"); + + // 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; + STBI_FREE(scanline); + goto main_decode_loop; // yes, this makes no sense + } + len <<= 8; + len |= stbi__get8(s); + if (len != width) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("invalid decoded scanline length", "corrupt HDR"); } + if (scanline == NULL) { + scanline = (stbi_uc*)stbi__malloc_mad2(width, 4, 0); + if (!scanline) { + STBI_FREE(hdr_data); + return stbi__errpf("outofmem", "Out of memory"); + } + } + + for (k = 0; k < 4; ++k) { + int nleft; + i = 0; + while ((nleft = width - i) > 0) { + count = stbi__get8(s); + if (count > 128) { + // Run + value = stbi__get8(s); + count -= 128; + if (count > nleft) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } + for (z = 0; z < count; ++z) + scanline[i++ * 4 + k] = value; + } + else { + // Dump + if (count > nleft) { STBI_FREE(hdr_data); STBI_FREE(scanline); return stbi__errpf("corrupt", "bad RLE data in HDR"); } + 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); + } + if (scanline) + STBI_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; + int dummy; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + if (stbi__hdr_test(s) == 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 + +#ifndef STBI_NO_BMP +static int stbi__bmp_info(stbi__context* s, int* x, int* y, int* comp) +{ + void* p; + stbi__bmp_data info; + + info.all_a = 255; + p = stbi__bmp_parse_header(s, &info); + stbi__rewind(s); + if (p == NULL) + return 0; + if (x)* x = s->img_x; + if (y)* y = s->img_y; + if (comp)* comp = info.ma ? 4 : 3; + return 1; +} +#endif + +#ifndef STBI_NO_PSD +static int stbi__psd_info(stbi__context* s, int* x, int* y, int* comp) +{ + int channelCount, dummy, depth; + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + 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); + depth = stbi__get16be(s); + if (depth != 8 && depth != 16) { + stbi__rewind(s); + return 0; + } + if (stbi__get16be(s) != 3) { + stbi__rewind(s); + return 0; + } + *comp = 4; + return 1; +} + +static int stbi__psd_is16(stbi__context* s) +{ + int channelCount, depth; + 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; + } + (void)stbi__get32be(s); + (void)stbi__get32be(s); + depth = stbi__get16be(s); + if (depth != 16) { + stbi__rewind(s); + return 0; + } + return 1; +} +#endif + +#ifndef STBI_NO_PIC +static int stbi__pic_info(stbi__context* s, int* x, int* y, int* comp) +{ + int act_comp = 0, num_packets = 0, chained, dummy; + stbi__pic_packet packets[10]; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + if (!stbi__pic_is4(s, "\x53\x80\xF6\x34")) { + stbi__rewind(s); + return 0; + } + + stbi__skip(s, 88); + + *x = stbi__get16be(s); + *y = stbi__get16be(s); + if (stbi__at_eof(s)) { + stbi__rewind(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; +} +#endif + +// ************************************************************************************************* +// Portable Gray Map and Portable Pixel Map loader +// by Ken Miller +// +// PGM: http://netpbm.sourceforge.net/doc/pgm.html +// PPM: http://netpbm.sourceforge.net/doc/ppm.html +// +// Known limitations: +// Does not support comments in the header section +// Does not support ASCII image data (formats P2 and P3) +// Does not support 16-bit-per-channel + +#ifndef STBI_NO_PNM + +static int stbi__pnm_test(stbi__context* s) +{ + char p, t; + p = (char)stbi__get8(s); + t = (char)stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind(s); + return 0; + } + return 1; +} + +static void* stbi__pnm_load(stbi__context* s, int* x, int* y, int* comp, int req_comp, stbi__result_info* ri) +{ + stbi_uc* out; + STBI_NOTUSED(ri); + + if (!stbi__pnm_info(s, (int*)& s->img_x, (int*)& s->img_y, (int*)& s->img_n)) + return 0; + + *x = s->img_x; + *y = s->img_y; + if (comp)* comp = s->img_n; + + if (!stbi__mad3sizes_valid(s->img_n, s->img_x, s->img_y, 0)) + return stbi__errpuc("too large", "PNM too large"); + + out = (stbi_uc*)stbi__malloc_mad3(s->img_n, s->img_x, s->img_y, 0); + if (!out) return stbi__errpuc("outofmem", "Out of memory"); + stbi__getn(s, out, s->img_n * s->img_x * s->img_y); + + if (req_comp && req_comp != s->img_n) { + out = stbi__convert_format(out, s->img_n, req_comp, s->img_x, s->img_y); + if (out == NULL) return out; // stbi__convert_format frees input on failure + } + return out; +} + +static int stbi__pnm_isspace(char c) +{ + return c == ' ' || c == '\t' || c == '\n' || c == '\v' || c == '\f' || c == '\r'; +} + +static void stbi__pnm_skip_whitespace(stbi__context* s, char* c) +{ + for (;;) { + while (!stbi__at_eof(s) && stbi__pnm_isspace(*c)) + * c = (char)stbi__get8(s); + + if (stbi__at_eof(s) || *c != '#') + break; + + while (!stbi__at_eof(s) && *c != '\n' && *c != '\r') + * c = (char)stbi__get8(s); + } +} + +static int stbi__pnm_isdigit(char c) +{ + return c >= '0' && c <= '9'; +} + +static int stbi__pnm_getinteger(stbi__context* s, char* c) +{ + int value = 0; + + while (!stbi__at_eof(s) && stbi__pnm_isdigit(*c)) { + value = value * 10 + (*c - '0'); + *c = (char)stbi__get8(s); + } + + return value; +} + +static int stbi__pnm_info(stbi__context* s, int* x, int* y, int* comp) +{ + int maxv, dummy; + char c, p, t; + + if (!x) x = &dummy; + if (!y) y = &dummy; + if (!comp) comp = &dummy; + + stbi__rewind(s); + + // Get identifier + p = (char)stbi__get8(s); + t = (char)stbi__get8(s); + if (p != 'P' || (t != '5' && t != '6')) { + stbi__rewind(s); + return 0; + } + + *comp = (t == '6') ? 3 : 1; // '5' is 1-component .pgm; '6' is 3-component .ppm + + c = (char)stbi__get8(s); + stbi__pnm_skip_whitespace(s, &c); + + *x = stbi__pnm_getinteger(s, &c); // read width + stbi__pnm_skip_whitespace(s, &c); + + *y = stbi__pnm_getinteger(s, &c); // read height + stbi__pnm_skip_whitespace(s, &c); + + maxv = stbi__pnm_getinteger(s, &c); // read max value + + if (maxv > 255) + return stbi__err("max value > 255", "PPM image not 8-bit"); + else + return 1; +} +#endif + +static int stbi__info_main(stbi__context* s, int* x, int* y, int* comp) +{ +#ifndef STBI_NO_JPEG + if (stbi__jpeg_info(s, x, y, comp)) return 1; +#endif + +#ifndef STBI_NO_PNG + if (stbi__png_info(s, x, y, comp)) return 1; +#endif + +#ifndef STBI_NO_GIF + if (stbi__gif_info(s, x, y, comp)) return 1; +#endif + +#ifndef STBI_NO_BMP + if (stbi__bmp_info(s, x, y, comp)) return 1; +#endif + +#ifndef STBI_NO_PSD + if (stbi__psd_info(s, x, y, comp)) return 1; +#endif + +#ifndef STBI_NO_PIC + if (stbi__pic_info(s, x, y, comp)) return 1; +#endif + +#ifndef STBI_NO_PNM + if (stbi__pnm_info(s, x, y, comp)) return 1; +#endif + +#ifndef STBI_NO_HDR + if (stbi__hdr_info(s, x, y, comp)) return 1; +#endif + + // test tga last because it's a crappy test! +#ifndef STBI_NO_TGA + if (stbi__tga_info(s, x, y, comp)) + return 1; +#endif + return stbi__err("unknown image type", "Image not of any known type, or corrupt"); +} + +static int stbi__is_16_main(stbi__context* s) +{ +#ifndef STBI_NO_PNG + if (stbi__png_is16(s)) return 1; +#endif + +#ifndef STBI_NO_PSD + if (stbi__psd_is16(s)) return 1; +#endif + + return 0; +} + +#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; +} + +STBIDEF int stbi_is_16_bit(char const* filename) +{ + FILE* f = stbi__fopen(filename, "rb"); + int result; + if (!f) return stbi__err("can't fopen", "Unable to open file"); + result = stbi_is_16_bit_from_file(f); + fclose(f); + return result; +} + +STBIDEF int stbi_is_16_bit_from_file(FILE* f) +{ + int r; + stbi__context s; + long pos = ftell(f); + stbi__start_file(&s, f); + r = stbi__is_16_main(&s); + 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); +} + +STBIDEF int stbi_is_16_bit_from_memory(stbi_uc const* buffer, int len) +{ + stbi__context s; + stbi__start_mem(&s, buffer, len); + return stbi__is_16_main(&s); +} + +STBIDEF int stbi_is_16_bit_from_callbacks(stbi_io_callbacks const* c, void* user) +{ + stbi__context s; + stbi__start_callbacks(&s, (stbi_io_callbacks*)c, user); + return stbi__is_16_main(&s); +} + +#endif // STB_IMAGE_IMPLEMENTATION + +/* + revision history: + 2.19 (2018-02-11) fix warning + 2.18 (2018-01-30) fix warnings + 2.17 (2018-01-29) change sbti__shiftsigned to avoid clang -O2 bug + 1-bit BMP + *_is_16_bit api + avoid warnings + 2.16 (2017-07-23) all functions have 16-bit variants; + STBI_NO_STDIO works again; + compilation fixes; + fix rounding in unpremultiply; + optimize vertical flip; + disable raw_len validation; + documentation fixes + 2.15 (2017-03-18) fix png-1,2,4 bug; now all Imagenet JPGs decode; + warning fixes; disable run-time SSE detection on gcc; + uniform handling of optional "return" values; + thread-safe initialization of zlib tables + 2.14 (2017-03-03) remove deprecated STBI_JPEG_OLD; fixes for Imagenet JPGs + 2.13 (2016-11-29) add 16-bit API, only supported for PNG right now + 2.12 (2016-04-02) fix typo in 2.11 PSD fix that caused crashes + 2.11 (2016-04-02) allocate large structures on the stack + remove white matting for transparent PSD + fix reported channel count for PNG & BMP + re-enable SSE2 in non-gcc 64-bit + support RGB-formatted JPEG + read 16-bit PNGs (only as 8-bit) + 2.10 (2016-01-22) avoid warning introduced in 2.09 by STBI_REALLOC_SIZED + 2.09 (2016-01-16) allow comments in PNM files + 16-bit-per-pixel TGA (not bit-per-component) + info() for TGA could break due to .hdr handling + info() for BMP to shares code instead of sloppy parse + can use STBI_REALLOC_SIZED if allocator doesn't support realloc + code cleanup + 2.08 (2015-09-13) fix to 2.07 cleanup, reading RGB PSD as RGBA + 2.07 (2015-09-13) fix compiler warnings + partial animated GIF support + limited 16-bpc PSD support + #ifdef unused functions + bug with < 92 byte PIC,PNM,HDR,TGA + 2.06 (2015-04-19) fix bug where PSD returns wrong '*comp' value + 2.05 (2015-04-19) fix bug in progressive JPEG handling, fix warning + 2.04 (2015-04-15) try to re-enable SIMD on MinGW 64-bit + 2.03 (2015-04-12) extra corruption checking (mmozeiko) + stbi_set_flip_vertically_on_load (nguillemot) + fix NEON support; fix mingw support + 2.02 (2015-01-19) fix incorrect assert, fix warning + 2.01 (2015-01-17) fix various warnings; suppress SIMD on gcc 32-bit without -msse2 + 2.00b (2014-12-25) fix STBI_MALLOC in progressive JPEG + 2.00 (2014-12-25) optimize JPG, including x86 SSE2 & NEON SIMD (ryg) + progressive JPEG (stb) + PGM/PPM support (Ken Miller) + STBI_MALLOC,STBI_REALLOC,STBI_FREE + GIF bugfix -- seemingly never worked + STBI_NO_*, STBI_ONLY_* + 1.48 (2014-12-14) fix incorrectly-named assert() + 1.47 (2014-12-14) 1/2/4-bit PNG support, both direct and paletted (Omar Cornut & stb) + optimize PNG (ryg) + fix bug in interlaced PNG with user-specified channel count (stb) + 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 (2006-11-19) + first released version +*/ + + +/* +------------------------------------------------------------------------------ +This software is available under 2 licenses -- choose whichever you prefer. +------------------------------------------------------------------------------ +ALTERNATIVE A - MIT License +Copyright (c) 2017 Sean Barrett +Permission is hereby granted, free of charge, to any person obtaining a copy of +this software and associated documentation files (the "Software"), to deal in +the Software without restriction, including without limitation the rights to +use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies +of the Software, and to permit persons to whom the Software is furnished to do +so, subject to the following conditions: +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. +------------------------------------------------------------------------------ +ALTERNATIVE B - Public Domain (www.unlicense.org) +This is free and unencumbered software released into the public domain. +Anyone is free to copy, modify, publish, use, compile, sell, or distribute this +software, either in source code form or as a compiled binary, for any purpose, +commercial or non-commercial, and by any means. +In jurisdictions that recognize copyright laws, the author or authors of this +software dedicate any and all copyright interest in the software to the public +domain. We make this dedication for the benefit of the public at large and to +the detriment of our heirs and successors. We intend this dedication to be an +overt act of relinquishment in perpetuity of all present and future rights to +this software under copyright law. +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN +ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION +WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. +------------------------------------------------------------------------------ +*/ diff --git a/src/display.cpp b/src/display.cpp new file mode 100644 index 0000000..79c680e --- /dev/null +++ b/src/display.cpp @@ -0,0 +1,120 @@ +#include "display.hpp" + +Display::Display( int w, int h, std::string title ) + : mLogger() +{ + + mLogger << LOGGER_INFO << "Initializing display" << LOGGER_ENDL; + SDL_Init( SDL_INIT_VIDEO | SDL_INIT_AUDIO ); + + SDL_GL_SetAttribute( SDL_GL_RED_SIZE, 8 ); + SDL_GL_SetAttribute( SDL_GL_GREEN_SIZE, 8 ); + SDL_GL_SetAttribute( SDL_GL_BLUE_SIZE, 8 ); + SDL_GL_SetAttribute( SDL_GL_ALPHA_SIZE, 8 ); + SDL_GL_SetAttribute( SDL_GL_BUFFER_SIZE, 32 ); + SDL_GL_SetAttribute( SDL_GL_DOUBLEBUFFER, 1 ); + + SDL_GL_SetAttribute( SDL_GL_MULTISAMPLEBUFFERS, 1 ); + SDL_GL_SetAttribute( SDL_GL_MULTISAMPLESAMPLES, 4 ); + + SDL_GL_SetAttribute( SDL_GL_CONTEXT_MAJOR_VERSION, 4 ); + SDL_GL_SetAttribute( SDL_GL_CONTEXT_MINOR_VERSION, 5 ); + + // Create GL window + mLogger << LOGGER_INFO << "Creating window" << LOGGER_ENDL; + mWindow = SDL_CreateWindow( title.c_str(), + SDL_WINDOWPOS_CENTERED, + SDL_WINDOWPOS_CENTERED, w, h, + SDL_WINDOW_OPENGL|SDL_WINDOW_RESIZABLE ); + + // Create GL context + mLogger << LOGGER_INFO << "Creating OpenGL context" << LOGGER_ENDL; + mGlContext = SDL_GL_CreateContext( mWindow ); + + SDL_SetRelativeMouseMode( SDL_TRUE ); + + // Set VSYNC swap interval + SDL_GL_SetSwapInterval( 1 ); + + mLogger << LOGGER_INFO << "Display set up" << LOGGER_ENDL; + + // Load OpenGL + gladLoadGLLoader( SDL_GL_GetProcAddress ); + glEnable( GL_MULTISAMPLE ); + // glEnable(GL_CULL_FACE); + glCullFace( GL_BACK ); + glEnable( GL_DEPTH_TEST ); + // glPolygonMode(GL_FRONT_AND_BACK, GL_LINE); + + mLogger << LOGGER_INFO << "Loaded OpenGL" << LOGGER_ENDL; + mLogger << LOGGER_ENDL; + + IsWindowOpen = true; + +} + +void Display::Input( SDL_Event* e ) +{ + + Uint8* state = (Uint8*) SDL_GetKeyboardState( NULL ); + + while ( SDL_PollEvent( e ) ) + { + switch ( e->type ) + { + + case SDL_KEYDOWN: + { + if ( e->key.keysym.sym == SDLK_ESCAPE ) + { + IsMouseActive = !IsMouseActive; + + if ( IsMouseActive ) + SDL_SetRelativeMouseMode( SDL_TRUE ); + else + SDL_SetRelativeMouseMode( SDL_FALSE ); + } + + break; + } + + case SDL_WINDOWEVENT: + { + if ( e->window.event == SDL_WINDOWEVENT_RESIZED ) + { + mW = e->window.data1; mH = e->window.data2; + // CameraUpdateProjection( mW, mH ); + glViewport( 0, 0, mW, mH ); + } + + break; + } + + case SDL_QUIT: + { + IsWindowOpen = false; + break; + } + + } + + // if ( IsMouseActive ) HandleMouseSDL( *e ); + } + + // m_player->MoveSDL( state ); + +} + +void Display::PrepareFrame() +{ + static const float clear[] = { 186.0f / 255.0f, 214.0f / 255.0f, 254.0f / 255.0f }; + + glClear( GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT ); + glClearBufferfv( GL_COLOR, 0, clear ); +} + +void Display::NextFrame() +{ + SDL_GL_SwapWindow( mWindow ); +} + diff --git a/src/display.hpp b/src/display.hpp new file mode 100644 index 0000000..487be4b --- /dev/null +++ b/src/display.hpp @@ -0,0 +1,42 @@ +#ifndef MINECRAFT_DISPLAY_H_ +#define MINECRAFT_DISPLAY_H_ + +#include + +#include + +#if _WIN32 +#include +#else +#include +#endif + +#include +#include + +class Display +{ +public: + + Display( int w, int h, std::string title ); + + void Input( SDL_Event* e ); + + void PrepareFrame(); + void NextFrame(); + + bool IsWindowOpen = false; + bool IsMouseActive = true; + +private: + + Logger mLogger; + + SDL_Window* mWindow = nullptr; + SDL_GLContext mGlContext = nullptr; + + int mW, mH; + +}; + +#endif diff --git a/src/settings.hpp b/src/settings.hpp new file mode 100644 index 0000000..5ab11c0 --- /dev/null +++ b/src/settings.hpp @@ -0,0 +1,14 @@ +#ifndef MINECRAFT_SETTINGS_H_ +#define MINECRAFT_SETTINGS_H_ + +#include + +// TODO: import settings and stuff +// for now this works + +static const int WindowWidth = 1000; +static const int WindowHeight = 600; + +static const std::string ResourceBase = MC_RESOURCES; + +#endif diff --git a/src/threadpool.hpp b/src/threadpool.hpp new file mode 100644 index 0000000..764544e --- /dev/null +++ b/src/threadpool.hpp @@ -0,0 +1,3 @@ +// Threadpool for asset management and other such tasks + + diff --git a/src/utilities.hpp b/src/utilities.hpp new file mode 100644 index 0000000..21e44ed --- /dev/null +++ b/src/utilities.hpp @@ -0,0 +1,10 @@ +#include +#include + +inline std::string LoadTextFromFile( std::string file ) +{ + std::ifstream t( file ); + std::string text( (std::istreambuf_iterator( t )), + std::istreambuf_iterator() ); + return text; +}