Initial Update

diff --git a/Makefile b/Makefile
new file mode 100644 (file)
index 0000000..8b95cc6
--- /dev/null
+++ b/Makefile
@@ -0,0 +1,19 @@
+OBJS = main.o stb_image.o helpers.o tiny_obj_loader.o baseclass.o old_objects.o 
+CXXFLAGS = -g -Wall 
+LDFLAGS = -lGL -lglfw -lGLEW -pthread -g 
+
+# For FreeBSD (doesn't seem to hurt anything on Linux)
+CXXFLAGS += -I /usr/local/include
+LDFLAGS += -L /usr/local/lib
+
+all: ${OBJS}
+       ${CXX} ${OBJS}  ${LDFLAGS}
+
+clean:
+       rm a.out ${OBJS}
+
+test:  all
+       ./a.out
+
+${OBJS} : game.h
+main.o : projectiles.h activation_area.h turret.h tile_floor.h old_objects.h moving_platform.h

diff --git a/activation_area.h b/activation_area.h
new file mode 100644 (file)
index 0000000..d059527
--- /dev/null
+++ b/activation_area.h
@@ -0,0 +1,26 @@
+#pragma once
+
+class activation_area : public gameobject {
+       public:
+               std::vector<void (*)()> callbacks;
+               activation_area() {
+                       collision_check = false;
+               }
+               void add_area(glm::vec3 location, void (*callback_function)()){
+                       locations.push_back(location);
+                       callbacks.push_back(callback_function);
+               }
+               ssize_t collision_index(glm::vec3 position, float distance = 0){
+                       for(size_t i = 0; i < locations.size(); i++){
+                               glm::vec3 l = locations[i]; // This'll get optimized out
+                               if(     size.x/2.0f + distance > abs(l.x-position.x) && 
+                                               size.y/2.0f + distance > abs(l.y-position.y) && 
+                                               size.z/2.0f + distance > abs(l.z-position.z)){
+                                       callbacks[i]();
+                                       return i;
+                               }
+                       }
+                       return -1;
+               }
+};
+

diff --git a/baseclass.cpp b/baseclass.cpp
new file mode 100644 (file)
index 0000000..dd43f6d
--- /dev/null
+++ b/baseclass.cpp
@@ -0,0 +1,130 @@
+#include "game.h"
+
+int loaded_object::init() {
+       // Initialization part
+       std::vector<vertex> vertices;
+       std::vector<uint32_t> indices; // Consider unified terminology
+
+       load_model(vertices, indices, objectfile, scale, swap_yz, texture_scale);
+
+       glGenBuffers(1, &vbuf);
+       glBindBuffer(GL_SHADER_STORAGE_BUFFER, vbuf);
+       glBufferData(GL_SHADER_STORAGE_BUFFER, sizeof(vertex) * vertices.size(), vertices.data(), GL_STATIC_DRAW);
+       // TODO:  Remember to explain the layout later
+
+       glGenBuffers(1, &ebuf);
+       glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, ebuf);
+       glBufferData(GL_ELEMENT_ARRAY_BUFFER, sizeof(uint32_t) * indices.size(), indices.data(), GL_STATIC_DRAW);
+
+       glGenBuffers(1, &models_buffer);
+
+       tex = load_texture(texturefile);
+
+       program = make_program("loaded_object_vertex_shader.glsl",0, 0, 0, "loaded_object_fragment_shader.glsl");
+       if (!program)
+               return 1;
+
+       v_attrib = glGetAttribLocation(program, "in_vertex");
+       t_attrib = glGetAttribLocation(program, "in_texcoord");
+       mvp_uniform = glGetUniformLocation(program, "vp");
+       return 0;
+}
+
+void loaded_object::draw(glm::mat4 vp) {
+       glUseProgram(program);
+       std::vector<glm::mat4> models;
+       data_mutex.lock();
+       models.reserve(locations.size());
+       for(glm::vec3 l : locations){
+               glm::mat4 new_model = glm::mat4(1.0f);
+               new_model = translate(new_model, l);
+               models.push_back(new_model);
+       }
+       data_mutex.unlock();
+       glBindBuffer(GL_SHADER_STORAGE_BUFFER, models_buffer);
+       glBufferData(GL_SHADER_STORAGE_BUFFER, models.size() * sizeof(glm::mat4), models.data(), GL_STATIC_DRAW);
+       glBindBufferBase(GL_SHADER_STORAGE_BUFFER, 0, models_buffer);
+
+       glEnableVertexAttribArray(v_attrib);
+       glBindBuffer(GL_ARRAY_BUFFER, vbuf);
+       glVertexAttribPointer(v_attrib, 3, GL_FLOAT, GL_FALSE, 20, 0);
+
+       glEnableVertexAttribArray(t_attrib);
+       glVertexAttribPointer(t_attrib, 2, GL_FLOAT, GL_FALSE, 20, (const void*)12);
+
+       glActiveTexture(GL_TEXTURE0);
+       glBindTexture(GL_TEXTURE_2D, tex);
+
+       int size;
+       glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, ebuf);
+       glGetBufferParameteriv(GL_ELEMENT_ARRAY_BUFFER, GL_BUFFER_SIZE, &size);
+
+       glUniformMatrix4fv(mvp_uniform, 1, 0, glm::value_ptr(vp));
+
+       glDrawElementsInstanced(GL_TRIANGLES, size / sizeof(GLuint), GL_UNSIGNED_INT, 0, models.size());
+}
+
+bool block_object::is_on_idx(glm::vec3 position, size_t index, float height){
+       return (0.0f < (position.y - locations[index].y) && 
+                       1.0f > (position.y - height) - (locations[index].y + size.y/2) &&
+                       size.x/2 > fabs(position.x - locations[index].x) && 
+                       size.z/2 > fabs(position.z - locations[index].z));
+
+}
+
+ssize_t block_object::is_on(glm::vec3 position, float height) {
+       for(size_t i = 0; i < locations.size(); i++)
+               if(is_on_idx(position, i, height))
+                       return i;
+       return -1;
+}
+
+ssize_t block_object::collision_index(glm::vec3 position, float distance){
+       for(size_t i = 0; i < locations.size(); i++){
+               glm::vec3 l = locations[i]; // This'll get optimized out
+               if(     size.x/2.0f + distance > abs(l.x-position.x) && 
+                               size.y/2.0f + distance > abs(l.y-position.y) && 
+                               size.z/2.0f + distance > abs(l.z-position.z)){
+                       return i;       
+               }
+       }
+       return -1;
+}
+
+/* Leaving y for later, so we finish today */
+glm::vec3 block_object::collision_normal(glm::vec3 move_to, glm::vec3 old_position, ssize_t index, float distance){
+       glm::vec3 l = locations[index]; // This'll get optimized out
+       if(     old_position.z > l.z + (size.z/2 + distance) &&
+                       old_position.x >= l.x - (size.x/2 + distance) &&
+                       old_position.x <= l.x + (size.x/2 + distance)){
+               return glm::vec3(0, 0, 1);
+       }
+       if(     old_position.z < l.z - (size.z/2 + distance) &&
+                       old_position.x >= l.x - (size.x/2 + distance) &&
+                       old_position.x <= l.x + (size.x/2 + distance)){
+               return glm::vec3(0, 0, -1);
+       }
+       if(     old_position.x < l.x - (size.x/2 + distance) &&
+                       old_position.z >= l.z - (size.z/2 + distance) &&
+                       old_position.z <= l.z + (size.z/2 + distance)){
+               return glm::vec3(1, 0, 0);
+       }
+       if(     old_position.x > l.x + (size.x/2 + distance) &&
+                       old_position.z >= l.z - (size.z/2 + distance) &&
+                       old_position.z <= l.z + (size.z/2 + distance)){
+               return glm::vec3(-1, 0, 0);
+       }
+       puts("Ended collision normal without returning");
+       return glm::vec3(0, 0, 0);
+}
+
+bool block_object::collision_with_index(glm::vec3 position, size_t index, float distance){     
+       glm::vec3 l = locations[index]; // This variable will get optimized out
+       if(size.x/2.0f + distance > abs(l.x-position.x) && 
+                       size.y/2.0f + distance > abs(l.y-position.y) &&
+                       size.z/2.0f + distance > abs(l.z-position.z)){
+               return true;
+       }
+       return false;
+}
+

diff --git a/brick.jpg b/brick.jpg
new file mode 100644 (file)
index 0000000..c2a2f12
Binary files /dev/null and b/brick.jpg differ

diff --git a/cone.obj b/cone.obj
new file mode 100644 (file)
index 0000000..0a49558
--- /dev/null
+++ b/cone.obj
@@ -0,0 +1,170 @@
+# Blender v3.0.1 OBJ File: ''
+# www.blender.org
+mtllib cone.mtl
+o Cone
+v 0.000000 -1.000000 -1.000000
+v 0.195090 -1.000000 -0.980785
+v 0.382683 -1.000000 -0.923880
+v 0.555570 -1.000000 -0.831470
+v 0.707107 -1.000000 -0.707107
+v 0.831470 -1.000000 -0.555570
+v 0.923880 -1.000000 -0.382683
+v 0.980785 -1.000000 -0.195090
+v 1.000000 -1.000000 0.000000
+v 0.980785 -1.000000 0.195090
+v 0.923880 -1.000000 0.382683
+v 0.831470 -1.000000 0.555570
+v 0.707107 -1.000000 0.707107
+v 0.555570 -1.000000 0.831470
+v 0.382683 -1.000000 0.923880
+v 0.195090 -1.000000 0.980785
+v -0.000000 -1.000000 1.000000
+v -0.195090 -1.000000 0.980785
+v -0.382683 -1.000000 0.923880
+v -0.555570 -1.000000 0.831470
+v -0.707107 -1.000000 0.707107
+v -0.831469 -1.000000 0.555570
+v -0.923880 -1.000000 0.382684
+v -0.980785 -1.000000 0.195090
+v -1.000000 -1.000000 -0.000000
+v -0.980785 -1.000000 -0.195090
+v -0.923879 -1.000000 -0.382684
+v -0.831470 -1.000000 -0.555570
+v -0.707107 -1.000000 -0.707107
+v -0.555570 -1.000000 -0.831470
+v -0.382683 -1.000000 -0.923880
+v -0.195090 -1.000000 -0.980785
+v 0.000000 1.000000 0.000000
+vt 0.250000 0.490000
+vt 0.250000 0.250000
+vt 0.296822 0.485388
+vt 0.341844 0.471731
+vt 0.383337 0.449553
+vt 0.419706 0.419706
+vt 0.449553 0.383337
+vt 0.471731 0.341844
+vt 0.485388 0.296822
+vt 0.490000 0.250000
+vt 0.485388 0.203178
+vt 0.471731 0.158156
+vt 0.449553 0.116663
+vt 0.419706 0.080294
+vt 0.383337 0.050447
+vt 0.341844 0.028269
+vt 0.296822 0.014612
+vt 0.250000 0.010000
+vt 0.203178 0.014612
+vt 0.158156 0.028269
+vt 0.116663 0.050447
+vt 0.080294 0.080294
+vt 0.050447 0.116663
+vt 0.028269 0.158156
+vt 0.014612 0.203178
+vt 0.010000 0.250000
+vt 0.014612 0.296822
+vt 0.028269 0.341844
+vt 0.050447 0.383337
+vt 0.080294 0.419706
+vt 0.116663 0.449553
+vt 0.158156 0.471731
+vt 0.750000 0.490000
+vt 0.796822 0.485388
+vt 0.841844 0.471731
+vt 0.883337 0.449553
+vt 0.919706 0.419706
+vt 0.949553 0.383337
+vt 0.971731 0.341844
+vt 0.985388 0.296822
+vt 0.990000 0.250000
+vt 0.985388 0.203178
+vt 0.971731 0.158156
+vt 0.949553 0.116663
+vt 0.919706 0.080294
+vt 0.883337 0.050447
+vt 0.841844 0.028269
+vt 0.796822 0.014612
+vt 0.750000 0.010000
+vt 0.703178 0.014612
+vt 0.658156 0.028269
+vt 0.616663 0.050447
+vt 0.580294 0.080294
+vt 0.550447 0.116663
+vt 0.528269 0.158156
+vt 0.514612 0.203178
+vt 0.510000 0.250000
+vt 0.514612 0.296822
+vt 0.528269 0.341844
+vt 0.550447 0.383337
+vt 0.580294 0.419706
+vt 0.616663 0.449553
+vt 0.658156 0.471731
+vt 0.703178 0.485388
+vt 0.203178 0.485388
+vn 0.0878 0.4455 -0.8910
+vn 0.2599 0.4455 -0.8567
+vn 0.4220 0.4455 -0.7896
+vn 0.5680 0.4455 -0.6921
+vn 0.6921 0.4455 -0.5680
+vn 0.7896 0.4455 -0.4220
+vn 0.8567 0.4455 -0.2599
+vn 0.8910 0.4455 -0.0878
+vn 0.8910 0.4455 0.0878
+vn 0.8567 0.4455 0.2599
+vn 0.7896 0.4455 0.4220
+vn 0.6921 0.4455 0.5680
+vn 0.5680 0.4455 0.6921
+vn 0.4220 0.4455 0.7896
+vn 0.2599 0.4455 0.8567
+vn 0.0878 0.4455 0.8910
+vn -0.0878 0.4455 0.8910
+vn -0.2599 0.4455 0.8567
+vn -0.4220 0.4455 0.7896
+vn -0.5680 0.4455 0.6921
+vn -0.6921 0.4455 0.5680
+vn -0.7896 0.4455 0.4220
+vn -0.8567 0.4455 0.2599
+vn -0.8910 0.4455 0.0878
+vn -0.8910 0.4455 -0.0878
+vn -0.8567 0.4455 -0.2599
+vn -0.7896 0.4455 -0.4220
+vn -0.6921 0.4455 -0.5680
+vn -0.5680 0.4455 -0.6921
+vn -0.4220 0.4455 -0.7896
+vn 0.0000 -1.0000 0.0000
+vn -0.2599 0.4455 -0.8567
+vn -0.0878 0.4455 -0.8910
+usemtl None
+s off
+f 1/1/1 33/2/1 2/3/1
+f 2/3/2 33/2/2 3/4/2
+f 3/4/3 33/2/3 4/5/3
+f 4/5/4 33/2/4 5/6/4
+f 5/6/5 33/2/5 6/7/5
+f 6/7/6 33/2/6 7/8/6
+f 7/8/7 33/2/7 8/9/7
+f 8/9/8 33/2/8 9/10/8
+f 9/10/9 33/2/9 10/11/9
+f 10/11/10 33/2/10 11/12/10
+f 11/12/11 33/2/11 12/13/11
+f 12/13/12 33/2/12 13/14/12
+f 13/14/13 33/2/13 14/15/13
+f 14/15/14 33/2/14 15/16/14
+f 15/16/15 33/2/15 16/17/15
+f 16/17/16 33/2/16 17/18/16
+f 17/18/17 33/2/17 18/19/17
+f 18/19/18 33/2/18 19/20/18
+f 19/20/19 33/2/19 20/21/19
+f 20/21/20 33/2/20 21/22/20
+f 21/22/21 33/2/21 22/23/21
+f 22/23/22 33/2/22 23/24/22
+f 23/24/23 33/2/23 24/25/23
+f 24/25/24 33/2/24 25/26/24
+f 25/26/25 33/2/25 26/27/25
+f 26/27/26 33/2/26 27/28/26
+f 27/28/27 33/2/27 28/29/27
+f 28/29/28 33/2/28 29/30/28
+f 29/30/29 33/2/29 30/31/29
+f 30/31/30 33/2/30 31/32/30
+f 1/33/31 2/34/31 3/35/31 4/36/31 5/37/31 6/38/31 7/39/31 8/40/31 9/41/31 10/42/31 11/43/31 12/44/31 13/45/31 14/46/31 15/47/31 16/48/31 17/49/31 18/50/31 19/51/31 20/52/31 21/53/31 22/54/31 23/55/31 24/56/31 25/57/31 26/58/31 27/59/31 28/60/31 29/61/31 30/62/31 31/63/31 32/64/31
+f 31/32/32 33/2/32 32/65/32
+f 32/65/33 33/2/33 1/1/33

diff --git a/cube.obj b/cube.obj
new file mode 100644 (file)
index 0000000..6b83915
--- /dev/null
+++ b/cube.obj
@@ -0,0 +1,30 @@
+# Blender v3.0.1 OBJ File: ''
+# www.blender.org
+mtllib cube.mtl
+o Cube
+v 1.000000 1.000000 -1.000000
+v 1.000000 -1.000000 -1.000000
+v 1.000000 1.000000 1.000000
+v 1.000000 -1.000000 1.000000
+v -1.000000 1.000000 -1.000000
+v -1.000000 -1.000000 -1.000000
+v -1.000000 1.000000 1.000000
+v -1.000000 -1.000000 1.000000
+vt 0.0 1.0 
+vt 0.0 0.0 
+vt 1.0 1.0 
+vt 1.0 0.0 
+vn 0.0000 1.0000 0.0000
+vn 0.0000 0.0000 1.0000
+vn -1.0000 0.0000 0.0000
+vn 0.0000 -1.0000 0.0000
+vn 1.0000 0.0000 0.0000
+vn 0.0000 0.0000 -1.0000
+usemtl Material
+s off
+f 1/2/1 5/1/1 7/3/1 3/4/1
+f 4/2/2 3/1/2 7/3/2 8/4/2
+f 8/2/3 7/1/3 5/3/3 6/4/3
+f 6/2/4 2/1/4 4/3/4 8/4/4
+f 2/2/5 1/1/5 3/3/5 4/4/5
+f 6/2/6 5/1/6 1/3/6 2/4/6

diff --git a/cube_backup.obj b/cube_backup.obj
new file mode 100644 (file)
index 0000000..0c58ea6
--- /dev/null
+++ b/cube_backup.obj
@@ -0,0 +1,40 @@
+# Blender v3.0.1 OBJ File: ''
+# www.blender.org
+mtllib cube.mtl
+o Cube
+v 1.000000 1.000000 -1.000000
+v 1.000000 -1.000000 -1.000000
+v 1.000000 1.000000 1.000000
+v 1.000000 -1.000000 1.000000
+v -1.000000 1.000000 -1.000000
+v -1.000000 -1.000000 -1.000000
+v -1.000000 1.000000 1.000000
+v -1.000000 -1.000000 1.000000
+vt 0.625000 0.500000
+vt 0.875000 0.500000
+vt 0.875000 0.750000
+vt 0.625000 0.750000
+vt 0.375000 0.750000
+vt 0.625000 1.000000
+vt 0.375000 1.000000
+vt 0.375000 0.000000
+vt 0.625000 0.000000
+vt 0.625000 0.250000
+vt 0.375000 0.250000
+vt 0.125000 0.500000
+vt 0.375000 0.500000
+vt 0.125000 0.750000
+vn 0.0000 1.0000 0.0000
+vn 0.0000 0.0000 1.0000
+vn -1.0000 0.0000 0.0000
+vn 0.0000 -1.0000 0.0000
+vn 1.0000 0.0000 0.0000
+vn 0.0000 0.0000 -1.0000
+usemtl Material
+s off
+f 1/1/1 5/2/1 7/3/1 3/4/1
+f 4/5/2 3/4/2 7/6/2 8/7/2
+f 8/8/3 7/9/3 5/10/3 6/11/3
+f 6/12/4 2/13/4 4/5/4 8/14/4
+f 2/13/5 1/1/5 3/4/5 4/5/5
+f 6/11/6 5/10/6 1/1/6 2/13/6

diff --git a/floor_fragment_shader.glsl b/floor_fragment_shader.glsl
new file mode 100644 (file)
index 0000000..ac6612a
--- /dev/null
+++ b/floor_fragment_shader.glsl
@@ -0,0 +1,8 @@
+#version 460
+
+in vec4 fcolor;
+out vec4 outcolor;
+
+void main(void) {
+  outcolor = vec4(fcolor.xyz, 1);
+}

diff --git a/floor_vertex_shader.glsl b/floor_vertex_shader.glsl
new file mode 100644 (file)
index 0000000..f051319
--- /dev/null
+++ b/floor_vertex_shader.glsl
@@ -0,0 +1,15 @@
+#version 460\r
+\r
+in vec3 in_vertex;\r
+uniform mat4 mvp;\r
+out vec4 fcolor;\r
+out vec4 gl_Position;\r
+\r
+void main(void) {      \r
+       int x_offset = 2 * (gl_InstanceID / 100) - 100;\r
+       int z_offset = 2 * (gl_InstanceID % 100) - 100;\r
+       vec4 instance_point = vec4(in_vertex.x + float(x_offset), in_vertex.y, in_vertex.z + float(z_offset), 1.0);\r
+       instance_point.xz *= 5;\r
+       gl_Position = mvp * instance_point;\r
+       fcolor = vec4((1 + in_vertex.x) * 0.5, (1 + 0.5 * (in_vertex.z + in_vertex.x)) * 0.5, (1 + in_vertex.x) * 0.5, 1.0);\r
+}\r

diff --git a/fragment_shader.glsl b/fragment_shader.glsl
new file mode 100644 (file)
index 0000000..ac6612a
--- /dev/null
+++ b/fragment_shader.glsl
@@ -0,0 +1,8 @@
+#version 460
+
+in vec4 fcolor;
+out vec4 outcolor;
+
+void main(void) {
+  outcolor = vec4(fcolor.xyz, 1);
+}

diff --git a/game.h b/game.h
new file mode 100644 (file)
index 0000000..720f25a
--- /dev/null
+++ b/game.h
@@ -0,0 +1,106 @@
+
+#ifndef GAME_H
+#define GAME_H
+
+#include<stdio.h>
+#include<stdlib.h>
+#include<fcntl.h>
+#include<GL/glew.h>
+#include<GLFW/glfw3.h>
+#include<glm/glm.hpp>
+#include<glm/gtc/matrix_transform.hpp>
+#include<glm/gtc/type_ptr.hpp>
+#include<vector>
+#include<thread>
+#include<chrono>
+#include<mutex>
+#include "scolor.hpp"
+
+#define _USE_MATH_DEFINES
+#define GRAVITY 0.0001f
+
+struct vertex {
+       glm::vec3 pos;
+       glm::vec2 tex_coord;
+
+       bool operator==(const vertex& other) const {
+               return pos == other.pos && tex_coord == other.tex_coord;
+       }
+}; // __attribute__((packed)); // TODO:  Alignment
+
+unsigned int load_texture(const char* filename);
+int load_model(std::vector<vertex> &verticies, std::vector<uint32_t> &indices, const char* filename, float scale = 1.0f, bool swap_yz = false, glm::vec2 texture_scale = glm::vec2(1, 1));
+GLuint make_program(const char* v_file, const char* tcs_file, const char* tes_file, const char* g_file, const char* f_file);
+GLuint make_shader(const char* filename, GLenum shaderType);
+float randvel(float speed);
+void init_helpers();
+void free_helpers();
+
+class gameobject {
+       public:
+               std::mutex data_mutex;
+               unsigned int mvp_uniform, v_attrib, t_attrib, program, vbuf, cbuf, ebuf, tex, models_buffer;
+               bool collision_check = false; // Consider separating out player from projectile collision here
+               std::vector<glm::vec3> locations;
+               glm::vec3 size; // What about non-square objects?
+               virtual int init() { return 0; }
+               virtual void deinit() {};
+               virtual void draw(glm::mat4) {}
+               virtual void move() {}
+               virtual void animate() {}
+               virtual bool is_on_idx(glm::vec3 position, size_t index, float height) {return false;}
+               virtual ssize_t is_on(glm::vec3 position, float height) {return -1;}
+               virtual ssize_t collision_index(glm::vec3 position, float distance = 0) {
+                       return -1;
+               }
+               virtual glm::vec3 collision_normal(glm::vec3 move_to, glm::vec3 old_position, ssize_t index, float distance = 0) {
+                       return glm::vec3(0, 0, 0);
+               }
+               virtual bool collision_with_index(glm::vec3 position, size_t index, float distance = 0) { return false;}
+               virtual void hit_index(size_t index) {}
+};
+
+class loaded_object : virtual  public gameobject {
+public:
+               const char *objectfile, *texturefile;
+               float scale = 1.0f;
+               glm::vec2 texture_scale = glm::vec2(1, 1);
+               bool swap_yz = false;
+               loaded_object(const char* of, const char* tf, glm::vec3 s) : objectfile(of), texturefile(tf) { size = s;        }
+               
+               int init() override ;
+               void draw(glm::mat4 vp) override;
+};
+
+
+class block_object : virtual  public gameobject {
+       public:
+               bool is_on_idx(glm::vec3 position, size_t index, float height) override;
+               ssize_t is_on(glm::vec3 position, float height) override;
+               ssize_t collision_index(glm::vec3 position, float distance = 0) override;
+               glm::vec3 collision_normal(glm::vec3 move_to, glm::vec3 old_position, ssize_t index, float distance = 0) override;
+               bool collision_with_index(glm::vec3 position, size_t index, float distance = 0) override;
+};
+
+
+class wall_block : public loaded_object, public block_object {
+public:
+               wall_block(const char* of, const char* tf, glm::vec3 s) : loaded_object(of, tf, s) {
+                       collision_check = true;
+               }
+};
+
+/* Global variables that are really all the way global */
+/*
+#ifdef MAIN
+#define EXTERN
+#define INIT(X) = X
+#else
+#define EXTERN extern
+#define INIT(x)
+#endif
+
+EXTERN char* general_buffer;
+EXTERN int framecount = INIT(0);
+*/
+#endif

diff --git a/helpers.cpp b/helpers.cpp
new file mode 100644 (file)
index 0000000..3a0595c
--- /dev/null
+++ b/helpers.cpp
@@ -0,0 +1,178 @@
+#define GLM_ENABLE_EXPERIMENTAL
+
+#include<GL/glew.h>
+#include<glm/glm.hpp>
+#include<glm/gtc/matrix_transform.hpp>
+#include<glm/gtc/type_ptr.hpp>
+#include <glm/gtx/hash.hpp>
+#include <algorithm>
+#include <unordered_map>
+#include <vector>
+#include "scolor.hpp"
+#include "game.h"
+#include "tiny_obj_loader.h"
+#include "stb_image.h"
+
+#define GBLEN 100000
+char *general_buffer;
+
+void init_helpers(){
+       general_buffer = (char*)malloc(GBLEN);
+}
+
+void free_helpers() {
+       free(general_buffer);
+}
+
+float randvel(float speed){
+       long min = -100;
+       long max = 100;
+       return speed * (min + random() % (max + 1 - min));
+}
+
+GLuint make_shader(const char* filename, GLenum shaderType) {
+       FILE* fd = fopen(filename, "r");
+       if (fd == 0) {
+               printf("File not found:  %s\n", filename);
+               return 0;
+       }
+       size_t readlen = fread(general_buffer, 1, GBLEN, fd);
+       fclose(fd);
+       if (readlen == GBLEN) {
+               printf(RED("Buffer Length of %d bytes Inadequate for File %s\n").c_str(), GBLEN, filename);
+               return 0;
+       }
+       if (readlen == 0) {
+               puts(RED("File read problem, read 0 bytes").c_str());
+               return 0;
+       }
+       general_buffer[readlen] = 0;
+       printf(DGREEN("Read shader in file %s (%d bytes)\n").c_str(), filename, readlen);
+       puts(general_buffer);
+       unsigned int s_reference = glCreateShader(shaderType);
+       glShaderSource(s_reference, 1, (const char**)&general_buffer, 0);
+       glCompileShader(s_reference);
+       glGetShaderInfoLog(s_reference, GBLEN, NULL, general_buffer);
+       puts(general_buffer);
+       GLint compile_ok;
+       glGetShaderiv(s_reference, GL_COMPILE_STATUS, &compile_ok);
+       if (compile_ok) {
+               puts(GREEN("Compile Success").c_str());
+               return s_reference;
+       }
+       puts(RED("Compile Failed\n").c_str());
+       return 0;
+}
+
+GLuint make_program(const char* v_file, const char* tcs_file, const char* tes_file, const char* g_file, const char* f_file) {
+       unsigned int vs_reference = make_shader(v_file, GL_VERTEX_SHADER);
+       unsigned int tcs_reference = 0, tes_reference = 0;
+       if (tcs_file)
+               if (!(tcs_reference = make_shader(tcs_file, GL_TESS_CONTROL_SHADER)))
+                       return 0;
+       if (tes_file)
+               if (!(tes_reference = make_shader(tes_file, GL_TESS_EVALUATION_SHADER)))
+                       return 0;
+       unsigned int gs_reference = 0;
+       if (g_file)
+               gs_reference = make_shader(g_file, GL_GEOMETRY_SHADER);
+       unsigned int fs_reference = make_shader(f_file, GL_FRAGMENT_SHADER);
+       if (!(vs_reference && fs_reference))
+               return 0;
+       if (g_file && !gs_reference)
+               return 0;
+
+
+       unsigned int program = glCreateProgram();
+       glAttachShader(program, vs_reference);
+       if (g_file)
+               glAttachShader(program, gs_reference);
+       if (tcs_file)
+               glAttachShader(program, tcs_reference);
+       if (tes_file)
+               glAttachShader(program, tes_reference);
+       glAttachShader(program, fs_reference);
+       glLinkProgram(program);
+       GLint link_ok;
+       glGetProgramiv(program, GL_LINK_STATUS, &link_ok);
+       if (!link_ok) {
+               glGetProgramInfoLog(program, GBLEN, NULL, general_buffer);
+               puts(general_buffer);
+               puts(RED("Link Failed").c_str());
+               return 0;
+       }
+
+       return program;
+}
+
+unsigned int load_texture(const char* filename){
+       int width, height, channels;
+       unsigned char *image_data = stbi_load(filename, &width, &height, &channels, 4);
+       if(image_data){
+               printf("Loaded image, %d by %d\n", width, height);
+               unsigned int tex;
+               glGenTextures(1, &tex);
+               glBindTexture(GL_TEXTURE_2D, tex);
+               // If you want to set parameters, call glTexParameteri (or similar)
+               glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
+               glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_REPEAT);
+               glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, width, height, 0, GL_RGBA, GL_UNSIGNED_BYTE, image_data);
+               free(image_data);
+               return tex;
+       } else {
+               printf(RED("Image failed to load:  %s\n").c_str(), filename);
+               return 0;
+       }
+}
+
+namespace std {
+       template<> struct hash<vertex> {
+               size_t operator()(vertex const& v) const {
+                       return hash<glm::vec3>()(v.pos) ^ (hash<glm::vec2>()(v.tex_coord) << 1);
+               }
+       };
+}
+
+int load_model(std::vector<vertex> &vertices, std::vector<uint32_t> &indices, const char *filename, float scale, bool swap_yz, glm::vec2 texture_scale){
+       tinyobj::attrib_t attrib;
+       std::vector<tinyobj::shape_t> shapes;
+       std::vector<tinyobj::material_t> materials;
+       std::string warn, err;
+       if(!tinyobj::LoadObj(&attrib, &shapes, &materials, &err, filename)){
+               printf("Loading failed!  %s\n", err.c_str());
+               return 1;
+       }
+
+       std::unordered_map<vertex, uint32_t> uniqueVertices = {};
+
+       for(const auto& shape : shapes){
+               for(const auto& index : shape.mesh.indices){
+                       vertex new_vertex = {};
+                       new_vertex.pos = {
+                               attrib.vertices[3 * index.vertex_index + 0],
+                               attrib.vertices[3 * index.vertex_index + 1],
+                               attrib.vertices[3 * index.vertex_index + 2]
+
+                       };      
+                       /* Transform vertices if we need to */
+                       new_vertex.pos *= scale;
+                       if(swap_yz){
+                               float tmp = new_vertex.pos.y;
+                               new_vertex.pos.y = new_vertex.pos.z;
+                               new_vertex.pos.z = tmp;
+                       }
+                       
+                       new_vertex.tex_coord = {
+                               attrib.texcoords[2 * index.texcoord_index + 0],
+                               1.0f - attrib.texcoords[2 * index.texcoord_index + 1]
+                       };
+                       new_vertex.tex_coord *= texture_scale;
+                       if(uniqueVertices.count(new_vertex) == 0){
+                               uniqueVertices[new_vertex] = (uint32_t)vertices.size();
+                               vertices.push_back(new_vertex);
+                       }
+                       indices.push_back(uniqueVertices[new_vertex]);
+               }
+       }
+       return 0;
+}

diff --git a/loaded_object_fragment_shader.glsl b/loaded_object_fragment_shader.glsl
new file mode 100644 (file)
index 0000000..52613af
--- /dev/null
+++ b/loaded_object_fragment_shader.glsl
@@ -0,0 +1,9 @@
+#version 460
+
+in vec2 frag_texcoord;
+out vec4 outcolor;
+uniform sampler2D tex;
+
+void main(void) {
+  outcolor = texture(tex, frag_texcoord);
+}

diff --git a/loaded_object_vertex_shader.glsl b/loaded_object_vertex_shader.glsl
new file mode 100644 (file)
index 0000000..0e92338
--- /dev/null
+++ b/loaded_object_vertex_shader.glsl
@@ -0,0 +1,15 @@
+#version 460
+
+layout(packed, binding=0) buffer model_list {
+       mat4 models[];
+};
+in vec3 in_vertex;
+in vec2 in_texcoord;
+uniform mat4 vp;
+out vec2 frag_texcoord;
+out vec4 gl_Position;
+
+void main(void) {      
+       gl_Position = vp * models[gl_InstanceID] * vec4(in_vertex, 1.0);
+       frag_texcoord = in_texcoord;
+}

diff --git a/main.cpp b/main.cpp
new file mode 100644 (file)
index 0000000..e9c30d5
--- /dev/null
+++ b/main.cpp
@@ -0,0 +1,407 @@
+/*
+   Simplified OpenGL 4.5 demo
+ * Seth Long, Fall 2020
+ * This is a *very* short demo which displays a triangle
+ * Many of these functions can fail, and return error values
+ * I doubt it can be done much shorter without leaving vertices in only main memory or some such
+ */
+
+#include "game.h"
+
+#include "activation_area.h"
+#include "projectiles.h"
+#include "tile_floor.h"
+#include "moving_platform.h"
+#include "turret.h"
+
+std::mutex grand_mutex;
+
+float height = 1550;
+float width = 2600;
+
+// Will be used for a lot of stuff throughout the demo
+// NOTE:  general_buffer is NOT thread safe.  Don't try to load shaders in parallel!
+// NOTE on the NOTE:  You probably shouldn't do that anyway!
+#define GBLEN (1024*32)
+/* Global section */
+int time_resolution = 10;
+int framecount = 0;
+
+/* Player globals */
+glm::vec3 player_position;
+float player_heading;
+float player_height = 2;
+float player_elevation;
+float player_fall_speed = 0;
+gameobject *player_platform = 0;
+size_t player_platform_index = 0;
+
+std::vector<gameobject*> objects;
+projectile ice_balls;
+
+fragment brick_fragments;
+class target : public loaded_object, public block_object {
+public:
+       target() : loaded_object("monkey.obj", "brick.jpg", glm::vec3(15.0f, 10.0f, 15.0f)) {
+               collision_check = true;
+       }
+       void hit_index(size_t index){
+               // Make fragments
+               brick_fragments.create_burst(100, locations[index], 0.01f);
+               locations.erase(locations.begin() + index);
+       }
+       
+};
+
+struct key_status {
+       int forward, backward, left, right;
+};
+struct key_status player_key_status;
+
+void fire(bool burst = false){
+       if(burst)
+               ice_balls.add_projectile(player_position, player_heading, player_elevation, 0.4f, 10000.0f, 1.0f, true);
+       else
+               ice_balls.add_projectile(player_position, player_heading, player_elevation, 0.05f, 100000.0f, 1.0f, false);
+
+}
+
+void mouse_click_callback(GLFWwindow* window, int button, int action, int mods){
+       if(button == GLFW_MOUSE_BUTTON_LEFT && action == GLFW_PRESS)
+               fire();
+       if(button == GLFW_MOUSE_BUTTON_RIGHT && action == GLFW_PRESS)
+               fire(true);
+}
+
+void key_callback(GLFWwindow* window, int key, int scancode, int action, int mods){
+       if(GLFW_KEY_W == key && 1 == action){
+               player_key_status.forward = 1;
+       }       
+       else if(GLFW_KEY_W == key && 0 == action){
+               player_key_status.forward = 0;
+       }       
+       if(GLFW_KEY_S == key && 1 == action){
+               player_key_status.backward = 1;
+       }       
+       else if(GLFW_KEY_S == key && 0 == action){
+               player_key_status.backward = 0;
+       }
+       if(GLFW_KEY_A == key)
+               player_key_status.left = action;
+       if(GLFW_KEY_D == key)
+               player_key_status.right = action;
+       if(GLFW_KEY_SPACE == key && 1 == action){
+               if(player_platform){
+                       player_fall_speed = 0.04f;
+                       player_position.y += 1.0f;
+                       player_platform = 0;
+               }
+       }
+}
+
+ssize_t is_empty(glm::vec3 position, float distance){
+        for(gameobject* o : objects) {
+                ssize_t collide_index = o->collision_index(position, 0.2f);
+                if(collide_index != -1)
+                        return false;
+        }
+        return true;
+
+}
+
+int shutdown_engine = 0;
+/* Must be called at a consistent rate */
+void player_movement(){
+       while(!shutdown_engine){
+//             grand_mutex.lock();
+               auto start = std::chrono::system_clock::now();
+               glm::vec3 step_to_point = player_position;
+               if(player_key_status.forward){
+                       step_to_point += 0.1f * glm::vec3(sinf(player_heading), 0, cosf(player_heading));
+               }
+               if(player_key_status.backward){
+                       step_to_point += 0.1f * glm::vec3(-sinf(player_heading), 0, -cosf(player_heading));
+               }
+               if(player_key_status.left){
+                       step_to_point += 0.05f * glm::vec3(sinf(player_heading + M_PI/2), 0, cosf(player_heading + M_PI/2));
+               }
+               if(player_key_status.right){
+                       step_to_point += 0.05f * glm::vec3(-sinf(player_heading + M_PI/2), 0, -cosf(player_heading + M_PI/2));
+               }
+               for(gameobject* o : objects) {
+                       ssize_t collide_index = o->collision_index(step_to_point, 0.2f);
+                       if(collide_index != -1) {
+                               if(is_empty(glm::vec3(player_position.x, step_to_point.y, step_to_point.z), 0.2f)) {
+                                       step_to_point.x = player_position.x;
+                                       break;
+                               }
+                               else if(is_empty(glm::vec3(step_to_point.x, step_to_point.y, player_position.z), 0.2f)) {
+                                       step_to_point.z = player_position.z;
+                                       break;
+                               }
+                               else {
+                                       step_to_point = player_position;
+                                       break;
+                               }
+               
+               
+                       }
+               }
+               player_position = step_to_point;
+
+               if(player_platform){
+                       if(!player_platform->is_on_idx(player_position, player_platform_index, player_height))
+                               player_platform = 0;
+               } else {
+                       float floor_height = 0;
+                       for(gameobject* o : objects) {
+                               ssize_t ppi = o->is_on(player_position, player_height);
+                               if(ppi != -1) {
+                                       player_platform_index = ppi;
+                                       player_platform = o;    
+                                       floor_height = player_platform->locations[player_platform_index].y + (player_platform->size.y / 2);
+                                       player_fall_speed = 0;
+                                       player_position.y = floor_height + player_height; 
+                               }
+                       }
+                       if(player_position.y - player_height > floor_height) {
+                               player_position.y += player_fall_speed;
+                               player_fall_speed -= GRAVITY;
+                       } else {
+                               player_fall_speed = 0;
+                               player_position.y = floor_height + player_height; 
+                       }
+               }
+//             grand_mutex.unlock();
+               auto end = std::chrono::system_clock::now();
+               //              double difference = std::chrono::duration_cast<std::chrono::milliseconds>(start - end).count();
+               //              printf("Time difference:  %lf\n", difference);
+               std::this_thread::sleep_for(std::chrono::microseconds(1000) - (start - end));
+       }
+}
+
+void object_movement(){
+       while(!shutdown_engine){
+               auto start = std::chrono::system_clock::now();
+//             grand_mutex.lock();
+               if(player_platform){
+                       glm::vec3 pltloc = player_platform->locations[player_platform_index];
+                       float floor_height = pltloc.y + (player_platform->size.y / 2);
+                       player_position.y = floor_height + player_height;
+               }
+//             grand_mutex.unlock();
+               for(gameobject* o : objects)
+                       o->move();
+               auto end = std::chrono::system_clock::now();
+               //              double difference = std::chrono::duration_cast<std::chrono::milliseconds>(start - end).count();
+               //              printf("Time difference:  %lf\n", difference);
+               std::this_thread::sleep_for(std::chrono::microseconds(1000) - (start - end));
+       }
+}
+
+void animation(){
+       while(!shutdown_engine){
+               auto start = std::chrono::system_clock::now();
+               for(gameobject* o : objects)
+                       o->animate();
+               auto end = std::chrono::system_clock::now();
+               //              double difference = std::chrono::duration_cast<std::chrono::milliseconds>(start - end).count();
+               //              printf("Time difference:  %lf\n", difference);
+               std::this_thread::sleep_for(std::chrono::microseconds(10000) - (start - end));
+       }
+}
+
+void collision_detection(){
+       while(!shutdown_engine){
+               auto start = std::chrono::system_clock::now();
+               ice_balls.data_mutex.lock();
+               size_t nproj = ice_balls.locations.size();
+               for(size_t proj_index = 0; proj_index < nproj; proj_index++){
+                       glm::vec3 l = ice_balls.locations[proj_index];
+                       for(auto o : objects){
+                               if(o->collision_check){
+                                       ssize_t index = o->collision_index(l);
+                                       if(index != -1) {
+                                               o->hit_index(index);
+                                               ice_balls.hit_index(proj_index);
+                                               break;
+                                       }
+                               }
+                       }
+               }       
+               ice_balls.data_mutex.unlock();
+               auto end = std::chrono::system_clock::now();
+               //              double difference = std::chrono::duration_cast<std::chrono::milliseconds>(start - end).count();
+               //              printf("Time difference:  %lf\n", difference);
+               std::this_thread::sleep_for(std::chrono::microseconds(1000) - (start - end));
+       }
+}
+void pos_callback(GLFWwindow* window, double xpos, double ypos){
+       double center_x = width/2;
+       double diff_x = xpos - center_x;
+       double center_y = height/2;
+       double diff_y = ypos - center_y;
+       glfwSetCursorPos(window, center_x, center_y);
+       player_heading -= diff_x / 1000.0; // Is this too fast or slow?
+       player_elevation -= diff_y / 1000.0;
+}
+
+void resize(GLFWwindow*, int new_width, int new_height){
+       width = new_width;
+       height = new_height;
+       printf("Window resized, now %f by %f\n", width, height);
+       glViewport(0, 0, width, height);
+}
+
+
+target targets;
+void bob(){
+       static bool bob_happened = false;
+       if(!bob_happened){
+               bob_happened = true;
+               targets.locations.push_back(glm::vec3(-10, 5, 10));
+       }
+};
+
+void debug_callback(GLenum, GLenum, GLuint, GLenum severity, GLsizei, const GLchar* message, const void*) {
+       if(severity == GL_DEBUG_SEVERITY_HIGH)
+               puts(RED(message).c_str());
+       else if(severity == GL_DEBUG_SEVERITY_MEDIUM)
+               puts(YELLOW(message).c_str());
+/* Only uncomment if you want a lot of messages */
+/*
+       else
+               puts(message);
+*/
+}
+
+int main(int argc, char** argv) {
+       init_helpers();
+       if(!glfwInit()) {
+               puts(RED("GLFW init error, terminating\n").c_str());
+               return 1;
+       }
+       GLFWwindow* window = glfwCreateWindow(width, height, "Simple OpenGL 4.0+ Demo", 0, 0);
+       if(!window){
+               puts(RED("Window creation failed, terminating\n").c_str());
+               return 1;
+       }
+       glfwSetInputMode(window, GLFW_CURSOR, GLFW_CURSOR_DISABLED);
+       glfwMakeContextCurrent(window);
+       glewInit();
+
+       unsigned supported_threads = std::thread::hardware_concurrency();
+       printf("Supported threads:  %u\n", supported_threads);
+
+       /* Set up callbacks */
+       glEnable(GL_DEBUG_OUTPUT);
+       glDebugMessageCallback(debug_callback, 0);
+       glfwSetKeyCallback(window, key_callback);
+       glfwSetCursorPosCallback(window, pos_callback);
+       glfwSetFramebufferSizeCallback(window, resize);
+       glfwSetMouseButtonCallback(window, mouse_click_callback);
+
+       /* Set starting point */
+       player_position = glm::vec3(53, 10, 50);
+       player_heading = M_PI;
+
+
+       /* Level Loading (hardcoded at the moment) */
+       tile_floor fl;
+       objects.push_back(&ice_balls);
+       objects.push_back(&fl);
+
+
+       targets.scale = 10.0f;
+       for(int height = 30; height < 300; height+= 30){
+               targets.locations.push_back(glm::vec3(0, height, 0));
+               targets.locations.push_back(glm::vec3(30, height, 0));
+               targets.locations.push_back(glm::vec3(-30, height, 0));
+       }
+       for(int z = -40; z > -300; z += -10)
+               targets.locations.push_back(glm::vec3(10, 3, z));
+       objects.push_back(&targets);
+       objects.push_back(&brick_fragments);
+
+       wall_block wallblock("cube.obj", "brick.jpg", glm::vec3(2, 2, 2));
+       objects.push_back(&wallblock);
+       for(int x = 0; x < 20; x += 2)
+               for(int y = -10; y < 10; y += 2)
+                       for(int z = 0; z < 4; z += 2)
+                               wallblock.locations.push_back(glm::vec3(x, y, z));
+
+
+       activation_area target_spawning;
+       target_spawning.size = glm::vec3(10, 10, 10);
+       target_spawning.add_area(glm::vec3(10, 0, 10), bob);
+       objects.push_back(&target_spawning);
+
+       wall_block monster_box("cube.obj", "monster1.png", glm::vec3(10, 10, 10));
+       monster_box.scale = 5.0;
+       monster_box.locations.push_back(glm::vec3(-40, 0, 20));
+       objects.push_back(&monster_box);
+
+       // What if it wasn't a cube?
+       wall_block bwb("cube.obj", "brick.jpg", glm::vec3(20, 20, 20));
+       bwb.scale = 10;
+       bwb.texture_scale = glm::vec2(10, 10);
+       bwb.locations.push_back(glm::vec3(-30, 0, 0));
+       objects.push_back(&bwb);
+       
+       turret t;
+       t.locations.push_back(glm::vec3(100, 0, 300));
+       t.current_target = &targets;
+       t.current_projectile = &ice_balls;
+       objects.push_back(&t);
+
+
+       /* Initialize game objects */
+       for(gameobject* o : objects){
+               if(o->init()){
+                       puts(RED("Compile Failed, giving up!").c_str());
+                       return 1;
+               }
+       }
+
+       /* Start Other Threads */
+       std::thread player_movement_thread(player_movement);
+       std::thread object_movement_thread(object_movement);
+       std::thread animation_thread(animation);
+       std::thread collision_detection_thread(collision_detection);
+
+       glEnable(GL_DEPTH_TEST);
+       while (!glfwWindowShouldClose(window)) {
+               framecount++;
+               glfwPollEvents();
+               glClearColor(0, 0, 0, 1.0);
+               glClear(GL_COLOR_BUFFER_BIT);
+               glClear(GL_DEPTH_BUFFER_BIT);
+
+//             grand_mutex.lock();
+
+               /* Where are we?  A:  player_position
+                * What are we looking at?
+                */
+               glm::vec3 look_at_point = player_position;
+               look_at_point.x += cosf(player_elevation) * sinf(player_heading);
+               look_at_point.y += sinf(player_elevation);
+               look_at_point.z += cosf(player_elevation) * cosf(player_heading);
+               glm::mat4 view = glm::lookAt(player_position, look_at_point, glm::vec3(0, 1, 0));
+               glm::mat4 projection = glm::perspective(45.0f, width / height, 0.1f, 10000.0f);
+               glm::mat4 vp = projection * view;
+
+               for(gameobject* o : objects)
+                       o->draw(vp);
+//             grand_mutex.unlock();
+
+               glfwSwapBuffers(window);
+       }
+       shutdown_engine = 1;
+       player_movement_thread.join();
+       // TODO:  join other threads
+       glfwDestroyWindow(window);
+       glfwTerminate();
+       free_helpers();
+}
+
+

diff --git a/monkey.obj b/monkey.obj
new file mode 100644 (file)
index 0000000..c0d40be
--- /dev/null
+++ b/monkey.obj
@@ -0,0 +1,2068 @@
+# Blender v3.0.1 OBJ File: ''
+# www.blender.org
+mtllib monkey.mtl
+o Suzanne
+v 0.437500 0.164062 0.765625
+v -0.437500 0.164062 0.765625
+v 0.500000 0.093750 0.687500
+v -0.500000 0.093750 0.687500
+v 0.546875 0.054688 0.578125
+v -0.546875 0.054688 0.578125
+v 0.351562 -0.023438 0.617188
+v -0.351562 -0.023438 0.617188
+v 0.351562 0.031250 0.718750
+v -0.351562 0.031250 0.718750
+v 0.351562 0.132812 0.781250
+v -0.351562 0.132812 0.781250
+v 0.273438 0.164062 0.796875
+v -0.273438 0.164062 0.796875
+v 0.203125 0.093750 0.742188
+v -0.203125 0.093750 0.742188
+v 0.156250 0.054688 0.648438
+v -0.156250 0.054688 0.648438
+v 0.078125 0.242188 0.656250
+v -0.078125 0.242188 0.656250
+v 0.140625 0.242188 0.742188
+v -0.140625 0.242188 0.742188
+v 0.242188 0.242188 0.796875
+v -0.242188 0.242188 0.796875
+v 0.273438 0.328125 0.796875
+v -0.273438 0.328125 0.796875
+v 0.203125 0.390625 0.742188
+v -0.203125 0.390625 0.742188
+v 0.156250 0.437500 0.648438
+v -0.156250 0.437500 0.648438
+v 0.351562 0.515625 0.617188
+v -0.351562 0.515625 0.617188
+v 0.351562 0.453125 0.718750
+v -0.351562 0.453125 0.718750
+v 0.351562 0.359375 0.781250
+v -0.351562 0.359375 0.781250
+v 0.437500 0.328125 0.765625
+v -0.437500 0.328125 0.765625
+v 0.500000 0.390625 0.687500
+v -0.500000 0.390625 0.687500
+v 0.546875 0.437500 0.578125
+v -0.546875 0.437500 0.578125
+v 0.625000 0.242188 0.562500
+v -0.625000 0.242188 0.562500
+v 0.562500 0.242188 0.671875
+v -0.562500 0.242188 0.671875
+v 0.468750 0.242188 0.757812
+v -0.468750 0.242188 0.757812
+v 0.476562 0.242188 0.773438
+v -0.476562 0.242188 0.773438
+v 0.445312 0.335938 0.781250
+v -0.445312 0.335938 0.781250
+v 0.351562 0.375000 0.804688
+v -0.351562 0.375000 0.804688
+v 0.265625 0.335938 0.820312
+v -0.265625 0.335938 0.820312
+v 0.226562 0.242188 0.820312
+v -0.226562 0.242188 0.820312
+v 0.265625 0.156250 0.820312
+v -0.265625 0.156250 0.820312
+v 0.351562 0.242188 0.828125
+v -0.351562 0.242188 0.828125
+v 0.351562 0.117188 0.804688
+v -0.351562 0.117188 0.804688
+v 0.445312 0.156250 0.781250
+v -0.445312 0.156250 0.781250
+v 0.000000 0.429688 0.742188
+v 0.000000 0.351562 0.820312
+v 0.000000 -0.679688 0.734375
+v 0.000000 -0.320312 0.781250
+v 0.000000 -0.187500 0.796875
+v 0.000000 -0.773438 0.718750
+v 0.000000 0.406250 0.601562
+v 0.000000 0.570312 0.570312
+v 0.000000 0.898438 -0.546875
+v 0.000000 0.562500 -0.851562
+v 0.000000 0.070312 -0.828125
+v 0.000000 -0.382812 -0.351562
+v 0.203125 -0.187500 0.562500
+v -0.203125 -0.187500 0.562500
+v 0.312500 -0.437500 0.570312
+v -0.312500 -0.437500 0.570312
+v 0.351562 -0.695312 0.570312
+v -0.351562 -0.695312 0.570312
+v 0.367188 -0.890625 0.531250
+v -0.367188 -0.890625 0.531250
+v 0.328125 -0.945312 0.523438
+v -0.328125 -0.945312 0.523438
+v 0.179688 -0.968750 0.554688
+v -0.179688 -0.968750 0.554688
+v 0.000000 -0.984375 0.578125
+v 0.437500 -0.140625 0.531250
+v -0.437500 -0.140625 0.531250
+v 0.632812 -0.039062 0.539062
+v -0.632812 -0.039062 0.539062
+v 0.828125 0.148438 0.445312
+v -0.828125 0.148438 0.445312
+v 0.859375 0.429688 0.593750
+v -0.859375 0.429688 0.593750
+v 0.710938 0.484375 0.625000
+v -0.710938 0.484375 0.625000
+v 0.492188 0.601562 0.687500
+v -0.492188 0.601562 0.687500
+v 0.320312 0.757812 0.734375
+v -0.320312 0.757812 0.734375
+v 0.156250 0.718750 0.757812
+v -0.156250 0.718750 0.757812
+v 0.062500 0.492188 0.750000
+v -0.062500 0.492188 0.750000
+v 0.164062 0.414062 0.773438
+v -0.164062 0.414062 0.773438
+v 0.125000 0.304688 0.765625
+v -0.125000 0.304688 0.765625
+v 0.203125 0.093750 0.742188
+v -0.203125 0.093750 0.742188
+v 0.375000 0.015625 0.703125
+v -0.375000 0.015625 0.703125
+v 0.492188 0.062500 0.671875
+v -0.492188 0.062500 0.671875
+v 0.625000 0.187500 0.648438
+v -0.625000 0.187500 0.648438
+v 0.640625 0.296875 0.648438
+v -0.640625 0.296875 0.648438
+v 0.601562 0.375000 0.664062
+v -0.601562 0.375000 0.664062
+v 0.429688 0.437500 0.718750
+v -0.429688 0.437500 0.718750
+v 0.250000 0.468750 0.757812
+v -0.250000 0.468750 0.757812
+v 0.000000 -0.765625 0.734375
+v 0.109375 -0.718750 0.734375
+v -0.109375 -0.718750 0.734375
+v 0.117188 -0.835938 0.710938
+v -0.117188 -0.835938 0.710938
+v 0.062500 -0.882812 0.695312
+v -0.062500 -0.882812 0.695312
+v 0.000000 -0.890625 0.687500
+v 0.000000 -0.195312 0.750000
+v 0.000000 -0.140625 0.742188
+v 0.101562 -0.148438 0.742188
+v -0.101562 -0.148438 0.742188
+v 0.125000 -0.226562 0.750000
+v -0.125000 -0.226562 0.750000
+v 0.085938 -0.289062 0.742188
+v -0.085938 -0.289062 0.742188
+v 0.398438 -0.046875 0.671875
+v -0.398438 -0.046875 0.671875
+v 0.617188 0.054688 0.625000
+v -0.617188 0.054688 0.625000
+v 0.726562 0.203125 0.601562
+v -0.726562 0.203125 0.601562
+v 0.742188 0.375000 0.656250
+v -0.742188 0.375000 0.656250
+v 0.687500 0.414062 0.726562
+v -0.687500 0.414062 0.726562
+v 0.437500 0.546875 0.796875
+v -0.437500 0.546875 0.796875
+v 0.312500 0.640625 0.835938
+v -0.312500 0.640625 0.835938
+v 0.203125 0.617188 0.851562
+v -0.203125 0.617188 0.851562
+v 0.101562 0.429688 0.843750
+v -0.101562 0.429688 0.843750
+v 0.125000 -0.101562 0.812500
+v -0.125000 -0.101562 0.812500
+v 0.210938 -0.445312 0.710938
+v -0.210938 -0.445312 0.710938
+v 0.250000 -0.703125 0.687500
+v -0.250000 -0.703125 0.687500
+v 0.265625 -0.820312 0.664062
+v -0.265625 -0.820312 0.664062
+v 0.234375 -0.914062 0.632812
+v -0.234375 -0.914062 0.632812
+v 0.164062 -0.929688 0.632812
+v -0.164062 -0.929688 0.632812
+v 0.000000 -0.945312 0.640625
+v 0.000000 0.046875 0.726562
+v 0.000000 0.210938 0.765625
+v 0.328125 0.476562 0.742188
+v -0.328125 0.476562 0.742188
+v 0.164062 0.140625 0.750000
+v -0.164062 0.140625 0.750000
+v 0.132812 0.210938 0.757812
+v -0.132812 0.210938 0.757812
+v 0.117188 -0.687500 0.734375
+v -0.117188 -0.687500 0.734375
+v 0.078125 -0.445312 0.750000
+v -0.078125 -0.445312 0.750000
+v 0.000000 -0.445312 0.750000
+v 0.000000 -0.328125 0.742188
+v 0.093750 -0.273438 0.781250
+v -0.093750 -0.273438 0.781250
+v 0.132812 -0.226562 0.796875
+v -0.132812 -0.226562 0.796875
+v 0.109375 -0.132812 0.781250
+v -0.109375 -0.132812 0.781250
+v 0.039062 -0.125000 0.781250
+v -0.039062 -0.125000 0.781250
+v 0.000000 -0.203125 0.828125
+v 0.046875 -0.148438 0.812500
+v -0.046875 -0.148438 0.812500
+v 0.093750 -0.156250 0.812500
+v -0.093750 -0.156250 0.812500
+v 0.109375 -0.226562 0.828125
+v -0.109375 -0.226562 0.828125
+v 0.078125 -0.250000 0.804688
+v -0.078125 -0.250000 0.804688
+v 0.000000 -0.289062 0.804688
+v 0.257812 -0.312500 0.554688
+v -0.257812 -0.312500 0.554688
+v 0.164062 -0.242188 0.710938
+v -0.164062 -0.242188 0.710938
+v 0.179688 -0.312500 0.710938
+v -0.179688 -0.312500 0.710938
+v 0.234375 -0.250000 0.554688
+v -0.234375 -0.250000 0.554688
+v 0.000000 -0.875000 0.687500
+v 0.046875 -0.867188 0.687500
+v -0.046875 -0.867188 0.687500
+v 0.093750 -0.820312 0.710938
+v -0.093750 -0.820312 0.710938
+v 0.093750 -0.742188 0.726562
+v -0.093750 -0.742188 0.726562
+v 0.000000 -0.781250 0.656250
+v 0.093750 -0.750000 0.664062
+v -0.093750 -0.750000 0.664062
+v 0.093750 -0.812500 0.640625
+v -0.093750 -0.812500 0.640625
+v 0.046875 -0.851562 0.632812
+v -0.046875 -0.851562 0.632812
+v 0.000000 -0.859375 0.632812
+v 0.171875 0.218750 0.781250
+v -0.171875 0.218750 0.781250
+v 0.187500 0.156250 0.773438
+v -0.187500 0.156250 0.773438
+v 0.335938 0.429688 0.757812
+v -0.335938 0.429688 0.757812
+v 0.273438 0.421875 0.773438
+v -0.273438 0.421875 0.773438
+v 0.421875 0.398438 0.773438
+v -0.421875 0.398438 0.773438
+v 0.562500 0.351562 0.695312
+v -0.562500 0.351562 0.695312
+v 0.585938 0.289062 0.687500
+v -0.585938 0.289062 0.687500
+v 0.578125 0.195312 0.679688
+v -0.578125 0.195312 0.679688
+v 0.476562 0.101562 0.718750
+v -0.476562 0.101562 0.718750
+v 0.375000 0.062500 0.742188
+v -0.375000 0.062500 0.742188
+v 0.226562 0.109375 0.781250
+v -0.226562 0.109375 0.781250
+v 0.179688 0.296875 0.781250
+v -0.179688 0.296875 0.781250
+v 0.210938 0.375000 0.781250
+v -0.210938 0.375000 0.781250
+v 0.234375 0.359375 0.757812
+v -0.234375 0.359375 0.757812
+v 0.195312 0.296875 0.757812
+v -0.195312 0.296875 0.757812
+v 0.242188 0.125000 0.757812
+v -0.242188 0.125000 0.757812
+v 0.375000 0.085938 0.726562
+v -0.375000 0.085938 0.726562
+v 0.460938 0.117188 0.703125
+v -0.460938 0.117188 0.703125
+v 0.546875 0.210938 0.671875
+v -0.546875 0.210938 0.671875
+v 0.554688 0.281250 0.671875
+v -0.554688 0.281250 0.671875
+v 0.531250 0.335938 0.679688
+v -0.531250 0.335938 0.679688
+v 0.414062 0.390625 0.750000
+v -0.414062 0.390625 0.750000
+v 0.281250 0.398438 0.765625
+v -0.281250 0.398438 0.765625
+v 0.335938 0.406250 0.750000
+v -0.335938 0.406250 0.750000
+v 0.203125 0.171875 0.750000
+v -0.203125 0.171875 0.750000
+v 0.195312 0.226562 0.750000
+v -0.195312 0.226562 0.750000
+v 0.109375 0.460938 0.609375
+v -0.109375 0.460938 0.609375
+v 0.195312 0.664062 0.617188
+v -0.195312 0.664062 0.617188
+v 0.335938 0.687500 0.593750
+v -0.335938 0.687500 0.593750
+v 0.484375 0.554688 0.554688
+v -0.484375 0.554688 0.554688
+v 0.679688 0.453125 0.492188
+v -0.679688 0.453125 0.492188
+v 0.796875 0.406250 0.460938
+v -0.796875 0.406250 0.460938
+v 0.773438 0.164062 0.375000
+v -0.773438 0.164062 0.375000
+v 0.601562 0.000000 0.414062
+v -0.601562 0.000000 0.414062
+v 0.437500 -0.093750 0.468750
+v -0.437500 -0.093750 0.468750
+v 0.000000 0.898438 0.289062
+v 0.000000 0.984375 -0.078125
+v 0.000000 -0.195312 -0.671875
+v 0.000000 -0.460938 0.187500
+v 0.000000 -0.976562 0.460938
+v 0.000000 -0.804688 0.343750
+v 0.000000 -0.570312 0.320312
+v 0.000000 -0.484375 0.281250
+v 0.851562 0.234375 0.054688
+v -0.851562 0.234375 0.054688
+v 0.859375 0.320312 -0.046875
+v -0.859375 0.320312 -0.046875
+v 0.773438 0.265625 -0.437500
+v -0.773438 0.265625 -0.437500
+v 0.460938 0.437500 -0.703125
+v -0.460938 0.437500 -0.703125
+v 0.734375 -0.046875 0.070312
+v -0.734375 -0.046875 0.070312
+v 0.593750 -0.125000 -0.164062
+v -0.593750 -0.125000 -0.164062
+v 0.640625 -0.007812 -0.429688
+v -0.640625 -0.007812 -0.429688
+v 0.335938 0.054688 -0.664062
+v -0.335938 0.054688 -0.664062
+v 0.234375 -0.351562 0.406250
+v -0.234375 -0.351562 0.406250
+v 0.179688 -0.414062 0.257812
+v -0.179688 -0.414062 0.257812
+v 0.289062 -0.710938 0.382812
+v -0.289062 -0.710938 0.382812
+v 0.250000 -0.500000 0.390625
+v -0.250000 -0.500000 0.390625
+v 0.328125 -0.914062 0.398438
+v -0.328125 -0.914062 0.398438
+v 0.140625 -0.757812 0.367188
+v -0.140625 -0.757812 0.367188
+v 0.125000 -0.539062 0.359375
+v -0.125000 -0.539062 0.359375
+v 0.164062 -0.945312 0.437500
+v -0.164062 -0.945312 0.437500
+v 0.218750 -0.281250 0.429688
+v -0.218750 -0.281250 0.429688
+v 0.210938 -0.226562 0.468750
+v -0.210938 -0.226562 0.468750
+v 0.203125 -0.171875 0.500000
+v -0.203125 -0.171875 0.500000
+v 0.210938 -0.390625 0.164062
+v -0.210938 -0.390625 0.164062
+v 0.296875 -0.312500 -0.265625
+v -0.296875 -0.312500 -0.265625
+v 0.343750 -0.148438 -0.539062
+v -0.343750 -0.148438 -0.539062
+v 0.453125 0.867188 -0.382812
+v -0.453125 0.867188 -0.382812
+v 0.453125 0.929688 -0.070312
+v -0.453125 0.929688 -0.070312
+v 0.453125 0.851562 0.234375
+v -0.453125 0.851562 0.234375
+v 0.460938 0.523438 0.429688
+v -0.460938 0.523438 0.429688
+v 0.726562 0.406250 0.335938
+v -0.726562 0.406250 0.335938
+v 0.632812 0.453125 0.281250
+v -0.632812 0.453125 0.281250
+v 0.640625 0.703125 0.054688
+v -0.640625 0.703125 0.054688
+v 0.796875 0.562500 0.125000
+v -0.796875 0.562500 0.125000
+v 0.796875 0.617188 -0.117188
+v -0.796875 0.617188 -0.117188
+v 0.640625 0.750000 -0.195312
+v -0.640625 0.750000 -0.195312
+v 0.640625 0.679688 -0.445312
+v -0.640625 0.679688 -0.445312
+v 0.796875 0.539062 -0.359375
+v -0.796875 0.539062 -0.359375
+v 0.617188 0.328125 -0.585938
+v -0.617188 0.328125 -0.585938
+v 0.484375 0.023438 -0.546875
+v -0.484375 0.023438 -0.546875
+v 0.820312 0.328125 -0.203125
+v -0.820312 0.328125 -0.203125
+v 0.406250 -0.171875 0.148438
+v -0.406250 -0.171875 0.148438
+v 0.429688 -0.195312 -0.210938
+v -0.429688 -0.195312 -0.210938
+v 0.890625 0.406250 -0.234375
+v -0.890625 0.406250 -0.234375
+v 0.773438 -0.140625 -0.125000
+v -0.773438 -0.140625 -0.125000
+v 1.039062 -0.101562 -0.328125
+v -1.039062 -0.101562 -0.328125
+v 1.281250 0.054688 -0.429688
+v -1.281250 0.054688 -0.429688
+v 1.351562 0.320312 -0.421875
+v -1.351562 0.320312 -0.421875
+v 1.234375 0.507812 -0.421875
+v -1.234375 0.507812 -0.421875
+v 1.023438 0.476562 -0.312500
+v -1.023438 0.476562 -0.312500
+v 1.015625 0.414062 -0.289062
+v -1.015625 0.414062 -0.289062
+v 1.187500 0.437500 -0.390625
+v -1.187500 0.437500 -0.390625
+v 1.265625 0.289062 -0.406250
+v -1.265625 0.289062 -0.406250
+v 1.210938 0.078125 -0.406250
+v -1.210938 0.078125 -0.406250
+v 1.031250 -0.039062 -0.304688
+v -1.031250 -0.039062 -0.304688
+v 0.828125 -0.070312 -0.132812
+v -0.828125 -0.070312 -0.132812
+v 0.921875 0.359375 -0.218750
+v -0.921875 0.359375 -0.218750
+v 0.945312 0.304688 -0.289062
+v -0.945312 0.304688 -0.289062
+v 0.882812 -0.023438 -0.210938
+v -0.882812 -0.023438 -0.210938
+v 1.039062 0.000000 -0.367188
+v -1.039062 0.000000 -0.367188
+v 1.187500 0.093750 -0.445312
+v -1.187500 0.093750 -0.445312
+v 1.234375 0.250000 -0.445312
+v -1.234375 0.250000 -0.445312
+v 1.171875 0.359375 -0.437500
+v -1.171875 0.359375 -0.437500
+v 1.023438 0.343750 -0.359375
+v -1.023438 0.343750 -0.359375
+v 0.843750 0.289062 -0.210938
+v -0.843750 0.289062 -0.210938
+v 0.835938 0.171875 -0.273438
+v -0.835938 0.171875 -0.273438
+v 0.757812 0.093750 -0.273438
+v -0.757812 0.093750 -0.273438
+v 0.820312 0.085938 -0.273438
+v -0.820312 0.085938 -0.273438
+v 0.843750 0.015625 -0.273438
+v -0.843750 0.015625 -0.273438
+v 0.812500 -0.015625 -0.273438
+v -0.812500 -0.015625 -0.273438
+v 0.726562 0.000000 -0.070312
+v -0.726562 0.000000 -0.070312
+v 0.718750 -0.023438 -0.171875
+v -0.718750 -0.023438 -0.171875
+v 0.718750 0.039062 -0.187500
+v -0.718750 0.039062 -0.187500
+v 0.796875 0.203125 -0.210938
+v -0.796875 0.203125 -0.210938
+v 0.890625 0.242188 -0.265625
+v -0.890625 0.242188 -0.265625
+v 0.890625 0.234375 -0.320312
+v -0.890625 0.234375 -0.320312
+v 0.812500 -0.015625 -0.320312
+v -0.812500 -0.015625 -0.320312
+v 0.851562 0.015625 -0.320312
+v -0.851562 0.015625 -0.320312
+v 0.828125 0.078125 -0.320312
+v -0.828125 0.078125 -0.320312
+v 0.765625 0.093750 -0.320312
+v -0.765625 0.093750 -0.320312
+v 0.843750 0.171875 -0.320312
+v -0.843750 0.171875 -0.320312
+v 1.039062 0.328125 -0.414062
+v -1.039062 0.328125 -0.414062
+v 1.187500 0.343750 -0.484375
+v -1.187500 0.343750 -0.484375
+v 1.257812 0.242188 -0.492188
+v -1.257812 0.242188 -0.492188
+v 1.210938 0.085938 -0.484375
+v -1.210938 0.085938 -0.484375
+v 1.046875 0.000000 -0.421875
+v -1.046875 0.000000 -0.421875
+v 0.882812 -0.015625 -0.265625
+v -0.882812 -0.015625 -0.265625
+v 0.953125 0.289062 -0.343750
+v -0.953125 0.289062 -0.343750
+v 0.890625 0.109375 -0.328125
+v -0.890625 0.109375 -0.328125
+v 0.937500 0.062500 -0.335938
+v -0.937500 0.062500 -0.335938
+v 1.000000 0.125000 -0.367188
+v -1.000000 0.125000 -0.367188
+v 0.960938 0.171875 -0.351562
+v -0.960938 0.171875 -0.351562
+v 1.015625 0.234375 -0.375000
+v -1.015625 0.234375 -0.375000
+v 1.054688 0.187500 -0.382812
+v -1.054688 0.187500 -0.382812
+v 1.109375 0.210938 -0.390625
+v -1.109375 0.210938 -0.390625
+v 1.085938 0.273438 -0.390625
+v -1.085938 0.273438 -0.390625
+v 1.023438 0.437500 -0.484375
+v -1.023438 0.437500 -0.484375
+v 1.250000 0.468750 -0.546875
+v -1.250000 0.468750 -0.546875
+v 1.367188 0.296875 -0.500000
+v -1.367188 0.296875 -0.500000
+v 1.312500 0.054688 -0.531250
+v -1.312500 0.054688 -0.531250
+v 1.039062 -0.085938 -0.492188
+v -1.039062 -0.085938 -0.492188
+v 0.789062 -0.125000 -0.328125
+v -0.789062 -0.125000 -0.328125
+v 0.859375 0.382812 -0.382812
+v -0.859375 0.382812 -0.382812
+vt 0.890955 0.590063
+vt 0.870622 0.589649
+vt 0.860081 0.560115
+vt 0.904571 0.559404
+vt 0.856226 0.850547
+vt 0.868067 0.821510
+vt 0.888398 0.821999
+vt 0.900640 0.853232
+vt 0.853018 0.521562
+vt 0.920166 0.524546
+vt 0.847458 0.888748
+vt 0.914672 0.888748
+vt 0.828900 0.590771
+vt 0.798481 0.569535
+vt 0.795104 0.838402
+vt 0.826436 0.818537
+vt 0.854402 0.604754
+vt 0.852534 0.805700
+vt 0.854107 0.625459
+vt 0.828171 0.633354
+vt 0.827598 0.775964
+vt 0.853157 0.785002
+vt 0.791018 0.645443
+vt 0.791018 0.762238
+vt 0.855181 0.668527
+vt 0.842358 0.702491
+vt 0.844839 0.707525
+vt 0.856142 0.742025
+vt 0.867508 0.642291
+vt 0.867293 0.768782
+vt 0.890474 0.641909
+vt 0.900375 0.666964
+vt 0.901223 0.745592
+vt 0.890219 0.770183
+vt 0.918898 0.699697
+vt 0.921180 0.713713
+vt 0.931889 0.636832
+vt 0.968392 0.645333
+vt 0.968213 0.770220
+vt 0.931368 0.777093
+vt 0.905882 0.627902
+vt 0.904990 0.784860
+vt 0.906232 0.605742
+vt 0.933717 0.593037
+vt 0.931250 0.820926
+vt 0.904357 0.807013
+vt 0.968392 0.573812
+vt 0.965038 0.841671
+vt 0.902359 0.607909
+vt 0.889591 0.593275
+vt 0.900583 0.804677
+vt 0.887178 0.818729
+vt 0.899781 0.626257
+vt 0.898822 0.786233
+vt 0.887842 0.636527
+vt 0.887351 0.775442
+vt 0.870908 0.635245
+vt 0.870376 0.775972
+vt 0.859881 0.623942
+vt 0.858859 0.786774
+vt 0.859664 0.608186
+vt 0.857942 0.802505
+vt 0.871664 0.593961
+vt 0.869299 0.817249
+vt 0.879400 0.616512
+vt 0.878029 0.795063
+vt 0.540260 0.053805
+vt 0.536419 0.062072
+vt 0.518925 0.059681
+vt 0.518916 0.050294
+vt 0.501452 0.062043
+vt 0.497626 0.053770
+vt 0.551930 0.058338
+vt 0.542788 0.064089
+vt 0.495083 0.064047
+vt 0.485955 0.058273
+vt 0.555073 0.061900
+vt 0.546290 0.072669
+vt 0.491565 0.072625
+vt 0.482805 0.061829
+vt 0.563812 0.076586
+vt 0.548333 0.084893
+vt 0.489507 0.084858
+vt 0.474014 0.076511
+vt 0.583135 0.108495
+vt 0.555621 0.121749
+vt 0.482177 0.121781
+vt 0.454527 0.108481
+vt 0.605512 0.165134
+vt 0.647395 0.200502
+vt 0.621513 0.227818
+vt 0.553118 0.209599
+vt 0.416514 0.229490
+vt 0.389677 0.201890
+vt 0.432024 0.165644
+vt 0.485339 0.210053
+vt 0.676379 0.233241
+vt 0.664761 0.253225
+vt 0.372747 0.256357
+vt 0.360308 0.235899
+vt 0.715342 0.265392
+vt 0.683908 0.279995
+vt 0.353696 0.284606
+vt 0.320452 0.270303
+vt 0.707254 0.310054
+vt 0.687515 0.311539
+vt 0.351187 0.317440
+vt 0.330721 0.316853
+vt 0.697446 0.332673
+vt 0.676824 0.323937
+vt 0.362723 0.329722
+vt 0.341964 0.339667
+vt 0.662817 0.372521
+vt 0.639050 0.357330
+vt 0.402772 0.362131
+vt 0.379297 0.378686
+vt 0.626842 0.395792
+vt 0.618316 0.375151
+vt 0.424583 0.379267
+vt 0.416915 0.400552
+vt 0.604826 0.397804
+vt 0.600808 0.377857
+vt 0.442396 0.381222
+vt 0.439252 0.401540
+vt 0.553095 0.390512
+vt 0.559674 0.357011
+vt 0.482938 0.358497
+vt 0.490934 0.391862
+vt 0.521923 0.386009
+vt 0.521086 0.343868
+vt 0.577279 0.340156
+vt 0.599845 0.344815
+vt 0.441977 0.347815
+vt 0.464579 0.342230
+vt 0.615546 0.342005
+vt 0.425972 0.345582
+vt 0.634472 0.332311
+vt 0.406362 0.336480
+vt 0.662406 0.312804
+vt 0.377061 0.317685
+vt 0.668440 0.297958
+vt 0.370304 0.302644
+vt 0.664101 0.277872
+vt 0.374100 0.281778
+vt 0.639236 0.253047
+vt 0.398938 0.255633
+vt 0.613992 0.242662
+vt 0.424464 0.244473
+vt 0.572941 0.258564
+vt 0.466409 0.259709
+vt 0.563905 0.272007
+vt 0.519760 0.248864
+vt 0.475886 0.273078
+vt 0.558527 0.316594
+vt 0.482619 0.317843
+vt 0.520277 0.294764
+vt 0.556923 0.291214
+vt 0.483433 0.292249
+vt 0.525483 0.068967
+vt 0.518928 0.067899
+vt 0.512375 0.068956
+vt 0.531231 0.073829
+vt 0.506626 0.073811
+vt 0.531019 0.087431
+vt 0.506827 0.087416
+vt 0.532042 0.127713
+vt 0.532669 0.090920
+vt 0.505177 0.090908
+vt 0.505828 0.127728
+vt 0.538112 0.158382
+vt 0.518981 0.151749
+vt 0.518941 0.128358
+vt 0.499851 0.158434
+vt 0.518925 0.093952
+vt 0.518927 0.085180
+vt 0.548362 0.173560
+vt 0.537959 0.175966
+vt 0.535214 0.166808
+vt 0.502799 0.166857
+vt 0.500100 0.176033
+vt 0.489683 0.173693
+vt 0.544281 0.193366
+vt 0.537248 0.187577
+vt 0.500890 0.187571
+vt 0.493996 0.193428
+vt 0.519841 0.200843
+vt 0.528757 0.191785
+vt 0.509219 0.191626
+vt 0.517577 0.190607
+vt 0.519132 0.185382
+vt 0.518998 0.159028
+vt 0.531131 0.171631
+vt 0.519016 0.165599
+vt 0.506910 0.171667
+vt 0.519099 0.179457
+vt 0.528222 0.186316
+vt 0.509787 0.186260
+vt 0.533528 0.184215
+vt 0.504547 0.184206
+vt 0.533449 0.176739
+vt 0.504604 0.176791
+vt 0.561572 0.167779
+vt 0.476363 0.167996
+vt 0.559475 0.149319
+vt 0.478371 0.149447
+vt 0.596138 0.133426
+vt 0.441395 0.133592
+vt 0.601169 0.147885
+vt 0.436337 0.148194
+vt 0.518925 0.083865
+vt 0.528933 0.084957
+vt 0.508915 0.084945
+vt 0.529036 0.075429
+vt 0.508820 0.075415
+vt 0.523751 0.070508
+vt 0.514106 0.070501
+vt 0.518929 0.069468
+vt 0.521560 0.074970
+vt 0.518928 0.074259
+vt 0.516297 0.074966
+vt 0.524236 0.076691
+vt 0.513619 0.076684
+vt 0.524601 0.079886
+vt 0.513252 0.079879
+vt 0.518926 0.079331
+vt 0.571787 0.277295
+vt 0.568351 0.292904
+vt 0.468070 0.278617
+vt 0.471978 0.294282
+vt 0.573085 0.311386
+vt 0.467790 0.313081
+vt 0.584855 0.327708
+vt 0.456477 0.329961
+vt 0.580734 0.266620
+vt 0.458737 0.268049
+vt 0.611720 0.255725
+vt 0.427062 0.257728
+vt 0.632494 0.262853
+vt 0.406068 0.265508
+vt 0.653658 0.279971
+vt 0.384904 0.283634
+vt 0.656064 0.297636
+vt 0.383015 0.301864
+vt 0.652752 0.310186
+vt 0.386858 0.314615
+vt 0.629040 0.323864
+vt 0.411556 0.327673
+vt 0.614408 0.331972
+vt 0.426727 0.335361
+vt 0.601033 0.333624
+vt 0.440344 0.336537
+vt 0.590644 0.321516
+vt 0.601799 0.328453
+vt 0.450408 0.323919
+vt 0.439372 0.331331
+vt 0.613335 0.327083
+vt 0.427623 0.330358
+vt 0.626851 0.320513
+vt 0.413648 0.324175
+vt 0.646248 0.306421
+vt 0.393381 0.310510
+vt 0.649541 0.296225
+vt 0.389662 0.300183
+vt 0.647785 0.283486
+vt 0.391040 0.287071
+vt 0.629829 0.267263
+vt 0.408893 0.269959
+vt 0.612641 0.261560
+vt 0.426254 0.263693
+vt 0.585166 0.270991
+vt 0.454369 0.272583
+vt 0.578124 0.281900
+vt 0.461798 0.283441
+vt 0.579548 0.309340
+vt 0.461204 0.311233
+vt 0.577524 0.293776
+vt 0.462754 0.295432
+vt 0.553209 0.433063
+vt 0.523031 0.433628
+vt 0.492809 0.434538
+vt 0.609819 0.431516
+vt 0.435860 0.435740
+vt 0.648174 0.419316
+vt 0.396518 0.425416
+vt 0.692106 0.388274
+vt 0.350292 0.396229
+vt 0.726332 0.341754
+vt 0.312756 0.350588
+vt 0.735879 0.312112
+vt 0.301067 0.320593
+vt 0.729900 0.256393
+vt 0.304876 0.261087
+vt 0.698172 0.216906
+vt 0.337414 0.219179
+vt 0.663103 0.190671
+vt 0.373474 0.191872
+vt 0.626908 0.015608
+vt 0.649444 0.022378
+vt 0.660451 0.076084
+vt 0.621440 0.048089
+vt 0.376796 0.075296
+vt 0.388827 0.021586
+vt 0.411318 0.015131
+vt 0.416419 0.047631
+vt 0.567460 0.000144
+vt 0.577206 0.032801
+vt 0.470636 0.000144
+vt 0.460782 0.032656
+vt 0.518922 0.024886
+vt 0.547413 0.041724
+vt 0.490511 0.041669
+vt 0.558059 0.053871
+vt 0.479842 0.053785
+vt 0.576951 0.057998
+vt 0.460920 0.057845
+vt 0.611687 0.078268
+vt 0.425932 0.077985
+vt 0.626663 0.111357
+vt 0.410618 0.111244
+vt 0.629482 0.130456
+vt 0.623495 0.146796
+vt 0.413741 0.147158
+vt 0.407648 0.130594
+vt 0.619303 0.159841
+vt 0.418035 0.160361
+vt 0.945900 0.079569
+vt 0.886245 0.121777
+vt 0.849114 0.099732
+vt 0.891780 0.036916
+vt 0.183115 0.092127
+vt 0.141314 0.112482
+vt 0.078961 0.060719
+vt 0.142277 0.021467
+vt 0.788458 0.080826
+vt 0.805584 0.010786
+vt 0.246353 0.076510
+vt 0.232648 0.003484
+vt 0.687018 0.077204
+vt 0.672384 0.022201
+vt 0.349875 0.075955
+vt 0.365979 0.020991
+vt 0.760215 0.193244
+vt 0.789046 0.233323
+vt 0.271553 0.193871
+vt 0.241255 0.236977
+vt 0.994525 0.167705
+vt 0.909112 0.183261
+vt 0.107928 0.179083
+vt 0.011829 0.155367
+vt 0.911671 0.402429
+vt 0.862868 0.338556
+vt 0.894128 0.301884
+vt 0.962901 0.344752
+vt 0.123776 0.315519
+vt 0.160557 0.356821
+vt 0.106400 0.432652
+vt 0.043968 0.367038
+vt 0.915360 0.259804
+vt 0.999856 0.254640
+vt 0.098965 0.266968
+vt 0.000144 0.259113
+vt 0.749542 0.334683
+vt 0.766337 0.300809
+vt 0.789162 0.313727
+vt 0.267408 0.310142
+vt 0.288183 0.346496
+vt 0.242992 0.325552
+vt 0.815314 0.276388
+vt 0.846174 0.293397
+vt 0.213065 0.285164
+vt 0.178537 0.304983
+vt 0.845007 0.256352
+vt 0.873517 0.265922
+vt 0.179662 0.263312
+vt 0.147089 0.274284
+vt 0.859075 0.228168
+vt 0.886999 0.233769
+vt 0.162803 0.231720
+vt 0.131514 0.237587
+vt 0.842355 0.195160
+vt 0.875030 0.184705
+vt 0.145224 0.182749
+vt 0.176788 0.196179
+vt 0.794286 0.364062
+vt 0.239776 0.382592
+vt 0.770185 0.379538
+vt 0.268122 0.398737
+vt 0.845499 0.449967
+vt 0.185281 0.484099
+vt 0.815858 0.445381
+vt 0.770572 0.444261
+vt 0.755700 0.418603
+vt 0.287033 0.442912
+vt 0.271364 0.473316
+vt 0.219260 0.477186
+vt 0.819845 0.468071
+vt 0.215894 0.503605
+vt 0.809631 0.233887
+vt 0.219168 0.237388
+vt 0.829287 0.219562
+vt 0.199067 0.222464
+vt 0.786480 0.117591
+vt 0.715482 0.139727
+vt 0.246666 0.114850
+vt 0.319538 0.139409
+vt 0.785486 0.152330
+vt 0.245969 0.151002
+vt 0.837382 0.156361
+vt 0.858171 0.137775
+vt 0.171653 0.132294
+vt 0.196622 0.155241
+vt 0.506166 0.904851
+vt 0.432388 0.894943
+vt 0.438797 0.870229
+vt 0.491058 0.881714
+vt 0.315867 0.868209
+vt 0.321637 0.893225
+vt 0.247207 0.901159
+vt 0.263032 0.878321
+vt 0.572792 0.860484
+vt 0.604825 0.879946
+vt 0.181486 0.854693
+vt 0.148729 0.873349
+vt 0.586396 0.793977
+vt 0.619962 0.791615
+vt 0.169745 0.787474
+vt 0.136063 0.784093
+vt 0.549027 0.746412
+vt 0.563786 0.739211
+vt 0.208656 0.740879
+vt 0.194086 0.733241
+vt 0.500314 0.711729
+vt 0.508270 0.697693
+vt 0.258399 0.707497
+vt 0.250811 0.693249
+vt 0.438641 0.680683
+vt 0.434803 0.658882
+vt 0.320962 0.677959
+vt 0.325318 0.656224
+vt 0.505666 0.730944
+vt 0.452955 0.700023
+vt 0.306136 0.696976
+vt 0.252524 0.726592
+vt 0.542850 0.755753
+vt 0.214575 0.750414
+vt 0.568148 0.787367
+vt 0.188269 0.781375
+vt 0.555495 0.826352
+vt 0.199850 0.820889
+vt 0.501231 0.844356
+vt 0.253846 0.840502
+vt 0.457832 0.840040
+vt 0.297562 0.837358
+vt 0.796021 0.176969
+vt 0.783193 0.187449
+vt 0.233625 0.175620
+vt 0.246955 0.187075
+vt 0.391039 0.611891
+vt 0.394766 0.686125
+vt 0.369913 0.610196
+vt 0.364838 0.684445
+vt 0.391747 0.862097
+vt 0.401605 0.841460
+vt 0.354026 0.840297
+vt 0.363377 0.861308
+vt 0.435018 0.718280
+vt 0.323658 0.715731
+vt 0.433669 0.729661
+vt 0.384658 0.710299
+vt 0.374400 0.708969
+vt 0.324726 0.727177
+vt 0.410995 0.747662
+vt 0.427812 0.742828
+vt 0.347028 0.745816
+vt 0.330270 0.740536
+vt 0.418086 0.784946
+vt 0.384657 0.795423
+vt 0.372270 0.794472
+vt 0.338952 0.783073
+vt 0.431333 0.817535
+vt 0.324790 0.815460
+vt 0.816266 0.203086
+vt 0.825107 0.209762
+vt 0.199767 0.214827
+vt 0.209828 0.206161
+vt 0.802192 0.184609
+vt 0.226485 0.183086
+vt 0.448505 0.804621
+vt 0.473386 0.824700
+vt 0.307886 0.802031
+vt 0.282357 0.821525
+vt 0.435868 0.779569
+vt 0.321237 0.777208
+vt 0.423718 0.754191
+vt 0.334089 0.752045
+vt 0.437950 0.749777
+vt 0.319919 0.747250
+vt 0.445392 0.731997
+vt 0.312907 0.729222
+vt 0.440995 0.724383
+vt 0.317510 0.721697
+vt 0.455277 0.713731
+vt 0.303460 0.710657
+vt 0.512485 0.828811
+vt 0.242975 0.824574
+vt 0.550942 0.811814
+vt 0.204839 0.806417
+vt 0.552139 0.787682
+vt 0.204331 0.782156
+vt 0.539407 0.764539
+vt 0.217774 0.759319
+vt 0.508439 0.743135
+vt 0.249419 0.738732
+vt 0.470841 0.748408
+vt 0.454776 0.761665
+vt 0.286960 0.745020
+vt 0.302729 0.758742
+vt 0.488870 0.770464
+vt 0.475403 0.783904
+vt 0.268291 0.766661
+vt 0.281439 0.780511
+vt 0.503673 0.787562
+vt 0.494476 0.802470
+vt 0.252972 0.783410
+vt 0.261790 0.798626
+vt 0.518562 0.791602
+vt 0.516802 0.807339
+vt 0.237920 0.787045
+vt 0.239243 0.802891
+vt 0.484068 0.628776
+vt 0.543385 0.683538
+vt 0.276936 0.625067
+vt 0.216123 0.678120
+vt 0.581052 0.726933
+vt 0.177176 0.720426
+vt 0.616701 0.759965
+vt 0.140379 0.752377
+vt 0.707492 0.759884
+vt 0.660647 0.741167
+vt 0.049526 0.748824
+vt 0.097038 0.732052
+vt 0.745511 0.652100
+vt 0.677256 0.670436
+vt 0.019409 0.639749
+vt 0.083564 0.662038
+vt 0.740843 0.572428
+vt 0.671403 0.592656
+vt 0.033664 0.564403
+vt 0.092820 0.589862
+vt 0.834578 0.206879
+vt 0.834705 0.206959
+vt 0.051216 0.522659
+vt 0.145041 0.562595
+vt 0.620420 0.565675
+vt 0.498072 0.552315
+vt 0.264218 0.550140
+vn 0.6650 -0.2008 0.7194
+vn -0.6650 -0.2008 0.7194
+vn 0.8294 -0.3036 0.4689
+vn -0.8294 -0.3036 0.4689
+vn 0.4155 -0.7933 0.4449
+vn -0.4155 -0.7933 0.4449
+vn 0.3600 -0.5089 0.7820
+vn -0.3600 -0.5089 0.7820
+vn -0.0787 -0.5394 0.8384
+vn 0.0787 -0.5394 0.8384
+vn -0.2696 -0.8413 0.4685
+vn 0.2696 -0.8413 0.4685
+vn -0.7707 -0.3352 0.5420
+vn 0.7707 -0.3352 0.5420
+vn -0.4689 -0.1940 0.8617
+vn 0.4689 -0.1940 0.8617
+vn -0.4767 0.1907 0.8581
+vn 0.4767 0.1907 0.8581
+vn -0.7672 0.3264 0.5521
+vn 0.7672 0.3264 0.5521
+vn -0.2519 0.8173 0.5182
+vn 0.2519 0.8173 0.5182
+vn -0.0949 0.5696 0.8164
+vn 0.0949 0.5696 0.8164
+vn 0.3667 0.5370 0.7597
+vn -0.3667 0.5370 0.7597
+vn 0.4141 0.7672 0.4898
+vn -0.4141 0.7672 0.4898
+vn 0.8277 0.2952 0.4771
+vn -0.8277 0.2952 0.4771
+vn 0.6713 0.1971 0.7145
+vn -0.6713 0.1971 0.7145
+vn 0.8111 0.3244 -0.4867
+vn -0.8111 0.3244 -0.4867
+vn 0.2052 0.8206 -0.5334
+vn -0.2052 0.8206 -0.5334
+vn -0.4223 0.7806 -0.4607
+vn 0.4223 0.7806 -0.4607
+vn -0.8241 0.3225 -0.4658
+vn 0.8241 0.3225 -0.4658
+vn -0.8137 -0.3487 -0.4650
+vn 0.8137 -0.3487 -0.4650
+vn -0.4223 -0.7806 -0.4607
+vn 0.4223 -0.7806 -0.4607
+vn 0.2052 -0.8206 -0.5334
+vn -0.2052 -0.8206 -0.5334
+vn 0.7995 -0.3510 -0.4875
+vn -0.7995 -0.3510 -0.4875
+vn 0.4000 -0.0623 0.9144
+vn -0.4000 -0.0623 0.9144
+vn 0.3069 -0.1754 0.9354
+vn -0.3069 -0.1754 0.9354
+vn 0.0945 -0.1835 0.9785
+vn -0.0945 -0.1835 0.9785
+vn -0.0624 -0.0283 0.9977
+vn 0.0624 -0.0283 0.9977
+vn -0.0624 0.0260 0.9977
+vn 0.0624 0.0260 0.9977
+vn 0.0996 0.1729 0.9799
+vn -0.0996 0.1729 0.9799
+vn 0.3036 0.1656 0.9383
+vn -0.3036 0.1656 0.9383
+vn 0.4002 0.0572 0.9147
+vn -0.4002 0.0572 0.9147
+vn 0.1231 -0.8616 0.4924
+vn -0.1231 -0.8616 0.4924
+vn 0.2190 -0.8647 0.4520
+vn -0.2190 -0.8647 0.4520
+vn 0.5902 -0.4550 0.6668
+vn -0.5902 -0.4550 0.6668
+vn 0.7689 -0.0506 0.6374
+vn -0.7689 -0.0506 0.6374
+vn 0.7796 0.0900 0.6197
+vn -0.7796 0.0900 0.6197
+vn 0.3241 -0.8188 0.4739
+vn -0.3241 -0.8188 0.4739
+vn 0.3857 -0.6629 0.6417
+vn -0.3857 -0.6629 0.6417
+vn 0.6895 -0.4193 0.5906
+vn -0.6895 -0.4193 0.5906
+vn 0.6588 -0.3634 0.6588
+vn -0.6588 -0.3634 0.6588
+vn 0.5465 0.3707 0.7509
+vn -0.5465 0.3707 0.7509
+vn 0.5064 0.6464 0.5706
+vn -0.5064 0.6464 0.5706
+vn 0.6092 0.5167 0.6015
+vn -0.6092 0.5167 0.6015
+vn -0.0441 0.6610 0.7491
+vn 0.0441 0.6610 0.7491
+vn -0.7246 0.3187 0.6110
+vn 0.7246 0.3187 0.6110
+vn -0.5880 0.5554 0.5880
+vn 0.5880 0.5554 0.5880
+vn 0.5361 -0.3909 0.7482
+vn -0.5361 -0.3909 0.7482
+vn 0.2207 -0.4690 0.8552
+vn -0.2207 -0.4690 0.8552
+vn -0.0794 -0.5321 0.8429
+vn 0.0794 -0.5321 0.8429
+vn -0.0825 -0.6575 0.7490
+vn 0.0825 -0.6575 0.7490
+vn 0.0457 -0.5667 0.8226
+vn -0.0457 -0.5667 0.8226
+vn 0.2784 -0.2130 0.9365
+vn -0.2784 -0.2130 0.9365
+vn 0.3813 -0.1824 0.9063
+vn -0.3813 -0.1824 0.9063
+vn 0.3357 -0.2878 0.8969
+vn -0.3357 -0.2878 0.8969
+vn 0.3762 0.0603 0.9246
+vn -0.3762 0.0603 0.9246
+vn -0.1352 0.2680 0.9539
+vn 0.1352 0.2680 0.9539
+vn 0.3961 -0.4321 0.8102
+vn -0.3961 -0.4321 0.8102
+vn 0.1856 -0.2474 0.9510
+vn -0.1856 -0.2474 0.9510
+vn 0.0099 -0.1948 0.9808
+vn -0.0099 -0.1948 0.9808
+vn 0.0721 -0.6966 0.7138
+vn -0.0721 -0.6966 0.7138
+vn 0.1863 -0.5723 0.7986
+vn -0.1863 -0.5723 0.7986
+vn 0.3157 -0.2708 0.9094
+vn -0.3157 -0.2708 0.9094
+vn 0.3063 -0.0265 0.9516
+vn -0.3063 -0.0265 0.9516
+vn 0.3266 -0.1306 0.9361
+vn -0.3266 -0.1306 0.9361
+vn -0.0137 0.0574 0.9983
+vn 0.0137 0.0574 0.9983
+vn -0.0026 -0.0656 0.9978
+vn 0.0026 -0.0656 0.9978
+vn 0.0000 0.0000 1.0000
+vn 0.8174 -0.5744 -0.0442
+vn -0.8174 -0.5744 -0.0442
+vn 0.9494 0.2297 -0.2144
+vn -0.9494 0.2297 -0.2144
+vn 0.0825 0.9073 -0.4124
+vn -0.0825 0.9073 -0.4124
+vn -0.8836 0.3555 0.3047
+vn 0.8836 0.3555 0.3047
+vn 0.4207 -0.8797 0.2218
+vn -0.4207 -0.8797 0.2218
+vn 0.2873 -0.5747 0.7663
+vn -0.2873 -0.5747 0.7663
+vn -0.6542 0.6019 0.4580
+vn 0.6542 0.6019 0.4580
+vn 0.1052 0.7892 0.6051
+vn -0.1052 0.7892 0.6051
+vn 0.7582 0.2916 0.5832
+vn -0.7582 0.2916 0.5832
+vn 0.3889 -0.7130 0.5834
+vn -0.3889 -0.7130 0.5834
+vn 0.0463 0.2314 0.9718
+vn -0.0463 0.2314 0.9718
+vn 0.0335 -0.4018 0.9151
+vn -0.0335 -0.4018 0.9151
+vn -0.4452 -0.1610 0.8809
+vn 0.4452 -0.1610 0.8809
+vn -0.2182 -0.4364 0.8729
+vn 0.2182 -0.4364 0.8729
+vn 0.4341 -0.1290 0.8916
+vn -0.4341 -0.1290 0.8916
+vn 0.3008 0.0501 0.9524
+vn -0.3008 0.0501 0.9524
+vn 0.8123 0.3010 0.4996
+vn -0.8123 0.3010 0.4996
+vn 0.8753 0.2574 0.4093
+vn -0.8753 0.2574 0.4093
+vn 0.9385 0.1601 0.3060
+vn -0.9385 0.1601 0.3060
+vn 0.2237 -0.6539 0.7227
+vn -0.2237 -0.6539 0.7227
+vn -0.1536 -0.1997 0.9677
+vn 0.1536 -0.1997 0.9677
+vn -0.2733 -0.1025 0.9565
+vn 0.2733 -0.1025 0.9565
+vn -0.0976 0.1952 0.9759
+vn 0.0976 0.1952 0.9759
+vn -0.1582 0.9494 0.2713
+vn 0.1582 0.9494 0.2713
+vn -0.6934 0.7082 0.1328
+vn 0.6934 0.7082 0.1328
+vn -1.0000 0.0000 0.0000
+vn 1.0000 0.0000 0.0000
+vn 0.3051 -0.9450 0.1181
+vn -0.3051 -0.9450 0.1181
+vn 0.0298 -0.2981 0.9541
+vn -0.0298 -0.2981 0.9541
+vn 0.1353 -0.3479 0.9277
+vn -0.1353 -0.3479 0.9277
+vn -0.5085 -0.2755 0.8158
+vn 0.5085 -0.2755 0.8158
+vn -0.3843 -0.0419 0.9223
+vn 0.3843 -0.0419 0.9223
+vn -0.2083 0.0374 0.9774
+vn 0.2083 0.0374 0.9774
+vn -0.5721 -0.4767 0.6674
+vn 0.5721 -0.4767 0.6674
+vn -0.1369 -0.7531 0.6435
+vn 0.1369 -0.7531 0.6435
+vn 0.4088 -0.6071 0.6814
+vn -0.4088 -0.6071 0.6814
+vn 0.5740 -0.4130 0.7070
+vn -0.5740 -0.4130 0.7070
+vn 0.5665 -0.0968 0.8183
+vn -0.5665 -0.0968 0.8183
+vn 0.5703 0.1180 0.8129
+vn -0.5703 0.1180 0.8129
+vn 0.4823 0.5621 0.6719
+vn -0.4823 0.5621 0.6719
+vn 0.2604 0.6114 0.7473
+vn -0.2604 0.6114 0.7473
+vn 0.1640 0.3607 0.9182
+vn -0.1640 0.3607 0.9182
+vn -0.0178 0.2495 0.9682
+vn 0.0178 0.2495 0.9682
+vn 0.3273 -0.4166 0.8481
+vn -0.3273 -0.4166 0.8481
+vn 0.2811 -0.2610 0.9235
+vn -0.2811 -0.2610 0.9235
+vn -0.2542 -0.6514 0.7149
+vn 0.2542 -0.6514 0.7149
+vn -0.0260 -0.8455 0.5333
+vn 0.0260 -0.8455 0.5333
+vn -0.3518 -0.2606 0.8991
+vn 0.3518 -0.2606 0.8991
+vn -0.3523 -0.0110 0.9358
+vn 0.3523 -0.0110 0.9358
+vn -0.1317 0.4608 0.8777
+vn 0.1317 0.4608 0.8777
+vn -0.0342 0.6159 0.7870
+vn 0.0342 0.6159 0.7870
+vn 0.3603 0.5836 0.7277
+vn -0.3603 0.5836 0.7277
+vn 0.4988 0.5300 0.6858
+vn -0.4988 0.5300 0.6858
+vn 0.6667 -0.3333 0.6667
+vn -0.6667 -0.3333 0.6667
+vn 0.8165 -0.0731 0.5727
+vn -0.8165 -0.0731 0.5727
+vn 0.7840 0.1161 0.6098
+vn -0.7840 0.1161 0.6098
+vn -0.5306 0.8111 -0.2461
+vn 0.5306 0.8111 -0.2461
+vn -0.8511 0.3695 -0.3730
+vn 0.8511 0.3695 -0.3730
+vn -0.2446 0.8675 -0.4331
+vn 0.2446 0.8675 -0.4331
+vn 0.5924 0.7465 -0.3030
+vn -0.5924 0.7465 -0.3030
+vn 0.3685 0.8758 -0.3118
+vn -0.3685 0.8758 -0.3118
+vn 0.2821 0.9151 -0.2880
+vn -0.2821 0.9151 -0.2880
+vn 0.8561 0.1340 -0.4991
+vn -0.8561 0.1340 -0.4991
+vn 0.5342 -0.7233 -0.4376
+vn -0.5342 -0.7233 -0.4376
+vn 0.3849 -0.8131 -0.4368
+vn -0.3849 -0.8131 -0.4368
+vn 0.2335 -0.5806 -0.7800
+vn -0.2335 -0.5806 -0.7800
+vn 0.2449 -0.0583 -0.9678
+vn -0.2449 -0.0583 -0.9678
+vn 0.1163 -0.4535 -0.8837
+vn -0.1163 -0.4535 -0.8837
+vn 0.1152 -0.9836 -0.1388
+vn -0.1152 -0.9836 -0.1388
+vn 0.1184 -0.9669 -0.2260
+vn -0.1184 -0.9669 -0.2260
+vn 0.9597 -0.0085 -0.2808
+vn -0.9597 -0.0085 -0.2808
+vn 0.9319 0.1629 -0.3242
+vn -0.9319 0.1629 -0.3242
+vn 0.1626 0.0207 -0.9865
+vn -0.1626 0.0207 -0.9865
+vn -0.0188 -0.2177 -0.9758
+vn 0.0188 -0.2177 -0.9758
+vn 0.7538 -0.2926 -0.5884
+vn -0.7538 -0.2926 -0.5884
+vn 0.9196 0.1379 -0.3678
+vn -0.9196 0.1379 -0.3678
+vn 0.9297 0.3127 -0.1944
+vn -0.9297 0.3127 -0.1944
+vn 0.9120 0.3376 -0.2329
+vn -0.9120 0.3376 -0.2329
+vn 0.9407 0.3338 -0.0607
+vn -0.9407 0.3338 -0.0607
+vn 0.1761 -0.8805 -0.4402
+vn -0.1761 -0.8805 -0.4402
+vn 0.3708 -0.4733 -0.7991
+vn -0.3708 -0.4733 -0.7991
+vn 0.3107 -0.8284 -0.4660
+vn -0.3107 -0.8284 -0.4660
+vn 0.2793 -0.9515 -0.1287
+vn -0.2793 -0.9515 -0.1287
+vn 0.3139 -0.9321 -0.1807
+vn -0.3139 -0.9321 -0.1807
+vn 0.9762 -0.2083 -0.0609
+vn -0.9762 -0.2083 -0.0609
+vn 0.8267 -0.5066 0.2447
+vn -0.8267 -0.5066 0.2447
+vn 0.3449 -0.1158 -0.9315
+vn -0.3449 -0.1158 -0.9315
+vn 0.1203 0.9644 0.2355
+vn -0.1203 0.9644 0.2355
+vn 0.1275 0.9744 -0.1851
+vn -0.1275 0.9744 -0.1851
+vn 0.3492 0.5947 -0.7241
+vn -0.3492 0.5947 -0.7241
+vn 0.4153 0.8981 -0.1449
+vn -0.4153 0.8981 -0.1449
+vn 0.1845 0.7036 0.6863
+vn -0.1845 0.7036 0.6863
+vn 0.6056 0.7794 0.1608
+vn -0.6056 0.7794 0.1608
+vn 0.7033 0.6806 -0.2053
+vn -0.7033 0.6806 -0.2053
+vn 0.6679 0.2007 -0.7166
+vn -0.6679 0.2007 -0.7166
+vn 0.4948 0.4342 -0.7528
+vn -0.4948 0.4342 -0.7528
+vn 0.6423 0.7459 -0.1761
+vn -0.6423 0.7459 -0.1761
+vn 0.7182 0.6788 0.1530
+vn -0.7182 0.6788 0.1530
+vn 0.7388 0.3972 0.5444
+vn -0.7388 0.3972 0.5444
+vn 0.3428 0.9261 -0.1579
+vn -0.3428 0.9261 -0.1579
+vn 0.2270 0.5740 0.7867
+vn -0.2270 0.5740 0.7867
+vn -0.1722 0.1046 -0.9795
+vn 0.1722 0.1046 -0.9795
+vn 0.0425 0.9150 0.4013
+vn -0.0425 0.9150 0.4013
+vn -0.1616 0.1847 0.9694
+vn 0.1616 0.1847 0.9694
+vn 0.9791 0.1973 0.0483
+vn -0.9791 0.1973 0.0483
+vn 0.9470 0.0918 0.3079
+vn -0.9470 0.0918 0.3079
+vn 0.9794 0.1905 -0.0661
+vn -0.9794 0.1905 -0.0661
+vn 0.9938 0.0312 -0.1070
+vn -0.9938 0.0312 -0.1070
+vn 0.7116 -0.7008 0.0501
+vn -0.7116 -0.7008 0.0501
+vn 0.3722 -0.9243 0.0847
+vn -0.3722 -0.9243 0.0847
+vn 0.4465 -0.8644 0.2310
+vn -0.4465 -0.8644 0.2310
+vn 0.6066 -0.7578 0.2405
+vn -0.6066 -0.7578 0.2405
+vn 0.7325 -0.6368 0.2407
+vn -0.7325 -0.6368 0.2407
+vn 0.2637 -0.4499 0.8533
+vn -0.2637 -0.4499 0.8533
+vn 0.5568 -0.3181 -0.7673
+vn -0.5568 -0.3181 -0.7673
+vn 0.5004 -0.2807 -0.8190
+vn -0.5004 -0.2807 -0.8190
+vn 0.3190 -0.8494 -0.4205
+vn -0.3190 -0.8494 -0.4205
+vn 0.7198 -0.6356 -0.2793
+vn -0.7198 -0.6356 -0.2793
+vn 0.4972 -0.4408 -0.7473
+vn -0.4972 -0.4408 -0.7473
+vn 0.3506 0.3807 0.8557
+vn -0.3506 0.3807 0.8557
+vn 0.4566 0.1715 0.8730
+vn -0.4566 0.1715 0.8730
+vn 0.2583 0.1055 0.9603
+vn -0.2583 0.1055 0.9603
+vn 0.2455 -0.0802 0.9661
+vn -0.2455 -0.0802 0.9661
+vn 0.4643 -0.0599 0.8837
+vn -0.4643 -0.0599 0.8837
+vn 0.6225 -0.3045 0.7210
+vn -0.6225 -0.3045 0.7210
+vn 0.4500 0.6590 0.6027
+vn -0.4500 0.6590 0.6027
+vn -0.2667 0.8309 0.4884
+vn 0.2667 0.8309 0.4884
+vn -0.8284 0.2291 0.5111
+vn 0.8284 0.2291 0.5111
+vn -0.5251 -0.3566 0.7727
+vn 0.5251 -0.3566 0.7727
+vn 0.4546 -0.5665 0.6873
+vn -0.4546 -0.5665 0.6873
+vn 0.6996 -0.4497 0.5552
+vn -0.6996 -0.4497 0.5552
+vn 0.7220 -0.6827 -0.1126
+vn -0.7220 -0.6827 -0.1126
+vn -0.1919 0.2860 0.9388
+vn 0.1919 0.2860 0.9388
+vn 0.9048 -0.3734 -0.2047
+vn -0.9048 -0.3734 -0.2047
+vn 0.1034 0.1551 0.9825
+vn -0.1034 0.1551 0.9825
+vn 0.0841 0.9318 0.3530
+vn -0.0841 0.9318 0.3530
+vn 0.6446 -0.0883 0.7594
+vn -0.6446 -0.0883 0.7594
+vn 0.4309 0.4740 0.7678
+vn -0.4309 0.4740 0.7678
+vn 0.8032 -0.4847 0.3462
+vn -0.8032 -0.4847 0.3462
+vn 0.5811 -0.4128 0.7014
+vn -0.5811 -0.4128 0.7014
+vn 0.5910 -0.4305 0.6822
+vn -0.5910 -0.4305 0.6822
+vn 0.9818 -0.1804 -0.0591
+vn -0.9818 -0.1804 -0.0591
+vn 0.9105 -0.3965 -0.1175
+vn -0.9105 -0.3965 -0.1175
+vn 0.9972 -0.0181 -0.0725
+vn -0.9972 -0.0181 -0.0725
+vn 0.7313 -0.6543 0.1925
+vn -0.7313 -0.6543 0.1925
+vn 0.7867 -0.6079 0.1073
+vn -0.7867 -0.6079 0.1073
+vn 0.7022 -0.7022 0.1170
+vn -0.7022 -0.7022 0.1170
+vn 0.1840 0.9816 -0.0511
+vn -0.1840 0.9816 -0.0511
+vn 0.9352 0.3301 0.1284
+vn -0.9352 0.3301 0.1284
+vn 0.6633 -0.7463 0.0553
+vn -0.6633 -0.7463 0.0553
+vn -0.0085 0.9970 0.0767
+vn 0.0085 0.9970 0.0767
+vn 0.6237 -0.7061 0.3354
+vn -0.6237 -0.7061 0.3354
+vn 0.2733 -0.8925 0.3587
+vn -0.2733 -0.8925 0.3587
+vn -0.8328 -0.5080 -0.2200
+vn 0.8328 -0.5080 -0.2200
+vn -0.8339 0.2377 -0.4981
+vn 0.8339 0.2377 -0.4981
+vn -0.5655 0.7847 -0.2539
+vn 0.5655 0.7847 -0.2539
+vn -0.0560 0.9962 0.0672
+vn 0.0560 0.9962 0.0672
+vn 0.1445 0.0222 0.9893
+vn -0.1445 0.0222 0.9893
+vn 0.3275 0.0645 0.9427
+vn -0.3275 0.0645 0.9427
+vn 0.3127 0.0232 0.9496
+vn -0.3127 0.0232 0.9496
+vn 0.1710 0.0274 0.9849
+vn -0.1710 0.0274 0.9849
+vn 0.3487 0.2849 0.8929
+vn -0.3487 0.2849 0.8929
+vn 0.4006 -0.0343 0.9156
+vn -0.4006 -0.0343 0.9156
+vn 0.2572 -0.0603 0.9645
+vn -0.2572 -0.0603 0.9645
+vn 0.0637 -0.0106 0.9979
+vn -0.0637 -0.0106 0.9979
+vn -0.3637 0.7039 0.6101
+vn 0.3637 0.7039 0.6101
+vn 0.6299 0.0355 0.7759
+vn -0.6299 0.0355 0.7759
+vn 0.4472 -0.2002 0.8717
+vn -0.4472 -0.2002 0.8717
+vn 0.5072 -0.2141 0.8348
+vn -0.5072 -0.2141 0.8348
+vn 0.5258 0.2619 0.8093
+vn -0.5258 0.2619 0.8093
+vn 0.2980 0.5802 0.7580
+vn -0.2980 0.5802 0.7580
+vn 0.0930 -0.9924 -0.0805
+vn -0.0930 -0.9924 -0.0805
+vn 0.5006 -0.8657 0.0080
+vn -0.5006 -0.8657 0.0080
+vn 0.9285 -0.2497 0.2748
+vn -0.9285 -0.2497 0.2748
+vn 0.8393 0.5424 -0.0378
+vn -0.8393 0.5424 -0.0378
+vn -0.2355 0.9367 -0.2589
+vn 0.2355 0.9367 -0.2589
+vn -0.4499 0.8838 -0.1285
+vn 0.4499 0.8838 -0.1285
+vn -0.5384 -0.0098 -0.8427
+vn 0.5384 -0.0098 -0.8427
+vn -0.1910 -0.0241 -0.9813
+vn 0.1910 -0.0241 -0.9813
+vn 0.4046 0.0266 -0.9141
+vn -0.4046 0.0266 -0.9141
+vn -0.7819 0.6231 0.0197
+vn 0.7819 0.6231 0.0197
+vn 0.5428 -0.2063 -0.8142
+vn -0.5428 -0.2063 -0.8142
+vn -0.2474 -0.9231 -0.2945
+vn 0.2474 -0.9231 -0.2945
+usemtl None
+s off
+f 47/1/1 1/2/1 3/3/1 45/4/1
+f 4/5/2 2/6/2 48/7/2 46/8/2
+f 45/4/3 3/3/3 5/9/3 43/10/3
+f 6/11/4 4/5/4 46/8/4 44/12/4
+f 3/3/5 9/13/5 7/14/5 5/9/5
+f 8/15/6 10/16/6 4/5/6 6/11/6
+f 1/2/7 11/17/7 9/13/7 3/3/7
+f 10/16/8 12/18/8 2/6/8 4/5/8
+f 11/17/9 13/19/9 15/20/9 9/13/9
+f 16/21/10 14/22/10 12/18/10 10/16/10
+f 9/13/11 15/20/11 17/23/11 7/14/11
+f 18/24/12 16/21/12 10/16/12 8/15/12
+f 15/20/13 21/25/13 19/26/13 17/23/13
+f 20/27/14 22/28/14 16/21/14 18/24/14
+f 13/19/15 23/29/15 21/25/15 15/20/15
+f 22/28/16 24/30/16 14/22/16 16/21/16
+f 23/29/17 25/31/17 27/32/17 21/25/17
+f 28/33/18 26/34/18 24/30/18 22/28/18
+f 21/25/19 27/32/19 29/35/19 19/26/19
+f 30/36/20 28/33/20 22/28/20 20/27/20
+f 27/32/21 33/37/21 31/38/21 29/35/21
+f 32/39/22 34/40/22 28/33/22 30/36/22
+f 25/31/23 35/41/23 33/37/23 27/32/23
+f 34/40/24 36/42/24 26/34/24 28/33/24
+f 35/41/25 37/43/25 39/44/25 33/37/25
+f 40/45/26 38/46/26 36/42/26 34/40/26
+f 33/37/27 39/44/27 41/47/27 31/38/27
+f 42/48/28 40/45/28 34/40/28 32/39/28
+f 39/44/29 45/4/29 43/10/29 41/47/29
+f 44/12/30 46/8/30 40/45/30 42/48/30
+f 37/43/31 47/1/31 45/4/31 39/44/31
+f 46/8/32 48/7/32 38/46/32 40/45/32
+f 47/1/33 37/43/33 51/49/33 49/50/33
+f 52/51/34 38/46/34 48/7/34 50/52/34
+f 37/43/35 35/41/35 53/53/35 51/49/35
+f 54/54/36 36/42/36 38/46/36 52/51/36
+f 35/41/37 25/31/37 55/55/37 53/53/37
+f 56/56/38 26/34/38 36/42/38 54/54/38
+f 25/31/39 23/29/39 57/57/39 55/55/39
+f 58/58/40 24/30/40 26/34/40 56/56/40
+f 23/29/41 13/19/41 59/59/41 57/57/41
+f 60/60/42 14/22/42 24/30/42 58/58/42
+f 13/19/43 11/17/43 63/61/43 59/59/43
+f 64/62/44 12/18/44 14/22/44 60/60/44
+f 11/17/45 1/2/45 65/63/45 63/61/45
+f 66/64/46 2/6/46 12/18/46 64/62/46
+f 1/2/47 47/1/47 49/50/47 65/63/47
+f 50/52/48 48/7/48 2/6/48 66/64/48
+f 61/65/49 65/63/49 49/50/49
+f 50/52/50 66/64/50 62/66/50
+f 63/61/51 65/63/51 61/65/51
+f 62/66/52 66/64/52 64/62/52
+f 61/65/53 59/59/53 63/61/53
+f 64/62/54 60/60/54 62/66/54
+f 61/65/55 57/57/55 59/59/55
+f 60/60/56 58/58/56 62/66/56
+f 61/65/57 55/55/57 57/57/57
+f 58/58/58 56/56/58 62/66/58
+f 61/65/59 53/53/59 55/55/59
+f 56/56/60 54/54/60 62/66/60
+f 61/65/61 51/49/61 53/53/61
+f 54/54/62 52/51/62 62/66/62
+f 61/65/63 49/50/63 51/49/63
+f 52/51/64 50/52/64 62/66/64
+f 89/67/65 174/68/65 176/69/65 91/70/65
+f 176/69/66 175/71/66 90/72/66 91/70/66
+f 87/73/67 172/74/67 174/68/67 89/67/67
+f 175/71/68 173/75/68 88/76/68 90/72/68
+f 85/77/69 170/78/69 172/74/69 87/73/69
+f 173/75/70 171/79/70 86/80/70 88/76/70
+f 83/81/71 168/82/71 170/78/71 85/77/71
+f 171/79/72 169/83/72 84/84/72 86/80/72
+f 81/85/73 166/86/73 168/82/73 83/81/73
+f 169/83/74 167/87/74 82/88/74 84/84/74
+f 79/89/75 92/90/75 146/91/75 164/92/75
+f 147/93/76 93/94/76 80/95/76 165/96/76
+f 92/90/77 94/97/77 148/98/77 146/91/77
+f 149/99/78 95/100/78 93/94/78 147/93/78
+f 94/97/79 96/101/79 150/102/79 148/98/79
+f 151/103/80 97/104/80 95/100/80 149/99/80
+f 96/101/81 98/105/81 152/106/81 150/102/81
+f 153/107/82 99/108/82 97/104/82 151/103/82
+f 98/105/83 100/109/83 154/110/83 152/106/83
+f 155/111/84 101/112/84 99/108/84 153/107/84
+f 100/109/85 102/113/85 156/114/85 154/110/85
+f 157/115/86 103/116/86 101/112/86 155/111/86
+f 102/113/87 104/117/87 158/118/87 156/114/87
+f 159/119/88 105/120/88 103/116/88 157/115/88
+f 104/117/89 106/121/89 160/122/89 158/118/89
+f 161/123/90 107/124/90 105/120/90 159/119/90
+f 106/121/91 108/125/91 162/126/91 160/122/91
+f 163/127/92 109/128/92 107/124/92 161/123/92
+f 108/125/93 67/129/93 68/130/93 162/126/93
+f 68/130/94 67/129/94 109/128/94 163/127/94
+f 110/131/95 128/132/95 160/122/95 162/126/95
+f 161/123/96 129/133/96 111/134/96 163/127/96
+f 128/132/97 179/135/97 158/118/97 160/122/97
+f 159/119/98 180/136/98 129/133/98 161/123/98
+f 126/137/99 156/114/99 158/118/99 179/135/99
+f 159/119/100 157/115/100 127/138/100 180/136/100
+f 124/139/101 154/110/101 156/114/101 126/137/101
+f 157/115/102 155/111/102 125/140/102 127/138/102
+f 122/141/103 152/106/103 154/110/103 124/139/103
+f 155/111/104 153/107/104 123/142/104 125/140/104
+f 120/143/105 150/102/105 152/106/105 122/141/105
+f 153/107/106 151/103/106 121/144/106 123/142/106
+f 118/145/107 148/98/107 150/102/107 120/143/107
+f 151/103/108 149/99/108 119/146/108 121/144/108
+f 116/147/109 146/91/109 148/98/109 118/145/109
+f 149/99/110 147/93/110 117/148/110 119/146/110
+f 114/149/111 164/92/111 146/91/111 116/147/111
+f 147/93/112 165/96/112 115/150/112 117/148/112
+f 114/149/113 181/151/113 177/152/113 164/92/113
+f 177/152/114 182/153/114 115/150/114 165/96/114
+f 110/131/115 162/126/115 68/130/115 112/154/115
+f 68/130/116 163/127/116 111/134/116 113/155/116
+f 112/154/117 68/130/117 178/156/117 183/157/117
+f 178/156/118 68/130/118 113/155/118 184/158/118
+f 177/152/119 181/151/119 183/157/119 178/156/119
+f 184/158/120 182/153/120 177/152/120 178/156/120
+f 135/159/121 137/160/121 176/69/121 174/68/121
+f 176/69/122 137/160/122 136/161/122 175/71/122
+f 133/162/123 135/159/123 174/68/123 172/74/123
+f 175/71/124 136/161/124 134/163/124 173/75/124
+f 131/164/125 133/162/125 172/74/125 170/78/125
+f 173/75/126 134/163/126 132/165/126 171/79/126
+f 166/86/127 187/166/127 185/167/127 168/82/127
+f 186/168/128 188/169/128 167/87/128 169/83/128
+f 131/164/129 170/78/129 168/82/129 185/167/129
+f 169/83/130 171/79/130 132/165/130 186/168/130
+f 144/170/131 190/171/131 189/172/131 187/166/131
+f 189/172/132 190/171/132 145/173/132 188/169/132
+f 185/167/133 187/166/133 189/172/133 69/174/133
+f 189/172/134 188/169/134 186/168/134 69/174/134
+f 130/175/135 131/164/135 185/167/135 69/174/135
+f 186/168/135 132/165/135 130/175/135 69/174/135
+f 142/176/136 193/177/136 191/178/136 144/170/136
+f 192/179/137 194/180/137 143/181/137 145/173/137
+f 140/182/138 195/183/138 193/177/138 142/176/138
+f 194/180/139 196/184/139 141/185/139 143/181/139
+f 139/186/140 197/187/140 195/183/140 140/182/140
+f 196/184/141 198/188/141 139/186/141 141/185/141
+f 138/189/142 71/190/142 197/187/142 139/186/142
+f 198/188/143 71/190/143 138/189/143 139/186/143
+f 190/171/144 144/170/144 191/178/144 70/191/144
+f 192/179/145 145/173/145 190/171/145 70/191/145
+f 70/191/146 191/178/146 206/192/146 208/193/146
+f 207/194/147 192/179/147 70/191/147 208/193/147
+f 71/190/148 199/195/148 200/196/148 197/187/148
+f 201/197/149 199/195/149 71/190/149 198/188/149
+f 197/187/150 200/196/150 202/198/150 195/183/150
+f 203/199/151 201/197/151 198/188/151 196/184/151
+f 195/183/152 202/198/152 204/200/152 193/177/152
+f 205/201/153 203/199/153 196/184/153 194/180/153
+f 193/177/154 204/200/154 206/192/154 191/178/154
+f 207/194/155 205/201/155 194/180/155 192/179/155
+f 199/195/156 204/200/156 202/198/156 200/196/156
+f 203/199/157 205/201/157 199/195/157 201/197/157
+f 199/195/158 208/193/158 206/192/158 204/200/158
+f 207/194/159 208/193/159 199/195/159 205/201/159
+f 139/186/160 140/182/160 164/92/160 177/152/160
+f 165/96/161 141/185/161 139/186/161 177/152/161
+f 140/182/162 142/176/162 211/202/162 164/92/162
+f 212/203/163 143/181/163 141/185/163 165/96/163
+f 142/176/164 144/170/164 213/204/164 211/202/164
+f 214/205/165 145/173/165 143/181/165 212/203/165
+f 144/170/166 187/166/166 166/86/166 213/204/166
+f 167/87/167 188/169/167 145/173/167 214/205/167
+f 81/85/168 209/206/168 213/204/168 166/86/168
+f 214/205/169 210/207/169 82/88/169 167/87/169
+f 209/206/170 215/208/170 211/202/170 213/204/170
+f 212/203/171 216/209/171 210/207/171 214/205/171
+f 79/89/172 164/92/172 211/202/172 215/208/172
+f 212/203/173 165/96/173 80/95/173 216/209/173
+f 131/164/174 130/175/174 72/210/174 222/211/174
+f 72/210/175 130/175/175 132/165/175 223/212/175
+f 133/162/176 131/164/176 222/211/176 220/213/176
+f 223/212/177 132/165/177 134/163/177 221/214/177
+f 135/159/178 133/162/178 220/213/178 218/215/178
+f 221/214/179 134/163/179 136/161/179 219/216/179
+f 137/160/180 135/159/180 218/215/180 217/217/180
+f 219/216/181 136/161/181 137/160/181 217/217/181
+f 217/217/182 218/215/182 229/218/182 231/219/182
+f 230/220/183 219/216/183 217/217/183 231/219/183
+f 218/215/184 220/213/184 227/221/184 229/218/184
+f 228/222/185 221/214/185 219/216/185 230/220/185
+f 220/213/186 222/211/186 225/223/186 227/221/186
+f 226/224/187 223/212/187 221/214/187 228/222/187
+f 222/211/188 72/210/188 224/225/188 225/223/188
+f 224/225/189 72/210/189 223/212/189 226/224/189
+f 224/225/190 231/219/190 229/218/190 225/223/190
+f 230/220/191 231/219/191 224/225/191 226/224/191
+f 225/223/192 229/218/192 227/221/192
+f 228/222/193 230/220/193 226/224/193
+f 183/157/194 181/151/194 234/226/194 232/227/194
+f 235/228/195 182/153/195 184/158/195 233/229/195
+f 112/154/196 183/157/196 232/227/196 254/230/196
+f 233/229/197 184/158/197 113/155/197 255/231/197
+f 110/131/198 112/154/198 254/230/198 256/232/198
+f 255/231/199 113/155/199 111/134/199 257/233/199
+f 181/151/200 114/149/200 252/234/200 234/226/200
+f 253/235/201 115/150/201 182/153/201 235/228/201
+f 114/149/202 116/147/202 250/236/202 252/234/202
+f 251/237/203 117/148/203 115/150/203 253/235/203
+f 116/147/204 118/145/204 248/238/204 250/236/204
+f 249/239/205 119/146/205 117/148/205 251/237/205
+f 118/145/206 120/143/206 246/240/206 248/238/206
+f 247/241/207 121/144/207 119/146/207 249/239/207
+f 120/143/208 122/141/208 244/242/208 246/240/208
+f 245/243/209 123/142/209 121/144/209 247/241/209
+f 122/141/210 124/139/210 242/244/210 244/242/210
+f 243/245/211 125/140/211 123/142/211 245/243/211
+f 124/139/212 126/137/212 240/246/212 242/244/212
+f 241/247/213 127/138/213 125/140/213 243/245/213
+f 126/137/214 179/135/214 236/248/214 240/246/214
+f 237/249/215 180/136/215 127/138/215 241/247/215
+f 179/135/216 128/132/216 238/250/216 236/248/216
+f 239/251/217 129/133/217 180/136/217 237/249/217
+f 128/132/218 110/131/218 256/232/218 238/250/218
+f 257/233/219 111/134/219 129/133/219 239/251/219
+f 238/250/220 256/232/220 258/252/220 276/253/220
+f 259/254/221 257/233/221 239/251/221 277/255/221
+f 236/248/222 238/250/222 276/253/222 278/256/222
+f 277/255/223 239/251/223 237/249/223 279/257/223
+f 240/246/224 236/248/224 278/256/224 274/258/224
+f 279/257/225 237/249/225 241/247/225 275/259/225
+f 242/244/226 240/246/226 274/258/226 272/260/226
+f 275/259/227 241/247/227 243/245/227 273/261/227
+f 244/242/228 242/244/228 272/260/228 270/262/228
+f 273/261/229 243/245/229 245/243/229 271/263/229
+f 246/240/230 244/242/230 270/262/230 268/264/230
+f 271/263/231 245/243/231 247/241/231 269/265/231
+f 248/238/232 246/240/232 268/264/232 266/266/232
+f 269/265/233 247/241/233 249/239/233 267/267/233
+f 250/236/234 248/238/234 266/266/234 264/268/234
+f 267/267/235 249/239/235 251/237/235 265/269/235
+f 252/234/236 250/236/236 264/268/236 262/270/236
+f 265/269/237 251/237/237 253/235/237 263/271/237
+f 234/226/238 252/234/238 262/270/238 280/272/238
+f 263/271/239 253/235/239 235/228/239 281/273/239
+f 256/232/240 254/230/240 260/274/240 258/252/240
+f 261/275/241 255/231/241 257/233/241 259/254/241
+f 254/230/242 232/227/242 282/276/242 260/274/242
+f 283/277/243 233/229/243 255/231/243 261/275/243
+f 232/227/244 234/226/244 280/272/244 282/276/244
+f 281/273/245 235/228/245 233/229/245 283/277/245
+f 67/129/246 108/125/246 284/278/246 73/279/246
+f 285/280/247 109/128/247 67/129/247 73/279/247
+f 108/125/248 106/121/248 286/281/248 284/278/248
+f 287/282/249 107/124/249 109/128/249 285/280/249
+f 106/121/250 104/117/250 288/283/250 286/281/250
+f 289/284/251 105/120/251 107/124/251 287/282/251
+f 104/117/252 102/113/252 290/285/252 288/283/252
+f 291/286/253 103/116/253 105/120/253 289/284/253
+f 102/113/254 100/109/254 292/287/254 290/285/254
+f 293/288/255 101/112/255 103/116/255 291/286/255
+f 100/109/256 98/105/256 294/289/256 292/287/256
+f 295/290/257 99/108/257 101/112/257 293/288/257
+f 98/105/258 96/101/258 296/291/258 294/289/258
+f 297/292/259 97/104/259 99/108/259 295/290/259
+f 96/101/260 94/97/260 298/293/260 296/291/260
+f 299/294/261 95/100/261 97/104/261 297/292/261
+f 94/97/262 92/90/262 300/295/262 298/293/262
+f 301/296/263 93/94/263 95/100/263 299/294/263
+f 308/297/264 309/298/264 328/299/264 338/300/264
+f 329/301/265 309/302/265 308/303/265 339/304/265
+f 307/305/266 308/297/266 338/300/266 336/306/266
+f 339/304/267 308/303/267 307/307/267 337/308/267
+f 306/309/268 307/305/268 336/306/268 340/310/268
+f 337/308/269 307/307/269 306/309/269 341/311/269
+f 89/67/270 91/70/270 306/309/270 340/310/270
+f 306/309/271 91/70/271 90/72/271 341/311/271
+f 87/73/272 89/67/272 340/310/272 334/312/272
+f 341/311/273 90/72/273 88/76/273 335/313/273
+f 85/77/274 87/73/274 334/312/274 330/314/274
+f 335/313/275 88/76/275 86/80/275 331/315/275
+f 83/81/276 85/77/276 330/314/276 332/316/276
+f 331/315/277 86/80/277 84/84/277 333/317/277
+f 330/314/278 336/306/278 338/300/278 332/316/278
+f 339/304/279 337/308/279 331/315/279 333/317/279
+f 330/314/280 334/312/280 340/310/280 336/306/280
+f 341/311/281 335/313/281 331/315/281 337/308/281
+f 326/318/282 332/316/282 338/300/282 328/299/282
+f 339/304/283 333/317/283 327/319/283 329/301/283
+f 81/85/284 83/81/284 332/316/284 326/318/284
+f 333/317/285 84/84/285 82/88/285 327/319/285
+f 209/206/286 342/320/286 344/321/286 215/208/286
+f 345/322/287 343/323/287 210/207/287 216/209/287
+f 81/85/288 326/318/288 342/320/288 209/206/288
+f 343/323/289 327/319/289 82/88/289 210/207/289
+f 79/89/290 215/208/290 344/321/290 346/324/290
+f 345/322/291 216/209/291 80/95/291 347/325/291
+f 79/89/292 346/324/292 300/295/292 92/90/292
+f 301/296/293 347/325/293 80/95/293 93/94/293
+f 77/326/294 324/327/294 352/328/294 304/329/294
+f 353/330/295 325/331/295 77/332/295 304/333/295
+f 304/329/296 352/328/296 350/334/296 78/335/296
+f 351/336/297 353/330/297 304/333/297 78/337/297
+f 78/335/298 350/334/298 348/338/298 305/339/298
+f 349/340/299 351/336/299 78/337/299 305/341/299
+f 305/339/300 348/338/300 328/299/300 309/298/300
+f 329/301/301 349/340/301 305/341/301 309/302/301
+f 326/318/302 328/299/302 348/338/302 342/320/302
+f 349/340/303 329/301/303 327/319/303 343/323/303
+f 296/291/304 298/293/304 318/342/304 310/343/304
+f 319/344/305 299/294/305 297/292/305 311/345/305
+f 76/346/306 316/347/306 324/327/306 77/326/306
+f 325/331/307 317/348/307 76/349/307 77/332/307
+f 302/350/308 358/351/308 356/352/308 303/353/308
+f 357/354/309 359/355/309 302/356/309 303/357/309
+f 303/353/310 356/352/310 354/358/310 75/359/310
+f 355/360/311 357/354/311 303/357/311 75/361/311
+f 75/359/312 354/358/312 316/347/312 76/346/312
+f 317/348/313 355/360/313 75/361/313 76/349/313
+f 292/362/314 294/289/314 362/363/314 364/364/314
+f 363/365/315 295/290/315 293/366/315 365/367/315
+f 364/364/316 362/363/316 368/368/316 366/369/316
+f 369/370/317 363/365/317 365/367/317 367/371/317
+f 366/369/318 368/368/318 370/372/318 372/373/318
+f 371/374/319 369/370/319 367/371/319 373/375/319
+f 372/373/320 370/372/320 376/376/320 374/377/320
+f 377/378/321 371/374/321 373/375/321 375/379/321
+f 314/380/322 378/381/322 374/377/322 376/376/322
+f 375/379/323 379/382/323 315/383/323 377/378/323
+f 316/347/324 354/358/324 374/377/324 378/381/324
+f 375/379/325 355/360/325 317/348/325 379/382/325
+f 354/358/326 356/352/326 372/373/326 374/377/326
+f 373/375/327 357/354/327 355/360/327 375/379/327
+f 356/352/328 358/351/328 366/369/328 372/373/328
+f 367/371/329 359/355/329 357/354/329 373/375/329
+f 358/351/330 360/384/330 364/364/330 366/369/330
+f 365/367/331 361/385/331 359/355/331 367/371/331
+f 290/386/332 292/362/332 364/364/332 360/384/332
+f 365/367/333 293/366/333 291/387/333 361/385/333
+f 74/388/334 360/384/334 358/351/334 302/350/334
+f 359/355/335 361/385/335 74/389/335 302/356/335
+f 284/390/336 286/391/336 288/392/336 290/386/336
+f 289/393/337 287/394/337 285/395/337 291/387/337
+f 284/390/338 290/386/338 360/384/338 74/388/338
+f 361/385/339 291/387/339 285/395/339 74/389/339
+f 73/396/340 284/390/340 74/388/340
+f 74/389/341 285/395/341 73/397/341
+f 294/289/342 296/291/342 310/343/342 362/363/342
+f 311/345/343 297/292/343 295/290/343 363/365/343
+f 310/343/344 312/398/344 368/368/344 362/363/344
+f 369/370/345 313/399/345 311/345/345 363/365/345
+f 312/398/346 382/400/346 370/372/346 368/368/346
+f 371/374/347 383/401/347 313/399/347 369/370/347
+f 314/380/348 376/376/348 370/372/348 382/400/348
+f 371/374/349 377/378/349 315/383/349 383/401/349
+f 348/338/350 350/334/350 386/402/350 384/403/350
+f 387/404/351 351/336/351 349/340/351 385/405/351
+f 318/342/352 384/403/352 386/402/352 320/406/352
+f 387/404/353 385/405/353 319/344/353 321/407/353
+f 298/293/354 300/295/354 384/403/354 318/342/354
+f 385/405/355 301/296/355 299/294/355 319/344/355
+f 300/295/356 344/321/356 342/320/356 384/403/356
+f 343/323/357 345/322/357 301/296/357 385/405/357
+f 342/320/358 348/338/358 384/403/358
+f 385/405/359 349/340/359 343/323/359
+f 300/295/360 346/324/360 344/321/360
+f 345/322/361 347/325/361 301/296/361
+f 314/380/362 322/408/362 380/409/362 378/381/362
+f 381/410/363 323/411/363 315/383/363 379/382/363
+f 316/347/364 378/381/364 380/409/364 324/327/364
+f 381/410/365 379/382/365 317/348/365 325/331/365
+f 320/406/366 386/402/366 380/409/366 322/408/366
+f 381/410/367 387/404/367 321/407/367 323/411/367
+f 350/334/368 352/328/368 380/409/368 386/402/368
+f 381/410/369 353/330/369 351/336/369 387/404/369
+f 324/327/370 380/409/370 352/328/370
+f 353/330/371 381/410/371 325/331/371
+f 400/412/372 388/413/372 414/414/372 402/415/372
+f 415/416/373 389/417/373 401/418/373 403/419/373
+f 400/412/374 402/415/374 404/420/374 398/421/374
+f 405/422/375 403/419/375 401/418/375 399/423/375
+f 398/421/376 404/420/376 406/424/376 396/425/376
+f 407/426/377 405/422/377 399/423/377 397/427/377
+f 396/425/378 406/424/378 408/428/378 394/429/378
+f 409/430/379 407/426/379 397/427/379 395/431/379
+f 394/429/380 408/428/380 410/432/380 392/433/380
+f 411/434/381 409/430/381 395/431/381 393/435/381
+f 392/433/382 410/432/382 412/436/382 390/437/382
+f 413/438/383 411/434/383 393/435/383 391/439/383
+f 410/432/384 420/440/384 418/441/384 412/436/384
+f 419/442/385 421/443/385 411/434/385 413/438/385
+f 408/428/386 422/444/386 420/440/386 410/432/386
+f 421/443/387 423/445/387 409/430/387 411/434/387
+f 406/424/388 424/446/388 422/444/388 408/428/388
+f 423/445/389 425/447/389 407/426/389 409/430/389
+f 404/420/390 426/448/390 424/446/390 406/424/390
+f 425/447/391 427/449/391 405/422/391 407/426/391
+f 402/415/392 428/450/392 426/448/392 404/420/392
+f 427/449/393 429/451/393 403/419/393 405/422/393
+f 402/415/394 414/414/394 416/452/394 428/450/394
+f 417/453/395 415/416/395 403/419/395 429/451/395
+f 318/342/396 320/406/396 444/454/396 442/455/396
+f 445/456/397 321/407/397 319/344/397 443/457/397
+f 320/458/398 390/437/398 412/436/398 444/459/398
+f 413/438/399 391/439/399 321/460/399 445/461/399
+f 310/343/400 318/342/400 442/455/400 312/398/400
+f 443/457/401 319/344/401 311/345/401 313/399/401
+f 382/462/402 430/463/402 414/414/402 388/413/402
+f 415/416/403 431/464/403 383/465/403 389/417/403
+f 412/436/404 418/441/404 440/466/404 444/459/404
+f 441/467/405 419/442/405 413/438/405 445/461/405
+f 438/468/406 446/469/406 444/459/406 440/466/406
+f 445/461/407 447/470/407 439/471/407 441/467/407
+f 434/472/408 446/469/408 438/468/408 436/473/408
+f 439/471/409 447/470/409 435/474/409 437/475/409
+f 432/476/410 448/477/410 446/469/410 434/472/410
+f 447/470/411 449/478/411 433/479/411 435/474/411
+f 430/463/412 448/477/412 432/476/412 450/480/412
+f 433/479/413 449/478/413 431/464/413 451/481/413
+f 414/414/414 430/463/414 450/480/414 416/452/414
+f 451/481/415 431/464/415 415/416/415 417/453/415
+f 312/398/416 448/482/416 430/483/416 382/400/416
+f 431/484/417 449/485/417 313/399/417 383/401/417
+f 312/398/418 442/455/418 446/486/418 448/482/418
+f 447/487/419 443/457/419 313/399/419 449/485/419
+f 442/455/420 444/454/420 446/486/420
+f 447/487/421 445/456/421 443/457/421
+f 416/452/422 450/480/422 452/488/422 476/489/422
+f 453/490/423 451/481/423 417/453/423 477/491/423
+f 450/480/424 432/476/424 462/492/424 452/488/424
+f 463/493/425 433/479/425 451/481/425 453/490/425
+f 432/476/426 434/472/426 460/494/426 462/492/426
+f 461/495/427 435/474/427 433/479/427 463/493/427
+f 434/472/428 436/473/428 458/496/428 460/494/428
+f 459/497/429 437/475/429 435/474/429 461/495/429
+f 436/473/430 438/468/430 456/498/430 458/496/430
+f 457/499/431 439/471/431 437/475/431 459/497/431
+f 438/468/432 440/466/432 454/500/432 456/498/432
+f 455/501/433 441/467/433 439/471/433 457/499/433
+f 440/466/434 418/441/434 474/502/434 454/500/434
+f 475/503/435 419/442/435 441/467/435 455/501/435
+f 428/450/436 416/452/436 476/489/436 464/504/436
+f 477/491/437 417/453/437 429/451/437 465/505/437
+f 426/448/438 428/450/438 464/504/438 466/506/438
+f 465/505/439 429/451/439 427/449/439 467/507/439
+f 424/446/440 426/448/440 466/506/440 468/508/440
+f 467/507/441 427/449/441 425/447/441 469/509/441
+f 422/444/442 424/446/442 468/508/442 470/510/442
+f 469/509/443 425/447/443 423/445/443 471/511/443
+f 420/440/444 422/444/444 470/510/444 472/512/444
+f 471/511/445 423/445/445 421/443/445 473/513/445
+f 418/441/446 420/440/446 472/512/446 474/502/446
+f 473/513/447 421/443/447 419/442/447 475/503/447
+f 458/496/448 456/498/448 480/514/448 478/515/448
+f 481/516/449 457/499/449 459/497/449 479/517/449
+f 478/515/450 480/514/450 482/518/450 484/519/450
+f 483/520/451 481/516/451 479/517/451 485/521/451
+f 484/519/452 482/518/452 488/522/452 486/523/452
+f 489/524/453 483/520/453 485/521/453 487/525/453
+f 486/523/454 488/522/454 490/526/454 492/527/454
+f 491/528/455 489/524/455 487/525/455 493/529/455
+f 464/504/456 476/489/456 486/523/456 492/527/456
+f 487/525/457 477/491/457 465/505/457 493/529/457
+f 452/488/458 484/519/458 486/523/458 476/489/458
+f 487/525/459 485/521/459 453/490/459 477/491/459
+f 452/488/460 462/492/460 478/515/460 484/519/460
+f 479/517/461 463/493/461 453/490/461 485/521/461
+f 458/496/462 478/515/462 462/492/462 460/494/462
+f 463/493/463 479/517/463 459/497/463 461/495/463
+f 454/500/464 474/502/464 480/514/464 456/498/464
+f 481/516/465 475/503/465 455/501/465 457/499/465
+f 472/512/466 482/518/466 480/514/466 474/502/466
+f 481/516/467 483/520/467 473/513/467 475/503/467
+f 470/510/468 488/522/468 482/518/468 472/512/468
+f 483/520/469 489/524/469 471/511/469 473/513/469
+f 468/508/470 490/526/470 488/522/470 470/510/470
+f 489/524/471 491/528/471 469/509/471 471/511/471
+f 466/506/472 492/527/472 490/526/472 468/508/472
+f 491/528/473 493/529/473 467/507/473 469/509/473
+f 464/504/474 492/527/474 466/506/474
+f 467/507/475 493/529/475 465/505/475
+f 392/433/476 390/437/476 504/530/476 502/531/476
+f 505/532/477 391/439/477 393/435/477 503/533/477
+f 394/429/478 392/433/478 502/531/478 500/534/478
+f 503/533/479 393/435/479 395/431/479 501/535/479
+f 396/425/480 394/429/480 500/534/480 498/536/480
+f 501/535/481 395/431/481 397/427/481 499/537/481
+f 398/538/482 396/425/482 498/536/482 496/539/482
+f 499/537/483 397/427/483 399/540/483 497/541/483
+f 400/542/484 398/538/484 496/539/484 494/543/484
+f 497/541/485 399/540/485 401/544/485 495/545/485
+f 388/546/486 400/542/486 494/543/486 506/547/486
+f 495/545/487 401/544/487 389/548/487 507/549/487
+f 494/543/488 502/531/488 504/530/488 506/547/488
+f 505/532/489 503/533/489 495/545/489 507/549/489
+f 494/543/490 496/539/490 500/534/490 502/531/490
+f 501/535/491 497/541/491 495/545/491 503/533/491
+f 496/539/492 498/536/492 500/534/492
+f 501/535/493 499/537/493 497/541/493
+f 314/380/494 382/400/494 388/550/494 506/551/494
+f 389/548/495 383/552/495 315/553/495 507/549/495
+f 314/554/496 506/547/496 504/530/496 322/555/496
+f 505/532/497 507/549/497 315/553/497 323/556/497
+f 320/458/498 322/555/498 504/530/498 390/437/498
+f 505/532/499 323/556/499 321/460/499 391/439/499

diff --git a/moving_platform.h b/moving_platform.h
new file mode 100644 (file)
index 0000000..19c4bbf
--- /dev/null
+++ b/moving_platform.h
@@ -0,0 +1,24 @@
+#pragma once
+
+class elevator : public loaded_object, public block_object {
+       public:
+               bool up = true;
+
+               elevator(const char* of, const char* tf, glm::vec3 s) : loaded_object(of, tf, s) {}
+               void move(){
+                       // Just one elevator for now
+                       if(up) {
+                               locations[0].y += .1;
+                               if(locations[0].y > 100)
+                                       up = false;
+                       } else {
+                               locations[0].y -= .1;
+                               if(locations[0].y <= 0)
+                                       up = true;
+                       }
+               }
+               void draw(glm::mat4 vp){
+                       loaded_object::draw(vp);
+               }
+};
+

diff --git a/old_objects.cpp b/old_objects.cpp
new file mode 100644 (file)
index 0000000..943a5fc
--- /dev/null
+++ b/old_objects.cpp
@@ -0,0 +1,192 @@
+#include "game.h"
+#include "old_objects.h"
+
+int weirdbox::init()  {
+       // Initialization part
+       float vertices[] = {
+               // front
+               -1.0, -1.0,  1.0,
+               1.0, -1.0,  1.0,
+               1.0,  1.0,  1.0,
+               -1.0,  1.0,  1.0,
+               // back
+               -1.0, -1.0, -1.0,
+               1.0, -1.0, -1.0,
+               1.0,  1.0, -1.0,
+               -1.0,  1.0, -1.0,
+       };
+       glGenBuffers(1, &vbuf);
+       glBindBuffer(GL_SHADER_STORAGE_BUFFER, vbuf);
+       glBufferData(GL_SHADER_STORAGE_BUFFER, sizeof(vertices), vertices, GL_STATIC_DRAW);
+
+       float colors[] = {
+               // front colors
+               1.0, 0.0, 0.0,
+               0.0, 1.0, 0.0,
+               0.0, 0.0, 1.0,
+               1.0, 1.0, 1.0,
+               // back colors
+               1.0, 0.0, 0.0,
+               0.0, 1.0, 0.0,
+               0.0, 0.0, 1.0,
+               1.0, 1.0, 1.0,
+       };
+       glGenBuffers(1, &cbuf);
+       glBindBuffer(GL_SHADER_STORAGE_BUFFER, cbuf);
+       glBufferData(GL_SHADER_STORAGE_BUFFER, sizeof(colors), colors, GL_STATIC_DRAW);
+
+       GLushort cube_elements[] = {
+               // front
+               0, 1, 2,
+               2, 3, 0,
+               // top
+               1, 5, 6,
+               6, 2, 1,
+               // back
+               7, 6, 5,
+               5, 4, 7,
+               // bottom
+               4, 0, 3,
+               3, 7, 4,
+               // left
+               4, 5, 1,
+               1, 0, 4,
+               // right
+               3, 2, 6,
+               6, 7, 3,
+       };
+       glGenBuffers(1, &ebuf);
+       glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, ebuf);
+       glBufferData(GL_ELEMENT_ARRAY_BUFFER, sizeof(cube_elements), cube_elements, GL_STATIC_DRAW);
+
+       program = make_program("vertex_shader.glsl",0, 0, 0, "fragment_shader.glsl");
+       if (!program)
+               return 1;
+
+       v_attrib = glGetAttribLocation(program, "in_vertex");
+       c_attrib = glGetAttribLocation(program, "in_color");
+       mvp_uniform = glGetUniformLocation(program, "mvp");
+       anim_uniform = glGetUniformLocation(program, "animation");
+       return 0;
+}
+void weirdbox::draw(glm::mat4 vp)  {
+       glUseProgram(program);
+
+       glEnableVertexAttribArray(v_attrib);
+       glBindBuffer(GL_ARRAY_BUFFER, vbuf);
+       glVertexAttribPointer(v_attrib, 3, GL_FLOAT, GL_FALSE, 0, 0);
+
+       glEnableVertexAttribArray(c_attrib);
+       glBindBuffer(GL_ARRAY_BUFFER, cbuf);
+       glVertexAttribPointer(c_attrib, 3, GL_FLOAT, GL_FALSE, 0, 0);
+
+       int size;
+       glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, ebuf);
+       glGetBufferParameteriv(GL_ELEMENT_ARRAY_BUFFER, GL_BUFFER_SIZE, &size);
+
+
+       glUniformMatrix4fv(mvp_uniform, 1, 0, glm::value_ptr(vp));
+       glUniformMatrix4fv(anim_uniform, 1, 0, (float*)&anim);
+
+       glDrawElementsInstanced(GL_TRIANGLES, size / sizeof(GLushort), GL_UNSIGNED_SHORT, 0, 10);
+}
+void weirdbox::animate()  {
+       acount++;
+       anim = glm::rotate(glm::mat4(1.0f), acount / 20.0f, glm::vec3(0, 1, 0));
+       anim = glm::scale(anim, glm::vec3(1 - .5f*sinf(acount/10.0f), 3 + 2*sinf(acount / 10.0f), 1 - 0.5*sinf(acount/10.0f)));
+}
+
+void weirdbox::deinit()  {
+       glDeleteProgram(program);
+       // And the vertex, color, and element buffers
+       // And the shaders      
+}
+
+int lightbox::init()  {
+       // Initialization part
+       float vertices[] = {
+               // front
+               -1.0, -1.0,  1.0,
+               1.0, -1.0,  1.0,
+               1.0,  1.0,  1.0,
+               -1.0,  1.0,  1.0,
+               // back
+               -1.0, -1.0, -1.0,
+               1.0, -1.0, -1.0,
+               1.0,  1.0, -1.0,
+               -1.0,  1.0, -1.0,
+       };
+       glGenBuffers(1, &vbuf);
+       glBindBuffer(GL_SHADER_STORAGE_BUFFER, vbuf);
+       glBufferData(GL_SHADER_STORAGE_BUFFER, sizeof(vertices), vertices, GL_STATIC_DRAW);
+
+       float colors[] = {
+               // front colors
+               1.0, 0.0, 1.0,
+               1.0, 0.5, 1.0,
+               1.0, 1.0, 1.0,
+               1.0, 1.0, 1.0,
+               // back colors
+               0.5, 1.0, 0.5,
+               0.5, 1.0, 0.5,
+               1.0, 1.0, 1.0,
+               1.0, 1.0, 1.0,
+       };
+       glGenBuffers(1, &cbuf);
+       glBindBuffer(GL_SHADER_STORAGE_BUFFER, cbuf);
+       glBufferData(GL_SHADER_STORAGE_BUFFER, sizeof(colors), colors, GL_STATIC_DRAW);
+
+       GLushort cube_elements[] = {
+               // front
+               0, 1, 2,
+               2, 3, 0,
+               // top
+               1, 5, 6,
+               6, 2, 1,
+               // back
+               7, 6, 5,
+               5, 4, 7,
+               // bottom
+               4, 0, 3,
+               3, 7, 4,
+               // left
+               4, 5, 1,
+               1, 0, 4,
+               // right
+               3, 2, 6,
+               6, 7, 3,
+       };
+       glGenBuffers(1, &ebuf);
+       glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, ebuf);
+       glBufferData(GL_ELEMENT_ARRAY_BUFFER, sizeof(cube_elements), cube_elements, GL_STATIC_DRAW);
+
+       program = make_program("other_vertex_shader.glsl",0, 0, 0, "fragment_shader.glsl");
+       if (!program)
+               return 1;
+
+       v_attrib = glGetAttribLocation(program, "in_vertex");
+       c_attrib = glGetAttribLocation(program, "in_color");
+       mvp_uniform = glGetUniformLocation(program, "mvp");
+       return 0;
+}
+
+void lightbox::draw(glm::mat4 vp)  {
+       glUseProgram(program);
+
+       glEnableVertexAttribArray(v_attrib);
+       glBindBuffer(GL_ARRAY_BUFFER, vbuf);
+       glVertexAttribPointer(v_attrib, 3, GL_FLOAT, GL_FALSE, 0, 0);
+
+       glEnableVertexAttribArray(c_attrib);
+       glBindBuffer(GL_ARRAY_BUFFER, cbuf);
+       glVertexAttribPointer(c_attrib, 3, GL_FLOAT, GL_FALSE, 0, 0);
+
+       int size;
+       glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, ebuf);
+       glGetBufferParameteriv(GL_ELEMENT_ARRAY_BUFFER, GL_BUFFER_SIZE, &size);
+
+       glUniformMatrix4fv(mvp_uniform, 1, 0, glm::value_ptr(vp));
+
+       glDrawElementsInstanced(GL_TRIANGLES, size / sizeof(GLushort), GL_UNSIGNED_SHORT, 0, 1);
+}
+

diff --git a/old_objects.h b/old_objects.h
new file mode 100644 (file)
index 0000000..da78b3c
--- /dev/null
+++ b/old_objects.h
@@ -0,0 +1,19 @@
+#pragma once
+
+class weirdbox : public gameobject {
+       public:
+               unsigned int mvp_uniform, anim_uniform, v_attrib, c_attrib, program, vbuf, cbuf, ebuf, acount;
+               glm::mat4 anim;
+               int init() override;
+               void draw(glm::mat4 vp) override;
+               void animate() override;
+               void deinit() override;
+};
+
+class lightbox : public gameobject {
+       public:
+               unsigned int mvp_uniform, anim_uniform, v_attrib, c_attrib, program, vbuf, cbuf, ebuf;
+               int init() override;
+               void draw(glm::mat4 vp) override;
+};
+

diff --git a/other_vertex_shader.glsl b/other_vertex_shader.glsl
new file mode 100644 (file)
index 0000000..b80f976
--- /dev/null
+++ b/other_vertex_shader.glsl
@@ -0,0 +1,13 @@
+#version 460\r
+\r
+in vec3 in_vertex;\r
+in vec3 in_color;\r
+uniform mat4 mvp;\r
+out vec4 fcolor;\r
+out vec4 gl_Position;\r
+\r
+void main(void) {\r
+       vec4 tmp = vec4(in_vertex.x, in_vertex.y + 6, in_vertex.z, 1.0);\r
+       gl_Position = mvp * tmp;\r
+       fcolor = vec4(in_color, 1.0);\r
+}\r

diff --git a/platform.obj b/platform.obj
new file mode 100644 (file)
index 0000000..83d1143
--- /dev/null
+++ b/platform.obj
@@ -0,0 +1,40 @@
+# Blender v3.0.1 OBJ File: ''
+# www.blender.org
+mtllib cube.mtl
+o Cube
+v 4.000000 1.000000 -4.000000
+v 4.000000 -1.000000 -4.000000
+v 4.000000 1.000000 4.000000
+v 4.000000 -1.000000 4.000000
+v -4.000000 1.000000 -4.000000
+v -4.000000 -1.000000 -4.000000
+v -4.000000 1.000000 4.000000
+v -4.000000 -1.000000 4.000000
+vt 0.625000 0.500000
+vt 0.875000 0.500000
+vt 0.875000 0.750000
+vt 0.625000 0.750000
+vt 0.375000 0.750000
+vt 0.625000 1.000000
+vt 0.375000 1.000000
+vt 0.375000 0.000000
+vt 0.625000 0.000000
+vt 0.625000 0.250000
+vt 0.375000 0.250000
+vt 0.125000 0.500000
+vt 0.375000 0.500000
+vt 0.125000 0.750000
+vn 0.0000 1.0000 0.0000
+vn 0.0000 0.0000 1.0000
+vn -1.0000 0.0000 0.0000
+vn 0.0000 -1.0000 0.0000
+vn 1.0000 0.0000 0.0000
+vn 0.0000 0.0000 -1.0000
+usemtl Material
+s off
+f 1/1/1 5/2/1 7/3/1 3/4/1
+f 4/5/2 3/4/2 7/6/2 8/7/2
+f 8/8/3 7/9/3 5/10/3 6/11/3
+f 6/12/4 2/13/4 4/5/4 8/14/4
+f 2/13/5 1/1/5 3/4/5 4/5/5
+f 6/11/6 5/10/6 1/1/6 2/13/6

diff --git a/projectile.jpg b/projectile.jpg
new file mode 100644 (file)
index 0000000..4e86c8c
Binary files /dev/null and b/projectile.jpg differ

diff --git a/projectile.obj b/projectile.obj
new file mode 100644 (file)
index 0000000..7d40d0c
--- /dev/null
+++ b/projectile.obj
@@ -0,0 +1,271 @@
+# Blender v3.0.1 OBJ File: ''
+# www.blender.org
+mtllib projectile.mtl
+o Icosphere
+v 0.000000 -1.000000 0.000000
+v 0.723607 -0.447220 0.525725
+v -0.276388 -0.447220 0.850649
+v -0.894426 -0.447216 0.000000
+v -0.276388 -0.447220 -0.850649
+v 0.723607 -0.447220 -0.525725
+v 0.276388 0.447220 0.850649
+v -0.723607 0.447220 0.525725
+v -0.723607 0.447220 -0.525725
+v 0.276388 0.447220 -0.850649
+v 0.894426 0.447216 0.000000
+v 0.000000 1.000000 0.000000
+v -0.162456 -0.850654 0.499995
+v 0.425323 -0.850654 0.309011
+v 0.262869 -0.525738 0.809012
+v 0.850648 -0.525736 0.000000
+v 0.425323 -0.850654 -0.309011
+v -0.525730 -0.850652 0.000000
+v -0.688189 -0.525736 0.499997
+v -0.162456 -0.850654 -0.499995
+v -0.688189 -0.525736 -0.499997
+v 0.262869 -0.525738 -0.809012
+v 0.951058 0.000000 0.309013
+v 0.951058 0.000000 -0.309013
+v 0.000000 0.000000 1.000000
+v 0.587786 0.000000 0.809017
+v -0.951058 0.000000 0.309013
+v -0.587786 0.000000 0.809017
+v -0.587786 0.000000 -0.809017
+v -0.951058 0.000000 -0.309013
+v 0.587786 0.000000 -0.809017
+v 0.000000 0.000000 -1.000000
+v 0.688189 0.525736 0.499997
+v -0.262869 0.525738 0.809012
+v -0.850648 0.525736 0.000000
+v -0.262869 0.525738 -0.809012
+v 0.688189 0.525736 -0.499997
+v 0.162456 0.850654 0.499995
+v 0.525730 0.850652 0.000000
+v -0.425323 0.850654 0.309011
+v -0.425323 0.850654 -0.309011
+v 0.162456 0.850654 -0.499995
+vt 0.181819 0.000000
+vt 0.227273 0.078731
+vt 0.136365 0.078731
+vt 0.272728 0.157461
+vt 0.318182 0.078731
+vt 0.363637 0.157461
+vt 0.909091 0.000000
+vt 0.954545 0.078731
+vt 0.863636 0.078731
+vt 0.727273 0.000000
+vt 0.772727 0.078731
+vt 0.681818 0.078731
+vt 0.545455 0.000000
+vt 0.590909 0.078731
+vt 0.500000 0.078731
+vt 0.318182 0.236191
+vt 0.090910 0.157461
+vt 0.181819 0.157461
+vt 0.136365 0.236191
+vt 0.818182 0.157461
+vt 0.909091 0.157461
+vt 0.863636 0.236191
+vt 0.636364 0.157461
+vt 0.727273 0.157461
+vt 0.681818 0.236191
+vt 0.454546 0.157461
+vt 0.545455 0.157461
+vt 0.500000 0.236191
+vt 0.227273 0.236191
+vt 0.045455 0.236191
+vt 0.772727 0.236191
+vt 0.590909 0.236191
+vt 0.409092 0.236191
+vt 0.181819 0.314921
+vt 0.272728 0.314921
+vt 0.227273 0.393651
+vt 0.000000 0.314921
+vt 0.090910 0.314921
+vt 0.045455 0.393651
+vt 0.727273 0.314921
+vt 0.818182 0.314921
+vt 0.772727 0.393651
+vt 0.545455 0.314921
+vt 0.636364 0.314921
+vt 0.590909 0.393651
+vt 0.363637 0.314921
+vt 0.454546 0.314921
+vt 0.409092 0.393651
+vt 0.500000 0.393651
+vt 0.454546 0.472382
+vt 0.681818 0.393651
+vt 0.636364 0.472382
+vt 0.863636 0.393651
+vt 0.818182 0.472382
+vt 0.909091 0.314921
+vt 0.136365 0.393651
+vt 0.090910 0.472382
+vt 0.318182 0.393651
+vt 0.272728 0.472382
+vt 0.954545 0.236191
+vt 1.000000 0.157461
+vt 0.409092 0.078731
+vt 0.363637 0.000000
+vn 0.1024 -0.9435 0.3151
+vn 0.7002 -0.6617 0.2680
+vn -0.2680 -0.9435 0.1947
+vn -0.2680 -0.9435 -0.1947
+vn 0.1024 -0.9435 -0.3151
+vn 0.9050 -0.3304 0.2680
+vn 0.0247 -0.3304 0.9435
+vn -0.8897 -0.3304 0.3151
+vn -0.5746 -0.3304 -0.7488
+vn 0.5346 -0.3304 -0.7779
+vn 0.8026 -0.1256 0.5831
+vn -0.3066 -0.1256 0.9435
+vn -0.9921 -0.1256 0.0000
+vn -0.3066 -0.1256 -0.9435
+vn 0.8026 -0.1256 -0.5831
+vn 0.4089 0.6617 0.6284
+vn -0.4713 0.6617 0.5831
+vn -0.7002 0.6617 -0.2680
+vn 0.0385 0.6617 -0.7488
+vn 0.7240 0.6617 -0.1947
+vn 0.2680 0.9435 -0.1947
+vn 0.4911 0.7947 -0.3568
+vn 0.4089 0.6617 -0.6284
+vn -0.1024 0.9435 -0.3151
+vn -0.1876 0.7947 -0.5773
+vn -0.4713 0.6617 -0.5831
+vn -0.3313 0.9435 0.0000
+vn -0.6071 0.7947 0.0000
+vn -0.7002 0.6617 0.2680
+vn -0.1024 0.9435 0.3151
+vn -0.1876 0.7947 0.5773
+vn 0.0385 0.6617 0.7488
+vn 0.2680 0.9435 0.1947
+vn 0.4911 0.7947 0.3568
+vn 0.7240 0.6617 0.1947
+vn 0.8897 0.3304 -0.3151
+vn 0.7947 0.1876 -0.5773
+vn 0.5746 0.3304 -0.7488
+vn -0.0247 0.3304 -0.9435
+vn -0.3035 0.1876 -0.9342
+vn -0.5346 0.3304 -0.7779
+vn -0.9050 0.3304 -0.2680
+vn -0.9822 0.1876 0.0000
+vn -0.9050 0.3304 0.2680
+vn -0.5346 0.3304 0.7779
+vn -0.3035 0.1876 0.9342
+vn -0.0247 0.3304 0.9435
+vn 0.5746 0.3304 0.7488
+vn 0.7947 0.1876 0.5773
+vn 0.8897 0.3304 0.3151
+vn 0.3066 0.1256 -0.9435
+vn 0.3035 -0.1876 -0.9342
+vn 0.0247 -0.3304 -0.9435
+vn -0.8026 0.1256 -0.5831
+vn -0.7947 -0.1876 -0.5773
+vn -0.8897 -0.3304 -0.3151
+vn -0.8026 0.1256 0.5831
+vn -0.7947 -0.1876 0.5773
+vn -0.5746 -0.3304 0.7488
+vn 0.3066 0.1256 0.9435
+vn 0.3035 -0.1876 0.9342
+vn 0.5346 -0.3304 0.7779
+vn 0.9921 0.1256 0.0000
+vn 0.9822 -0.1876 0.0000
+vn 0.9050 -0.3304 -0.2680
+vn 0.4713 -0.6617 -0.5831
+vn 0.1876 -0.7947 -0.5773
+vn -0.0385 -0.6617 -0.7488
+vn -0.4089 -0.6617 -0.6284
+vn -0.4911 -0.7947 -0.3568
+vn -0.7240 -0.6617 -0.1947
+vn -0.7240 -0.6617 0.1947
+vn -0.4911 -0.7947 0.3568
+vn -0.4089 -0.6617 0.6284
+vn 0.7002 -0.6617 -0.2680
+vn 0.6071 -0.7947 0.0000
+vn 0.3313 -0.9435 0.0000
+vn -0.0385 -0.6617 0.7488
+vn 0.1876 -0.7947 0.5773
+vn 0.4713 -0.6617 0.5831
+usemtl None
+s off
+f 1/1/1 14/2/1 13/3/1
+f 2/4/2 14/5/2 16/6/2
+f 1/7/3 13/8/3 18/9/3
+f 1/10/4 18/11/4 20/12/4
+f 1/13/5 20/14/5 17/15/5
+f 2/4/6 16/6/6 23/16/6
+f 3/17/7 15/18/7 25/19/7
+f 4/20/8 19/21/8 27/22/8
+f 5/23/9 21/24/9 29/25/9
+f 6/26/10 22/27/10 31/28/10
+f 2/4/11 23/16/11 26/29/11
+f 3/17/12 25/19/12 28/30/12
+f 4/20/13 27/22/13 30/31/13
+f 5/23/14 29/25/14 32/32/14
+f 6/26/15 31/28/15 24/33/15
+f 7/34/16 33/35/16 38/36/16
+f 8/37/17 34/38/17 40/39/17
+f 9/40/18 35/41/18 41/42/18
+f 10/43/19 36/44/19 42/45/19
+f 11/46/20 37/47/20 39/48/20
+f 39/48/21 42/49/21 12/50/21
+f 39/48/22 37/47/22 42/49/22
+f 37/47/23 10/43/23 42/49/23
+f 42/45/24 41/51/24 12/52/24
+f 42/45/25 36/44/25 41/51/25
+f 36/44/26 9/40/26 41/51/26
+f 41/42/27 40/53/27 12/54/27
+f 41/42/28 35/41/28 40/53/28
+f 35/41/29 8/55/29 40/53/29
+f 40/39/30 38/56/30 12/57/30
+f 40/39/31 34/38/31 38/56/31
+f 34/38/32 7/34/32 38/56/32
+f 38/36/33 39/58/33 12/59/33
+f 38/36/34 33/35/34 39/58/34
+f 33/35/35 11/46/35 39/58/35
+f 24/33/36 37/47/36 11/46/36
+f 24/33/37 31/28/37 37/47/37
+f 31/28/38 10/43/38 37/47/38
+f 32/32/39 36/44/39 10/43/39
+f 32/32/40 29/25/40 36/44/40
+f 29/25/41 9/40/41 36/44/41
+f 30/31/42 35/41/42 9/40/42
+f 30/31/43 27/22/43 35/41/43
+f 27/22/44 8/55/44 35/41/44
+f 28/30/45 34/38/45 8/37/45
+f 28/30/46 25/19/46 34/38/46
+f 25/19/47 7/34/47 34/38/47
+f 26/29/48 33/35/48 7/34/48
+f 26/29/49 23/16/49 33/35/49
+f 23/16/50 11/46/50 33/35/50
+f 31/28/51 32/32/51 10/43/51
+f 31/28/52 22/27/52 32/32/52
+f 22/27/53 5/23/53 32/32/53
+f 29/25/54 30/31/54 9/40/54
+f 29/25/55 21/24/55 30/31/55
+f 21/24/56 4/20/56 30/31/56
+f 27/22/57 28/60/57 8/55/57
+f 27/22/58 19/21/58 28/60/58
+f 19/21/59 3/61/59 28/60/59
+f 25/19/60 26/29/60 7/34/60
+f 25/19/61 15/18/61 26/29/61
+f 15/18/62 2/4/62 26/29/62
+f 23/16/63 24/33/63 11/46/63
+f 23/16/64 16/6/64 24/33/64
+f 16/6/65 6/26/65 24/33/65
+f 17/15/66 22/27/66 6/26/66
+f 17/15/67 20/14/67 22/27/67
+f 20/14/68 5/23/68 22/27/68
+f 20/12/69 21/24/69 5/23/69
+f 20/12/70 18/11/70 21/24/70
+f 18/11/71 4/20/71 21/24/71
+f 18/9/72 19/21/72 4/20/72
+f 18/9/73 13/8/73 19/21/73
+f 13/8/74 3/61/74 19/21/74
+f 16/6/75 17/62/75 6/26/75
+f 16/6/76 14/5/76 17/62/76
+f 14/5/77 1/63/77 17/62/77
+f 13/3/78 15/18/78 3/17/78
+f 13/3/79 14/2/79 15/18/79
+f 14/2/80 2/4/80 15/18/80

diff --git a/projectiles.h b/projectiles.h
new file mode 100644 (file)
index 0000000..045c51c
--- /dev/null
+++ b/projectiles.h
@@ -0,0 +1,148 @@
+#pragma once
+
+class projectile : public loaded_object {
+public:
+       std::vector<glm::vec3> directions;
+       std::vector<float> lifetimes;
+       std::vector<bool> bursting;
+       projectile() : loaded_object("projectile.obj", "projectile.jpg", glm::vec3(0.1, 0.1, 0.1)) {
+               collision_check = false;
+       }
+       void create_burst(float quantity, glm::vec3 origin, float speed){
+               for(size_t i = 0; i < quantity; i++){
+                       locations.push_back(origin);
+                       lifetimes.push_back(10000.0f);
+                       // One note:  This does create a cube of projectiles
+                       directions.push_back(glm::vec3(randvel(speed), randvel(speed), randvel(speed)));
+                       bursting.push_back(false);
+               }
+       }
+       void move() {
+               data_mutex.lock();
+               for(size_t i = 0; i < locations.size(); i++){
+                       if(bursting[i])
+                               directions[i].y -= 0.0002;
+                       locations[i] += directions[i];
+                       lifetimes[i] -= 10; // TODO:  Manage time resolutions better
+                       if(lifetimes[i] <= 0.0f) {
+                               if(bursting[i])
+                                       create_burst(200, locations[i], 0.003);
+                               remove_projectile(i);
+                       }
+               }
+               data_mutex.unlock();
+       }
+       void remove_projectile(size_t index){
+               locations.erase(locations.begin() + index);
+               directions.erase(directions.begin() + index);
+               lifetimes.erase(lifetimes.begin() + index);
+               bursting.erase(bursting.begin() + index);
+       }
+       
+       void add_projectile(glm::vec3 location, glm::vec3 direction, float lifetime, bool burst = false){
+               data_mutex.lock();
+               locations.push_back(location);
+               directions.push_back(direction);
+               lifetimes.push_back(lifetime);
+               bursting.push_back(burst);
+               data_mutex.unlock();
+       }
+       void add_projectile(glm::vec3 location, float heading, float elevation, float speed, float lifetime, float offset = 0.0f, bool burst = false){
+               glm::vec3 direction;
+               direction.x = cosf(elevation) * sinf(heading);
+               direction.y = sinf(elevation);
+               direction.z = cosf(elevation) * cosf(heading);
+               location += offset * direction;
+               if(!burst)
+                       speed *= 2;
+               direction *= speed;
+               add_projectile(location, direction, lifetime, burst);
+       }
+       void hit_index(size_t index) override {
+               directions[index] = glm::vec3(0, 0, 0);
+       } 
+};
+
+class fragment : public loaded_object {
+public:
+       std::vector<float> life_counts;
+       std::vector<glm::vec3> trajectories;
+       fragment() : loaded_object("projectile.obj", "brick.jpg", glm::vec3(1.0f, 1.0f, 1.0f)){
+               collision_check = false;
+       }
+       
+       void create_burst(float quantity, glm::vec3 origin, float speed){
+               for(size_t i = 0; i < quantity; i++){
+                       locations.push_back(origin);
+                       life_counts.push_back(1000.0f);
+                       // One note:  This does create a cube of projectiles
+                       trajectories.push_back(glm::vec3(randvel(speed), randvel(speed), randvel(speed)) * 0.2f);
+               }
+       }
+
+       void move() override {
+               for(size_t i = 0; i < locations.size(); i++){
+                       life_counts[i] -= 0.01f;
+                       locations[i] += trajectories[i];
+                       // Is it on the ground?
+                       // Import player fall code to make this more elaborate and probably buggy
+                       if(locations[i].y <= -9.0){
+                               trajectories[i].y = fabs(trajectories[i].y);
+
+                               if(fabs(trajectories[i].x) < 0.02)
+                                       trajectories[i].x = 0.0f;
+                               else 
+                                       trajectories[i].x *= 0.95f;
+
+                               if(trajectories[i].y < 0.2)
+                                       trajectories[i].y = 0.0f;
+                               else
+                                       trajectories[i].y *= 0.8f;
+
+                               if(fabs(trajectories[i].z) < 0.02)
+                                       trajectories[i].z = 0.0f;
+                               else
+                                       trajectories[i].z *= 0.95f;
+
+                       } else { 
+                               trajectories[i].y -= 0.001;
+                       }
+               }
+       }
+               void draw(glm::mat4 vp) override {
+                       glUseProgram(program);
+                       std::vector<glm::mat4> models;
+                       models.reserve(locations.size());
+                       for(size_t i = 0; i < locations.size(); i++){
+                               glm::mat4 new_model = glm::mat4(1.0f);
+                               new_model = translate(new_model, locations[i]);
+                               if(fabs(trajectories[i].x) > 0.0f || fabs(trajectories[i].z) > 0.0f)
+                                       new_model = rotate(new_model, life_counts[i], glm::vec3(-trajectories[i].z, 0, trajectories[i].x));
+                               models.push_back(new_model);
+                       }
+                       glBindBuffer(GL_SHADER_STORAGE_BUFFER, models_buffer);
+                       glBufferData(GL_SHADER_STORAGE_BUFFER, models.size() * sizeof(glm::mat4), models.data(), GL_STATIC_DRAW);
+                       glBindBufferBase(GL_SHADER_STORAGE_BUFFER, 0, models_buffer);
+
+                       glEnableVertexAttribArray(v_attrib);
+                       glBindBuffer(GL_ARRAY_BUFFER, vbuf);
+                       glVertexAttribPointer(v_attrib, 3, GL_FLOAT, GL_FALSE, 20, 0);
+
+                       glEnableVertexAttribArray(t_attrib);
+                       glVertexAttribPointer(t_attrib, 2, GL_FLOAT, GL_FALSE, 20, (const void*)12);
+
+                       glActiveTexture(GL_TEXTURE0);
+                       glBindTexture(GL_TEXTURE_2D, tex);
+
+                       int size;
+                       glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, ebuf);
+                       glGetBufferParameteriv(GL_ELEMENT_ARRAY_BUFFER, GL_BUFFER_SIZE, &size);
+
+                       glUniformMatrix4fv(mvp_uniform, 1, 0, glm::value_ptr(vp));
+
+                       glDrawElementsInstanced(GL_TRIANGLES, size / sizeof(GLuint), GL_UNSIGNED_INT, 0, locations.size());
+               }
+       
+};
+
+

diff --git a/stb_image.cpp b/stb_image.cpp
new file mode 100644 (file)
index 0000000..6e5aa1c
--- /dev/null
+++ b/stb_image.cpp
@@ -0,0 +1,3 @@
+#define STB_IMAGE_IMPLEMENTATION
+#include "stb_image.h"
+

diff --git a/stb_image.h b/stb_image.h
new file mode 100644 (file)
index 0000000..a6202a3
--- /dev/null
+++ b/stb_image.h
@@ -0,0 +1,7547 @@
+/* stb_image - v2.22 - 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:
+
+      2.22  (2019-03-04) gif fixes, fix warnings
+      2.21  (2019-02-25) fix typo in comment
+      2.20  (2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs 
+      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
+    Carmelo J Fdez-Aguera
+
+ 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      JR Smith           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.
+//
+// ===========================================================================
+//
+// UNICODE:
+//
+//   If compiling for Windows and you wish to use Unicode filenames, compile
+//   with
+//       #define STBI_WINDOWS_UTF8
+//   and pass utf8-encoded filenames. Call stbi_convert_wchar_to_utf8 to convert
+//   Windows wchar_t filenames to utf8.
+//
+// ===========================================================================
+//
+// 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
+// provide 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 supports loading HDR images in general, and currently the Radiance
+// .HDR file format specifically. 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 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 <stdio.h>
+#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
+};
+
+#include <stdlib.h>
+typedef unsigned char stbi_uc;
+typedef unsigned short stbi_us;
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifndef STBIDEF
+#ifdef STB_IMAGE_STATIC
+#define STBIDEF static
+#else
+#define STBIDEF extern
+#endif
+#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_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
+
+#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
+
+#ifdef STBI_WINDOWS_UTF8
+STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input);
+#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 <stdarg.h>
+#include <stddef.h> // ptrdiff_t on osx
+#include <stdlib.h>
+#include <string.h>
+#include <limits.h>
+
+#if !defined(STBI_NO_LINEAR) || !defined(STBI_NO_HDR)
+#include <math.h>  // ldexp, pow
+#endif
+
+#ifndef STBI_NO_STDIO
+#include <stdio.h>
+#endif
+
+#ifndef STBI_ASSERT
+#include <assert.h>
+#define STBI_ASSERT(x) assert(x)
+#endif
+
+#ifdef __cplusplus
+#define STBI_EXTERN extern "C"
+#else
+#define STBI_EXTERN extern
+#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 <stdint.h>
+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 <emmintrin.h>
+
+#ifdef _MSC_VER
+
+#if _MSC_VER >= 1400  // not VC6
+#include <intrin.h> // __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
+
+#if !defined(STBI_NO_JPEG) && defined(STBI_SSE2)
+static int stbi__sse2_available(void)
+{
+   int info3 = stbi__cpuid3();
+   return ((info3 >> 26) & 1) != 0;
+}
+#endif
+
+#else // assume GCC-style if not VC++
+#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16)))
+
+#if !defined(STBI_NO_JPEG) && defined(STBI_SSE2)
+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
+#endif
+
+// ARM NEON
+#if defined(STBI_NO_SIMD) && defined(STBI_NEON)
+#undef STBI_NEON
+#endif
+
+#ifdef STBI_NEON
+#include <arm_neon.h>
+// 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;
+      }
+   }
+}
+
+#ifndef STBI_NO_GIF
+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; 
+   }
+}
+#endif
+
+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
+
+#if defined(_MSC_VER) && defined(STBI_WINDOWS_UTF8)
+STBI_EXTERN __declspec(dllimport) int __stdcall MultiByteToWideChar(unsigned int cp, unsigned long flags, const char *str, int cbmb, wchar_t *widestr, int cchwide);
+STBI_EXTERN __declspec(dllimport) int __stdcall WideCharToMultiByte(unsigned int cp, unsigned long flags, const wchar_t *widestr, int cchwide, char *str, int cbmb, const char *defchar, int *used_default);
+#endif
+
+#if defined(_MSC_VER) && defined(STBI_WINDOWS_UTF8)
+STBIDEF int stbi_convert_wchar_to_utf8(char *buffer, size_t bufferlen, const wchar_t* input)
+{
+       return WideCharToMultiByte(65001 /* UTF8 */, 0, input, -1, buffer, (int) bufferlen, NULL, NULL);
+}
+#endif
+
+static FILE *stbi__fopen(char const *filename, char const *mode)
+{
+   FILE *f;
+#if defined(_MSC_VER) && defined(STBI_WINDOWS_UTF8)
+   wchar_t wMode[64];
+   wchar_t wFilename[1024];
+       if (0 == MultiByteToWideChar(65001 /* UTF8 */, 0, filename, -1, wFilename, sizeof(wFilename)))
+      return 0;
+       
+       if (0 == MultiByteToWideChar(65001 /* UTF8 */, 0, mode, -1, wMode, sizeof(wMode)))
+      return 0;
+
+#if _MSC_VER >= 1400
+       if (0 != _wfopen_s(&f, wFilename, wMode))
+               f = 0;
+#else
+   f = _wfopen(wFilename, wMode);
+#endif
+
+#elif 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 (n < comp) {
+      for (i=0; i < x*y; ++i) {
+         output[i*comp + n] = data[i*comp + n]/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<<n) + 1
+static const int stbi__jbias[16] = {0,-1,-3,-7,-15,-31,-63,-127,-255,-511,-1023,-2047,-4095,-8191,-16383,-32767};
+
+// combined JPEG 'receive' and JPEG 'extend', since baseline
+// always extends everything it receives.
+stbi_inline static int stbi__extend_receive(stbi__jpeg *j, int n)
+{
+   unsigned int k;
+   int sgn;
+   if (j->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] = { NULL, NULL, NULL, NULL };
+
+      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]={0};
+   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(unsigned 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 (target == 4) out[z++] = 255;
+               if (i+1 == (int) s->img_x) break;
+               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 preceded 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; 
+   STBI_NOTUSED(req_comp);
+
+   // 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
+      if (!stbi__mad3sizes_valid(4, g->w, g->h, 0))
+         return stbi__errpuc("too large", "GIF image is too large");
+      pcount = g->w * g->h;
+      g->out = (stbi_uc *) stbi__malloc(4 * pcount);
+      g->background = (stbi_uc *) stbi__malloc(4 * pcount);
+      g->history = (stbi_uc *) stbi__malloc(pcount);
+      if (!g->out || !g->background || !g->history)
+         return stbi__errpuc("outofmem", "Out of memory");
+
+      // image is treated as "transparent" 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 the color that was there the previous frame. 
+      memset(g->out, 0x00, 4 * pcount);
+      memset(g->background, 0x00, 4 * pcount); // state of the background (starts transparent)
+      memset(g->history, 0x00, pcount);        // 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;
+
+            // if the width of the specified rectangle is 0, that means
+            // we may not see *any* pixels or the image is malformed;
+            // to make sure this is caught, move the current y down to
+            // max_y (which is what out_gif_code checks).
+            if (w == 0)
+               g->cur_y = g->max_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) 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));
+   STBI_NOTUSED(ri);
+
+   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);
+   } else if (g.out) {
+      // if there was an error and we allocated an image buffer, free it!
+      STBI_FREE(g.out);
+   }
+
+   // 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:
+      2.20  (2019-02-07) support utf8 filenames in Windows; fix warnings and platform ifdefs 
+      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/tile_floor.h b/tile_floor.h
new file mode 100644 (file)
index 0000000..592b6c9
--- /dev/null
+++ b/tile_floor.h
@@ -0,0 +1,53 @@
+#pragma once
+
+class tile_floor : public gameobject {
+       public:
+               unsigned int mvp_uniform, anim_uniform, v_attrib, c_attrib, program, vbuf, cbuf, ebuf;
+               int init() override {
+                       // Initialization part
+                       float vertices[] = {
+                               1.0, -10,  1.0,
+                               1.0, -10,  -1.0,
+                               -1.0, -10,  -1.0,
+                               -1.0, -10,  1.0,
+                       };
+                       glGenBuffers(1, &vbuf);
+                       glBindBuffer(GL_SHADER_STORAGE_BUFFER, vbuf);
+                       glBufferData(GL_SHADER_STORAGE_BUFFER, sizeof(vertices), vertices, GL_STATIC_DRAW);
+
+                       GLushort floor_elements[] = {
+                               0, 1, 2, 2, 3, 0,
+
+                       };
+                       glGenBuffers(1, &ebuf);
+                       glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, ebuf);
+                       glBufferData(GL_ELEMENT_ARRAY_BUFFER, sizeof(floor_elements), floor_elements, GL_STATIC_DRAW);
+
+                       program = make_program("floor_vertex_shader.glsl",0, 0, 0, "floor_fragment_shader.glsl");
+                       if (!program)
+                               return 1;
+
+                       v_attrib = glGetAttribLocation(program, "in_vertex");
+                       c_attrib = glGetAttribLocation(program, "in_color");
+                       mvp_uniform = glGetUniformLocation(program, "mvp");
+                       return 0;
+               }
+               void draw(glm::mat4 vp) override {
+                       glUseProgram(program);
+
+                       glEnableVertexAttribArray(v_attrib);
+                       glBindBuffer(GL_ARRAY_BUFFER, vbuf);
+                       glVertexAttribPointer(v_attrib, 3, GL_FLOAT, GL_FALSE, 0, 0);
+
+                       int size;
+                       glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, ebuf);
+                       glGetBufferParameteriv(GL_ELEMENT_ARRAY_BUFFER, GL_BUFFER_SIZE, &size);
+
+                       glUniformMatrix4fv(mvp_uniform, 1, 0, glm::value_ptr(vp));
+
+                       glDrawElementsInstanced(GL_TRIANGLES, size / sizeof(uint16_t), GL_UNSIGNED_SHORT, 0, 10000);
+               }
+};
+
+
+

diff --git a/tiny_obj_loader.cc b/tiny_obj_loader.cc
new file mode 100644 (file)
index 0000000..e57d044
--- /dev/null
+++ b/tiny_obj_loader.cc
@@ -0,0 +1,2 @@
+#define TINYOBJLOADER_IMPLEMENTATION
+#include "tiny_obj_loader.h"

diff --git a/tiny_obj_loader.h b/tiny_obj_loader.h
new file mode 100644 (file)
index 0000000..4e2f6b5
--- /dev/null
+++ b/tiny_obj_loader.h
@@ -0,0 +1,2020 @@
+/*
+The MIT License (MIT)
+
+Copyright (c) 2012-2017 Syoyo Fujita and many contributors.
+
+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.
+*/
+
+//
+// version 1.0.8 : Fix parsing `g` tag just after `usemtl`(#138)
+// version 1.0.7 : Support multiple tex options(#126)
+// version 1.0.6 : Add TINYOBJLOADER_USE_DOUBLE option(#124)
+// version 1.0.5 : Ignore `Tr` when `d` exists in MTL(#43)
+// version 1.0.4 : Support multiple filenames for 'mtllib'(#112)
+// version 1.0.3 : Support parsing texture options(#85)
+// version 1.0.2 : Improve parsing speed by about a factor of 2 for large
+// files(#105)
+// version 1.0.1 : Fixes a shape is lost if obj ends with a 'usemtl'(#104)
+// version 1.0.0 : Change data structure. Change license from BSD to MIT.
+//
+
+//
+// Use this in *one* .cc
+//   #define TINYOBJLOADER_IMPLEMENTATION
+//   #include "tiny_obj_loader.h"
+//
+
+#ifndef TINY_OBJ_LOADER_H_
+#define TINY_OBJ_LOADER_H_
+
+#include <map>
+#include <string>
+#include <vector>
+
+namespace tinyobj {
+
+// https://en.wikipedia.org/wiki/Wavefront_.obj_file says ...
+//
+//  -blendu on | off                       # set horizontal texture blending
+//  (default on)
+//  -blendv on | off                       # set vertical texture blending
+//  (default on)
+//  -boost real_value                      # boost mip-map sharpness
+//  -mm base_value gain_value              # modify texture map values (default
+//  0 1)
+//                                         #     base_value = brightness,
+//                                         gain_value = contrast
+//  -o u [v [w]]                           # Origin offset             (default
+//  0 0 0)
+//  -s u [v [w]]                           # Scale                     (default
+//  1 1 1)
+//  -t u [v [w]]                           # Turbulence                (default
+//  0 0 0)
+//  -texres resolution                     # texture resolution to create
+//  -clamp on | off                        # only render texels in the clamped
+//  0-1 range (default off)
+//                                         #   When unclamped, textures are
+//                                         repeated across a surface,
+//                                         #   when clamped, only texels which
+//                                         fall within the 0-1
+//                                         #   range are rendered.
+//  -bm mult_value                         # bump multiplier (for bump maps
+//  only)
+//
+//  -imfchan r | g | b | m | l | z         # specifies which channel of the file
+//  is used to
+//                                         # create a scalar or bump texture.
+//                                         r:red, g:green,
+//                                         # b:blue, m:matte, l:luminance,
+//                                         z:z-depth..
+//                                         # (the default for bump is 'l' and
+//                                         for decal is 'm')
+//  bump -imfchan r bumpmap.tga            # says to use the red channel of
+//  bumpmap.tga as the bumpmap
+//
+// For reflection maps...
+//
+//   -type sphere                           # specifies a sphere for a "refl"
+//   reflection map
+//   -type cube_top    | cube_bottom |      # when using a cube map, the texture
+//   file for each
+//         cube_front  | cube_back   |      # side of the cube is specified
+//         separately
+//         cube_left   | cube_right
+
+#ifdef TINYOBJLOADER_USE_DOUBLE
+  //#pragma message "using double"
+  typedef double real_t;
+#else
+  //#pragma message "using float"
+  typedef float real_t;
+#endif
+
+typedef enum {
+  TEXTURE_TYPE_NONE,  // default
+  TEXTURE_TYPE_SPHERE,
+  TEXTURE_TYPE_CUBE_TOP,
+  TEXTURE_TYPE_CUBE_BOTTOM,
+  TEXTURE_TYPE_CUBE_FRONT,
+  TEXTURE_TYPE_CUBE_BACK,
+  TEXTURE_TYPE_CUBE_LEFT,
+  TEXTURE_TYPE_CUBE_RIGHT
+} texture_type_t;
+
+typedef struct {
+  texture_type_t type;     // -type (default TEXTURE_TYPE_NONE)
+  real_t sharpness;         // -boost (default 1.0?)
+  real_t brightness;        // base_value in -mm option (default 0)
+  real_t contrast;          // gain_value in -mm option (default 1)
+  real_t origin_offset[3];  // -o u [v [w]] (default 0 0 0)
+  real_t scale[3];          // -s u [v [w]] (default 1 1 1)
+  real_t turbulence[3];     // -t u [v [w]] (default 0 0 0)
+  // int   texture_resolution; // -texres resolution (default = ?) TODO
+  bool clamp;    // -clamp (default false)
+  char imfchan;  // -imfchan (the default for bump is 'l' and for decal is 'm')
+  bool blendu;   // -blendu (default on)
+  bool blendv;   // -blendv (default on)
+  real_t bump_multiplier;  // -bm (for bump maps only, default 1.0)
+} texture_option_t;
+
+typedef struct {
+  std::string name;
+
+  real_t ambient[3];
+  real_t diffuse[3];
+  real_t specular[3];
+  real_t transmittance[3];
+  real_t emission[3];
+  real_t shininess;
+  real_t ior;       // index of refraction
+  real_t dissolve;  // 1 == opaque; 0 == fully transparent
+  // illumination model (see http://www.fileformat.info/format/material/)
+  int illum;
+
+  int dummy;  // Suppress padding warning.
+
+  std::string ambient_texname;             // map_Ka
+  std::string diffuse_texname;             // map_Kd
+  std::string specular_texname;            // map_Ks
+  std::string specular_highlight_texname;  // map_Ns
+  std::string bump_texname;                // map_bump, map_Bump, bump
+  std::string displacement_texname;        // disp
+  std::string alpha_texname;               // map_d
+  std::string reflection_texname;          // refl
+
+  texture_option_t ambient_texopt;
+  texture_option_t diffuse_texopt;
+  texture_option_t specular_texopt;
+  texture_option_t specular_highlight_texopt;
+  texture_option_t bump_texopt;
+  texture_option_t displacement_texopt;
+  texture_option_t alpha_texopt;
+  texture_option_t reflection_texopt;
+
+  // PBR extension
+  // http://exocortex.com/blog/extending_wavefront_mtl_to_support_pbr
+  real_t roughness;            // [0, 1] default 0
+  real_t metallic;             // [0, 1] default 0
+  real_t sheen;                // [0, 1] default 0
+  real_t clearcoat_thickness;  // [0, 1] default 0
+  real_t clearcoat_roughness;  // [0, 1] default 0
+  real_t anisotropy;           // aniso. [0, 1] default 0
+  real_t anisotropy_rotation;  // anisor. [0, 1] default 0
+  real_t pad0;
+  std::string roughness_texname;  // map_Pr
+  std::string metallic_texname;   // map_Pm
+  std::string sheen_texname;      // map_Ps
+  std::string emissive_texname;   // map_Ke
+  std::string normal_texname;     // norm. For normal mapping.
+
+  texture_option_t roughness_texopt;
+  texture_option_t metallic_texopt;
+  texture_option_t sheen_texopt;
+  texture_option_t emissive_texopt;
+  texture_option_t normal_texopt;
+
+  int pad2;
+
+  std::map<std::string, std::string> unknown_parameter;
+} material_t;
+
+typedef struct {
+  std::string name;
+
+  std::vector<int> intValues;
+  std::vector<real_t> floatValues;
+  std::vector<std::string> stringValues;
+} tag_t;
+
+// Index struct to support different indices for vtx/normal/texcoord.
+// -1 means not used.
+typedef struct {
+  int vertex_index;
+  int normal_index;
+  int texcoord_index;
+} index_t;
+
+typedef struct {
+  std::vector<index_t> indices;
+  std::vector<unsigned char> num_face_vertices;  // The number of vertices per
+                                                 // face. 3 = polygon, 4 = quad,
+                                                 // ... Up to 255.
+  std::vector<int> material_ids;                 // per-face material ID
+  std::vector<tag_t> tags;                       // SubD tag
+} mesh_t;
+
+typedef struct {
+  std::string name;
+  mesh_t mesh;
+} shape_t;
+
+// Vertex attributes
+typedef struct {
+  std::vector<real_t> vertices;   // 'v'
+  std::vector<real_t> normals;    // 'vn'
+  std::vector<real_t> texcoords;  // 'vt'
+} attrib_t;
+
+typedef struct callback_t_ {
+  // W is optional and set to 1 if there is no `w` item in `v` line
+  void (*vertex_cb)(void *user_data, real_t x, real_t y, real_t z, real_t w);
+  void (*normal_cb)(void *user_data, real_t x, real_t y, real_t z);
+
+  // y and z are optional and set to 0 if there is no `y` and/or `z` item(s) in
+  // `vt` line.
+  void (*texcoord_cb)(void *user_data, real_t x, real_t y, real_t z);
+
+  // called per 'f' line. num_indices is the number of face indices(e.g. 3 for
+  // triangle, 4 for quad)
+  // 0 will be passed for undefined index in index_t members.
+  void (*index_cb)(void *user_data, index_t *indices, int num_indices);
+  // `name` material name, `material_id` = the array index of material_t[]. -1
+  // if
+  // a material not found in .mtl
+  void (*usemtl_cb)(void *user_data, const char *name, int material_id);
+  // `materials` = parsed material data.
+  void (*mtllib_cb)(void *user_data, const material_t *materials,
+                    int num_materials);
+  // There may be multiple group names
+  void (*group_cb)(void *user_data, const char **names, int num_names);
+  void (*object_cb)(void *user_data, const char *name);
+
+  callback_t_()
+      : vertex_cb(NULL),
+        normal_cb(NULL),
+        texcoord_cb(NULL),
+        index_cb(NULL),
+        usemtl_cb(NULL),
+        mtllib_cb(NULL),
+        group_cb(NULL),
+        object_cb(NULL) {}
+} callback_t;
+
+class MaterialReader {
+ public:
+  MaterialReader() {}
+  virtual ~MaterialReader();
+
+  virtual bool operator()(const std::string &matId,
+                          std::vector<material_t> *materials,
+                          std::map<std::string, int> *matMap,
+                          std::string *err) = 0;
+};
+
+class MaterialFileReader : public MaterialReader {
+ public:
+  explicit MaterialFileReader(const std::string &mtl_basedir)
+      : m_mtlBaseDir(mtl_basedir) {}
+  virtual ~MaterialFileReader() {}
+  virtual bool operator()(const std::string &matId,
+                          std::vector<material_t> *materials,
+                          std::map<std::string, int> *matMap, std::string *err);
+
+ private:
+  std::string m_mtlBaseDir;
+};
+
+class MaterialStreamReader : public MaterialReader {
+ public:
+  explicit MaterialStreamReader(std::istream &inStream)
+      : m_inStream(inStream) {}
+  virtual ~MaterialStreamReader() {}
+  virtual bool operator()(const std::string &matId,
+                          std::vector<material_t> *materials,
+                          std::map<std::string, int> *matMap, std::string *err);
+
+ private:
+  std::istream &m_inStream;
+};
+
+/// Loads .obj from a file.
+/// 'attrib', 'shapes' and 'materials' will be filled with parsed shape data
+/// 'shapes' will be filled with parsed shape data
+/// Returns true when loading .obj become success.
+/// Returns warning and error message into `err`
+/// 'mtl_basedir' is optional, and used for base directory for .mtl file.
+/// In default(`NULL'), .mtl file is searched from an application's working
+/// directory.
+/// 'triangulate' is optional, and used whether triangulate polygon face in .obj
+/// or not.
+bool LoadObj(attrib_t *attrib, std::vector<shape_t> *shapes,
+             std::vector<material_t> *materials, std::string *err,
+             const char *filename, const char *mtl_basedir = NULL,
+             bool triangulate = true);
+
+/// Loads .obj from a file with custom user callback.
+/// .mtl is loaded as usual and parsed material_t data will be passed to
+/// `callback.mtllib_cb`.
+/// Returns true when loading .obj/.mtl become success.
+/// Returns warning and error message into `err`
+/// See `examples/callback_api/` for how to use this function.
+bool LoadObjWithCallback(std::istream &inStream, const callback_t &callback,
+                         void *user_data = NULL,
+                         MaterialReader *readMatFn = NULL,
+                         std::string *err = NULL);
+
+/// Loads object from a std::istream, uses GetMtlIStreamFn to retrieve
+/// std::istream for materials.
+/// Returns true when loading .obj become success.
+/// Returns warning and error message into `err`
+bool LoadObj(attrib_t *attrib, std::vector<shape_t> *shapes,
+             std::vector<material_t> *materials, std::string *err,
+             std::istream *inStream, MaterialReader *readMatFn = NULL,
+             bool triangulate = true);
+
+/// Loads materials into std::map
+void LoadMtl(std::map<std::string, int> *material_map,
+             std::vector<material_t> *materials, std::istream *inStream,
+             std::string *warning);
+
+}  // namespace tinyobj
+
+#endif  // TINY_OBJ_LOADER_H_
+
+#ifdef TINYOBJLOADER_IMPLEMENTATION
+#include <cassert>
+#include <cctype>
+#include <cmath>
+#include <cstddef>
+#include <cstdlib>
+#include <cstring>
+#include <utility>
+
+#include <fstream>
+#include <sstream>
+
+namespace tinyobj {
+
+MaterialReader::~MaterialReader() {}
+
+
+struct vertex_index {
+  int v_idx, vt_idx, vn_idx;
+  vertex_index() : v_idx(-1), vt_idx(-1), vn_idx(-1) {}
+  explicit vertex_index(int idx) : v_idx(idx), vt_idx(idx), vn_idx(idx) {}
+  vertex_index(int vidx, int vtidx, int vnidx)
+      : v_idx(vidx), vt_idx(vtidx), vn_idx(vnidx) {}
+};
+
+struct tag_sizes {
+  tag_sizes() : num_ints(0), num_reals(0), num_strings(0) {}
+  int num_ints;
+  int num_reals;
+  int num_strings;
+};
+
+struct obj_shape {
+  std::vector<real_t> v;
+  std::vector<real_t> vn;
+  std::vector<real_t> vt;
+};
+
+// See
+// http://stackoverflow.com/questions/6089231/getting-std-ifstream-to-handle-lf-cr-and-crlf
+static std::istream &safeGetline(std::istream &is, std::string &t) {
+  t.clear();
+
+  // The characters in the stream are read one-by-one using a std::streambuf.
+  // That is faster than reading them one-by-one using the std::istream.
+  // Code that uses streambuf this way must be guarded by a sentry object.
+  // The sentry object performs various tasks,
+  // such as thread synchronization and updating the stream state.
+
+  std::istream::sentry se(is, true);
+  std::streambuf *sb = is.rdbuf();
+
+  if (se) {
+    for (;;) {
+      int c = sb->sbumpc();
+      switch (c) {
+        case '\n':
+          return is;
+        case '\r':
+          if (sb->sgetc() == '\n') sb->sbumpc();
+          return is;
+        case EOF:
+          // Also handle the case when the last line has no line ending
+          if (t.empty()) is.setstate(std::ios::eofbit);
+          return is;
+        default:
+          t += static_cast<char>(c);
+      }
+    }
+  }
+
+  return is;
+}
+
+#define IS_SPACE(x) (((x) == ' ') || ((x) == '\t'))
+#define IS_DIGIT(x) \
+  (static_cast<unsigned int>((x) - '0') < static_cast<unsigned int>(10))
+#define IS_NEW_LINE(x) (((x) == '\r') || ((x) == '\n') || ((x) == '\0'))
+
+// Make index zero-base, and also support relative index.
+static inline int fixIndex(int idx, int n) {
+  if (idx > 0) return idx - 1;
+  if (idx == 0) return 0;
+  return n + idx;  // negative value = relative
+}
+
+static inline std::string parseString(const char **token) {
+  std::string s;
+  (*token) += strspn((*token), " \t");
+  size_t e = strcspn((*token), " \t\r");
+  s = std::string((*token), &(*token)[e]);
+  (*token) += e;
+  return s;
+}
+
+static inline int parseInt(const char **token) {
+  (*token) += strspn((*token), " \t");
+  int i = atoi((*token));
+  (*token) += strcspn((*token), " \t\r");
+  return i;
+}
+
+// Tries to parse a floating point number located at s.
+//
+// s_end should be a location in the string where reading should absolutely
+// stop. For example at the end of the string, to prevent buffer overflows.
+//
+// Parses the following EBNF grammar:
+//   sign    = "+" | "-" ;
+//   END     = ? anything not in digit ?
+//   digit   = "0" | "1" | "2" | "3" | "4" | "5" | "6" | "7" | "8" | "9" ;
+//   integer = [sign] , digit , {digit} ;
+//   decimal = integer , ["." , integer] ;
+//   float   = ( decimal , END ) | ( decimal , ("E" | "e") , integer , END ) ;
+//
+//  Valid strings are for example:
+//   -0  +3.1417e+2  -0.0E-3  1.0324  -1.41   11e2
+//
+// If the parsing is a success, result is set to the parsed value and true
+// is returned.
+//
+// The function is greedy and will parse until any of the following happens:
+//  - a non-conforming character is encountered.
+//  - s_end is reached.
+//
+// The following situations triggers a failure:
+//  - s >= s_end.
+//  - parse failure.
+//
+static bool tryParseDouble(const char *s, const char *s_end, double *result) {
+  if (s >= s_end) {
+    return false;
+  }
+
+  double mantissa = 0.0;
+  // This exponent is base 2 rather than 10.
+  // However the exponent we parse is supposed to be one of ten,
+  // thus we must take care to convert the exponent/and or the
+  // mantissa to a * 2^E, where a is the mantissa and E is the
+  // exponent.
+  // To get the final double we will use ldexp, it requires the
+  // exponent to be in base 2.
+  int exponent = 0;
+
+  // NOTE: THESE MUST BE DECLARED HERE SINCE WE ARE NOT ALLOWED
+  // TO JUMP OVER DEFINITIONS.
+  char sign = '+';
+  char exp_sign = '+';
+  char const *curr = s;
+
+  // How many characters were read in a loop.
+  int read = 0;
+  // Tells whether a loop terminated due to reaching s_end.
+  bool end_not_reached = false;
+
+  /*
+          BEGIN PARSING.
+  */
+
+  // Find out what sign we've got.
+  if (*curr == '+' || *curr == '-') {
+    sign = *curr;
+    curr++;
+  } else if (IS_DIGIT(*curr)) { /* Pass through. */
+  } else {
+    goto fail;
+  }
+
+  // Read the integer part.
+  end_not_reached = (curr != s_end);
+  while (end_not_reached && IS_DIGIT(*curr)) {
+    mantissa *= 10;
+    mantissa += static_cast<int>(*curr - 0x30);
+    curr++;
+    read++;
+    end_not_reached = (curr != s_end);
+  }
+
+  // We must make sure we actually got something.
+  if (read == 0) goto fail;
+  // We allow numbers of form "#", "###" etc.
+  if (!end_not_reached) goto assemble;
+
+  // Read the decimal part.
+  if (*curr == '.') {
+    curr++;
+    read = 1;
+    end_not_reached = (curr != s_end);
+    while (end_not_reached && IS_DIGIT(*curr)) {
+      static const double pow_lut[] = {
+          1.0, 0.1, 0.01, 0.001, 0.0001, 0.00001, 0.000001, 0.0000001,
+      };
+      const int lut_entries = sizeof pow_lut / sizeof pow_lut[0];
+
+      // NOTE: Don't use powf here, it will absolutely murder precision.
+      mantissa += static_cast<int>(*curr - 0x30) *
+                  (read < lut_entries ? pow_lut[read] : std::pow(10.0, -read));
+      read++;
+      curr++;
+      end_not_reached = (curr != s_end);
+    }
+  } else if (*curr == 'e' || *curr == 'E') {
+  } else {
+    goto assemble;
+  }
+
+  if (!end_not_reached) goto assemble;
+
+  // Read the exponent part.
+  if (*curr == 'e' || *curr == 'E') {
+    curr++;
+    // Figure out if a sign is present and if it is.
+    end_not_reached = (curr != s_end);
+    if (end_not_reached && (*curr == '+' || *curr == '-')) {
+      exp_sign = *curr;
+      curr++;
+    } else if (IS_DIGIT(*curr)) { /* Pass through. */
+    } else {
+      // Empty E is not allowed.
+      goto fail;
+    }
+
+    read = 0;
+    end_not_reached = (curr != s_end);
+    while (end_not_reached && IS_DIGIT(*curr)) {
+      exponent *= 10;
+      exponent += static_cast<int>(*curr - 0x30);
+      curr++;
+      read++;
+      end_not_reached = (curr != s_end);
+    }
+    exponent *= (exp_sign == '+' ? 1 : -1);
+    if (read == 0) goto fail;
+  }
+
+assemble:
+  *result =
+      (sign == '+' ? 1 : -1) *
+      (exponent ? std::ldexp(mantissa * std::pow(5.0, exponent), exponent) : mantissa);
+  return true;
+fail:
+  return false;
+}
+
+static inline real_t parseReal(const char **token, double default_value = 0.0) {
+  (*token) += strspn((*token), " \t");
+  const char *end = (*token) + strcspn((*token), " \t\r");
+  double val = default_value;
+  tryParseDouble((*token), end, &val);
+  real_t f = static_cast<real_t>(val);
+  (*token) = end;
+  return f;
+}
+
+static inline void parseReal2(real_t *x, real_t *y, const char **token,
+                               const double default_x = 0.0,
+                               const double default_y = 0.0) {
+  (*x) = parseReal(token, default_x);
+  (*y) = parseReal(token, default_y);
+}
+
+static inline void parseReal3(real_t *x, real_t *y, real_t *z, const char **token,
+                               const double default_x = 0.0,
+                               const double default_y = 0.0,
+                               const double default_z = 0.0) {
+  (*x) = parseReal(token, default_x);
+  (*y) = parseReal(token, default_y);
+  (*z) = parseReal(token, default_z);
+}
+
+static inline void parseV(real_t *x, real_t *y, real_t *z, real_t *w,
+                          const char **token, const double default_x = 0.0,
+                          const double default_y = 0.0,
+                          const double default_z = 0.0,
+                          const double default_w = 1.0) {
+  (*x) = parseReal(token, default_x);
+  (*y) = parseReal(token, default_y);
+  (*z) = parseReal(token, default_z);
+  (*w) = parseReal(token, default_w);
+}
+
+static inline bool parseOnOff(const char **token, bool default_value = true) {
+  (*token) += strspn((*token), " \t");
+  const char *end = (*token) + strcspn((*token), " \t\r");
+
+  bool ret = default_value;
+  if ((0 == strncmp((*token), "on", 2))) {
+    ret = true;
+  } else if ((0 == strncmp((*token), "off", 3))) {
+    ret = false;
+  }
+
+  (*token) = end;
+  return ret;
+}
+
+static inline texture_type_t parseTextureType(
+    const char **token, texture_type_t default_value = TEXTURE_TYPE_NONE) {
+  (*token) += strspn((*token), " \t");
+  const char *end = (*token) + strcspn((*token), " \t\r");
+  texture_type_t ty = default_value;
+
+  if ((0 == strncmp((*token), "cube_top", strlen("cube_top")))) {
+    ty = TEXTURE_TYPE_CUBE_TOP;
+  } else if ((0 == strncmp((*token), "cube_bottom", strlen("cube_bottom")))) {
+    ty = TEXTURE_TYPE_CUBE_BOTTOM;
+  } else if ((0 == strncmp((*token), "cube_left", strlen("cube_left")))) {
+    ty = TEXTURE_TYPE_CUBE_LEFT;
+  } else if ((0 == strncmp((*token), "cube_right", strlen("cube_right")))) {
+    ty = TEXTURE_TYPE_CUBE_RIGHT;
+  } else if ((0 == strncmp((*token), "cube_front", strlen("cube_front")))) {
+    ty = TEXTURE_TYPE_CUBE_FRONT;
+  } else if ((0 == strncmp((*token), "cube_back", strlen("cube_back")))) {
+    ty = TEXTURE_TYPE_CUBE_BACK;
+  } else if ((0 == strncmp((*token), "sphere", strlen("sphere")))) {
+    ty = TEXTURE_TYPE_SPHERE;
+  }
+
+  (*token) = end;
+  return ty;
+}
+
+static tag_sizes parseTagTriple(const char **token) {
+  tag_sizes ts;
+
+  ts.num_ints = atoi((*token));
+  (*token) += strcspn((*token), "/ \t\r");
+  if ((*token)[0] != '/') {
+    return ts;
+  }
+  (*token)++;
+
+  ts.num_reals = atoi((*token));
+  (*token) += strcspn((*token), "/ \t\r");
+  if ((*token)[0] != '/') {
+    return ts;
+  }
+  (*token)++;
+
+  ts.num_strings = atoi((*token));
+  (*token) += strcspn((*token), "/ \t\r") + 1;
+
+  return ts;
+}
+
+// Parse triples with index offsets: i, i/j/k, i//k, i/j
+static vertex_index parseTriple(const char **token, int vsize, int vnsize,
+                                int vtsize) {
+  vertex_index vi(-1);
+
+  vi.v_idx = fixIndex(atoi((*token)), vsize);
+  (*token) += strcspn((*token), "/ \t\r");
+  if ((*token)[0] != '/') {
+    return vi;
+  }
+  (*token)++;
+
+  // i//k
+  if ((*token)[0] == '/') {
+    (*token)++;
+    vi.vn_idx = fixIndex(atoi((*token)), vnsize);
+    (*token) += strcspn((*token), "/ \t\r");
+    return vi;
+  }
+
+  // i/j/k or i/j
+  vi.vt_idx = fixIndex(atoi((*token)), vtsize);
+  (*token) += strcspn((*token), "/ \t\r");
+  if ((*token)[0] != '/') {
+    return vi;
+  }
+
+  // i/j/k
+  (*token)++;  // skip '/'
+  vi.vn_idx = fixIndex(atoi((*token)), vnsize);
+  (*token) += strcspn((*token), "/ \t\r");
+  return vi;
+}
+
+// Parse raw triples: i, i/j/k, i//k, i/j
+static vertex_index parseRawTriple(const char **token) {
+  vertex_index vi(static_cast<int>(0));  // 0 is an invalid index in OBJ
+
+  vi.v_idx = atoi((*token));
+  (*token) += strcspn((*token), "/ \t\r");
+  if ((*token)[0] != '/') {
+    return vi;
+  }
+  (*token)++;
+
+  // i//k
+  if ((*token)[0] == '/') {
+    (*token)++;
+    vi.vn_idx = atoi((*token));
+    (*token) += strcspn((*token), "/ \t\r");
+    return vi;
+  }
+
+  // i/j/k or i/j
+  vi.vt_idx = atoi((*token));
+  (*token) += strcspn((*token), "/ \t\r");
+  if ((*token)[0] != '/') {
+    return vi;
+  }
+
+  // i/j/k
+  (*token)++;  // skip '/'
+  vi.vn_idx = atoi((*token));
+  (*token) += strcspn((*token), "/ \t\r");
+  return vi;
+}
+
+static bool ParseTextureNameAndOption(std::string *texname,
+                                      texture_option_t *texopt,
+                                      const char *linebuf, const bool is_bump) {
+  // @todo { write more robust lexer and parser. }
+  bool found_texname = false;
+  std::string texture_name;
+
+  // Fill with default value for texopt.
+  if (is_bump) {
+    texopt->imfchan = 'l';
+  } else {
+    texopt->imfchan = 'm';
+  }
+  texopt->bump_multiplier = 1.0f;
+  texopt->clamp = false;
+  texopt->blendu = true;
+  texopt->blendv = true;
+  texopt->sharpness = 1.0f;
+  texopt->brightness = 0.0f;
+  texopt->contrast = 1.0f;
+  texopt->origin_offset[0] = 0.0f;
+  texopt->origin_offset[1] = 0.0f;
+  texopt->origin_offset[2] = 0.0f;
+  texopt->scale[0] = 1.0f;
+  texopt->scale[1] = 1.0f;
+  texopt->scale[2] = 1.0f;
+  texopt->turbulence[0] = 0.0f;
+  texopt->turbulence[1] = 0.0f;
+  texopt->turbulence[2] = 0.0f;
+  texopt->type = TEXTURE_TYPE_NONE;
+
+  const char *token = linebuf;  // Assume line ends with NULL
+
+  while (!IS_NEW_LINE((*token))) {
+    token += strspn(token, " \t");  // skip space
+    if ((0 == strncmp(token, "-blendu", 7)) && IS_SPACE((token[7]))) {
+      token += 8;
+      texopt->blendu = parseOnOff(&token, /* default */ true);
+    } else if ((0 == strncmp(token, "-blendv", 7)) && IS_SPACE((token[7]))) {
+      token += 8;
+      texopt->blendv = parseOnOff(&token, /* default */ true);
+    } else if ((0 == strncmp(token, "-clamp", 6)) && IS_SPACE((token[6]))) {
+      token += 7;
+      texopt->clamp = parseOnOff(&token, /* default */ true);
+    } else if ((0 == strncmp(token, "-boost", 6)) && IS_SPACE((token[6]))) {
+      token += 7;
+      texopt->sharpness = parseReal(&token, 1.0);
+    } else if ((0 == strncmp(token, "-bm", 3)) && IS_SPACE((token[3]))) {
+      token += 4;
+      texopt->bump_multiplier = parseReal(&token, 1.0);
+    } else if ((0 == strncmp(token, "-o", 2)) && IS_SPACE((token[2]))) {
+      token += 3;
+      parseReal3(&(texopt->origin_offset[0]), &(texopt->origin_offset[1]),
+                  &(texopt->origin_offset[2]), &token);
+    } else if ((0 == strncmp(token, "-s", 2)) && IS_SPACE((token[2]))) {
+      token += 3;
+      parseReal3(&(texopt->scale[0]), &(texopt->scale[1]), &(texopt->scale[2]),
+                  &token, 1.0, 1.0, 1.0);
+    } else if ((0 == strncmp(token, "-t", 2)) && IS_SPACE((token[2]))) {
+      token += 3;
+      parseReal3(&(texopt->turbulence[0]), &(texopt->turbulence[1]),
+                  &(texopt->turbulence[2]), &token);
+    } else if ((0 == strncmp(token, "-type", 5)) && IS_SPACE((token[5]))) {
+      token += 5;
+      texopt->type = parseTextureType((&token), TEXTURE_TYPE_NONE);
+    } else if ((0 == strncmp(token, "-imfchan", 8)) && IS_SPACE((token[8]))) {
+      token += 9;
+      token += strspn(token, " \t");
+      const char *end = token + strcspn(token, " \t\r");
+      if ((end - token) == 1) {  // Assume one char for -imfchan
+        texopt->imfchan = (*token);
+      }
+      token = end;
+    } else if ((0 == strncmp(token, "-mm", 3)) && IS_SPACE((token[3]))) {
+      token += 4;
+      parseReal2(&(texopt->brightness), &(texopt->contrast), &token, 0.0, 1.0);
+    } else {
+      // Assume texture filename
+      size_t len = strcspn(token, " \t\r");  // untile next space
+      texture_name = std::string(token, token + len);
+      token += len;
+
+      token += strspn(token, " \t");  // skip space
+
+      found_texname = true;
+    }
+  }
+
+  if (found_texname) {
+    (*texname) = texture_name;
+    return true;
+  } else {
+    return false;
+  }
+}
+
+static void InitMaterial(material_t *material) {
+  material->name = "";
+  material->ambient_texname = "";
+  material->diffuse_texname = "";
+  material->specular_texname = "";
+  material->specular_highlight_texname = "";
+  material->bump_texname = "";
+  material->displacement_texname = "";
+  material->reflection_texname = "";
+  material->alpha_texname = "";
+  for (int i = 0; i < 3; i++) {
+    material->ambient[i] = 0.f;
+    material->diffuse[i] = 0.f;
+    material->specular[i] = 0.f;
+    material->transmittance[i] = 0.f;
+    material->emission[i] = 0.f;
+  }
+  material->illum = 0;
+  material->dissolve = 1.f;
+  material->shininess = 1.f;
+  material->ior = 1.f;
+
+  material->roughness = 0.f;
+  material->metallic = 0.f;
+  material->sheen = 0.f;
+  material->clearcoat_thickness = 0.f;
+  material->clearcoat_roughness = 0.f;
+  material->anisotropy_rotation = 0.f;
+  material->anisotropy = 0.f;
+  material->roughness_texname = "";
+  material->metallic_texname = "";
+  material->sheen_texname = "";
+  material->emissive_texname = "";
+  material->normal_texname = "";
+
+  material->unknown_parameter.clear();
+}
+
+static bool exportFaceGroupToShape(
+    shape_t *shape, const std::vector<std::vector<vertex_index> > &faceGroup,
+    const std::vector<tag_t> &tags, const int material_id,
+    const std::string &name, bool triangulate) {
+  if (faceGroup.empty()) {
+    return false;
+  }
+
+  // Flatten vertices and indices
+  for (size_t i = 0; i < faceGroup.size(); i++) {
+    const std::vector<vertex_index> &face = faceGroup[i];
+
+    vertex_index i0 = face[0];
+    vertex_index i1(-1);
+    vertex_index i2 = face[1];
+
+    size_t npolys = face.size();
+
+    if (triangulate) {
+      // Polygon -> triangle fan conversion
+      for (size_t k = 2; k < npolys; k++) {
+        i1 = i2;
+        i2 = face[k];
+
+        index_t idx0, idx1, idx2;
+        idx0.vertex_index = i0.v_idx;
+        idx0.normal_index = i0.vn_idx;
+        idx0.texcoord_index = i0.vt_idx;
+        idx1.vertex_index = i1.v_idx;
+        idx1.normal_index = i1.vn_idx;
+        idx1.texcoord_index = i1.vt_idx;
+        idx2.vertex_index = i2.v_idx;
+        idx2.normal_index = i2.vn_idx;
+        idx2.texcoord_index = i2.vt_idx;
+
+        shape->mesh.indices.push_back(idx0);
+        shape->mesh.indices.push_back(idx1);
+        shape->mesh.indices.push_back(idx2);
+
+        shape->mesh.num_face_vertices.push_back(3);
+        shape->mesh.material_ids.push_back(material_id);
+      }
+    } else {
+      for (size_t k = 0; k < npolys; k++) {
+        index_t idx;
+        idx.vertex_index = face[k].v_idx;
+        idx.normal_index = face[k].vn_idx;
+        idx.texcoord_index = face[k].vt_idx;
+        shape->mesh.indices.push_back(idx);
+      }
+
+      shape->mesh.num_face_vertices.push_back(
+          static_cast<unsigned char>(npolys));
+      shape->mesh.material_ids.push_back(material_id);  // per face
+    }
+  }
+
+  shape->name = name;
+  shape->mesh.tags = tags;
+
+  return true;
+}
+
+// Split a string with specified delimiter character.
+// http://stackoverflow.com/questions/236129/split-a-string-in-c
+static void SplitString(const std::string &s, char delim,
+                        std::vector<std::string> &elems) {
+  std::stringstream ss;
+  ss.str(s);
+  std::string item;
+  while (std::getline(ss, item, delim)) {
+    elems.push_back(item);
+  }
+}
+
+void LoadMtl(std::map<std::string, int> *material_map,
+             std::vector<material_t> *materials, std::istream *inStream,
+             std::string *warning) {
+  // Create a default material anyway.
+  material_t material;
+  InitMaterial(&material);
+
+  // Issue 43. `d` wins against `Tr` since `Tr` is not in the MTL specification.
+  bool has_d = false;
+  bool has_tr = false;
+
+  std::stringstream ss;
+
+  std::string linebuf;
+  while (inStream->peek() != -1) {
+    safeGetline(*inStream, linebuf);
+
+    // Trim trailing whitespace.
+    if (linebuf.size() > 0) {
+      linebuf = linebuf.substr(0, linebuf.find_last_not_of(" \t") + 1);
+    }
+
+    // Trim newline '\r\n' or '\n'
+    if (linebuf.size() > 0) {
+      if (linebuf[linebuf.size() - 1] == '\n')
+        linebuf.erase(linebuf.size() - 1);
+    }
+    if (linebuf.size() > 0) {
+      if (linebuf[linebuf.size() - 1] == '\r')
+        linebuf.erase(linebuf.size() - 1);
+    }
+
+    // Skip if empty line.
+    if (linebuf.empty()) {
+      continue;
+    }
+
+    // Skip leading space.
+    const char *token = linebuf.c_str();
+    token += strspn(token, " \t");
+
+    assert(token);
+    if (token[0] == '\0') continue;  // empty line
+
+    if (token[0] == '#') continue;  // comment line
+
+    // new mtl
+    if ((0 == strncmp(token, "newmtl", 6)) && IS_SPACE((token[6]))) {
+      // flush previous material.
+      if (!material.name.empty()) {
+        material_map->insert(std::pair<std::string, int>(
+            material.name, static_cast<int>(materials->size())));
+        materials->push_back(material);
+      }
+
+      // initial temporary material
+      InitMaterial(&material);
+
+      has_d = false;
+      has_tr = false;
+
+      // set new mtl name
+      token += 7;
+      {
+        std::stringstream sstr;
+        sstr << token;
+        material.name = sstr.str();
+      }
+      continue;
+    }
+
+    // ambient
+    if (token[0] == 'K' && token[1] == 'a' && IS_SPACE((token[2]))) {
+      token += 2;
+      real_t r, g, b;
+      parseReal3(&r, &g, &b, &token);
+      material.ambient[0] = r;
+      material.ambient[1] = g;
+      material.ambient[2] = b;
+      continue;
+    }
+
+    // diffuse
+    if (token[0] == 'K' && token[1] == 'd' && IS_SPACE((token[2]))) {
+      token += 2;
+      real_t r, g, b;
+      parseReal3(&r, &g, &b, &token);
+      material.diffuse[0] = r;
+      material.diffuse[1] = g;
+      material.diffuse[2] = b;
+      continue;
+    }
+
+    // specular
+    if (token[0] == 'K' && token[1] == 's' && IS_SPACE((token[2]))) {
+      token += 2;
+      real_t r, g, b;
+      parseReal3(&r, &g, &b, &token);
+      material.specular[0] = r;
+      material.specular[1] = g;
+      material.specular[2] = b;
+      continue;
+    }
+
+    // transmittance
+    if ((token[0] == 'K' && token[1] == 't' && IS_SPACE((token[2]))) ||
+        (token[0] == 'T' && token[1] == 'f' && IS_SPACE((token[2])))) {
+      token += 2;
+      real_t r, g, b;
+      parseReal3(&r, &g, &b, &token);
+      material.transmittance[0] = r;
+      material.transmittance[1] = g;
+      material.transmittance[2] = b;
+      continue;
+    }
+
+    // ior(index of refraction)
+    if (token[0] == 'N' && token[1] == 'i' && IS_SPACE((token[2]))) {
+      token += 2;
+      material.ior = parseReal(&token);
+      continue;
+    }
+
+    // emission
+    if (token[0] == 'K' && token[1] == 'e' && IS_SPACE(token[2])) {
+      token += 2;
+      real_t r, g, b;
+      parseReal3(&r, &g, &b, &token);
+      material.emission[0] = r;
+      material.emission[1] = g;
+      material.emission[2] = b;
+      continue;
+    }
+
+    // shininess
+    if (token[0] == 'N' && token[1] == 's' && IS_SPACE(token[2])) {
+      token += 2;
+      material.shininess = parseReal(&token);
+      continue;
+    }
+
+    // illum model
+    if (0 == strncmp(token, "illum", 5) && IS_SPACE(token[5])) {
+      token += 6;
+      material.illum = parseInt(&token);
+      continue;
+    }
+
+    // dissolve
+    if ((token[0] == 'd' && IS_SPACE(token[1]))) {
+      token += 1;
+      material.dissolve = parseReal(&token);
+
+      if (has_tr) {
+        ss << "WARN: Both `d` and `Tr` parameters defined for \""
+           << material.name << "\". Use the value of `d` for dissolve."
+           << std::endl;
+      }
+      has_d = true;
+      continue;
+    }
+    if (token[0] == 'T' && token[1] == 'r' && IS_SPACE(token[2])) {
+      token += 2;
+      if (has_d) {
+        // `d` wins. Ignore `Tr` value.
+        ss << "WARN: Both `d` and `Tr` parameters defined for \""
+           << material.name << "\". Use the value of `d` for dissolve."
+           << std::endl;
+      } else {
+        // We invert value of Tr(assume Tr is in range [0, 1])
+        // NOTE: Interpretation of Tr is application(exporter) dependent. For
+        // some application(e.g. 3ds max obj exporter), Tr = d(Issue 43)
+        material.dissolve = 1.0f - parseReal(&token);
+      }
+      has_tr = true;
+      continue;
+    }
+
+    // PBR: roughness
+    if (token[0] == 'P' && token[1] == 'r' && IS_SPACE(token[2])) {
+      token += 2;
+      material.roughness = parseReal(&token);
+      continue;
+    }
+
+    // PBR: metallic
+    if (token[0] == 'P' && token[1] == 'm' && IS_SPACE(token[2])) {
+      token += 2;
+      material.metallic = parseReal(&token);
+      continue;
+    }
+
+    // PBR: sheen
+    if (token[0] == 'P' && token[1] == 's' && IS_SPACE(token[2])) {
+      token += 2;
+      material.sheen = parseReal(&token);
+      continue;
+    }
+
+    // PBR: clearcoat thickness
+    if (token[0] == 'P' && token[1] == 'c' && IS_SPACE(token[2])) {
+      token += 2;
+      material.clearcoat_thickness = parseReal(&token);
+      continue;
+    }
+
+    // PBR: clearcoat roughness
+    if ((0 == strncmp(token, "Pcr", 3)) && IS_SPACE(token[3])) {
+      token += 4;
+      material.clearcoat_roughness = parseReal(&token);
+      continue;
+    }
+
+    // PBR: anisotropy
+    if ((0 == strncmp(token, "aniso", 5)) && IS_SPACE(token[5])) {
+      token += 6;
+      material.anisotropy = parseReal(&token);
+      continue;
+    }
+
+    // PBR: anisotropy rotation
+    if ((0 == strncmp(token, "anisor", 6)) && IS_SPACE(token[6])) {
+      token += 7;
+      material.anisotropy_rotation = parseReal(&token);
+      continue;
+    }
+
+    // ambient texture
+    if ((0 == strncmp(token, "map_Ka", 6)) && IS_SPACE(token[6])) {
+      token += 7;
+      ParseTextureNameAndOption(&(material.ambient_texname),
+                                &(material.ambient_texopt), token,
+                                /* is_bump */ false);
+      continue;
+    }
+
+    // diffuse texture
+    if ((0 == strncmp(token, "map_Kd", 6)) && IS_SPACE(token[6])) {
+      token += 7;
+      ParseTextureNameAndOption(&(material.diffuse_texname),
+                                &(material.diffuse_texopt), token,
+                                /* is_bump */ false);
+      continue;
+    }
+
+    // specular texture
+    if ((0 == strncmp(token, "map_Ks", 6)) && IS_SPACE(token[6])) {
+      token += 7;
+      ParseTextureNameAndOption(&(material.specular_texname),
+                                &(material.specular_texopt), token,
+                                /* is_bump */ false);
+      continue;
+    }
+
+    // specular highlight texture
+    if ((0 == strncmp(token, "map_Ns", 6)) && IS_SPACE(token[6])) {
+      token += 7;
+      ParseTextureNameAndOption(&(material.specular_highlight_texname),
+                                &(material.specular_highlight_texopt), token,
+                                /* is_bump */ false);
+      continue;
+    }
+
+    // bump texture
+    if ((0 == strncmp(token, "map_bump", 8)) && IS_SPACE(token[8])) {
+      token += 9;
+      ParseTextureNameAndOption(&(material.bump_texname),
+                                &(material.bump_texopt), token,
+                                /* is_bump */ true);
+      continue;
+    }
+
+    // bump texture
+    if ((0 == strncmp(token, "map_Bump", 8)) && IS_SPACE(token[8])) {
+      token += 9;
+      ParseTextureNameAndOption(&(material.bump_texname),
+                                &(material.bump_texopt), token,
+                                /* is_bump */ true);
+      continue;
+    }
+
+    // bump texture
+    if ((0 == strncmp(token, "bump", 4)) && IS_SPACE(token[4])) {
+      token += 5;
+      ParseTextureNameAndOption(&(material.bump_texname),
+                                &(material.bump_texopt), token,
+                                /* is_bump */ true);
+      continue;
+    }
+
+    // alpha texture
+    if ((0 == strncmp(token, "map_d", 5)) && IS_SPACE(token[5])) {
+      token += 6;
+      material.alpha_texname = token;
+      ParseTextureNameAndOption(&(material.alpha_texname),
+                                &(material.alpha_texopt), token,
+                                /* is_bump */ false);
+      continue;
+    }
+
+    // displacement texture
+    if ((0 == strncmp(token, "disp", 4)) && IS_SPACE(token[4])) {
+      token += 5;
+      ParseTextureNameAndOption(&(material.displacement_texname),
+                                &(material.displacement_texopt), token,
+                                /* is_bump */ false);
+      continue;
+    }
+
+    // reflection map
+    if ((0 == strncmp(token, "refl", 4)) && IS_SPACE(token[4])) {
+      token += 5;
+      ParseTextureNameAndOption(&(material.reflection_texname),
+                                &(material.reflection_texopt), token,
+                                /* is_bump */ false);
+      continue;
+    }
+
+    // PBR: roughness texture
+    if ((0 == strncmp(token, "map_Pr", 6)) && IS_SPACE(token[6])) {
+      token += 7;
+      ParseTextureNameAndOption(&(material.roughness_texname),
+                                &(material.roughness_texopt), token,
+                                /* is_bump */ false);
+      continue;
+    }
+
+    // PBR: metallic texture
+    if ((0 == strncmp(token, "map_Pm", 6)) && IS_SPACE(token[6])) {
+      token += 7;
+      ParseTextureNameAndOption(&(material.metallic_texname),
+                                &(material.metallic_texopt), token,
+                                /* is_bump */ false);
+      continue;
+    }
+
+    // PBR: sheen texture
+    if ((0 == strncmp(token, "map_Ps", 6)) && IS_SPACE(token[6])) {
+      token += 7;
+      ParseTextureNameAndOption(&(material.sheen_texname),
+                                &(material.sheen_texopt), token,
+                                /* is_bump */ false);
+      continue;
+    }
+
+    // PBR: emissive texture
+    if ((0 == strncmp(token, "map_Ke", 6)) && IS_SPACE(token[6])) {
+      token += 7;
+      ParseTextureNameAndOption(&(material.emissive_texname),
+                                &(material.emissive_texopt), token,
+                                /* is_bump */ false);
+      continue;
+    }
+
+    // PBR: normal map texture
+    if ((0 == strncmp(token, "norm", 4)) && IS_SPACE(token[4])) {
+      token += 5;
+      ParseTextureNameAndOption(
+          &(material.normal_texname), &(material.normal_texopt), token,
+          /* is_bump */ false);  // @fixme { is_bump will be true? }
+      continue;
+    }
+
+    // unknown parameter
+    const char *_space = strchr(token, ' ');
+    if (!_space) {
+      _space = strchr(token, '\t');
+    }
+    if (_space) {
+      std::ptrdiff_t len = _space - token;
+      std::string key(token, static_cast<size_t>(len));
+      std::string value = _space + 1;
+      material.unknown_parameter.insert(
+          std::pair<std::string, std::string>(key, value));
+    }
+  }
+  // flush last material.
+  material_map->insert(std::pair<std::string, int>(
+      material.name, static_cast<int>(materials->size())));
+  materials->push_back(material);
+
+  if (warning) {
+    (*warning) = ss.str();
+  }
+}
+
+bool MaterialFileReader::operator()(const std::string &matId,
+                                    std::vector<material_t> *materials,
+                                    std::map<std::string, int> *matMap,
+                                    std::string *err) {
+  std::string filepath;
+
+  if (!m_mtlBaseDir.empty()) {
+    filepath = std::string(m_mtlBaseDir) + matId;
+  } else {
+    filepath = matId;
+  }
+
+  std::ifstream matIStream(filepath.c_str());
+  if (!matIStream) {
+    std::stringstream ss;
+    ss << "WARN: Material file [ " << filepath << " ] not found." << std::endl;
+    if (err) {
+      (*err) += ss.str();
+    }
+    return false;
+  }
+
+  std::string warning;
+  LoadMtl(matMap, materials, &matIStream, &warning);
+
+  if (!warning.empty()) {
+    if (err) {
+      (*err) += warning;
+    }
+  }
+
+  return true;
+}
+
+bool MaterialStreamReader::operator()(const std::string &matId,
+                                      std::vector<material_t> *materials,
+                                      std::map<std::string, int> *matMap,
+                                      std::string *err) {
+  (void)matId;
+  if (!m_inStream) {
+    std::stringstream ss;
+    ss << "WARN: Material stream in error state. " << std::endl;
+    if (err) {
+      (*err) += ss.str();
+    }
+    return false;
+  }
+
+  std::string warning;
+  LoadMtl(matMap, materials, &m_inStream, &warning);
+
+  if (!warning.empty()) {
+    if (err) {
+      (*err) += warning;
+    }
+  }
+
+  return true;
+}
+
+bool LoadObj(attrib_t *attrib, std::vector<shape_t> *shapes,
+             std::vector<material_t> *materials, std::string *err,
+             const char *filename, const char *mtl_basedir, bool trianglulate) {
+  attrib->vertices.clear();
+  attrib->normals.clear();
+  attrib->texcoords.clear();
+  shapes->clear();
+
+  std::stringstream errss;
+
+  std::ifstream ifs(filename);
+  if (!ifs) {
+    errss << "Cannot open file [" << filename << "]" << std::endl;
+    if (err) {
+      (*err) = errss.str();
+    }
+    return false;
+  }
+
+  std::string baseDir;
+  if (mtl_basedir) {
+    baseDir = mtl_basedir;
+  }
+  MaterialFileReader matFileReader(baseDir);
+
+  return LoadObj(attrib, shapes, materials, err, &ifs, &matFileReader,
+                 trianglulate);
+}
+
+bool LoadObj(attrib_t *attrib, std::vector<shape_t> *shapes,
+             std::vector<material_t> *materials, std::string *err,
+             std::istream *inStream, MaterialReader *readMatFn /*= NULL*/,
+             bool triangulate) {
+  std::stringstream errss;
+
+  std::vector<real_t> v;
+  std::vector<real_t> vn;
+  std::vector<real_t> vt;
+  std::vector<tag_t> tags;
+  std::vector<std::vector<vertex_index> > faceGroup;
+  std::string name;
+
+  // material
+  std::map<std::string, int> material_map;
+  int material = -1;
+
+  shape_t shape;
+
+  std::string linebuf;
+  while (inStream->peek() != -1) {
+    safeGetline(*inStream, linebuf);
+
+    // Trim newline '\r\n' or '\n'
+    if (linebuf.size() > 0) {
+      if (linebuf[linebuf.size() - 1] == '\n')
+        linebuf.erase(linebuf.size() - 1);
+    }
+    if (linebuf.size() > 0) {
+      if (linebuf[linebuf.size() - 1] == '\r')
+        linebuf.erase(linebuf.size() - 1);
+    }
+
+    // Skip if empty line.
+    if (linebuf.empty()) {
+      continue;
+    }
+
+    // Skip leading space.
+    const char *token = linebuf.c_str();
+    token += strspn(token, " \t");
+
+    assert(token);
+    if (token[0] == '\0') continue;  // empty line
+
+    if (token[0] == '#') continue;  // comment line
+
+    // vertex
+    if (token[0] == 'v' && IS_SPACE((token[1]))) {
+      token += 2;
+      real_t x, y, z;
+      parseReal3(&x, &y, &z, &token);
+      v.push_back(x);
+      v.push_back(y);
+      v.push_back(z);
+      continue;
+    }
+
+    // normal
+    if (token[0] == 'v' && token[1] == 'n' && IS_SPACE((token[2]))) {
+      token += 3;
+      real_t x, y, z;
+      parseReal3(&x, &y, &z, &token);
+      vn.push_back(x);
+      vn.push_back(y);
+      vn.push_back(z);
+      continue;
+    }
+
+    // texcoord
+    if (token[0] == 'v' && token[1] == 't' && IS_SPACE((token[2]))) {
+      token += 3;
+      real_t x, y;
+      parseReal2(&x, &y, &token);
+      vt.push_back(x);
+      vt.push_back(y);
+      continue;
+    }
+
+    // face
+    if (token[0] == 'f' && IS_SPACE((token[1]))) {
+      token += 2;
+      token += strspn(token, " \t");
+
+      std::vector<vertex_index> face;
+      face.reserve(3);
+
+      while (!IS_NEW_LINE(token[0])) {
+        vertex_index vi = parseTriple(&token, static_cast<int>(v.size() / 3),
+                                      static_cast<int>(vn.size() / 3),
+                                      static_cast<int>(vt.size() / 2));
+        face.push_back(vi);
+        size_t n = strspn(token, " \t\r");
+        token += n;
+      }
+
+      // replace with emplace_back + std::move on C++11
+      faceGroup.push_back(std::vector<vertex_index>());
+      faceGroup[faceGroup.size() - 1].swap(face);
+
+      continue;
+    }
+
+    // use mtl
+    if ((0 == strncmp(token, "usemtl", 6)) && IS_SPACE((token[6]))) {
+      token += 7;
+      std::stringstream ss;
+      ss << token;
+      std::string namebuf = ss.str();
+
+      int newMaterialId = -1;
+      if (material_map.find(namebuf) != material_map.end()) {
+        newMaterialId = material_map[namebuf];
+      } else {
+        // { error!! material not found }
+      }
+
+      if (newMaterialId != material) {
+        // Create per-face material. Thus we don't add `shape` to `shapes` at
+        // this time.
+        // just clear `faceGroup` after `exportFaceGroupToShape()` call.
+        exportFaceGroupToShape(&shape, faceGroup, tags, material, name,
+                               triangulate);
+        faceGroup.clear();
+        material = newMaterialId;
+      }
+
+      continue;
+    }
+
+    // load mtl
+    if ((0 == strncmp(token, "mtllib", 6)) && IS_SPACE((token[6]))) {
+      if (readMatFn) {
+        token += 7;
+
+        std::vector<std::string> filenames;
+        SplitString(std::string(token), ' ', filenames);
+
+        if (filenames.empty()) {
+          if (err) {
+            (*err) +=
+                "WARN: Looks like empty filename for mtllib. Use default "
+                "material. \n";
+          }
+        } else {
+          bool found = false;
+          for (size_t s = 0; s < filenames.size(); s++) {
+            std::string err_mtl;
+            bool ok = (*readMatFn)(filenames[s].c_str(), materials,
+                                   &material_map, &err_mtl);
+            if (err && (!err_mtl.empty())) {
+              (*err) += err_mtl;  // This should be warn message.
+            }
+
+            if (ok) {
+              found = true;
+              break;
+            }
+          }
+
+          if (!found) {
+            if (err) {
+              (*err) +=
+                  "WARN: Failed to load material file(s). Use default "
+                  "material.\n";
+            }
+          }
+        }
+      }
+
+      continue;
+    }
+
+    // group name
+    if (token[0] == 'g' && IS_SPACE((token[1]))) {
+      // flush previous face group.
+      bool ret = exportFaceGroupToShape(&shape, faceGroup, tags, material, name,
+                                        triangulate);
+      (void)ret; // return value not used.
+
+      if (shape.mesh.indices.size() > 0) {
+        shapes->push_back(shape);
+      }
+
+      shape = shape_t();
+
+      // material = -1;
+      faceGroup.clear();
+
+      std::vector<std::string> names;
+      names.reserve(2);
+
+      while (!IS_NEW_LINE(token[0])) {
+        std::string str = parseString(&token);
+        names.push_back(str);
+        token += strspn(token, " \t\r");  // skip tag
+      }
+
+      assert(names.size() > 0);
+
+      // names[0] must be 'g', so skip the 0th element.
+      if (names.size() > 1) {
+        name = names[1];
+      } else {
+        name = "";
+      }
+
+      continue;
+    }
+
+    // object name
+    if (token[0] == 'o' && IS_SPACE((token[1]))) {
+      // flush previous face group.
+      bool ret = exportFaceGroupToShape(&shape, faceGroup, tags, material, name,
+                                        triangulate);
+      if (ret) {
+        shapes->push_back(shape);
+      }
+
+      // material = -1;
+      faceGroup.clear();
+      shape = shape_t();
+
+      // @todo { multiple object name? }
+      token += 2;
+      std::stringstream ss;
+      ss << token;
+      name = ss.str();
+
+      continue;
+    }
+
+    if (token[0] == 't' && IS_SPACE(token[1])) {
+      tag_t tag;
+
+      token += 2;
+      std::stringstream ss;
+      ss << token;
+      tag.name = ss.str();
+
+      token += tag.name.size() + 1;
+
+      tag_sizes ts = parseTagTriple(&token);
+
+      tag.intValues.resize(static_cast<size_t>(ts.num_ints));
+
+      for (size_t i = 0; i < static_cast<size_t>(ts.num_ints); ++i) {
+        tag.intValues[i] = atoi(token);
+        token += strcspn(token, "/ \t\r") + 1;
+      }
+
+      tag.floatValues.resize(static_cast<size_t>(ts.num_reals));
+      for (size_t i = 0; i < static_cast<size_t>(ts.num_reals); ++i) {
+        tag.floatValues[i] = parseReal(&token);
+        token += strcspn(token, "/ \t\r") + 1;
+      }
+
+      tag.stringValues.resize(static_cast<size_t>(ts.num_strings));
+      for (size_t i = 0; i < static_cast<size_t>(ts.num_strings); ++i) {
+        std::stringstream sstr;
+        sstr << token;
+        tag.stringValues[i] = sstr.str();
+        token += tag.stringValues[i].size() + 1;
+      }
+
+      tags.push_back(tag);
+    }
+
+    // Ignore unknown command.
+  }
+
+  bool ret = exportFaceGroupToShape(&shape, faceGroup, tags, material, name,
+                                    triangulate);
+  // exportFaceGroupToShape return false when `usemtl` is called in the last
+  // line.
+  // we also add `shape` to `shapes` when `shape.mesh` has already some
+  // faces(indices)
+  if (ret || shape.mesh.indices.size()) {
+    shapes->push_back(shape);
+  }
+  faceGroup.clear();  // for safety
+
+  if (err) {
+    (*err) += errss.str();
+  }
+
+  attrib->vertices.swap(v);
+  attrib->normals.swap(vn);
+  attrib->texcoords.swap(vt);
+
+  return true;
+}
+
+bool LoadObjWithCallback(std::istream &inStream, const callback_t &callback,
+                         void *user_data /*= NULL*/,
+                         MaterialReader *readMatFn /*= NULL*/,
+                         std::string *err /*= NULL*/) {
+  std::stringstream errss;
+
+  // material
+  std::map<std::string, int> material_map;
+  int material_id = -1;  // -1 = invalid
+
+  std::vector<index_t> indices;
+  std::vector<material_t> materials;
+  std::vector<std::string> names;
+  names.reserve(2);
+  std::string name;
+  std::vector<const char *> names_out;
+
+  std::string linebuf;
+  while (inStream.peek() != -1) {
+    safeGetline(inStream, linebuf);
+
+    // Trim newline '\r\n' or '\n'
+    if (linebuf.size() > 0) {
+      if (linebuf[linebuf.size() - 1] == '\n')
+        linebuf.erase(linebuf.size() - 1);
+    }
+    if (linebuf.size() > 0) {
+      if (linebuf[linebuf.size() - 1] == '\r')
+        linebuf.erase(linebuf.size() - 1);
+    }
+
+    // Skip if empty line.
+    if (linebuf.empty()) {
+      continue;
+    }
+
+    // Skip leading space.
+    const char *token = linebuf.c_str();
+    token += strspn(token, " \t");
+
+    assert(token);
+    if (token[0] == '\0') continue;  // empty line
+
+    if (token[0] == '#') continue;  // comment line
+
+    // vertex
+    if (token[0] == 'v' && IS_SPACE((token[1]))) {
+      token += 2;
+      real_t x, y, z, w;  // w is optional. default = 1.0
+      parseV(&x, &y, &z, &w, &token);
+      if (callback.vertex_cb) {
+        callback.vertex_cb(user_data, x, y, z, w);
+      }
+      continue;
+    }
+
+    // normal
+    if (token[0] == 'v' && token[1] == 'n' && IS_SPACE((token[2]))) {
+      token += 3;
+      real_t x, y, z;
+      parseReal3(&x, &y, &z, &token);
+      if (callback.normal_cb) {
+        callback.normal_cb(user_data, x, y, z);
+      }
+      continue;
+    }
+
+    // texcoord
+    if (token[0] == 'v' && token[1] == 't' && IS_SPACE((token[2]))) {
+      token += 3;
+      real_t x, y, z;  // y and z are optional. default = 0.0
+      parseReal3(&x, &y, &z, &token);
+      if (callback.texcoord_cb) {
+        callback.texcoord_cb(user_data, x, y, z);
+      }
+      continue;
+    }
+
+    // face
+    if (token[0] == 'f' && IS_SPACE((token[1]))) {
+      token += 2;
+      token += strspn(token, " \t");
+
+      indices.clear();
+      while (!IS_NEW_LINE(token[0])) {
+        vertex_index vi = parseRawTriple(&token);
+
+        index_t idx;
+        idx.vertex_index = vi.v_idx;
+        idx.normal_index = vi.vn_idx;
+        idx.texcoord_index = vi.vt_idx;
+
+        indices.push_back(idx);
+        size_t n = strspn(token, " \t\r");
+        token += n;
+      }
+
+      if (callback.index_cb && indices.size() > 0) {
+        callback.index_cb(user_data, &indices.at(0),
+                          static_cast<int>(indices.size()));
+      }
+
+      continue;
+    }
+
+    // use mtl
+    if ((0 == strncmp(token, "usemtl", 6)) && IS_SPACE((token[6]))) {
+      token += 7;
+      std::stringstream ss;
+      ss << token;
+      std::string namebuf = ss.str();
+
+      int newMaterialId = -1;
+      if (material_map.find(namebuf) != material_map.end()) {
+        newMaterialId = material_map[namebuf];
+      } else {
+        // { error!! material not found }
+      }
+
+      if (newMaterialId != material_id) {
+        material_id = newMaterialId;
+      }
+
+      if (callback.usemtl_cb) {
+        callback.usemtl_cb(user_data, namebuf.c_str(), material_id);
+      }
+
+      continue;
+    }
+
+    // load mtl
+    if ((0 == strncmp(token, "mtllib", 6)) && IS_SPACE((token[6]))) {
+      if (readMatFn) {
+        token += 7;
+
+        std::vector<std::string> filenames;
+        SplitString(std::string(token), ' ', filenames);
+
+        if (filenames.empty()) {
+          if (err) {
+            (*err) +=
+                "WARN: Looks like empty filename for mtllib. Use default "
+                "material. \n";
+          }
+        } else {
+          bool found = false;
+          for (size_t s = 0; s < filenames.size(); s++) {
+            std::string err_mtl;
+            bool ok = (*readMatFn)(filenames[s].c_str(), &materials,
+                                   &material_map, &err_mtl);
+            if (err && (!err_mtl.empty())) {
+              (*err) += err_mtl;  // This should be warn message.
+            }
+
+            if (ok) {
+              found = true;
+              break;
+            }
+          }
+
+          if (!found) {
+            if (err) {
+              (*err) +=
+                  "WARN: Failed to load material file(s). Use default "
+                  "material.\n";
+            }
+          } else {
+            if (callback.mtllib_cb) {
+              callback.mtllib_cb(user_data, &materials.at(0),
+                                 static_cast<int>(materials.size()));
+            }
+          }
+        }
+      }
+
+      continue;
+    }
+
+    // group name
+    if (token[0] == 'g' && IS_SPACE((token[1]))) {
+      names.clear();
+
+      while (!IS_NEW_LINE(token[0])) {
+        std::string str = parseString(&token);
+        names.push_back(str);
+        token += strspn(token, " \t\r");  // skip tag
+      }
+
+      assert(names.size() > 0);
+
+      // names[0] must be 'g', so skip the 0th element.
+      if (names.size() > 1) {
+        name = names[1];
+      } else {
+        name.clear();
+      }
+
+      if (callback.group_cb) {
+        if (names.size() > 1) {
+          // create const char* array.
+          names_out.resize(names.size() - 1);
+          for (size_t j = 0; j < names_out.size(); j++) {
+            names_out[j] = names[j + 1].c_str();
+          }
+          callback.group_cb(user_data, &names_out.at(0),
+                            static_cast<int>(names_out.size()));
+
+        } else {
+          callback.group_cb(user_data, NULL, 0);
+        }
+      }
+
+      continue;
+    }
+
+    // object name
+    if (token[0] == 'o' && IS_SPACE((token[1]))) {
+      // @todo { multiple object name? }
+      token += 2;
+
+      std::stringstream ss;
+      ss << token;
+      std::string object_name = ss.str();
+
+      if (callback.object_cb) {
+        callback.object_cb(user_data, object_name.c_str());
+      }
+
+      continue;
+    }
+
+#if 0  // @todo
+    if (token[0] == 't' && IS_SPACE(token[1])) {
+      tag_t tag;
+
+      token += 2;
+      std::stringstream ss;
+      ss << token;
+      tag.name = ss.str();
+
+      token += tag.name.size() + 1;
+
+      tag_sizes ts = parseTagTriple(&token);
+
+      tag.intValues.resize(static_cast<size_t>(ts.num_ints));
+
+      for (size_t i = 0; i < static_cast<size_t>(ts.num_ints); ++i) {
+        tag.intValues[i] = atoi(token);
+        token += strcspn(token, "/ \t\r") + 1;
+      }
+
+      tag.floatValues.resize(static_cast<size_t>(ts.num_reals));
+      for (size_t i = 0; i < static_cast<size_t>(ts.num_reals); ++i) {
+        tag.floatValues[i] = parseReal(&token);
+        token += strcspn(token, "/ \t\r") + 1;
+      }
+
+      tag.stringValues.resize(static_cast<size_t>(ts.num_strings));
+      for (size_t i = 0; i < static_cast<size_t>(ts.num_strings); ++i) {
+        std::stringstream ss;
+        ss << token;
+        tag.stringValues[i] = ss.str();
+        token += tag.stringValues[i].size() + 1;
+      }
+
+      tags.push_back(tag);
+    }
+#endif
+
+    // Ignore unknown command.
+  }
+
+  if (err) {
+    (*err) += errss.str();
+  }
+
+  return true;
+}
+}  // namespace tinyobj
+
+#endif

diff --git a/turret.h b/turret.h
new file mode 100644 (file)
index 0000000..fe22eec
--- /dev/null
+++ b/turret.h
@@ -0,0 +1,31 @@
+#pragma once
+#include "game.h"
+#include "projectiles.h"
+
+class turret : public loaded_object {
+public:
+       size_t target_idx = 0;
+       gameobject *current_target; // Might be nice if this was a list
+       projectile *current_projectile;
+       int countdown = 10000;
+       
+       turret() : loaded_object("turret.obj", "turret.png", glm::vec3(1, 1, 1)) {}
+
+       void move() {
+/* At some rate (every so many calls to this function)
+ * Pick a target  (How do we know what targets we have?)
+ * Launch a projectile at it (How do we launch a projectile from here?)
+ */
+               if(countdown > 0){
+                       countdown--;
+                       return;
+               }       
+               if(!current_target->locations.size())
+                       return;
+               glm::vec3 target_location = current_target->locations[target_idx];
+               current_projectile->add_projectile(locations[0], 0.001f * (target_location - locations[0]), 10000); 
+               target_idx++;
+               if(target_idx >= current_target->locations.size())
+                       target_idx = 0;
+       }
+};

diff --git a/vertex_shader.glsl b/vertex_shader.glsl
new file mode 100644 (file)
index 0000000..f1cc9a1
--- /dev/null
+++ b/vertex_shader.glsl
@@ -0,0 +1,15 @@
+#version 460\r
+\r
+in vec3 in_vertex;\r
+in vec3 in_color;\r
+uniform mat4 mvp;\r
+uniform mat4 animation;\r
+out vec4 fcolor;\r
+out vec4 gl_Position;\r
+\r
+void main(void) {\r
+       vec4 tmp = animation * vec4(in_vertex, 1.0);\r
+       tmp.x += 25 - 5 * gl_InstanceID;\r
+       gl_Position = mvp * tmp;\r
+       fcolor = vec4(in_color, 1.0);\r
+}\r