commit f047c7e95a5f9ad9902a90ac6b1e9e1a03ad172d
parent c9ba0adbd7064240488f173bde6b565b1b59367b
Author: Samdal <samdal@protonmail.com>
Date: Sat, 22 Mar 2025 01:10:13 +0100
retained mode rendering, xkb, textures...
Diffstat:
33 files changed, 12513 insertions(+), 717 deletions(-)
diff --git a/.editorconfig b/.editorconfig
@@ -0,0 +1,16 @@
+root = true
+
+[*]
+end_of_line = lf
+insert_final_newline = true
+
+[*.{c,h,cpp,hpp}]
+indent_style = space
+indent_size = 4
+
+[*.mdesk]
+indent_style = space
+indent_size = 4
+
+[Makefile]
+indent_style = tab
diff --git a/build.sh b/build.sh
@@ -2,31 +2,58 @@
set -e
-CC="clang"
-output=(./bin/App)
sources=(./examples/test.c)
+
+for arg in "$@"
+do
+ case ${arg} in
+ "events") sources=(./examples/events.c) ;;
+ "test") sources=(./examples/test.c) ;;
+ "uniform") sources=(./examples/uniform.c) ;;
+ "simple_triangle") sources=(./examples/simple_triangle.c) ;;
+ "simple_texture") sources=(./examples/simple_texture.c) ;;
+ esac
+done
+
+if [ "$1" = "events" ]; then
+ sources=(./examples/events.c)
+fi
+
+output=(./bin/App)
+CC="clang"
compiler_flags=(
-g
-O0
- -pipe
+ #-march=native -ffast-math
-Wall
-Wno-missing-braces -Wno-nan-infinity-disabled -Wno-unused-function -Wno-unused-variable -Wno-unused-but-set-variable -Wno-initializer-overrides -Wfatal-errors
- -std=c11 -pedantic
+ -std=c11
-Isrc/
+
+ -D_GNU_SOURCE=1
+)
+common_flags=(
+ #-fsanitize=address,float-divide-by-zero,float-cast-overflow -fno-sanitize=null,alignment -fno-sanitize-recover=all -fno-omit-frame-pointer
)
-common_flags=(-pthread -fopenmp)
linker_flags=(
- -lm -lxcb -lEGL -lGL
+ -lm -lxcb -lxcb-xkb -lEGL -lGL
-fuse-ld=mold
)
mkdir -p $(dirname "${output}")
+echo "Compiling..."
${CC} ${compiler_flags[*]} ${common_flags[*]} ${sources[*]} ${linker_flags[*]} -o ${output}
+echo "Done"
-if [ "$1" = "run" ]; then
- ./${output}
-fi
+
+for arg in "$@"
+do
+ if [ ${arg} = "run" ]; then
+ echo "Running"
+ ./${output}
+ fi
+done
diff --git a/examples/events.c b/examples/events.c
@@ -0,0 +1,45 @@
+//////////////////////////////////////////////////////////////////
+// events.c
+#define RV_RENDER_OPENGL 0
+#include "revolver_inc.h"
+#include "revolver_inc.c"
+
+int main(void) {
+ rv_window_desc_t desc = {.name = S("App"), .attach_render = true};
+ rv_window_handle_t* window = rv_create_window(desc);
+
+ while(1) {
+ rv_temp_arena scratch = rv_scratch_begin(0, 0);
+ for(rv_event_t* e = rv_get_events(scratch.arena, 0); e; e = e->next) {
+ switch(e->type) {
+ case RV_EVENT_INVALID: {
+ rv_assert(false);
+ } break;
+ case RV_EVENT_BUTTON_PRESS: {
+ //rv_destroy_window(&window);
+ } break;
+ case RV_EVENT_KEY_PRESS: {
+ printf("pressed key %d '%.*s' (mods: %d) on window %p\n", e->key_press.keycode, rv_s8v(e->key_press.string), e->key_press.mods, e->key_press.window);
+ } break;
+ case RV_EVENT_KEY_RELEASE: {
+ printf("pressed key %d '%.*s' (mods: %d) on window %p\n", e->key_release.keycode, rv_s8v(e->key_release.string), e->key_release.mods, e->key_release.window);
+ } break;
+ case RV_EVENT_WINDOW_CLOSE: {
+ if (e->window_close == window) {
+ rv_abort_msg(0, "close request");
+ }
+ } break;
+ case RV_EVENT_WINDOW_RESIZE: {
+ rv_vec2 size = rv_window_size(window);
+ printf("size: <w:%f, h:%f> on window %p\n", size.x, size.y, e->window_resize);
+ } break;
+ default: break;
+ }
+ }
+ rv_vec2 ptr = rv_mouse_pos(window);
+ //printf("ptr: <w:%f, h:%f>\n", ptr.x, ptr.y);
+ rv_scratch_end(scratch);
+ }
+
+ return 0;
+}
diff --git a/examples/simple_texture.c b/examples/simple_texture.c
@@ -0,0 +1,139 @@
+//////////////////////////////////////////////////////////////////
+// simple_texture.c
+
+#include "revolver_inc.h"
+#include "revolver_inc.c"
+
+#define ROW_COL_CT 10
+
+//////////////////////////////////////////////////////////////////
+// data
+
+// Vertex data for quad
+f32 v_data[] = {
+ // Positions UVs
+ -0.5f, -0.5f, 0.0f, 0.0f, // Top Left
+ 0.5f, -0.5f, 1.0f, 0.0f, // Top Right
+ -0.5f, 0.5f, 0.0f, 1.0f, // Bottom Left
+ 0.5f, 0.5f, 1.0f, 1.0f // Bottom Right
+};
+
+// Index data for quad
+uint32_t i_data[] = {
+ 0, 3, 2, // First Triangle
+ 0, 1, 3 // Second Triangle
+};
+
+rv_str8 v_src = S("#version 330 core\n" rv_strify(
+ layout(location = 0) in vec2 a_pos;
+ layout(location = 1) in vec2 a_uv;
+ precision mediump float;
+ out vec2 uv;
+ void main()
+ {
+ gl_Position = vec4(a_pos, 0.0, 1.0);
+ uv = a_uv;
+ }
+));
+
+rv_str8 f_src = S("#version 330 core\n" rv_strify(
+ precision mediump float;
+ uniform sampler2D u_tex;
+ in vec2 uv;
+ out vec4 frag_color;
+ void main()
+ {
+ frag_color = texture(u_tex, uv);
+ }
+));
+
+//////////////////////////////////////////////////////////////////
+
+
+int main(void) {
+ rv_window_desc_t desc = {.name = S("App"), .attach_render = true};
+ rv_window_handle_t* window = rv_create_window(desc);
+
+ rv_arena* arena = rv_arena_alloc();
+ rv_render_pass_list_t rpass_list = {0};
+
+
+ // Generate procedural texture data (checkered texture)
+ rv_color_t c0 = RV_COLOR_BLACK;
+ rv_color_t c1 = RV_COLOR_PURPLE;
+ rv_color_t pixels[ROW_COL_CT * ROW_COL_CT] = {0};
+ for (uint32_t r = 0; r < ROW_COL_CT; ++r) {
+ for (uint32_t c = 0; c < ROW_COL_CT; ++c) {
+ const bool32 re = (r % 2) == 0;
+ const bool32 ce = (c % 2) == 0;
+ uint32_t idx = r * ROW_COL_CT + c;
+ pixels[idx] = (re && ce) ? c0 : (re) ? c1 : (ce) ? c1 : c0;
+ }
+ }
+
+ rv_texture_t* tex = rv_push_compound(arena, rv_texture_t, {.data = pixels, .size = rv_v2s(ROW_COL_CT)});
+ rv_vbo_t* vbo = rv_push_compound(arena, rv_vbo_t, {.data = v_data, .size = sizeof(v_data)});
+ rv_vbo_t* ibo = rv_push_compound(arena, rv_ibo_t, {.data = i_data, .size = sizeof(i_data), .elem_size = sizeof(i_data[0])});
+ rv_shader_t* vertex = rv_push_compound(arena, rv_shader_t, {.source = v_src, .type = RV_SHADER_TYPE_VERTEX});
+ rv_shader_t* fragment = rv_push_compound(arena, rv_shader_t, {.source = f_src, .type = RV_SHADER_TYPE_FRAGMENT});
+ rv_program_t* prog = rv_push(arena, rv_program_t);
+
+ rv_render_push_shader_to_prog(arena, prog, vertex);
+ rv_render_push_shader_to_prog(arena, prog, fragment);
+ rv_uniform_t* u_tex = rv_render_push_uniform_to_prog_texture(arena, prog, S("u_tex"), tex);
+ rv_render_push_vattr_to_prog(arena, prog, RV_VATTR_TYPE_FLOAT2);
+ rv_render_push_vattr_to_prog(arena, prog, RV_VATTR_TYPE_FLOAT2);
+
+ rv_render_command_list_t create_shader_instructions = {0};
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_TEXTURE, RV_RENDER_OBJ_OP_CREATE, tex);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_VERTEX, RV_RENDER_OBJ_OP_CREATE, vbo);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_INDEX, RV_RENDER_OBJ_OP_CREATE, ibo);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_SHADER, RV_RENDER_OBJ_OP_CREATE, vertex);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_SHADER, RV_RENDER_OBJ_OP_CREATE, fragment);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_PROGRAM, RV_RENDER_OBJ_OP_CREATE, prog);
+
+ rv_render_command_list_t render_shader_instructions = {0};
+ rv_render_push_obj_command(arena, &render_shader_instructions, RV_RENDER_COMMAND_OBJ_VERTEX, RV_RENDER_OBJ_OP_BIND, vbo);
+ rv_render_push_obj_command(arena, &render_shader_instructions, RV_RENDER_COMMAND_OBJ_INDEX, RV_RENDER_OBJ_OP_BIND, ibo);
+ rv_render_push_obj_command(arena, &render_shader_instructions, RV_RENDER_COMMAND_OBJ_PROGRAM, RV_RENDER_OBJ_OP_BIND, prog);
+ rv_render_push_command_compound(arena, &render_shader_instructions, RV_RENDER_COMMAND_DRAW, { .draw = {.count = 6 } });
+
+ // copy over create commands
+ rv_render_pass_t* create_rpass = rv_render_push_render_pass(arena, &rpass_list);
+ rv_render_copy_commands(arena, &create_rpass->commands, &create_shader_instructions);
+
+
+ while(1) {
+
+ if (window) { // render screen
+
+ rv_temp_arena scratch = rv_scratch_begin(0, 0);
+
+ // pick renderpass
+ rv_render_pass_t* rpass = rv_render_push_render_pass(scratch.arena, &rpass_list);
+
+ { // set viewport
+ rv_render_command_t* viewport = rv_render_push_command(scratch.arena, &rpass->commands, RV_RENDER_COMMAND_SET_VIEWPORT);
+ viewport->viewport = rv_v4(.xy = {0, 0}, .zw = rv_window_size(window));
+ }
+
+ { // clear
+ rv_render_command_t* clear = rv_render_push_command(scratch.arena, &rpass->commands, RV_RENDER_COMMAND_CLEAR);
+ rv_render_clear_desc_t* clear_desc = rv_render_push_clear_desc(scratch.arena, clear, RV_RENDER_CLEAR_FLAG_COLOR);
+ clear_desc->color_v = rv_v4(0.1, 0.1, 0.1, 0.1);
+ }
+
+ { // draw custom shader
+ // copy render commands
+ rv_render_copy_commands(scratch.arena, &rpass->commands, &render_shader_instructions);
+ }
+
+ rv_window_render_commit(window, &rpass_list);
+ rpass_list = (rv_render_pass_list_t){0};
+
+ rv_scratch_end(scratch);
+ }
+ }
+
+ return 0;
+}
diff --git a/examples/simple_triangle.c b/examples/simple_triangle.c
@@ -0,0 +1,105 @@
+//////////////////////////////////////////////////////////////////
+// simple_triangle.c
+
+#include "revolver_inc.h"
+#include "revolver_inc.c"
+
+//////////////////////////////////////////////////////////////////
+// data
+
+f32 v_data[] = {
+ 0.0, 0.5, 1, 0, 0,
+ -0.5, -0.5, 0, 1, 0,
+ 0.5, -0.5, 0, 0, 1,
+};
+
+rv_str8 v_src = S("#version 330 core\n" rv_strify(
+ layout(location = 0) in vec2 a_pos;
+ layout(location = 1) in vec3 a_color;
+ precision mediump float;
+ out vec3 f_color;
+ void main()
+ {
+ gl_Position = vec4(a_pos, 0.0, 1.0);
+ f_color = a_color;
+ }
+));
+
+rv_str8 f_src = S("#version 330 core\n" rv_strify(
+ precision mediump float;
+ in vec3 f_color;
+ out vec4 frag_color;
+ void main()
+ {
+ frag_color = vec4(f_color, 1.0);
+ }
+));
+
+//////////////////////////////////////////////////////////////////
+
+int main(void) {
+ rv_window_desc_t desc = {.name = S("App"), .attach_render = true};
+ rv_window_handle_t* window = rv_create_window(desc);
+
+ rv_arena* arena = rv_arena_alloc();
+ rv_render_pass_list_t rpass_list = {0};
+
+ rv_vbo_t* vbo = rv_push_compound(arena, rv_vbo_t, {.data = v_data, .size = sizeof(v_data)});
+ rv_shader_t* vertex = rv_push_compound(arena, rv_shader_t, {.source = v_src, .type = RV_SHADER_TYPE_VERTEX});
+ rv_shader_t* fragment = rv_push_compound(arena, rv_shader_t, {.source = f_src, .type = RV_SHADER_TYPE_FRAGMENT});
+ rv_program_t* prog = rv_push(arena, rv_program_t);
+
+ rv_render_push_shader_to_prog(arena, prog, vertex);
+ rv_render_push_shader_to_prog(arena, prog, fragment);
+ rv_render_push_vattr_to_prog(arena, prog, RV_VATTR_TYPE_FLOAT2);
+ rv_render_push_vattr_to_prog(arena, prog, RV_VATTR_TYPE_FLOAT3);
+
+ rv_render_command_list_t create_shader_instructions = {0};
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_VERTEX, RV_RENDER_OBJ_OP_CREATE, vbo);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_SHADER, RV_RENDER_OBJ_OP_CREATE, vertex);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_SHADER, RV_RENDER_OBJ_OP_CREATE, fragment);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_PROGRAM, RV_RENDER_OBJ_OP_CREATE, prog);
+
+ rv_render_command_list_t render_shader_instructions = {0};
+ rv_render_push_obj_command(arena, &render_shader_instructions, RV_RENDER_COMMAND_OBJ_VERTEX, RV_RENDER_OBJ_OP_BIND, vbo);
+ rv_render_push_obj_command(arena, &render_shader_instructions, RV_RENDER_COMMAND_OBJ_PROGRAM, RV_RENDER_OBJ_OP_BIND, prog);
+ rv_render_push_command_compound(arena, &render_shader_instructions, RV_RENDER_COMMAND_DRAW, { .draw = {.count = 3 } });
+
+ // copy over create commands
+ rv_render_pass_t* create_rpass = rv_render_push_render_pass(arena, &rpass_list);
+ rv_render_copy_commands(arena, &create_rpass->commands, &create_shader_instructions);
+
+ while(1) {
+
+ if (window) { // render screen
+
+ rv_temp_arena scratch = rv_scratch_begin(0, 0);
+
+ // pick renderpass
+ rv_render_pass_t* rpass = rv_render_push_render_pass(scratch.arena, &rpass_list);
+
+ { // set viewport
+ rv_render_command_t* viewport = rv_render_push_command(scratch.arena, &rpass->commands, RV_RENDER_COMMAND_SET_VIEWPORT);
+ viewport->viewport = rv_v4(.xy = {0, 0}, .zw = rv_window_size(window));
+ }
+
+ { // clear
+ rv_render_command_t* clear = rv_render_push_command(scratch.arena, &rpass->commands, RV_RENDER_COMMAND_CLEAR);
+ rv_render_clear_desc_t* clear_desc = rv_render_push_clear_desc(scratch.arena, clear, RV_RENDER_CLEAR_FLAG_COLOR);
+ clear_desc->color_v = rv_v4(0.1, 0.1, 0.1, 0.1);
+ }
+
+ { // draw custom shader
+ // copy render commands
+ rv_render_copy_commands(scratch.arena, &rpass->commands, &render_shader_instructions);
+ }
+
+ rv_window_render_commit(window, &rpass_list);
+ rpass_list = (rv_render_pass_list_t){0};
+
+ rv_scratch_end(scratch);
+ }
+ }
+
+ return 0;
+}
diff --git a/examples/test.c b/examples/test.c
@@ -1,62 +1,162 @@
#include "revolver_inc.h"
#include "revolver_inc.c"
+#define ROW_COL_CT 10
+
+// Vertex data for quad
+f32 v_data[] = {
+ // Positions UVs
+ -0.5f, -0.5f, 0.0f, 0.0f, // Top Left
+ 0.5f, -0.5f, 1.0f, 0.0f, // Top Right
+ -0.5f, 0.5f, 0.0f, 1.0f, // Bottom Left
+ 0.5f, 0.5f, 1.0f, 1.0f // Bottom Right
+};
+
+// Index data for quad
+uint32_t i_data[] = {
+ 0, 3, 2, // First Triangle
+ 0, 1, 3 // Second Triangle
+};
+
+rv_str8 v_src = S("#version 330 core\n" rv_strify(
+ layout(location = 0) in vec2 a_pos;
+ layout(location = 1) in vec2 a_uv;
+ precision mediump float;
+ out vec2 uv;
+ void main()
+ {
+ gl_Position = vec4(a_pos, 0.0, 1.0);
+ uv = a_uv;
+ }
+));
+
+rv_str8 f_src = S("#version 330 core\n" rv_strify(
+ precision mediump float;
+ uniform sampler2D u_tex;
+ in vec2 uv;
+ out vec4 frag_color;
+ void main()
+ {
+ frag_color = texture(u_tex, uv);
+ }
+));
+
+
int main(void) {
- rv_window_desc_t desc = {.name = "App", .attach_render = true};
- rv_window_handle_t* window = rv_create_window(desc);
-
- while(1) {
-
- { // process events
- rv_temp_arena scratch = rv_scratch_begin(0, 0);
- for(rv_window_event_t* e = rv_get_events(scratch.arena); e; e = e->next) {
- switch(e->type) {
- case RV_EVENT_INVALID: {
- rv_assert(false);
- } break;
- case RV_EVENT_BUTTON_PRESS: {
- rv_abort_msg(0, "button press");
- } break;
- case RV_EVENT_WINDOW_CLOSE: {
- if (e->window_close == window) {
- rv_abort_msg(0, "close request");
- }
- } break;
- case RV_EVENT_WINDOW_CHANGE: {
- puts("move");
- } break;
- default: break;
- }
- }
- rv_scratch_end(scratch);
- }
-
- rv_render_pass_list_t rpass_list = {0};
-
- { // render screen
-
- rv_temp_arena scratch = rv_scratch_begin(0, 0);
-
- // pick renderpass
- rv_render_pass_t* rpass = rv_render_push_render_pass(scratch.arena, &rpass_list);
-
- { // set viewport
- rv_render_command_t* viewport = rv_render_push_command(scratch.arena, rpass, RV_RENDER_COMMAND_SET_VIEWPORT);
- viewport->viewport = rv_v4(.xy = {0, 0}, .zw = rv_window_size(window));
- }
-
- { // clear
- rv_render_command_t* clear = rv_render_push_command(scratch.arena, rpass, RV_RENDER_COMMAND_CLEAR);
- rv_render_clear_desc_t* clear_desc = rv_render_push_clear_desc(scratch.arena, clear, RV_RENDER_CLEAR_FLAG_COLOR);
- clear_desc->color[0] = fabsf(sinf(rv_time()));
- }
-
- rv_window_render_commit(window, &rpass_list);
- rpass_list = (rv_render_pass_list_t){0};
-
- rv_scratch_end(scratch);
- }
- }
-
- return 0;
+ rv_window_desc_t desc = {.name = S("App"), .attach_render = true};
+ rv_window_handle_t* window = rv_create_window(desc);
+
+ rv_arena* arena = rv_arena_alloc();
+ rv_render_pass_list_t rpass_list = {0};
+
+
+ // Generate procedural texture data (checkered texture)
+ rv_color_t c0 = RV_COLOR_BLACK;
+ rv_color_t c1 = RV_COLOR_PURPLE;
+ rv_color_t pixels[ROW_COL_CT * ROW_COL_CT] = {0};
+ for (uint32_t r = 0; r < ROW_COL_CT; ++r) {
+ for (uint32_t c = 0; c < ROW_COL_CT; ++c) {
+ const bool32 re = (r % 2) == 0;
+ const bool32 ce = (c % 2) == 0;
+ uint32_t idx = r * ROW_COL_CT + c;
+ pixels[idx] = (re && ce) ? c0 : (re) ? c1 : (ce) ? c1 : c0;
+ }
+ }
+
+ rv_texture_t* tex = rv_push_compound(arena, rv_texture_t, {.data = pixels, .size = rv_v2s(ROW_COL_CT)});
+ rv_vbo_t* vbo = rv_push_compound(arena, rv_vbo_t, {.data = v_data, .size = sizeof(v_data)});
+ rv_vbo_t* ibo = rv_push_compound(arena, rv_ibo_t, {.data = i_data, .size = sizeof(i_data), .elem_size = sizeof(i_data[0])});
+ rv_shader_t* vertex = rv_push_compound(arena, rv_shader_t, {.source = v_src, .type = RV_SHADER_TYPE_VERTEX});
+ rv_shader_t* fragment = rv_push_compound(arena, rv_shader_t, {.source = f_src, .type = RV_SHADER_TYPE_FRAGMENT});
+
+ rv_program_t* prog = rv_push(arena, rv_program_t);
+ rv_render_push_shader_to_prog(arena, prog, vertex);
+ rv_render_push_shader_to_prog(arena, prog, fragment);
+ rv_uniform_t* u_tex = rv_render_push_uniform_to_prog_texture(arena, prog, S("u_tex"), tex);
+ rv_render_push_vattr_to_prog(arena, prog, RV_VATTR_TYPE_FLOAT2);
+ rv_render_push_vattr_to_prog(arena, prog, RV_VATTR_TYPE_FLOAT2);
+
+ rv_render_command_list_t create_shader_instructions = {0};
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_TEXTURE, RV_RENDER_OBJ_OP_CREATE, tex);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_VERTEX, RV_RENDER_OBJ_OP_CREATE, vbo);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_INDEX, RV_RENDER_OBJ_OP_CREATE, ibo);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_SHADER, RV_RENDER_OBJ_OP_CREATE, vertex);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_SHADER, RV_RENDER_OBJ_OP_CREATE, fragment);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_PROGRAM, RV_RENDER_OBJ_OP_CREATE, prog);
+
+ rv_render_command_list_t render_shader_instructions = {0};
+ rv_render_push_obj_command(arena, &render_shader_instructions, RV_RENDER_COMMAND_OBJ_VERTEX, RV_RENDER_OBJ_OP_BIND, vbo);
+ rv_render_push_obj_command(arena, &render_shader_instructions, RV_RENDER_COMMAND_OBJ_INDEX, RV_RENDER_OBJ_OP_BIND, ibo);
+ rv_render_push_obj_command(arena, &render_shader_instructions, RV_RENDER_COMMAND_OBJ_PROGRAM, RV_RENDER_OBJ_OP_BIND, prog);
+ rv_render_push_command_compound(arena, &render_shader_instructions, RV_RENDER_COMMAND_DRAW, { .draw = {.count = 6 } });
+
+ // copy over create commands
+ rv_render_pass_t* create_rpass = rv_render_push_render_pass(arena, &rpass_list);
+ rv_render_copy_commands(arena, &create_rpass->commands, &create_shader_instructions);
+
+
+ while(1) {
+
+ { // process events
+ rv_temp_arena scratch = rv_scratch_begin(0, 0);
+ for(rv_event_t* e = rv_get_events(scratch.arena, 0); e; e = e->next) {
+ switch(e->type) {
+ case RV_EVENT_INVALID: {
+ rv_assert(false);
+ } break;
+ case RV_EVENT_BUTTON_PRESS: {
+ //rv_destroy_window(&window);
+ } break;
+ case RV_EVENT_KEY_PRESS: {
+ printf("pressed key %d '%.*s' (mods: %d) on window %p\n", e->key_press.keycode, rv_s8v(e->key_press.string), e->key_press.mods, e->key_press.window);
+ } break;
+ case RV_EVENT_KEY_RELEASE: {
+ printf("pressed key %d '%.*s' (mods: %d) on window %p\n", e->key_release.keycode, rv_s8v(e->key_release.string), e->key_release.mods, e->key_release.window);
+ } break;
+ case RV_EVENT_WINDOW_CLOSE: {
+ if (e->window_close == window) {
+ rv_abort_msg(0, "close request");
+ }
+ } break;
+ case RV_EVENT_WINDOW_RESIZE: {
+ puts("move");
+ } break;
+ default: break;
+ }
+ }
+ rv_scratch_end(scratch);
+ }
+
+
+ if (window) { // render screen
+
+ rv_temp_arena scratch = rv_scratch_begin(0, 0);
+
+ // pick renderpass
+ rv_render_pass_t* rpass = rv_render_push_render_pass(scratch.arena, &rpass_list);
+
+ { // set viewport
+ rv_render_command_t* viewport = rv_render_push_command(scratch.arena, &rpass->commands, RV_RENDER_COMMAND_SET_VIEWPORT);
+ viewport->viewport = rv_v4(.xy = {0, 0}, .zw = rv_window_size(window));
+ }
+
+ { // clear
+ rv_render_command_t* clear = rv_render_push_command(scratch.arena, &rpass->commands, RV_RENDER_COMMAND_CLEAR);
+ rv_render_clear_desc_t* clear_desc = rv_render_push_clear_desc(scratch.arena, clear, RV_RENDER_CLEAR_FLAG_COLOR);
+ clear_desc->color_v = rv_v4(0.1, 0.1, 0.1, 0.1);
+ }
+
+ { // draw custom shader
+ // copy render commands
+ rv_render_copy_commands(scratch.arena, &rpass->commands, &render_shader_instructions);
+ }
+
+ rv_window_render_commit(window, &rpass_list);
+ rpass_list = (rv_render_pass_list_t){0};
+
+ rv_scratch_end(scratch);
+ }
+ }
+
+ return 0;
}
diff --git a/examples/uniform.c b/examples/uniform.c
@@ -0,0 +1,133 @@
+//////////////////////////////////////////////////////////////////
+// uniform.c
+
+#include "revolver_inc.h"
+#include "revolver_inc.c"
+
+//////////////////////////////////////////////////////////////////
+// data
+
+f32 v_data[] = {
+ 0.0f, 0.5f, 1, 0, 0,
+ -0.5f, -0.5f, 0, 1, 0,
+ 0.5f, -0.5f, 0, 0, 1,
+};
+
+rv_str8 v_src = S("#version 330 core\n" rv_strify(
+ layout(location = 0) in vec2 a_pos;
+ layout(location = 1) in vec3 a_color;
+ precision mediump float;
+ uniform mat4 u_model;
+ out vec3 f_color;
+ void main()
+ {
+ gl_Position = u_model * vec4(a_pos, 0.0, 1.0);
+ f_color = a_color;
+ }
+));
+
+rv_str8 f_src = S("#version 330 core\n" rv_strify(
+ precision mediump float;
+ in vec3 f_color;
+ out vec4 frag_color;
+ uniform vec3 u_color;
+ void main()
+ {
+ frag_color = vec4(u_color * f_color, 1.0);
+ }
+));
+
+//////////////////////////////////////////////////////////////////
+
+int main(void) {
+ rv_window_desc_t desc = {.name = S("App"), .attach_render = true};
+ rv_window_handle_t* window = rv_create_window(desc);
+
+ rv_arena* arena = rv_arena_alloc();
+ rv_render_pass_list_t rpass_list = {0};
+
+ rv_vbo_t* vbo = rv_push_compound(arena, rv_vbo_t, {.data = v_data, .size = sizeof(v_data)});
+ rv_shader_t* vertex = rv_push_compound(arena, rv_shader_t, {.source = v_src, .type = RV_SHADER_TYPE_VERTEX});
+ rv_shader_t* fragment = rv_push_compound(arena, rv_shader_t, {.source = f_src, .type = RV_SHADER_TYPE_FRAGMENT});
+ rv_program_t* prog = rv_push(arena, rv_program_t);
+
+ rv_render_push_shader_to_prog(arena, prog, vertex);
+ rv_render_push_shader_to_prog(arena, prog, fragment);
+ rv_uniform_t* u_color = rv_render_push_uniform_to_prog(arena, prog, S("u_color"), RV_UNIFORM_VEC3);
+ rv_uniform_t* u_model = rv_render_push_uniform_to_prog(arena, prog, S("u_model"), RV_UNIFORM_MAT4);
+ rv_render_push_vattr_to_prog(arena, prog, RV_VATTR_TYPE_FLOAT2);
+ rv_render_push_vattr_to_prog(arena, prog, RV_VATTR_TYPE_FLOAT3);
+
+ rv_render_command_list_t create_shader_instructions = {0};
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_VERTEX, RV_RENDER_OBJ_OP_CREATE, vbo);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_SHADER, RV_RENDER_OBJ_OP_CREATE, vertex);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_SHADER, RV_RENDER_OBJ_OP_CREATE, fragment);
+ rv_render_push_obj_command(arena, &create_shader_instructions, RV_RENDER_COMMAND_OBJ_PROGRAM, RV_RENDER_OBJ_OP_CREATE, prog);
+
+ rv_render_command_list_t render_shader_instructions = {0};
+ rv_render_push_obj_command(arena, &render_shader_instructions, RV_RENDER_COMMAND_OBJ_VERTEX, RV_RENDER_OBJ_OP_BIND, vbo);
+ rv_render_push_obj_command(arena, &render_shader_instructions, RV_RENDER_COMMAND_OBJ_PROGRAM, RV_RENDER_OBJ_OP_BIND, prog);
+ rv_render_push_command_compound(arena, &render_shader_instructions, RV_RENDER_COMMAND_DRAW, { .draw = {.count = 3 } });
+
+ // copy over create commands
+ rv_render_pass_t* create_rpass = rv_render_push_render_pass(arena, &rpass_list);
+ rv_render_copy_commands(arena, &create_rpass->commands, &create_shader_instructions);
+
+
+ while(1) {
+
+ if (window) { // render screen
+
+ rv_temp_arena scratch = rv_scratch_begin(0, 0);
+
+ // pick renderpass
+ rv_render_pass_t* rpass = rv_render_push_render_pass(scratch.arena, &rpass_list);
+
+ { // set viewport
+ rv_render_command_t* viewport = rv_render_push_command(scratch.arena, &rpass->commands, RV_RENDER_COMMAND_SET_VIEWPORT);
+ viewport->viewport = rv_v4(.xy = {0, 0}, .zw = rv_window_size(window));
+ }
+
+ { // clear
+ rv_render_command_t* clear = rv_render_push_command(scratch.arena, &rpass->commands, RV_RENDER_COMMAND_CLEAR);
+ rv_render_clear_desc_t* clear_desc = rv_render_push_clear_desc(scratch.arena, clear, RV_RENDER_CLEAR_FLAG_COLOR);
+ clear_desc->color_v = rv_v4(0.1, 0.1, 0.1, 0.1);
+ }
+
+ { // draw custom shader
+
+ // copy render commands
+ rv_render_command_t* rcmd = rv_render_copy_commands(scratch.arena, &rpass->commands, &render_shader_instructions);
+
+ // find the program command
+ rv_render_command_t* prog_cmd = rcmd;
+ for (; prog_cmd; prog_cmd = prog_cmd->next) {
+ if (rv_render_command_has_program(prog_cmd, prog)) break;
+ }
+
+ if (prog_cmd) {
+ // add uniform updates
+ const f64 t = rv_time();
+ const f32 r = sinf(t) * 0.5f + 0.5f;
+ const f32 g = cosf(t * 6.f) * 0.5f + 0.5f;
+ const f32 b = sinf(t * 3.f) * 0.5f + 0.5f;
+ rv_render_update_uniform_compound(scratch.arena, prog_cmd, u_color, {.v_vec3 = rv_v3(r, g, b)});
+
+ const f32 st = sin(t);
+ rv_mat4 rot = rv_mat4_rotatev(t, RV_ZAXIS);
+ rv_mat4 scl = rv_mat4_scalev(rv_v3(st, st, st));
+ rv_mat4 model = rv_mat4_mul_list(2, scl, rot);
+ rv_render_update_uniform_compound(scratch.arena, prog_cmd, u_model, {.v_mat4 = model});
+
+ }
+ }
+
+ rv_window_render_commit(window, &rpass_list);
+ rpass_list = (rv_render_pass_list_t){0};
+
+ rv_scratch_end(scratch);
+ }
+ }
+
+ return 0;
+}
diff --git a/examples/verify_compile.c b/examples/verify_compile.c
@@ -2,5 +2,5 @@
#include "revolver_inc.c"
int main(void) {
- return 0;
+ return 0;
}
diff --git a/src/platform/gfx/KHR/khrplatform.h b/src/external/KHR/khrplatform.h
diff --git a/src/platform/gfx/glad.c b/src/external/glad.c
diff --git a/src/platform/gfx/glad.h b/src/external/glad.h
diff --git a/src/external/stb_image.h b/src/external/stb_image.h
@@ -0,0 +1,7988 @@
+/* stb_image - v2.30 - 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:
+
+ 2.30 (2024-05-31) avoid erroneous gcc warning
+ 2.29 (2023-05-xx) optimizations
+ 2.28 (2023-01-29) many error fixes, security errors, just tons of stuff
+ 2.27 (2021-07-11) document stbi_info better, 16-bit PNM support, bug fixes
+ 2.26 (2020-07-13) many minor fixes
+ 2.25 (2020-02-02) fix warnings
+ 2.24 (2020-02-02) fix warnings; thread-local failure_reason and flip_vertically
+ 2.23 (2019-08-11) fix clang static analysis warning
+ 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 Simon Breuss (16-bit PNM)
+ John-Mark Allen
+ Carmelo J Fdez-Aguera
+
+ Bug & warning fixes
+ Marc LeBlanc David Woo Guillaume George Martins Mozeiko
+ Christpher Lloyd Jerry Jansson Joseph Thomson Blazej Dariusz Roszkowski
+ 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
+ Eugene Golushkov Laurent Gomila Cort Stratton github:snagar
+ Aruelien Pocheville Sergio Gonzalez Thibault Reuille github:Zelex
+ Cass Everitt Ryamond Barbiero github:grim210
+ Paul Du Bois Engin Manap Aldo Culquicondor github:sammyhw
+ Philipp Wiesemann Dale Weiler Oriol Ferrer Mesia github:phprus
+ Josh Tobin Neil Bickford Matthew Gregan github:poppolopoppo
+ Julian Raschke Gregory Mullen Christian Floisand github:darealshinji
+ Baldur Karlsson Kevin Schmidt JR Smith github:Michaelangel007
+ Brad Weinberger Matvey Cherevko github:mosra
+ Luca Sas Alexander Veselov Zack Middleton [reserved]
+ Ryan C. Gordon [reserved] [reserved]
+ DO NOT ADD YOUR NAME HERE
+
+ Jacko Dirks
+
+ To add your name to the credits, pick a random blank space in the middle and fill it.
+ 80% of merge conflicts on stb PRs are due to people adding their name at the end
+ of the credits.
+*/
+
+#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.
+//
+// To query the width, height and component count of an image without having to
+// decode the full file, you can use the stbi_info family of functions:
+//
+// int x,y,n,ok;
+// ok = stbi_info(filename, &x, &y, &n);
+// // returns ok=1 and sets x, y, n if image is a supported format,
+// // 0 otherwise.
+//
+// Note that stb_image pervasively uses ints in its public API for sizes,
+// including sizes of memory buffers. This is now part of the API and thus
+// hard to change without causing breakage. As a result, the various image
+// loaders all have certain limits on image size; these differ somewhat
+// by format but generally boil down to either just under 2GB or just under
+// 1GB. When the decoded image would be larger than this, stb_image decoding
+// will fail.
+//
+// Additionally, stb_image will reject image files that have any of their
+// dimensions set to a larger value than the configurable STBI_MAX_DIMENSIONS,
+// which defaults to 2**24 = 16777216 pixels. Due to the above memory limit,
+// the only way to have an image with such dimensions load correctly
+// is for it to have a rather extreme aspect ratio. Either way, the
+// assumption here is that such larger images are likely to be malformed
+// or malicious. If you do need to load an image with individual dimensions
+// larger than that, and it still fits in the overall size limit, you can
+// #define STBI_MAX_DIMENSIONS on your own to be something larger.
+//
+// ===========================================================================
+//
+// 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:
+//
+// We optionally support converting iPhone-formatted PNGs (which store
+// premultiplied BGRA) back to RGB, even though they're internally encoded
+// differently. To enable this conversion, call
+// stbi_convert_iphone_png_to_rgb(1).
+//
+// 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
+//
+// - If you define STBI_MAX_DIMENSIONS, stb_image will reject images greater
+// than that size (in either width or height) without further processing.
+// This is to let programs in the wild set an upper bound to prevent
+// denial-of-service attacks on untrusted data, as one could generate a
+// valid image of gigantic dimensions and force stb_image to allocate a
+// huge block of memory and spend disproportionate time decoding it. By
+// default this is set to (1 << 24), which is 16777216, but that's still
+// very big.
+
+#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
+// on most compilers (and ALL modern mainstream compilers) this is 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);
+
+// as above, but only applies to images loaded on the thread that calls the function
+// this function is only available if your compiler supports thread-local variables;
+// calling it will fail to link if your compiler doesn't
+STBIDEF void stbi_set_unpremultiply_on_load_thread(int flag_true_if_should_unpremultiply);
+STBIDEF void stbi_convert_iphone_png_to_rgb_thread(int flag_true_if_should_convert);
+STBIDEF void stbi_set_flip_vertically_on_load_thread(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
+
+#ifndef STBI_NO_THREAD_LOCALS
+ #if defined(__cplusplus) && __cplusplus >= 201103L
+ #define STBI_THREAD_LOCAL thread_local
+ #elif defined(__GNUC__) && __GNUC__ < 5
+ #define STBI_THREAD_LOCAL __thread
+ #elif defined(_MSC_VER)
+ #define STBI_THREAD_LOCAL __declspec(thread)
+ #elif defined (__STDC_VERSION__) && __STDC_VERSION__ >= 201112L && !defined(__STDC_NO_THREADS__)
+ #define STBI_THREAD_LOCAL _Thread_local
+ #endif
+
+ #ifndef STBI_THREAD_LOCAL
+ #if defined(__GNUC__)
+ #define STBI_THREAD_LOCAL __thread
+ #endif
+ #endif
+#endif
+
+#if defined(_MSC_VER) || defined(__SYMBIAN32__)
+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) >> (-(y) & 31)))
+#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>
+#ifdef _MSC_VER
+#define STBI_SIMD_ALIGN(type, name) __declspec(align(16)) type name
+#else
+#define STBI_SIMD_ALIGN(type, name) type name __attribute__((aligned(16)))
+#endif
+#endif
+
+#ifndef STBI_SIMD_ALIGN
+#define STBI_SIMD_ALIGN(type, name) type name
+#endif
+
+#ifndef STBI_MAX_DIMENSIONS
+#define STBI_MAX_DIMENSIONS (1 << 24)
+#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];
+ int callback_already_read;
+
+ 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->callback_already_read = 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->callback_already_read = 0;
+ s->img_buffer = 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)
+{
+ int ch;
+ fseek((FILE*) user, n, SEEK_CUR);
+ ch = fgetc((FILE*) user); /* have to read a byte to reset feof()'s flag */
+ if (ch != EOF) {
+ ungetc(ch, (FILE *) user); /* push byte back onto stream if valid. */
+ }
+}
+
+static int stbi__stdio_eof(void *user)
+{
+ return feof((FILE*) user) || ferror((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);
+static int stbi__pnm_is16(stbi__context *s);
+#endif
+
+static
+#ifdef STBI_THREAD_LOCAL
+STBI_THREAD_LOCAL
+#endif
+const char *stbi__g_failure_reason;
+
+STBIDEF const char *stbi_failure_reason(void)
+{
+ return stbi__g_failure_reason;
+}
+
+#ifndef STBI_NO_FAILURE_STRINGS
+static int stbi__err(const char *str)
+{
+ stbi__g_failure_reason = str;
+ return 0;
+}
+#endif
+
+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;
+}
+
+#if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR)
+// 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);
+}
+#endif
+
+// 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) || !defined(STBI_NO_PNM)
+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
+
+#if !defined(STBI_NO_JPEG) || !defined(STBI_NO_PNG) || !defined(STBI_NO_TGA) || !defined(STBI_NO_HDR)
+// 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);
+}
+#endif
+
+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) || !defined(STBI_NO_PNM)
+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
+
+// returns 1 if the sum of two signed ints is valid (between -2^31 and 2^31-1 inclusive), 0 on overflow.
+static int stbi__addints_valid(int a, int b)
+{
+ if ((a >= 0) != (b >= 0)) return 1; // a and b have different signs, so no overflow
+ if (a < 0 && b < 0) return a >= INT_MIN - b; // same as a + b >= INT_MIN; INT_MIN - b cannot overflow since b < 0.
+ return a <= INT_MAX - b;
+}
+
+// returns 1 if the product of two ints fits in a signed short, 0 on overflow.
+static int stbi__mul2shorts_valid(int a, int b)
+{
+ if (b == 0 || b == -1) return 1; // multiplication by 0 is always 0; check for -1 so SHRT_MIN/b doesn't overflow
+ if ((a >= 0) == (b >= 0)) return a <= SHRT_MAX/b; // product is positive, so similar to mul2sizes_valid
+ if (b < 0) return a <= SHRT_MIN / b; // same as a * b >= SHRT_MIN
+ return a >= SHRT_MIN / b;
+}
+
+// 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_global = 0;
+
+STBIDEF void stbi_set_flip_vertically_on_load(int flag_true_if_should_flip)
+{
+ stbi__vertically_flip_on_load_global = flag_true_if_should_flip;
+}
+
+#ifndef STBI_THREAD_LOCAL
+#define stbi__vertically_flip_on_load stbi__vertically_flip_on_load_global
+#else
+static STBI_THREAD_LOCAL int stbi__vertically_flip_on_load_local, stbi__vertically_flip_on_load_set;
+
+STBIDEF void stbi_set_flip_vertically_on_load_thread(int flag_true_if_should_flip)
+{
+ stbi__vertically_flip_on_load_local = flag_true_if_should_flip;
+ stbi__vertically_flip_on_load_set = 1;
+}
+
+#define stbi__vertically_flip_on_load (stbi__vertically_flip_on_load_set \
+ ? stbi__vertically_flip_on_load_local \
+ : stbi__vertically_flip_on_load_global)
+#endif // STBI_THREAD_LOCAL
+
+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;
+
+ // test the formats with a very explicit header first (at least a FOURCC
+ // or distinctive magic number first)
+ #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);
+ #else
+ STBI_NOTUSED(bpc);
+ #endif
+ #ifndef STBI_NO_PIC
+ if (stbi__pic_test(s)) return stbi__pic_load(s,x,y,comp,req_comp, ri);
+ #endif
+
+ // then the formats that can end up attempting to load with just 1 or 2
+ // bytes matching expectations; these are prone to false positives, so
+ // try them later
+ #ifndef STBI_NO_JPEG
+ if (stbi__jpeg_test(s)) return stbi__jpeg_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;
+
+ // it is the responsibility of the loaders to make sure we get either 8 or 16 bit.
+ STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16);
+
+ if (ri.bits_per_channel != 8) {
+ 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;
+
+ // it is the responsibility of the loaders to make sure we get either 8 or 16 bit.
+ STBI_ASSERT(ri.bits_per_channel == 8 || ri.bits_per_channel == 16);
+
+ if (ri.bits_per_channel != 16) {
+ 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(_WIN32) && 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(_WIN32) && 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(_WIN32) && defined(STBI_WINDOWS_UTF8)
+ wchar_t wMode[64];
+ wchar_t wFilename[1024];
+ if (0 == MultiByteToWideChar(65001 /* UTF8 */, 0, filename, -1, wFilename, sizeof(wFilename)/sizeof(*wFilename)))
+ return 0;
+
+ if (0 == MultiByteToWideChar(65001 /* UTF8 */, 0, mode, -1, wMode, sizeof(wMode)/sizeof(*wMode)))
+ return 0;
+
+#if defined(_MSC_VER) && _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);
+ s->callback_already_read += (int) (s->img_buffer - s->img_buffer_original);
+ 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;
+}
+
+#if defined(STBI_NO_JPEG) && defined(STBI_NO_HDR) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM)
+// nothing
+#else
+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;
+}
+#endif
+
+#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC)
+// nothing
+#else
+static void stbi__skip(stbi__context *s, int n)
+{
+ if (n == 0) return; // already there!
+ 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;
+}
+#endif
+
+#if defined(STBI_NO_PNG) && defined(STBI_NO_TGA) && defined(STBI_NO_HDR) && defined(STBI_NO_PNM)
+// nothing
+#else
+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;
+}
+#endif
+
+#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && defined(STBI_NO_PIC)
+// nothing
+#else
+static int stbi__get16be(stbi__context *s)
+{
+ int z = stbi__get8(s);
+ return (z << 8) + stbi__get8(s);
+}
+#endif
+
+#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD) && defined(STBI_NO_PIC)
+// nothing
+#else
+static stbi__uint32 stbi__get32be(stbi__context *s)
+{
+ stbi__uint32 z = stbi__get16be(s);
+ return (z << 16) + stbi__get16be(s);
+}
+#endif
+
+#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);
+ z += (stbi__uint32)stbi__get16le(s) << 16;
+ return z;
+}
+#endif
+
+#define STBI__BYTECAST(x) ((stbi_uc) ((x) & 255)) // truncate int to byte without warnings
+
+#if defined(STBI_NO_JPEG) && defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM)
+// nothing
+#else
+//////////////////////////////////////////////////////////////////////////////
+//
+// 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);
+}
+#endif
+
+#if defined(STBI_NO_PNG) && defined(STBI_NO_BMP) && defined(STBI_NO_PSD) && defined(STBI_NO_TGA) && defined(STBI_NO_GIF) && defined(STBI_NO_PIC) && defined(STBI_NO_PNM)
+// nothing
+#else
+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); STBI_FREE(data); STBI_FREE(good); return stbi__errpuc("unsupported", "Unsupported format conversion");
+ }
+ #undef STBI__CASE
+ }
+
+ STBI_FREE(data);
+ return good;
+}
+#endif
+
+#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD)
+// nothing
+#else
+static stbi__uint16 stbi__compute_y_16(int r, int g, int b)
+{
+ return (stbi__uint16) (((r*77) + (g*150) + (29*b)) >> 8);
+}
+#endif
+
+#if defined(STBI_NO_PNG) && defined(STBI_NO_PSD)
+// nothing
+#else
+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); STBI_FREE(data); STBI_FREE(good); return (stbi__uint16*) stbi__errpuc("unsupported", "Unsupported format conversion");
+ }
+ #undef STBI__CASE
+ }
+
+ STBI_FREE(data);
+ return good;
+}
+#endif
+
+#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);
+ if(k >= 257) return stbi__err("bad size list","Corrupt JPEG");
+ }
+ }
+ 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];
+ if(c < 0 || c >= 256) // symbol id out of bounds!
+ return -1;
+ 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);
+ if (j->code_bits < n) return 0; // ran out of bits from stream, return 0s intead of continuing
+
+ sgn = j->code_buffer >> 31; // sign bit always in MSB; 0 if MSB clear (positive), 1 if MSB set (negative)
+ 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__jbias[n] & (sgn - 1));
+}
+
+// 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);
+ if (j->code_bits < n) return 0; // ran out of bits from stream, return 0s intead of continuing
+ 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);
+ if (j->code_bits < 1) return 0; // ran out of bits from stream, return 0s intead of continuing
+ 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 || t > 15) 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;
+ if (!stbi__addints_valid(j->img_comp[b].dc_pred, diff)) return stbi__err("bad delta","Corrupt JPEG");
+ dc = j->img_comp[b].dc_pred + diff;
+ j->img_comp[b].dc_pred = dc;
+ if (!stbi__mul2shorts_valid(dc, dequant[0])) return stbi__err("can't merge dc and ac", "Corrupt JPEG");
+ 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
+ if (s > j->code_bits) return stbi__err("bad huffman code", "Combined length longer than code bits available");
+ 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);
+ if (t < 0 || t > 15) return stbi__err("can't merge dc and ac", "Corrupt JPEG");
+ diff = t ? stbi__extend_receive(j, t) : 0;
+
+ if (!stbi__addints_valid(j->img_comp[b].dc_pred, diff)) return stbi__err("bad delta", "Corrupt JPEG");
+ dc = j->img_comp[b].dc_pred + diff;
+ j->img_comp[b].dc_pred = dc;
+ if (!stbi__mul2shorts_valid(dc, 1 << j->succ_low)) return stbi__err("can't merge dc and ac", "Corrupt JPEG");
+ data[0] = (short) (dc * (1 << 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
+ if (s > j->code_bits) return stbi__err("bad huffman code", "Combined length longer than code bits available");
+ j->code_buffer <<= s;
+ j->code_bits -= s;
+ zig = stbi__jpeg_dezigzag[k++];
+ data[zig] = (short) ((r >> 8) * (1 << 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) * (1 << 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];
+ }
+ if(n > 256) return stbi__err("bad DHT header","Corrupt JPEG"); // Loop over i < n would write past end of values!
+ 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
+ if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)");
+ if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)");
+ 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;
+ }
+
+ // check that plane subsampling factors are integer ratios; our resamplers can't deal with fractional ratios
+ // and I've never seen a non-corrupted JPEG file actually use them
+ for (i=0; i < s->img_n; ++i) {
+ if (h_max % z->img_comp[i].h != 0) return stbi__err("bad H","Corrupt JPEG");
+ if (v_max % z->img_comp[i].v != 0) return stbi__err("bad V","Corrupt JPEG");
+ }
+
+ // 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;
+}
+
+static stbi_uc stbi__skip_jpeg_junk_at_end(stbi__jpeg *j)
+{
+ // some JPEGs have junk at end, skip over it but if we find what looks
+ // like a valid marker, resume there
+ while (!stbi__at_eof(j->s)) {
+ stbi_uc x = stbi__get8(j->s);
+ while (x == 0xff) { // might be a marker
+ if (stbi__at_eof(j->s)) return STBI__MARKER_none;
+ x = stbi__get8(j->s);
+ if (x != 0x00 && x != 0xff) {
+ // not a stuffed zero or lead-in to another marker, looks
+ // like an actual marker, return it
+ return x;
+ }
+ // stuffed zero has x=0 now which ends the loop, meaning we go
+ // back to regular scan loop.
+ // repeated 0xff keeps trying to read the next byte of the marker.
+ }
+ }
+ return STBI__MARKER_none;
+}
+
+// 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 ) {
+ j->marker = stbi__skip_jpeg_junk_at_end(j);
+ // if we reach eof without hitting a marker, stbi__get_marker() below will fail and we'll eventually return 0
+ }
+ m = stbi__get_marker(j);
+ if (STBI__RESTART(m))
+ m = stbi__get_marker(j);
+ } 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");
+ m = stbi__get_marker(j);
+ } else {
+ if (!stbi__process_marker(j, m)) return 1;
+ 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;
+
+ // nothing to do if no components requested; check this now to avoid
+ // accessing uninitialized coutput[0] later
+ if (decode_n <= 0) { stbi__cleanup_jpeg(z); return NULL; }
+
+ // 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));
+ if (!j) return stbi__errpuc("outofmem", "Out of memory");
+ memset(j, 0, 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));
+ if (!j) return stbi__err("outofmem", "Out of memory");
+ memset(j, 0, 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)));
+ if (!j) return stbi__err("outofmem", "Out of memory");
+ memset(j, 0, 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)
+#define STBI__ZNSYMS 288 // number of symbols in literal/length alphabet
+
+// 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[STBI__ZNSYMS];
+ stbi__uint16 value[STBI__ZNSYMS];
+} 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;
+ int hit_zeof_once;
+ 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 int stbi__zeof(stbi__zbuf *z)
+{
+ return (z->zbuffer >= z->zbuffer_end);
+}
+
+stbi_inline static stbi_uc stbi__zget8(stbi__zbuf *z)
+{
+ return stbi__zeof(z) ? 0 : *z->zbuffer++;
+}
+
+static void stbi__fill_bits(stbi__zbuf *z)
+{
+ do {
+ if (z->code_buffer >= (1U << z->num_bits)) {
+ z->zbuffer = z->zbuffer_end; /* treat this as EOF so we fail. */
+ return;
+ }
+ 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];
+ if (b >= STBI__ZNSYMS) return -1; // some data was corrupt somewhere!
+ if (z->size[b] != s) return -1; // was originally an assert, but report failure instead.
+ 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) {
+ if (stbi__zeof(a)) {
+ if (!a->hit_zeof_once) {
+ // This is the first time we hit eof, insert 16 extra padding btis
+ // to allow us to keep going; if we actually consume any of them
+ // though, that is invalid data. This is caught later.
+ a->hit_zeof_once = 1;
+ a->num_bits += 16; // add 16 implicit zero bits
+ } else {
+ // We already inserted our extra 16 padding bits and are again
+ // out, this stream is actually prematurely terminated.
+ return -1;
+ }
+ } else {
+ 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;
+ unsigned int cur, limit, old_limit;
+ z->zout = zout;
+ if (!z->z_expandable) return stbi__err("output buffer limit","Corrupt PNG");
+ cur = (unsigned int) (z->zout - z->zout_start);
+ limit = old_limit = (unsigned) (z->zout_end - z->zout_start);
+ if (UINT_MAX - cur < (unsigned) n) return stbi__err("outofmem", "Out of memory");
+ while (cur + n > limit) {
+ if(limit > UINT_MAX / 2) return stbi__err("outofmem", "Out of memory");
+ 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;
+ if (a->hit_zeof_once && a->num_bits < 16) {
+ // The first time we hit zeof, we inserted 16 extra zero bits into our bit
+ // buffer so the decoder can just do its speculative decoding. But if we
+ // actually consumed any of those bits (which is the case when num_bits < 16),
+ // the stream actually read past the end so it is malformed.
+ return stbi__err("unexpected end","Corrupt PNG");
+ }
+ return 1;
+ }
+ if (z >= 286) return stbi__err("bad huffman code","Corrupt PNG"); // per DEFLATE, length codes 286 and 287 must not appear in compressed data
+ 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 || z >= 30) return stbi__err("bad huffman code","Corrupt PNG"); // per DEFLATE, distance codes 30 and 31 must not appear in compressed data
+ 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 (len > a->zout_end - zout) {
+ 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 if (c == 18) {
+ c = stbi__zreceive(a,7)+11;
+ } else {
+ return stbi__err("bad codelengths", "Corrupt PNG");
+ }
+ 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;
+ }
+ if (a->num_bits < 0) return stbi__err("zlib corrupt","Corrupt PNG");
+ // 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 (stbi__zeof(a)) return stbi__err("bad zlib header","Corrupt PNG"); // zlib spec
+ 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[STBI__ZNSYMS] =
+{
+ 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
+ 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
+ 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
+ 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,
+ 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;
+ a->hit_zeof_once = 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 , STBI__ZNSYMS)) 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 filter used for first scanline to avoid needing a dummy row of 0s
+ STBI__F_avg_first
+};
+
+static stbi_uc first_row_filter[5] =
+{
+ STBI__F_none,
+ STBI__F_sub,
+ STBI__F_none,
+ STBI__F_avg_first,
+ STBI__F_sub // Paeth with b=c=0 turns out to be equivalent to sub
+};
+
+static int stbi__paeth(int a, int b, int c)
+{
+ // This formulation looks very different from the reference in the PNG spec, but is
+ // actually equivalent and has favorable data dependencies and admits straightforward
+ // generation of branch-free code, which helps performance significantly.
+ int thresh = c*3 - (a + b);
+ int lo = a < b ? a : b;
+ int hi = a < b ? b : a;
+ int t0 = (hi <= thresh) ? lo : c;
+ int t1 = (thresh <= lo) ? hi : t0;
+ return t1;
+}
+
+static const stbi_uc stbi__depth_scale_table[9] = { 0, 0xff, 0x55, 0, 0x11, 0,0,0, 0x01 };
+
+// adds an extra all-255 alpha channel
+// dest == src is legal
+// img_n must be 1 or 3
+static void stbi__create_png_alpha_expand8(stbi_uc *dest, stbi_uc *src, stbi__uint32 x, int img_n)
+{
+ int i;
+ // must process data backwards since we allow dest==src
+ if (img_n == 1) {
+ for (i=x-1; i >= 0; --i) {
+ dest[i*2+1] = 255;
+ dest[i*2+0] = src[i];
+ }
+ } else {
+ STBI_ASSERT(img_n == 3);
+ for (i=x-1; i >= 0; --i) {
+ dest[i*4+3] = 255;
+ dest[i*4+2] = src[i*3+2];
+ dest[i*4+1] = src[i*3+1];
+ dest[i*4+0] = src[i*3+0];
+ }
+ }
+}
+
+// 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;
+ stbi_uc *filter_buf;
+ int all_ok = 1;
+ 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");
+
+ // note: error exits here don't need to clean up a->out individually,
+ // stbi__do_png always does on error.
+ 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);
+ if (!stbi__mad2sizes_valid(img_width_bytes, y, img_width_bytes)) return stbi__err("too large", "Corrupt PNG");
+ 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");
+
+ // Allocate two scan lines worth of filter workspace buffer.
+ filter_buf = (stbi_uc *) stbi__malloc_mad2(img_width_bytes, 2, 0);
+ if (!filter_buf) return stbi__err("outofmem", "Out of memory");
+
+ // Filtering for low-bit-depth images
+ if (depth < 8) {
+ filter_bytes = 1;
+ width = img_width_bytes;
+ }
+
+ for (j=0; j < y; ++j) {
+ // cur/prior filter buffers alternate
+ stbi_uc *cur = filter_buf + (j & 1)*img_width_bytes;
+ stbi_uc *prior = filter_buf + (~j & 1)*img_width_bytes;
+ stbi_uc *dest = a->out + stride*j;
+ int nk = width * filter_bytes;
+ int filter = *raw++;
+
+ // check filter type
+ if (filter > 4) {
+ all_ok = stbi__err("invalid filter","Corrupt PNG");
+ break;
+ }
+
+ // if first row, use special filter that doesn't sample previous row
+ if (j == 0) filter = first_row_filter[filter];
+
+ // perform actual filtering
+ switch (filter) {
+ case STBI__F_none:
+ memcpy(cur, raw, nk);
+ break;
+ case STBI__F_sub:
+ memcpy(cur, raw, filter_bytes);
+ for (k = filter_bytes; k < nk; ++k)
+ cur[k] = STBI__BYTECAST(raw[k] + cur[k-filter_bytes]);
+ break;
+ case STBI__F_up:
+ for (k = 0; k < nk; ++k)
+ cur[k] = STBI__BYTECAST(raw[k] + prior[k]);
+ break;
+ case STBI__F_avg:
+ for (k = 0; k < filter_bytes; ++k)
+ cur[k] = STBI__BYTECAST(raw[k] + (prior[k]>>1));
+ for (k = filter_bytes; k < nk; ++k)
+ cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k-filter_bytes])>>1));
+ break;
+ case STBI__F_paeth:
+ for (k = 0; k < filter_bytes; ++k)
+ cur[k] = STBI__BYTECAST(raw[k] + prior[k]); // prior[k] == stbi__paeth(0,prior[k],0)
+ for (k = filter_bytes; k < nk; ++k)
+ cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-filter_bytes], prior[k], prior[k-filter_bytes]));
+ break;
+ case STBI__F_avg_first:
+ memcpy(cur, raw, filter_bytes);
+ for (k = filter_bytes; k < nk; ++k)
+ cur[k] = STBI__BYTECAST(raw[k] + (cur[k-filter_bytes] >> 1));
+ break;
+ }
+
+ raw += nk;
+
+ // expand decoded bits in cur to dest, also adding an extra alpha channel if desired
+ if (depth < 8) {
+ stbi_uc scale = (color == 0) ? stbi__depth_scale_table[depth] : 1; // scale grayscale values to 0..255 range
+ stbi_uc *in = cur;
+ stbi_uc *out = dest;
+ stbi_uc inb = 0;
+ stbi__uint32 nsmp = x*img_n;
+
+ // expand bits to bytes first
+ if (depth == 4) {
+ for (i=0; i < nsmp; ++i) {
+ if ((i & 1) == 0) inb = *in++;
+ *out++ = scale * (inb >> 4);
+ inb <<= 4;
+ }
+ } else if (depth == 2) {
+ for (i=0; i < nsmp; ++i) {
+ if ((i & 3) == 0) inb = *in++;
+ *out++ = scale * (inb >> 6);
+ inb <<= 2;
+ }
+ } else {
+ STBI_ASSERT(depth == 1);
+ for (i=0; i < nsmp; ++i) {
+ if ((i & 7) == 0) inb = *in++;
+ *out++ = scale * (inb >> 7);
+ inb <<= 1;
+ }
+ }
+
+ // insert alpha=255 values if desired
+ if (img_n != out_n)
+ stbi__create_png_alpha_expand8(dest, dest, x, img_n);
+ } else if (depth == 8) {
+ if (img_n == out_n)
+ memcpy(dest, cur, x*img_n);
+ else
+ stbi__create_png_alpha_expand8(dest, cur, x, img_n);
+ } else if (depth == 16) {
+ // convert the image data from big-endian to platform-native
+ stbi__uint16 *dest16 = (stbi__uint16*)dest;
+ stbi__uint32 nsmp = x*img_n;
+
+ if (img_n == out_n) {
+ for (i = 0; i < nsmp; ++i, ++dest16, cur += 2)
+ *dest16 = (cur[0] << 8) | cur[1];
+ } else {
+ STBI_ASSERT(img_n+1 == out_n);
+ if (img_n == 1) {
+ for (i = 0; i < x; ++i, dest16 += 2, cur += 2) {
+ dest16[0] = (cur[0] << 8) | cur[1];
+ dest16[1] = 0xffff;
+ }
+ } else {
+ STBI_ASSERT(img_n == 3);
+ for (i = 0; i < x; ++i, dest16 += 4, cur += 6) {
+ dest16[0] = (cur[0] << 8) | cur[1];
+ dest16[1] = (cur[2] << 8) | cur[3];
+ dest16[2] = (cur[4] << 8) | cur[5];
+ dest16[3] = 0xffff;
+ }
+ }
+ }
+ }
+ }
+
+ STBI_FREE(filter_buf);
+ if (!all_ok) return 0;
+
+ 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);
+ if (!final) return stbi__err("outofmem", "Out of memory");
+ 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_global = 0;
+static int stbi__de_iphone_flag_global = 0;
+
+STBIDEF void stbi_set_unpremultiply_on_load(int flag_true_if_should_unpremultiply)
+{
+ stbi__unpremultiply_on_load_global = flag_true_if_should_unpremultiply;
+}
+
+STBIDEF void stbi_convert_iphone_png_to_rgb(int flag_true_if_should_convert)
+{
+ stbi__de_iphone_flag_global = flag_true_if_should_convert;
+}
+
+#ifndef STBI_THREAD_LOCAL
+#define stbi__unpremultiply_on_load stbi__unpremultiply_on_load_global
+#define stbi__de_iphone_flag stbi__de_iphone_flag_global
+#else
+static STBI_THREAD_LOCAL int stbi__unpremultiply_on_load_local, stbi__unpremultiply_on_load_set;
+static STBI_THREAD_LOCAL int stbi__de_iphone_flag_local, stbi__de_iphone_flag_set;
+
+STBIDEF void stbi_set_unpremultiply_on_load_thread(int flag_true_if_should_unpremultiply)
+{
+ stbi__unpremultiply_on_load_local = flag_true_if_should_unpremultiply;
+ stbi__unpremultiply_on_load_set = 1;
+}
+
+STBIDEF void stbi_convert_iphone_png_to_rgb_thread(int flag_true_if_should_convert)
+{
+ stbi__de_iphone_flag_local = flag_true_if_should_convert;
+ stbi__de_iphone_flag_set = 1;
+}
+
+#define stbi__unpremultiply_on_load (stbi__unpremultiply_on_load_set \
+ ? stbi__unpremultiply_on_load_local \
+ : stbi__unpremultiply_on_load_global)
+#define stbi__de_iphone_flag (stbi__de_iphone_flag_set \
+ ? stbi__de_iphone_flag_local \
+ : stbi__de_iphone_flag_global)
+#endif // STBI_THREAD_LOCAL
+
+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);
+ s->img_y = stbi__get32be(s);
+ if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)");
+ if (s->img_x > STBI_MAX_DIMENSIONS) 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");
+ } 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");
+ }
+ // even with 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;
+ // non-paletted with tRNS = constant alpha. if header-scanning, we can stop now.
+ if (scan == STBI__SCAN_header) { ++s->img_n; return 1; }
+ if (z->depth == 16) {
+ for (k = 0; k < s->img_n && k < 3; ++k) // extra loop test to suppress false GCC warning
+ tc16[k] = (stbi__uint16)stbi__get16be(s); // copy the values as-is
+ } else {
+ for (k = 0; k < s->img_n && k < 3; ++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) {
+ // header scan definitely stops at first IDAT
+ if (pal_img_n)
+ s->img_n = pal_img_n;
+ return 1;
+ }
+ if (c.length > (1u << 30)) return stbi__err("IDAT size limit", "IDAT section larger than 2^30 bytes");
+ 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;
+ // end of PNG chunk, read and skip CRC
+ stbi__get32be(s);
+ 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 if (p->depth == 16)
+ ri->bits_per_channel = 16;
+ else
+ return stbi__errpuc("bad bits_per_channel", "PNG not supported: unsupported color 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;/* >>= 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 < 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;
+ int extra_read;
+} stbi__bmp_data;
+
+static int stbi__bmp_set_mask_defaults(stbi__bmp_data *info, int compress)
+{
+ // BI_BITFIELDS specifies masks explicitly, don't override
+ if (compress == 3)
+ return 1;
+
+ if (compress == 0) {
+ if (info->bpp == 16) {
+ info->mr = 31u << 10;
+ info->mg = 31u << 5;
+ info->mb = 31u << 0;
+ } else 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 {
+ // otherwise, use defaults, which is all-0
+ info->mr = info->mg = info->mb = info->ma = 0;
+ }
+ return 1;
+ }
+ return 0; // error
+}
+
+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;
+ info->extra_read = 14;
+
+ if (info->offset < 0) return stbi__errpuc("bad BMP", "bad BMP");
+
+ 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");
+ if (compress >= 4) return stbi__errpuc("BMP JPEG/PNG", "BMP type not supported: unsupported compression"); // this includes PNG/JPEG modes
+ if (compress == 3 && info->bpp != 16 && info->bpp != 32) return stbi__errpuc("bad BMP", "bad BMP"); // bitfields requires 16 or 32 bits/pixel
+ 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) {
+ stbi__bmp_set_mask_defaults(info, compress);
+ } else if (compress == 3) {
+ info->mr = stbi__get32le(s);
+ info->mg = stbi__get32le(s);
+ info->mb = stbi__get32le(s);
+ info->extra_read += 12;
+ // 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 {
+ // V4/V5 header
+ 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);
+ if (compress != 3) // override mr/mg/mb unless in BI_BITFIELDS mode, as per docs
+ stbi__bmp_set_mask_defaults(info, compress);
+ 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);
+
+ if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)");
+ if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)");
+
+ 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 - info.extra_read - 24) / 3;
+ } else {
+ if (info.bpp < 16)
+ psize = (info.offset - info.extra_read - info.hsz) >> 2;
+ }
+ if (psize == 0) {
+ // accept some number of extra bytes after the header, but if the offset points either to before
+ // the header ends or implies a large amount of extra data, reject the file as malformed
+ int bytes_read_so_far = s->callback_already_read + (int)(s->img_buffer - s->img_buffer_original);
+ int header_limit = 1024; // max we actually read is below 256 bytes currently.
+ int extra_data_limit = 256*4; // what ordinarily goes here is a palette; 256 entries*4 bytes is its max size.
+ if (bytes_read_so_far <= 0 || bytes_read_so_far > header_limit) {
+ return stbi__errpuc("bad header", "Corrupt BMP");
+ }
+ // we established that bytes_read_so_far is positive and sensible.
+ // the first half of this test rejects offsets that are either too small positives, or
+ // negative, and guarantees that info.offset >= bytes_read_so_far > 0. this in turn
+ // ensures the number computed in the second half of the test can't overflow.
+ if (info.offset < bytes_read_so_far || info.offset - bytes_read_so_far > extra_data_limit) {
+ return stbi__errpuc("bad offset", "Corrupt BMP");
+ } else {
+ stbi__skip(s, info.offset - bytes_read_so_far);
+ }
+ }
+
+ if (info.bpp == 24 && ma == 0xff000000)
+ s->img_n = 3;
+ else
+ 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 - info.extra_read - 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 - info.extra_read - 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);
+ if (rcount > 8 || gcount > 8 || bcount > 8 || acount > 8) { STBI_FREE(out); return stbi__errpuc("bad masks", "Corrupt BMP"); }
+ }
+ 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);
+ STBI_NOTUSED(tga_x_origin); // @TODO
+ STBI_NOTUSED(tga_y_origin); // @TODO
+
+ if (tga_height > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)");
+ if (tga_width > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)");
+
+ // 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)
+ {
+ if (tga_palette_len == 0) { /* you have to have at least one entry! */
+ STBI_FREE(tga_data);
+ return stbi__errpuc("bad palette", "Corrupt TGA");
+ }
+
+ // 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;
+ STBI_NOTUSED(tga_palette_start);
+ // 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);
+
+ if (h > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)");
+ if (w > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)");
+
+ // 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 (y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)");
+ if (x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)");
+
+ 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);
+ if (!result) return stbi__errpuc("outofmem", "Out of memory");
+ 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 (g->w > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)");
+ if (g->h > STBI_MAX_DIMENSIONS) return stbi__err("too large","Very large image (corrupt?)");
+
+ 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 (!g) return stbi__err("outofmem", "Out of memory");
+ 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 dispose 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_outofmem(stbi__gif *g, stbi_uc *out, int **delays)
+{
+ STBI_FREE(g->out);
+ STBI_FREE(g->history);
+ STBI_FREE(g->background);
+
+ if (out) STBI_FREE(out);
+ if (delays && *delays) STBI_FREE(*delays);
+ return stbi__errpuc("outofmem", "Out of memory");
+}
+
+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;
+ int out_size = 0;
+ int delays_size = 0;
+
+ STBI_NOTUSED(out_size);
+ STBI_NOTUSED(delays_size);
+
+ 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) {
+ void *tmp = (stbi_uc*) STBI_REALLOC_SIZED( out, out_size, layers * stride );
+ if (!tmp)
+ return stbi__load_gif_main_outofmem(&g, out, delays);
+ else {
+ out = (stbi_uc*) tmp;
+ out_size = layers * stride;
+ }
+
+ if (delays) {
+ int *new_delays = (int*) STBI_REALLOC_SIZED( *delays, delays_size, sizeof(int) * layers );
+ if (!new_delays)
+ return stbi__load_gif_main_outofmem(&g, out, delays);
+ *delays = new_delays;
+ delays_size = layers * sizeof(int);
+ }
+ } else {
+ out = (stbi_uc*)stbi__malloc( layers * stride );
+ if (!out)
+ return stbi__load_gif_main_outofmem(&g, out, delays);
+ out_size = layers * stride;
+ if (delays) {
+ *delays = (int*) stbi__malloc( layers * sizeof(int) );
+ if (!*delays)
+ return stbi__load_gif_main_outofmem(&g, out, delays);
+ delays_size = 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);
+
+ if (height > STBI_MAX_DIMENSIONS) return stbi__errpf("too large","Very large image (corrupt?)");
+ if (width > STBI_MAX_DIMENSIONS) return stbi__errpf("too large","Very large image (corrupt?)");
+
+ *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 == 0) || (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 == 0) || (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);
+ if (p == NULL) {
+ stbi__rewind( s );
+ return 0;
+ }
+ if (x) *x = s->img_x;
+ if (y) *y = s->img_y;
+ if (comp) {
+ if (info.bpp == 24 && info.ma == 0xff000000)
+ *comp = 3;
+ else
+ *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;
+ }
+ STBI_NOTUSED(stbi__get32be(s));
+ STBI_NOTUSED(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)
+
+#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);
+
+ ri->bits_per_channel = stbi__pnm_info(s, (int *)&s->img_x, (int *)&s->img_y, (int *)&s->img_n);
+ if (ri->bits_per_channel == 0)
+ return 0;
+
+ if (s->img_y > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)");
+ if (s->img_x > STBI_MAX_DIMENSIONS) return stbi__errpuc("too large","Very large image (corrupt?)");
+
+ *x = s->img_x;
+ *y = s->img_y;
+ if (comp) *comp = s->img_n;
+
+ if (!stbi__mad4sizes_valid(s->img_n, s->img_x, s->img_y, ri->bits_per_channel / 8, 0))
+ return stbi__errpuc("too large", "PNM too large");
+
+ out = (stbi_uc *) stbi__malloc_mad4(s->img_n, s->img_x, s->img_y, ri->bits_per_channel / 8, 0);
+ if (!out) return stbi__errpuc("outofmem", "Out of memory");
+ if (!stbi__getn(s, out, s->img_n * s->img_x * s->img_y * (ri->bits_per_channel / 8))) {
+ STBI_FREE(out);
+ return stbi__errpuc("bad PNM", "PNM file truncated");
+ }
+
+ if (req_comp && req_comp != s->img_n) {
+ if (ri->bits_per_channel == 16) {
+ out = (stbi_uc *) stbi__convert_format16((stbi__uint16 *) out, s->img_n, req_comp, s->img_x, s->img_y);
+ } else {
+ 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);
+ if((value > 214748364) || (value == 214748364 && *c > '7'))
+ return stbi__err("integer parse overflow", "Parsing an integer in the PPM header overflowed a 32-bit int");
+ }
+
+ 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
+ if(*x == 0)
+ return stbi__err("invalid width", "PPM image header had zero or overflowing width");
+ stbi__pnm_skip_whitespace(s, &c);
+
+ *y = stbi__pnm_getinteger(s, &c); // read height
+ if (*y == 0)
+ return stbi__err("invalid width", "PPM image header had zero or overflowing width");
+ stbi__pnm_skip_whitespace(s, &c);
+
+ maxv = stbi__pnm_getinteger(s, &c); // read max value
+ if (maxv > 65535)
+ return stbi__err("max value > 65535", "PPM image supports only 8-bit and 16-bit images");
+ else if (maxv > 255)
+ return 16;
+ else
+ return 8;
+}
+
+static int stbi__pnm_is16(stbi__context *s)
+{
+ if (stbi__pnm_info(s, NULL, NULL, NULL) == 16)
+ return 1;
+ return 0;
+}
+#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
+
+ #ifndef STBI_NO_PNM
+ if (stbi__pnm_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/src/external/xkb-keysym-utf.c b/src/external/xkb-keysym-utf.c
@@ -0,0 +1,1043 @@
+#define XKB_KEY_KP_Space 0xff80
+#define XKB_KEY_space 0x0020
+#define XKB_KEY_BackSpace 65288
+#define XKB_KEY_Clear 65291
+#define XKB_KEY_KP_Multiply 65450
+#define XKB_KEY_KP_9 65465
+#define XKB_KEY_Return 65293
+#define XKB_KEY_Escape 65307
+#define XKB_KEY_Delete 65535
+#define XKB_KEY_KP_Tab 65289
+#define XKB_KEY_KP_Enter 65421
+#define XKB_KEY_KP_Equal 65469
+
+#define XKB_KEYSYM_UNICODE_OFFSET 0x01000000
+#define XKB_KEYSYM_UNICODE_MAX 0x0110ffff
+
+#define XKB_KEYSYM_UTF8_MAX_SIZE 5
+
+#define ARRAY_SIZE(__ARR) sizeof(__ARR) / sizeof(__ARR[0])
+
+#define XKB_KEY_NoSymbol 0
+
+/* The table and comments below along with the function xkb_keysym_to_ucs4
+ * are under the public domain and are derived as described below.
+ */
+/* This module converts keysym values into the corresponding ISO 10646
+ * (UCS, Unicode) values.
+ *
+ * The array keysymtab[] contains pairs of X11 keysym values for graphical
+ * characters and the corresponding Unicode value. The function
+ * keysym2ucs() maps a keysym onto a Unicode value using a binary search,
+ * therefore keysymtab[] must remain SORTED by keysym value.
+ *
+ * The keysym -> UTF-8 conversion will hopefully one day be provided
+ * by Xlib via XmbLookupString() and should ideally not have to be
+ * done in X applications. But we are not there yet.
+ *
+ * We allow to represent any UCS character in the range U-00000000 to
+ * U-00FFFFFF by a keysym value in the range 0x01000000 to 0x01ffffff.
+ * This admittedly does not cover the entire 31-bit space of UCS, but
+ * it does cover all of the characters up to U-10FFFF, which can be
+ * represented by UTF-16, and more, and it is very unlikely that higher
+ * UCS codes will ever be assigned by ISO. So to get Unicode character
+ * U+ABCD you can directly use keysym 0x0100abcd.
+ *
+ * NOTE: The comments in the table below contain the actual character
+ * encoded in UTF-8, so for viewing and editing best use an editor in
+ * UTF-8 mode.
+ *
+ * Author: Markus G. Kuhn <http://www.cl.cam.ac.uk/~mgk25/>,
+ * University of Cambridge, April 2001
+ *
+ * Special thanks to Richard Verhoeven <river@win.tue.nl> for preparing
+ * an initial draft of the mapping table.
+ *
+ * This software is in the public domain. Share and enjoy!
+ *
+ */
+
+#define NO_KEYSYM_UNICODE_CONVERSION 0
+
+/* We don't use the uint32_t types here, to save some space. */
+struct codepair {
+ uint16_t keysym;
+ uint16_t ucs;
+};
+
+static const struct codepair keysymtab[] = {
+ { 0x01a1, 0x0104 }, /* Aogonek Ą LATIN CAPITAL LETTER A WITH OGONEK */
+ { 0x01a2, 0x02d8 }, /* breve ˘ BREVE */
+ { 0x01a3, 0x0141 }, /* Lstroke Ł LATIN CAPITAL LETTER L WITH STROKE */
+ { 0x01a5, 0x013d }, /* Lcaron Ľ LATIN CAPITAL LETTER L WITH CARON */
+ { 0x01a6, 0x015a }, /* Sacute Ś LATIN CAPITAL LETTER S WITH ACUTE */
+ { 0x01a9, 0x0160 }, /* Scaron Š LATIN CAPITAL LETTER S WITH CARON */
+ { 0x01aa, 0x015e }, /* Scedilla Ş LATIN CAPITAL LETTER S WITH CEDILLA */
+ { 0x01ab, 0x0164 }, /* Tcaron Ť LATIN CAPITAL LETTER T WITH CARON */
+ { 0x01ac, 0x0179 }, /* Zacute Ź LATIN CAPITAL LETTER Z WITH ACUTE */
+ { 0x01ae, 0x017d }, /* Zcaron Ž LATIN CAPITAL LETTER Z WITH CARON */
+ { 0x01af, 0x017b }, /* Zabovedot Ż LATIN CAPITAL LETTER Z WITH DOT ABOVE */
+ { 0x01b1, 0x0105 }, /* aogonek ą LATIN SMALL LETTER A WITH OGONEK */
+ { 0x01b2, 0x02db }, /* ogonek ˛ OGONEK */
+ { 0x01b3, 0x0142 }, /* lstroke ł LATIN SMALL LETTER L WITH STROKE */
+ { 0x01b5, 0x013e }, /* lcaron ľ LATIN SMALL LETTER L WITH CARON */
+ { 0x01b6, 0x015b }, /* sacute ś LATIN SMALL LETTER S WITH ACUTE */
+ { 0x01b7, 0x02c7 }, /* caron ˇ CARON */
+ { 0x01b9, 0x0161 }, /* scaron š LATIN SMALL LETTER S WITH CARON */
+ { 0x01ba, 0x015f }, /* scedilla ş LATIN SMALL LETTER S WITH CEDILLA */
+ { 0x01bb, 0x0165 }, /* tcaron ť LATIN SMALL LETTER T WITH CARON */
+ { 0x01bc, 0x017a }, /* zacute ź LATIN SMALL LETTER Z WITH ACUTE */
+ { 0x01bd, 0x02dd }, /* doubleacute ˝ DOUBLE ACUTE ACCENT */
+ { 0x01be, 0x017e }, /* zcaron ž LATIN SMALL LETTER Z WITH CARON */
+ { 0x01bf, 0x017c }, /* zabovedot ż LATIN SMALL LETTER Z WITH DOT ABOVE */
+ { 0x01c0, 0x0154 }, /* Racute Ŕ LATIN CAPITAL LETTER R WITH ACUTE */
+ { 0x01c3, 0x0102 }, /* Abreve Ă LATIN CAPITAL LETTER A WITH BREVE */
+ { 0x01c5, 0x0139 }, /* Lacute Ĺ LATIN CAPITAL LETTER L WITH ACUTE */
+ { 0x01c6, 0x0106 }, /* Cacute Ć LATIN CAPITAL LETTER C WITH ACUTE */
+ { 0x01c8, 0x010c }, /* Ccaron Č LATIN CAPITAL LETTER C WITH CARON */
+ { 0x01ca, 0x0118 }, /* Eogonek Ę LATIN CAPITAL LETTER E WITH OGONEK */
+ { 0x01cc, 0x011a }, /* Ecaron Ě LATIN CAPITAL LETTER E WITH CARON */
+ { 0x01cf, 0x010e }, /* Dcaron Ď LATIN CAPITAL LETTER D WITH CARON */
+ { 0x01d0, 0x0110 }, /* Dstroke Đ LATIN CAPITAL LETTER D WITH STROKE */
+ { 0x01d1, 0x0143 }, /* Nacute Ń LATIN CAPITAL LETTER N WITH ACUTE */
+ { 0x01d2, 0x0147 }, /* Ncaron Ň LATIN CAPITAL LETTER N WITH CARON */
+ { 0x01d5, 0x0150 }, /* Odoubleacute Ő LATIN CAPITAL LETTER O WITH DOUBLE ACUTE */
+ { 0x01d8, 0x0158 }, /* Rcaron Ř LATIN CAPITAL LETTER R WITH CARON */
+ { 0x01d9, 0x016e }, /* Uring Ů LATIN CAPITAL LETTER U WITH RING ABOVE */
+ { 0x01db, 0x0170 }, /* Udoubleacute Ű LATIN CAPITAL LETTER U WITH DOUBLE ACUTE */
+ { 0x01de, 0x0162 }, /* Tcedilla Ţ LATIN CAPITAL LETTER T WITH CEDILLA */
+ { 0x01e0, 0x0155 }, /* racute ŕ LATIN SMALL LETTER R WITH ACUTE */
+ { 0x01e3, 0x0103 }, /* abreve ă LATIN SMALL LETTER A WITH BREVE */
+ { 0x01e5, 0x013a }, /* lacute ĺ LATIN SMALL LETTER L WITH ACUTE */
+ { 0x01e6, 0x0107 }, /* cacute ć LATIN SMALL LETTER C WITH ACUTE */
+ { 0x01e8, 0x010d }, /* ccaron č LATIN SMALL LETTER C WITH CARON */
+ { 0x01ea, 0x0119 }, /* eogonek ę LATIN SMALL LETTER E WITH OGONEK */
+ { 0x01ec, 0x011b }, /* ecaron ě LATIN SMALL LETTER E WITH CARON */
+ { 0x01ef, 0x010f }, /* dcaron ď LATIN SMALL LETTER D WITH CARON */
+ { 0x01f0, 0x0111 }, /* dstroke đ LATIN SMALL LETTER D WITH STROKE */
+ { 0x01f1, 0x0144 }, /* nacute ń LATIN SMALL LETTER N WITH ACUTE */
+ { 0x01f2, 0x0148 }, /* ncaron ň LATIN SMALL LETTER N WITH CARON */
+ { 0x01f5, 0x0151 }, /* odoubleacute ő LATIN SMALL LETTER O WITH DOUBLE ACUTE */
+ { 0x01f8, 0x0159 }, /* rcaron ř LATIN SMALL LETTER R WITH CARON */
+ { 0x01f9, 0x016f }, /* uring ů LATIN SMALL LETTER U WITH RING ABOVE */
+ { 0x01fb, 0x0171 }, /* udoubleacute ű LATIN SMALL LETTER U WITH DOUBLE ACUTE */
+ { 0x01fe, 0x0163 }, /* tcedilla ţ LATIN SMALL LETTER T WITH CEDILLA */
+ { 0x01ff, 0x02d9 }, /* abovedot ˙ DOT ABOVE */
+ { 0x02a1, 0x0126 }, /* Hstroke Ħ LATIN CAPITAL LETTER H WITH STROKE */
+ { 0x02a6, 0x0124 }, /* Hcircumflex Ĥ LATIN CAPITAL LETTER H WITH CIRCUMFLEX */
+ { 0x02a9, 0x0130 }, /* Iabovedot İ LATIN CAPITAL LETTER I WITH DOT ABOVE */
+ { 0x02ab, 0x011e }, /* Gbreve Ğ LATIN CAPITAL LETTER G WITH BREVE */
+ { 0x02ac, 0x0134 }, /* Jcircumflex Ĵ LATIN CAPITAL LETTER J WITH CIRCUMFLEX */
+ { 0x02b1, 0x0127 }, /* hstroke ħ LATIN SMALL LETTER H WITH STROKE */
+ { 0x02b6, 0x0125 }, /* hcircumflex ĥ LATIN SMALL LETTER H WITH CIRCUMFLEX */
+ { 0x02b9, 0x0131 }, /* idotless ı LATIN SMALL LETTER DOTLESS I */
+ { 0x02bb, 0x011f }, /* gbreve ğ LATIN SMALL LETTER G WITH BREVE */
+ { 0x02bc, 0x0135 }, /* jcircumflex ĵ LATIN SMALL LETTER J WITH CIRCUMFLEX */
+ { 0x02c5, 0x010a }, /* Cabovedot Ċ LATIN CAPITAL LETTER C WITH DOT ABOVE */
+ { 0x02c6, 0x0108 }, /* Ccircumflex Ĉ LATIN CAPITAL LETTER C WITH CIRCUMFLEX */
+ { 0x02d5, 0x0120 }, /* Gabovedot Ġ LATIN CAPITAL LETTER G WITH DOT ABOVE */
+ { 0x02d8, 0x011c }, /* Gcircumflex Ĝ LATIN CAPITAL LETTER G WITH CIRCUMFLEX */
+ { 0x02dd, 0x016c }, /* Ubreve Ŭ LATIN CAPITAL LETTER U WITH BREVE */
+ { 0x02de, 0x015c }, /* Scircumflex Ŝ LATIN CAPITAL LETTER S WITH CIRCUMFLEX */
+ { 0x02e5, 0x010b }, /* cabovedot ċ LATIN SMALL LETTER C WITH DOT ABOVE */
+ { 0x02e6, 0x0109 }, /* ccircumflex ĉ LATIN SMALL LETTER C WITH CIRCUMFLEX */
+ { 0x02f5, 0x0121 }, /* gabovedot ġ LATIN SMALL LETTER G WITH DOT ABOVE */
+ { 0x02f8, 0x011d }, /* gcircumflex ĝ LATIN SMALL LETTER G WITH CIRCUMFLEX */
+ { 0x02fd, 0x016d }, /* ubreve ŭ LATIN SMALL LETTER U WITH BREVE */
+ { 0x02fe, 0x015d }, /* scircumflex ŝ LATIN SMALL LETTER S WITH CIRCUMFLEX */
+ { 0x03a2, 0x0138 }, /* kra ĸ LATIN SMALL LETTER KRA */
+ { 0x03a3, 0x0156 }, /* Rcedilla Ŗ LATIN CAPITAL LETTER R WITH CEDILLA */
+ { 0x03a5, 0x0128 }, /* Itilde Ĩ LATIN CAPITAL LETTER I WITH TILDE */
+ { 0x03a6, 0x013b }, /* Lcedilla Ļ LATIN CAPITAL LETTER L WITH CEDILLA */
+ { 0x03aa, 0x0112 }, /* Emacron Ē LATIN CAPITAL LETTER E WITH MACRON */
+ { 0x03ab, 0x0122 }, /* Gcedilla Ģ LATIN CAPITAL LETTER G WITH CEDILLA */
+ { 0x03ac, 0x0166 }, /* Tslash Ŧ LATIN CAPITAL LETTER T WITH STROKE */
+ { 0x03b3, 0x0157 }, /* rcedilla ŗ LATIN SMALL LETTER R WITH CEDILLA */
+ { 0x03b5, 0x0129 }, /* itilde ĩ LATIN SMALL LETTER I WITH TILDE */
+ { 0x03b6, 0x013c }, /* lcedilla ļ LATIN SMALL LETTER L WITH CEDILLA */
+ { 0x03ba, 0x0113 }, /* emacron ē LATIN SMALL LETTER E WITH MACRON */
+ { 0x03bb, 0x0123 }, /* gcedilla ģ LATIN SMALL LETTER G WITH CEDILLA */
+ { 0x03bc, 0x0167 }, /* tslash ŧ LATIN SMALL LETTER T WITH STROKE */
+ { 0x03bd, 0x014a }, /* ENG Ŋ LATIN CAPITAL LETTER ENG */
+ { 0x03bf, 0x014b }, /* eng ŋ LATIN SMALL LETTER ENG */
+ { 0x03c0, 0x0100 }, /* Amacron Ā LATIN CAPITAL LETTER A WITH MACRON */
+ { 0x03c7, 0x012e }, /* Iogonek Į LATIN CAPITAL LETTER I WITH OGONEK */
+ { 0x03cc, 0x0116 }, /* Eabovedot Ė LATIN CAPITAL LETTER E WITH DOT ABOVE */
+ { 0x03cf, 0x012a }, /* Imacron Ī LATIN CAPITAL LETTER I WITH MACRON */
+ { 0x03d1, 0x0145 }, /* Ncedilla Ņ LATIN CAPITAL LETTER N WITH CEDILLA */
+ { 0x03d2, 0x014c }, /* Omacron Ō LATIN CAPITAL LETTER O WITH MACRON */
+ { 0x03d3, 0x0136 }, /* Kcedilla Ķ LATIN CAPITAL LETTER K WITH CEDILLA */
+ { 0x03d9, 0x0172 }, /* Uogonek Ų LATIN CAPITAL LETTER U WITH OGONEK */
+ { 0x03dd, 0x0168 }, /* Utilde Ũ LATIN CAPITAL LETTER U WITH TILDE */
+ { 0x03de, 0x016a }, /* Umacron Ū LATIN CAPITAL LETTER U WITH MACRON */
+ { 0x03e0, 0x0101 }, /* amacron ā LATIN SMALL LETTER A WITH MACRON */
+ { 0x03e7, 0x012f }, /* iogonek į LATIN SMALL LETTER I WITH OGONEK */
+ { 0x03ec, 0x0117 }, /* eabovedot ė LATIN SMALL LETTER E WITH DOT ABOVE */
+ { 0x03ef, 0x012b }, /* imacron ī LATIN SMALL LETTER I WITH MACRON */
+ { 0x03f1, 0x0146 }, /* ncedilla ņ LATIN SMALL LETTER N WITH CEDILLA */
+ { 0x03f2, 0x014d }, /* omacron ō LATIN SMALL LETTER O WITH MACRON */
+ { 0x03f3, 0x0137 }, /* kcedilla ķ LATIN SMALL LETTER K WITH CEDILLA */
+ { 0x03f9, 0x0173 }, /* uogonek ų LATIN SMALL LETTER U WITH OGONEK */
+ { 0x03fd, 0x0169 }, /* utilde ũ LATIN SMALL LETTER U WITH TILDE */
+ { 0x03fe, 0x016b }, /* umacron ū LATIN SMALL LETTER U WITH MACRON */
+ { 0x047e, 0x203e }, /* overline ‾ OVERLINE */
+ { 0x04a1, 0x3002 }, /* kana_fullstop 。 IDEOGRAPHIC FULL STOP */
+ { 0x04a2, 0x300c }, /* kana_openingbracket 「 LEFT CORNER BRACKET */
+ { 0x04a3, 0x300d }, /* kana_closingbracket 」 RIGHT CORNER BRACKET */
+ { 0x04a4, 0x3001 }, /* kana_comma 、 IDEOGRAPHIC COMMA */
+ { 0x04a5, 0x30fb }, /* kana_conjunctive ・ KATAKANA MIDDLE DOT */
+ { 0x04a6, 0x30f2 }, /* kana_WO ヲ KATAKANA LETTER WO */
+ { 0x04a7, 0x30a1 }, /* kana_a ァ KATAKANA LETTER SMALL A */
+ { 0x04a8, 0x30a3 }, /* kana_i ィ KATAKANA LETTER SMALL I */
+ { 0x04a9, 0x30a5 }, /* kana_u ゥ KATAKANA LETTER SMALL U */
+ { 0x04aa, 0x30a7 }, /* kana_e ェ KATAKANA LETTER SMALL E */
+ { 0x04ab, 0x30a9 }, /* kana_o ォ KATAKANA LETTER SMALL O */
+ { 0x04ac, 0x30e3 }, /* kana_ya ャ KATAKANA LETTER SMALL YA */
+ { 0x04ad, 0x30e5 }, /* kana_yu ュ KATAKANA LETTER SMALL YU */
+ { 0x04ae, 0x30e7 }, /* kana_yo ョ KATAKANA LETTER SMALL YO */
+ { 0x04af, 0x30c3 }, /* kana_tsu ッ KATAKANA LETTER SMALL TU */
+ { 0x04b0, 0x30fc }, /* prolongedsound ー KATAKANA-HIRAGANA PROLONGED SOUND MARK */
+ { 0x04b1, 0x30a2 }, /* kana_A ア KATAKANA LETTER A */
+ { 0x04b2, 0x30a4 }, /* kana_I イ KATAKANA LETTER I */
+ { 0x04b3, 0x30a6 }, /* kana_U ウ KATAKANA LETTER U */
+ { 0x04b4, 0x30a8 }, /* kana_E エ KATAKANA LETTER E */
+ { 0x04b5, 0x30aa }, /* kana_O オ KATAKANA LETTER O */
+ { 0x04b6, 0x30ab }, /* kana_KA カ KATAKANA LETTER KA */
+ { 0x04b7, 0x30ad }, /* kana_KI キ KATAKANA LETTER KI */
+ { 0x04b8, 0x30af }, /* kana_KU ク KATAKANA LETTER KU */
+ { 0x04b9, 0x30b1 }, /* kana_KE ケ KATAKANA LETTER KE */
+ { 0x04ba, 0x30b3 }, /* kana_KO コ KATAKANA LETTER KO */
+ { 0x04bb, 0x30b5 }, /* kana_SA サ KATAKANA LETTER SA */
+ { 0x04bc, 0x30b7 }, /* kana_SHI シ KATAKANA LETTER SI */
+ { 0x04bd, 0x30b9 }, /* kana_SU ス KATAKANA LETTER SU */
+ { 0x04be, 0x30bb }, /* kana_SE セ KATAKANA LETTER SE */
+ { 0x04bf, 0x30bd }, /* kana_SO ソ KATAKANA LETTER SO */
+ { 0x04c0, 0x30bf }, /* kana_TA タ KATAKANA LETTER TA */
+ { 0x04c1, 0x30c1 }, /* kana_CHI チ KATAKANA LETTER TI */
+ { 0x04c2, 0x30c4 }, /* kana_TSU ツ KATAKANA LETTER TU */
+ { 0x04c3, 0x30c6 }, /* kana_TE テ KATAKANA LETTER TE */
+ { 0x04c4, 0x30c8 }, /* kana_TO ト KATAKANA LETTER TO */
+ { 0x04c5, 0x30ca }, /* kana_NA ナ KATAKANA LETTER NA */
+ { 0x04c6, 0x30cb }, /* kana_NI ニ KATAKANA LETTER NI */
+ { 0x04c7, 0x30cc }, /* kana_NU ヌ KATAKANA LETTER NU */
+ { 0x04c8, 0x30cd }, /* kana_NE ネ KATAKANA LETTER NE */
+ { 0x04c9, 0x30ce }, /* kana_NO ノ KATAKANA LETTER NO */
+ { 0x04ca, 0x30cf }, /* kana_HA ハ KATAKANA LETTER HA */
+ { 0x04cb, 0x30d2 }, /* kana_HI ヒ KATAKANA LETTER HI */
+ { 0x04cc, 0x30d5 }, /* kana_FU フ KATAKANA LETTER HU */
+ { 0x04cd, 0x30d8 }, /* kana_HE ヘ KATAKANA LETTER HE */
+ { 0x04ce, 0x30db }, /* kana_HO ホ KATAKANA LETTER HO */
+ { 0x04cf, 0x30de }, /* kana_MA マ KATAKANA LETTER MA */
+ { 0x04d0, 0x30df }, /* kana_MI ミ KATAKANA LETTER MI */
+ { 0x04d1, 0x30e0 }, /* kana_MU ム KATAKANA LETTER MU */
+ { 0x04d2, 0x30e1 }, /* kana_ME メ KATAKANA LETTER ME */
+ { 0x04d3, 0x30e2 }, /* kana_MO モ KATAKANA LETTER MO */
+ { 0x04d4, 0x30e4 }, /* kana_YA ヤ KATAKANA LETTER YA */
+ { 0x04d5, 0x30e6 }, /* kana_YU ユ KATAKANA LETTER YU */
+ { 0x04d6, 0x30e8 }, /* kana_YO ヨ KATAKANA LETTER YO */
+ { 0x04d7, 0x30e9 }, /* kana_RA ラ KATAKANA LETTER RA */
+ { 0x04d8, 0x30ea }, /* kana_RI リ KATAKANA LETTER RI */
+ { 0x04d9, 0x30eb }, /* kana_RU ル KATAKANA LETTER RU */
+ { 0x04da, 0x30ec }, /* kana_RE レ KATAKANA LETTER RE */
+ { 0x04db, 0x30ed }, /* kana_RO ロ KATAKANA LETTER RO */
+ { 0x04dc, 0x30ef }, /* kana_WA ワ KATAKANA LETTER WA */
+ { 0x04dd, 0x30f3 }, /* kana_N ン KATAKANA LETTER N */
+ { 0x04de, 0x309b }, /* voicedsound ゛ KATAKANA-HIRAGANA VOICED SOUND MARK */
+ { 0x04df, 0x309c }, /* semivoicedsound ゜ KATAKANA-HIRAGANA SEMI-VOICED SOUND MARK */
+ { 0x05ac, 0x060c }, /* Arabic_comma ، ARABIC COMMA */
+ { 0x05bb, 0x061b }, /* Arabic_semicolon ؛ ARABIC SEMICOLON */
+ { 0x05bf, 0x061f }, /* Arabic_question_mark ؟ ARABIC QUESTION MARK */
+ { 0x05c1, 0x0621 }, /* Arabic_hamza ء ARABIC LETTER HAMZA */
+ { 0x05c2, 0x0622 }, /* Arabic_maddaonalef آ ARABIC LETTER ALEF WITH MADDA ABOVE */
+ { 0x05c3, 0x0623 }, /* Arabic_hamzaonalef أ ARABIC LETTER ALEF WITH HAMZA ABOVE */
+ { 0x05c4, 0x0624 }, /* Arabic_hamzaonwaw ؤ ARABIC LETTER WAW WITH HAMZA ABOVE */
+ { 0x05c5, 0x0625 }, /* Arabic_hamzaunderalef إ ARABIC LETTER ALEF WITH HAMZA BELOW */
+ { 0x05c6, 0x0626 }, /* Arabic_hamzaonyeh ئ ARABIC LETTER YEH WITH HAMZA ABOVE */
+ { 0x05c7, 0x0627 }, /* Arabic_alef ا ARABIC LETTER ALEF */
+ { 0x05c8, 0x0628 }, /* Arabic_beh ب ARABIC LETTER BEH */
+ { 0x05c9, 0x0629 }, /* Arabic_tehmarbuta ة ARABIC LETTER TEH MARBUTA */
+ { 0x05ca, 0x062a }, /* Arabic_teh ت ARABIC LETTER TEH */
+ { 0x05cb, 0x062b }, /* Arabic_theh ث ARABIC LETTER THEH */
+ { 0x05cc, 0x062c }, /* Arabic_jeem ج ARABIC LETTER JEEM */
+ { 0x05cd, 0x062d }, /* Arabic_hah ح ARABIC LETTER HAH */
+ { 0x05ce, 0x062e }, /* Arabic_khah خ ARABIC LETTER KHAH */
+ { 0x05cf, 0x062f }, /* Arabic_dal د ARABIC LETTER DAL */
+ { 0x05d0, 0x0630 }, /* Arabic_thal ذ ARABIC LETTER THAL */
+ { 0x05d1, 0x0631 }, /* Arabic_ra ر ARABIC LETTER REH */
+ { 0x05d2, 0x0632 }, /* Arabic_zain ز ARABIC LETTER ZAIN */
+ { 0x05d3, 0x0633 }, /* Arabic_seen س ARABIC LETTER SEEN */
+ { 0x05d4, 0x0634 }, /* Arabic_sheen ش ARABIC LETTER SHEEN */
+ { 0x05d5, 0x0635 }, /* Arabic_sad ص ARABIC LETTER SAD */
+ { 0x05d6, 0x0636 }, /* Arabic_dad ض ARABIC LETTER DAD */
+ { 0x05d7, 0x0637 }, /* Arabic_tah ط ARABIC LETTER TAH */
+ { 0x05d8, 0x0638 }, /* Arabic_zah ظ ARABIC LETTER ZAH */
+ { 0x05d9, 0x0639 }, /* Arabic_ain ع ARABIC LETTER AIN */
+ { 0x05da, 0x063a }, /* Arabic_ghain غ ARABIC LETTER GHAIN */
+ { 0x05e0, 0x0640 }, /* Arabic_tatweel ـ ARABIC TATWEEL */
+ { 0x05e1, 0x0641 }, /* Arabic_feh ف ARABIC LETTER FEH */
+ { 0x05e2, 0x0642 }, /* Arabic_qaf ق ARABIC LETTER QAF */
+ { 0x05e3, 0x0643 }, /* Arabic_kaf ك ARABIC LETTER KAF */
+ { 0x05e4, 0x0644 }, /* Arabic_lam ل ARABIC LETTER LAM */
+ { 0x05e5, 0x0645 }, /* Arabic_meem م ARABIC LETTER MEEM */
+ { 0x05e6, 0x0646 }, /* Arabic_noon ن ARABIC LETTER NOON */
+ { 0x05e7, 0x0647 }, /* Arabic_ha ه ARABIC LETTER HEH */
+ { 0x05e8, 0x0648 }, /* Arabic_waw و ARABIC LETTER WAW */
+ { 0x05e9, 0x0649 }, /* Arabic_alefmaksura ى ARABIC LETTER ALEF MAKSURA */
+ { 0x05ea, 0x064a }, /* Arabic_yeh ي ARABIC LETTER YEH */
+ { 0x05eb, 0x064b }, /* Arabic_fathatan ً ARABIC FATHATAN */
+ { 0x05ec, 0x064c }, /* Arabic_dammatan ٌ ARABIC DAMMATAN */
+ { 0x05ed, 0x064d }, /* Arabic_kasratan ٍ ARABIC KASRATAN */
+ { 0x05ee, 0x064e }, /* Arabic_fatha َ ARABIC FATHA */
+ { 0x05ef, 0x064f }, /* Arabic_damma ُ ARABIC DAMMA */
+ { 0x05f0, 0x0650 }, /* Arabic_kasra ِ ARABIC KASRA */
+ { 0x05f1, 0x0651 }, /* Arabic_shadda ّ ARABIC SHADDA */
+ { 0x05f2, 0x0652 }, /* Arabic_sukun ْ ARABIC SUKUN */
+ { 0x06a1, 0x0452 }, /* Serbian_dje ђ CYRILLIC SMALL LETTER DJE */
+ { 0x06a2, 0x0453 }, /* Macedonia_gje ѓ CYRILLIC SMALL LETTER GJE */
+ { 0x06a3, 0x0451 }, /* Cyrillic_io ё CYRILLIC SMALL LETTER IO */
+ { 0x06a4, 0x0454 }, /* Ukrainian_ie є CYRILLIC SMALL LETTER UKRAINIAN IE */
+ { 0x06a5, 0x0455 }, /* Macedonia_dse ѕ CYRILLIC SMALL LETTER DZE */
+ { 0x06a6, 0x0456 }, /* Ukrainian_i і CYRILLIC SMALL LETTER BYELORUSSIAN-UKRAINIAN I */
+ { 0x06a7, 0x0457 }, /* Ukrainian_yi ї CYRILLIC SMALL LETTER YI */
+ { 0x06a8, 0x0458 }, /* Cyrillic_je ј CYRILLIC SMALL LETTER JE */
+ { 0x06a9, 0x0459 }, /* Cyrillic_lje љ CYRILLIC SMALL LETTER LJE */
+ { 0x06aa, 0x045a }, /* Cyrillic_nje њ CYRILLIC SMALL LETTER NJE */
+ { 0x06ab, 0x045b }, /* Serbian_tshe ћ CYRILLIC SMALL LETTER TSHE */
+ { 0x06ac, 0x045c }, /* Macedonia_kje ќ CYRILLIC SMALL LETTER KJE */
+ { 0x06ad, 0x0491 }, /* Ukrainian_ghe_with_upturn ґ CYRILLIC SMALL LETTER GHE WITH UPTURN */
+ { 0x06ae, 0x045e }, /* Byelorussian_shortu ў CYRILLIC SMALL LETTER SHORT U */
+ { 0x06af, 0x045f }, /* Cyrillic_dzhe џ CYRILLIC SMALL LETTER DZHE */
+ { 0x06b0, 0x2116 }, /* numerosign № NUMERO SIGN */
+ { 0x06b1, 0x0402 }, /* Serbian_DJE Ђ CYRILLIC CAPITAL LETTER DJE */
+ { 0x06b2, 0x0403 }, /* Macedonia_GJE Ѓ CYRILLIC CAPITAL LETTER GJE */
+ { 0x06b3, 0x0401 }, /* Cyrillic_IO Ё CYRILLIC CAPITAL LETTER IO */
+ { 0x06b4, 0x0404 }, /* Ukrainian_IE Є CYRILLIC CAPITAL LETTER UKRAINIAN IE */
+ { 0x06b5, 0x0405 }, /* Macedonia_DSE Ѕ CYRILLIC CAPITAL LETTER DZE */
+ { 0x06b6, 0x0406 }, /* Ukrainian_I І CYRILLIC CAPITAL LETTER BYELORUSSIAN-UKRAINIAN I */
+ { 0x06b7, 0x0407 }, /* Ukrainian_YI Ї CYRILLIC CAPITAL LETTER YI */
+ { 0x06b8, 0x0408 }, /* Cyrillic_JE Ј CYRILLIC CAPITAL LETTER JE */
+ { 0x06b9, 0x0409 }, /* Cyrillic_LJE Љ CYRILLIC CAPITAL LETTER LJE */
+ { 0x06ba, 0x040a }, /* Cyrillic_NJE Њ CYRILLIC CAPITAL LETTER NJE */
+ { 0x06bb, 0x040b }, /* Serbian_TSHE Ћ CYRILLIC CAPITAL LETTER TSHE */
+ { 0x06bc, 0x040c }, /* Macedonia_KJE Ќ CYRILLIC CAPITAL LETTER KJE */
+ { 0x06bd, 0x0490 }, /* Ukrainian_GHE_WITH_UPTURN Ґ CYRILLIC CAPITAL LETTER GHE WITH UPTURN */
+ { 0x06be, 0x040e }, /* Byelorussian_SHORTU Ў CYRILLIC CAPITAL LETTER SHORT U */
+ { 0x06bf, 0x040f }, /* Cyrillic_DZHE Џ CYRILLIC CAPITAL LETTER DZHE */
+ { 0x06c0, 0x044e }, /* Cyrillic_yu ю CYRILLIC SMALL LETTER YU */
+ { 0x06c1, 0x0430 }, /* Cyrillic_a а CYRILLIC SMALL LETTER A */
+ { 0x06c2, 0x0431 }, /* Cyrillic_be б CYRILLIC SMALL LETTER BE */
+ { 0x06c3, 0x0446 }, /* Cyrillic_tse ц CYRILLIC SMALL LETTER TSE */
+ { 0x06c4, 0x0434 }, /* Cyrillic_de д CYRILLIC SMALL LETTER DE */
+ { 0x06c5, 0x0435 }, /* Cyrillic_ie е CYRILLIC SMALL LETTER IE */
+ { 0x06c6, 0x0444 }, /* Cyrillic_ef ф CYRILLIC SMALL LETTER EF */
+ { 0x06c7, 0x0433 }, /* Cyrillic_ghe г CYRILLIC SMALL LETTER GHE */
+ { 0x06c8, 0x0445 }, /* Cyrillic_ha х CYRILLIC SMALL LETTER HA */
+ { 0x06c9, 0x0438 }, /* Cyrillic_i и CYRILLIC SMALL LETTER I */
+ { 0x06ca, 0x0439 }, /* Cyrillic_shorti й CYRILLIC SMALL LETTER SHORT I */
+ { 0x06cb, 0x043a }, /* Cyrillic_ka к CYRILLIC SMALL LETTER KA */
+ { 0x06cc, 0x043b }, /* Cyrillic_el л CYRILLIC SMALL LETTER EL */
+ { 0x06cd, 0x043c }, /* Cyrillic_em м CYRILLIC SMALL LETTER EM */
+ { 0x06ce, 0x043d }, /* Cyrillic_en н CYRILLIC SMALL LETTER EN */
+ { 0x06cf, 0x043e }, /* Cyrillic_o о CYRILLIC SMALL LETTER O */
+ { 0x06d0, 0x043f }, /* Cyrillic_pe п CYRILLIC SMALL LETTER PE */
+ { 0x06d1, 0x044f }, /* Cyrillic_ya я CYRILLIC SMALL LETTER YA */
+ { 0x06d2, 0x0440 }, /* Cyrillic_er р CYRILLIC SMALL LETTER ER */
+ { 0x06d3, 0x0441 }, /* Cyrillic_es с CYRILLIC SMALL LETTER ES */
+ { 0x06d4, 0x0442 }, /* Cyrillic_te т CYRILLIC SMALL LETTER TE */
+ { 0x06d5, 0x0443 }, /* Cyrillic_u у CYRILLIC SMALL LETTER U */
+ { 0x06d6, 0x0436 }, /* Cyrillic_zhe ж CYRILLIC SMALL LETTER ZHE */
+ { 0x06d7, 0x0432 }, /* Cyrillic_ve в CYRILLIC SMALL LETTER VE */
+ { 0x06d8, 0x044c }, /* Cyrillic_softsign ь CYRILLIC SMALL LETTER SOFT SIGN */
+ { 0x06d9, 0x044b }, /* Cyrillic_yeru ы CYRILLIC SMALL LETTER YERU */
+ { 0x06da, 0x0437 }, /* Cyrillic_ze з CYRILLIC SMALL LETTER ZE */
+ { 0x06db, 0x0448 }, /* Cyrillic_sha ш CYRILLIC SMALL LETTER SHA */
+ { 0x06dc, 0x044d }, /* Cyrillic_e э CYRILLIC SMALL LETTER E */
+ { 0x06dd, 0x0449 }, /* Cyrillic_shcha щ CYRILLIC SMALL LETTER SHCHA */
+ { 0x06de, 0x0447 }, /* Cyrillic_che ч CYRILLIC SMALL LETTER CHE */
+ { 0x06df, 0x044a }, /* Cyrillic_hardsign ъ CYRILLIC SMALL LETTER HARD SIGN */
+ { 0x06e0, 0x042e }, /* Cyrillic_YU Ю CYRILLIC CAPITAL LETTER YU */
+ { 0x06e1, 0x0410 }, /* Cyrillic_A А CYRILLIC CAPITAL LETTER A */
+ { 0x06e2, 0x0411 }, /* Cyrillic_BE Б CYRILLIC CAPITAL LETTER BE */
+ { 0x06e3, 0x0426 }, /* Cyrillic_TSE Ц CYRILLIC CAPITAL LETTER TSE */
+ { 0x06e4, 0x0414 }, /* Cyrillic_DE Д CYRILLIC CAPITAL LETTER DE */
+ { 0x06e5, 0x0415 }, /* Cyrillic_IE Е CYRILLIC CAPITAL LETTER IE */
+ { 0x06e6, 0x0424 }, /* Cyrillic_EF Ф CYRILLIC CAPITAL LETTER EF */
+ { 0x06e7, 0x0413 }, /* Cyrillic_GHE Г CYRILLIC CAPITAL LETTER GHE */
+ { 0x06e8, 0x0425 }, /* Cyrillic_HA Х CYRILLIC CAPITAL LETTER HA */
+ { 0x06e9, 0x0418 }, /* Cyrillic_I И CYRILLIC CAPITAL LETTER I */
+ { 0x06ea, 0x0419 }, /* Cyrillic_SHORTI Й CYRILLIC CAPITAL LETTER SHORT I */
+ { 0x06eb, 0x041a }, /* Cyrillic_KA К CYRILLIC CAPITAL LETTER KA */
+ { 0x06ec, 0x041b }, /* Cyrillic_EL Л CYRILLIC CAPITAL LETTER EL */
+ { 0x06ed, 0x041c }, /* Cyrillic_EM М CYRILLIC CAPITAL LETTER EM */
+ { 0x06ee, 0x041d }, /* Cyrillic_EN Н CYRILLIC CAPITAL LETTER EN */
+ { 0x06ef, 0x041e }, /* Cyrillic_O О CYRILLIC CAPITAL LETTER O */
+ { 0x06f0, 0x041f }, /* Cyrillic_PE П CYRILLIC CAPITAL LETTER PE */
+ { 0x06f1, 0x042f }, /* Cyrillic_YA Я CYRILLIC CAPITAL LETTER YA */
+ { 0x06f2, 0x0420 }, /* Cyrillic_ER Р CYRILLIC CAPITAL LETTER ER */
+ { 0x06f3, 0x0421 }, /* Cyrillic_ES С CYRILLIC CAPITAL LETTER ES */
+ { 0x06f4, 0x0422 }, /* Cyrillic_TE Т CYRILLIC CAPITAL LETTER TE */
+ { 0x06f5, 0x0423 }, /* Cyrillic_U У CYRILLIC CAPITAL LETTER U */
+ { 0x06f6, 0x0416 }, /* Cyrillic_ZHE Ж CYRILLIC CAPITAL LETTER ZHE */
+ { 0x06f7, 0x0412 }, /* Cyrillic_VE В CYRILLIC CAPITAL LETTER VE */
+ { 0x06f8, 0x042c }, /* Cyrillic_SOFTSIGN Ь CYRILLIC CAPITAL LETTER SOFT SIGN */
+ { 0x06f9, 0x042b }, /* Cyrillic_YERU Ы CYRILLIC CAPITAL LETTER YERU */
+ { 0x06fa, 0x0417 }, /* Cyrillic_ZE З CYRILLIC CAPITAL LETTER ZE */
+ { 0x06fb, 0x0428 }, /* Cyrillic_SHA Ш CYRILLIC CAPITAL LETTER SHA */
+ { 0x06fc, 0x042d }, /* Cyrillic_E Э CYRILLIC CAPITAL LETTER E */
+ { 0x06fd, 0x0429 }, /* Cyrillic_SHCHA Щ CYRILLIC CAPITAL LETTER SHCHA */
+ { 0x06fe, 0x0427 }, /* Cyrillic_CHE Ч CYRILLIC CAPITAL LETTER CHE */
+ { 0x06ff, 0x042a }, /* Cyrillic_HARDSIGN Ъ CYRILLIC CAPITAL LETTER HARD SIGN */
+ { 0x07a1, 0x0386 }, /* Greek_ALPHAaccent Ά GREEK CAPITAL LETTER ALPHA WITH TONOS */
+ { 0x07a2, 0x0388 }, /* Greek_EPSILONaccent Έ GREEK CAPITAL LETTER EPSILON WITH TONOS */
+ { 0x07a3, 0x0389 }, /* Greek_ETAaccent Ή GREEK CAPITAL LETTER ETA WITH TONOS */
+ { 0x07a4, 0x038a }, /* Greek_IOTAaccent Ί GREEK CAPITAL LETTER IOTA WITH TONOS */
+ { 0x07a5, 0x03aa }, /* Greek_IOTAdiaeresis Ϊ GREEK CAPITAL LETTER IOTA WITH DIALYTIKA */
+ { 0x07a7, 0x038c }, /* Greek_OMICRONaccent Ό GREEK CAPITAL LETTER OMICRON WITH TONOS */
+ { 0x07a8, 0x038e }, /* Greek_UPSILONaccent Ύ GREEK CAPITAL LETTER UPSILON WITH TONOS */
+ { 0x07a9, 0x03ab }, /* Greek_UPSILONdieresis Ϋ GREEK CAPITAL LETTER UPSILON WITH DIALYTIKA */
+ { 0x07ab, 0x038f }, /* Greek_OMEGAaccent Ώ GREEK CAPITAL LETTER OMEGA WITH TONOS */
+ { 0x07ae, 0x0385 }, /* Greek_accentdieresis ΅ GREEK DIALYTIKA TONOS */
+ { 0x07af, 0x2015 }, /* Greek_horizbar ― HORIZONTAL BAR */
+ { 0x07b1, 0x03ac }, /* Greek_alphaaccent ά GREEK SMALL LETTER ALPHA WITH TONOS */
+ { 0x07b2, 0x03ad }, /* Greek_epsilonaccent έ GREEK SMALL LETTER EPSILON WITH TONOS */
+ { 0x07b3, 0x03ae }, /* Greek_etaaccent ή GREEK SMALL LETTER ETA WITH TONOS */
+ { 0x07b4, 0x03af }, /* Greek_iotaaccent ί GREEK SMALL LETTER IOTA WITH TONOS */
+ { 0x07b5, 0x03ca }, /* Greek_iotadieresis ϊ GREEK SMALL LETTER IOTA WITH DIALYTIKA */
+ { 0x07b6, 0x0390 }, /* Greek_iotaaccentdieresis ΐ GREEK SMALL LETTER IOTA WITH DIALYTIKA AND TONOS */
+ { 0x07b7, 0x03cc }, /* Greek_omicronaccent ό GREEK SMALL LETTER OMICRON WITH TONOS */
+ { 0x07b8, 0x03cd }, /* Greek_upsilonaccent ύ GREEK SMALL LETTER UPSILON WITH TONOS */
+ { 0x07b9, 0x03cb }, /* Greek_upsilondieresis ϋ GREEK SMALL LETTER UPSILON WITH DIALYTIKA */
+ { 0x07ba, 0x03b0 }, /* Greek_upsilonaccentdieresis ΰ GREEK SMALL LETTER UPSILON WITH DIALYTIKA AND TONOS */
+ { 0x07bb, 0x03ce }, /* Greek_omegaaccent ώ GREEK SMALL LETTER OMEGA WITH TONOS */
+ { 0x07c1, 0x0391 }, /* Greek_ALPHA Α GREEK CAPITAL LETTER ALPHA */
+ { 0x07c2, 0x0392 }, /* Greek_BETA Β GREEK CAPITAL LETTER BETA */
+ { 0x07c3, 0x0393 }, /* Greek_GAMMA Γ GREEK CAPITAL LETTER GAMMA */
+ { 0x07c4, 0x0394 }, /* Greek_DELTA Δ GREEK CAPITAL LETTER DELTA */
+ { 0x07c5, 0x0395 }, /* Greek_EPSILON Ε GREEK CAPITAL LETTER EPSILON */
+ { 0x07c6, 0x0396 }, /* Greek_ZETA Ζ GREEK CAPITAL LETTER ZETA */
+ { 0x07c7, 0x0397 }, /* Greek_ETA Η GREEK CAPITAL LETTER ETA */
+ { 0x07c8, 0x0398 }, /* Greek_THETA Θ GREEK CAPITAL LETTER THETA */
+ { 0x07c9, 0x0399 }, /* Greek_IOTA Ι GREEK CAPITAL LETTER IOTA */
+ { 0x07ca, 0x039a }, /* Greek_KAPPA Κ GREEK CAPITAL LETTER KAPPA */
+ { 0x07cb, 0x039b }, /* Greek_LAMBDA Λ GREEK CAPITAL LETTER LAMDA */
+ { 0x07cc, 0x039c }, /* Greek_MU Μ GREEK CAPITAL LETTER MU */
+ { 0x07cd, 0x039d }, /* Greek_NU Ν GREEK CAPITAL LETTER NU */
+ { 0x07ce, 0x039e }, /* Greek_XI Ξ GREEK CAPITAL LETTER XI */
+ { 0x07cf, 0x039f }, /* Greek_OMICRON Ο GREEK CAPITAL LETTER OMICRON */
+ { 0x07d0, 0x03a0 }, /* Greek_PI Π GREEK CAPITAL LETTER PI */
+ { 0x07d1, 0x03a1 }, /* Greek_RHO Ρ GREEK CAPITAL LETTER RHO */
+ { 0x07d2, 0x03a3 }, /* Greek_SIGMA Σ GREEK CAPITAL LETTER SIGMA */
+ { 0x07d4, 0x03a4 }, /* Greek_TAU Τ GREEK CAPITAL LETTER TAU */
+ { 0x07d5, 0x03a5 }, /* Greek_UPSILON Υ GREEK CAPITAL LETTER UPSILON */
+ { 0x07d6, 0x03a6 }, /* Greek_PHI Φ GREEK CAPITAL LETTER PHI */
+ { 0x07d7, 0x03a7 }, /* Greek_CHI Χ GREEK CAPITAL LETTER CHI */
+ { 0x07d8, 0x03a8 }, /* Greek_PSI Ψ GREEK CAPITAL LETTER PSI */
+ { 0x07d9, 0x03a9 }, /* Greek_OMEGA Ω GREEK CAPITAL LETTER OMEGA */
+ { 0x07e1, 0x03b1 }, /* Greek_alpha α GREEK SMALL LETTER ALPHA */
+ { 0x07e2, 0x03b2 }, /* Greek_beta β GREEK SMALL LETTER BETA */
+ { 0x07e3, 0x03b3 }, /* Greek_gamma γ GREEK SMALL LETTER GAMMA */
+ { 0x07e4, 0x03b4 }, /* Greek_delta δ GREEK SMALL LETTER DELTA */
+ { 0x07e5, 0x03b5 }, /* Greek_epsilon ε GREEK SMALL LETTER EPSILON */
+ { 0x07e6, 0x03b6 }, /* Greek_zeta ζ GREEK SMALL LETTER ZETA */
+ { 0x07e7, 0x03b7 }, /* Greek_eta η GREEK SMALL LETTER ETA */
+ { 0x07e8, 0x03b8 }, /* Greek_theta θ GREEK SMALL LETTER THETA */
+ { 0x07e9, 0x03b9 }, /* Greek_iota ι GREEK SMALL LETTER IOTA */
+ { 0x07ea, 0x03ba }, /* Greek_kappa κ GREEK SMALL LETTER KAPPA */
+ { 0x07eb, 0x03bb }, /* Greek_lambda λ GREEK SMALL LETTER LAMDA */
+ { 0x07ec, 0x03bc }, /* Greek_mu μ GREEK SMALL LETTER MU */
+ { 0x07ed, 0x03bd }, /* Greek_nu ν GREEK SMALL LETTER NU */
+ { 0x07ee, 0x03be }, /* Greek_xi ξ GREEK SMALL LETTER XI */
+ { 0x07ef, 0x03bf }, /* Greek_omicron ο GREEK SMALL LETTER OMICRON */
+ { 0x07f0, 0x03c0 }, /* Greek_pi π GREEK SMALL LETTER PI */
+ { 0x07f1, 0x03c1 }, /* Greek_rho ρ GREEK SMALL LETTER RHO */
+ { 0x07f2, 0x03c3 }, /* Greek_sigma σ GREEK SMALL LETTER SIGMA */
+ { 0x07f3, 0x03c2 }, /* Greek_finalsmallsigma ς GREEK SMALL LETTER FINAL SIGMA */
+ { 0x07f4, 0x03c4 }, /* Greek_tau τ GREEK SMALL LETTER TAU */
+ { 0x07f5, 0x03c5 }, /* Greek_upsilon υ GREEK SMALL LETTER UPSILON */
+ { 0x07f6, 0x03c6 }, /* Greek_phi φ GREEK SMALL LETTER PHI */
+ { 0x07f7, 0x03c7 }, /* Greek_chi χ GREEK SMALL LETTER CHI */
+ { 0x07f8, 0x03c8 }, /* Greek_psi ψ GREEK SMALL LETTER PSI */
+ { 0x07f9, 0x03c9 }, /* Greek_omega ω GREEK SMALL LETTER OMEGA */
+ { 0x08a1, 0x23b7 }, /* leftradical ⎷ ??? */
+ { 0x08a2, 0x250c }, /* topleftradical ┌ BOX DRAWINGS LIGHT DOWN AND RIGHT */
+ { 0x08a3, 0x2500 }, /* horizconnector ─ BOX DRAWINGS LIGHT HORIZONTAL */
+ { 0x08a4, 0x2320 }, /* topintegral ⌠ TOP HALF INTEGRAL */
+ { 0x08a5, 0x2321 }, /* botintegral ⌡ BOTTOM HALF INTEGRAL */
+ { 0x08a6, 0x2502 }, /* vertconnector │ BOX DRAWINGS LIGHT VERTICAL */
+ { 0x08a7, 0x23a1 }, /* topleftsqbracket ⎡ ??? */
+ { 0x08a8, 0x23a3 }, /* botleftsqbracket ⎣ ??? */
+ { 0x08a9, 0x23a4 }, /* toprightsqbracket ⎤ ??? */
+ { 0x08aa, 0x23a6 }, /* botrightsqbracket ⎦ ??? */
+ { 0x08ab, 0x239b }, /* topleftparens ⎛ ??? */
+ { 0x08ac, 0x239d }, /* botleftparens ⎝ ??? */
+ { 0x08ad, 0x239e }, /* toprightparens ⎞ ??? */
+ { 0x08ae, 0x23a0 }, /* botrightparens ⎠ ??? */
+ { 0x08af, 0x23a8 }, /* leftmiddlecurlybrace ⎨ ??? */
+ { 0x08b0, 0x23ac }, /* rightmiddlecurlybrace ⎬ ??? */
+ /* 0x08b1 topleftsummation ? ??? */
+ /* 0x08b2 botleftsummation ? ??? */
+ /* 0x08b3 topvertsummationconnector ? ??? */
+ /* 0x08b4 botvertsummationconnector ? ??? */
+ /* 0x08b5 toprightsummation ? ??? */
+ /* 0x08b6 botrightsummation ? ??? */
+ /* 0x08b7 rightmiddlesummation ? ??? */
+ { 0x08bc, 0x2264 }, /* lessthanequal ≤ LESS-THAN OR EQUAL TO */
+ { 0x08bd, 0x2260 }, /* notequal ≠ NOT EQUAL TO */
+ { 0x08be, 0x2265 }, /* greaterthanequal ≥ GREATER-THAN OR EQUAL TO */
+ { 0x08bf, 0x222b }, /* integral ∫ INTEGRAL */
+ { 0x08c0, 0x2234 }, /* therefore ∴ THEREFORE */
+ { 0x08c1, 0x221d }, /* variation ∝ PROPORTIONAL TO */
+ { 0x08c2, 0x221e }, /* infinity ∞ INFINITY */
+ { 0x08c5, 0x2207 }, /* nabla ∇ NABLA */
+ { 0x08c8, 0x223c }, /* approximate ∼ TILDE OPERATOR */
+ { 0x08c9, 0x2243 }, /* similarequal ≃ ASYMPTOTICALLY EQUAL TO */
+ { 0x08cd, 0x21d4 }, /* ifonlyif ⇔ LEFT RIGHT DOUBLE ARROW */
+ { 0x08ce, 0x21d2 }, /* implies ⇒ RIGHTWARDS DOUBLE ARROW */
+ { 0x08cf, 0x2261 }, /* identical ≡ IDENTICAL TO */
+ { 0x08d6, 0x221a }, /* radical √ SQUARE ROOT */
+ { 0x08da, 0x2282 }, /* includedin ⊂ SUBSET OF */
+ { 0x08db, 0x2283 }, /* includes ⊃ SUPERSET OF */
+ { 0x08dc, 0x2229 }, /* intersection ∩ INTERSECTION */
+ { 0x08dd, 0x222a }, /* union ∪ UNION */
+ { 0x08de, 0x2227 }, /* logicaland ∧ LOGICAL AND */
+ { 0x08df, 0x2228 }, /* logicalor ∨ LOGICAL OR */
+ { 0x08ef, 0x2202 }, /* partialderivative ∂ PARTIAL DIFFERENTIAL */
+ { 0x08f6, 0x0192 }, /* function ƒ LATIN SMALL LETTER F WITH HOOK */
+ { 0x08fb, 0x2190 }, /* leftarrow ← LEFTWARDS ARROW */
+ { 0x08fc, 0x2191 }, /* uparrow ↑ UPWARDS ARROW */
+ { 0x08fd, 0x2192 }, /* rightarrow → RIGHTWARDS ARROW */
+ { 0x08fe, 0x2193 }, /* downarrow ↓ DOWNWARDS ARROW */
+/* 0x09df blank ? ??? */
+ { 0x09e0, 0x25c6 }, /* soliddiamond ◆ BLACK DIAMOND */
+ { 0x09e1, 0x2592 }, /* checkerboard ▒ MEDIUM SHADE */
+ { 0x09e2, 0x2409 }, /* ht ␉ SYMBOL FOR HORIZONTAL TABULATION */
+ { 0x09e3, 0x240c }, /* ff ␌ SYMBOL FOR FORM FEED */
+ { 0x09e4, 0x240d }, /* cr ␍ SYMBOL FOR CARRIAGE RETURN */
+ { 0x09e5, 0x240a }, /* lf ␊ SYMBOL FOR LINE FEED */
+ { 0x09e8, 0x2424 }, /* nl  SYMBOL FOR NEWLINE */
+ { 0x09e9, 0x240b }, /* vt ␋ SYMBOL FOR VERTICAL TABULATION */
+ { 0x09ea, 0x2518 }, /* lowrightcorner ┘ BOX DRAWINGS LIGHT UP AND LEFT */
+ { 0x09eb, 0x2510 }, /* uprightcorner ┐ BOX DRAWINGS LIGHT DOWN AND LEFT */
+ { 0x09ec, 0x250c }, /* upleftcorner ┌ BOX DRAWINGS LIGHT DOWN AND RIGHT */
+ { 0x09ed, 0x2514 }, /* lowleftcorner └ BOX DRAWINGS LIGHT UP AND RIGHT */
+ { 0x09ee, 0x253c }, /* crossinglines ┼ BOX DRAWINGS LIGHT VERTICAL AND HORIZONTAL */
+ { 0x09ef, 0x23ba }, /* horizlinescan1 ⎺ HORIZONTAL SCAN LINE-1 (Unicode 3.2 draft) */
+ { 0x09f0, 0x23bb }, /* horizlinescan3 ⎻ HORIZONTAL SCAN LINE-3 (Unicode 3.2 draft) */
+ { 0x09f1, 0x2500 }, /* horizlinescan5 ─ BOX DRAWINGS LIGHT HORIZONTAL */
+ { 0x09f2, 0x23bc }, /* horizlinescan7 ⎼ HORIZONTAL SCAN LINE-7 (Unicode 3.2 draft) */
+ { 0x09f3, 0x23bd }, /* horizlinescan9 ⎽ HORIZONTAL SCAN LINE-9 (Unicode 3.2 draft) */
+ { 0x09f4, 0x251c }, /* leftt ├ BOX DRAWINGS LIGHT VERTICAL AND RIGHT */
+ { 0x09f5, 0x2524 }, /* rightt ┤ BOX DRAWINGS LIGHT VERTICAL AND LEFT */
+ { 0x09f6, 0x2534 }, /* bott ┴ BOX DRAWINGS LIGHT UP AND HORIZONTAL */
+ { 0x09f7, 0x252c }, /* topt ┬ BOX DRAWINGS LIGHT DOWN AND HORIZONTAL */
+ { 0x09f8, 0x2502 }, /* vertbar │ BOX DRAWINGS LIGHT VERTICAL */
+ { 0x0aa1, 0x2003 }, /* emspace EM SPACE */
+ { 0x0aa2, 0x2002 }, /* enspace EN SPACE */
+ { 0x0aa3, 0x2004 }, /* em3space THREE-PER-EM SPACE */
+ { 0x0aa4, 0x2005 }, /* em4space FOUR-PER-EM SPACE */
+ { 0x0aa5, 0x2007 }, /* digitspace FIGURE SPACE */
+ { 0x0aa6, 0x2008 }, /* punctspace PUNCTUATION SPACE */
+ { 0x0aa7, 0x2009 }, /* thinspace THIN SPACE */
+ { 0x0aa8, 0x200a }, /* hairspace HAIR SPACE */
+ { 0x0aa9, 0x2014 }, /* emdash — EM DASH */
+ { 0x0aaa, 0x2013 }, /* endash – EN DASH */
+ { 0x0aac, 0x2423 }, /* signifblank ␣ OPEN BOX */
+ { 0x0aae, 0x2026 }, /* ellipsis … HORIZONTAL ELLIPSIS */
+ { 0x0aaf, 0x2025 }, /* doubbaselinedot ‥ TWO DOT LEADER */
+ { 0x0ab0, 0x2153 }, /* onethird ⅓ VULGAR FRACTION ONE THIRD */
+ { 0x0ab1, 0x2154 }, /* twothirds ⅔ VULGAR FRACTION TWO THIRDS */
+ { 0x0ab2, 0x2155 }, /* onefifth ⅕ VULGAR FRACTION ONE FIFTH */
+ { 0x0ab3, 0x2156 }, /* twofifths ⅖ VULGAR FRACTION TWO FIFTHS */
+ { 0x0ab4, 0x2157 }, /* threefifths ⅗ VULGAR FRACTION THREE FIFTHS */
+ { 0x0ab5, 0x2158 }, /* fourfifths ⅘ VULGAR FRACTION FOUR FIFTHS */
+ { 0x0ab6, 0x2159 }, /* onesixth ⅙ VULGAR FRACTION ONE SIXTH */
+ { 0x0ab7, 0x215a }, /* fivesixths ⅚ VULGAR FRACTION FIVE SIXTHS */
+ { 0x0ab8, 0x2105 }, /* careof ℅ CARE OF */
+ { 0x0abb, 0x2012 }, /* figdash ‒ FIGURE DASH */
+ { 0x0abc, 0x27e8 }, /* leftanglebracket ⟨ MATHEMATICAL LEFT ANGLE BRACKET */
+ { 0x0abd, 0x002e }, /* decimalpoint . FULL STOP */
+ { 0x0abe, 0x27e9 }, /* rightanglebracket ⟩ MATHEMATICAL RIGHT ANGLE BRACKET */
+ /* 0x0abf marker ? ??? */
+ { 0x0ac3, 0x215b }, /* oneeighth ⅛ VULGAR FRACTION ONE EIGHTH */
+ { 0x0ac4, 0x215c }, /* threeeighths ⅜ VULGAR FRACTION THREE EIGHTHS */
+ { 0x0ac5, 0x215d }, /* fiveeighths ⅝ VULGAR FRACTION FIVE EIGHTHS */
+ { 0x0ac6, 0x215e }, /* seveneighths ⅞ VULGAR FRACTION SEVEN EIGHTHS */
+ { 0x0ac9, 0x2122 }, /* trademark ™ TRADE MARK SIGN */
+ { 0x0aca, 0x2613 }, /* signaturemark ☓ SALTIRE */
+ /* 0x0acb trademarkincircle ? ??? */
+ { 0x0acc, 0x25c1 }, /* leftopentriangle ◁ WHITE LEFT-POINTING TRIANGLE */
+ { 0x0acd, 0x25b7 }, /* rightopentriangle ▷ WHITE RIGHT-POINTING TRIANGLE */
+ { 0x0ace, 0x25cb }, /* emopencircle ○ WHITE CIRCLE */
+ { 0x0acf, 0x25af }, /* emopenrectangle ▯ WHITE VERTICAL RECTANGLE */
+ { 0x0ad0, 0x2018 }, /* leftsinglequotemark ‘ LEFT SINGLE QUOTATION MARK */
+ { 0x0ad1, 0x2019 }, /* rightsinglequotemark ’ RIGHT SINGLE QUOTATION MARK */
+ { 0x0ad2, 0x201c }, /* leftdoublequotemark “ LEFT DOUBLE QUOTATION MARK */
+ { 0x0ad3, 0x201d }, /* rightdoublequotemark ” RIGHT DOUBLE QUOTATION MARK */
+ { 0x0ad4, 0x211e }, /* prescription ℞ PRESCRIPTION TAKE */
+ { 0x0ad5, 0x2030 }, /* permille ‰ PER MILLE SIGN */
+ { 0x0ad6, 0x2032 }, /* minutes ′ PRIME */
+ { 0x0ad7, 0x2033 }, /* seconds ″ DOUBLE PRIME */
+ { 0x0ad9, 0x271d }, /* latincross ✝ LATIN CROSS */
+ /* 0x0ada hexagram ? ??? */
+ { 0x0adb, 0x25ac }, /* filledrectbullet ▬ BLACK RECTANGLE */
+ { 0x0adc, 0x25c0 }, /* filledlefttribullet ◀ BLACK LEFT-POINTING TRIANGLE */
+ { 0x0add, 0x25b6 }, /* filledrighttribullet ▶ BLACK RIGHT-POINTING TRIANGLE */
+ { 0x0ade, 0x25cf }, /* emfilledcircle ● BLACK CIRCLE */
+ { 0x0adf, 0x25ae }, /* emfilledrect ▮ BLACK VERTICAL RECTANGLE */
+ { 0x0ae0, 0x25e6 }, /* enopencircbullet ◦ WHITE BULLET */
+ { 0x0ae1, 0x25ab }, /* enopensquarebullet ▫ WHITE SMALL SQUARE */
+ { 0x0ae2, 0x25ad }, /* openrectbullet ▭ WHITE RECTANGLE */
+ { 0x0ae3, 0x25b3 }, /* opentribulletup △ WHITE UP-POINTING TRIANGLE */
+ { 0x0ae4, 0x25bd }, /* opentribulletdown ▽ WHITE DOWN-POINTING TRIANGLE */
+ { 0x0ae5, 0x2606 }, /* openstar ☆ WHITE STAR */
+ { 0x0ae6, 0x2022 }, /* enfilledcircbullet • BULLET */
+ { 0x0ae7, 0x25aa }, /* enfilledsqbullet ▪ BLACK SMALL SQUARE */
+ { 0x0ae8, 0x25b2 }, /* filledtribulletup ▲ BLACK UP-POINTING TRIANGLE */
+ { 0x0ae9, 0x25bc }, /* filledtribulletdown ▼ BLACK DOWN-POINTING TRIANGLE */
+ { 0x0aea, 0x261c }, /* leftpointer ☜ WHITE LEFT POINTING INDEX */
+ { 0x0aeb, 0x261e }, /* rightpointer ☞ WHITE RIGHT POINTING INDEX */
+ { 0x0aec, 0x2663 }, /* club ♣ BLACK CLUB SUIT */
+ { 0x0aed, 0x2666 }, /* diamond ♦ BLACK DIAMOND SUIT */
+ { 0x0aee, 0x2665 }, /* heart ♥ BLACK HEART SUIT */
+ { 0x0af0, 0x2720 }, /* maltesecross ✠ MALTESE CROSS */
+ { 0x0af1, 0x2020 }, /* dagger † DAGGER */
+ { 0x0af2, 0x2021 }, /* doubledagger ‡ DOUBLE DAGGER */
+ { 0x0af3, 0x2713 }, /* checkmark ✓ CHECK MARK */
+ { 0x0af4, 0x2717 }, /* ballotcross ✗ BALLOT X */
+ { 0x0af5, 0x266f }, /* musicalsharp ♯ MUSIC SHARP SIGN */
+ { 0x0af6, 0x266d }, /* musicalflat ♭ MUSIC FLAT SIGN */
+ { 0x0af7, 0x2642 }, /* malesymbol ♂ MALE SIGN */
+ { 0x0af8, 0x2640 }, /* femalesymbol ♀ FEMALE SIGN */
+ { 0x0af9, 0x260e }, /* telephone ☎ BLACK TELEPHONE */
+ { 0x0afa, 0x2315 }, /* telephonerecorder ⌕ TELEPHONE RECORDER */
+ { 0x0afb, 0x2117 }, /* phonographcopyright ℗ SOUND RECORDING COPYRIGHT */
+ { 0x0afc, 0x2038 }, /* caret ‸ CARET */
+ { 0x0afd, 0x201a }, /* singlelowquotemark ‚ SINGLE LOW-9 QUOTATION MARK */
+ { 0x0afe, 0x201e }, /* doublelowquotemark „ DOUBLE LOW-9 QUOTATION MARK */
+ /* 0x0aff cursor ? ??? */
+ { 0x0ba3, 0x003c }, /* leftcaret < LESS-THAN SIGN */
+ { 0x0ba6, 0x003e }, /* rightcaret > GREATER-THAN SIGN */
+ { 0x0ba8, 0x2228 }, /* downcaret ∨ LOGICAL OR */
+ { 0x0ba9, 0x2227 }, /* upcaret ∧ LOGICAL AND */
+ { 0x0bc0, 0x00af }, /* overbar ¯ MACRON */
+ { 0x0bc2, 0x22a4 }, /* downtack ⊤ DOWN TACK */
+ { 0x0bc3, 0x2229 }, /* upshoe ∩ INTERSECTION */
+ { 0x0bc4, 0x230a }, /* downstile ⌊ LEFT FLOOR */
+ { 0x0bc6, 0x005f }, /* underbar _ LOW LINE */
+ { 0x0bca, 0x2218 }, /* jot ∘ RING OPERATOR */
+ { 0x0bcc, 0x2395 }, /* quad ⎕ APL FUNCTIONAL SYMBOL QUAD (Unicode 3.0) */
+ { 0x0bce, 0x22a5 }, /* uptack ⊥ UP TACK */
+ { 0x0bcf, 0x25cb }, /* circle ○ WHITE CIRCLE */
+ { 0x0bd3, 0x2308 }, /* upstile ⌈ LEFT CEILING */
+ { 0x0bd6, 0x222a }, /* downshoe ∪ UNION */
+ { 0x0bd8, 0x2283 }, /* rightshoe ⊃ SUPERSET OF */
+ { 0x0bda, 0x2282 }, /* leftshoe ⊂ SUBSET OF */
+ { 0x0bdc, 0x22a3 }, /* lefttack ⊣ LEFT TACK */
+ { 0x0bfc, 0x22a2 }, /* righttack ⊢ RIGHT TACK */
+ { 0x0cdf, 0x2017 }, /* hebrew_doublelowline ‗ DOUBLE LOW LINE */
+ { 0x0ce0, 0x05d0 }, /* hebrew_aleph א HEBREW LETTER ALEF */
+ { 0x0ce1, 0x05d1 }, /* hebrew_bet ב HEBREW LETTER BET */
+ { 0x0ce2, 0x05d2 }, /* hebrew_gimel ג HEBREW LETTER GIMEL */
+ { 0x0ce3, 0x05d3 }, /* hebrew_dalet ד HEBREW LETTER DALET */
+ { 0x0ce4, 0x05d4 }, /* hebrew_he ה HEBREW LETTER HE */
+ { 0x0ce5, 0x05d5 }, /* hebrew_waw ו HEBREW LETTER VAV */
+ { 0x0ce6, 0x05d6 }, /* hebrew_zain ז HEBREW LETTER ZAYIN */
+ { 0x0ce7, 0x05d7 }, /* hebrew_chet ח HEBREW LETTER HET */
+ { 0x0ce8, 0x05d8 }, /* hebrew_tet ט HEBREW LETTER TET */
+ { 0x0ce9, 0x05d9 }, /* hebrew_yod י HEBREW LETTER YOD */
+ { 0x0cea, 0x05da }, /* hebrew_finalkaph ך HEBREW LETTER FINAL KAF */
+ { 0x0ceb, 0x05db }, /* hebrew_kaph כ HEBREW LETTER KAF */
+ { 0x0cec, 0x05dc }, /* hebrew_lamed ל HEBREW LETTER LAMED */
+ { 0x0ced, 0x05dd }, /* hebrew_finalmem ם HEBREW LETTER FINAL MEM */
+ { 0x0cee, 0x05de }, /* hebrew_mem מ HEBREW LETTER MEM */
+ { 0x0cef, 0x05df }, /* hebrew_finalnun ן HEBREW LETTER FINAL NUN */
+ { 0x0cf0, 0x05e0 }, /* hebrew_nun נ HEBREW LETTER NUN */
+ { 0x0cf1, 0x05e1 }, /* hebrew_samech ס HEBREW LETTER SAMEKH */
+ { 0x0cf2, 0x05e2 }, /* hebrew_ayin ע HEBREW LETTER AYIN */
+ { 0x0cf3, 0x05e3 }, /* hebrew_finalpe ף HEBREW LETTER FINAL PE */
+ { 0x0cf4, 0x05e4 }, /* hebrew_pe פ HEBREW LETTER PE */
+ { 0x0cf5, 0x05e5 }, /* hebrew_finalzade ץ HEBREW LETTER FINAL TSADI */
+ { 0x0cf6, 0x05e6 }, /* hebrew_zade צ HEBREW LETTER TSADI */
+ { 0x0cf7, 0x05e7 }, /* hebrew_qoph ק HEBREW LETTER QOF */
+ { 0x0cf8, 0x05e8 }, /* hebrew_resh ר HEBREW LETTER RESH */
+ { 0x0cf9, 0x05e9 }, /* hebrew_shin ש HEBREW LETTER SHIN */
+ { 0x0cfa, 0x05ea }, /* hebrew_taw ת HEBREW LETTER TAV */
+ { 0x0da1, 0x0e01 }, /* Thai_kokai ก THAI CHARACTER KO KAI */
+ { 0x0da2, 0x0e02 }, /* Thai_khokhai ข THAI CHARACTER KHO KHAI */
+ { 0x0da3, 0x0e03 }, /* Thai_khokhuat ฃ THAI CHARACTER KHO KHUAT */
+ { 0x0da4, 0x0e04 }, /* Thai_khokhwai ค THAI CHARACTER KHO KHWAI */
+ { 0x0da5, 0x0e05 }, /* Thai_khokhon ฅ THAI CHARACTER KHO KHON */
+ { 0x0da6, 0x0e06 }, /* Thai_khorakhang ฆ THAI CHARACTER KHO RAKHANG */
+ { 0x0da7, 0x0e07 }, /* Thai_ngongu ง THAI CHARACTER NGO NGU */
+ { 0x0da8, 0x0e08 }, /* Thai_chochan จ THAI CHARACTER CHO CHAN */
+ { 0x0da9, 0x0e09 }, /* Thai_choching ฉ THAI CHARACTER CHO CHING */
+ { 0x0daa, 0x0e0a }, /* Thai_chochang ช THAI CHARACTER CHO CHANG */
+ { 0x0dab, 0x0e0b }, /* Thai_soso ซ THAI CHARACTER SO SO */
+ { 0x0dac, 0x0e0c }, /* Thai_chochoe ฌ THAI CHARACTER CHO CHOE */
+ { 0x0dad, 0x0e0d }, /* Thai_yoying ญ THAI CHARACTER YO YING */
+ { 0x0dae, 0x0e0e }, /* Thai_dochada ฎ THAI CHARACTER DO CHADA */
+ { 0x0daf, 0x0e0f }, /* Thai_topatak ฏ THAI CHARACTER TO PATAK */
+ { 0x0db0, 0x0e10 }, /* Thai_thothan ฐ THAI CHARACTER THO THAN */
+ { 0x0db1, 0x0e11 }, /* Thai_thonangmontho ฑ THAI CHARACTER THO NANGMONTHO */
+ { 0x0db2, 0x0e12 }, /* Thai_thophuthao ฒ THAI CHARACTER THO PHUTHAO */
+ { 0x0db3, 0x0e13 }, /* Thai_nonen ณ THAI CHARACTER NO NEN */
+ { 0x0db4, 0x0e14 }, /* Thai_dodek ด THAI CHARACTER DO DEK */
+ { 0x0db5, 0x0e15 }, /* Thai_totao ต THAI CHARACTER TO TAO */
+ { 0x0db6, 0x0e16 }, /* Thai_thothung ถ THAI CHARACTER THO THUNG */
+ { 0x0db7, 0x0e17 }, /* Thai_thothahan ท THAI CHARACTER THO THAHAN */
+ { 0x0db8, 0x0e18 }, /* Thai_thothong ธ THAI CHARACTER THO THONG */
+ { 0x0db9, 0x0e19 }, /* Thai_nonu น THAI CHARACTER NO NU */
+ { 0x0dba, 0x0e1a }, /* Thai_bobaimai บ THAI CHARACTER BO BAIMAI */
+ { 0x0dbb, 0x0e1b }, /* Thai_popla ป THAI CHARACTER PO PLA */
+ { 0x0dbc, 0x0e1c }, /* Thai_phophung ผ THAI CHARACTER PHO PHUNG */
+ { 0x0dbd, 0x0e1d }, /* Thai_fofa ฝ THAI CHARACTER FO FA */
+ { 0x0dbe, 0x0e1e }, /* Thai_phophan พ THAI CHARACTER PHO PHAN */
+ { 0x0dbf, 0x0e1f }, /* Thai_fofan ฟ THAI CHARACTER FO FAN */
+ { 0x0dc0, 0x0e20 }, /* Thai_phosamphao ภ THAI CHARACTER PHO SAMPHAO */
+ { 0x0dc1, 0x0e21 }, /* Thai_moma ม THAI CHARACTER MO MA */
+ { 0x0dc2, 0x0e22 }, /* Thai_yoyak ย THAI CHARACTER YO YAK */
+ { 0x0dc3, 0x0e23 }, /* Thai_rorua ร THAI CHARACTER RO RUA */
+ { 0x0dc4, 0x0e24 }, /* Thai_ru ฤ THAI CHARACTER RU */
+ { 0x0dc5, 0x0e25 }, /* Thai_loling ล THAI CHARACTER LO LING */
+ { 0x0dc6, 0x0e26 }, /* Thai_lu ฦ THAI CHARACTER LU */
+ { 0x0dc7, 0x0e27 }, /* Thai_wowaen ว THAI CHARACTER WO WAEN */
+ { 0x0dc8, 0x0e28 }, /* Thai_sosala ศ THAI CHARACTER SO SALA */
+ { 0x0dc9, 0x0e29 }, /* Thai_sorusi ษ THAI CHARACTER SO RUSI */
+ { 0x0dca, 0x0e2a }, /* Thai_sosua ส THAI CHARACTER SO SUA */
+ { 0x0dcb, 0x0e2b }, /* Thai_hohip ห THAI CHARACTER HO HIP */
+ { 0x0dcc, 0x0e2c }, /* Thai_lochula ฬ THAI CHARACTER LO CHULA */
+ { 0x0dcd, 0x0e2d }, /* Thai_oang อ THAI CHARACTER O ANG */
+ { 0x0dce, 0x0e2e }, /* Thai_honokhuk ฮ THAI CHARACTER HO NOKHUK */
+ { 0x0dcf, 0x0e2f }, /* Thai_paiyannoi ฯ THAI CHARACTER PAIYANNOI */
+ { 0x0dd0, 0x0e30 }, /* Thai_saraa ะ THAI CHARACTER SARA A */
+ { 0x0dd1, 0x0e31 }, /* Thai_maihanakat ั THAI CHARACTER MAI HAN-AKAT */
+ { 0x0dd2, 0x0e32 }, /* Thai_saraaa า THAI CHARACTER SARA AA */
+ { 0x0dd3, 0x0e33 }, /* Thai_saraam ำ THAI CHARACTER SARA AM */
+ { 0x0dd4, 0x0e34 }, /* Thai_sarai ิ THAI CHARACTER SARA I */
+ { 0x0dd5, 0x0e35 }, /* Thai_saraii ี THAI CHARACTER SARA II */
+ { 0x0dd6, 0x0e36 }, /* Thai_saraue ึ THAI CHARACTER SARA UE */
+ { 0x0dd7, 0x0e37 }, /* Thai_sarauee ื THAI CHARACTER SARA UEE */
+ { 0x0dd8, 0x0e38 }, /* Thai_sarau ุ THAI CHARACTER SARA U */
+ { 0x0dd9, 0x0e39 }, /* Thai_sarauu ู THAI CHARACTER SARA UU */
+ { 0x0dda, 0x0e3a }, /* Thai_phinthu ฺ THAI CHARACTER PHINTHU */
+ { 0x0dde, 0x0e3e }, /* Thai_maihanakat_maitho ??? */
+ { 0x0ddf, 0x0e3f }, /* Thai_baht ฿ THAI CURRENCY SYMBOL BAHT */
+ { 0x0de0, 0x0e40 }, /* Thai_sarae เ THAI CHARACTER SARA E */
+ { 0x0de1, 0x0e41 }, /* Thai_saraae แ THAI CHARACTER SARA AE */
+ { 0x0de2, 0x0e42 }, /* Thai_sarao โ THAI CHARACTER SARA O */
+ { 0x0de3, 0x0e43 }, /* Thai_saraaimaimuan ใ THAI CHARACTER SARA AI MAIMUAN */
+ { 0x0de4, 0x0e44 }, /* Thai_saraaimaimalai ไ THAI CHARACTER SARA AI MAIMALAI */
+ { 0x0de5, 0x0e45 }, /* Thai_lakkhangyao ๅ THAI CHARACTER LAKKHANGYAO */
+ { 0x0de6, 0x0e46 }, /* Thai_maiyamok ๆ THAI CHARACTER MAIYAMOK */
+ { 0x0de7, 0x0e47 }, /* Thai_maitaikhu ็ THAI CHARACTER MAITAIKHU */
+ { 0x0de8, 0x0e48 }, /* Thai_maiek ่ THAI CHARACTER MAI EK */
+ { 0x0de9, 0x0e49 }, /* Thai_maitho ้ THAI CHARACTER MAI THO */
+ { 0x0dea, 0x0e4a }, /* Thai_maitri ๊ THAI CHARACTER MAI TRI */
+ { 0x0deb, 0x0e4b }, /* Thai_maichattawa ๋ THAI CHARACTER MAI CHATTAWA */
+ { 0x0dec, 0x0e4c }, /* Thai_thanthakhat ์ THAI CHARACTER THANTHAKHAT */
+ { 0x0ded, 0x0e4d }, /* Thai_nikhahit ํ THAI CHARACTER NIKHAHIT */
+ { 0x0df0, 0x0e50 }, /* Thai_leksun ๐ THAI DIGIT ZERO */
+ { 0x0df1, 0x0e51 }, /* Thai_leknung ๑ THAI DIGIT ONE */
+ { 0x0df2, 0x0e52 }, /* Thai_leksong ๒ THAI DIGIT TWO */
+ { 0x0df3, 0x0e53 }, /* Thai_leksam ๓ THAI DIGIT THREE */
+ { 0x0df4, 0x0e54 }, /* Thai_leksi ๔ THAI DIGIT FOUR */
+ { 0x0df5, 0x0e55 }, /* Thai_lekha ๕ THAI DIGIT FIVE */
+ { 0x0df6, 0x0e56 }, /* Thai_lekhok ๖ THAI DIGIT SIX */
+ { 0x0df7, 0x0e57 }, /* Thai_lekchet ๗ THAI DIGIT SEVEN */
+ { 0x0df8, 0x0e58 }, /* Thai_lekpaet ๘ THAI DIGIT EIGHT */
+ { 0x0df9, 0x0e59 }, /* Thai_lekkao ๙ THAI DIGIT NINE */
+ { 0x0ea1, 0x3131 }, /* Hangul_Kiyeog ㄱ HANGUL LETTER KIYEOK */
+ { 0x0ea2, 0x3132 }, /* Hangul_SsangKiyeog ㄲ HANGUL LETTER SSANGKIYEOK */
+ { 0x0ea3, 0x3133 }, /* Hangul_KiyeogSios ㄳ HANGUL LETTER KIYEOK-SIOS */
+ { 0x0ea4, 0x3134 }, /* Hangul_Nieun ㄴ HANGUL LETTER NIEUN */
+ { 0x0ea5, 0x3135 }, /* Hangul_NieunJieuj ㄵ HANGUL LETTER NIEUN-CIEUC */
+ { 0x0ea6, 0x3136 }, /* Hangul_NieunHieuh ㄶ HANGUL LETTER NIEUN-HIEUH */
+ { 0x0ea7, 0x3137 }, /* Hangul_Dikeud ㄷ HANGUL LETTER TIKEUT */
+ { 0x0ea8, 0x3138 }, /* Hangul_SsangDikeud ㄸ HANGUL LETTER SSANGTIKEUT */
+ { 0x0ea9, 0x3139 }, /* Hangul_Rieul ㄹ HANGUL LETTER RIEUL */
+ { 0x0eaa, 0x313a }, /* Hangul_RieulKiyeog ㄺ HANGUL LETTER RIEUL-KIYEOK */
+ { 0x0eab, 0x313b }, /* Hangul_RieulMieum ㄻ HANGUL LETTER RIEUL-MIEUM */
+ { 0x0eac, 0x313c }, /* Hangul_RieulPieub ㄼ HANGUL LETTER RIEUL-PIEUP */
+ { 0x0ead, 0x313d }, /* Hangul_RieulSios ㄽ HANGUL LETTER RIEUL-SIOS */
+ { 0x0eae, 0x313e }, /* Hangul_RieulTieut ㄾ HANGUL LETTER RIEUL-THIEUTH */
+ { 0x0eaf, 0x313f }, /* Hangul_RieulPhieuf ㄿ HANGUL LETTER RIEUL-PHIEUPH */
+ { 0x0eb0, 0x3140 }, /* Hangul_RieulHieuh ㅀ HANGUL LETTER RIEUL-HIEUH */
+ { 0x0eb1, 0x3141 }, /* Hangul_Mieum ㅁ HANGUL LETTER MIEUM */
+ { 0x0eb2, 0x3142 }, /* Hangul_Pieub ㅂ HANGUL LETTER PIEUP */
+ { 0x0eb3, 0x3143 }, /* Hangul_SsangPieub ㅃ HANGUL LETTER SSANGPIEUP */
+ { 0x0eb4, 0x3144 }, /* Hangul_PieubSios ㅄ HANGUL LETTER PIEUP-SIOS */
+ { 0x0eb5, 0x3145 }, /* Hangul_Sios ㅅ HANGUL LETTER SIOS */
+ { 0x0eb6, 0x3146 }, /* Hangul_SsangSios ㅆ HANGUL LETTER SSANGSIOS */
+ { 0x0eb7, 0x3147 }, /* Hangul_Ieung ㅇ HANGUL LETTER IEUNG */
+ { 0x0eb8, 0x3148 }, /* Hangul_Jieuj ㅈ HANGUL LETTER CIEUC */
+ { 0x0eb9, 0x3149 }, /* Hangul_SsangJieuj ㅉ HANGUL LETTER SSANGCIEUC */
+ { 0x0eba, 0x314a }, /* Hangul_Cieuc ㅊ HANGUL LETTER CHIEUCH */
+ { 0x0ebb, 0x314b }, /* Hangul_Khieuq ㅋ HANGUL LETTER KHIEUKH */
+ { 0x0ebc, 0x314c }, /* Hangul_Tieut ㅌ HANGUL LETTER THIEUTH */
+ { 0x0ebd, 0x314d }, /* Hangul_Phieuf ㅍ HANGUL LETTER PHIEUPH */
+ { 0x0ebe, 0x314e }, /* Hangul_Hieuh ㅎ HANGUL LETTER HIEUH */
+ { 0x0ebf, 0x314f }, /* Hangul_A ㅏ HANGUL LETTER A */
+ { 0x0ec0, 0x3150 }, /* Hangul_AE ㅐ HANGUL LETTER AE */
+ { 0x0ec1, 0x3151 }, /* Hangul_YA ㅑ HANGUL LETTER YA */
+ { 0x0ec2, 0x3152 }, /* Hangul_YAE ㅒ HANGUL LETTER YAE */
+ { 0x0ec3, 0x3153 }, /* Hangul_EO ㅓ HANGUL LETTER EO */
+ { 0x0ec4, 0x3154 }, /* Hangul_E ㅔ HANGUL LETTER E */
+ { 0x0ec5, 0x3155 }, /* Hangul_YEO ㅕ HANGUL LETTER YEO */
+ { 0x0ec6, 0x3156 }, /* Hangul_YE ㅖ HANGUL LETTER YE */
+ { 0x0ec7, 0x3157 }, /* Hangul_O ㅗ HANGUL LETTER O */
+ { 0x0ec8, 0x3158 }, /* Hangul_WA ㅘ HANGUL LETTER WA */
+ { 0x0ec9, 0x3159 }, /* Hangul_WAE ㅙ HANGUL LETTER WAE */
+ { 0x0eca, 0x315a }, /* Hangul_OE ㅚ HANGUL LETTER OE */
+ { 0x0ecb, 0x315b }, /* Hangul_YO ㅛ HANGUL LETTER YO */
+ { 0x0ecc, 0x315c }, /* Hangul_U ㅜ HANGUL LETTER U */
+ { 0x0ecd, 0x315d }, /* Hangul_WEO ㅝ HANGUL LETTER WEO */
+ { 0x0ece, 0x315e }, /* Hangul_WE ㅞ HANGUL LETTER WE */
+ { 0x0ecf, 0x315f }, /* Hangul_WI ㅟ HANGUL LETTER WI */
+ { 0x0ed0, 0x3160 }, /* Hangul_YU ㅠ HANGUL LETTER YU */
+ { 0x0ed1, 0x3161 }, /* Hangul_EU ㅡ HANGUL LETTER EU */
+ { 0x0ed2, 0x3162 }, /* Hangul_YI ㅢ HANGUL LETTER YI */
+ { 0x0ed3, 0x3163 }, /* Hangul_I ㅣ HANGUL LETTER I */
+ { 0x0ed4, 0x11a8 }, /* Hangul_J_Kiyeog ᆨ HANGUL JONGSEONG KIYEOK */
+ { 0x0ed5, 0x11a9 }, /* Hangul_J_SsangKiyeog ᆩ HANGUL JONGSEONG SSANGKIYEOK */
+ { 0x0ed6, 0x11aa }, /* Hangul_J_KiyeogSios ᆪ HANGUL JONGSEONG KIYEOK-SIOS */
+ { 0x0ed7, 0x11ab }, /* Hangul_J_Nieun ᆫ HANGUL JONGSEONG NIEUN */
+ { 0x0ed8, 0x11ac }, /* Hangul_J_NieunJieuj ᆬ HANGUL JONGSEONG NIEUN-CIEUC */
+ { 0x0ed9, 0x11ad }, /* Hangul_J_NieunHieuh ᆭ HANGUL JONGSEONG NIEUN-HIEUH */
+ { 0x0eda, 0x11ae }, /* Hangul_J_Dikeud ᆮ HANGUL JONGSEONG TIKEUT */
+ { 0x0edb, 0x11af }, /* Hangul_J_Rieul ᆯ HANGUL JONGSEONG RIEUL */
+ { 0x0edc, 0x11b0 }, /* Hangul_J_RieulKiyeog ᆰ HANGUL JONGSEONG RIEUL-KIYEOK */
+ { 0x0edd, 0x11b1 }, /* Hangul_J_RieulMieum ᆱ HANGUL JONGSEONG RIEUL-MIEUM */
+ { 0x0ede, 0x11b2 }, /* Hangul_J_RieulPieub ᆲ HANGUL JONGSEONG RIEUL-PIEUP */
+ { 0x0edf, 0x11b3 }, /* Hangul_J_RieulSios ᆳ HANGUL JONGSEONG RIEUL-SIOS */
+ { 0x0ee0, 0x11b4 }, /* Hangul_J_RieulTieut ᆴ HANGUL JONGSEONG RIEUL-THIEUTH */
+ { 0x0ee1, 0x11b5 }, /* Hangul_J_RieulPhieuf ᆵ HANGUL JONGSEONG RIEUL-PHIEUPH */
+ { 0x0ee2, 0x11b6 }, /* Hangul_J_RieulHieuh ᆶ HANGUL JONGSEONG RIEUL-HIEUH */
+ { 0x0ee3, 0x11b7 }, /* Hangul_J_Mieum ᆷ HANGUL JONGSEONG MIEUM */
+ { 0x0ee4, 0x11b8 }, /* Hangul_J_Pieub ᆸ HANGUL JONGSEONG PIEUP */
+ { 0x0ee5, 0x11b9 }, /* Hangul_J_PieubSios ᆹ HANGUL JONGSEONG PIEUP-SIOS */
+ { 0x0ee6, 0x11ba }, /* Hangul_J_Sios ᆺ HANGUL JONGSEONG SIOS */
+ { 0x0ee7, 0x11bb }, /* Hangul_J_SsangSios ᆻ HANGUL JONGSEONG SSANGSIOS */
+ { 0x0ee8, 0x11bc }, /* Hangul_J_Ieung ᆼ HANGUL JONGSEONG IEUNG */
+ { 0x0ee9, 0x11bd }, /* Hangul_J_Jieuj ᆽ HANGUL JONGSEONG CIEUC */
+ { 0x0eea, 0x11be }, /* Hangul_J_Cieuc ᆾ HANGUL JONGSEONG CHIEUCH */
+ { 0x0eeb, 0x11bf }, /* Hangul_J_Khieuq ᆿ HANGUL JONGSEONG KHIEUKH */
+ { 0x0eec, 0x11c0 }, /* Hangul_J_Tieut ᇀ HANGUL JONGSEONG THIEUTH */
+ { 0x0eed, 0x11c1 }, /* Hangul_J_Phieuf ᇁ HANGUL JONGSEONG PHIEUPH */
+ { 0x0eee, 0x11c2 }, /* Hangul_J_Hieuh ᇂ HANGUL JONGSEONG HIEUH */
+ { 0x0eef, 0x316d }, /* Hangul_RieulYeorinHieuh ㅭ HANGUL LETTER RIEUL-YEORINHIEUH */
+ { 0x0ef0, 0x3171 }, /* Hangul_SunkyeongeumMieum ㅱ HANGUL LETTER KAPYEOUNMIEUM */
+ { 0x0ef1, 0x3178 }, /* Hangul_SunkyeongeumPieub ㅸ HANGUL LETTER KAPYEOUNPIEUP */
+ { 0x0ef2, 0x317f }, /* Hangul_PanSios ㅿ HANGUL LETTER PANSIOS */
+ { 0x0ef3, 0x3181 }, /* Hangul_KkogjiDalrinIeung ㆁ HANGUL LETTER YESIEUNG */
+ { 0x0ef4, 0x3184 }, /* Hangul_SunkyeongeumPhieuf ㆄ HANGUL LETTER KAPYEOUNPHIEUPH */
+ { 0x0ef5, 0x3186 }, /* Hangul_YeorinHieuh ㆆ HANGUL LETTER YEORINHIEUH */
+ { 0x0ef6, 0x318d }, /* Hangul_AraeA ㆍ HANGUL LETTER ARAEA */
+ { 0x0ef7, 0x318e }, /* Hangul_AraeAE ㆎ HANGUL LETTER ARAEAE */
+ { 0x0ef8, 0x11eb }, /* Hangul_J_PanSios ᇫ HANGUL JONGSEONG PANSIOS */
+ { 0x0ef9, 0x11f0 }, /* Hangul_J_KkogjiDalrinIeung ᇰ HANGUL JONGSEONG YESIEUNG */
+ { 0x0efa, 0x11f9 }, /* Hangul_J_YeorinHieuh ᇹ HANGUL JONGSEONG YEORINHIEUH */
+ { 0x0eff, 0x20a9 }, /* Korean_Won ₩ WON SIGN */
+ { 0x13bc, 0x0152 }, /* OE ΠLATIN CAPITAL LIGATURE OE */
+ { 0x13bd, 0x0153 }, /* oe œ LATIN SMALL LIGATURE OE */
+ { 0x13be, 0x0178 }, /* Ydiaeresis Ÿ LATIN CAPITAL LETTER Y WITH DIAERESIS */
+ { 0x20ac, 0x20ac }, /* EuroSign € EURO SIGN */
+};
+
+/* binary search with range check */
+static uint32_t
+bin_search(const struct codepair *table, size_t length, xcb_keysym_t keysym)
+{
+ size_t first = 0;
+ size_t last = length;
+
+ if (keysym < table[0].keysym || keysym > table[length].keysym)
+ return 0;
+
+ /* binary search in table */
+ while (last >= first) {
+ size_t mid = (first + last) / 2;
+ if (table[mid].keysym < keysym)
+ first = mid + 1;
+ else if (table[mid].keysym > keysym)
+ last = mid - 1;
+ else /* found it */
+ return table[mid].ucs;
+ }
+
+ /* no matching Unicode value found in table */
+ return NO_KEYSYM_UNICODE_CONVERSION;
+}
+
+uint32_t
+xkb_keysym_to_utf32(xcb_keysym_t keysym)
+{
+ /* first check for Latin-1 characters (1:1 mapping) */
+ if ((keysym >= 0x0020 && keysym <= 0x007e) ||
+ (keysym >= 0x00a0 && keysym <= 0x00ff))
+ return keysym;
+
+ /* patch encoding botch */
+ if (keysym == XKB_KEY_KP_Space)
+ return XKB_KEY_space & 0x7f;
+ /* special keysyms */
+ if ((keysym >= XKB_KEY_BackSpace && keysym <= XKB_KEY_Clear) ||
+ (keysym >= XKB_KEY_KP_Multiply && keysym <= XKB_KEY_KP_9) ||
+ keysym == XKB_KEY_Return || keysym == XKB_KEY_Escape ||
+ keysym == XKB_KEY_Delete || keysym == XKB_KEY_KP_Tab ||
+ keysym == XKB_KEY_KP_Enter || keysym == XKB_KEY_KP_Equal)
+ return keysym & 0x7f;
+
+ /* also check for directly encoded Unicode codepoints */
+
+ /* Exclude surrogates: they are invalid in UTF-32.
+ * See https://www.unicode.org/versions/Unicode15.0.0/ch03.pdf#G28875
+ * for further details.
+ */
+ if (0x0100d800 <= keysym && keysym <= 0x0100dfff)
+ return NO_KEYSYM_UNICODE_CONVERSION;
+ /*
+ * In theory, this is supposed to start from 0x100100, such that the ASCII
+ * range, which is already covered by 0x00-0xff, can't be encoded in two
+ * ways. However, changing this after a couple of decades probably won't
+ * go well, so it stays as it is.
+ */
+ if (XKB_KEYSYM_UNICODE_OFFSET <= keysym && keysym <= XKB_KEYSYM_UNICODE_MAX)
+ return keysym - XKB_KEYSYM_UNICODE_OFFSET;
+
+ /* search main table */
+ return bin_search(keysymtab, ARRAY_SIZE(keysymtab) - 1, keysym);
+}
+
+xcb_keysym_t
+xkb_utf32_to_keysym(uint32_t ucs)
+{
+ /* first check for Latin-1 characters (1:1 mapping) */
+ if ((ucs >= 0x0020 && ucs <= 0x007e) ||
+ (ucs >= 0x00a0 && ucs <= 0x00ff))
+ return ucs;
+
+ /* special keysyms */
+ if ((ucs >= (XKB_KEY_BackSpace & 0x7f) && ucs <= (XKB_KEY_Clear & 0x7f)) ||
+ ucs == (XKB_KEY_Return & 0x7f) || ucs == (XKB_KEY_Escape & 0x7f))
+ return ucs | 0xff00;
+ if (ucs == (XKB_KEY_Delete & 0x7f))
+ return XKB_KEY_Delete;
+
+ /* Unicode non-symbols and code points outside Unicode planes */
+ if ((ucs >= 0xd800 && ucs <= 0xdfff) ||
+ (ucs >= 0xfdd0 && ucs <= 0xfdef) ||
+ ucs > 0x10ffff || (ucs & 0xfffe) == 0xfffe)
+ return XKB_KEY_NoSymbol;
+
+ /* search main table */
+ for (size_t i = 0; i < ARRAY_SIZE(keysymtab); i++)
+ if (keysymtab[i].ucs == ucs)
+ return keysymtab[i].keysym;
+
+ /* Use direct encoding if everything else fails */
+ return ucs | XKB_KEYSYM_UNICODE_OFFSET;
+}
+
+int
+utf32_to_utf8(uint32_t unichar, char *buffer)
+{
+ int count, shift, length;
+ uint8_t head;
+
+ /* NOLINTBEGIN(bugprone-branch-clone) */
+ if (unichar <= 0x007f) {
+ buffer[0] = (char) unichar;
+
+ buffer[1] = '\0';
+ return 2;
+ }
+ else if (unichar <= 0x07FF) {
+ length = 2;
+ head = 0xc0;
+ }
+ /* Handle surrogates */
+ else if (0xd800 <= unichar && unichar <= 0xdfff) {
+ goto ill_formed_code_unit_subsequence;
+ }
+ else if (unichar <= 0xffff) {
+ length = 3;
+ head = 0xe0;
+ }
+ else if (unichar <= 0x10ffff) {
+ length = 4;
+ head = 0xf0;
+ }
+ else {
+ goto ill_formed_code_unit_subsequence;
+ }
+ /* NOLINTEND(bugprone-branch-clone) */
+
+ for (count = length - 1, shift = 0; count > 0; count--, shift += 6)
+ buffer[count] = (char)(0x80 | ((unichar >> shift) & 0x3f));
+
+ buffer[0] = (char)(head | ((unichar >> shift) & 0x3f));
+ buffer[length] = '\0';
+
+ return length + 1;
+
+ill_formed_code_unit_subsequence:
+ buffer[0] = '\0';
+ return 0;
+}
+
+/*
+ * SPDX-SnippetBegin
+ * Copyright © 2012 Intel Corporation
+ * SPDX-License-Identifier: MIT
+ *
+ * Author: Rob Bradford <rob@linux.intel.com>
+ */
+
+int
+xkb_keysym_to_utf8(xcb_keysym_t keysym, char *buffer, size_t size)
+{
+ uint32_t codepoint;
+
+ /* Unicode code point takes up to 4 bytes in UTF-8, plus the
+ * NULL-terminating byte. */
+ if (size < XKB_KEYSYM_UTF8_MAX_SIZE)
+ return -1;
+
+ codepoint = xkb_keysym_to_utf32(keysym);
+
+ if (codepoint == NO_KEYSYM_UNICODE_CONVERSION)
+ return 0;
+
+ return utf32_to_utf8(codepoint, buffer);
+}
+
+/* SPDX-SnippetEnd */
+
+
+
+
+
+
+
+
+
+static char
+XkbToControl(char ch)
+{
+ char c = ch;
+
+ if ((c >= '@' && c < '\177') || c == ' ')
+ c &= 0x1F;
+ else if (c == '2')
+ c = '\000';
+ else if (c >= '3' && c <= '7')
+ c -= ('3' - '\033');
+ else if (c == '8')
+ c = '\177';
+ else if (c == '/')
+ c = '_' & 0x1F;
+ return c;
+}
+
+
+
diff --git a/src/platform/gfx/platform_gfx.c b/src/platform/gfx/platform_gfx.c
@@ -0,0 +1,48 @@
+//////////////////////////////////////////////////////////////////
+// platform_gfx.c
+
+RV_GLOBAL bool32 rv_key_pressed(rv_keycode_t keycode)
+{
+ return rv_keyboard_context.current_state[keycode] && !rv_keyboard_context.last_state[keycode];
+}
+
+RV_GLOBAL bool32 rv_key_down(rv_keycode_t keycode)
+{
+ return rv_keyboard_context.current_state[keycode];
+}
+
+RV_GLOBAL bool32 rv_key_released(rv_keycode_t keycode)
+{
+ return !rv_keyboard_context.current_state[keycode] && rv_keyboard_context.last_state[keycode];
+}
+
+RV_GLOBAL bool32 rv_key_changed(rv_keycode_t keycode)
+{
+ return rv_keyboard_context.current_state[keycode] != rv_keyboard_context.last_state[keycode];
+}
+
+RV_GLOBAL bool32 rv_key_mod(rv_keycode_t keycode, u8 mask)
+{
+ return (rv_keyboard_context.mods & mask) != 0;
+}
+
+RV_GLOBAL void rv_keyboard_context_update(rv_keyboard_context_t* kbd_ctx, rv_event_t* events)
+{
+ rv_mem_copy(kbd_ctx->last_state, kbd_ctx->current_state, sizeof(kbd_ctx->current_state));
+
+ for (rv_event_t* e = events; e; e = e->next) {
+ if (e->key_press.keycode == RV_KEYCODE_INVALID) continue;
+ switch(e->type) {
+ case RV_EVENT_KEY_PRESS: {
+ kbd_ctx->current_state[e->key_press.keycode] = true;
+ kbd_ctx->mods = e->key_press.mods;
+ kbd_ctx->group = e->key_press.group;
+ } break;
+ case RV_EVENT_KEY_RELEASE: {
+ kbd_ctx->current_state[e->key_release.keycode] = false;
+ kbd_ctx->group = e->key_press.group;
+ } break;
+ default: break;
+ }
+ }
+}
diff --git a/src/platform/gfx/platform_gfx.h b/src/platform/gfx/platform_gfx.h
@@ -2,11 +2,17 @@
// platform_gfx.h
//////////////////////////////////////////////////////////////////
+// forward declares
+
+typedef struct rv_event_t rv_event_t;
+
+//////////////////////////////////////////////////////////////////
// Window creation
typedef struct {
- const char* name;
- bool32 attach_render;
+ rv_str8 name;
+ bool32 attach_render;
+ rv_rect rect;
} rv_window_desc_t;
typedef struct rv_window_handle_t rv_window_handle_t;
@@ -15,45 +21,237 @@ RV_GLOBAL rv_window_handle_t* rv_create_window(rv_window_desc_t desc);
RV_GLOBAL void rv_destroy_window(rv_window_handle_t** handle);
//////////////////////////////////////////////////////////////////
+// Keyboard Input
+
+typedef enum rv_keycode_t {
+ RV_KEYCODE_INVALID,
+ RV_KEYCODE_SPACE,
+ RV_KEYCODE_APOSTROPHE, /* ' */
+ RV_KEYCODE_COMMA, /* , */
+ RV_KEYCODE_MINUS, /* - */
+ RV_KEYCODE_PERIOD, /* . */
+ RV_KEYCODE_SLASH, /* / */
+ RV_KEYCODE_0,
+ RV_KEYCODE_1,
+ RV_KEYCODE_2,
+ RV_KEYCODE_3,
+ RV_KEYCODE_4,
+ RV_KEYCODE_5,
+ RV_KEYCODE_6,
+ RV_KEYCODE_7,
+ RV_KEYCODE_8,
+ RV_KEYCODE_9,
+ RV_KEYCODE_SEMICOLON, /* ; */
+ RV_KEYCODE_EQUAL, /* = */
+ RV_KEYCODE_A,
+ RV_KEYCODE_B,
+ RV_KEYCODE_C,
+ RV_KEYCODE_D,
+ RV_KEYCODE_E,
+ RV_KEYCODE_F,
+ RV_KEYCODE_G,
+ RV_KEYCODE_H,
+ RV_KEYCODE_I,
+ RV_KEYCODE_J,
+ RV_KEYCODE_K,
+ RV_KEYCODE_L,
+ RV_KEYCODE_M,
+ RV_KEYCODE_N,
+ RV_KEYCODE_O,
+ RV_KEYCODE_P,
+ RV_KEYCODE_Q,
+ RV_KEYCODE_R,
+ RV_KEYCODE_S,
+ RV_KEYCODE_T,
+ RV_KEYCODE_U,
+ RV_KEYCODE_V,
+ RV_KEYCODE_W,
+ RV_KEYCODE_X,
+ RV_KEYCODE_Y,
+ RV_KEYCODE_Z,
+ RV_KEYCODE_BAR,
+ RV_KEYCODE_LEFT_BRACKET, /* [ */
+ RV_KEYCODE_BACKSLASH, /* \ */
+ RV_KEYCODE_RIGHT_BRACKET, /* ] */
+ RV_KEYCODE_GRAVE_ACCENT, /* ` */
+ RV_KEYCODE_ACUTE_ACCENT, /* ´ */
+ RV_KEYCODE_DIAERESIS, /* ¨ */
+ RV_KEYCODE_CIRCUMFLEX, /* ^ */
+ RV_KEYCODE_TILDE, /* ~ */
+ RV_KEYCODE_WORLD_1, /* non-US #1 */
+ RV_KEYCODE_WORLD_2, /* non-US #2 */
+ RV_KEYCODE_ESC,
+ RV_KEYCODE_ENTER,
+ RV_KEYCODE_TAB,
+ RV_KEYCODE_BACKSPACE,
+ RV_KEYCODE_INSERT,
+ RV_KEYCODE_DELETE,
+ RV_KEYCODE_RIGHT,
+ RV_KEYCODE_LEFT,
+ RV_KEYCODE_DOWN,
+ RV_KEYCODE_UP,
+ RV_KEYCODE_PAGE_UP,
+ RV_KEYCODE_PAGE_DOWN,
+ RV_KEYCODE_HOME,
+ RV_KEYCODE_END,
+ RV_KEYCODE_CAPS_LOCK,
+ RV_KEYCODE_SCROLL_LOCK,
+ RV_KEYCODE_NUM_LOCK,
+ RV_KEYCODE_PRINT_SCREEN,
+ RV_KEYCODE_PAUSE,
+ RV_KEYCODE_F1,
+ RV_KEYCODE_F2,
+ RV_KEYCODE_F3,
+ RV_KEYCODE_F4,
+ RV_KEYCODE_F5,
+ RV_KEYCODE_F6,
+ RV_KEYCODE_F7,
+ RV_KEYCODE_F8,
+ RV_KEYCODE_F9,
+ RV_KEYCODE_F10,
+ RV_KEYCODE_F11,
+ RV_KEYCODE_F12,
+ RV_KEYCODE_F13,
+ RV_KEYCODE_F14,
+ RV_KEYCODE_F15,
+ RV_KEYCODE_F16,
+ RV_KEYCODE_F17,
+ RV_KEYCODE_F18,
+ RV_KEYCODE_F19,
+ RV_KEYCODE_F20,
+ RV_KEYCODE_F21,
+ RV_KEYCODE_F22,
+ RV_KEYCODE_F23,
+ RV_KEYCODE_F24,
+ RV_KEYCODE_F25,
+ RV_KEYCODE_KP_0,
+ RV_KEYCODE_KP_1,
+ RV_KEYCODE_KP_2,
+ RV_KEYCODE_KP_3,
+ RV_KEYCODE_KP_4,
+ RV_KEYCODE_KP_5,
+ RV_KEYCODE_KP_6,
+ RV_KEYCODE_KP_7,
+ RV_KEYCODE_KP_8,
+ RV_KEYCODE_KP_9,
+ RV_KEYCODE_KP_DECIMAL,
+ RV_KEYCODE_KP_DIVIDE,
+ RV_KEYCODE_KP_MULTIPLY,
+ RV_KEYCODE_KP_SUBTRACT,
+ RV_KEYCODE_KP_ADD,
+ RV_KEYCODE_KP_ENTER,
+ RV_KEYCODE_KP_EQUAL,
+ RV_KEYCODE_LEFT_SHIFT,
+ RV_KEYCODE_LEFT_CONTROL,
+ RV_KEYCODE_LEFT_ALT,
+ RV_KEYCODE_LEFT_SUPER,
+ RV_KEYCODE_RIGHT_SHIFT,
+ RV_KEYCODE_RIGHT_CONTROL,
+ RV_KEYCODE_RIGHT_ALT,
+ RV_KEYCODE_RIGHT_SUPER,
+ RV_KEYCODE_MENU,
+
+ RV_KEYCODE_COUNT
+} rv_keycode_t;
+
+typedef enum {
+ RV_KEY_MOD_NONE = 0,
+ RV_KEY_MOD_1 = (1 << 0),
+ RV_KEY_MOD_2 = (1 << 1),
+ RV_KEY_MOD_3 = (1 << 2),
+ RV_KEY_MOD_4 = (1 << 3),
+ RV_KEY_MOD_5 = (1 << 4),
+ RV_KEY_MOD_7 = (1 << 5),
+ RV_KEY_MOD_8 = (1 << 7),
+
+ RV_KEY_MOD_SHIFT = RV_KEY_MOD_1,
+ RV_KEY_MOD_LOCK = RV_KEY_MOD_2,
+ RV_KEY_MOD_CTRL = RV_KEY_MOD_3,
+ RV_KEY_MOD_ALT = RV_KEY_MOD_4,
+ RV_KEY_MOD_NUM_LOCK = RV_KEY_MOD_5,
+ RV_KEY_MOD_SUPER = RV_KEY_MOD_7,
+ RV_KEY_MOD_ALTGR = RV_KEY_MOD_8,
+} rv_key_mods_t;
+
+typedef struct {
+ bool8 last_state[RV_KEYCODE_COUNT];
+ bool8 current_state[RV_KEYCODE_COUNT];
+ u8 mods;
+ u8 group;
+} rv_keyboard_context_t;
+
+// updated when rv_get_events() is called
+RV_GLOBAL rv_keyboard_context_t rv_keyboard_context;
+
+RV_GLOBAL void rv_keyboard_context_update(rv_keyboard_context_t* kbd_ctx, rv_event_t* events);
+
+RV_GLOBAL bool32 rv_key_pressed(rv_keycode_t keycode);
+RV_GLOBAL bool32 rv_key_down(rv_keycode_t keycode);
+RV_GLOBAL bool32 rv_key_released(rv_keycode_t keycode);
+RV_GLOBAL bool32 rv_key_changed(rv_keycode_t keycode);
+
+RV_GLOBAL bool32 rv_key_mod(rv_keycode_t keycode, u8 mask);
+
+
+//////////////////////////////////////////////////////////////////
// Event polling
typedef enum {
- RV_EVENT_INVALID,
+ RV_EVENT_INVALID,
- RV_EVENT_WINDOW_CHANGE, // move and resize
- RV_EVENT_WINDOW_CLOSE,
+ RV_EVENT_WINDOW_RESIZE,
+ RV_EVENT_WINDOW_CLOSE,
- RV_EVENT_KEY_PRESS,
- RV_EVENT_KEY_RELEASE,
+ RV_EVENT_KEY_PRESS,
+ RV_EVENT_KEY_RELEASE,
- RV_EVENT_BUTTON_PRESS,
- RV_EVENT_BUTTON_RELEASE,
+ RV_EVENT_BUTTON_PRESS,
+ RV_EVENT_BUTTON_RELEASE,
- //RV_EVENT_FOCUS,
- //RV_EVENT_DEFOCUS,
+ RV_EVENT_WINDOW_ENTER,
+ RV_EVENT_WINDOW_LEAVE,
} rv_window_event_type_t;
-typedef struct rv_window_event_t rv_window_event_t;
-struct rv_window_event_t {
- rv_window_event_t* next;
- rv_window_event_type_t type;
+struct rv_event_t {
+ rv_event_t* next;
+ rv_window_event_type_t type;
+
+ union {
+ struct _rv_key_press_releaase {
+ rv_keycode_t keycode;
+ rv_window_handle_t* window;
+ rv_str8 string;
+ u8 mods;
+ u8 group;
+ } key_press;
+ struct _rv_key_press_releaase key_release;
- union {
- rv_window_handle_t* window_close;
- };
+ rv_window_handle_t* window_resize;
+ rv_window_handle_t* window_close;
+ rv_window_handle_t* window_enter;
+ rv_window_handle_t* window_leave;
+ };
};
-RV_GLOBAL rv_window_event_t* rv_get_events(rv_arena* arena); // non-blocking
+typedef enum {
+ RV_EVENT_FLAGS_NONE = 0,
+} rv_event_flags_t;
+
+RV_GLOBAL rv_event_t* rv_get_events(rv_arena* arena, rv_event_flags_t flags); // non-blocking
// TODO(Samdal): blocking event function?
//////////////////////////////////////////////////////////////////
// Querying
+// if NULL, will get the last created window
RV_GLOBAL rv_vec2 rv_window_size(rv_window_handle_t* window);
+RV_GLOBAL rv_vec2 rv_mouse_pos(rv_window_handle_t* window);
//////////////////////////////////////////////////////////////////
// Rendering
+// if NULL, will get the last created window
RV_GLOBAL void rv_window_render_begin(rv_window_handle_t* window);
RV_GLOBAL void rv_window_render_end(rv_window_handle_t* window);
+
diff --git a/src/platform/gfx/xcb_impl.c b/src/platform/gfx/xcb_impl.c
@@ -2,14 +2,61 @@
// xcb_impl.c
#include <xcb/xcb.h>
+#include <xcb/xkb.h>
+
+#include "../../external/xkb-keysym-utf.c"
+
+#define RV_XKB_DEBUG 0
+
+
+//////////////////////////////////////////////////////////////////
+// Window
struct rv_window_handle_t {
- rv_window_handle_t* next;
- xcb_window_t window_id;
- xcb_intern_atom_reply_t* close_reply;
- void* render;
+ rv_window_handle_t* next;
+ xcb_window_t window_id;
+ rv_vec2 last_size;
+ xcb_intern_atom_reply_t* close_reply;
+ void* render;
};
+//////////////////////////////////////////////////////////////////
+// Context
+
+#define KEYSYM_LOOKUP_SIZE 6
+
+RV_INTERNAL struct {
+ bool32 is_init;
+ xcb_connection_t* connection;
+
+ rv_window_handle_t* window_list;
+
+
+ const xcb_query_extension_reply_t *xkb_extension_reply;
+
+ struct {
+ xcb_keysym_t keysyms[KEYSYM_LOOKUP_SIZE];
+ s32 num_groups;
+ u8 kt_index[4];
+ } keycode_converter_list[256];
+
+ u8 mods;
+ u8 group;
+
+ struct {
+ u32 mask;
+ struct {
+ bool32 active;
+ u32 mask;
+ s32 level;
+ } entries[64];
+ s32 entries_size;
+ } type_map[64];
+ s32 type_map_size;
+} xcb_context;
+
+#define xcb_maybe_window(w) ((w) ? (w) : (xcb_context.window_list))
+
//////////////////////////////////////////////////////////////////
// EGL
@@ -20,134 +67,134 @@ struct rv_window_handle_t {
#include <EGL/eglext.h>
RV_INTERNAL struct {
- bool32 is_init;
- EGLContext* egl_context;
- EGLDisplay egl_display;
- EGLConfig config;
+ bool32 is_init;
+ EGLContext* egl_context;
+ EGLDisplay egl_display;
+ EGLConfig config;
} egl_context;
RV_GLOBAL void* egl_attach(xcb_window_t window, xcb_connection_t* connection)
{
- { // make egl context
- if (!egl_context.is_init) {
- egl_context.is_init = true;
-
- { // egl_display
- PFNEGLGETPLATFORMDISPLAYEXTPROC GetPlatformDisplayEXT = (PFNEGLGETPLATFORMDISPLAYEXTPROC)eglGetProcAddress("eglGetPlatformDisplayEXT");
- // initialize EGL
- {
- egl_context.egl_display = GetPlatformDisplayEXT(EGL_PLATFORM_XCB_EXT, connection,
- (const EGLint[]){
- EGL_PLATFORM_XCB_SCREEN_EXT,
- 0, // This is a screenp that you can get from 2nd argument of xcb_connect
- EGL_NONE,
- });
- rv_assert(egl_context.egl_display != EGL_NO_DISPLAY);
-
- EGLint major, minor;
- if (!eglInitialize(egl_context.egl_display, &major, &minor))
- {
- rv_abort_msg(1, "Cannot initialize EGL display");
- }
- if (major < 1 || (major == 1 && minor < 5))
- {
- rv_abort_msg(1, "EGL version 1.5 or higher required");
- }
- }
- }
-
- // choose OpenGL API for EGL, by default it uses OpenGL ES
- EGLBoolean ok = eglBindAPI(EGL_OPENGL_API);
- rv_assert(ok);
-
- { // egl_context
- // choose EGL configuration
- {
- EGLint attr[] =
- {
- EGL_SURFACE_TYPE, EGL_WINDOW_BIT,
- EGL_CONFORMANT, EGL_OPENGL_BIT,
- EGL_RENDERABLE_TYPE, EGL_OPENGL_BIT,
- EGL_COLOR_BUFFER_TYPE, EGL_RGB_BUFFER,
-
- EGL_RED_SIZE, 8,
- EGL_GREEN_SIZE, 8,
- EGL_BLUE_SIZE, 8,
- EGL_DEPTH_SIZE, 24,
- EGL_STENCIL_SIZE, 8,
-
- // uncomment for multisampled framebuffer
- //EGL_SAMPLE_BUFFERS, 1,
- //EGL_SAMPLES, 4, // 4x MSAA
-
- EGL_NONE,
- };
-
- EGLint count;
- if (!eglChooseConfig(egl_context.egl_display, attr, &egl_context.config, 1, &count) || count != 1)
- {
- rv_abort_msg(1, "Cannot choose EGL config");
- }
- }
-
- {
- EGLint attr[] =
- {
- EGL_CONTEXT_MAJOR_VERSION, 3,
- EGL_CONTEXT_MINOR_VERSION, 3,
- EGL_CONTEXT_OPENGL_PROFILE_MASK, EGL_CONTEXT_OPENGL_CORE_PROFILE_BIT,
- // ask for debug context for non "Release" builds
- // this is so we can enable debug callback
- EGL_CONTEXT_OPENGL_DEBUG, EGL_TRUE,
- EGL_NONE,
- };
-
- egl_context.egl_context = eglCreateContext(egl_context.egl_display, egl_context.config, EGL_NO_CONTEXT, attr);
- if (egl_context.egl_context == EGL_NO_CONTEXT)
- {
- rv_abort_msg(1, "Cannot create EGL context, OpenGL 4.5 not supported?");
- }
- }
- }
- }
- }
-
- eglSwapInterval(egl_context.egl_display, 1);
-
- EGLint attr[] = {
- EGL_GL_COLORSPACE, EGL_GL_COLORSPACE_LINEAR, // or use EGL_GL_COLORSPACE_SRGB for sRGB framebuffer
- EGL_RENDER_BUFFER, EGL_BACK_BUFFER,
- EGL_NONE,
- };
+ { // make egl context
+ if (!egl_context.is_init) {
+ egl_context.is_init = true;
+
+ { // egl_display
+ PFNEGLGETPLATFORMDISPLAYEXTPROC GetPlatformDisplayEXT = (PFNEGLGETPLATFORMDISPLAYEXTPROC)eglGetProcAddress("eglGetPlatformDisplayEXT");
+ // initialize EGL
+ {
+ egl_context.egl_display = GetPlatformDisplayEXT(EGL_PLATFORM_XCB_EXT, connection,
+ (const EGLint[]){
+ EGL_PLATFORM_XCB_SCREEN_EXT,
+ 0, // This is a screenp that you can get from 2nd argument of xcb_connect
+ EGL_NONE,
+ });
+ rv_assert(egl_context.egl_display != EGL_NO_DISPLAY);
+
+ EGLint major, minor;
+ if (!eglInitialize(egl_context.egl_display, &major, &minor))
+ {
+ rv_abort_msg(1, "Cannot initialize EGL display");
+ }
+ if (major < 1 || (major == 1 && minor < 5))
+ {
+ rv_abort_msg(1, "EGL version 1.5 or higher required");
+ }
+ }
+ }
+
+ // choose OpenGL API for EGL, by default it uses OpenGL ES
+ EGLBoolean ok = eglBindAPI(EGL_OPENGL_API);
+ rv_assert(ok);
+
+ { // egl_context
+ // choose EGL configuration
+ {
+ EGLint attr[] =
+ {
+ EGL_SURFACE_TYPE, EGL_WINDOW_BIT,
+ EGL_CONFORMANT, EGL_OPENGL_BIT,
+ EGL_RENDERABLE_TYPE, EGL_OPENGL_BIT,
+ EGL_COLOR_BUFFER_TYPE, EGL_RGB_BUFFER,
+
+ EGL_RED_SIZE, 8,
+ EGL_GREEN_SIZE, 8,
+ EGL_BLUE_SIZE, 8,
+ EGL_DEPTH_SIZE, 24,
+ EGL_STENCIL_SIZE, 8,
+
+ // uncomment for multisampled framebuffer
+ //EGL_SAMPLE_BUFFERS, 1,
+ //EGL_SAMPLES, 4, // 4x MSAA
+
+ EGL_NONE,
+ };
+
+ EGLint count;
+ if (!eglChooseConfig(egl_context.egl_display, attr, &egl_context.config, 1, &count) || count != 1)
+ {
+ rv_abort_msg(1, "Cannot choose EGL config");
+ }
+ }
+
+ {
+ EGLint attr[] =
+ {
+ EGL_CONTEXT_MAJOR_VERSION, 3,
+ EGL_CONTEXT_MINOR_VERSION, 3,
+ EGL_CONTEXT_OPENGL_PROFILE_MASK, EGL_CONTEXT_OPENGL_CORE_PROFILE_BIT,
+ // ask for debug context for non "Release" builds
+ // this is so we can enable debug callback
+ EGL_CONTEXT_OPENGL_DEBUG, EGL_TRUE,
+ EGL_NONE,
+ };
+
+ egl_context.egl_context = eglCreateContext(egl_context.egl_display, egl_context.config, EGL_NO_CONTEXT, attr);
+ if (egl_context.egl_context == EGL_NO_CONTEXT)
+ {
+ rv_abort_msg(1, "Cannot create EGL context, OpenGL 4.5 not supported?");
+ }
+ }
+ }
+ }
+ }
+
+ eglSwapInterval(egl_context.egl_display, 1);
+
+ EGLint attr[] = {
+ EGL_GL_COLORSPACE, EGL_GL_COLORSPACE_LINEAR, // or use EGL_GL_COLORSPACE_SRGB for sRGB framebuffer
+ EGL_RENDER_BUFFER, EGL_BACK_BUFFER,
+ EGL_NONE,
+ };
EGLSurface* surface = eglCreateWindowSurface(egl_context.egl_display, egl_context.config, window, attr);
- if (surface == EGL_NO_SURFACE) {
- rv_abort_msg(1, "Cannot create EGL surface");
- }
-
- { // load opengl functions
- RV_LOCAL_PERSIST bool32 glad_loaded = false;
- if (!glad_loaded) {
- glad_loaded = true;
- eglMakeCurrent(egl_context.egl_display, surface, surface, egl_context.egl_context);
- gladLoadGLLoader((GLADloadproc)eglGetProcAddress);
- }
- }
-
- return surface;
+ if (surface == EGL_NO_SURFACE) {
+ rv_abort_msg(1, "Cannot create EGL surface");
+ }
+
+ { // load opengl functions
+ RV_LOCAL_PERSIST bool32 glad_loaded = false;
+ if (!glad_loaded) {
+ glad_loaded = true;
+ eglMakeCurrent(egl_context.egl_display, surface, surface, egl_context.egl_context);
+ gladLoadGLLoader((GLADloadproc)eglGetProcAddress);
+ }
+ }
+
+ return surface;
}
RV_GLOBAL void rv_window_render_begin(rv_window_handle_t* window)
{
- EGLSurface* surface = window->render;
- eglMakeCurrent(egl_context.egl_display, surface, surface, egl_context.egl_context);
+ EGLSurface* surface = xcb_maybe_window(window)->render;
+ eglMakeCurrent(egl_context.egl_display, surface, surface, egl_context.egl_context);
}
RV_GLOBAL void rv_window_render_end(rv_window_handle_t* window)
{
- EGLSurface* surface = window->render;
- eglSwapBuffers(egl_context.egl_display, surface);
+ EGLSurface* surface = xcb_maybe_window(window)->render;
+ eglSwapBuffers(egl_context.egl_display, surface);
}
@@ -155,195 +202,710 @@ RV_GLOBAL void rv_window_render_end(rv_window_handle_t* window)
//////////////////////////////////////////////////////////////////
-// Context
-
-RV_INTERNAL struct {
- bool32 is_init;
- xcb_connection_t* connection;
+// Context Creation
- rv_window_handle_t* window_list;
-} xcb_context;
+u32 mods_from_real_virtual(u32 rmod, u32 vmod_low, u32 vmod_high) {
+ return (rmod) | (vmod_low << 8) | (vmod_high << 16);
+}
+#define XKB_COMPONENTS_MASK \
+ (XCB_XKB_MAP_PART_KEY_TYPES | \
+ XCB_XKB_MAP_PART_KEY_SYMS )
RV_INTERNAL void make_xcb_context(void) {
- if (xcb_context.is_init) return;
- xcb_context.is_init = true;
+ if (xcb_context.is_init) return;
+ xcb_context.is_init = true;
+
+ xcb_context.connection = xcb_connect(NULL, NULL);
+
+ { // xkb
+
+ // NOTE(Samdal):
+ // we only set up the keyboard map once
+ // this does not support changing the keyboard layout during runtime...
+ // do we care?
+
+ // NOTE(Samdal):
+ // Sources
+ // https://gist.github.com/bluetech/6038239
+ // https://www.cl.cam.ac.uk/~mgk25/ucs/keysymdef.h
+ // https://www.x.org/docs/XKB/XKBlib.pdf
+ // https://tronche.com/gui/x/xlib/
+
+ { // set up xkb extension
+ {
+ const xcb_query_extension_reply_t *reply;
+
+ xcb_context.xkb_extension_reply = xcb_get_extension_data(xcb_context.connection, &xcb_xkb_id);
+ if (!xcb_context.xkb_extension_reply)
+ rv_abort_msg(1, "no XKB in X server");
+ }
+
+ {
+ xcb_xkb_use_extension_cookie_t cookie;
+ xcb_xkb_use_extension_reply_t *reply = NULL;
+
+ cookie = xcb_xkb_use_extension(xcb_context.connection,
+ XCB_XKB_MAJOR_VERSION,
+ XCB_XKB_MINOR_VERSION);
+ reply = xcb_xkb_use_extension_reply(xcb_context.connection, cookie, NULL);
+ if (!reply)
+ rv_abort_msg(1, "couldn't use XKB extension");
+
+ if (!reply->supported)
+ rv_abort_msg(1, "the XKB extension is not supported in X server");
+ }
+ }
+
+
+ xcb_xkb_get_map_reply_t *reply = NULL;
+
+ {
+ xcb_xkb_get_map_cookie_t cookie;
+ xcb_generic_error_t *error = NULL;
+
+ // Send the XkbGetMap request.
+ cookie = xcb_xkb_get_map(xcb_context.connection,
+ XCB_XKB_ID_USE_CORE_KBD,
+ XKB_COMPONENTS_MASK,
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
+ reply = xcb_xkb_get_map_reply(xcb_context.connection, cookie, &error);
+ if (!reply || error)
+ rv_abort_msg(1, "couldn't get get_map reply: error_code %d", error->error_code);
+
+ if ((reply->present & XKB_COMPONENTS_MASK) != XKB_COMPONENTS_MASK)
+ rv_abort_msg(1, "didn't get all components");
+
+ free(error);
+ }
+
+
+ // Get the map itself, with all the details.
+ xcb_xkb_get_map_map_t map;
+ {
+ void *buffer;
+
+ buffer = xcb_xkb_get_map_map(reply);
+ xcb_xkb_get_map_map_unpack(buffer,
+ reply->nTypes,
+ reply->nKeySyms,
+ reply->nKeyActions,
+ reply->totalActions,
+ reply->totalKeyBehaviors,
+ reply->nVModMapKeys,
+ reply->totalKeyExplicit,
+ reply->totalModMapKeys,
+ reply->totalVModMapKeys,
+ reply->present,
+ &map);
+ }
+
+ rv_mem_zero(xcb_context.keycode_converter_list, sizeof(xcb_context.keycode_converter_list));
+
+ // Get types
+ {
+ s32 length = xcb_xkb_get_map_map_types_rtrn_length(reply, &map);
+
+ rv_assert(length <= rv_array_size(xcb_context.type_map));
+ length = rv_min(length, rv_array_size(xcb_context.type_map));
+ xcb_context.type_map_size = length;
+
+ xcb_xkb_key_type_iterator_t iter = xcb_xkb_get_map_map_types_rtrn_iterator(reply, &map);
+
+ for (int i = 0; i < length; i++) {
+ xcb_xkb_key_type_t *type = iter.data;
+
+ xcb_context.type_map[i].mask = type->mods_mask;
+#if RV_XKB_DEBUG
+ printf("type (%d):\n", i);
+#endif
+
+ {
+ s32 length2 = xcb_xkb_key_type_map_length(type);
+
+ rv_assert(length2 <= rv_array_size(xcb_context.type_map[i].entries));
+ length2 = rv_min(length2, rv_array_size(xcb_context.type_map[i].entries));
+ xcb_context.type_map[i].entries_size = length2;
+
+ xcb_xkb_kt_map_entry_iterator_t iter2 = xcb_xkb_key_type_map_iterator(type);
+
+ for (int j = 0; j < length2; j++) {
+ xcb_xkb_kt_map_entry_t *entry = iter2.data;
+
+
+#if RV_XKB_DEBUG
+ printf("\tentry(%d) (active: %d | mask: %d | lvl: %d):\n", j,
+ entry->active,
+ entry->mods_mask,
+ entry->level
+ );
+#endif
+ xcb_context.type_map[i].entries[j].active = entry->active;
+ xcb_context.type_map[i].entries[j].mask = entry->mods_mask;
+ xcb_context.type_map[i].entries[j].level = entry->level;
+
+ xcb_xkb_kt_map_entry_next(&iter2);
+ }
+ }
+
+ xcb_xkb_key_type_next(&iter);
+ }
+ }
+
+ { // map keysyms
+ int length;
+ xcb_xkb_key_sym_map_iterator_t iter;
+
+ length = xcb_xkb_get_map_map_syms_rtrn_length(reply, &map);
+ iter = xcb_xkb_get_map_map_syms_rtrn_iterator(reply, &map);
+
+ rv_assert(length <= 256);
+ length = rv_min(length, 256);
+ for (int i = 0; i < length; i++) {
+ xcb_xkb_key_sym_map_t *sym_map = iter.data;
+
+ s32 num_groups = sym_map->groupInfo;
+ xcb_context.keycode_converter_list[i + reply->firstKeySym].num_groups = num_groups;
+ for (s32 g = 0; g < rv_min(4, num_groups); g++) {
+ xcb_context.keycode_converter_list[i + reply->firstKeySym].kt_index[g] = sym_map->kt_index[g];
+ }
+
+#if RV_XKB_DEBUG
+ printf("ksyms(%d) [%d]=>[%d, %d, %d, %d]:\n", i, sym_map->groupInfo,
+ sym_map->kt_index[0],
+ sym_map->kt_index[1],
+ sym_map->kt_index[2],
+ sym_map->kt_index[3]
+ );
+#endif
+
+ int length2;
+ xcb_keysym_t *syms;
+
+ length2 = xcb_xkb_key_sym_map_syms_length(sym_map);
+ syms = xcb_xkb_key_sym_map_syms(sym_map);
+
+ rv_assert(length2 <= KEYSYM_LOOKUP_SIZE);
+ length2 = rv_min(length2, KEYSYM_LOOKUP_SIZE);
+ for (int j = 0; j < length2; j++) {
+ xcb_keysym_t keysym = syms[j];
+
+#if RV_XKB_DEBUG
+ printf("\tsym(%d): 0x%x\n", j, keysym);
+#endif
+
+ xcb_context.keycode_converter_list[i + reply->firstKeySym].keysyms[j] = keysym;
+ }
+
+ xcb_xkb_key_sym_map_next(&iter);
+ }
+ }
+
+ { // add xkb events
+ xcb_void_cookie_t cookie = xcb_xkb_select_events(xcb_context.connection,
+ XCB_XKB_ID_USE_CORE_KBD,
+ XCB_XKB_EVENT_TYPE_STATE_NOTIFY,
+ 0,
+ XCB_XKB_EVENT_TYPE_STATE_NOTIFY,
+ 0xff,
+ 0xff,
+ NULL);
+ }
+
+ {
+ xcb_xkb_get_state_cookie_t cookie = xcb_xkb_get_state(xcb_context.connection, XCB_XKB_ID_USE_CORE_KBD);
+ xcb_xkb_get_state_reply_t* reply = xcb_xkb_get_state_reply(xcb_context.connection, cookie, NULL);
+
+ xcb_context.mods = reply->mods;
+ xcb_context.group = reply->group;
+
+ free(reply);
+ }
+
+ }
- xcb_context.connection = xcb_connect(NULL, NULL);
}
+RV_INTERNAL xcb_keysym_t keycode_to_keysym(u8 keycode, u32 mods)
+{
+ s32 level = 0;
+ { // look up key level
+
+ s32 group = rv_min(xcb_context.keycode_converter_list[keycode].num_groups, xcb_context.group);
+
+ if (xcb_context.keycode_converter_list[keycode].num_groups) {
+ s32 key_type = xcb_context.keycode_converter_list[keycode].kt_index[group];
+
+ mods &= xcb_context.type_map[key_type].mask;
+ for (s32 e = 0; e < xcb_context.type_map[key_type].entries_size; e++) {
+ if (xcb_context.type_map[key_type].entries[e].active &&
+ xcb_context.type_map[key_type].entries[e].mask == mods) {
+ level = xcb_context.type_map[key_type].entries[e].level;
+ break;
+ }
+ }
+ }
+ }
+
+ return xcb_context.keycode_converter_list[keycode].keysyms[level];
+}
//////////////////////////////////////////////////////////////////
// window creation/destruction
RV_GLOBAL rv_window_handle_t* rv_create_window(rv_window_desc_t desc)
{
- make_xcb_context();
-
- rv_window_handle_t* res = rv_mem_global_alloc(rv_window_handle_t, 1);
-
- xcb_screen_t *screen;
- xcb_gcontext_t foreground;
- xcb_gcontext_t background;
- xcb_generic_event_t *e;
- uint32_t mask = 0;
- uint32_t values[2];
-
- // get the first screen
- screen = xcb_setup_roots_iterator(xcb_get_setup(xcb_context.connection)).data;
-
- // root window
- xcb_drawable_t root_window = screen->root;
-
- // create black (foreground) graphic context
- foreground = xcb_generate_id(xcb_context.connection);
- mask = XCB_GC_FOREGROUND | XCB_GC_GRAPHICS_EXPOSURES;
- values[0] = screen->black_pixel;
- values[1] = 0;
- xcb_create_gc(xcb_context.connection, foreground, root_window, mask, values);
-
- // create white (background) graphic context
- background = xcb_generate_id (xcb_context.connection);
- mask = XCB_GC_BACKGROUND | XCB_GC_GRAPHICS_EXPOSURES;
- values[0] = screen->white_pixel;
- values[1] = 0;
- xcb_create_gc(xcb_context.connection, background, root_window, mask, values);
-
- // create the window
- res->window_id = xcb_generate_id(xcb_context.connection);
-
- mask = XCB_CW_BACK_PIXEL | XCB_CW_EVENT_MASK;
- values[0] = screen->white_pixel;
-
- values[1] = XCB_EVENT_MASK_EXPOSURE | XCB_EVENT_MASK_STRUCTURE_NOTIFY |
- XCB_EVENT_MASK_KEY_PRESS | XCB_EVENT_MASK_KEY_RELEASE |
- XCB_EVENT_MASK_BUTTON_PRESS | XCB_EVENT_MASK_BUTTON_RELEASE |
- XCB_EVENT_MASK_ENTER_WINDOW | XCB_EVENT_MASK_LEAVE_WINDOW ;
-
- xcb_create_window(xcb_context.connection, // xcb_context.connection
- XCB_COPY_FROM_PARENT, // depth
- res->window_id, // window Id
- screen->root, // parent window
- 0, 0, // x, y
- 150, 150, // width, height
- 10, // border_width
- XCB_WINDOW_CLASS_INPUT_OUTPUT, // class
- screen->root_visual, // visual
- mask, values); // masks
-
- // map the window on the screen
- xcb_map_window(xcb_context.connection, res->window_id);
-
- { // change window title
- xcb_intern_atom_cookie_t utf8_string_cookie = xcb_intern_atom(xcb_context.connection, 0, 11, "UTF8_STRING");
- xcb_intern_atom_reply_t* utf8_string_reply = xcb_intern_atom_reply(xcb_context.connection, utf8_string_cookie, 0);
- xcb_atom_t m_utf8_string_atom = utf8_string_reply->atom;
- free(utf8_string_reply);
-
- xcb_change_property(xcb_context.connection, XCB_PROP_MODE_REPLACE, res->window_id,
- XCB_ATOM_WM_NAME, m_utf8_string_atom, 8,
- strlen(desc.name), desc.name);
- }
-
- { // setup close handler event
- xcb_intern_atom_cookie_t protocolCookie = xcb_intern_atom_unchecked(xcb_context.connection, 1, 12, "WM_PROTOCOLS");
- xcb_intern_atom_reply_t* protocolReply = xcb_intern_atom_reply(xcb_context.connection, protocolCookie, 0);
- xcb_intern_atom_cookie_t closeCookie = xcb_intern_atom_unchecked(xcb_context.connection, 0, 16, "WM_DELETE_WINDOW");
- res->close_reply = xcb_intern_atom_reply(xcb_context.connection, closeCookie, 0);
- xcb_change_property(xcb_context.connection, XCB_PROP_MODE_REPLACE, res->window_id, protocolReply->atom, 4, 32, 1, &(res->close_reply->atom));
-
- free(protocolReply);
- }
-
- if (desc.attach_render) {
+ make_xcb_context();
+
+ rv_window_handle_t* res = rv_mem_global_alloc(rv_window_handle_t, 1);
+
+ { // create the window
+
+ // get the first screen
+ xcb_screen_t* screen = xcb_setup_roots_iterator(xcb_get_setup(xcb_context.connection)).data;
+
+ // set up masks
+ uint32_t mask = 0;
+ uint32_t values[2];
+ res->window_id = xcb_generate_id(xcb_context.connection);
+
+ mask = XCB_CW_BACK_PIXEL | XCB_CW_EVENT_MASK;
+
+ values[1] = XCB_EVENT_MASK_EXPOSURE | XCB_EVENT_MASK_STRUCTURE_NOTIFY |
+ XCB_EVENT_MASK_KEY_PRESS | XCB_EVENT_MASK_KEY_RELEASE |
+ XCB_EVENT_MASK_BUTTON_PRESS | XCB_EVENT_MASK_BUTTON_RELEASE |
+ XCB_EVENT_MASK_ENTER_WINDOW | XCB_EVENT_MASK_LEAVE_WINDOW ;
+
+ // default window size
+ rv_vec2 size = desc.rect.wh;
+ if (size.x == 0) size.x = 150;
+ if (size.y == 0) size.y = 150;
+
+ xcb_create_window(xcb_context.connection, // xcb_context.connection
+ XCB_COPY_FROM_PARENT, // depth
+ res->window_id, // window Id
+ screen->root, // parent window
+ desc.rect.x, desc.rect.y, // x, y
+ size.x, size.y, // width, height
+ 10, // border_width
+ XCB_WINDOW_CLASS_INPUT_OUTPUT, // class
+ screen->root_visual, // visual
+ mask, values); // masks
+
+ // map the window on the screen
+ xcb_map_window(xcb_context.connection, res->window_id);
+ }
+
+ { // change window title
+ xcb_intern_atom_cookie_t utf8_string_cookie = xcb_intern_atom(xcb_context.connection, 0, 11, "UTF8_STRING");
+ xcb_intern_atom_reply_t* utf8_string_reply = xcb_intern_atom_reply(xcb_context.connection, utf8_string_cookie, 0);
+ xcb_atom_t m_utf8_string_atom = utf8_string_reply->atom;
+ free(utf8_string_reply);
+
+ xcb_change_property(xcb_context.connection, XCB_PROP_MODE_REPLACE, res->window_id,
+ XCB_ATOM_WM_NAME, m_utf8_string_atom, 8,
+ desc.name.size, desc.name.str);
+ }
+
+ { // setup close handler event
+ xcb_intern_atom_cookie_t protocolCookie = xcb_intern_atom_unchecked(xcb_context.connection, 1, 12, "WM_PROTOCOLS");
+ xcb_intern_atom_reply_t* protocolReply = xcb_intern_atom_reply(xcb_context.connection, protocolCookie, 0);
+ xcb_intern_atom_cookie_t closeCookie = xcb_intern_atom_unchecked(xcb_context.connection, 0, 16, "WM_DELETE_WINDOW");
+ res->close_reply = xcb_intern_atom_reply(xcb_context.connection, closeCookie, 0);
+ xcb_change_property(xcb_context.connection, XCB_PROP_MODE_REPLACE, res->window_id, protocolReply->atom, 4, 32, 1, &(res->close_reply->atom));
+
+ free(protocolReply);
+ }
+
+ if (desc.attach_render) {
#if RV_RENDER_OPENGL
- res->render = egl_attach(res->window_id, xcb_context.connection);
+ res->render = egl_attach(res->window_id, xcb_context.connection);
#endif
- }
-
+ }
- xcb_flush(xcb_context.connection);
+ xcb_flush(xcb_context.connection);
- while ((e = xcb_wait_for_event(xcb_context.connection))) {
- bool32 done = false;
+ xcb_generic_event_t *e;
+ while ((e = xcb_wait_for_event(xcb_context.connection))) {
+ bool32 done = false;
- if ((e->response_type & ~0x80) == XCB_EXPOSE) {
- done = true;
- }
+ if ((e->response_type & ~0x80) == XCB_EXPOSE) {
+ done = true;
+ }
- free(e);
+ free(e);
- if (done) break;
- }
+ if (done) break;
+ }
- rv_llist_stack_push(xcb_context.window_list, res);
+ rv_llist_stack_push(xcb_context.window_list, res);
- return res;
+ return res;
}
RV_GLOBAL void rv_destroy_window(rv_window_handle_t** handle)
{
- rv_abort_msg(1, "need to implement removing windows from the window list!");
+ if (!*handle) {
+ rv_assert(false);
+ return;
+ }
- if (!*handle) {
- rv_assert(*handle);
- return;
- }
+#if RV_RENDER_OPENGL
+ {
+ eglMakeCurrent(egl_context.egl_display, NULL, NULL, egl_context.egl_context);
+ EGLSurface* surface = handle[0]->render;
+ eglDestroySurface(egl_context.egl_display, surface);
+ }
+#endif
+
+ free(handle[0]->close_reply);
+
+ xcb_destroy_window(xcb_context.connection, handle[0]->window_id);
- xcb_destroy_window(xcb_context.connection, handle[0]->window_id);
+ xcb_flush(xcb_context.connection);
+
+ for (rv_window_handle_t** window_iter = &xcb_context.window_list; *window_iter; window_iter = &window_iter[0]->next) {
+ if (*window_iter == *handle) {
+ *window_iter = handle[0]->next;
+ puts("remove");
+ break;
+ }
+ }
+
+ rv_mem_global_alloc_free(*handle, sizeof(**handle));
+ *handle = 0;
+}
- rv_mem_global_alloc_free(*handle, sizeof(**handle));
- *handle = 0;
+rv_window_handle_t* xcb_get_window(xcb_window_t window) {
+ for (rv_window_handle_t* w = xcb_context.window_list; w; w = w->next) {
+ if (window == w->window_id) {
+ return w;
+ }
+ }
+ return NULL;
}
//////////////////////////////////////////////////////////////////
// event polling
-RV_GLOBAL rv_window_event_t* rv_get_events(rv_arena* arena)
+RV_GLOBAL rv_event_t* rv_get_events(rv_arena* arena, rv_event_flags_t flags)
{
- rv_window_event_t* res = NULL;
+ rv_event_t* res = NULL;
xcb_generic_event_t *event;
- while ((event = xcb_poll_for_event(xcb_context.connection))) {
-
- rv_window_event_t new_ev = {0};
-
- switch (event->response_type & ~0x80) {
- case XCB_EXPOSE: {
- xcb_flush(xcb_context.connection);
- } break;
-
- case XCB_CLIENT_MESSAGE: {
- for (rv_window_handle_t* w = xcb_context.window_list; w; w = w->next) {
- if (((xcb_client_message_event_t*)event)->data.data32[0] == w->close_reply->atom) {
- new_ev.type = RV_EVENT_WINDOW_CLOSE;
- new_ev.window_close = w;
- break;
- }
- }
- } break;
-
- // TODO: handle closing
-
- case XCB_KEY_PRESS: { new_ev.type = RV_EVENT_KEY_PRESS; } break;
- case XCB_KEY_RELEASE: { new_ev.type = RV_EVENT_KEY_RELEASE; } break;
- case XCB_BUTTON_PRESS: { new_ev.type = RV_EVENT_BUTTON_PRESS; } break;
- case XCB_BUTTON_RELEASE: { new_ev.type = RV_EVENT_BUTTON_RELEASE; } break;
- case XCB_CONFIGURE_NOTIFY: { new_ev.type = RV_EVENT_WINDOW_CHANGE; } break;
-
- default: { new_ev.type = RV_EVENT_INVALID; } break;
- }
-
- if (new_ev.type != RV_EVENT_INVALID) {
- rv_window_event_t* n = rv_push_no_zero(arena, rv_window_event_t);
- *n = new_ev;
- rv_llist_stack_push(res, n);
- }
-
- free(event);
- }
-
- return res;
+ while ((event = xcb_poll_for_event(xcb_context.connection))) {
+
+ rv_event_t new_ev = {0};
+
+ bool32 key_is_press = false;
+
+ switch (event->response_type & ~0x80) {
+ case XCB_EXPOSE: {
+ xcb_flush(xcb_context.connection);
+ } break;
+
+ case XCB_CLIENT_MESSAGE: {
+ for (rv_window_handle_t* w = xcb_context.window_list; w; w = w->next) {
+ if (((xcb_client_message_event_t*)event)->data.data32[0] == w->close_reply->atom) {
+ new_ev.type = RV_EVENT_WINDOW_CLOSE;
+ new_ev.window_close = w;
+ break;
+ }
+ }
+ } break;
+
+ // TODO: handle closing
+
+ case XCB_KEY_PRESS: {
+ _Static_assert(sizeof(xcb_key_release_event_t) == sizeof(xcb_key_press_event_t), "press/release were not the same");
+ key_is_press = true;
+ case XCB_KEY_RELEASE:;
+
+ if (key_is_press) new_ev.type = RV_EVENT_KEY_PRESS;
+ else new_ev.type = RV_EVENT_KEY_RELEASE;
+
+ xcb_key_press_event_t* key_event = (xcb_key_press_event_t*)event;
+
+ u8 keycode = key_event->detail;
+
+ new_ev.key_press.window = xcb_get_window(key_event->event);
+
+ xcb_keysym_t keysym = keycode_to_keysym(keycode, xcb_context.mods);
+
+ // TODO(Samdal): look up conversion map for multi-keys and dead keys
+ { // convert (some) dead keys to real keys...
+ switch (keysym) {
+ case 0xfe50: { keysym = 0x0060; } break;
+ case 0xfe51: { keysym = 0x00b4; } break;
+ case 0xfe57: { keysym = 0x00a8; } break;
+ case 0xfe52: { keysym = 0x005e; } break;
+ case 0xfe53: { keysym = 0x007e; } break;
+ }
+ }
+
+ rv_keycode_t kc = RV_KEYCODE_INVALID;
+ { // convert to RV keycode
+ switch (keysym) {
+ case 0x0020: { kc = RV_KEYCODE_SPACE; } break;
+ case 0x0027: { kc = RV_KEYCODE_APOSTROPHE; } break;
+ case 0x002c: { kc = RV_KEYCODE_COMMA; } break;
+ case 0x002d: { kc = RV_KEYCODE_MINUS; } break;
+ case 0x002e: { kc = RV_KEYCODE_PERIOD; } break;
+ case 0x002f: { kc = RV_KEYCODE_SLASH; } break;
+ case 0x0030: { kc = RV_KEYCODE_0; } break;
+ case 0x0031: { kc = RV_KEYCODE_1; } break;
+ case 0x0032: { kc = RV_KEYCODE_2; } break;
+ case 0x0033: { kc = RV_KEYCODE_3; } break;
+ case 0x0034: { kc = RV_KEYCODE_4; } break;
+ case 0x0035: { kc = RV_KEYCODE_5; } break;
+ case 0x0036: { kc = RV_KEYCODE_6; } break;
+ case 0x0037: { kc = RV_KEYCODE_7; } break;
+ case 0x0038: { kc = RV_KEYCODE_8; } break;
+ case 0x0039: { kc = RV_KEYCODE_9; } break;
+ case 0x003b: { kc = RV_KEYCODE_SEMICOLON; } break;
+ case 0x003d: { kc = RV_KEYCODE_EQUAL; } break;
+
+ // capital letters
+ case 0x0041: { kc = RV_KEYCODE_A; } break;
+ case 0x0042: { kc = RV_KEYCODE_B; } break;
+ case 0x0043: { kc = RV_KEYCODE_C; } break;
+ case 0x0044: { kc = RV_KEYCODE_D; } break;
+ case 0x0045: { kc = RV_KEYCODE_E; } break;
+ case 0x0046: { kc = RV_KEYCODE_F; } break;
+ case 0x0047: { kc = RV_KEYCODE_G; } break;
+ case 0x0048: { kc = RV_KEYCODE_H; } break;
+ case 0x0049: { kc = RV_KEYCODE_I; } break;
+ case 0x004a: { kc = RV_KEYCODE_J; } break;
+ case 0x004b: { kc = RV_KEYCODE_K; } break;
+ case 0x004c: { kc = RV_KEYCODE_L; } break;
+ case 0x004d: { kc = RV_KEYCODE_M; } break;
+ case 0x004e: { kc = RV_KEYCODE_N; } break;
+ case 0x004f: { kc = RV_KEYCODE_O; } break;
+ case 0x0050: { kc = RV_KEYCODE_P; } break;
+ case 0x0051: { kc = RV_KEYCODE_Q; } break;
+ case 0x0052: { kc = RV_KEYCODE_R; } break;
+ case 0x0053: { kc = RV_KEYCODE_S; } break;
+ case 0x0054: { kc = RV_KEYCODE_T; } break;
+ case 0x0055: { kc = RV_KEYCODE_U; } break;
+ case 0x0056: { kc = RV_KEYCODE_V; } break;
+ case 0x0057: { kc = RV_KEYCODE_W; } break;
+ case 0x0058: { kc = RV_KEYCODE_X; } break;
+ case 0x0059: { kc = RV_KEYCODE_Y; } break;
+ case 0x005a: { kc = RV_KEYCODE_Z; } break;
+
+ // small letters
+ case 0x0061: { kc = RV_KEYCODE_A; } break;
+ case 0x0062: { kc = RV_KEYCODE_B; } break;
+ case 0x0063: { kc = RV_KEYCODE_C; } break;
+ case 0x0064: { kc = RV_KEYCODE_D; } break;
+ case 0x0065: { kc = RV_KEYCODE_E; } break;
+ case 0x0066: { kc = RV_KEYCODE_F; } break;
+ case 0x0067: { kc = RV_KEYCODE_G; } break;
+ case 0x0068: { kc = RV_KEYCODE_H; } break;
+ case 0x0069: { kc = RV_KEYCODE_I; } break;
+ case 0x006a: { kc = RV_KEYCODE_J; } break;
+ case 0x006b: { kc = RV_KEYCODE_K; } break;
+ case 0x006c: { kc = RV_KEYCODE_L; } break;
+ case 0x006d: { kc = RV_KEYCODE_M; } break;
+ case 0x006e: { kc = RV_KEYCODE_N; } break;
+ case 0x006f: { kc = RV_KEYCODE_O; } break;
+ case 0x0070: { kc = RV_KEYCODE_P; } break;
+ case 0x0071: { kc = RV_KEYCODE_Q; } break;
+ case 0x0072: { kc = RV_KEYCODE_R; } break;
+ case 0x0073: { kc = RV_KEYCODE_S; } break;
+ case 0x0074: { kc = RV_KEYCODE_T; } break;
+ case 0x0075: { kc = RV_KEYCODE_U; } break;
+ case 0x0076: { kc = RV_KEYCODE_V; } break;
+ case 0x0077: { kc = RV_KEYCODE_W; } break;
+ case 0x0078: { kc = RV_KEYCODE_X; } break;
+ case 0x0079: { kc = RV_KEYCODE_Y; } break;
+ case 0x007a: { kc = RV_KEYCODE_Z; } break;
+
+ case 0x007c: { kc = RV_KEYCODE_BAR; } break;
+ case 0x005b: { kc = RV_KEYCODE_LEFT_BRACKET; } break;
+ case 0x005c: { kc = RV_KEYCODE_BACKSLASH; } break;
+ case 0x005d: { kc = RV_KEYCODE_RIGHT_BRACKET; } break;
+ case 0x0060: { kc = RV_KEYCODE_GRAVE_ACCENT; } break; // ascii grave
+ case 0x00b4: { kc = RV_KEYCODE_ACUTE_ACCENT; } break; // ascii acute
+ case 0x00a8: { kc = RV_KEYCODE_DIAERESIS; } break; // ascii diaeresis
+ case 0x005e: { kc = RV_KEYCODE_CIRCUMFLEX; } break; // ascii circumflex
+ case 0x007e: { kc = RV_KEYCODE_TILDE; } break; // ascii tilde
+ //case 0x0000: { kc = RV_KEYCODE_WORLD_1; } break;
+ //case 0x0000: { kc = RV_KEYCODE_WORLD_2; } break;
+ case 0xff1b: { kc = RV_KEYCODE_ESC; } break;
+ case 0xff0d: { kc = RV_KEYCODE_ENTER; } break;
+ case 0xff09: { kc = RV_KEYCODE_TAB; } break;
+ case 0xff08: { kc = RV_KEYCODE_BACKSPACE; } break;
+
+ // nav cluster
+ case 0xff63: { kc = RV_KEYCODE_INSERT; } break;
+ case 0xffff: { kc = RV_KEYCODE_DELETE; } break;
+ case 0xff53: { kc = RV_KEYCODE_RIGHT; } break;
+ case 0xff51: { kc = RV_KEYCODE_LEFT; } break;
+ case 0xff54: { kc = RV_KEYCODE_DOWN; } break;
+ case 0xff52: { kc = RV_KEYCODE_UP; } break;
+ case 0xff55: { kc = RV_KEYCODE_PAGE_UP; } break;
+ case 0xff56: { kc = RV_KEYCODE_PAGE_DOWN; } break;
+ case 0xff50: { kc = RV_KEYCODE_HOME; } break;
+ case 0xff57: { kc = RV_KEYCODE_END; } break;
+ // non-numlock nav
+ case 0xff9e: { kc = RV_KEYCODE_INSERT; } break;
+ case 0xff9f: { kc = RV_KEYCODE_DELETE; } break;
+ case 0xff98: { kc = RV_KEYCODE_RIGHT; } break;
+ case 0xff96: { kc = RV_KEYCODE_LEFT; } break;
+ case 0xff99: { kc = RV_KEYCODE_DOWN; } break;
+ case 0xff97: { kc = RV_KEYCODE_UP; } break;
+ case 0xff9a: { kc = RV_KEYCODE_PAGE_UP; } break;
+ case 0xff9b: { kc = RV_KEYCODE_PAGE_DOWN; } break;
+ case 0xff95: { kc = RV_KEYCODE_HOME; } break;
+ case 0xff9c: { kc = RV_KEYCODE_END; } break;
+
+ case 0xffe5: { kc = RV_KEYCODE_CAPS_LOCK; } break;
+ case 0xff14: { kc = RV_KEYCODE_SCROLL_LOCK; } break;
+ case 0xff7f: { kc = RV_KEYCODE_NUM_LOCK; } break;
+ case 0xff61: { kc = RV_KEYCODE_PRINT_SCREEN; } break;
+ case 0xff13: { kc = RV_KEYCODE_PAUSE; } break;
+ case 0xffbe: { kc = RV_KEYCODE_F1; } break;
+ case 0xffbf: { kc = RV_KEYCODE_F2; } break;
+ case 0xffc0: { kc = RV_KEYCODE_F3; } break;
+ case 0xffc1: { kc = RV_KEYCODE_F4; } break;
+ case 0xffc2: { kc = RV_KEYCODE_F5; } break;
+ case 0xffc3: { kc = RV_KEYCODE_F6; } break;
+ case 0xffc4: { kc = RV_KEYCODE_F7; } break;
+ case 0xffc5: { kc = RV_KEYCODE_F8; } break;
+ case 0xffc6: { kc = RV_KEYCODE_F9; } break;
+ case 0xffc7: { kc = RV_KEYCODE_F10; } break;
+ case 0xffc8: { kc = RV_KEYCODE_F11; } break;
+ case 0xffc9: { kc = RV_KEYCODE_F12; } break;
+ case 0xffca: { kc = RV_KEYCODE_F13; } break;
+ case 0xffcb: { kc = RV_KEYCODE_F14; } break;
+ case 0xffcc: { kc = RV_KEYCODE_F15; } break;
+ case 0xffcd: { kc = RV_KEYCODE_F16; } break;
+ case 0xffce: { kc = RV_KEYCODE_F17; } break;
+ case 0xffcf: { kc = RV_KEYCODE_F18; } break;
+ case 0xffd0: { kc = RV_KEYCODE_F19; } break;
+ case 0xffd1: { kc = RV_KEYCODE_F20; } break;
+ case 0xffd2: { kc = RV_KEYCODE_F21; } break;
+ case 0xffd3: { kc = RV_KEYCODE_F22; } break;
+ case 0xffd4: { kc = RV_KEYCODE_F23; } break;
+ case 0xffd5: { kc = RV_KEYCODE_F24; } break;
+ case 0xffd6: { kc = RV_KEYCODE_F25; } break;
+ case 0xffb0: { kc = RV_KEYCODE_KP_0; } break;
+ case 0xffb1: { kc = RV_KEYCODE_KP_1; } break;
+ case 0xffb2: { kc = RV_KEYCODE_KP_2; } break;
+ case 0xffb3: { kc = RV_KEYCODE_KP_3; } break;
+ case 0xffb4: { kc = RV_KEYCODE_KP_4; } break;
+ case 0xffb5: { kc = RV_KEYCODE_KP_5; } break;
+ case 0xffb6: { kc = RV_KEYCODE_KP_6; } break;
+ case 0xffb7: { kc = RV_KEYCODE_KP_7; } break;
+ case 0xffb8: { kc = RV_KEYCODE_KP_8; } break;
+ case 0xffb9: { kc = RV_KEYCODE_KP_9; } break;
+ case 0xffae: { kc = RV_KEYCODE_KP_DECIMAL; } break;
+ case 0xffaf: { kc = RV_KEYCODE_KP_DIVIDE; } break;
+ case 0xffaa: { kc = RV_KEYCODE_KP_MULTIPLY; } break;
+ case 0xffad: { kc = RV_KEYCODE_KP_SUBTRACT; } break;
+ case 0xffab: { kc = RV_KEYCODE_KP_ADD; } break;
+ case 0xff8d: { kc = RV_KEYCODE_KP_ENTER; } break; // same as normal enter...
+ case 0xffbd: { kc = RV_KEYCODE_KP_EQUAL; } break;
+ case 0xffe1: { kc = RV_KEYCODE_LEFT_SHIFT; } break;
+ case 0xffe3: { kc = RV_KEYCODE_LEFT_CONTROL; } break;
+ case 0xffe9: { kc = RV_KEYCODE_LEFT_ALT; } break;
+ case 0xffeb: { kc = RV_KEYCODE_LEFT_SUPER; } break;
+ case 0xffe2: { kc = RV_KEYCODE_RIGHT_SHIFT; } break;
+ case 0xffe4: { kc = RV_KEYCODE_RIGHT_CONTROL; } break;
+ case 0xffea: { kc = RV_KEYCODE_RIGHT_ALT; } break;
+ case 0xfe03: { kc = RV_KEYCODE_RIGHT_ALT; } break; // ISO level 3 shift
+ case 0xffec: { kc = RV_KEYCODE_RIGHT_SUPER; } break;
+ case 0xff67: { kc = RV_KEYCODE_MENU; } break;
+ default: { kc = RV_KEYCODE_INVALID; } break;
+ }
+ }
+ new_ev.key_press.keycode = kc;
+ new_ev.key_press.mods = xcb_context.mods;
+ new_ev.key_press.group = xcb_context.group;
+
+ {
+ char buf[XKB_KEYSYM_UTF8_MAX_SIZE];
+ s32 len = xkb_keysym_to_utf8(keysym, buf, sizeof(buf));
+ if (len) {
+ new_ev.key_press.string = rv_str8_from_cstr(buf);
+ new_ev.key_press.string = rv_str8_copy(arena, new_ev.key_press.string);
+ if (new_ev.key_press.string.size == 1) {
+ if (new_ev.key_press.string.str[0] == '\r') {
+ new_ev.key_press.string.str[0] = '\n';
+ }
+ }
+ }
+ }
+
+ // invalidate event if nothing gave any results
+ if (new_ev.key_press.keycode == RV_KEYCODE_INVALID && new_ev.key_press.string.size == 0) {
+ new_ev.type = RV_EVENT_INVALID;
+ }
+ } break;
+ case XCB_BUTTON_PRESS: { new_ev.type = RV_EVENT_BUTTON_PRESS; } break;
+ case XCB_BUTTON_RELEASE: { new_ev.type = RV_EVENT_BUTTON_RELEASE; } break;
+ case XCB_CONFIGURE_NOTIFY: {
+ xcb_configure_notify_event_t* conf_event = (xcb_configure_notify_event_t*)event;
+ new_ev.window_resize = xcb_get_window(conf_event->event);
+ if (new_ev.window_resize) {
+ if (new_ev.window_resize->last_size.x != conf_event->width || new_ev.window_resize->last_size.y != conf_event->height) {
+ new_ev.type = RV_EVENT_WINDOW_RESIZE;
+ new_ev.window_resize->last_size = rv_v2(conf_event->width, conf_event->height);
+ }
+ }
+ } break;
+ case XCB_ENTER_NOTIFY: {
+ xcb_enter_notify_event_t* enter_event = (xcb_leave_notify_event_t*)event;
+ new_ev.type = RV_EVENT_WINDOW_ENTER;
+ new_ev.window_enter = xcb_get_window(enter_event->event);
+ } break;
+ case XCB_LEAVE_NOTIFY: {
+ xcb_leave_notify_event_t* leave_event = (xcb_leave_notify_event_t*)event;
+ new_ev.type = RV_EVENT_WINDOW_LEAVE;
+ new_ev.window_leave = xcb_get_window(leave_event->event);
+ } break;
+
+ default: {
+ if (xcb_context.xkb_extension_reply->first_event == (event->response_type & ~0x80)) {
+
+ xcb_xkb_state_notify_event_t *state = (xcb_xkb_state_notify_event_t *)event;
+ xcb_context.mods = state->mods;
+ xcb_context.group = state->group;
+
+ } else {
+ printf("unhandled event: %d\n", event->response_type & ~0x80);
+ }
+ } break;
+ }
+
+ if (new_ev.type != RV_EVENT_INVALID) {
+ rv_event_t* n = rv_push_no_zero(arena, rv_event_t);
+ *n = new_ev;
+ rv_llist_stack_push(res, n);
+ }
+
+ free(event);
+ }
+
+ // make a resize event if it's not up to date
+ for (rv_window_handle_t* w = xcb_context.window_list; w; w = w->next) {
+ rv_event_t new_ev = {0};
+ new_ev.window_resize = w;
+
+ rv_vec2 size = rv_window_size(w);
+ if (new_ev.window_resize->last_size.x != size.x || new_ev.window_resize->last_size.y != size.y) {
+ new_ev.type = RV_EVENT_WINDOW_RESIZE;
+ new_ev.window_resize->last_size = rv_v2(size.x, size.y);
+
+ rv_event_t* n = rv_push_copy(arena, rv_event_t, &new_ev);
+ rv_llist_stack_push(res, n);
+ }
+ }
+
+ rv_keyboard_context_update(&rv_keyboard_context, res);
+
+ return res;
}
//////////////////////////////////////////////////////////////////
@@ -351,20 +913,30 @@ RV_GLOBAL rv_window_event_t* rv_get_events(rv_arena* arena)
RV_GLOBAL rv_vec2 rv_window_size(rv_window_handle_t* window)
{
- rv_vec2 ret = {0};
- xcb_get_geometry_cookie_t geom_cookie = xcb_get_geometry(xcb_context.connection, window->window_id);
- xcb_get_geometry_reply_t* reply = xcb_get_geometry_reply(xcb_context.connection, geom_cookie, NULL);
- if (reply) {
- ret = rv_v2(
- reply->width,
- reply->height,
- );
- free(reply);
- }
- return ret;
+ rv_vec2 ret = {0};
+ xcb_get_geometry_cookie_t geom_cookie = xcb_get_geometry(xcb_context.connection, xcb_maybe_window(window)->window_id);
+ xcb_get_geometry_reply_t* reply = xcb_get_geometry_reply(xcb_context.connection, geom_cookie, NULL);
+ if (reply) {
+ ret = rv_v2(
+ reply->width,
+ reply->height,
+ );
+ free(reply);
+ }
+ return ret;
}
-RV_GLOBAL void* rv_window_render(rv_window_handle_t* window)
+RV_GLOBAL rv_vec2 rv_mouse_pos(rv_window_handle_t* window)
{
- return window->render;
+ rv_vec2 ret = {0};
+ xcb_query_pointer_cookie_t cookie = xcb_query_pointer(xcb_context.connection, xcb_maybe_window(window)->window_id);
+ xcb_query_pointer_reply_t* reply = xcb_query_pointer_reply(xcb_context.connection, cookie, NULL);
+ if (reply) {
+ ret = rv_v2(
+ reply->win_x,
+ reply->win_y,
+ );
+ free(reply);
+ }
+ return ret;
}
diff --git a/src/platform/platform_arena.c b/src/platform/platform_arena.c
@@ -5,53 +5,53 @@
RV_INTERNAL rv_arena*
rv_arena_alloc_(rv_arena_params* params) {
- // round up reserve/commit sizes
- u64 page_size = rv_mem_get_page_size(params->flags & rv_arena_flag_large_pages);
- u64 reserve_size = rv_align_pow2(params->reserve_size, page_size);
- u64 commit_size = rv_align_pow2(params->commit_size, page_size);
-
- // reserve/commit initial block
- void* base = params->optional_backing_buffer;
- if (base == 0) {
- if (params->flags & rv_arena_flag_large_pages) {
- base = rv_mem_reserve_large(reserve_size);
- rv_mem_commit_large(base, commit_size);
- } else {
- base = rv_mem_reserve(reserve_size);
- rv_mem_commit(base, commit_size);
- }
- }
-
- if (base == NULL) {
- rv_abort_msg(1, "Fatal Allocation Failure");
- }
-
- // extract arena header & fill
- rv_arena* arena = (rv_arena*)base;
- arena->current = arena;
- arena->flags = params->flags;
- arena->cmt_size = params->commit_size;
- arena->res_size = params->reserve_size;
- arena->base_pos = 0;
- arena->pos = RV_ARENA_HEADER_SIZE;
- arena->cmt = commit_size;
- arena->res = reserve_size;
+ // round up reserve/commit sizes
+ u64 page_size = rv_mem_get_page_size(params->flags & rv_arena_flag_large_pages);
+ u64 reserve_size = rv_align_pow2(params->reserve_size, page_size);
+ u64 commit_size = rv_align_pow2(params->commit_size, page_size);
+
+ // reserve/commit initial block
+ void* base = params->optional_backing_buffer;
+ if (base == 0) {
+ if (params->flags & rv_arena_flag_large_pages) {
+ base = rv_mem_reserve_large(reserve_size);
+ rv_mem_commit_large(base, commit_size);
+ } else {
+ base = rv_mem_reserve(reserve_size);
+ rv_mem_commit(base, commit_size);
+ }
+ }
+
+ if (base == NULL) {
+ rv_abort_msg(1, "Fatal Allocation Failure");
+ }
+
+ // extract arena header & fill
+ rv_arena* arena = (rv_arena*)base;
+ arena->current = arena;
+ arena->flags = params->flags;
+ arena->cmt_size = params->commit_size;
+ arena->res_size = params->reserve_size;
+ arena->base_pos = 0;
+ arena->pos = RV_ARENA_HEADER_SIZE;
+ arena->cmt = commit_size;
+ arena->res = reserve_size;
#if RV_ARENA_FREE_LIST
- arena->free_size = 0;
- arena->free_last = 0;
+ arena->free_size = 0;
+ arena->free_last = 0;
#endif
- RV_ASAN_POISOIN(base, commit_size);
- RV_ASAN_UNPOISOIN(base, RV_ARENA_HEADER_SIZE);
- return arena;
+ RV_ASAN_POISOIN(base, commit_size);
+ RV_ASAN_UNPOISOIN(base, RV_ARENA_HEADER_SIZE);
+ return arena;
}
RV_INTERNAL void
rv_arena_release(rv_arena* arena)
{
- for (rv_arena* n = arena->current, *prev = 0; n != 0; n = prev) {
- prev = n->prev;
- rv_mem_release(n, n->res);
- }
+ for (rv_arena* n = arena->current, *prev = 0; n != 0; n = prev) {
+ prev = n->prev;
+ rv_mem_release(n, n->res);
+ }
}
// arena push/pop core functions
@@ -59,113 +59,113 @@ rv_arena_release(rv_arena* arena)
RV_INTERNAL void*
rv_arena_push(rv_arena* arena, u64 size, u64 align)
{
- rv_arena* current = arena->current;
- u64 pos_pre = rv_align_pow2(current->pos, align);
- u64 pos_pst = pos_pre + size;
+ rv_arena* current = arena->current;
+ u64 pos_pre = rv_align_pow2(current->pos, align);
+ u64 pos_pst = pos_pre + size;
- // Chain, if needed
- if (current->res < pos_pst && !(arena->flags & rv_arena_flag_no_chain)) {
- rv_arena* new_block = 0;
+ // Chain, if needed
+ if (current->res < pos_pst && !(arena->flags & rv_arena_flag_no_chain)) {
+ rv_arena* new_block = 0;
#if RV_ARENA_FREE_LIST
- rv_arena* prev_block;
- for (new_block = arena->free_last, prev_block = 0; new_block != 0; prev_block = new_block, new_block = new_block->prev) {
- if (new_block->res >= rv_align_pow2(size, align)) {
- if (prev_block) {
- prev_block->prev = new_block->prev;
- } else {
- arena->free_last = new_block->prev;
- }
- arena->free_size -= new_block->res_size;
- RV_ASAN_UNPOISOIN((u8*)new_block + RV_ARENA_HEADER_SIZE, new_block->res_size - RV_ARENA_HEADER_SIZE);
- break;
- }
- }
+ rv_arena* prev_block;
+ for (new_block = arena->free_last, prev_block = 0; new_block != 0; prev_block = new_block, new_block = new_block->prev) {
+ if (new_block->res >= rv_align_pow2(size, align)) {
+ if (prev_block) {
+ prev_block->prev = new_block->prev;
+ } else {
+ arena->free_last = new_block->prev;
+ }
+ arena->free_size -= new_block->res_size;
+ RV_ASAN_UNPOISOIN((u8*)new_block + RV_ARENA_HEADER_SIZE, new_block->res_size - RV_ARENA_HEADER_SIZE);
+ break;
+ }
+ }
#endif
- if (new_block == 0) {
- u64 res_size = current->res_size;
- u64 cmt_size = current->cmt_size;
- if (size + RV_ARENA_HEADER_SIZE > res_size) {
- res_size = rv_align_pow2(size + RV_ARENA_HEADER_SIZE, align);
- cmt_size = rv_align_pow2(size + RV_ARENA_HEADER_SIZE, align);
- }
- new_block = rv_arena_alloc(.reserve_size = res_size,
- .commit_size = cmt_size,
- .flags = current->flags);
- }
-
- new_block->base_pos = current->base_pos + current->res;
- RV_STACK_PUSH_N(arena->current, new_block, prev);
-
- current = new_block;
- pos_pre = rv_align_pow2(current->pos, align);
- pos_pst = pos_pre + size;
- }
-
- // commit new pages, if needed
- if (current->cmt < pos_pst) {
- u64 cmt_pst_aligned = pos_pst + current->cmt_size-1;
- cmt_pst_aligned -= cmt_pst_aligned%current->cmt_size;
- u64 cmt_pst_clamped = rv_min(cmt_pst_aligned, current->res);
- u64 cmt_size = cmt_pst_clamped - current->cmt;
- u8 *cmt_ptr = (u8 *)current + current->cmt;
- if (current->flags & rv_arena_flag_large_pages) {
- rv_mem_commit_large(cmt_ptr, cmt_size);
- } else {
- rv_mem_commit(cmt_ptr, cmt_size);
- }
- current->cmt = cmt_pst_clamped;
- }
-
- // push onto current block
- void* result = 0;
- if (current->cmt >= pos_pst) {
- result = (u8 *)current+pos_pre;
- current->pos = pos_pst;
- RV_ASAN_UNPOISOIN(result, size);
- }
-
- if (result == 0) {
- rv_abort_msg(1, "Fatal Allocation Failure");
- }
-
- return result;
+ if (new_block == 0) {
+ u64 res_size = current->res_size;
+ u64 cmt_size = current->cmt_size;
+ if (size + RV_ARENA_HEADER_SIZE > res_size) {
+ res_size = rv_align_pow2(size + RV_ARENA_HEADER_SIZE, align);
+ cmt_size = rv_align_pow2(size + RV_ARENA_HEADER_SIZE, align);
+ }
+ new_block = rv_arena_alloc(.reserve_size = res_size,
+ .commit_size = cmt_size,
+ .flags = current->flags);
+ }
+
+ new_block->base_pos = current->base_pos + current->res;
+ RV_STACK_PUSH_N(arena->current, new_block, prev);
+
+ current = new_block;
+ pos_pre = rv_align_pow2(current->pos, align);
+ pos_pst = pos_pre + size;
+ }
+
+ // commit new pages, if needed
+ if (current->cmt < pos_pst) {
+ u64 cmt_pst_aligned = pos_pst + current->cmt_size-1;
+ cmt_pst_aligned -= cmt_pst_aligned%current->cmt_size;
+ u64 cmt_pst_clamped = rv_min(cmt_pst_aligned, current->res);
+ u64 cmt_size = cmt_pst_clamped - current->cmt;
+ u8 *cmt_ptr = (u8 *)current + current->cmt;
+ if (current->flags & rv_arena_flag_large_pages) {
+ rv_mem_commit_large(cmt_ptr, cmt_size);
+ } else {
+ rv_mem_commit(cmt_ptr, cmt_size);
+ }
+ current->cmt = cmt_pst_clamped;
+ }
+
+ // push onto current block
+ void* result = 0;
+ if (current->cmt >= pos_pst) {
+ result = (u8 *)current+pos_pre;
+ current->pos = pos_pst;
+ RV_ASAN_UNPOISOIN(result, size);
+ }
+
+ if (result == 0) {
+ rv_abort_msg(1, "Fatal Allocation Failure");
+ }
+
+ return result;
}
RV_INTERNAL u64
rv_arena_pos(rv_arena* arena)
{
- rv_arena* current = arena->current;
- u64 pos = current->base_pos + current->pos;
- return pos;
+ rv_arena* current = arena->current;
+ u64 pos = current->base_pos + current->pos;
+ return pos;
}
RV_INTERNAL void
rv_arena_pop_to(rv_arena* arena, u64 pos)
{
- u64 big_pos = rv_max(RV_ARENA_HEADER_SIZE, pos);
- rv_arena* current = arena->current;
+ u64 big_pos = rv_max(RV_ARENA_HEADER_SIZE, pos);
+ rv_arena* current = arena->current;
#if RV_ARENA_FREE_LIST
- for (rv_arena* prev = 0; current->base_pos >= big_pos; current = prev) {
- prev = current->prev;
- current->pos = RV_ARENA_HEADER_SIZE;
- arena->free_size += current->res_size;
- RV_STACK_PUSH_N(arena->free_last, current, prev);
- RV_ASAN_POISOIN((u8*)current + RV_ARENA_HEADER_SIZE, current->res_size - RV_ARENA_HEADER_SIZE);
- }
+ for (rv_arena* prev = 0; current->base_pos >= big_pos; current = prev) {
+ prev = current->prev;
+ current->pos = RV_ARENA_HEADER_SIZE;
+ arena->free_size += current->res_size;
+ RV_STACK_PUSH_N(arena->free_last, current, prev);
+ RV_ASAN_POISOIN((u8*)current + RV_ARENA_HEADER_SIZE, current->res_size - RV_ARENA_HEADER_SIZE);
+ }
#else
- for (rv_arena* prev = 0; current->base_pos >= big_pos; current = prev) {
- prev = current->prev;
- rv_mem_release(current, current->res);
- }
+ for (rv_arena* prev = 0; current->base_pos >= big_pos; current = prev) {
+ prev = current->prev;
+ rv_mem_release(current, current->res);
+ }
#endif
- arena->current = current;
- u64 new_pos = big_pos - current->base_pos;
- rv_assert(new_pos <= current->pos);
- RV_ASAN_POISOIN((u8*)current + new_pos, (current->pos - new_pos));
- current->pos = new_pos;
+ arena->current = current;
+ u64 new_pos = big_pos - current->base_pos;
+ rv_assert(new_pos <= current->pos);
+ RV_ASAN_POISOIN((u8*)current + new_pos, (current->pos - new_pos));
+ current->pos = new_pos;
}
// arena push/pop helpers
@@ -173,18 +173,18 @@ rv_arena_pop_to(rv_arena* arena, u64 pos)
RV_INTERNAL void
rv_arena_clear(rv_arena* arena)
{
- rv_arena_pop_to(arena, 0);
+ rv_arena_pop_to(arena, 0);
}
RV_INTERNAL void
rv_arena_pop(rv_arena* arena, u64 amt)
{
- u64 pos_old = rv_arena_pos(arena);
- u64 pos_new = pos_old;
- if (amt < pos_old) {
- pos_new = pos_old - amt;
- }
- rv_arena_pop_to(arena, pos_new);
+ u64 pos_old = rv_arena_pos(arena);
+ u64 pos_new = pos_old;
+ if (amt < pos_old) {
+ pos_new = pos_old - amt;
+ }
+ rv_arena_pop_to(arena, pos_new);
}
// Temporary arena scopes
@@ -192,45 +192,45 @@ rv_arena_pop(rv_arena* arena, u64 amt)
RV_INTERNAL rv_temp_arena
rv_temp_begin(rv_arena* arena)
{
- u64 pos = rv_arena_pos(arena);
- rv_temp_arena temp = {arena, pos};
- return temp;
+ u64 pos = rv_arena_pos(arena);
+ rv_temp_arena temp = {arena, pos};
+ return temp;
}
RV_INTERNAL void
rv_temp_end(rv_temp_arena temp)
{
- rv_arena_pop_to(temp.arena, temp.pos);
+ rv_arena_pop_to(temp.arena, temp.pos);
}
#if !defined(_rv_scratch_count)
- #define _rv_scratch_count 2
+#define _rv_scratch_count 2
#endif
-RV_THREAD_LOCAL rv_arena* _rv_thread_scratch_pool[_rv_scratch_count] = {0};
+RV_THREAD_LOCAL RV_INTERNAL rv_arena* _rv_thread_scratch_pool[_rv_scratch_count] = {0};
RV_GLOBAL rv_arena* rv_scratch_begin_(rv_arena **conflicts, u64 count)
{
- rv_arena *result = 0;
- rv_arena **arena_ptr = _rv_thread_scratch_pool;
- for (u64 i = 0; i < _rv_scratch_count; i += 1, arena_ptr += 1) {
- rv_arena **conflict_ptr = conflicts;
- bool32 has_conflict = 0;
- for (u64 j = 0; j < count; j += 1, conflict_ptr += 1) {
- if (*arena_ptr == *conflict_ptr) {
- has_conflict = 1;
- break;
- }
- }
- if (!has_conflict) {
- if (*arena_ptr == NULL) {
- *arena_ptr = rv_arena_alloc();
- }
- result = *arena_ptr;
- break;
- }
- }
-
- return result;
+ rv_arena *result = 0;
+ rv_arena **arena_ptr = _rv_thread_scratch_pool;
+ for (u64 i = 0; i < _rv_scratch_count; i += 1, arena_ptr += 1) {
+ rv_arena **conflict_ptr = conflicts;
+ bool32 has_conflict = 0;
+ for (u64 j = 0; j < count; j += 1, conflict_ptr += 1) {
+ if (*arena_ptr == *conflict_ptr) {
+ has_conflict = 1;
+ break;
+ }
+ }
+ if (!has_conflict) {
+ if (*arena_ptr == NULL) {
+ *arena_ptr = rv_arena_alloc();
+ }
+ result = *arena_ptr;
+ break;
+ }
+ }
+
+ return result;
}
diff --git a/src/platform/platform_arena.h b/src/platform/platform_arena.h
@@ -11,33 +11,33 @@
typedef u64 rv_arena_flags;
enum {
- rv_arena_flag_no_chain = (1<<0),
- rv_arena_flag_large_pages = (1<<1),
+ rv_arena_flag_no_chain = (1<<0),
+ rv_arena_flag_large_pages = (1<<1),
};
typedef struct rv_arena_params rv_arena_params;
struct rv_arena_params {
- rv_arena_flags flags;
- u64 reserve_size;
- u64 commit_size;
- void *optional_backing_buffer;
+ rv_arena_flags flags;
+ u64 reserve_size;
+ u64 commit_size;
+ void *optional_backing_buffer;
};
typedef struct rv_arena rv_arena;
struct rv_arena
{
- rv_arena *prev; // previous arena in chain
- rv_arena *current; // current arena in chain
- rv_arena_flags flags;
- u64 cmt_size;
- u64 res_size;
- u64 base_pos;
- u64 pos;
- u64 cmt;
- u64 res;
+ rv_arena *prev; // previous arena in chain
+ rv_arena *current; // current arena in chain
+ rv_arena_flags flags;
+ u64 cmt_size;
+ u64 res_size;
+ u64 base_pos;
+ u64 pos;
+ u64 cmt;
+ u64 res;
#if RV_ARENA_FREE_LIST
- u64 free_size;
- rv_arena *free_last;
+ u64 free_size;
+ rv_arena *free_last;
#endif
};
//StaticAssert(sizeof(rv_arena) <= RV_ARENA_HEADER_SIZE, arena header too large);
@@ -45,8 +45,8 @@ struct rv_arena
typedef struct rv_temp_arena rv_temp_arena;
struct rv_temp_arena
{
- rv_arena *arena;
- u64 pos;
+ rv_arena *arena;
+ u64 pos;
};
//////////////////////////////////////////////////////////////////
@@ -83,11 +83,16 @@ RV_INTERNAL void rv_temp_end(rv_temp_arena temp);
#define rv_push_aligned(a, T, align) (T*)rv_mem_zero(rv_push_no_zero_aligned(a, T, align), sizeof(T))
#define rv_push(a, T) rv_push_aligned(a, T, rv_max(8, rv_alignof(T)))
+#define rv_push_compound(a, T, ...) rv_mem_copy(rv_push_no_zero_aligned(a, T, rv_max(8, rv_alignof(T))), &(T)__VA_ARGS__, sizeof(T))
+#define rv_push_copy(a, T, v) rv_mem_copy(rv_push_no_zero_aligned(a, T, rv_max(8, rv_alignof(T))), (T*)v, sizeof(T))
+
#define rv_push_array_no_zero_aligned(a, T, c, align) (T*)rv_arena_push((a), sizeof(T)*(c), (align))
#define rv_push_array_no_zero(a, T, c) rv_push_array_no_zero_aligned(a, T, c, rv_max(8, rv_alignof(T)))
#define rv_push_array_aligned(a, T, c, align) (T*)rv_mem_zero(rv_push_array_no_zero_aligned(a, T, c, align), sizeof(T)*(c))
#define rv_push_array(a, T, c) rv_push_array_aligned(a, T, c, rv_max(8, rv_alignof(T)))
+#define rv_push_array_compound(a, T, c, ...) rv_mem_copy(rv_push_array_no_zero_aligned(a, T, c, rv_max(8, rv_alignof(T))), (T[c])__VA_ARGS__, sizeof(T)*c)
+
// Scratch
RV_GLOBAL rv_arena* rv_scratch_begin_(rv_arena **conflicts, u64 count);
#define rv_scratch_begin(conflicts, count) rv_temp_begin(rv_scratch_begin_((conflicts), (count)));
diff --git a/src/platform/platform_core.c b/src/platform/platform_core.c
@@ -4,9 +4,18 @@
//////////////////////////////////////////////////////////////////
// Utils
-RV_GLOBAL void rv_abort_msg_(const char* file, const char* func, s32 line_no, const char* message, s32 exit_code)
+RV_GLOBAL void rv_abort_msg_(const char* file, const char* func, s32 line_no, s32 exit_code, const char* message, ...)
{
- fprintf(stderr, "rv_abort[%d]: '%s' at %s:%d (%s)\n", exit_code, message, file, line_no, func);
+
+ char buf[512] = {0};
+ va_list args;
+ va_start(args, message);
+ vsnprintf(buf, sizeof(buf), message, args);
+ va_end(args);
+
+ fprintf(stderr, "rv_abort[%d]: '%s' at %s:%d (%s)\n", exit_code, buf, file, line_no, func);
+
+
#if RV_OS_WINDOWS
ExitProcess(exit_code);
@@ -37,7 +46,7 @@ f64 rv_time(void) {
//////////////////////////////////////////////////////////////////
// Memory
-RV_GLOBAL void* rv_mem_reserve(u64 size)
+RV_GLOBAL void* rv_mem_reserve(s64 size)
{
void* result = NULL;
#if RV_OS_WINDOWS
@@ -48,7 +57,7 @@ RV_GLOBAL void* rv_mem_reserve(u64 size)
#endif
return result;
}
-RV_GLOBAL void* rv_mem_reserve_large(u64 size)
+RV_GLOBAL void* rv_mem_reserve_large(s64 size)
{
void* result = NULL;
#if RV_OS_WINDOWS
@@ -61,7 +70,7 @@ RV_GLOBAL void* rv_mem_reserve_large(u64 size)
return result;
}
-RV_GLOBAL void rv_mem_commit(void* ptr, u64 size)
+RV_GLOBAL void rv_mem_commit(void* ptr, s64 size)
{
#if RV_OS_WINDOWS
VirtualAlloc(ptr, size, MEM_COMMIT, PAGE_READWRITE);
@@ -69,7 +78,7 @@ RV_GLOBAL void rv_mem_commit(void* ptr, u64 size)
mprotect(ptr, size, PROT_READ|PROT_WRITE);
#endif
}
-RV_GLOBAL void rv_mem_commit_large(void* ptr, u64 size)
+RV_GLOBAL void rv_mem_commit_large(void* ptr, s64 size)
{
#if RV_OS_WINDOWS
// do nothing, already commited
@@ -78,7 +87,7 @@ RV_GLOBAL void rv_mem_commit_large(void* ptr, u64 size)
#endif
}
-RV_GLOBAL void rv_mem_decommit(void* ptr, u64 size)
+RV_GLOBAL void rv_mem_decommit(void* ptr, s64 size)
{
#if RV_OS_WINDOWS
VirtualFree(ptr, size, MEM_DECOMMIT);
@@ -88,7 +97,7 @@ RV_GLOBAL void rv_mem_decommit(void* ptr, u64 size)
#endif
}
-RV_GLOBAL void rv_mem_release(void* ptr, u64 size)
+RV_GLOBAL void rv_mem_release(void* ptr, s64 size)
{
#if RV_OS_WINDOWS
VirtualFree(ptr, 0, MEM_RELEASE);
@@ -97,9 +106,9 @@ RV_GLOBAL void rv_mem_release(void* ptr, u64 size)
#endif
}
-RV_GLOBAL u64 rv_mem_get_page_size(bool32 is_large)
+RV_GLOBAL s64 rv_mem_get_page_size(bool32 is_large)
{
- u64 result = 0;
+ s64 result = 0;
#if RV_OS_WINDOWS
@@ -116,37 +125,40 @@ RV_GLOBAL u64 rv_mem_get_page_size(bool32 is_large)
if (is_large) {
// I hate this
- const char* file = "/proc/meminfo";
- char buf[2048];
+ rv_str8 filename = S("/proc/meminfo");
int readerr;
long val;
- int fd = open(file, O_RDONLY);
- int len = 0;
-
- bool32 read_file = false;
- if (fd) {
- len = read(fd, buf, sizeof(buf));
- if (len > 0 && len <= sizeof(buf)) {
- buf[len] = '\0';
- read_file = true;
- }
- close(fd);
- }
+ rv_temp_arena scratch = rv_scratch_begin(0, 0);
- if (read_file) {
- char* p = strstr(buf, "Hugepagesize:");
- if (p) p += strlen( "Hugepagesize:");
-
- char* q = "";
- result = strtol(p, &q, 0);
- if (!isspace(*q)) {
- result = 0;
- } else {
- result = KB(result);
- }
+ char* buffer = rv_push_array(scratch.arena, char, KB(4));
+ FILE* f = fopen("/proc/meminfo", "rb");
+ if (f) {
+ fread(buffer, KB(4) - 1, 1, f);
}
+ rv_str8 proc_meminfo = rv_str8_from_cstr(buffer);
+
+ rv_str8 tag = S("Hugepagesize:");
+ rv_str8 found = rv_str8_skip(proc_meminfo, rv_str8_find_needle(proc_meminfo, 0, tag, 0) + tag.size);
+
+ // chop around number + byte size
+ found = rv_str8_skip_chop_whitespace(found);
+ found = rv_str8_prefix(found, rv_str8_find_needle(found, 0, S("\n"), 0));
+
+ bool32 kb = rv_str8_find_needle(found, 0, S("kb"), rv_str8_match_flag_case_insensetive) < found.size;
+ bool32 mb = rv_str8_find_needle(found, 0, S("mb"), rv_str8_match_flag_case_insensetive) < found.size;
+ bool32 gb = rv_str8_find_needle(found, 0, S("gb"), rv_str8_match_flag_case_insensetive) < found.size;
+
+ // remove byte part
+ found = rv_str8_prefix(found, rv_str8_find_needle(found, 0, S(" "), 0));
+
+ result = rv_s64_from_str8(found, 10);
+ if (kb) result = KB(result);
+ if (mb) result = MB(result);
+ if (gb) result = GB(result);
+
+ rv_scratch_end(scratch);
} else {
result = getpagesize();
}
@@ -154,16 +166,46 @@ RV_GLOBAL u64 rv_mem_get_page_size(bool32 is_large)
return result;
}
-RV_GLOBAL void* rv_mem_zero(void* mem, s64 size)
+RV_GLOBAL void* rv_mem_set(void* mem, u8 set_byte, s64 size)
{
- if (mem && size > 0) {
- bzero(mem, size);
+ u8* bytes = mem;
+ for (s64 i = 0; i < size; i++) {
+ bytes[i] = set_byte;
}
return mem;
}
+RV_GLOBAL void* rv_mem_zero(void* mem, s64 size)
+{
+ return rv_mem_set(mem, 0, size);
+}
+
+RV_GLOBAL void* rv_mem_copy(void* restrict target, const void* restrict source, s64 size)
+{
+ u8* t = target;
+ const u8* s = source;
+ for (s64 i = 0; i < size; i++) {
+ t[i] = s[i];
+ }
+ return target;
+}
+
+RV_GLOBAL void* rv_mem_move(void* target, const void* source, s64 size)
+{
+ rv_temp_arena scratch = rv_scratch_begin(0, 0);
+
+ u8* temp = rv_push_array_no_zero(scratch.arena, u8, size);
+
+ rv_mem_copy(temp, source, size);
+ rv_mem_copy(target, temp, size);
+
+ rv_scratch_end(scratch);
+
+ return target;
+}
+
#if !defined(RV_NO_GLOBAL_ALLOC)
-RV_GLOBAL void* rv_mem_global_alloc_nz_(s64 size, u64 align) { return malloc(size); }
-RV_GLOBAL void* rv_mem_global_realloc_nz_(void* old_ptr, s64 new_size, u64 align) { return realloc(old_ptr, new_size);}
+RV_GLOBAL void* rv_mem_global_alloc_nz_(s64 size, s64 align) { return malloc(size); }
+RV_GLOBAL void* rv_mem_global_realloc_nz_(void* old_ptr, s64 new_size, s64 align) { return realloc(old_ptr, new_size);}
RV_GLOBAL void rv_mem_global_alloc_free(void* ptr, s64 size) { free(ptr); }
#endif
diff --git a/src/platform/platform_core.h b/src/platform/platform_core.h
@@ -26,27 +26,10 @@
#endif
-#include <alloca.h>
#include <stdio.h>
-#include <string.h>
#include <stdlib.h>
-#include <ctype.h>
-#include <stddef.h>
#include <time.h>
-#include <math.h>
-
-//////////////////////////////////////////////////////////////////
-// Units
-
-#define KB(n) (((u64)(n)) << 10)
-#define MB(n) (((u64)(n)) << 20)
-#define GB(n) (((u64)(n)) << 30)
-#define TB(n) (((u64)(n)) << 40)
-
-#define rv_min(n, _min) ((n) > (_min) ? (n) : (_min))
-#define rv_max(n, _max) ((n) < (_max) ? (n) : (_max))
-
-#define rv_array_count(arr) sizeof(arr) / sizeof(arr[0])
+#include <stdarg.h>
//////////////////////////////////////////////////////////////////
// Linked List
@@ -96,15 +79,18 @@
#define rv_dll_insert(f,l,p,n) RV_DLL_INSERT_NPZ(f,l,p,n,next,prev,RV_CHECK_NULL,RV_SET_NULL)
#define rv_dll_remove(f,l,n) RV_DLL_REMOVE_NPZ(f,l,n,next,prev,RV_CHECK_NULL,RV_SET_NULL)
-
//////////////////////////////////////////////////////////////////
// Utils
+#define rv_strify(...) #__VA_ARGS__
+
#define rv_align_pow2(x,b) (((x) + (b) - 1)&(~((b) - 1)))
-#define rv_abort(exit_code) rv_abort_msg_(__FILE__, __FUNCTION__, __LINE__, "(no message)", exit_code)
-#define rv_abort_msg(exit_code, msg) rv_abort_msg_(__FILE__, __FUNCTION__, __LINE__, msg, exit_code)
-RV_GLOBAL void rv_abort_msg_(const char* file, const char* func, s32 line_no, const char* message, s32 exit_code);
+#define rv_array_size(__ARR) sizeof(__ARR) / sizeof(__ARR[0])
+
+#define rv_abort(exit_code) rv_abort_msg_(__FILE__, __FUNCTION__, __LINE__, exit_code, "(no message)")
+#define rv_abort_msg(exit_code, ...) rv_abort_msg_(__FILE__, __FUNCTION__, __LINE__, exit_code, __VA_ARGS__)
+RV_GLOBAL void rv_abort_msg_(const char* file, const char* func, s32 line_no, s32 exit_code, const char* message, ...);
f64 rv_time(void);
@@ -112,19 +98,22 @@ f64 rv_time(void);
//////////////////////////////////////////////////////////////////
// Memory
-RV_GLOBAL void* rv_mem_reserve(u64 size);
-RV_GLOBAL void* rv_mem_reserve_large(u64 size);
-RV_GLOBAL void rv_mem_commit(void* ptr, u64 size);
-RV_GLOBAL void rv_mem_commit_large(void* ptr, u64 size);
-RV_GLOBAL void rv_mem_decommit(void* ptr, u64 size);
-RV_GLOBAL void rv_mem_release(void* ptr, u64 size);
+RV_GLOBAL void* rv_mem_reserve(s64 size);
+RV_GLOBAL void* rv_mem_reserve_large(s64 size);
+RV_GLOBAL void rv_mem_commit(void* ptr, s64 size);
+RV_GLOBAL void rv_mem_commit_large(void* ptr, s64 size);
+RV_GLOBAL void rv_mem_decommit(void* ptr, s64 size);
+RV_GLOBAL void rv_mem_release(void* ptr, s64 size);
-RV_GLOBAL u64 rv_mem_get_page_size(bool32 is_large);
+RV_GLOBAL s64 rv_mem_get_page_size(bool32 is_large); // if is_large = true, it will return 0 if large pages are not supported
-RV_GLOBAL void* rv_mem_zero(void* mem, s64 size); // returns mem
+RV_GLOBAL void* rv_mem_set(void* mem, u8 set_byte, s64 size); // returns mem
+RV_GLOBAL void* rv_mem_zero(void* mem, s64 size); // returns mem
+RV_GLOBAL void* rv_mem_copy(void* restrict target, const void* restrict source, s64 size); // returns target
+RV_GLOBAL void* rv_mem_move(void* target, const void* source, s64 size); // returns target
-RV_GLOBAL void* rv_mem_global_alloc_nz_(s64 size, u64 align);
-RV_GLOBAL void* rv_mem_global_realloc_nz_(void* old_ptr, s64 new_size, u64 align);
+RV_GLOBAL void* rv_mem_global_alloc_nz_(s64 size, s64 align);
+RV_GLOBAL void* rv_mem_global_realloc_nz_(void* old_ptr, s64 new_size, s64 align);
RV_GLOBAL void rv_mem_global_alloc_free(void* ptr, s64 size);
// helper macros
diff --git a/src/platform/platform_cracker.h b/src/platform/platform_cracker.h
@@ -41,6 +41,7 @@
// RV_OS_WINDOWS
//
// Window System:
+// RV_WIN_ENABLED
// RV_WIN_X11 (default on linux)
// RV_WIN_WINDOWS (not implemented)
// RV_WIN_WAYLAND (not implemented)
@@ -343,6 +344,12 @@
#define RV_WIN_WINDOWS 0
#endif
+#if RV_WIN_X11 || RV_WIN_WAYLAND || RV_WIN_WINDOWS
+ #define RV_WIN_ENABLED 1
+#else
+ #define RV_WIN_ENABLED 0
+#endif
+
// language constructs
#if RV_LANG_C
@@ -415,7 +422,3 @@
#endif
-#if RV_OS_LINUX
- #define _GNU_SOURCE 1
-#endif
-
diff --git a/src/platform/platform_inc.c b/src/platform/platform_inc.c
@@ -2,16 +2,22 @@
// platform_inc.c
#include "platform_core.c"
+#include "platform_math.c"
#include "platform_arena.c"
+#include "platform_string.c"
+
+#if RV_WIN_ENABLED
+ #include "gfx/platform_gfx.c"
+#endif
#if RV_WIN_X11
- #include "gfx/xcb_impl.c"
+ #include "gfx/xcb_impl.c"
#elif RV_WIN_WAYLAND
- #error Wayland not implemented
+ #error Wayland not implemented
#elif RV_WIN_WINDOWS
- #error Windows GFX not implemented
+ #error Windows GFX not implemented
#endif
#if RV_RENDER_OPENGL
- #include "gfx/glad.c"
+ #include "../external/glad.c"
#endif
diff --git a/src/platform/platform_inc.h b/src/platform/platform_inc.h
@@ -4,9 +4,14 @@
#include "platform_cracker.h"
#include "platform_types.h"
#include "platform_core.h"
+#include "platform_math.h"
#include "platform_arena.h"
+#include "platform_string.h"
+
+#if RV_WIN_ENABLED
+ #include "gfx/platform_gfx.h"
+#endif
-#include "gfx/platform_gfx.h"
#if RV_RENDER_OPENGL
- #include "gfx/glad.h"
+ #include "../external/glad.h"
#endif
diff --git a/src/platform/platform_math.c b/src/platform/platform_math.c
@@ -0,0 +1,107 @@
+//////////////////////////////////////////////////////////////////
+// platform_math.c
+
+// range
+
+RV_GLOBAL s64 rv_range_diff(rv_range range)
+{
+ return range.max - range.min;
+}
+
+//////////////////////////////////////////////////////////////////
+// vector math
+
+
+RV_GLOBAL f32 rv_vec3_dot(const rv_vec3 a, const rv_vec3 b)
+{
+ return (a.x * b.x) + (a.y * b.y) + (a.z * b.z);
+}
+
+RV_GLOBAL f32 rv_vec3_len(const rv_vec3 v)
+{
+ return sqrtf(rv_vec3_dot(v, v));
+}
+
+RV_GLOBAL rv_vec3 rv_vec3_scale(const rv_vec3 v, const f32 s)
+{
+ return rv_v3(v.x * s, v.y * s, v.z * s);
+}
+
+RV_GLOBAL rv_vec3 rv_vec3_norm(const rv_vec3 v)
+{
+ f32 len = rv_vec3_len(v);
+ return len == 0.f ? v : rv_vec3_scale(v, 1.f / len);
+}
+
+
+//////////////////////////////////////////////////////////////////
+// matrix math
+
+RV_GLOBAL rv_mat4 rv_mat4_rotatev(const f32 angle, const rv_vec3 axis)
+{
+ rv_mat4 m_res = rv_mat4_identity();
+
+ f32 a = angle;
+ f32 c = cosf(a);
+ f32 s = sinf(a);
+
+ rv_vec3 naxis = rv_vec3_norm(axis);
+ f32 x = naxis.x;
+ f32 y = naxis.y;
+ f32 z = naxis.z;
+
+ // first column
+ m_res.elements[0 + 0 * 4] = x * x * (1 - c) + c;
+ m_res.elements[1 + 0 * 4] = x * y * (1 - c) + z * s;
+ m_res.elements[2 + 0 * 4] = x * z * (1 - c) - y * s;
+
+ //Second column
+ m_res.elements[0 + 1 * 4] = x * y * (1 - c) - z * s;
+ m_res.elements[1 + 1 * 4] = y * y * (1 - c) + c;
+ m_res.elements[2 + 1 * 4] = y * z * (1 - c) + x * s;
+
+ //Third column
+ m_res.elements[0 + 2 * 4] = x * z * (1 - c) + y * s;
+ m_res.elements[1 + 2 * 4] = y * z * (1 - c) - x * s;
+ m_res.elements[2 + 2 * 4] = z * z * (1 - c) + c;
+
+ return m_res;
+}
+
+RV_GLOBAL rv_mat4 rv_mat4_scalev(const rv_vec3 v)
+{
+ rv_mat4 m_res = rv_mat4_identity();
+ m_res.elements[0 + 0 * 4] = v.x;
+ m_res.elements[1 + 1 * 4] = v.y;
+ m_res.elements[2 + 2 * 4] = v.z;
+ return m_res;
+}
+
+RV_GLOBAL rv_mat4
+rv_mat4_mul(const rv_mat4 m0, const rv_mat4 m1)
+{
+ rv_mat4 m_res = {0};
+ for (s32 y = 0; y < 4; ++y) {
+ for (s32 x = 0; x < 4; ++x) {
+ f32 sum = 0.0f;
+ for (s32 e = 0; e < 4; ++e) {
+ sum += m0.elements[x + e * 4] * m1.elements[e + y * 4];
+ }
+ m_res.elements[x + y * 4] = sum;
+ }
+ }
+ return m_res;
+}
+
+RV_GLOBAL rv_mat4 rv_mat4_mul_list(const s32 count, ...)
+{
+
+ va_list ap;
+ rv_mat4 m = rv_mat4_identity();
+ va_start(ap, count);
+ for (uint32_t i = 0; i < count; ++i) {
+ m = rv_mat4_mul(m, va_arg(ap, rv_mat4));
+ }
+ va_end(ap);
+ return m;
+}
diff --git a/src/platform/platform_math.h b/src/platform/platform_math.h
@@ -0,0 +1,42 @@
+//////////////////////////////////////////////////////////////////
+// platform_math.h
+
+
+//////////////////////////////////////////////////////////////////
+// Misc
+
+#include <math.h>
+
+#define KB(n) (((u64)(n)) << 10)
+#define MB(n) (((u64)(n)) << 20)
+#define GB(n) (((u64)(n)) << 30)
+#define TB(n) (((u64)(n)) << 40)
+
+#define rv_min(A, B) ((A) < (B) ? (A) : (B))
+#define rv_max(A, B) ((A) > (B) ? (A) : (B))
+
+#define rv_array_count(arr) sizeof(arr) / sizeof(arr[0])
+
+#define RV_XAXIS rv_v3(1, 0, 0)
+#define RV_YAXIS rv_v3(0, 1, 0)
+#define RV_ZAXIS rv_v3(0, 0, 1)
+
+// range
+
+RV_GLOBAL s64 rv_range_diff(rv_range range);
+
+//////////////////////////////////////////////////////////////////
+// vector math
+
+RV_GLOBAL f32 rv_vec3_dot(const rv_vec3 a, const rv_vec3 b);
+RV_GLOBAL f32 rv_vec3_len(const rv_vec3 v);
+RV_GLOBAL rv_vec3 rv_vec3_scale(const rv_vec3 v, const f32 s);
+RV_GLOBAL rv_vec3 rv_vec3_norm(const rv_vec3 v);
+
+//////////////////////////////////////////////////////////////////
+// matrix math
+
+RV_GLOBAL rv_mat4 rv_mat4_rotatev(const f32 angle, const rv_vec3 axis);
+RV_GLOBAL rv_mat4 rv_mat4_scalev(const rv_vec3 v);
+RV_GLOBAL rv_mat4 rv_mat4_mul(const rv_mat4 m0, const rv_mat4 m1);
+RV_GLOBAL rv_mat4 rv_mat4_mul_list(const s32 count, ...);
diff --git a/src/platform/platform_string.c b/src/platform/platform_string.c
@@ -0,0 +1,315 @@
+//////////////////////////////////////////////////////////////////
+// platform_string.h
+
+//////////////////////////////////////////////////////////////////
+// char
+
+
+RV_READ_ONLY RV_INTERNAL u8 char_integer_symbol_reverse[128] = {
+ 0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,
+ 0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,
+ 0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,
+ 0x00,0x01,0x02,0x03,0x04,0x05,0x06,0x07,0x08,0x09,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,
+ 0xFF,0x0A,0x0B,0x0C,0x0D,0x0E,0x0F,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,
+ 0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,
+ 0xFF,0x0A,0x0B,0x0C,0x0D,0x0E,0x0F,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,
+ 0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,
+};
+
+RV_GLOBAL bool32 rv_char_is_space(u8 c)
+{
+ return c == ' ' || c == '\n' || c == '\t' || c == '\r' || c == '\f' || c == '\v';
+}
+
+RV_GLOBAL bool32 rv_char_is_upper(u8 c)
+{
+ return 'A' <= c && c <= 'Z';
+}
+
+RV_GLOBAL bool32 rv_char_is_lower(u8 c)
+{
+ return 'a' <= c && c <= 'z';
+}
+
+RV_GLOBAL bool32 rv_char_is_alpha(u8 c)
+{
+ return rv_char_is_upper(c) || rv_char_is_lower(c);
+}
+
+RV_GLOBAL bool32 rv_char_is_slash(u8 c)
+{
+ return c == '/' || c == '\\';
+}
+
+RV_GLOBAL bool32 rv_char_is_digit(u8 c, s32 base)
+{
+ bool32 result = 0;
+ if (base >= 0 && base <= 16) {
+ u8 val = char_integer_symbol_reverse[c];
+ if (val < base) {
+ result = 1;
+ }
+ }
+ return result;
+}
+
+RV_GLOBAL u8 rv_char_to_lower(u8 c)
+{
+ if (rv_char_is_upper(c)) {
+ c += ('a' - 'A');
+ }
+ return c;
+}
+
+RV_GLOBAL u8 rv_char_to_upper(u8 c)
+{
+ if (rv_char_is_lower(c)) {
+ c += ('A' - 'a');
+ }
+ return c;
+}
+
+RV_GLOBAL u8 rv_char_to_correct_slash(u8 c)
+{
+ if(rv_char_is_slash(c)) {
+ c = '/';
+ }
+ return c;
+}
+
+//////////////////////////////////////////////////////////////////
+// c strings
+
+RV_GLOBAL s64 rv_cstrlen(const char* c_str)
+{
+ s64 res = 0;
+ while (*c_str++) res++;
+ return res;
+}
+
+//////////////////////////////////////////////////////////////////
+// constructors
+
+RV_GLOBAL rv_str8 rv_str8_range(u8* first, u8* one_past_last)
+{
+ return rv_s8(first, (s64)(one_past_last - first));
+}
+
+
+//////////////////////////////////////////////////////////////////
+// conversions
+
+RV_GLOBAL s64 rv_sign_from_str8(rv_str8 string, rv_str8* string_tail)
+{
+ // count negative signs
+ u64 neg_count = 0;
+ u64 i = 0;
+ for (; i < string.size; i += 1) {
+ if (string.str[i] == '-') {
+ neg_count += 1;
+ }
+ else if (string.str[i] != '+') {
+ break;
+ }
+ }
+
+ // output part of string after signs
+ *string_tail = rv_str8_skip(string, i);
+
+ // output integer sign
+ s64 sign = (neg_count & 1) ? -1 : +1;
+ return sign;
+}
+
+
+RV_GLOBAL u64 rv_u64_from_str8(rv_str8 string, u32 radix)
+{
+ u64 x = 0;
+ if (1 < radix && radix <= 16) {
+ for (u64 i = 0; i < string.size; i += 1) {
+ x *= radix;
+ x += char_integer_symbol_reverse[string.str[i]&0x7F];
+ }
+ }
+ return x;
+}
+
+
+RV_GLOBAL s64 rv_s64_from_str8(rv_str8 string, s32 radix)
+{
+ s64 sign = rv_sign_from_str8(string, &string);
+ s64 x = (s64)rv_u64_from_str8(string, radix) * sign;
+ return x;
+}
+
+RV_GLOBAL rv_str8 str8_from_s64(rv_arena* arena, s64 integer, s32 radix, u8 min_digits, u8 digit_group_separator);
+
+//////////////////////////////////////////////////////////////////
+// matching
+
+RV_GLOBAL bool32 rv_str8_match(rv_str8 a, rv_str8 b, rv_str8_match_flags flags)
+{
+ bool32 result = 0;
+ if (a.size == b.size || (flags & rv_str8_match_flag_right_side_sloppy)){
+ bool32 case_insensitive = (flags & rv_str8_match_flag_case_insensetive);
+ bool32 slash_insensitive = (flags & rv_str8_match_flag_slash_insensetive);
+ s64 size = rv_min(a.size, b.size);
+ result = 1;
+ for (s64 i = 0; i < size; i += 1){
+ u8 at = a.str[i];
+ u8 bt = b.str[i];
+ if (case_insensitive){
+ at = rv_char_to_upper(at);
+ bt = rv_char_to_upper(bt);
+ }
+ if (slash_insensitive){
+ at = rv_char_to_correct_slash(at);
+ bt = rv_char_to_correct_slash(bt);
+ }
+ if (at != bt){
+ result = 0;
+ break;
+ }
+ }
+ }
+ return result;
+}
+
+RV_GLOBAL s64 rv_str8_find_needle(rv_str8 string, s64 start_pos, rv_str8 needle, rv_str8_match_flags flags)
+{
+ u8* p = string.str + start_pos;
+ s64 stop_offset = rv_max(string.size + 1, needle.size) - needle.size;
+ u8* stop_p = string.str + stop_offset;
+
+ if (string.size > 0 && needle.size > 0) {
+ u8* string_opl = string.str + string.size;
+
+ rv_str8 needle_tail = rv_str8_skip(needle, 1);
+ rv_str8_match_flags adjusted_flags = flags | rv_str8_match_flag_right_side_sloppy;
+ u8 needle_first_rv_char_adjusted = needle.str[0];
+
+ if (adjusted_flags & rv_str8_match_flag_case_insensetive) {
+ needle_first_rv_char_adjusted = rv_char_to_upper(needle_first_rv_char_adjusted);
+ }
+
+ for (;p < stop_p; p += 1) {
+ u8 haystack_rv_char_adjusted = *p;
+
+ if (adjusted_flags & rv_str8_match_flag_case_insensetive) {
+ haystack_rv_char_adjusted = rv_char_to_upper(haystack_rv_char_adjusted);
+ }
+
+ if (haystack_rv_char_adjusted == needle_first_rv_char_adjusted) {
+ if (rv_str8_match(rv_str8_range(p + 1, string_opl), needle_tail, adjusted_flags)) {
+ break;
+ }
+ }
+ }
+ }
+ s64 result = string.size;
+ if (p < stop_p) {
+ result = (s64)(p - string.str);
+ }
+ return result;
+}
+
+//////////////////////////////////////////////////////////////////
+// slicing
+
+RV_GLOBAL rv_str8 rv_str8_substr(rv_str8 str, rv_range range)
+{
+ range.min = rv_min(range.min, str.size);
+ range.max = rv_min(range.max, str.size);
+ str.str += range.min;
+ str.size = rv_range_diff(range);
+ return str;
+}
+
+RV_GLOBAL rv_str8 rv_str8_prefix(rv_str8 str, s64 size)
+{
+ str.size = rv_min(size, str.size);
+ return str;
+}
+
+RV_GLOBAL rv_str8 rv_str8_skip(rv_str8 str, s64 amt)
+{
+ amt = rv_min(amt, str.size);
+ str.str += amt;
+ str.size -= amt;
+ return str;
+}
+
+RV_GLOBAL rv_str8 rv_str8_postfix(rv_str8 str, s64 size)
+{
+ size = rv_min(size, str.size);
+ str.str = (str.str + str.size) - size;
+ str.size = size;
+ return str;
+}
+
+RV_GLOBAL rv_str8 rv_str8_chop(rv_str8 str, s64 amt)
+{
+ amt = rv_min(amt, str.size);
+ str.size -= amt;
+ return str;
+}
+
+RV_GLOBAL rv_str8 rv_str8_skip_chop_whitespace(rv_str8 string)
+{
+ u8* first = string.str;
+ u8* opl = first + string.size;
+ for (;first < opl; first += 1) {
+ if (!rv_char_is_space(*first)) {
+ break;
+ }
+ }
+ for (;opl > first;) {
+ opl -= 1;
+ if (!rv_char_is_space(*opl)) {
+ opl += 1;
+ break;
+ }
+ }
+ rv_str8 result = rv_str8_range(first, opl);
+ return result;
+}
+
+//////////////////////////////////////////////////////////////////
+// misc
+
+RV_GLOBAL rv_str8 rv_str8_copy(rv_arena* arena, rv_str8 s)
+{
+ rv_str8 str;
+ str.size = s.size;
+ str.str = rv_push_array_no_zero(arena, u8, str.size + 1);
+ rv_mem_copy(str.str, s.str, s.size);
+ str.str[str.size] = 0;
+ return str;
+}
+
+RV_GLOBAL rv_str8 rv_read_file(rv_arena* arena, rv_str8 filename)
+{
+ u8* buffer = NULL;
+ s64 size = 0;
+
+ FILE* fp = NULL;
+ {
+ rv_temp_arena scratch = rv_scratch_begin((rv_arena*[]){arena}, 1);
+ rv_str8 filename_copy = rv_str8_copy(scratch.arena, filename);
+ fp = fopen((char*)filename_copy.str, "rb");
+ rv_scratch_end(scratch);
+ }
+
+ if (fp) {
+ fseek(fp, 0, SEEK_END);
+ size = ftell(fp);
+ fseek(fp, 0, SEEK_SET);
+
+ buffer = rv_push_array_no_zero(arena, u8, size + 1);
+ fread(buffer, 1, size, fp);
+
+ fclose(fp);
+ }
+
+ return rv_s8(buffer, size);
+}
diff --git a/src/platform/platform_string.h b/src/platform/platform_string.h
@@ -0,0 +1,65 @@
+//////////////////////////////////////////////////////////////////
+// platform_string.h
+
+
+//////////////////////////////////////////////////////////////////
+// char
+
+
+RV_GLOBAL bool32 rv_char_is_space(u8 c);
+RV_GLOBAL bool32 rv_char_is_upper(u8 c);
+RV_GLOBAL bool32 rv_char_is_lower(u8 c);
+RV_GLOBAL bool32 rv_char_is_alpha(u8 c);
+RV_GLOBAL bool32 rv_char_is_slash(u8 c);
+RV_GLOBAL bool32 rv_char_is_digit(u8 c, s32 base);
+RV_GLOBAL u8 rv_char_to_lower(u8 c);
+RV_GLOBAL u8 rv_char_to_upper(u8 c);
+RV_GLOBAL u8 rv_char_to_correct_slash(u8 c);
+
+//////////////////////////////////////////////////////////////////
+// c strings
+
+RV_GLOBAL s64 rv_cstrlen(const char* c_str);
+
+//////////////////////////////////////////////////////////////////
+// constructors
+
+#define S(str) (rv_str8){(u8*)("" str ""), sizeof("" str "")-1}
+#define rv_s8(str, size) (rv_str8){str, size}
+#define rv_str8_from_cstr(cstr) rv_s8((u8*)cstr, rv_cstrlen((char*)cstr))
+#define rv_str8ify(...) S(#__VA_ARGS__)
+RV_GLOBAL rv_str8 rv_str8_range(u8* first, u8* one_past_last);
+
+// adds 0 terminator, use this to convert to c string
+RV_GLOBAL rv_str8 rv_str8_copy(rv_arena* arena, rv_str8 s);
+
+//////////////////////////////////////////////////////////////////
+// conversions
+
+#define rv_s8v(S) (int)((S).size), ((S).str)
+
+RV_GLOBAL s64 rv_sign_from_str8(rv_str8 string, rv_str8* string_tail);
+RV_GLOBAL u64 rv_u64_from_str8(rv_str8 string, u32 radix);
+RV_GLOBAL s64 rv_s64_from_str8(rv_str8 string, s32 radix);
+RV_GLOBAL rv_str8 rv_str8_from_s64(rv_arena* arena, s64 integer, s32 radix, u8 min_digits, u8 digit_group_separator);
+
+//////////////////////////////////////////////////////////////////
+// matching
+
+RV_GLOBAL bool32 rv_str8_match(rv_str8 a, rv_str8 b, rv_str8_match_flags flags);
+RV_GLOBAL s64 rv_str8_find_needle(rv_str8 string, s64 start_pos, rv_str8 needle, rv_str8_match_flags flags);
+
+//////////////////////////////////////////////////////////////////
+// slicing
+
+RV_GLOBAL rv_str8 rv_str8_substr(rv_str8 str, rv_range range);
+RV_GLOBAL rv_str8 rv_str8_prefix(rv_str8 str, s64 size);
+RV_GLOBAL rv_str8 rv_str8_skip(rv_str8 str, s64 amt);
+RV_GLOBAL rv_str8 rv_str8_postfix(rv_str8 str, s64 size);
+RV_GLOBAL rv_str8 rv_str8_chop(rv_str8 str, s64 amt);
+RV_GLOBAL rv_str8 rv_str8_skip_chop_whitespace(rv_str8 string);
+
+//////////////////////////////////////////////////////////////////
+// misc
+
+RV_GLOBAL rv_str8 rv_read_file(rv_arena* arena, rv_str8 filename);
diff --git a/src/platform/platform_types.h b/src/platform/platform_types.h
@@ -2,6 +2,7 @@
// platform_types.h
#include <stdint.h>
+#include <stddef.h>
typedef uint8_t u8;
typedef uint16_t u16;
@@ -18,11 +19,17 @@ typedef s64 bool64;
typedef float f32;
typedef double f64;
+typedef union {
+ struct {
+ s64 min, max;
+ };
+ s64 v[2];
+} rv_range;
+
+// vector / matrix math
-typedef struct
-{
- union
- {
+typedef struct {
+ union {
// f32
f32 xy[2];
struct {
@@ -30,7 +37,8 @@ typedef struct
};
};
} rv_vec2;
-#define rv_v2(...) ((rv_vec2){__VA_ARGS__})
+#define rv_v2(...) ((rv_vec2){__VA_ARGS__})
+#define rv_v2s(s) ((rv_vec2){(s), (s)})
typedef struct {
union {
@@ -53,6 +61,7 @@ typedef struct {
} rv_vec3;
#define rv_v3(...) ((rv_vec3){__VA_ARGS__})
+#define rv_v3s(s) ((rv_vec3){(s), (s), (s)})
typedef struct {
union {
@@ -84,4 +93,78 @@ typedef struct {
};
};
} rv_vec4;
-#define rv_v4(...) ((rv_vec4){__VA_ARGS__})
+#define rv_v4(...) ((rv_vec4){__VA_ARGS__})
+#define rv_v4s(...) (rv_vec4){(s), (s), (s), (s)})
+
+typedef struct gs_mat4 {
+ union {
+ rv_vec4 rows[4];
+ f32 m[4][4];
+ f32 elements[16];
+ struct {
+ rv_vec4 right, up, dir, position;
+ } v;
+ };
+} rv_mat4;
+#define rv_mat4_diag(v) (rv_mat4){ .elements = {\
+ v, 0, 0, 0,\
+ 0, v, 0, 0,\
+ 0, 0, v, 0,\
+ 0, 0, 0, v,\
+ }}
+#define rv_mat4_identity()\
+ rv_mat4_diag(1.0f)
+
+typedef union {
+ f32 xywh[4];
+ struct {
+ f32 x, y, w, h;
+ };
+ struct {
+ rv_vec2 xy;
+ rv_vec2 wh;
+ };
+ rv_vec4 vec4;
+} rv_rect;
+
+typedef struct {
+ union {
+ u8 rgba[4];
+ struct {
+ u8 r, g, b, a;
+ };
+ };
+} rv_color_t;
+#define rv_color(...) (rv_color_t){__VA_ARGS__}
+
+#define RV_COLOR_BLACK rv_color(0, 0, 0, 255)
+#define RV_COLOR_WHITE rv_color(255, 255, 255, 255)
+#define RV_COLOR_RED rv_color(255, 0, 0, 255)
+#define RV_COLOR_GREEN rv_color(0, 255, 0, 255)
+#define RV_COLOR_BLUE rv_color(0, 0, 255, 255)
+#define RV_COLOR_ORANGE rv_color(255, 100, 0, 255)
+#define RV_COLOR_PURPLE rv_color(128, 0, 128, 255)
+
+
+// strings
+
+typedef struct {
+ u8* str;
+ s64 size;
+} rv_str8;
+
+typedef struct rv_str8_node_t rv_str8_node_t;
+struct rv_str8_node_t {
+ rv_str8_node_t* next;
+ rv_str8 str;
+};
+typedef struct {
+ rv_str8_node_t* first;
+ rv_str8_node_t* last;
+} rv_str8_list_t;
+
+typedef enum {
+ rv_str8_match_flag_case_insensetive = (1 << 0),
+ rv_str8_match_flag_right_side_sloppy = (1 << 1),
+ rv_str8_match_flag_slash_insensetive = (1 << 2),
+} rv_str8_match_flags;
diff --git a/src/render/impl/opengl.c b/src/render/impl/opengl.c
@@ -1,46 +1,441 @@
//////////////////////////////////////////////////////////////////
// opengl.c
-#include <GL/glcorearb.h>
-#include <GL/gl.h>
-
+struct {
+ GLuint vao;
+ bool32 is_init;
+} ogl_ctx;
//////////////////////////////////////////////////////////////////
// Command buffer api
+RV_INTERNAL GLenum rv_usage_to_gl(rv_buffer_usage_t usage) {
+ GLenum mode = GL_STATIC_DRAW;
+ switch (usage) {
+ default:
+ case RV_BUFFER_USAGE_STATIC: mode = GL_STATIC_DRAW; break;
+ case RV_BUFFER_USAGE_DYNAMIC: mode = GL_DYNAMIC_DRAW; break;
+ case RV_BUFFER_USAGE_STREAM: mode = GL_STREAM_DRAW; break;
+ }
+ return mode;
+}
+
+RV_INTERNAL GLenum rv_shader_to_gl(rv_shader_type_t type) {
+ GLenum mode = 0;
+ switch (type) {
+ default:
+ case RV_SHADER_TYPE_VERTEX: mode = GL_VERTEX_SHADER; break;
+ case RV_SHADER_TYPE_FRAGMENT: mode = GL_FRAGMENT_SHADER; break;
+ }
+ return mode;
+}
+
+RV_INTERNAL GLenum rv_primitive_to_gl(rv_primitive_type_t type) {
+ GLenum res = 0;
+ switch (type) {
+ default:
+ case RV_PRIMITIVE_TYPE_TRIANGLES: res = GL_TRIANGLES; break;
+ }
+ return res;
+}
+
+RV_INTERNAL s64 rv_vattr_type_size(rv_vattr_type_t type) {
+ s64 size = 0;
+ switch (type) {
+ case RV_VATTR_TYPE_FLOAT4: { size = sizeof(f32) * 4; } break;
+ case RV_VATTR_TYPE_FLOAT3: { size = sizeof(f32) * 3; } break;
+ case RV_VATTR_TYPE_FLOAT2: { size = sizeof(f32) * 2; } break;
+ case RV_VATTR_TYPE_FLOAT: { size = sizeof(f32) * 1; } break;
+ case RV_VATTR_TYPE_UINT4: { size = sizeof(u32) * 4; } break;
+ case RV_VATTR_TYPE_UINT3: { size = sizeof(u32) * 3; } break;
+ case RV_VATTR_TYPE_UINT2: { size = sizeof(u32) * 2; } break;
+ case RV_VATTR_TYPE_UINT: { size = sizeof(u32) * 1; } break;
+ case RV_VATTR_TYPE_BYTE4: { size = sizeof(u8 ) * 4; } break;
+ case RV_VATTR_TYPE_BYTE3: { size = sizeof(u8 ) * 3; } break;
+ case RV_VATTR_TYPE_BYTE2: { size = sizeof(u8 ) * 2; } break;
+ case RV_VATTR_TYPE_BYTE: { size = sizeof(u8 ) * 1; } break;
+ }
+ return size;
+}
+
+
+RV_INTERNAL GLenum rv_tex_wrap_to_gl(rv_texture_wrap_t type) {
+ uint32_t wrap = GL_REPEAT;
+ switch (type) {
+ default:
+ case RV_TEXTURE_WRAP_REPEAT: wrap = GL_REPEAT; break;
+ case RV_TEXTURE_WRAP_MIRRORED_REPEAT: wrap = GL_MIRRORED_REPEAT; break;
+ case RV_TEXTURE_WRAP_CLAMP_TO_EDGE: wrap = GL_CLAMP_TO_EDGE; break;
+ };
+
+ return wrap;
+}
+
+
RV_GLOBAL void rv_window_render_commit(rv_window_handle_t* window, rv_render_pass_list_t* passes)
{
- rv_window_render_begin(window);
+ rv_window_render_begin(window);
+
+ if (!ogl_ctx.is_init) {
+ ogl_ctx.is_init = true;
+
+ glGenVertexArrays(1, &ogl_ctx.vao);
+ glBindVertexArray(ogl_ctx.vao);
+ }
+
+ struct {
+ rv_ibo_t* bound_ibo;
+ rv_vbo_t* bound_vbo;
+ } state = {0};
+
+ for (rv_render_pass_t* p = passes->first; p; p = p->next) {
+ for (rv_render_command_t* c = p->commands.first; c; c = c->next) {
+ switch (c->type) {
+ case RV_RENDER_COMMAND_INVALID: {
+ } break;
- for (rv_render_pass_t* p = passes->first; p; p = p->next) {
- for (rv_render_command_t* c = p->first; c; c = c->next) {
- switch (c->type) {
- case RV_RENDER_COMMAND_INVALID: {
- } break;
+ case RV_RENDER_COMMAND_CUSTOM: {
+ rv_assert(c->custom.func);
+ if (c->custom.func) {
+ c->custom.func(c->custom.data);
+ }
+ } break;
- case RV_RENDER_COMMAND_CLEAR: {
- for (rv_render_clear_desc_t* clear = c->clear.first; clear; clear = clear->next) {
- if (clear->flag == RV_RENDER_CLEAR_FLAG_NONE) continue;
+ case RV_RENDER_COMMAND_CLEAR: {
+ for (rv_render_clear_desc_t* clear = c->clear.first; clear; clear = clear->next) {
+ if (clear->flags == RV_RENDER_CLEAR_FLAG_NONE) continue;
- int opengl_clear_mask = 0;
- if (clear->flag & RV_RENDER_CLEAR_FLAG_COLOR) {
+ int opengl_clear_mask = 0;
+ if (clear->flags & RV_RENDER_CLEAR_FLAG_COLOR) {
glClearColor(clear->color[0], clear->color[1], clear->color[2], clear->color[3]);
- opengl_clear_mask |= GL_COLOR_BUFFER_BIT;
- }
- if (clear->flag & RV_RENDER_CLEAR_FLAG_DEPTH) opengl_clear_mask |= GL_DEPTH_BUFFER_BIT;
- if (clear->flag & RV_RENDER_CLEAR_FLAG_STENCIL) opengl_clear_mask |= GL_STENCIL_BUFFER_BIT;
+ opengl_clear_mask |= GL_COLOR_BUFFER_BIT;
+ }
+ if (clear->flags & RV_RENDER_CLEAR_FLAG_DEPTH) opengl_clear_mask |= GL_DEPTH_BUFFER_BIT;
+ if (clear->flags & RV_RENDER_CLEAR_FLAG_STENCIL) opengl_clear_mask |= GL_STENCIL_BUFFER_BIT;
- glClear(opengl_clear_mask);
- }
- } break;
+ glClear(opengl_clear_mask);
+ }
+ } break;
- case RV_RENDER_COMMAND_SET_VIEWPORT: {
+ case RV_RENDER_COMMAND_SET_VIEWPORT: {
glViewport(c->viewport.x, c->viewport.y, c->viewport.z, c->viewport.w);
- } break;
+ } break;
+
+ case RV_RENDER_COMMAND_OBJ_VERTEX: {
+ rv_vbo_t* vbo = c->obj.vbo;
+ switch (c->obj.operation) {
+ case RV_RENDER_OBJ_OP_CREATE: {
+ glGenBuffers(1, &vbo->handle.u);
+ glBindBuffer(GL_ARRAY_BUFFER, vbo->handle.u);
+ glBufferData(GL_ARRAY_BUFFER, vbo->size, vbo->data, rv_usage_to_gl(vbo->usage));
+ glBindBuffer(GL_ARRAY_BUFFER, 0);
+
+ } break;
+ case RV_RENDER_OBJ_OP_UPDATE: {
+ glBindBuffer(GL_ARRAY_BUFFER, vbo->handle.u);
+ glBufferData(GL_ARRAY_BUFFER, c->obj.vbo_update.size, c->obj.vbo_update.data, rv_usage_to_gl(vbo->usage));
+ if (state.bound_vbo) {
+ glBindBuffer(GL_ARRAY_BUFFER, state.bound_vbo->handle.u);
+ }
+ } break;
+ case RV_RENDER_OBJ_OP_BIND: {
+ if (vbo) {
+ glBindBuffer(GL_ARRAY_BUFFER, vbo->handle.u);
+ } else {
+ glBindBuffer(GL_ARRAY_BUFFER, vbo->handle.u);
+ }
+ state.bound_vbo = vbo;
+ } break;
+ case RV_RENDER_OBJ_OP_DELETE: {
+ glDeleteBuffers(1, &vbo->handle.u);
+ vbo->handle.u = 0;
+ } break;
+ }
+ } break;
+
+ case RV_RENDER_COMMAND_OBJ_INDEX: {
+ rv_ibo_t* ibo = c->obj.ibo;
+ switch (c->obj.operation) {
+ case RV_RENDER_OBJ_OP_CREATE: {
+ glGenBuffers(1, &ibo->handle.u);
+ glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, ibo->handle.u);
+ glBufferData(GL_ELEMENT_ARRAY_BUFFER, ibo->size, ibo->data, rv_usage_to_gl(ibo->usage));
+ glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, 0);
+ } break;
+ case RV_RENDER_OBJ_OP_UPDATE: {
+ glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, ibo->handle.u);
+ glBufferData(GL_ELEMENT_ARRAY_BUFFER, c->obj.ibo_update.size, c->obj.ibo_update.data, rv_usage_to_gl(ibo->usage));
+ if (state.bound_ibo) {
+ glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, state.bound_ibo->handle.u);
+ }
+ } break;
+ case RV_RENDER_OBJ_OP_BIND: {
+ if (ibo) {
+ glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, ibo->handle.u);
+ } else {
+ glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, 0);
+ }
+ state.bound_ibo = ibo;
+ } break;
+ case RV_RENDER_OBJ_OP_DELETE: {
+ glDeleteBuffers(1, &ibo->handle.u);
+ ibo->handle.u = 0;
+ } break;
+ }
+ } break;
+
+ case RV_RENDER_COMMAND_OBJ_SHADER: {
+ rv_shader_t* shader = c->obj.shader;
+ switch (c->obj.operation) {
+ case RV_RENDER_OBJ_OP_CREATE: {
+ shader->handle.u = glCreateShader(rv_shader_to_gl(shader->type));
+ glShaderSource(shader->handle.u, 1, (const char* const*)&shader->source.str, (s32*)&shader->source.size);
+ glCompileShader(shader->handle.u);
+
+ { // check for errors
+ int shader_compile_success;
+ glGetShaderiv(shader->handle.u, GL_COMPILE_STATUS, &shader_compile_success);
+
+ if (!shader_compile_success) {
+ char info_log[512];
+ glGetShaderInfoLog(shader->handle.u, 512, NULL, info_log);
+ fprintf(stderr, "-------------ERROR------------\n"
+ "::OpenGL Failed to compile shader::\n%s\n", info_log);
+ fprintf(stderr, "-------------SOURCE------------\n");
+ fprintf(stderr, "%.*s\n", rv_s8v(shader->source));
+ fprintf(stderr, "\n------------END_SOURCE----------\n");
+ rv_assert(shader_compile_success);
+ }
+ }
+ } break;
+ case RV_RENDER_OBJ_OP_UPDATE: {
+ rv_assert(false);
+ } break;
+ case RV_RENDER_OBJ_OP_BIND: {
+ rv_assert(false);
+ } break;
+ case RV_RENDER_OBJ_OP_DELETE: {
+ glDeleteShader(shader->handle.u);
+ shader->handle.u = 0;
+ } break;
+ }
+ } break;
+
+ case RV_RENDER_COMMAND_OBJ_PROGRAM: {
+ rv_program_t* program = c->obj.program;
+ switch (c->obj.operation) {
+ case RV_RENDER_OBJ_OP_CREATE: {
+ program->handle.u = glCreateProgram();
+ for (rv_shader_node_t* s = program->shader_first; s; s = s->next) {
+ glAttachShader(program->handle.u, s->shader->handle.u);
+ }
+ glLinkProgram(program->handle.u);
+
+ {
+ int program_link_success;
+ glGetProgramiv(program->handle.u, GL_LINK_STATUS, &program_link_success);
+
+ if (!program_link_success) {
+ char info_log[512];
+ glGetProgramInfoLog(program->handle.u, 512, NULL, info_log);
+ fprintf(stderr, "-------------ERROR------------\n"
+ "::OpenGL Failed to link program::\n%s\n", info_log);
+ rv_assert(program_link_success);
+ }
+ }
+
+ // find uniform locations
+ for (rv_uniform_t* u = program->uniform_desc_first; u; u = u->next) {
+ u->program_uniform_handle.u = glGetUniformLocation(program->handle.u, (const char*)u->name_null_terminated.str);
+ }
+ } break;
+ case RV_RENDER_OBJ_OP_UPDATE: {
+ rv_assert(false);
+ } break;
+ case RV_RENDER_OBJ_OP_BIND: {
+ rv_uniform_update_node_t* committed_uniforms = c->obj.program_bind.committed_uniforms;
+ glUseProgram(program->handle.u);
+ s32 sampler_channel;
+
+ // update last type
+ for (rv_uniform_update_node_t* u = committed_uniforms; u; u = u->next) {
+ u->desc->last_val = u->value;
+ }
+
+ // activate textures and sample2d, if any
+ sampler_channel = 0;
+ for (rv_uniform_t* u = program->uniform_desc_first; u; u = u->next) {
+ if (u->type == RV_UNIFORM_TEX) {
+ glActiveTexture(GL_TEXTURE0 + sampler_channel);
+ glBindTexture(GL_TEXTURE_2D, u->last_val.v_tex->handle.u);
+ if (!u->has_been_set) {
+ glUniform1i(u->program_uniform_handle.u, sampler_channel);
+ u->has_been_set = true;
+ }
+ u->last_val.v_tex->last_sampler_channel = sampler_channel;
+ sampler_channel++;
+ }
+ }
+
+ for (rv_uniform_update_node_t* u = committed_uniforms; u; u = u->next) {
+ switch (u->desc->type) {
+ case RV_UNIFORM_F32: glUniform1fv(u->desc->program_uniform_handle.u, 1, &u->value.v_f32); break;
+ case RV_UNIFORM_VEC2: glUniform2fv(u->desc->program_uniform_handle.u, 1, u->value.v_vec2.xy); break;
+ case RV_UNIFORM_VEC3: glUniform3fv(u->desc->program_uniform_handle.u, 1, u->value.v_vec3.xyz); break;
+ case RV_UNIFORM_VEC4: glUniform4fv(u->desc->program_uniform_handle.u, 1, u->value.v_vec4.xyzw); break;
+ case RV_UNIFORM_MAT4: glUniformMatrix4fv(u->desc->program_uniform_handle.u, 1, GL_FALSE, u->value.v_mat4.elements); break;
+ case RV_UNIFORM_TEX: break; // handled above
+ case RV_UNIFORM_INVALID: break;
+ }
+ u->desc->has_been_set = true;
+ }
+
+ s64 stride = 0;
+ for (rv_vattr_t* v = program->vattr_first; v; v = v->next) {
+ stride += rv_vattr_type_size(v->type);
+ }
+ s64 offset = 0;
+ s64 i = 0;
+ for (rv_vattr_t* v = program->vattr_first; v; v = v->next, i++) {
+ glEnableVertexAttribArray(i);
+ switch (v->type) {
+ case RV_VATTR_TYPE_FLOAT4: glVertexAttribPointer(i, 4, GL_FLOAT, GL_FALSE, stride, (void*)offset); break;
+ case RV_VATTR_TYPE_FLOAT3: glVertexAttribPointer(i, 3, GL_FLOAT, GL_FALSE, stride, (void*)offset); break;
+ case RV_VATTR_TYPE_FLOAT2: glVertexAttribPointer(i, 2, GL_FLOAT, GL_FALSE, stride, (void*)offset); break;
+ case RV_VATTR_TYPE_FLOAT: glVertexAttribPointer(i, 1, GL_FLOAT, GL_FALSE, stride, (void*)offset); break;
+ case RV_VATTR_TYPE_UINT4: glVertexAttribIPointer(i, 4, GL_UNSIGNED_INT, stride, (void*)offset); break;
+ case RV_VATTR_TYPE_UINT3: glVertexAttribIPointer(i, 3, GL_UNSIGNED_INT, stride, (void*)offset); break;
+ case RV_VATTR_TYPE_UINT2: glVertexAttribIPointer(i, 2, GL_UNSIGNED_INT, stride, (void*)offset); break;
+ case RV_VATTR_TYPE_UINT: glVertexAttribIPointer(i, 1, GL_UNSIGNED_INT, stride, (void*)offset); break;
+ case RV_VATTR_TYPE_BYTE: glVertexAttribPointer(i, 1, GL_UNSIGNED_BYTE, GL_TRUE, stride, (void*)offset); break;
+ case RV_VATTR_TYPE_BYTE2: glVertexAttribPointer(i, 2, GL_UNSIGNED_BYTE, GL_TRUE, stride, (void*)offset); break;
+ case RV_VATTR_TYPE_BYTE3: glVertexAttribPointer(i, 3, GL_UNSIGNED_BYTE, GL_TRUE, stride, (void*)offset); break;
+ case RV_VATTR_TYPE_BYTE4: glVertexAttribPointer(i, 4, GL_UNSIGNED_BYTE, GL_TRUE, stride, (void*)offset); break;
+ default: rv_assert(false); break;
+ }
+
+ offset += rv_vattr_type_size(v->type);
+ }
+ } break;
+ case RV_RENDER_OBJ_OP_DELETE: {
+ glDeleteProgram(program->handle.u);
+ program->handle.u = 0;
+ for (rv_uniform_t* u = program->uniform_desc_first; u; u = u->next) {
+ u->program_uniform_handle.u = 0;
+ }
+ } break;
+ }
+ } break;
+
+ case RV_RENDER_COMMAND_OBJ_TEXTURE: {
+ rv_texture_t* tex = c->obj.tex;
+ switch (c->obj.operation) {
+ case RV_RENDER_OBJ_OP_CREATE: {
+ glGenTextures(1, &tex->handle.u);
+ glActiveTexture(GL_TEXTURE0 + tex->last_sampler_channel);
+ glBindTexture(GL_TEXTURE_2D, tex->handle.u);
+
+ switch(tex->format)
+ {
+ case RV_TEXTURE_FORMAT_A8: glTexImage2D(GL_TEXTURE_2D, 0, GL_ALPHA, tex->size.x, tex->size.y, 0, GL_ALPHA, GL_UNSIGNED_BYTE, tex->data); break;
+ case RV_TEXTURE_FORMAT_R8: glTexImage2D(GL_TEXTURE_2D, 0, GL_RED, tex->size.x, tex->size.y, 0, GL_RED, GL_UNSIGNED_BYTE, tex->data); break;
+ case RV_TEXTURE_FORMAT_RGB8: glTexImage2D(GL_TEXTURE_2D, 0, GL_RGB8, tex->size.x, tex->size.y, 0, GL_RGB, GL_UNSIGNED_BYTE, tex->data); break;
+ case RV_TEXTURE_FORMAT_RGBA8: glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA8, tex->size.x, tex->size.y, 0, GL_RGBA, GL_UNSIGNED_BYTE, tex->data); break;
+ case RV_TEXTURE_FORMAT_RGBA16F: glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA16F, tex->size.x, tex->size.y, 0, GL_RGBA, GL_FLOAT, tex->data); break;
+ case RV_TEXTURE_FORMAT_RGBA32F: glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA32F, tex->size.x, tex->size.y, 0, GL_RGBA, GL_FLOAT, tex->data); break;
+ case RV_TEXTURE_FORMAT_DEPTH8: glTexImage2D(GL_TEXTURE_2D, 0, GL_DEPTH_COMPONENT, tex->size.x, tex->size.y, 0, GL_DEPTH_COMPONENT, GL_FLOAT, tex->data); break;
+ case RV_TEXTURE_FORMAT_DEPTH16: glTexImage2D(GL_TEXTURE_2D, 0, GL_DEPTH_COMPONENT16, tex->size.x, tex->size.y, 0, GL_DEPTH_COMPONENT, GL_FLOAT, tex->data); break;
+ case RV_TEXTURE_FORMAT_DEPTH24: glTexImage2D(GL_TEXTURE_2D, 0, GL_DEPTH_COMPONENT24, tex->size.x, tex->size.y, 0, GL_DEPTH_COMPONENT, GL_FLOAT, tex->data); break;
+ case RV_TEXTURE_FORMAT_DEPTH32F: glTexImage2D(GL_TEXTURE_2D, 0, GL_DEPTH_COMPONENT32F, tex->size.x, tex->size.y, 0, GL_DEPTH_COMPONENT, GL_FLOAT, tex->data); break;
+ case RV_TEXTURE_FORMAT_DEPTH24_STENCIL8: glTexImage2D(GL_TEXTURE_2D, 0, GL_DEPTH24_STENCIL8, tex->size.x, tex->size.y, 0, GL_DEPTH_STENCIL, GL_UNSIGNED_INT_24_8, tex->data); break;
+ case RV_TEXTURE_FORMAT_DEPTH32F_STENCIL8: glTexImage2D(GL_TEXTURE_2D, 0, GL_DEPTH32F_STENCIL8, tex->size.x, tex->size.y, 0, GL_DEPTH_STENCIL, GL_FLOAT_32_UNSIGNED_INT_24_8_REV, tex->data); break;
+ default: break;
+ }
+
+ int32_t mag_filter = tex->mag_filter == RV_TEXTURE_FILTER_NEAREST ? GL_NEAREST : GL_LINEAR;
+ int32_t min_filter = tex->min_filter == RV_TEXTURE_FILTER_NEAREST ? GL_NEAREST : GL_LINEAR;
+
+ if (tex->num_mips) {
+ if (tex->min_filter == RV_TEXTURE_FILTER_NEAREST) {
+ min_filter = tex->mip_filter == RV_TEXTURE_FILTER_NEAREST ? GL_NEAREST_MIPMAP_NEAREST :
+ GL_NEAREST_MIPMAP_LINEAR;
+ }
+ else {
+ min_filter = tex->mip_filter == RV_TEXTURE_FILTER_NEAREST ? GL_LINEAR_MIPMAP_NEAREST :
+ GL_NEAREST_MIPMAP_LINEAR;
+ }
+ }
+
+ uint32_t texture_wrap_s = rv_tex_wrap_to_gl(tex->wrap_s);
+ uint32_t texture_wrap_t = rv_tex_wrap_to_gl(tex->wrap_t);
+
+ if (tex->num_mips) {
+ glGenerateMipmap(GL_TEXTURE_2D);
+ }
+
+ glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, texture_wrap_s);
+ glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, texture_wrap_t);
+ glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, mag_filter);
+ glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, min_filter);
+
+ glBindTexture(GL_TEXTURE_2D, 0);
+
+ } break;
+ case RV_RENDER_OBJ_OP_UPDATE: {
+ glActiveTexture(GL_TEXTURE0 + tex->last_sampler_channel);
+ glBindTexture(GL_TEXTURE_2D, tex->handle.u);
+
+ switch(tex->format)
+ {
+ case RV_TEXTURE_FORMAT_A8: glTexSubImage2D(GL_TEXTURE_2D, 0, c->obj.tex_update.sub_part.x, c->obj.tex_update.sub_part.y, c->obj.tex_update.sub_part.w, c->obj.tex_update.sub_part.h, GL_ALPHA, GL_UNSIGNED_BYTE, c->obj.tex_update.data); break;
+ case RV_TEXTURE_FORMAT_R8: glTexSubImage2D(GL_TEXTURE_2D, 0, c->obj.tex_update.sub_part.x, c->obj.tex_update.sub_part.y, c->obj.tex_update.sub_part.w, c->obj.tex_update.sub_part.h, GL_RED, GL_UNSIGNED_BYTE, c->obj.tex_update.data); break;
+ case RV_TEXTURE_FORMAT_RGB8: glTexSubImage2D(GL_TEXTURE_2D, 0, c->obj.tex_update.sub_part.x, c->obj.tex_update.sub_part.y, c->obj.tex_update.sub_part.w, c->obj.tex_update.sub_part.h, GL_RGB, GL_UNSIGNED_BYTE, c->obj.tex_update.data); break;
+ case RV_TEXTURE_FORMAT_RGBA8: glTexSubImage2D(GL_TEXTURE_2D, 0, c->obj.tex_update.sub_part.x, c->obj.tex_update.sub_part.y, c->obj.tex_update.sub_part.w, c->obj.tex_update.sub_part.h, GL_RGBA, GL_UNSIGNED_BYTE, c->obj.tex_update.data); break;
+ case RV_TEXTURE_FORMAT_RGBA16F: glTexSubImage2D(GL_TEXTURE_2D, 0, c->obj.tex_update.sub_part.x, c->obj.tex_update.sub_part.y, c->obj.tex_update.sub_part.w, c->obj.tex_update.sub_part.h, GL_RGBA, GL_FLOAT, c->obj.tex_update.data); break;
+ case RV_TEXTURE_FORMAT_RGBA32F: glTexSubImage2D(GL_TEXTURE_2D, 0, c->obj.tex_update.sub_part.x, c->obj.tex_update.sub_part.y, c->obj.tex_update.sub_part.w, c->obj.tex_update.sub_part.h, GL_RGBA, GL_FLOAT, c->obj.tex_update.data); break;
+ case RV_TEXTURE_FORMAT_DEPTH8: glTexSubImage2D(GL_TEXTURE_2D, 0, c->obj.tex_update.sub_part.x, c->obj.tex_update.sub_part.y, c->obj.tex_update.sub_part.w, c->obj.tex_update.sub_part.h, GL_DEPTH_COMPONENT, GL_FLOAT, c->obj.tex_update.data); break;
+ case RV_TEXTURE_FORMAT_DEPTH16: glTexSubImage2D(GL_TEXTURE_2D, 0, c->obj.tex_update.sub_part.x, c->obj.tex_update.sub_part.y, c->obj.tex_update.sub_part.w, c->obj.tex_update.sub_part.h, GL_DEPTH_COMPONENT, GL_FLOAT, c->obj.tex_update.data); break;
+ case RV_TEXTURE_FORMAT_DEPTH24: glTexSubImage2D(GL_TEXTURE_2D, 0, c->obj.tex_update.sub_part.x, c->obj.tex_update.sub_part.y, c->obj.tex_update.sub_part.w, c->obj.tex_update.sub_part.h, GL_DEPTH_COMPONENT, GL_FLOAT, c->obj.tex_update.data); break;
+ case RV_TEXTURE_FORMAT_DEPTH32F: glTexSubImage2D(GL_TEXTURE_2D, 0, c->obj.tex_update.sub_part.x, c->obj.tex_update.sub_part.y, c->obj.tex_update.sub_part.w, c->obj.tex_update.sub_part.h, GL_DEPTH_COMPONENT, GL_FLOAT, c->obj.tex_update.data); break;
+ case RV_TEXTURE_FORMAT_DEPTH24_STENCIL8: glTexSubImage2D(GL_TEXTURE_2D, 0, c->obj.tex_update.sub_part.x, c->obj.tex_update.sub_part.y, c->obj.tex_update.sub_part.w, c->obj.tex_update.sub_part.h, GL_DEPTH_STENCIL, GL_UNSIGNED_INT_24_8, c->obj.tex_update.data); break;
+ case RV_TEXTURE_FORMAT_DEPTH32F_STENCIL8: glTexSubImage2D(GL_TEXTURE_2D, 0, c->obj.tex_update.sub_part.x, c->obj.tex_update.sub_part.y, c->obj.tex_update.sub_part.w, c->obj.tex_update.sub_part.h, GL_DEPTH_STENCIL, GL_FLOAT_32_UNSIGNED_INT_24_8_REV, c->obj.tex_update.data); break;
+ default: break;
+ }
+ } break;
+ case RV_RENDER_OBJ_OP_BIND: {
+ // uniforms bind textures automatically
+ // so you don't have to do this every time
+ glActiveTexture(GL_TEXTURE0 + c->obj.tex_bind.sampler_index);
+ glBindTexture(GL_TEXTURE_2D, tex->handle.u);
+ } break;
+ case RV_RENDER_OBJ_OP_DELETE: {
+ glDeleteTextures(1, &tex->handle.u);
+ tex->handle.u = 0;
+ } break;
+ }
+ } break;
+
+ case RV_RENDER_COMMAND_DRAW: {
+ GLenum prim = rv_primitive_to_gl(c->draw.primitive);
+ // TODO(Samdal): instancing...
+ if (state.bound_ibo) {
+
+ GLenum itype = GL_UNSIGNED_INT;
+ switch (state.bound_ibo->elem_size) {
+ default:
+ case 4: itype = GL_UNSIGNED_INT; break;
+ case 2: itype = GL_UNSIGNED_SHORT; break;
+ case 1: itype = GL_UNSIGNED_BYTE; break;
+ }
+ glDrawElements(prim, c->draw.count, itype, (void*)(s64)c->draw.first);
+ } else {
+ glDrawArrays(prim, c->draw.first, c->draw.count);
+ }
+ } break;
- }
- }
- }
+ }
+ }
+ }
- rv_window_render_end(window);
+ rv_window_render_end(window);
}
diff --git a/src/render/render.h b/src/render/render.h
@@ -4,12 +4,172 @@
//////////////////////////////////////////////////////////////////
// Render types
-typedef struct {void* internal;} render_handle_t;
+typedef union {void* vptr; u32 u;} rv_render_handle_t;
+
+typedef enum {
+ RV_TEXTURE_FORMAT_RGBA8,
+ RV_TEXTURE_FORMAT_RGB8,
+ RV_TEXTURE_FORMAT_RGBA16F,
+ RV_TEXTURE_FORMAT_RGBA32F,
+ RV_TEXTURE_FORMAT_A8,
+ RV_TEXTURE_FORMAT_R8,
+ RV_TEXTURE_FORMAT_DEPTH8,
+ RV_TEXTURE_FORMAT_DEPTH16,
+ RV_TEXTURE_FORMAT_DEPTH24,
+ RV_TEXTURE_FORMAT_DEPTH32F,
+ RV_TEXTURE_FORMAT_DEPTH24_STENCIL8,
+ RV_TEXTURE_FORMAT_DEPTH32F_STENCIL8,
+ RV_TEXTURE_FORMAT_STENCIL8
+} rv_texture_format_t;
+
+typedef enum {
+ RV_TEXTURE_WRAP_REPEAT,
+ RV_TEXTURE_WRAP_MIRRORED_REPEAT,
+ RV_TEXTURE_WRAP_CLAMP_TO_EDGE,
+} rv_texture_wrap_t;
+
+typedef enum {
+ RV_TEXTURE_FILTER_NEAREST,
+ RV_TEXTURE_FILTER_LINEAR
+} rv_texture_filter_t;
typedef struct {
- render_handle_t handle;
+ rv_vec2 size;
+ rv_texture_format_t format;
+ rv_texture_wrap_t wrap_s;
+ rv_texture_wrap_t wrap_t;
+ rv_texture_filter_t min_filter;
+ rv_texture_filter_t mag_filter;
+ rv_texture_filter_t mip_filter;
+ u32 num_mips;
+ void* data;
+
+ rv_render_handle_t handle;
+
+ s32 last_sampler_channel;
+} rv_texture_t;
+
+typedef struct {
+ rv_render_handle_t handle;
} rv_render_framebuffer_t;
+typedef enum {
+ RV_UNIFORM_INVALID,
+ RV_UNIFORM_F32,
+ RV_UNIFORM_VEC2,
+ RV_UNIFORM_VEC3,
+ RV_UNIFORM_VEC4,
+ RV_UNIFORM_MAT4,
+ RV_UNIFORM_TEX,
+} rv_uniform_type_t;
+typedef union {
+ f32 v_f32;
+ rv_vec2 v_vec2;
+ rv_vec3 v_vec3;
+ rv_vec4 v_vec4;
+ rv_mat4 v_mat4;
+ rv_texture_t* v_tex;
+} rv_uniform_variant_t;
+
+typedef struct rv_uniform_t rv_uniform_t;
+struct rv_uniform_t {
+ // NOTE(Samdal): uniforms are specific to their corresponding program
+ rv_uniform_t* next;
+
+ // create info
+ // uniforms are created when it's program is created
+ rv_str8 name_null_terminated; // MUST be null terminated
+
+ rv_uniform_type_t type;
+
+ rv_render_handle_t program_uniform_handle;
+ rv_uniform_variant_t last_val;
+ bool32 has_been_set;
+};
+
+typedef struct rv_uniform_update_node_t rv_uniform_update_node_t;
+struct rv_uniform_update_node_t {
+ rv_uniform_update_node_t* next;
+ rv_uniform_variant_t value;
+ rv_uniform_t* desc;
+};
+
+typedef enum {
+ RV_BUFFER_USAGE_STATIC,
+ RV_BUFFER_USAGE_DYNAMIC,
+ RV_BUFFER_USAGE_STREAM,
+} rv_buffer_usage_t;
+
+typedef struct rv_vbo_t rv_vbo_t;
+struct rv_vbo_t {
+ rv_vbo_t* next;
+ void* data;
+ s32 size;
+ rv_buffer_usage_t usage;
+ rv_render_handle_t handle;
+};
+
+typedef struct rv_ibo_t rv_ibo_t;
+struct rv_ibo_t {
+ rv_ibo_t* next;
+ void* data;
+ s32 size;
+ s32 elem_size;
+ rv_buffer_usage_t usage;
+ rv_render_handle_t handle;
+};
+
+typedef enum {
+ RV_SHADER_TYPE_INVALID,
+ RV_SHADER_TYPE_VERTEX,
+ RV_SHADER_TYPE_FRAGMENT,
+ //RV_SHADER_TYPE_COMPUTE
+} rv_shader_type_t;
+
+typedef struct {
+ rv_str8 source;
+
+ rv_shader_type_t type;
+ rv_render_handle_t handle;
+} rv_shader_t;
+typedef struct rv_shader_node_t rv_shader_node_t;
+struct rv_shader_node_t {
+ rv_shader_node_t* next;
+ rv_shader_t* shader;
+};
+
+typedef enum {
+ RV_VATTR_TYPE_FLOAT4,
+ RV_VATTR_TYPE_FLOAT3,
+ RV_VATTR_TYPE_FLOAT2,
+ RV_VATTR_TYPE_FLOAT,
+ RV_VATTR_TYPE_UINT4,
+ RV_VATTR_TYPE_UINT3,
+ RV_VATTR_TYPE_UINT2,
+ RV_VATTR_TYPE_UINT,
+ RV_VATTR_TYPE_BYTE4,
+ RV_VATTR_TYPE_BYTE3,
+ RV_VATTR_TYPE_BYTE2,
+ RV_VATTR_TYPE_BYTE,
+} rv_vattr_type_t;
+
+typedef struct rv_vattr_t rv_vattr_t;
+struct rv_vattr_t {
+ rv_vattr_t* next;
+ rv_vattr_type_t type;
+};
+
+typedef struct {
+ rv_uniform_t* uniform_desc_first;
+ rv_shader_node_t* shader_first;
+ rv_vattr_t* vattr_first;
+ rv_vattr_t* vattr_last;
+ rv_render_handle_t handle;
+} rv_program_t;
+
+typedef enum {
+ RV_PRIMITIVE_TYPE_TRIANGLES,
+} rv_primitive_type_t;
typedef enum {
RV_RENDER_CLEAR_FLAG_NONE = 0,
@@ -18,66 +178,154 @@ typedef enum {
RV_RENDER_CLEAR_FLAG_DEPTH = (1 << 1),
RV_RENDER_CLEAR_FLAG_STENCIL = (1 << 2),
- RV_RENDER_CLEAR_FLAG_ALL = RV_RENDER_CLEAR_FLAG_COLOR | RV_RENDER_CLEAR_FLAG_DEPTH | RV_RENDER_CLEAR_FLAG_STENCIL
+ RV_RENDER_CLEAR_FLAG_ALL = RV_RENDER_CLEAR_FLAG_COLOR | RV_RENDER_CLEAR_FLAG_DEPTH | RV_RENDER_CLEAR_FLAG_STENCIL
} rv_graphics_clear_flag_t;
typedef struct rv_render_clear_desc_t rv_render_clear_desc_t;
struct rv_render_clear_desc_t {
- rv_render_clear_desc_t* next;
-
- rv_graphics_clear_flag_t flag; // Flag to be set (clear color, clear depth, clear stencil, clear all)
- union {
- f32 color[4]; // Clear color value
- f32 depth; // Clear depth value
- s32 stencil; // Clear stencil value
- };
+ rv_render_clear_desc_t* next;
+
+ rv_graphics_clear_flag_t flags;
+ union {
+ f32 color[4];
+ rv_vec4 color_v;
+
+ f32 depth;
+ s32 stencil;
+ };
};
typedef enum {
- RV_RENDER_COMMAND_INVALID,
+ RV_RENDER_OBJ_OP_CREATE,
+ RV_RENDER_OBJ_OP_UPDATE,
+ RV_RENDER_OBJ_OP_BIND,
+ RV_RENDER_OBJ_OP_DELETE,
+} rv_render_command_obj_operation_t;
+
+typedef enum {
+ RV_RENDER_COMMAND_INVALID,
+
+ RV_RENDER_COMMAND_CUSTOM,
+
+ RV_RENDER_COMMAND_OBJ_VERTEX,
+ RV_RENDER_COMMAND_OBJ_INDEX,
+ RV_RENDER_COMMAND_OBJ_SHADER,
+ RV_RENDER_COMMAND_OBJ_PROGRAM,
+ RV_RENDER_COMMAND_OBJ_TEXTURE,
+
+ RV_RENDER_COMMAND_DRAW,
- RV_RENDER_COMMAND_CLEAR,
- RV_RENDER_COMMAND_SET_VIEWPORT,
- //RV_RENDER_COMMAND_PIPELINE_BIND,
+ RV_RENDER_COMMAND_CLEAR,
+ RV_RENDER_COMMAND_SET_VIEWPORT,
} rv_render_command_type_t;
typedef struct rv_render_command_t rv_render_command_t;
struct rv_render_command_t {
- rv_render_command_t* next;
+ rv_render_command_t* next;
- rv_render_command_type_t type;
+ rv_render_command_type_t type;
- union {
- struct {
- rv_render_clear_desc_t* first;
- rv_render_clear_desc_t* last;
- } clear;
- rv_vec4 viewport;
- };
+ union {
+ struct {
+ rv_render_clear_desc_t* first;
+ rv_render_clear_desc_t* last;
+ } clear;
+ rv_vec4 viewport;
+ struct {
+ void(*func)(void*);
+ void* data;
+ } custom;
+ struct {
+ rv_render_command_obj_operation_t operation;
+
+ // parameters
+ union {
+ struct {
+ rv_uniform_update_node_t* committed_uniforms;
+ } program_bind;
+
+ // TODO(Samdal): make these named structs and
+ // add ability to switch between different
+ // update methods
+ struct {void* data; s64 size;} ibo_update;
+ struct {void* data; s64 size;} vbo_update;
+
+ struct {
+ void* data;
+ rv_rect sub_part;
+ } tex_update;
+
+ struct {
+ // uniforms bind textures automatically,
+ // only for manual overriding or doing something special
+ s32 sampler_index;
+ } tex_bind;
+ };
+
+ // desc reference
+ union {
+ rv_program_t* program;
+ rv_shader_t* shader;
+ rv_vbo_t* vbo;
+ rv_ibo_t* ibo;
+ rv_texture_t* tex;
+ void* generic;
+ };
+ } obj;
+ struct {
+ rv_primitive_type_t primitive;
+ s32 first;
+ s64 count;
+ } draw;
+ };
};
+typedef struct {
+ rv_render_command_t* first;
+ rv_render_command_t* last;
+} rv_render_command_list_t;
+
typedef struct rv_render_pass_t rv_render_pass_t;
struct rv_render_pass_t {
- rv_render_pass_t* next;
- rv_render_command_t* first;
- rv_render_command_t* last;
+ rv_render_pass_t* next;
+ rv_render_command_list_t commands;
- rv_render_framebuffer_t* framebuffer;
+ rv_render_framebuffer_t* framebuffer;
};
typedef struct {
- rv_render_pass_t* first;
- rv_render_pass_t* last;
+ rv_render_pass_t* first;
+ rv_render_pass_t* last;
} rv_render_pass_list_t;
//////////////////////////////////////////////////////////////////
-// Command buffer api
-
// Helpers
-RV_GLOBAL rv_render_pass_t* rv_render_push_render_pass(rv_arena* arena, rv_render_pass_list_t* rpass_list);
-RV_GLOBAL rv_render_command_t* rv_render_push_command(rv_arena* arena, rv_render_pass_t* rpass, rv_render_command_type_t type);
-RV_GLOBAL rv_render_clear_desc_t* rv_render_push_clear_desc(rv_arena* arena, rv_render_command_t* clear, rv_graphics_clear_flag_t flags);
+// core command buffer helpers
+RV_GLOBAL rv_render_pass_t* rv_render_push_render_pass(rv_arena* arena, rv_render_pass_list_t* rpass_list);
+RV_GLOBAL rv_render_command_t* rv_render_push_command(rv_arena* arena, rv_render_command_list_t* commands, rv_render_command_type_t type);
+RV_GLOBAL rv_render_command_t* rv_render_push_command_copy(rv_arena* arena, rv_render_command_list_t* commands, rv_render_command_t* source);
+RV_GLOBAL rv_render_command_t* rv_render_push_obj_command(rv_arena* arena, rv_render_command_list_t* commands, rv_render_command_type_t type, rv_render_command_obj_operation_t operation, void* obj);
+#define rv_render_push_command_compound(arnea, commands, _type, ...) rv_render_push_command_copy(arena, commands, &(rv_render_command_t){.type = _type, __VA_ARGS__})
+RV_GLOBAL rv_render_command_t* rv_render_copy_commands(rv_arena* arena, rv_render_command_list_t* dest, rv_render_command_list_t* source); // reurns the first command addded
+
+// command modification helpers
+RV_GLOBAL rv_render_clear_desc_t* rv_render_push_clear_desc(rv_arena* arena, rv_render_command_t* clear, rv_graphics_clear_flag_t flags);
+RV_GLOBAL rv_uniform_update_node_t* rv_render_update_uniform(rv_arena* arena, rv_render_command_t* program, rv_uniform_t* uniform, rv_uniform_variant_t value);
+#define rv_render_update_uniform_compound(arena, prog, uniform, ...) rv_render_update_uniform(arena, prog, uniform, (rv_uniform_variant_t)__VA_ARGS__);
+RV_GLOBAL bool32 rv_render_command_has_program(rv_render_command_t* command, rv_program_t* prog);
+
+// shader modificatoin helpers
+RV_GLOBAL rv_shader_node_t* rv_render_push_shader_to_prog(rv_arena* arena, rv_program_t* prog, rv_shader_t* shader);
+RV_GLOBAL rv_uniform_t* rv_render_push_uniform_to_prog(rv_arena* arena, rv_program_t* prog, rv_str8 name, rv_uniform_type_t type);
+RV_GLOBAL rv_uniform_t* rv_render_push_uniform_to_prog_texture(rv_arena* arena, rv_program_t* prog, rv_str8 name, rv_texture_t* tex);
+RV_GLOBAL rv_vattr_t* rv_render_push_vattr_to_prog(rv_arena* arena, rv_program_t* prog, rv_vattr_type_t type);
+
+//////////////////////////////////////////////////////////////////
// Submit
RV_GLOBAL void rv_window_render_commit(rv_window_handle_t* window, rv_render_pass_list_t* passes);
+//////////////////////////////////////////////////////////////////
+// Util
+
+RV_GLOBAL void rv_tex_load_from_file(rv_str8 filename);
diff --git a/src/render/render_helpers.c b/src/render/render_helpers.c
@@ -4,24 +4,92 @@
RV_GLOBAL rv_render_pass_t* rv_render_push_render_pass(rv_arena* arena, rv_render_pass_list_t* rpass_list)
{
- rv_render_pass_t* res = rv_push(arena, rv_render_pass_t);
- rv_llist_queue_push(rpass_list->first, rpass_list->last, res);
- return res;
+ rv_render_pass_t* res = rv_push(arena, rv_render_pass_t);
+ rv_llist_queue_push(rpass_list->first, rpass_list->last, res);
+ return res;
}
-RV_GLOBAL rv_render_command_t* rv_render_push_command(rv_arena* arena, rv_render_pass_t* rpass, rv_render_command_type_t type)
+RV_GLOBAL rv_render_command_t* rv_render_push_command(rv_arena* arena, rv_render_command_list_t* commands, rv_render_command_type_t type)
{
- rv_render_command_t* res = rv_push(arena, rv_render_command_t);
- res->type = type;
- rv_llist_queue_push(rpass->first, rpass->last, res);
- return res;
+ rv_render_command_t* res = rv_push_compound(arena, rv_render_command_t, {.type = type});
+ rv_llist_queue_push(commands->first, commands->last, res);
+ return res;
+}
+
+RV_GLOBAL rv_render_command_t* rv_render_push_command_copy(rv_arena* arena, rv_render_command_list_t* commands, rv_render_command_t* source)
+{
+ rv_render_command_t* res = rv_push(arena, rv_render_command_t);
+ *res = *source;
+ rv_llist_queue_push(commands->first, commands->last, res);
+ return res;
+}
+
+RV_GLOBAL rv_render_command_t* rv_render_push_obj_command(rv_arena* arena, rv_render_command_list_t* commands, rv_render_command_type_t type, rv_render_command_obj_operation_t operation, void* obj)
+{
+ rv_assert(type >= RV_RENDER_COMMAND_OBJ_VERTEX && type <= RV_RENDER_COMMAND_OBJ_TEXTURE);
+
+ rv_render_command_t* res = rv_render_push_command(arena, commands, type);
+ res->obj.operation = operation;
+ res->obj.generic = obj;
+
+ return res;
+}
+
+RV_GLOBAL rv_render_command_t* rv_render_copy_commands(rv_arena* arena, rv_render_command_list_t* dest, rv_render_command_list_t* source)
+{
+ rv_render_command_t* res = NULL;
+ for (rv_render_command_t* c = source->first; c; c = c->next) {
+ if (!res) res = rv_render_push_command_copy(arena, dest, c);
+ else rv_render_push_command_copy(arena, dest, c);
+ }
+ return res;
}
RV_GLOBAL rv_render_clear_desc_t* rv_render_push_clear_desc(rv_arena* arena, rv_render_command_t* clear, rv_graphics_clear_flag_t flags)
{
- rv_render_clear_desc_t* res = rv_push(arena, rv_render_clear_desc_t);
- rv_assert(clear->type == RV_RENDER_COMMAND_CLEAR);
- res->flag = flags;
- rv_llist_queue_push(clear->clear.first, clear->clear.first, res);
- return res;
+ rv_assert(clear->type == RV_RENDER_COMMAND_CLEAR);
+ rv_render_clear_desc_t* res = rv_push_compound(arena, rv_render_clear_desc_t, {.flags = flags});
+ rv_llist_queue_push(clear->clear.first, clear->clear.last, res);
+ return res;
+}
+
+RV_GLOBAL rv_uniform_update_node_t* rv_render_update_uniform(rv_arena* arena, rv_render_command_t* program, rv_uniform_t* uniform, rv_uniform_variant_t value)
+{
+ rv_assert(program->type == RV_RENDER_COMMAND_OBJ_PROGRAM);
+ rv_uniform_update_node_t* res = rv_push_compound(arena, rv_uniform_update_node_t, {.desc = uniform, .value = value});
+ rv_llist_stack_push(program->obj.program_bind.committed_uniforms, res);
+ return res;
+}
+
+RV_GLOBAL bool32 rv_render_command_has_program(rv_render_command_t* command, rv_program_t* prog)
+{
+ return command->type == RV_RENDER_COMMAND_OBJ_PROGRAM && command->obj.program == prog;
+}
+
+RV_GLOBAL rv_shader_node_t* rv_render_push_shader_to_prog(rv_arena* arena, rv_program_t* prog, rv_shader_t* shader)
+{
+ rv_shader_node_t* res = rv_push_compound(arena, rv_shader_node_t, {.shader = shader});
+ rv_llist_stack_push(prog->shader_first, res);
+ return res;
+}
+
+RV_GLOBAL rv_uniform_t* rv_render_push_uniform_to_prog(rv_arena* arena, rv_program_t* prog, rv_str8 name, rv_uniform_type_t type)
+{
+ rv_uniform_t* res = rv_push_compound(arena, rv_uniform_t, {.name_null_terminated = rv_str8_copy(arena, name) , .type = type});
+ rv_llist_stack_push(prog->uniform_desc_first, res);
+ return res;
+}
+RV_GLOBAL rv_uniform_t* rv_render_push_uniform_to_prog_texture(rv_arena* arena, rv_program_t* prog, rv_str8 name, rv_texture_t* tex)
+{
+ rv_uniform_t* res = rv_push_compound(arena, rv_uniform_t, {.name_null_terminated = rv_str8_copy(arena, name) , .type = RV_UNIFORM_TEX, .last_val.v_tex = tex});
+ rv_llist_stack_push(prog->uniform_desc_first, res);
+ return res;
+
+}
+
+RV_GLOBAL rv_vattr_t* rv_render_push_vattr_to_prog(rv_arena* arena, rv_program_t* prog, rv_vattr_type_t type)
+{
+ rv_vattr_t* res = rv_push_compound(arena, rv_vattr_t, {.type = type});
+ rv_llist_queue_push(prog->vattr_first, prog->vattr_last, res);
+ return res;
}
diff --git a/src/render/render_inc.c b/src/render/render_inc.c
@@ -1,8 +1,15 @@
//////////////////////////////////////////////////////////////////
// render_inc.c
-#include "render_helpers.c"
+#if RV_WIN_ENABLED
+ #include "render_helpers.c"
-#if RV_RENDER_OPENGL
- #include "impl/opengl.c"
-#endif
+ #if RV_RENDER_OPENGL
+ #include "impl/opengl.c"
+ #else
+ RV_GLOBAL void rv_window_render_commit(rv_window_handle_t* window, rv_render_pass_list_t* passes)
+ {
+
+ }
+ #endif
+#endif // RV_WIN_ENABLED
diff --git a/src/render/render_inc.h b/src/render/render_inc.h
@@ -1,4 +1,6 @@
//////////////////////////////////////////////////////////////////
// render_inc.h
-#include "render.h"
+#if RV_WIN_ENABLED
+ #include "render.h"
+#endif // RV_WIN_ENABLED
