diff options
Diffstat (limited to 'src')
82 files changed, 20786 insertions, 0 deletions
diff --git a/src/editor/editor.cpp b/src/editor/editor.cpp new file mode 100644 index 0000000..d97164e --- /dev/null +++ b/src/editor/editor.cpp @@ -0,0 +1,138 @@ +#include "editor.h" + +#include "../game.h" + +GAME_EDITOR* editor = 0; + +GAME_EDITOR* editor_create( GAME_DATA* game ) { + if( editor ) { + dlog( "editor_create() : attempted to create editor when one already exists\n" ); + return 0; + } + + GAME_EDITOR* e = new GAME_EDITOR(); + editor = e; + + e->grid = 1.f; + e->spritesize = EDITOR_DEFAULT_SPRITE_SIZE; + + e->game = game; + gui_init( game ); + + e->wnd = gui_editorwindow( 800, 600 ); + + gui_end(); + return e; +} + +STAT editor_close( GAME_EDITOR* e ) { + game_unload_map( e->game ); + e->map = 0; + + gui_push_callback( e, pfn( void* data ) { + GAME_EDITOR* e = (GAME_EDITOR*)data; + I32 w = e->wnd->w, h = e->wnd->h; + + gui_free( e->wnd ); + e->wnd = gui_editorwindow( w, h ); + } ); + + return STAT_OK; +} + +STAT editor_load_map( GAME_EDITOR* e, const char* mapname ) { + if( e->map ) { + dlog( "editor_load() : map already loaded\n" ); + return STAT_ERR; + } + + dlog( "editor_load() : loading map %s\n", mapname ); + + WORLD_MAP* m = game_load_map( e->game, mapname ); + if( !m ) + return STAT_ERR; + + e->game->state.map = e->map = m; + gui_push_callback( e, pfn( void* d ) { + editor_create_map_view( (GAME_EDITOR*)d ); + } ); + return STAT_OK; +} + +STAT editor_new_map( GAME_EDITOR* e, const char* mapname ) { + WORLD_MAP* m = new WORLD_MAP; + defer( map_free( e->game, m ) ); + + strcpy( m->name, mapname ); + strcat( m->name, ".hmap" ); + + m->startpos = { 10.f, 10.f, 0.f }; + m->props.push( { .clr = { 1.f, 1.f, 1.f, 1.f } } ); + m->walls.push( { + .start = { 0 , 0, -40.f }, + .end = { 10.f, 0, 80.f }, + .propid = 0 + } ); + + CFG_SECTION* map_cfg = map_serialize( m ); + defer( cfg_free( map_cfg ) ); + + char full_path[256]; + sprintf( full_path, "../assets/maps/%s.hmap", mapname ); + if( !OK( cfg_save( map_cfg, full_path ) ) ) { + dlog( "editor_new_map() : error saving file %s\n", full_path ); + return STAT_ERR; + } + + GUI_LIST_ENTRY ne; + ne.val = e->map_list.size; + strcpy( ne.title, mapname ); + strcat( ne.title, ".hmap" ); + e->map_list.push( ne ); + + return STAT_OK; +} + +STAT editor_save_map( GAME_EDITOR* e ) { + if( !e->map ) return STAT_ERR; + + CFG_SECTION* serialized = map_serialize( e->map ); + if( !serialized ) return STAT_ERR; + defer( cfg_free( serialized ) ); + + char full_path[256]; + sprintf( full_path, "../assets/maps/%s", e->map->name ); + + if( !OK( cfg_save( serialized, full_path ) ) ) { + dlog( "failed to save map %s", full_path ); + return STAT_ERR; + } + + return STAT_OK; +} + +LIST<GUI_LIST_ENTRY>* editor_get_map_list( GAME_EDITOR* ge ) { + const char* mapsdir = "../assets/maps"; + const char* ext[] = { "hmap", 0 }; + LIST<FILE_ENTRY> files = assets_get_files_by_ext_dir( ext, mapsdir ); + LIST<GUI_LIST_ENTRY> list; + + I32 i = 0; + files.each( fn( FILE_ENTRY* e ) { + if( e->dir ) return; + + GUI_LIST_ENTRY ne; + ne.val = i++; + strcpy( ne.title, file_path_last_of( e->name ) ); + + list.push( ne ); + } ); + + ge->map_list = list; + return &ge->map_list; +} + +void editor_destroy( GAME_EDITOR* e ) { + delete e; + editor = 0; +} diff --git a/src/editor/editor.h b/src/editor/editor.h new file mode 100644 index 0000000..0bf2ab4 --- /dev/null +++ b/src/editor/editor.h @@ -0,0 +1,135 @@ +#pragma once +#include "../gui/base.h" +#include "../util/file.h" +#include "../game/world/map.h" + +const F32 EDITOR_DEFAULT_SPRITE_SIZE = 32.f; + +enum EditorTools_t { + EDITOR_TOOL_NONE, + EDITOR_TOOL_SELECT, + EDITOR_TOOL_WALL, + EDITOR_TOOL_POLY, + EDITOR_TOOL_SPRITE, + EDITOR_TOOL_ENT +}; + +enum EditorSelectType_t { + EDITOR_SELECT_NONE, + EDITOR_SELECT_WALL, + EDITOR_SELECT_POLY, + EDITOR_SELECT_WVERTEX, // wall vertex + EDITOR_SELECT_PVERTEX, // polygon vertex + EDITOR_SELECT_SPRITE, + EDITOR_SELECT_ENT, + EDITOR_SELECT_ORIGIN, + EDITOR_SELECT_SURFPROPS +}; + +struct GAME_EDITOR { + GAME_DATA* game; + + struct WORLD_MAP* map; + struct GUI_EDITORWINDOW* wnd; + + struct EDITOR_GUI { + struct GUI_WINDOW* new_map_popup; + + struct GUI_EDITOR_2DVIEW* v2d; + struct GUI_EDITOR_3DVIEW* v3d; + + struct GUI_LABEL* toollabel; + + struct GUI_LABEL* gridlabel; + + struct GUI_EDITOR_PROPVIEW* props; + I32 map_select{}; + } gui; + + + U8 wireframe{}; + U8 tool{}; + F32 grid{}; + U8 propgrid{}; + F32 spritesize{}; + U8 drawbsp{}; + + LIST<GUI_LIST_ENTRY> map_list{}; +}; + +extern GAME_EDITOR* editor_create( struct GAME_DATA* game ); +extern void editor_destroy( GAME_EDITOR* editor ); +extern STAT editor_load_map( GAME_EDITOR* e, const char* mapname ); +extern STAT editor_save_map( GAME_EDITOR* e ); +extern STAT editor_new_map( GAME_EDITOR* e, const char* mapname ); +extern STAT editor_close( GAME_EDITOR* e ); + +extern LIST<GUI_LIST_ENTRY>* editor_get_map_list( GAME_EDITOR* e ); + +extern void editor_load_map_cb( void* ); +extern void editor_new_map_cb( void* ); + +extern void editor_create_map_view( GAME_EDITOR* e ); + +struct GUI_EDITORWINDOW : GUI_WINDOW {}; +struct GUI_EDITOR_3DVIEW : GUI_VIEW {}; +struct GUI_EDITOR_2DVIEW : GUI_VIEW { + F32 scale; + F32 posx, posy; + + I32 moffx, moffy; + F32 voffx, voffy; + U8 dragheld; + U8 held; + U8 heldoutbounds; + + I32 oldmx, oldmy; + F32 mremainx, mremainy; + + void* curselect; + U8 seltype; + + void* curdrag; + U8 dragtype; + U8 dragmoved; +}; + +struct GUI_EDITOR_PROPVIEW : GUI_VIEW { + void* curselect; + U8 seltype; + + GUI_VIEW* itemview; +}; + +struct GUI_EDITOR_TEXTUREPICKER : GUI_WINDOW { + struct GL_TEX2D** target; + struct GL_TEX2D* curselect; + + I32 scrolloff; + + LIST<FILE_ENTRY> files; + LIST<struct GL_TEX2D*> textures; + I32 scrollheight; + + U32 rowcount; + char search[256]; + + U32 curload; + U8 loaded; + + GL_TEX2D* heldselect; + GUI_LABEL* densitylabel; + + GUI_CALLBACK cb; +}; + +extern GUI_EDITORWINDOW* gui_editorwindow( I32 w, I32 h ); +extern GUI_EDITOR_2DVIEW* gui_editor_2dview( I32 x, I32 y, I32 w, I32 h ); +extern GUI_EDITOR_3DVIEW* gui_editor_3dview( I32 x, I32 y, I32 w, I32 h ); +extern GUI_EDITOR_PROPVIEW* gui_editor_propview( I32 x, I32 y, I32 w, I32 h ); +extern GUI_EDITOR_TEXTUREPICKER* gui_editor_texturepicker( I32 x, I32 y, I32 w, I32 h, GL_TEX2D** target ); + +extern void gui_editor_propview_select( GUI_EDITOR_PROPVIEW* e, void* what, U8 seltype ); +extern void gui_editor_propview_update( GUI_EDITOR_PROPVIEW* e ); + +extern GAME_EDITOR* editor; diff --git a/src/editor/gui.cpp b/src/editor/gui.cpp new file mode 100644 index 0000000..71ce113 --- /dev/null +++ b/src/editor/gui.cpp @@ -0,0 +1,255 @@ +#include "editor.h" +#include "../game/world/bsp.h" + +void gui_editorwindow_draw_fn( void* ptr ) { + GUI_EDITORWINDOW* wnd = (GUI_EDITORWINDOW*)ptr; + + CLR clr = gui_is_fg_window( wnd )? ui_clr.border : ui_clr.border_inactive; + gui_draw_frect( wnd->x, wnd->y, wnd->w, wnd->h, clr ); + gui_draw_frect( wnd->x + 1, wnd->y + 1, wnd->w - 2, wnd->h - 2, ui_clr.bg ); + + wnd->children.each( fn( GUI_BASE** ptr ) { + GUI_BASE* it = *ptr; + if( !it->enabled ) return; + + if( it->draw_fn ) it->draw_fn( it ); + else dlog( "gui_editorwindow_draw_fn(): child %p has no draw_fn", it ); + } ); +} + +GUI_EDITORWINDOW* gui_editorwindow_create( I32 w, I32 h ) { + GUI_EDITORWINDOW* wnd = new GUI_EDITORWINDOW; + wnd->x = 0; + wnd->y = 0; + wnd->w = w; + wnd->h = h; + wnd->locked = 1; + wnd->draw_fn = gui_editorwindow_draw_fn; + strcpy( wnd->name, "EDITORWINDOW" ); + + _gui.windows.push( wnd ); + gui_set_window( wnd ); + + gui_set_view( 0 ); + gui_view( 1, 1, w - 2, h - 2 ); + return wnd; +} + +GUI_EDITORWINDOW* gui_editorwindow( I32 w, I32 h ) { + GUI_EDITORWINDOW* wnd = gui_editorwindow_create( w, h ); + + gui_title( "ray2Dscape (level editor)" ); + + GAME_EDITOR* e = editor; + LIST<GUI_LIST_ENTRY>* map_list = editor_get_map_list( e ); + + gui_list( wnd->w / 2 - 100, 40, 200, 200, "map list", map_list, &editor->gui.map_select ); + gui_button( wnd->w / 2 - 100, 260, 95, 25, "new map", editor_new_map_cb ); + gui_button( wnd->w / 2 + 5, 260, 95, 25, "load map", editor_load_map_cb ); + return wnd; +} + +void editor_update_active_tool_label( GAME_EDITOR* e ) { + char lstr[256]; + switch( e->tool ) { + case EDITOR_TOOL_SELECT: + sprintf( lstr, "current tool: select" ); break; + case EDITOR_TOOL_WALL: + sprintf( lstr, "current tool: wall" ); break; + case EDITOR_TOOL_POLY: + sprintf( lstr, "current tool: polygon" ); break; + case EDITOR_TOOL_SPRITE: + sprintf( lstr, "current tool: sprite" ); break; + case EDITOR_TOOL_ENT: + sprintf( lstr, "current tool: entity" ); break; + default: + sprintf( lstr, "current tool: none" ); break; + } + + strcpy( e->gui.toollabel->name, lstr ); +} + +void editor_update_properties_column( GAME_EDITOR* e ) { + GAME_EDITOR::EDITOR_GUI* egui = &e->gui; + gui_editor_propview_update( egui->props ); +} + +void editor_create_properties_column( GAME_EDITOR* e ) { + GAME_EDITOR::EDITOR_GUI* egui = &e->gui; + gui_button( 10, 10, 145, 20, "back", pfn( void* ) { editor_close( editor ); } ); + gui_button( 165, 10, 145, 20, "save", pfn( void* ) { editor_save_map( editor ); } ); + + egui->props = gui_editor_propview( 10, 38, 300, 460 ); + gui_editor_propview_select( egui->props, e->map, EDITOR_SELECT_ORIGIN ); +} + +void editor_create_game_view_column( GAME_EDITOR* e ) { + GAME_EDITOR::EDITOR_GUI* egui = &e->gui; + I32 x = 320, y = 10; + + gui_button( x, y, 229, 20, "2d view", pfn( void* ptr ) { + editor->gui.v3d->enabled = 0; + editor->gui.v2d->enabled = 1; + } ); + + gui_button( x + 239, y, 229, 20, "3d view", pfn( void* ptr ) { + editor->gui.v3d->enabled = 1; + editor->gui.v2d->enabled = 0; + } ); + + y += 28; + egui->v2d = gui_editor_2dview( x, y, 468, 370 ); + egui->v3d = gui_editor_3dview( x, y, 468, 370 ); + egui->v2d->enabled = 0; + egui->v3d->enabled = 1; +} + +void settool( U8 t ) { editor->tool = t; editor_update_active_tool_label( editor ); } +void editor_create_tools_row( GAME_EDITOR* e ) { + I32 x = 10, y = 520; + + e->gui.toollabel = gui_label( x, y, "current tool: none" ); + y += 16; + gui_button( x, y, 80, 20, "none", pfn( void* ) { settool( EDITOR_TOOL_NONE ); } ); x += 90; + gui_button( x, y, 80, 20, "select", pfn( void* ) { settool( EDITOR_TOOL_SELECT ); } ); x += 90; + gui_button( x, y, 80, 20, "wall", pfn( void* ) { settool( EDITOR_TOOL_WALL ); } ); x += 90; + + x = 10; y += 28; + gui_button( x, y, 80, 20, "polygon", pfn( void* ) { settool( EDITOR_TOOL_POLY ); } ); x += 90; + gui_button( x, y, 80, 20, "sprite", pfn( void* ) { settool( EDITOR_TOOL_SPRITE ); } ); x += 90; + gui_button( x, y, 80, 20, "entity", pfn( void* ) { settool( EDITOR_TOOL_ENT ); } ); x += 90; + editor_update_active_tool_label( e ); +} + +void editor_update_view_settings( GAME_EDITOR* e ) { + GAME_EDITOR::EDITOR_GUI* egui = &e->gui; + + sprintf( egui->gridlabel->name, "grid size: %1.2f", e->grid ); +} + +void editor_grid_increment_cb( void* ) { + if( editor->grid < 16.f ) editor->grid *= 2.f; + editor_update_view_settings( editor ); + + if( editor->propgrid ) + editor_update_properties_column( editor ); +} + +void editor_grid_decrement_cb( void* ) { + if( editor->grid > 0.25f ) editor->grid *= 0.5f; + editor_update_view_settings( editor ); + + if( editor->propgrid ) + editor_update_properties_column( editor ); +} + +void editor_grid_propgrid_cb( void* ) { + editor_update_properties_column( editor ); +} + +void editor_create_view_settings_row( GAME_EDITOR* e ) { + I32 x = 320, y = 426; + + gui_label( x, y, "view settings:" ); x += 95; + editor->gui.gridlabel = gui_label( x, y, "grid size: %1.2f", e->grid ); + x += 95; + gui_button( x, y, 20, 20, "+", editor_grid_increment_cb ); + gui_button( x + 25, y, 20, 20, "-", editor_grid_decrement_cb ); + x += 55; + GUI_CHECKBOX* check = gui_checkbox( x, y, "properties grid", &e->propgrid ); + check->cb = editor_grid_propgrid_cb; + x += 120; + gui_checkbox( x, y, "wireframe", &e->wireframe ); + + x = 320; + y = 440; + gui_button( x, y, 100, 20, "compile bsp", pfn( void* b ) { + if( editor->map->bsp ) + bsp_free( editor->map->bsp ); + editor->map->bsp = bsp_build_map( editor->map ); + } ); x += 110; + + gui_checkbox( x, y, "draw bsp", &e->drawbsp ); +} + +void editor_create_map_view( GAME_EDITOR* e ) { + if( !e->map ) { + dlog( "editor_create_map_views() : no map loaded\n" ); + return; + } + + GUI_EDITORWINDOW* w = e->wnd; + w->children.each( fn( GUI_BASE** ptr ) { + gui_free( *ptr ); + } ); + w->children.clear(); + + gui_set_window( w ); + gui_set_view( 0 ); + + gui_view( 1, 1, w->w - 2, w->h - 2 ); + + editor_create_properties_column( e ); + editor_create_game_view_column( e ); + editor_create_tools_row( e ); + editor_create_view_settings_row( e ); +} + +void close_new_map_popup( void* ) { + gui_push_callback( pfn( void* ) { + GUI_WINDOW* popup = editor->gui.new_map_popup; + if( !popup ) + return; + + gui_free( popup ); + editor->gui.new_map_popup = 0; + } ); +} + +void editor_new_map_cb( void* ptr ) { + if( editor->gui.new_map_popup ) return; + + GUI_WINDOW* cwnd = gui_get_window(); + GUI_VIEW* view = gui_get_view(); + + I32 wx = 250; + I32 wy = 140; + I32 ww = 300; + I32 wh = 200; + + editor->gui.new_map_popup = gui_window( wx, wy, ww, wh ); + gui_title( "new map" ); + GUI_TEXTBOX* tb = gui_textbox( 10, 20, ww - 20, 20, "name", 32 ); + tb->active = 1; + + gui_button( ww - 50 - 70, wh - 20 - 10, 50, 20, "ok", pfn( void* ptr ) { + GUI_BUTTON* btn = (GUI_BUTTON*)ptr; + GUI_TEXTBOX* name = (GUI_TEXTBOX*)gui_find_node( btn->parent, "name" ); + if( !name ) + return; + + editor_new_map( editor, name->value ); + close_new_map_popup( 0 ); + } ); + + gui_button( ww - 50 - 10, wh - 20 - 10, 50, 20, "cancel", close_new_map_popup ); + + gui_set_window( cwnd ); + gui_set_view( view ); +} + +void editor_load_map_cb( void* ptr ) { + GUI_LIST* list = (GUI_LIST*)gui_find_node( editor->wnd, "map list" ); + + GUI_LIST_ENTRY* e = gui_list_get_selected( list ); + if( !e ) + return; + + char full_path[256]; + sprintf( full_path, "../assets/maps/%s", e->title ); + + if( editor->map ) + editor_close( editor ); + + editor_load_map( editor, full_path ); +} diff --git a/src/editor/properties.cpp b/src/editor/properties.cpp new file mode 100644 index 0000000..ebae177 --- /dev/null +++ b/src/editor/properties.cpp @@ -0,0 +1,348 @@ +#include "editor.h" +#include "../render/gl.h" +#include "../game/assets.h" + +const I32 PROPVIEW_TITLE_OFFSET = 15; + +void gui_editor_propview_select( GUI_EDITOR_PROPVIEW* view, void* what, U8 seltype ) { + if( !editor->map ) + return; + + if( seltype == EDITOR_SELECT_WVERTEX ) { + MAP_WALL* s = editor->map->walls.where( fn( MAP_WALL* s ) { + return ( &s->start == what || &s->end == what ); + } ); + + if( !s ) + return gui_editor_propview_select( view, editor->map, EDITOR_SELECT_ORIGIN ); + + view->curselect = s; + view->seltype = EDITOR_SELECT_WALL; + return gui_editor_propview_update( view ); + } + + view->curselect = what; + view->seltype = seltype; + gui_editor_propview_update( view ); +} + +// returns the subentry height +I32 gui_editor_propview_surfprops_subentry( I32 x, I32 y, I32* propid, SURF_PROPS* props ) { + I32 oldy = y; + I32 space = 20; + gui_label( x, y, "prop id: %d", *propid ); + GUI_BUTTON* newprop = gui_button( x + 235, y, 20, 20, "+", pfn( void* ptr ) { + GUI_BUTTON* btn = (GUI_BUTTON*)ptr; + SURF_PROPS props; + props.tex = 0; + props.clr = { 1.f, 1.f, 1.f, 1.f }; + + editor->map->props.push( props ); + *(U32*)(btn->extra) = editor->map->props.size - 1; + gui_editor_propview_update( editor->gui.props ); + } ); + newprop->extra = propid; + GUI_BUTTON* goprop = gui_button( x + 260, y, 20, 20, "\x1A", pfn( void* ptr ) { + GUI_BUTTON* btn = (GUI_BUTTON*)ptr; + gui_editor_propview_select( editor->gui.props, btn->extra, EDITOR_SELECT_SURFPROPS ); + } ); + goprop->extra = props; + + y += space; + + if( props->tex ) + gui_label( x + 10, y, "texture: %s", props->tex->name ); + else + gui_label( x + 10, y, "texture: none" ); + + GUI_BUTTON* btn = gui_button( x + 260, y, 20, 20, "\x1A", pfn( void* ptr ) { + GUI_BUTTON* btn = (GUI_BUTTON*)ptr; + GL_TEX2D** ptex = (GL_TEX2D**)btn->extra; + GUI_EDITOR_TEXTUREPICKER* picker = gui_editor_texturepicker( 200, 100, 400, 400, ptex ); + picker->cb = pfn( void* ) { gui_editor_propview_update( editor->gui.props ); }; + } ); + + btn->extra = &props->tex; + y += space; + gui_colorinput( x + 10, y, 270, "color", &props->clr ); y += (space+18); + + return y - oldy; +} + +void gui_editor_propview_create_wallprops( GUI_EDITOR_PROPVIEW* view ) { + MAP_WALL* s = (MAP_WALL*)view->curselect; + WORLD_MAP* m = editor->map; + SURF_PROPS* props = wall_get_props( m, s ); + + I32 x = 10, y = 10; + I32 space = 20; + F32 step = editor->propgrid? editor->grid : 0.25f; + + I32 wall_idx = m->walls.idx_where( fn( MAP_WALL* ms ) { + return s == ms; + } ); + + gui_label( x, y, "idx: %d", wall_idx ); y += space; + y += gui_editor_propview_surfprops_subentry( x, y, &s->propid, props ); + + GUI_VECTORINPUT* posinput; + posinput = gui_vectorinput( x, y, 280, "start", (F32*)&s->start, 3, -INFINITY, INFINITY, step ); y += (space+18); + posinput->cb = pfn( void* ) { map_check_bounds( editor->map ); }; + posinput = gui_vectorinput( x, y, 280, "end", (F32*)&s->end, 3, -INFINITY, INFINITY, step ); y += (space+18); + posinput->cb = pfn( void* ) { map_check_bounds( editor->map ); }; +} + +void gui_editor_propview_create_polyprops( GUI_EDITOR_PROPVIEW* view ) { + MAP_POLYGON* p = (MAP_POLYGON*)view->curselect; + WORLD_MAP* m = editor->map; + SURF_PROPS* props = polygon_get_props( m, p ); + + I32 x = 10, y = 10; + I32 space = 20; + + I32 poly_idx = m->polygons.idx_where( fn( MAP_POLYGON* mp ) { + return p == mp; + } ); + + gui_label( x, y, "idx: %d", poly_idx ); y += space; + y += gui_editor_propview_surfprops_subentry( x, y, &p->propid, props ); + + gui_label( x, y, "vertices: %d", p->vertices.size ); y += space; + I32 idx = 0; + p->vertices.each( fn( MAP_VERTEX* v ) { + gui_label( x + 10, y, "[%d] -> { %.02f, %.02f, %.02f }", idx, v->pos.x, v->pos.y, v->pos.z ); + GUI_BUTTON* btn = gui_button( x + 260, y - 2, 20, 20, "\x1A", pfn( void* ptr ) { + GUI_EDITOR_PROPVIEW* view = editor->gui.props; + GUI_BUTTON* btn = (GUI_BUTTON*)ptr; + MAP_POLYGON* p = (MAP_POLYGON*)view->curselect; + I64 idx = (I64)btn->extra; + if( p ) // p100 + gui_editor_propview_select( view, &p->vertices[idx], EDITOR_SELECT_PVERTEX ); + } ); + + y += space; + btn->extra = (void*)( (I64)idx++ ); + } ); + gui_label( x, y, "mins: { %1.02f, %1.02f, %1.02f }", p->mins.x, p->mins.y, p->mins.z ); y += space; + gui_label( x, y, "maxs: { %1.02f, %1.02f, %1.02f }", p->maxs.x, p->maxs.y, p->maxs.z ); y += space; +} + +void gui_editor_propview_create_mapprops( GUI_EDITOR_PROPVIEW* view ) { + WORLD_MAP* m = (WORLD_MAP*)view->curselect; + + I32 x = 10, y = 10; + I32 space = 20; + + F32 step = editor->propgrid? editor->grid : 0.25f; + + gui_label( x, y, "name: %s", m->name ); y += space; + gui_label( x, y, "walls: %d", m->walls.size ); y += space; + gui_label( x, y, "polygons: %d", m->polygons.size ); y += space; + gui_label( x, y, "props: %d", m->props.size ); y += space; + I32 i = 0; + m->props.each( fn( SURF_PROPS* p ) { + if( p->tex ) + gui_label( x + 10, y, "[%d] -> %s", i++, assets_abspath( p->tex->name ) ); + else + gui_label( x + 10, y, "[%d] -> { %.02f, %.02f, %.02f, %.02f }", i++, p->clr.r, p->clr.g, p->clr.b, p->clr.a ); + + GUI_BUTTON* btn = gui_button( x + 260, y, 20, 20, "\x1A", pfn( void* ptr ) { + GUI_BUTTON* btn = (GUI_BUTTON*)ptr; + gui_editor_propview_select( editor->gui.props, btn->extra, EDITOR_SELECT_SURFPROPS ); + } ); + btn->extra = p; + y += space; + } ); + gui_label( x, y, "sprites: %d", m->sprites.size ); y += space; + gui_label( x, y, "loaded textures: %d", m->textures.size ); y += space; + gui_vectorinput( x, y, 280, "spawn position", (F32*)&m->startpos, 3, -INFINITY, INFINITY, step ); y += (space+18); + GUI_FLOATINPUT* ang = gui_floatinput( x, y, 280, "spawn angle", &m->startang, -180.f, 180.f, 1.f ); y += (space+18); + ang->wraparound = 1; +} + +void gui_editor_propview_create_pvertexprops( GUI_EDITOR_PROPVIEW* view ) { + MAP_VERTEX* v = (MAP_VERTEX*)view->curselect; + WORLD_MAP* m = editor->map; + + I32 x = 10, y = 10; + I32 space = 20; + + I32 vert_idx = -1; + I32 poly_idx = m->polygons.idx_where( fn( MAP_POLYGON* p ) { + vert_idx = p->vertices.idx_where( fn( MAP_VERTEX* pv ) { return (pv == v); } ); + return vert_idx != -1; + } ); + + F32 step = editor->propgrid? editor->grid : 0.25f; + + gui_label( x, y, "idx: %d", vert_idx ); y += space; + gui_label( x, y, "polygon idx: %d", poly_idx ); + + gui_button( x + 260, y - 2, 20, 20, "\x1A", pfn( void* ) { + GUI_EDITOR_PROPVIEW* view = editor->gui.props; + MAP_VERTEX* v = (MAP_VERTEX*)view->curselect; + WORLD_MAP* m = editor->map; + MAP_POLYGON* polygon = m->polygons.where( fn( MAP_POLYGON* p ) { + I32 vert_idx = p->vertices.idx_where( fn( MAP_VERTEX* pv ) { return (pv == v); } ); + return vert_idx != -1; + } ); + + if( polygon ) + gui_editor_propview_select( view, polygon, EDITOR_SELECT_POLY ); + } ); + y += space; + + GUI_VECTORINPUT* posinput = gui_vectorinput( x, y, 280, "position", (F32*)&v->pos, 3, -INFINITY, INFINITY, step ); y += (space+18); + posinput->cb = pfn( void* ptr ) { map_check_bounds( editor->map ); }; + gui_vectorinput( x, y, 280, "texture coordinates", (F32*)&v->uv, 2, 0.f, 1.f, 0.005f, "xy", "%.03f" ); y += (space+18); + gui_colorinput( x, y, 280, "color", &v->clr ); y += (space+18); +} + +void gui_editor_propview_create_spriteprops( GUI_EDITOR_PROPVIEW* view ) { + MAP_SPRITE* s = (MAP_SPRITE*)view->curselect; + WORLD_MAP* m = editor->map; + + I32 x = 10, y = 10; + I32 space = 20; + + I32 sprite_idx = m->sprites.idx_where( fn( MAP_SPRITE* ms ) { + return ms == s; + } ); + + F32 step = editor->propgrid? editor->grid : 0.25f; + + gui_label( x, y, "idx: %d", sprite_idx ); y += space; + gui_vectorinput( x, y, 280, "position", (F32*)&s->pos, 3, -INFINITY, INFINITY, step ); y += (space+18); + gui_vectorinput( x, y, 280, "size", (F32*)&s->size, 2, 1.f, INFINITY, 1.f, "wh" ); y += (space+18); + gui_colorinput( x, y, 280, "color", &s->clr ); y += (space+18); + if( s->tex ) + gui_label( x, y, "texture: %s", s->tex->name ); + else + gui_label( x, y, "texture: none" ); + + GUI_BUTTON* btn = gui_button( x + 260, y, 20, 20, "\x1A", pfn( void* ptr ) { + GUI_BUTTON* btn = (GUI_BUTTON*)ptr; + GL_TEX2D** ptex = (GL_TEX2D**)btn->extra; + GUI_EDITOR_TEXTUREPICKER* picker = gui_editor_texturepicker( 200, 100, 400, 400, ptex ); + picker->cb = pfn( void* ) { gui_editor_propview_update( editor->gui.props ); }; + } ); + btn->extra = &s->tex; + y += space; +} + +void gui_editor_propview_create_surfprops( GUI_EDITOR_PROPVIEW* view ) { + SURF_PROPS* p = (SURF_PROPS*)view->curselect; + WORLD_MAP* m = editor->map; + + I32 x = 10, y = 10; + I32 space = 20; + + I32 i = m->props.idx_where( fn( SURF_PROPS* mp ) { return mp == p; } ); + if( i == -1 ) + return; + + gui_label( x, y, "prop id: %d", i ); y += space; + if( p->tex ) + gui_label( x, y, "texture: %s", assets_abspath( p->tex->name ) ); + else + gui_label( x, y, "texture: none" ); + GUI_BUTTON* btn = gui_button( x + 260, y, 20, 20, "\x1A", pfn( void* ptr ) { + GUI_BUTTON* btn = (GUI_BUTTON*)ptr; + GL_TEX2D** ptex = (GL_TEX2D**)btn->extra; + GUI_EDITOR_TEXTUREPICKER* picker = gui_editor_texturepicker( 200, 100, 400, 400, ptex ); + picker->cb = pfn( void* ) { gui_editor_propview_update( editor->gui.props ); }; + } ); + btn->extra = &p->tex; + y += space; + gui_colorinput( x, y, 280, "color", &p->clr ); y += (space+18); +} + +void gui_editor_propview_create_entprops( GUI_EDITOR_PROPVIEW* view ) { + +} + +void gui_editor_propview_update( GUI_EDITOR_PROPVIEW* view ) { + if( !editor->map ) return; + + GUI_VIEW* target = gui_get_view(); + defer({ if( target ) gui_set_view( target ); }); + + gui_empty_children( view->itemview ); + gui_set_view( view->itemview ); + view->itemview->initheld = 1; + + switch( view->seltype ) { + case EDITOR_SELECT_WALL: gui_editor_propview_create_wallprops( view ); break; + case EDITOR_SELECT_POLY: gui_editor_propview_create_polyprops( view ); break; + case EDITOR_SELECT_ORIGIN: gui_editor_propview_create_mapprops( view ); break; + case EDITOR_SELECT_ENT: gui_editor_propview_create_entprops( view ); break; + case EDITOR_SELECT_PVERTEX: gui_editor_propview_create_pvertexprops( view ); break; + case EDITOR_SELECT_SPRITE: gui_editor_propview_create_spriteprops( view ); break; + case EDITOR_SELECT_SURFPROPS: gui_editor_propview_create_surfprops( view ); break; + default: break; + } +} + +void gui_editor_propview_get_title( GUI_EDITOR_PROPVIEW* view, char* buf ) { + switch( view->seltype ) { + case EDITOR_SELECT_NONE: sprintf( buf, "properties: " ); break; + case EDITOR_SELECT_POLY: sprintf( buf, "polygon properties: " ); break; + case EDITOR_SELECT_ORIGIN: sprintf( buf, "map [%s] properties: ", editor->map->name ); break; + case EDITOR_SELECT_ENT: sprintf( buf, "entity properties: " ); break; + case EDITOR_SELECT_SPRITE: sprintf( buf, "sprite properties: " ); break; + case EDITOR_SELECT_PVERTEX: sprintf( buf, "vertex properties: " ); break; + case EDITOR_SELECT_SURFPROPS: sprintf( buf, "surface properties: " ); break; + case EDITOR_SELECT_WVERTEX: + case EDITOR_SELECT_WALL: sprintf( buf, "wall properties: " ); break; + } +} + +void gui_editor_propview_draw_fn( void* ptr ) { + if( !editor->map ) return; + + GUI_EDITOR_PROPVIEW* view = (GUI_EDITOR_PROPVIEW*)ptr; + + I32 x = gui_relx( view ); + I32 y = gui_rely( view ); + I32 w = view->w; + I32 h = view->h; + + char title[64]; + gui_editor_propview_get_title( view, title ); + gui_draw_str( x, y, ALIGN_L, FNT_JPN12, ui_clr.txt, title ); + y += PROPVIEW_TITLE_OFFSET; + + CLR col = gui_is_fg_window( view )? ui_clr.border : ui_clr.border_inactive; + gui_draw_frect( x, y, w, h, col ); + gui_draw_frect( x+1, y+1, w-2, h-2, ui_clr.bg_sec ); + + view->itemview->draw_fn( view->itemview ); +} + +GUI_EDITOR_PROPVIEW* gui_editor_propview( I32 x, I32 y, I32 w, I32 h ) { + if( !gui_check_target() ) return 0; + + GUI_EDITOR_PROPVIEW* view = new GUI_EDITOR_PROPVIEW; + view->x = x; + view->y = y; + view->w = w; + view->h = h; + view->xbound = view->w; + view->ybound = view->h + PROPVIEW_TITLE_OFFSET; + strcpy( view->name, "EDITOR_PROP_VIEW" ); + view->draw_fn = gui_editor_propview_draw_fn; + view->input_fn = gui_base_input_fn; + + view->curselect = 0; + view->seltype = 0; + + GUI_VIEW* parent = gui_get_view(); + parent->children.push( view ); + view->parent = parent; + + gui_set_view( view ); + view->itemview = gui_view( 0, PROPVIEW_TITLE_OFFSET, w, h ); + + gui_set_view( parent ); + return view; +} diff --git a/src/editor/texturepicker.cpp b/src/editor/texturepicker.cpp new file mode 100644 index 0000000..a24e955 --- /dev/null +++ b/src/editor/texturepicker.cpp @@ -0,0 +1,356 @@ +#include "editor.h" +#include "../game/assets.h" +#include "../render/gl_2d.h" + +I32 TEXTUREVIEW_TOP_OFFSET = 20; +I32 TEXTUREVIEW_BOTTOM_OFFSET = 28; +const char* TEXTURE_EXTENSIONS[] = { + "png", + 0 +}; + +I32 gui_editor_texturepicker_get_scrollheight( GUI_EDITOR_TEXTUREPICKER* wnd ) { + return 0; +} + +void gui_editor_texturepicker_handle_scroll( GUI_EDITOR_TEXTUREPICKER* wnd ) { + I32 x = wnd->x; + I32 y = wnd->y + TEXTUREVIEW_TOP_OFFSET; + I32 w = wnd->w; + I32 h = wnd->h - TEXTUREVIEW_TOP_OFFSET - TEXTUREVIEW_BOTTOM_OFFSET; + + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + U8 inbounds = mx >= x && mx <= x + w && my >= y && my <= y + h; + if( !inbounds ) + return; + + U8 scroll = gui_mbutton_down( GUI_MBTNSCROLL ); + gui_capture_scroll(); + + if( scroll == 1 ) + wnd->scrolloff -= 20; +} + +void gui_editor_texturepicker_trim_texname( GUI_EDITOR_TEXTUREPICKER* wnd, char* trimname ) { + I32 size = (wnd->w - 20) / wnd->rowcount - 10; + I32 tw; + while( 1 ) { + gui_draw_get_str_bounds( &tw, 0, FNT_JPN12, trimname ); + if( tw >= size ) { + if( trimname[0] != '.' && trimname[1] != '.' && trimname[2] != '.' ) { + sprintf( trimname, "...%s", trimname ); + } + + for( U32 i = 3; !!trimname[i]; ++i ) + trimname[i] = trimname[i + 1]; + } else break; + } +} + +void gui_editor_texturepicker_draw_file_str( I32 x, I32 y, const char* str, U8 issel ) { + if( !issel ) { + return gui_draw_str( + x, y, + ALIGN_C, + FNT_JPN12, + ui_clr.txt, + str + ); + } + + I32 w, h; + gui_draw_get_str_bounds( &w, &h, FNT_JPN12, str ); + w += 2; + + gui_draw_frect( x - w / 2, y + 1, w, h, ui_clr.border ); + gui_draw_str( x, y, ALIGN_C, FNT_JPN12, ui_clr.bg_alt, str ); +} + +void gui_editor_texturepicker_draw_files( GUI_EDITOR_TEXTUREPICKER* wnd ) { + I32 x = wnd->x + 11; + I32 y = wnd->y + TEXTUREVIEW_TOP_OFFSET; + I32 w = wnd->w - 22; + I32 size = (w + 10) / wnd->rowcount - 10; + + I32 colcnt = (I32)floorf( (F32)w / size ); + CLR clr = gui_is_fg_window( wnd )? ui_clr.border : ui_clr.border_inactive; + + U32 i = 0; + wnd->textures.each( fn( GL_TEX2D** ptr ) { + GL_TEX2D* tex = *ptr; + + if( strlen( wnd->search ) > 0 && !strstr( tex->name, wnd->search ) ) + return; + + I32 tx = x + (i % colcnt) * (size + 10); + I32 ty = y + (i / colcnt) * (size + 18); + + gui_draw_frect( tx - 1, ty - 1, size + 2, size + 2, clr ); + gui_draw_frect( tx, ty, size, size, clr ); + gl_2d_textured_frect( + _gui.gl2d_font, + { (F32)tx, (F32)ty }, + { (F32)size, (F32)size }, + tex, + CLR::WHITE() + ); + + char trimname[256]; + strcpy( trimname, assets_abspath( tex->name ) ); + gui_editor_texturepicker_trim_texname( wnd, trimname ); + + gui_editor_texturepicker_draw_file_str( + tx + size / 2, + ty + size + 1, + trimname, + wnd->curselect == tex + ); + + ++i; + } ); +} + +void gui_editor_texturepicker_draw_fn( void* ptr ) { + GUI_EDITOR_TEXTUREPICKER* wnd = (GUI_EDITOR_TEXTUREPICKER*)ptr; + + CLR clr = gui_is_fg_window( wnd )? ui_clr.border : ui_clr.border_inactive; + gui_draw_frect( wnd->x, wnd->y, wnd->w, wnd->h, clr ); + gui_draw_frect( wnd->x + 1, wnd->y + 1, wnd->w - 2, wnd->h - 2, ui_clr.bg ); + + //gui_draw_str( wnd->x + 2, wnd->y + 2, ALIGN_LEFT, FNT_JPN12, ui_clr.txt, "texture list" ); + gui_editor_texturepicker_draw_files( wnd ); + + if( !wnd->loaded ) { + gui_draw_str( + wnd->x + wnd->w - 2, + wnd->y, + ALIGN_R, + FNT_JPN12, + ui_clr.txt, + "loading... [%d/%d]", wnd->curload + 1, wnd->files.size + ); + } + + wnd->children.each( fn( GUI_BASE** childptr ) { + GUI_BASE* child = *childptr; + if( !child->enabled ) return; + if( child->draw_fn ) + child->draw_fn( child ); + } ); +} + +void gui_editor_textureview_delete_textures( GUI_EDITOR_TEXTUREPICKER* wnd ) { + wnd->textures.each( fn( GL_TEX2D** ptr ) { + GL_TEX2D* tex = *ptr; + gl_texture_destroy( _gui.gl2d->gl, tex ); + } ); + + wnd->textures.clear(); +} + +void gui_editor_texturepicker_load_texture( void* ptr ) { + GUI_EDITOR_TEXTUREPICKER* wnd = (GUI_EDITOR_TEXTUREPICKER*)ptr; + + FILE_ENTRY* f = &wnd->files[wnd->curload]; + const char* name = assets_abspath( f->name ); + GL_TEX2D* tex = gl_texture_from_file( _gui.gl2d->gl, name ); + if( !tex ) { + dlog( "gui_editor_textureview_create_textures() : error creating texture %s", name ); + return; + } + wnd->textures.push( tex ); + + if( ++wnd->curload < wnd->files.size ) + gui_push_callback( wnd, gui_editor_texturepicker_load_texture ); + else + wnd->loaded = 1; +} + +void gui_editor_textureview_create_textures( GUI_EDITOR_TEXTUREPICKER* wnd ) { + wnd->curload = 0; + GL_TEX2D* emptytex = gl_texture_create( _gui.gl2d->gl, "" ); + strcpy( emptytex->name, "none" ); + emptytex->width = emptytex->height = 1; + + wnd->textures.push( emptytex ); + + gui_push_callback( wnd, gui_editor_texturepicker_load_texture ); +} + +void gui_editor_textureview_accept( GUI_EDITOR_TEXTUREPICKER* wnd ) { + if( !wnd->curselect ) + return; + + U8 isnone = !strcmp( wnd->curselect->name, "none" ); + wnd->textures.each( fn( GL_TEX2D** ptr ) { + GL_TEX2D* tex = *ptr; + if( !isnone && tex == wnd->curselect ) + return; + + gl_texture_destroy( _gui.gl2d->gl, tex ); + } ); + + if( !isnone ) { + if( map_add_texture_ref( editor->map, wnd->curselect ) == STAT_ALREADYEXISTS ) { + const char* name = assets_abspath( wnd->curselect->name ); + GL_TEX2D* tex = map_find_texture( editor->map, name ); + *wnd->target = tex; + + gl_texture_destroy( _gui.gl2d->gl, wnd->curselect ); + } else { + *wnd->target = wnd->curselect; + } + } else { + *wnd->target = 0; + } + + if( wnd->cb ) + wnd->cb( wnd ); + + wnd->textures.clear(); + gui_push_callback( wnd, pfn( void* p ) { gui_free( (GUI_BASE*)p ); } ); +} + +U8 gui_editor_texturepicker_input_files( GUI_EDITOR_TEXTUREPICKER* wnd ) { + I32 x = wnd->x + 11; + I32 y = wnd->y + TEXTUREVIEW_TOP_OFFSET; + I32 w = wnd->w - 22; + I32 size = (w + 10) / wnd->rowcount - 10; + + I32 colcnt = (I32)floorf( (F32)w / size ); + I32 i = 0; + + U8 m1 = gui_mbutton_down( 0 ); + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + wnd->textures.each( fn( GL_TEX2D** ptr ) { + GL_TEX2D* tex = *ptr; + + // yea technically this is slow as fuck but whatever + if( strlen( wnd->search ) > 0 && !strstr( tex->name, wnd->search ) ) + return; + + I32 tx = x + (i % colcnt) * (size + 10); + I32 ty = y + (i / colcnt) * (size + 18); + + U8 inbounds = ( (mx >= tx) && (mx < tx + size) && (my >= ty) && (my < ty + size + 18) ); + if( inbounds && m1 && !wnd->heldselect ) { + wnd->heldselect = tex; + } + if( !m1 && wnd->heldselect == tex ) { + wnd->curselect = tex; + } + + ++i; + } ); + + if( !m1 ) { + wnd->heldselect = 0; + } + + return 0; +} + +void gui_editor_texturepicker_input_fn( void* ptr ) { + GUI_EDITOR_TEXTUREPICKER* wnd = (GUI_EDITOR_TEXTUREPICKER*)ptr; + + if( !gui_editor_texturepicker_input_files( wnd ) ) + gui_base_input_fn( ptr ); +} + +void gui_editor_texturepicker_update_density( GUI_EDITOR_TEXTUREPICKER* wnd ) { + wnd->scrollheight = gui_editor_texturepicker_get_scrollheight( wnd ); + sprintf( wnd->densitylabel->name, "density: %d", wnd->rowcount ); +} + +void gui_editor_textureview_create_bottom_row( GUI_EDITOR_TEXTUREPICKER* wnd ) { + I32 x = 10; + + gui_label( x, 2, "filter:" ); + GUI_TEXTBOX* tb = gui_textbox( x + 50, -15, 92, 20, "", 256 ); + tb->cb = pfn( void* ptr ) { + GUI_TEXTBOX* tb = (GUI_TEXTBOX*)ptr; + GUI_EDITOR_TEXTUREPICKER* wnd = (GUI_EDITOR_TEXTUREPICKER*)tb->parent->parent; + strcpy( wnd->search, tb->value ); + }; + x += 150; + + wnd->densitylabel = gui_label( x, 2, "density: %d", wnd->rowcount ); + gui_button( x + 66, 0, 20, 20, "+", pfn( void* ptr ) { + GUI_BUTTON* btn = (GUI_BUTTON*)ptr; + GUI_EDITOR_TEXTUREPICKER* wnd = (GUI_EDITOR_TEXTUREPICKER*)btn->parent->parent; + if( wnd->rowcount < 9 ) wnd->rowcount++; + gui_editor_texturepicker_update_density( wnd ); + } ); + + gui_button( x + 91, 0, 20, 20, "-", pfn( void* ptr ) { + GUI_BUTTON* btn = (GUI_BUTTON*)ptr; + GUI_EDITOR_TEXTUREPICKER* wnd = (GUI_EDITOR_TEXTUREPICKER*)btn->parent->parent; + if( wnd->rowcount > 3 ) wnd->rowcount--; + gui_editor_texturepicker_update_density( wnd ); + } ); + + gui_button( wnd->w - 120, 0, 50, 20, "cancel", pfn( void* ptr ) { + GUI_BUTTON* btn = (GUI_BUTTON*)ptr; + GUI_EDITOR_TEXTUREPICKER* wnd = (GUI_EDITOR_TEXTUREPICKER*)btn->parent->parent; + gui_editor_textureview_delete_textures( wnd ); + gui_push_callback( wnd, pfn( void* p ) { gui_free( (GUI_BASE*)p ); } ); + } ); + + gui_button( wnd->w - 60, 0, 50, 20, "ok", pfn( void* ptr ) { + GUI_BUTTON* btn = (GUI_BUTTON*)ptr; + GUI_EDITOR_TEXTUREPICKER* wnd = (GUI_EDITOR_TEXTUREPICKER*)btn->parent->parent; + gui_editor_textureview_accept( wnd ); + } ); +} + +void gui_editor_texturepicker_init( void* ptr ) { + GUI_EDITOR_TEXTUREPICKER* wnd = (GUI_EDITOR_TEXTUREPICKER*)ptr; + + wnd->files = assets_get_files_by_ext( TEXTURE_EXTENSIONS ); + gui_editor_textureview_create_textures( wnd ); +} + +GUI_EDITOR_TEXTUREPICKER* gui_editor_texturepicker( I32 x, I32 y, I32 w, I32 h, GL_TEX2D **target ) { + GUI_EDITOR_TEXTUREPICKER* wnd = new GUI_EDITOR_TEXTUREPICKER; + wnd->x = x; + wnd->y = y; + wnd->w = w; + wnd->h = h; + wnd->ontop = 1; + wnd->locked = 0; + wnd->draw_fn = gui_editor_texturepicker_draw_fn; + wnd->input_fn = gui_editor_texturepicker_input_fn; + strcpy( wnd->name, "TEXTURE_PICKER_WINDOW" ); + wnd->cb = 0; + + _gui.windows.push( wnd ); + gui_set_window( wnd ); + + wnd->target = target; + wnd->rowcount = 3; + wnd->curselect = 0; + wnd->scrolloff = 0; + wnd->scrollheight = gui_editor_texturepicker_get_scrollheight( wnd ); + wnd->search[0] = 0; + + wnd->loaded = 0; + gui_push_callback( wnd, gui_editor_texturepicker_init ); + + GUI_VIEW* view = gui_get_view(); + + gui_set_view( 0 ); + gui_view( 0, 0, w, 20 ); + gui_title( "texture picker" ); + gui_set_view( 0 ); + + gui_view( 0, h - TEXTUREVIEW_BOTTOM_OFFSET - 1, w, TEXTUREVIEW_BOTTOM_OFFSET ); + gui_editor_textureview_create_bottom_row( wnd ); + + gui_set_view( view ); + + return wnd; +} diff --git a/src/editor/view2d.cpp b/src/editor/view2d.cpp new file mode 100644 index 0000000..35fd208 --- /dev/null +++ b/src/editor/view2d.cpp @@ -0,0 +1,938 @@ +#include <math.h> +#include "editor.h" +#include "../render/gl_2d.h" + +#include "../game/objlist.h" + +const I32 EDITORVIEW_TITLE_OFFSET = 15; +const I32 EDITORVIEW_GUTTERS_OFFSETX = 22; +const I32 EDITORVIEW_GUTTERS_OFFSETY = 16; + +F32 gui_editor_2dview_calc_scale( GUI_EDITOR_2DVIEW* view ) { + WORLD_MAP* m = editor->map; + + I32 w = view->w - EDITORVIEW_GUTTERS_OFFSETX - 1; + I32 h = view->h - EDITORVIEW_GUTTERS_OFFSETY - 1; + + F32 mapw = m->maxs.x - m->mins.x; + F32 maph = m->maxs.y - m->mins.y; + + F32 scale = min( w / mapw, h / maph ); + return scale * view->scale; +} + +VEC2 gui_editor_2dview_screen_to_world( GUI_EDITOR_2DVIEW* view, I32 x, I32 y ) { + WORLD_MAP* m = editor->map; + + I32 _x = gui_relx( view ) + EDITORVIEW_GUTTERS_OFFSETX; + I32 _y = gui_rely( view ) + EDITORVIEW_TITLE_OFFSET + EDITORVIEW_GUTTERS_OFFSETY; + x -= _x; y -= _y; + + F32 w = view->w - EDITORVIEW_GUTTERS_OFFSETX; + F32 h = view->h - EDITORVIEW_GUTTERS_OFFSETY; + + F32 tx = x / w; + F32 ty = y / h; + + F32 scaledx = m->mins.x + (m->maxs.x - m->mins.x) * tx; + F32 scaledy = m->mins.y + (m->maxs.y - m->mins.y) * ty; + + if( w > h ) + scaledx *= ( w / h ); + if( w < h ) + scaledy *= h / w; + + scaledx /= view->scale; + scaledx += view->posx; + scaledy /= view->scale; + scaledy += view->posy; + + return { scaledx, scaledy }; +} + +VEC2 gui_editor_2dview_world_to_screen( GUI_EDITOR_2DVIEW* view, VEC2 world ) { + WORLD_MAP* m = editor->map; + F32 scale = gui_editor_2dview_calc_scale( view ); + + F32 xoff = m->mins.x + view->posx; + F32 yoff = m->mins.y + view->posy; + + F32 vx = (F32)gui_relx( view ); + F32 vy = (F32)gui_rely( view ); + vy += EDITORVIEW_TITLE_OFFSET + EDITORVIEW_GUTTERS_OFFSETY; + vx += EDITORVIEW_GUTTERS_OFFSETX; + + F32 x = vx + (world.x - xoff) * scale; + F32 y = vy + (world.y - yoff) * scale; + + return { x, y }; +} + +void gui_editor_2dview_select( GUI_EDITOR_2DVIEW* view, void* what, U8 seltype ) { + view->curselect = what; + view->seltype = seltype; +} + +U8 gui_editor_2dview_is_gizmo_active( GUI_EDITOR_2DVIEW* view, void* what, U8 seltype ) { + GUI_EDITOR_PROPVIEW* props = editor->gui.props; + if( props->seltype == EDITOR_SELECT_WALL && (seltype == EDITOR_SELECT_WVERTEX) ) { + MAP_WALL* s = (MAP_WALL*)props->curselect; + if( what == &s->start || what == &s->end ) + return 1; + } + else if( props->seltype == seltype && props->curselect == what ) { + return 1; + } + + if( view->curdrag ) { + return what == view->curdrag && seltype == view->dragtype; + } + else { + return what == view->curselect && seltype == view->seltype; + } +} + +void gui_editor_2dview_draw_gizmo( I32 x, I32 y, CLR clr, U8 selected = 0 ) { + const I32 GIZMO_SIZE = selected ? 10 : 6; + const I32 ihalf = GIZMO_SIZE / 2; + + gui_draw_frect( x - ihalf - 1, y - ihalf - 1, GIZMO_SIZE, GIZMO_SIZE, { .7f, .7f, .7f, 1.f } ); + gui_draw_frect( x - ihalf + 1, y - ihalf + 1, GIZMO_SIZE, GIZMO_SIZE, { 0.f, 0.f, 0.f, 1.f } ); + gui_draw_frect( x - ihalf, y - ihalf, GIZMO_SIZE, GIZMO_SIZE, clr ); +} + +void gui_editor_2dview_draw_gutters( GUI_EDITOR_2DVIEW* view, I32 x, I32 y ) { + WORLD_MAP* m = editor->map; + F32 w = view->w; + F32 h = view->h; + + gui_draw_str( x + 2, y, ALIGN_L, FNT_JPN12, ui_clr.txt, "%1.f,%1.f", view->posx, view->posy ); + for( U32 i = 1; i <= 5; ++i ) { + F32 tx = x + i * (w / 5); + F32 ty = y; + + F32 tp = (tx - x) / w; + + F32 scaledx = m->mins.x + (m->maxs.x - m->mins.x) * tp; + if( w > h ) + scaledx *= ( w / h ); + scaledx /= view->scale; + scaledx += view->posx; + gui_draw_str( tx - 2, ty, ALIGN_R, FNT_JPN12, ui_clr.txt, "%1.f", scaledx ); + gui_draw_line( tx, ty, tx + 1, ty + 16, ui_clr.txt ); + } + + for( U32 i = 1; i <= 5; ++i ) { + F32 tx = x + 2; + F32 ty = y + i * (h / 5); + + F32 tp = (ty - y) / h; + + F32 scaledy = m->mins.y + (m->maxs.y - m->mins.y) * tp; + if( w < h ) + scaledy *= h / w; + scaledy /= view->scale; + scaledy += view->posy; + gui_draw_str( tx, ty - 16, ALIGN_L, FNT_JPN12, ui_clr.txt, "%1.f", scaledy ); + gui_draw_line( tx - 2, ty, tx + 20, ty, ui_clr.txt ); + } +} + +void gui_editor_2dview_draw_polygons( GUI_EDITOR_2DVIEW* view, I32 x, I32 y ) { + WORLD_MAP* m = editor->map; + + F32 scale = gui_editor_2dview_calc_scale( view ); + F32 xoff = m->mins.x + view->posx; + F32 yoff = m->mins.y + view->posy; + + m->polygons.each( fn( MAP_POLYGON* p ) { + SURF_PROPS* props = polygon_get_props( m, p ); + if( !editor->wireframe ) { + LIST<VERTEX> vertices; + p->vertices.each( fn( MAP_VERTEX* v ) { + VERTEX v2; + v2.uv = v->uv; + v2.pos.x = x + (v->pos.x - xoff) * scale; + v2.pos.y = y + (v->pos.y - yoff) * scale; + v2.clr = props->clr; + vertices.push( v2 ); + } ); + + if( props->tex ) + gl_textured_polygon( _gui.gl2d_font, vertices.data, vertices.size, props->tex ); + else + gl_polygon( _gui.gl2d, vertices.data, vertices.size ); + } + else { + for( U32 i = 0; i < p->vertices.size; ++i ) { + MAP_VERTEX* v1 = &p->vertices[i]; + MAP_VERTEX* v2 = &p->vertices[(i+1) % p->vertices.size]; + + I32 x0 = x + (I32)( (v1->pos.x - xoff) * scale ); + I32 y0 = y + (I32)( (v1->pos.y - yoff) * scale ); + I32 x1 = x + (I32)( (v2->pos.x - xoff) * scale ); + I32 y1 = y + (I32)( (v2->pos.y - yoff) * scale ); + + gui_draw_line( x0, y0, x1, y1, props->clr ); + } + } + + if( editor->tool != EDITOR_TOOL_POLY && editor->tool != EDITOR_TOOL_SELECT ) + return; + // draw gizmos + p->vertices.each( fn( MAP_VERTEX* v ) { + I32 vx = (I32)( x + (v->pos.x - xoff) * scale ); + I32 vy = (I32)( y + (v->pos.y - yoff) * scale ); + U8 selected = gui_editor_2dview_is_gizmo_active( view, v, EDITOR_SELECT_PVERTEX ) + || gui_editor_2dview_is_gizmo_active( view, p, EDITOR_SELECT_POLY ); + + gui_editor_2dview_draw_gizmo( vx, vy, CLR::CYAN(), selected ); + } ); + } ); +} + +void gui_editor_2dview_draw_walls( GUI_EDITOR_2DVIEW* view, I32 x, I32 y ) { + WORLD_MAP* m = editor->map; + + F32 scale = gui_editor_2dview_calc_scale( view ); + F32 xoff = m->mins.x + view->posx; + F32 yoff = m->mins.y + view->posy; + + m->walls.each( fn( MAP_WALL* s ) { + SURF_PROPS* props = wall_get_props( m, s ); + + I32 x0 = (I32)( x + (s->start.x - xoff) * scale ); + I32 y0 = (I32)( y + (s->start.y - yoff) * scale ); + I32 x1 = (I32)( x + (s->end.x - xoff) * scale ); + I32 y1 = (I32)( y + (s->end.y - yoff) * scale ); + + gui_draw_line( x0, y0, x1, y1, props->clr ); + if( gui_editor_2dview_is_gizmo_active( view, s, EDITOR_SELECT_WALL ) ) { + gui_draw_line( x0, y0 - 1, x1, y1 - 1, CLR::WHITE( .5f ) ); + gui_draw_line( x0, y0 + 1, x1, y1 + 1, CLR::WHITE( .5f ) ); + gui_draw_line( x0 - 1, y0, x1 - 1, y1, CLR::WHITE( .5f ) ); + gui_draw_line( x0 + 1, y0, x1 + 1, y1, CLR::WHITE( .5f ) ); + } + } ); + + if( editor->tool != EDITOR_TOOL_WALL && editor->tool != EDITOR_TOOL_SELECT ) + return; + + // gizmos + m->walls.each( fn( MAP_WALL* s ) { + I32 x0 = (I32)( x + (s->start.x - xoff) * scale ); + I32 y0 = (I32)( y + (s->start.y - yoff) * scale ); + I32 x1 = (I32)( x + (s->end.x - xoff) * scale ); + I32 y1 = (I32)( y + (s->end.y - yoff) * scale ); + + U8 sel = gui_editor_2dview_is_gizmo_active( view, s, EDITOR_SELECT_WALL ); + U8 sel1 = gui_editor_2dview_is_gizmo_active( view, &s->start, EDITOR_SELECT_WVERTEX ); + U8 sel2 = gui_editor_2dview_is_gizmo_active( view, &s->end, EDITOR_SELECT_WVERTEX ); + + gui_editor_2dview_draw_gizmo( x0, y0, CLR::GREEN(), sel || sel1 ); + gui_editor_2dview_draw_gizmo( x1, y1, CLR::GREEN(), sel || sel2 ); + } ); +} + +void gui_editor_2dview_draw_sprites( GUI_EDITOR_2DVIEW* view, I32 x, I32 y ) { + WORLD_MAP* m = editor->map; + + F32 scale = gui_editor_2dview_calc_scale( view ); + F32 xoff = m->mins.x + view->posx; + F32 yoff = m->mins.y + view->posy; + + m->sprites.each( fn( MAP_SPRITE* s ) { + F32 wantedsize = editor->spritesize; + F32 aspect = s->size.x / s->size.y; + F32 w = (aspect > 1.0f)? wantedsize : wantedsize * aspect; + F32 h = (aspect < 1.0f)? wantedsize : wantedsize / aspect; + VEC2 pos = { + (F32)x + (s->pos.x - xoff) * scale, + (F32)y + (s->pos.y - yoff) * scale + }; + + U8 sel= gui_editor_2dview_is_gizmo_active( view, s, EDITOR_SELECT_SPRITE ); + if( sel ) { + gui_draw_rect( pos.x - w/2 - 1, pos.y - h/2 - 1, w + 2, h + 2, CLR::WHITE( 0.5f ) ); + } + + gl_2d_textured_frect( + _gui.gl2d_font, + { (F32)((I32)pos.x - w/2), (F32)((I32)pos.y - h/2) }, + { (F32)w, (F32)h }, + s->tex, + s->clr + ); + } ); +} + +void gui_editor_2dview_draw_player( GUI_EDITOR_2DVIEW* view, I32 x, I32 y ) { + WORLD_MAP* m = editor->map; + + if( !objl->pl ) + return; + + F32 scale = gui_editor_2dview_calc_scale( view ); + F32 xoff = m->mins.x + view->posx; + F32 yoff = m->mins.y + view->posy; + + VEC3 pos = objl->pl->pos; + + VEC2 pos2d = { + (F32)x + (pos.x - xoff) * scale, + (F32)y + (pos.y - yoff) * scale + }; + + F32 yaw = objl->pl->rot.y; + VEC2 dir = m_radial_offset( yaw, 20.f ); + VEC2 ray = pos2d + dir; + + gui_draw_line( (I32)pos2d.x, (I32)pos2d.y, (I32)ray.x, (I32)ray.y, CLR::GREEN() ); + gui_draw_frect( (I32)pos2d.x-4, (I32)pos2d.y-4, 8, 8, CLR::WHITE() ); + gui_draw_frect( (I32)pos2d.x-3, (I32)pos2d.y-3, 6, 6, CLR::RED() ); +} + +void gui_editor_2dview_draw_origin( GUI_EDITOR_2DVIEW* view, I32 x, I32 y ) { + WORLD_MAP* m = editor->map; + + F32 scale = gui_editor_2dview_calc_scale( view ); + F32 xoff = m->mins.x + view->posx; + F32 yoff = m->mins.y + view->posy; + + VEC3 pos = m->startpos; + I32 screenx = (I32)( x + (pos.x - xoff) * scale ); + I32 screeny = (I32)( y + (pos.y - yoff) * scale ); + + if( gui_editor_2dview_is_gizmo_active( view, m, EDITOR_SELECT_ORIGIN ) ) { + I32 w, h; + gui_draw_get_str_bounds( &w, &h, FNT_JPN12, "(origin)" ); + gui_draw_frect( screenx - w / 2 - 1, screeny + h - 1, w, 1, CLR::BLACK() ); + gui_draw_frect( screenx - w / 2, screeny + h, w, 1, ui_clr.txt ); + gui_draw_str( screenx - 1, screeny - 1, ALIGN_C, FNT_JPN12, CLR::WHITE( .5f ), "(origin)" ); + } + + gui_draw_str( screenx + 1, screeny + 1, ALIGN_C, FNT_JPN12, CLR::BLACK(), "(origin)" ); + gui_draw_str( screenx, screeny, ALIGN_C, FNT_JPN12, ui_clr.txt, "(origin)" ); +} + +void gui_editor_2dview_draw_fn( void* ptr ) { + GUI_EDITOR_2DVIEW* view = (GUI_EDITOR_2DVIEW*)ptr; + if( !editor->map ) + return; + + I32 x = gui_relx( view ); + I32 y = gui_rely( view ); + I32 w = view->w; + I32 h = view->h; + + gui_draw_str( x, y, ALIGN_L, FNT_JPN12, ui_clr.txt, "2d view" ); + y += EDITORVIEW_TITLE_OFFSET; + + CLR col = gui_is_fg_window( view )? ui_clr.border : ui_clr.border_inactive; + gui_draw_frect( x-1, y-1, view->w+2, view->h+2, col ); + gui_draw_frect( x, y, view->w, view->h, CLR::BLACK() ); + + gui_editor_2dview_draw_gutters( view, x, y ); + + U32 offx = EDITORVIEW_GUTTERS_OFFSETX; + U32 offy = EDITORVIEW_GUTTERS_OFFSETY; + + gui_draw_push_clip( x + offx, y + offy, w - offx, h - offy ); + gui_editor_2dview_draw_polygons( view, x + offx, y + offy ); + gui_editor_2dview_draw_walls( view, x + offx, y + offy ); + gui_editor_2dview_draw_sprites( view, x + offx, y + offy ); + gui_editor_2dview_draw_player( view, x + offx, y + offy ); + gui_editor_2dview_draw_origin( view, x + offx, y + offy ); + gui_draw_pop_clip(); +} + +U8 gui_editor_2dview_input_drag( GUI_EDITOR_2DVIEW* view, U8 mouse ) { + if( !mouse ) { + view->dragheld = 0; + return 0; + } + + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + I32 x = gui_relx( view ); + I32 y = gui_rely( view ) + EDITORVIEW_TITLE_OFFSET; + I32 w = view->w - EDITORVIEW_GUTTERS_OFFSETX; + I32 h = view->h - EDITORVIEW_GUTTERS_OFFSETY; + x += EDITORVIEW_GUTTERS_OFFSETX; + y += EDITORVIEW_GUTTERS_OFFSETY; + + U8 inbound = mx >= x && mx <= x + w && my >= y && my <= y + h; + if( !inbound && !view->dragheld ) + return 0; + + I32 moffx = mx - x; + I32 moffy = my - y; + + if( !view->dragheld ) { + view->moffx = moffx; + view->moffy = moffy; + view->voffx = view->posx; + view->voffy = view->posy; + view->dragheld = 1; + return 1; + } + + F32 iscale = 1.f / gui_editor_2dview_calc_scale( view ); + + view->posx = view->voffx - (moffx - view->moffx) * iscale; + view->posy = view->voffy - (moffy - view->moffy) * iscale; + return 1; +} + +void gui_editor_2dview_input_tool_none( GUI_EDITOR_2DVIEW* view ) { + U8 mouse = gui_mbutton_down( 0 ) || gui_mbutton_down( 1 ); + + gui_editor_2dview_input_drag( view, mouse ); +} + +VEC2 gui_editor_2dview_input_get_drag_vec( GUI_EDITOR_2DVIEW* view ) { + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + F32 iscale = 1.f / gui_editor_2dview_calc_scale( view ); + F32 dx = (mx - view->oldmx) * iscale + view->mremainx; + F32 dy = (my - view->oldmy) * iscale + view->mremainy; + + F32 igridx = copysignf( floorf( fabsf( dx / editor->grid ) ), dx ); + F32 igridy = copysignf( floorf( fabsf( dy / editor->grid ) ), dy ); + + return { editor->grid * igridx, editor->grid * igridy }; +} + +void gui_editor_2dview_input_select_onmove( GUI_EDITOR_2DVIEW* view ) { + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + F32 iscale = 1.f / gui_editor_2dview_calc_scale( view ); + F32 dx = (mx - view->oldmx) * iscale + view->mremainx; + F32 dy = (my - view->oldmy) * iscale + view->mremainy; + + F32 rmx = copysignf( fmodf( fabsf( dx ), editor->grid ), dx ); + F32 rmy = copysignf( fmodf( fabsf( dy ), editor->grid ), dy ); + + view->oldmx = mx; + view->oldmy = my; + view->mremainx = rmx; + view->mremainy = rmy; + view->dragmoved = 1; + + GUI_EDITOR_PROPVIEW* props = editor->gui.props; + // special case for dragging wall vertices, just always update + U8 iswallv = props->seltype == EDITOR_SELECT_WALL && view->dragtype == EDITOR_SELECT_WVERTEX; + if( props->curselect == view->curdrag || iswallv ) + gui_editor_propview_update( editor->gui.props ); +} + +void gui_editor_2dview_input_select_drag_wall( GUI_EDITOR_2DVIEW* view ) { + MAP_WALL* s = (MAP_WALL*)view->curdrag; + + s->start.x = m_snap_to_grid( s->start.x, editor->grid ); + s->start.y = m_snap_to_grid( s->start.y, editor->grid ); + s->end.x = m_snap_to_grid( s->end.x, editor->grid ); + s->end.y = m_snap_to_grid( s->end.y, editor->grid ); + + VEC2 mv = gui_editor_2dview_input_get_drag_vec( view ); + if( !is_zero( mv ) ) { + s->start += mv; + s->end += mv; + map_check_bounds( editor->map ); + + gui_editor_2dview_input_select_onmove( view ); + } +} + +void gui_editor_2dview_input_select_drag_vertex( GUI_EDITOR_2DVIEW* view ) { + VEC2* v = (VEC2*)view->curdrag; + v->x = m_snap_to_grid( v->x, editor->grid ); + v->y = m_snap_to_grid( v->y, editor->grid ); + + VEC2 mv = gui_editor_2dview_input_get_drag_vec( view ); + if( !is_zero( mv ) ) { + *v += mv; + map_check_bounds( editor->map ); + + gui_editor_2dview_input_select_onmove( view ); + } +} + +void gui_editor_2dview_input_select_drag_polygon( GUI_EDITOR_2DVIEW* view ) { + MAP_POLYGON* p = (MAP_POLYGON*)view->curdrag; + + VEC2 mv = gui_editor_2dview_input_get_drag_vec( view ); + if( !is_zero( mv ) ) { + p->vertices.each( fn( MAP_VERTEX* v ) { + v->pos.x = m_snap_to_grid( v->pos.x, editor->grid ); + v->pos.y = m_snap_to_grid( v->pos.y, editor->grid ); + + v->pos.x += mv.x; + v->pos.y += mv.y; + } ); + + map_polygon_calc_bounds( p ); + map_check_bounds( editor->map ); + gui_editor_2dview_input_select_onmove( view ); + } +} + +void gui_editor_2dview_input_select_drag_sprite( GUI_EDITOR_2DVIEW* view ) { + MAP_SPRITE* s = (MAP_SPRITE*)view->curdrag; + + s->pos.x = m_snap_to_grid( s->pos.x, editor->grid ); + s->pos.y = m_snap_to_grid( s->pos.y, editor->grid ); + + VEC2 mv = gui_editor_2dview_input_get_drag_vec( view ); + if( !is_zero( mv ) ) { + s->pos += mv; + gui_editor_2dview_input_select_onmove( view ); + } +} + +void gui_editor_2dview_input_select_drag_origin( GUI_EDITOR_2DVIEW* view ) { + WORLD_MAP* m = editor->map; + + m->startpos.x = m_snap_to_grid( m->startpos.x, editor->grid ); + m->startpos.y = m_snap_to_grid( m->startpos.y, editor->grid ); + + VEC2 mv = gui_editor_2dview_input_get_drag_vec( view ); + if( !is_zero( mv ) ) { + m->startpos += mv; + gui_editor_2dview_input_select_onmove( view ); + } +} + +void gui_editor_2dview_input_select_ondrop( GUI_EDITOR_2DVIEW* view ) { + if( !view->dragmoved ) + gui_editor_propview_select( editor->gui.props, view->curdrag, view->dragtype ); + + view->curdrag = 0; +} + +void gui_editor_2dview_input_select_drag( GUI_EDITOR_2DVIEW* view ) { + U8 m1 = gui_mbutton_down( 0 ); + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + if( !m1 ) { + if( view->held ) { + gui_editor_2dview_input_select_ondrop( view ); + view->held = 0; + } + return; + } + + if( !view->held ) { + view->curdrag = view->curselect; + view->dragtype = view->seltype; + view->oldmx = mx; + view->oldmy = my; + view->mremainx = view->mremainy = 0; + view->dragmoved = 0; + view->held = 1; + return; + } + + switch( view->dragtype ) { + case EDITOR_SELECT_WALL: gui_editor_2dview_input_select_drag_wall( view ); break; + case EDITOR_SELECT_POLY: gui_editor_2dview_input_select_drag_polygon( view ); break; + case EDITOR_SELECT_SPRITE: gui_editor_2dview_input_select_drag_sprite( view ); break; + case EDITOR_SELECT_ORIGIN: gui_editor_2dview_input_select_drag_origin( view ); break; + case EDITOR_SELECT_WVERTEX: + case EDITOR_SELECT_PVERTEX: + gui_editor_2dview_input_select_drag_vertex( view ); break; + default: { + view->oldmx = mx; + view->oldmy = my; + } break; + } +} + +void gui_editor_2dview_input_select_walls( GUI_EDITOR_2DVIEW* view ) { + WORLD_MAP* m = editor->map; + + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + VEC2 mpos = { (F32)mx, (F32)my }; + + F32 mindist = 5.0001f; + for( U32 i = 0; i < m->walls.size; ++i ) { + MAP_WALL* w = &m->walls[i]; + + VEC2 w1 = { w->start.x, w->start.y }; + VEC2 w2 = { w->end.x, w->end.y }; + w1 = gui_editor_2dview_world_to_screen( view, w1 ); + w2 = gui_editor_2dview_world_to_screen( view, w2 ); + + F32 d1 = vec_dist( mpos, w1 ); + F32 d2 = vec_dist( mpos, w2 ); + if( d1 < mindist && d1 < d2 ) { + gui_editor_2dview_select( view, &w->start, EDITOR_SELECT_WVERTEX ); + mindist = d1; + continue; + } + else if( d2 < mindist ) { + gui_editor_2dview_select( view, &w->end, EDITOR_SELECT_WVERTEX ); + mindist = d2; + continue; + } + // give priority to vertices + if( view->seltype == EDITOR_SELECT_WVERTEX ) + continue; + + F32 dist = m_dist_line_to_point( w1, w2, mpos ); + if( dist < 3.f && dist < mindist ) { + gui_editor_2dview_select( view, w, EDITOR_SELECT_WALL ); + mindist = dist; + } + } + + if( !view->curselect && !view->held ) + return; + + gui_editor_2dview_input_select_drag( view ); +} + +void gui_editor_2dview_input_select_polygons( GUI_EDITOR_2DVIEW* view ) { + WORLD_MAP* m = editor->map; + + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + VEC2 mpos = { (F32)mx, (F32)my }; + + F32 mindist = 5.0001f; + for( U32 i = 0; i < m->polygons.size; ++i ) { + MAP_POLYGON* p = &m->polygons[i]; + LIST<VEC2> plist; + + p->vertices.each( fn( MAP_VERTEX* v ) { + VEC2 world = { v->pos.x, v->pos.y }; + VEC2 screen = gui_editor_2dview_world_to_screen( view, world ); + F32 d = vec_dist( mpos, screen ); + if( d < mindist ) { + gui_editor_2dview_select( view, v, EDITOR_SELECT_PVERTEX ); + mindist = d; + } + + plist.push( screen ); + } ); + + if( view->seltype == EDITOR_SELECT_PVERTEX ) + continue; + + if( m_point_in_polygon( mpos, plist.data, plist.size ) ) + gui_editor_2dview_select( view, p, EDITOR_SELECT_POLY ); + } + + if( !view->curselect && !view->held ) + return; + + gui_editor_2dview_input_select_drag( view ); +} + +void gui_editor_2dview_input_select_sprites( GUI_EDITOR_2DVIEW* view ) { + WORLD_MAP* m = editor->map; + + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + VEC2 mpos = { (F32)mx, (F32)my }; + + F32 hsize = editor->spritesize * .5f; + F32 mindist = editor->spritesize; + for( U32 i = 0; i < m->sprites.size; ++i ) { + MAP_SPRITE* s = &m->sprites[i]; + + VEC2 world = { s->pos.x, s->pos.y }; + VEC2 screen = gui_editor_2dview_world_to_screen( view, world ); + + if( mpos.x >= screen.x - hsize && mpos.x <= screen.x + hsize + && mpos.y >= screen.y - hsize && mpos.y <= screen.y + hsize ) + { + F32 dist = vec_dist( mpos, screen ); + if( dist < mindist ) { + mindist = dist; + gui_editor_2dview_select( view, s, EDITOR_SELECT_SPRITE ); + } + } + } + + if( !view->curselect && !view->held ) + return; + + gui_editor_2dview_input_select_drag( view ); +} + +void gui_editor_2dview_input_select_origin( GUI_EDITOR_2DVIEW* view ) { + WORLD_MAP* m = editor->map; + + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + VEC2 mpos = { (F32)mx, (F32)my }; + VEC2 world = { m->startpos.x, m->startpos.y }; + VEC2 screen = gui_editor_2dview_world_to_screen( view, world ); + + I32 w, h; + gui_draw_get_str_bounds( &w, &h, FNT_JPN12, "(origin)" ); + + if( fabsf( (F32)mx - screen.x ) < (w * .5f + 2.f) + && fabsf( (F32)my - h * .5f - screen.y ) < (h * .5f + 2.f) ) { + gui_editor_2dview_select( view, m, EDITOR_SELECT_ORIGIN ); + } + + if( !view->curselect && !view->held ) + return; + + gui_editor_2dview_input_select_drag( view ); +} + +void gui_editor_2dview_input_tool_select( GUI_EDITOR_2DVIEW* view ) { + U8 m1 = gui_mbutton_down( 0 ); + U8 m2 = !m1 && gui_mbutton_down( 1 ); + + if( gui_editor_2dview_input_drag( view, m2 ) ) { + return; + } + + view->curselect = 0; + view->seltype = EDITOR_SELECT_NONE; + gui_editor_2dview_input_select_origin( view ); + if( view->curselect ) return; + gui_editor_2dview_input_select_sprites( view ); + if( view->curselect ) return; + gui_editor_2dview_input_select_walls( view ); + if( view->curselect ) return; + gui_editor_2dview_input_select_polygons( view ); +} + +void gui_editor_2dview_input_scroll( GUI_EDITOR_2DVIEW* view ) { + U8 scroll = gui_mbutton_down( GUI_MBTNSCROLL ); + gui_capture_scroll(); + + if( scroll == (U8)-1 && view->scale > 0.5f ) { + view->scale *= 0.75f; + } + else if( scroll == 1 && view->scale < 16.f ) { + view->scale *= 1.25f; + } +} + +void gui_editor_2dview_input_tool_wall( GUI_EDITOR_2DVIEW* view ) { + U8 m1 = gui_mbutton_down( 0 ); + if( !m1 ) { + view->held = 0; + return; + } + + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + VEC2 world = gui_editor_2dview_screen_to_world( view, mx, my ); + if( !view->held ) { + MAP_WALL neww; + neww.start = neww.end = { + m_snap_to_grid( world.x, editor->grid ), + m_snap_to_grid( world.y, editor->grid ), + 0 + }; + neww.end.z = 80.f; + neww.propid = 0; + + I32 idx = editor->map->walls.size; + editor->map->walls.push( neww ); + view->curdrag = &editor->map->walls[idx].end; + view->held = 1; + + view->oldmx = mx; + view->oldmy = my; + view->mremainx = 0.0; + view->mremainy = 0.0; + return; + } + + gui_editor_2dview_input_select_drag_vertex( view ); +} + +void gui_editor_2dview_input_tool_poly( GUI_EDITOR_2DVIEW* view ) { + U8 m1 = gui_mbutton_down( 0 ); + if( !m1 ) { + view->held = 0; + return; + } + + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + VEC2 world = gui_editor_2dview_screen_to_world( view, mx, my ); + if( !view->held ) { + MAP_POLYGON newp; + newp.propid = 0; + newp.vertices.push( { .pos = { world.x, world.y, 0.f }, .uv = { 1, 0 } } ); + newp.vertices.push( { .pos = { world.x, world.y, 0.f }, .uv = { 0, 0 } } ); + newp.vertices.push( { .pos = { world.x, world.y, 0.f }, .uv = { 0, 1 } } ); + newp.vertices.push( { .pos = { world.x, world.y, 0.f }, .uv = { 1, 1 } } ); + + map_polygon_calc_bounds( &newp ); + + I32 idx = editor->map->polygons.size; + editor->map->polygons.push( newp ); + view->curdrag = &editor->map->polygons[idx]; + view->held = 1; + + view->oldmx = mx; + view->oldmy = my; + view->mremainx = 0.0; + view->mremainy = 0.0; + return; + } + + MAP_POLYGON* p = (MAP_POLYGON*)view->curdrag; + MAP_VERTEX + *tl = &p->vertices[1], + *tr = &p->vertices[0], + *bl = &p->vertices[2], + *br = &p->vertices[3]; + + VEC3 start = tl->pos; + VEC3 end = br->pos; + + for( U32 i = 0; i < 3; ++i ) { + start[i] = m_snap_to_grid( start[i], editor->grid ); + end[i] = m_snap_to_grid( end[i], editor->grid ); + } + + VEC2 move = gui_editor_2dview_input_get_drag_vec( view ); + if( !is_zero( move ) ) { + end += move; + tl->pos = { start.x, start.y, tl->pos.z }; + tr->pos = { end.x, start.y, tr->pos.z }; + bl->pos = { start.x, end.y, bl->pos.z }; + br->pos = { end.x, end.y, br->pos.z }; + + map_polygon_calc_bounds( p ); + gui_editor_2dview_input_select_onmove( view ); + } +} + +void gui_editor_2dview_input_tool_ent( GUI_EDITOR_2DVIEW* view ) { + +} + +void gui_editor_2dview_input_tool_sprite( GUI_EDITOR_2DVIEW* view ) { + U8 m1 = gui_mbutton_down( 0 ); + if( !m1 ) { + view->held = 0; + return; + } + + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + VEC2 world = gui_editor_2dview_screen_to_world( view, mx, my ); + if( !view->held ) { + MAP_SPRITE news; + news.pos = { world.x, world.y, 20.f }; + news.size = { 20.f, 20.f }; + news.clr = { 1.f, 1.f, 1.f, 1.f }; + news.tex = 0; + + I32 idx = editor->map->walls.size; + editor->map->sprites.push( news ); + view->curdrag = &editor->map->sprites[idx]; + view->held = 1; + + view->oldmx = mx; + view->oldmy = my; + view->mremainx = 0.0; + view->mremainy = 0.0; + return; + } + + gui_editor_2dview_input_select_drag_sprite( view ); +} + + +void gui_editor_2dview_input_tool_draw( GUI_EDITOR_2DVIEW* view ) { + U8 m1 = gui_mbutton_down( 0 ); + U8 m2 = !m1 && gui_mbutton_down( 1 ); + + if( gui_editor_2dview_input_drag( view, m2 ) ) { + return; + } + + switch( editor->tool ) { + case EDITOR_TOOL_WALL: return gui_editor_2dview_input_tool_wall( view ); + case EDITOR_TOOL_POLY: return gui_editor_2dview_input_tool_poly( view ); + case EDITOR_TOOL_ENT: return gui_editor_2dview_input_tool_ent( view ); + case EDITOR_TOOL_SPRITE: return gui_editor_2dview_input_tool_sprite( view ); + default: return; + } +} + +void gui_editor_2dview_input_fn( void* ptr ) { + GUI_EDITOR_2DVIEW* view = (GUI_EDITOR_2DVIEW*)ptr; + if( !editor->map ) + return; + + I32 x = gui_relx( view ); + I32 y = gui_rely( view ) + EDITORVIEW_TITLE_OFFSET; + I32 w = view->w; + I32 h = view->h; + + U8 mouse = gui_mbutton_down( 0 ) || gui_mbutton_down( 1 ); + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + U8 inbounds = mx >= x && mx <= x + w && my >= y && my <= y + h; + if( inbounds ) + gui_editor_2dview_input_scroll( view ); + + if( !mouse ) + view->heldoutbounds = 0; + else if( mouse && !inbounds ) + view->heldoutbounds = 1; + + if( view->heldoutbounds ) + return; + + switch( editor->tool ) { + case EDITOR_TOOL_SELECT: return gui_editor_2dview_input_tool_select( view ); + case EDITOR_TOOL_WALL: + case EDITOR_TOOL_POLY: + case EDITOR_TOOL_SPRITE: + case EDITOR_TOOL_ENT: + return gui_editor_2dview_input_tool_draw( view ); + default: return gui_editor_2dview_input_tool_none( view ); + } +} + +GUI_EDITOR_2DVIEW* gui_editor_2dview( I32 x, I32 y, I32 w, I32 h ) { + GUI_EDITOR_2DVIEW* view = new GUI_EDITOR_2DVIEW; + view->x = x; + view->y = y; + view->xbound = view->w = w; + view->h = h; + view->ybound = h + EDITORVIEW_TITLE_OFFSET; + + view->draw_fn = gui_editor_2dview_draw_fn; + view->input_fn = gui_editor_2dview_input_fn; + strcpy( view->name, "EDITOR_2D_VIEW" ); + + view->scale = 1.f; + + GUI_BASE* parent = gui_get_view(); + if( !parent ) + parent = gui_get_window(); + + parent->children.push( view ); + view->parent = parent; + + return view; +} diff --git a/src/editor/view3d.cpp b/src/editor/view3d.cpp new file mode 100644 index 0000000..f9bd60e --- /dev/null +++ b/src/editor/view3d.cpp @@ -0,0 +1,88 @@ +#include "editor.h" + +#include "../game/objlist.h" +#include "../game/world/draw.h" +#include "../game/world/bsp_draw.h" +#include "../game.h" + +const I32 EDITORVIEW_TITLE_OFFSET = 15; + +void gui_editor_3dview_draw_showpos( GUI_EDITOR_3DVIEW* view ) { + I32 x = gui_relx( view ); + I32 y = gui_rely( view ) + EDITORVIEW_TITLE_OFFSET; + + VEC3 pos = objl->pl->pos; + + gui_draw_str( + x + view->w - 2, y + 2, + ALIGN_R, + FNT_JPN12, + ui_clr.txt, + "pos: %.02f %.02f %.02f", + pos.x, pos.y, pos.z + ); + + gui_draw_str( + x + view->w - 2, y + 18, + ALIGN_R, + FNT_JPN12, + ui_clr.txt, + "rot: %.02f %.02f", + objl->pl->rot.y, + objl->pl->rot.x + ); +} + +void gui_editor_3dview_draw_fn( void* ptr ) { + GUI_EDITOR_3DVIEW* view = (GUI_EDITOR_3DVIEW*)ptr; + + if( !objl->pl ) + return; + + I32 x = gui_relx( view ); + I32 y = gui_rely( view ); + + gui_draw_str( x, y, ALIGN_L, FNT_JPN12, ui_clr.txt, "3d view" ); + y += EDITORVIEW_TITLE_OFFSET; + + VEC2 wnd = { (F32)x, (F32)y }; + VEC2 winsize = { (F32)view->w, (F32)view->h }; + + CLR col = gui_is_fg_window( view )? ui_clr.border : ui_clr.border_inactive; + gui_draw_frect( x-1, y-1, view->w+2, view->h+2, col ); + gui_draw_frect( x, y, view->w, view->h, CLR::BLACK() ); + if( editor->drawbsp && editor->map->bsp ) + bsp_draw( editor->map->bsp, editor->game, wnd, winsize ); + else + world_draw( editor->game, objl->world, wnd, winsize ); + + gui_editor_3dview_draw_showpos( view ); +} + +void gui_editor_3dview_input_fn( void* ptr ) { + if( !objl->pl ) + return; + + game_on_tick( editor->game ); +} + +GUI_EDITOR_3DVIEW* gui_editor_3dview( I32 x, I32 y, I32 w, I32 h ) { + GUI_EDITOR_3DVIEW* view = new GUI_EDITOR_3DVIEW; + view->x = x; + view->y = y; + view->xbound = view->w = w; + view->h = h; + view->ybound = h + EDITORVIEW_TITLE_OFFSET; + + view->draw_fn = gui_editor_3dview_draw_fn; + view->input_fn = gui_editor_3dview_input_fn; + strcpy( view->name, "EDITOR_3D_VIEW" ); + + GUI_BASE* parent = gui_get_view(); + if( !parent ) + parent = gui_get_window(); + + parent->children.push( view ); + view->parent = parent; + return view; +} diff --git a/src/game.cpp b/src/game.cpp new file mode 100644 index 0000000..bd3c4c8 --- /dev/null +++ b/src/game.cpp @@ -0,0 +1,149 @@ +#include <unistd.h> +#include "game.h" + +#include "game/world/draw.h" +#include "render/gl_2d.h" +#include "render/gl_3d.h" +#include "render/gl_2d_font.h" + +#include "editor/editor.h" +#include "game/assets.h" +#include "game/vars.h" +#include "game/objlist.h" +#include "game/world/map.h" + +#include "gui.h" +#include "render/gl_batch.h" +#include "util.h" + + +GAME_DATA* game_init( GL_DATA* gl ) { + GAME_DATA* game = (GAME_DATA*)malloc( sizeof( GAME_DATA ) ); + memset( game, 0, sizeof(GAME_DATA) ); + game->gl = gl; + + VEC2 screensize = { (F32)gl->canvas_size[0], (F32)gl->canvas_size[1] }; + + game->shaders.gl2d = gl_2d_init( gl, screensize, "2d" ); + game->shaders.gl2d_texcoord = gl_2d_init( gl, screensize, "2d_texcoord" ); + game->shaders.gl3d = gl_3d_init( gl, screensize, "3d" ); + + game_create_batches( game ); + + assets_init( game ); + assets_on_frame( game ); + varl = vars_init(); + objl = objl_init(); + +#if IS_EDITOR + game->editor = editor_create( game ); +#else + gui_create( game ); + game->state.ingame = 1; +#endif + + return game; +} + +void game_create_batches( GAME_DATA *game ) { + game->render.batch_3d = gl_batch_create( game->gl, game->shaders.gl3d, &gl_3d_batch_setup ); +} + +#ifdef DEBUG +void game_draw_fpsoverlay( GAME_DATA* game ) { + GL_DATA* gl = game->gl; + char buf[256]; + sprintf( buf, "fps: %1.3f (%2.5f ms) %dx%d", gl->fps, gl->frametime, canvas[0], canvas[1] ); + + VEC2 dim = gl_font_dim( game->assets.font, buf ); + gl_font_draw( + game->assets.font, + game->shaders.gl2d_texcoord, + s_tr({ -dim.x - 30.f, dim.y + 5.f }), + buf, + (CLR){ 1.f, 1.f, 1.f, 1.f } + ); +} + +void game_draw_fontoverlay( GAME_DATA* game ) { + for( U32 i = 0; i < 255; ++i ) { + char str[64]; + sprintf( str, "%d:%c", i, (char)i ); + gl_font_draw( + game->assets.jpn12, + game->shaders.gl2d_texcoord, + s_tl({ 10.f + ( i % 18 ) * 33, + 280.f + floorf( i / 18.f ) * 14.f }), + str, + CLR::WHITE() + ); + } +} +#endif + +WORLD_MAP* game_load_map( GAME_DATA* game, const char* mapname ) { + WORLD_MAP* m = map_from_file( game, mapname ); + if( !m ) { + dlog( "game_load_map() : failed to load file %s", mapname ); + return 0; + } + + objl_load_world( game, m ); + game->state.map = m; + return m; +} + +void game_unload_map( GAME_DATA* game ) { + map_free( game, game->state.map ); + game->state.map = 0; + + objl_clear( 1 ); +} + +void game_draw( GAME_DATA* game ) { + if( !objl->pl || !objl->world ) { + game_load_map( game, "../assets/maps/test.hmap" ); + } + + VEC2 window = { 270.f, 25.f }; + VEC2 winsize = { 480.f, 360.f }; + gl_2d_rect( game->shaders.gl2d, window, winsize, (CLR){ 0.3f, 0.3f, 0.3f, 1.f } ); + world_draw( game, objl->world, window, winsize ); +} + +void game_main_loop( GAME_DATA* game ) { + GL_DATA* gl = game->gl; + GL_SHADER_PROGRAM* gl2d = game->shaders.gl2d; + + if( !assets_on_frame( game ) ) + return; + + gl_beginframe( gl ); + gl_2d_frect( gl2d, s_tl(), s_br(), { 0.f, 0.f, 0.f, 1.f } ); + + if( game->state.ingame ) { + game_on_tick( game ); + game_draw( game ); + } + + gui_onframe( game ); + +#ifdef DEBUG + game_draw_fpsoverlay( game ); +#endif + + if( !OK( gl_endframe( gl ) ) ) + exit( 0 ); +} + +void game_on_tick( GAME_DATA* game ) { + U64 tick = u_tick(); + if( tick - game->state.last_tick > (U64)(TICK_INTERVAL * 10000) ) { + player_input( game, objl->pl ); + game->state.last_tick = tick; + } +} + +void game_destroy( GAME_DATA *game ) { + free( game ); +} diff --git a/src/game.h b/src/game.h new file mode 100644 index 0000000..b0307a1 --- /dev/null +++ b/src/game.h @@ -0,0 +1,60 @@ +#pragma once +#define IS_EDITOR 1 + +#include "game/assets.h" +#include "render/gl.h" +#include "render/gl_3d.h" + +#ifdef IS_EDITOR +#include "editor/editor.h" +#else +#include "gui.h" +#endif + +// HARDCODE BOSS +const U32 TICKS_PER_SEC = 64; +const F32 TICK_INTERVAL = 1.0f / (F32)TICKS_PER_SEC; + +typedef struct { + GL_SHADER_PROGRAM* gl2d; + GL_SHADER_PROGRAM* gl2d_texcoord; + + struct GL_3D* gl3d; +} GAME_SHADERS; + +typedef struct { + GL_BATCH3D* batch_3d; +} GAME_RENDER; + +typedef struct { + U8 ingame; + WORLD_MAP* map; + U64 last_tick; +} GAME_STATE; + +struct GAME_DATA { + GL_DATA* gl; + GAME_SHADERS shaders; + GAME_ASSETS assets; + + GAME_STATE state; + GAME_RENDER render; + +#ifdef IS_EDITOR + GAME_EDITOR* editor; +#endif +}; + +extern void ui_draw_background( GAME_DATA* game, VEC2 pos, VEC2 dim, F32 progress, CLR col ); + +extern GAME_DATA* game_init( GL_DATA* gl ); +extern void game_create_batches( GAME_DATA* game ); +extern void game_main_loop( GAME_DATA* game ); +extern void game_on_tick( GAME_DATA* game ); +extern void game_destroy( GAME_DATA* game ); + +// loads map and populates entities +extern WORLD_MAP* game_load_map( GAME_DATA* game, const char* mapname ); + +// unloads map and clears entity list including local player (no more level transitions can be made) +extern void game_unload_map( GAME_DATA* game ); diff --git a/src/game/assets.cpp b/src/game/assets.cpp new file mode 100644 index 0000000..7d755c0 --- /dev/null +++ b/src/game/assets.cpp @@ -0,0 +1,78 @@ +#include "assets.h" + +#include "../game.h" +#include "../render/gl_2d_font.h" + +void assets_init( GAME_DATA *game ) { + game->assets.test = gl_texture_create( game->gl, "assets/test.png" ); + game->assets.fonts_init = 0; +} + +void assets_create_fonts( struct GAME_DATA *game ) { + dlog( "creating fonts\n" ); + + game->gl->fonts.each( fn( GL_FONT** font ) { + gl_font_destroy( game->gl, *font ); + }); + + game->gl->fonts.clear(); + GAME_ASSETS* a = &game->assets; + a->font = gl_font_create( game->gl, "cour.ttf", 16 ); + a->jpn12 = gl_font_create( game->gl, "jpn12.ttf", 12 ); + a->jpn16 = gl_font_create( game->gl, "jpn16.ttf", 16 ); +} + +U8 assets_on_frame( struct GAME_DATA *game ) { + thread_mutex_lock( &font_mutex ); + if( !game->assets.fonts_init ) { + assets_create_fonts( game ); + game->assets.fonts_init = 1; + thread_mutex_unlock( &font_mutex ); + return 0; + } + + thread_mutex_unlock( &font_mutex ); + return 1; +} + +LIST<FILE_ENTRY> assets_get_files_by_ext_dir( const char** extensions, const char* dirname ) { + LIST<FILE_ENTRY> ret; + LIST<FILE_ENTRY> dir = dir_get_entries( dirname ); + + dir.each( fn( FILE_ENTRY* e ) { + char fullpath[256]; + sprintf( fullpath, "%s/%s", dirname, e->name ); + if( e->dir ) { + LIST<FILE_ENTRY> subdir = assets_get_files_by_ext_dir( extensions, fullpath ); + subdir.each( fn( FILE_ENTRY* se ) { ret.push( *se ); } ); + return; + } + + U32 len = strlen( e->name ); + for( U32 i = 0; !!extensions[i]; ++i ) { + const char* ext = extensions[i]; + U32 extlen = strlen( ext ); + + if( extlen < len && e->name[len - extlen - 1] == '.' ) { + if( !strcmp( e->name + len - extlen, ext ) ) { + FILE_ENTRY ne{ .dir = 0 }; + strcpy( ne.name, fullpath ); + ret.push( ne ); + } + } + } + } ); + + return ret; +} + +LIST<FILE_ENTRY> assets_get_files_by_ext( const char** extensions ) { + return assets_get_files_by_ext_dir( extensions, "../assets" ); +} + +const char* assets_abspath( const char* filename ) { + while( !strncmp( filename, "../assets/", 10 ) ) + filename += 10; + + return filename; +} diff --git a/src/game/assets.h b/src/game/assets.h new file mode 100644 index 0000000..d0fe270 --- /dev/null +++ b/src/game/assets.h @@ -0,0 +1,23 @@ +#pragma once +#include "../util/typedef.h" +#include "../util/file.h" + + +typedef struct { + struct GL_FONT* font; + struct GL_FONT* jpn12; + struct GL_FONT* jpn16; + struct GL_TEX2D* test; + + U8 fonts_init; +} GAME_ASSETS; + +extern void assets_init( struct GAME_DATA* game ); +extern void assets_create_fonts( struct GAME_DATA* game ); +// extensions is a null-terminated array of null-terminated strings. +extern LIST<FILE_ENTRY> assets_get_files_by_ext_dir( const char** extensions, const char* dir ); +// extensions is a null-terminated array of null-terminated strings. +extern LIST<FILE_ENTRY> assets_get_files_by_ext( const char** extensions ); +extern const char* assets_abspath( const char* filepath ); + +extern U8 assets_on_frame( struct GAME_DATA* game ); diff --git a/src/game/object.cpp b/src/game/object.cpp new file mode 100644 index 0000000..e69de29 --- /dev/null +++ b/src/game/object.cpp diff --git a/src/game/object.h b/src/game/object.h new file mode 100644 index 0000000..8191ab7 --- /dev/null +++ b/src/game/object.h @@ -0,0 +1,73 @@ +#pragma once +#include "../util/allocator.h" +#include "../util/vector.h" +#include "../util/fnv.h" + +typedef void ( *OBJECT_DEINIT_FN )( struct OBJECT* obj ); +inline const char* obj_classid_to_name( U32 classid ); + +enum ObjectClass_t { + OBJCLASS_NONE = 0, + OBJCLASS_WORLD, + OBJCLASS_TRIGGER, + OBJCLASS_PLAYER, + OBJCLASS_BASENPC, +}; + +static const char* OBJCLASS_NAMES[] = { + "none (invalid - FIXME)", + "world", + "trigger", + "player", + "base_npc", +}; + +// todo: come up with a good way to make these static +struct OBJECT_PROP { + OBJECT_PROP( const char* name, void* value, U32 size, LIST<OBJECT_PROP*>* list, void* parent ) { + this->name = name; + this->size = size; + this->hash = fnv1a( name ); + this->offset = (U32)( (U8*)value - (U8*)parent ); + list->push( this ); + } + + const char* name; + FNV1A hash; + U32 offset; + U32 size; +}; + +#define OBJVAR( type, name ) \ + type name; \ + OBJECT_PROP __##name##_props{ #name, (void*)&name, sizeof(type), &this->props, this } + +#define OBJVAR_STR( name, len ) \ + char name##[len]; \ + OBJECT_PROP __##name##_props{ #name, (void*)name, len, &this->props, this } + +struct OBJECT { + static const U32 CLASSID = OBJCLASS_NONE; + + U32 classid; + U8 keeponlevel{}; + char name[64]; + + OBJECT* parent; + LIST<OBJECT*> children{}; + LIST<OBJECT_PROP*> props{}; + OBJECT_DEINIT_FN deinit_fn{}; + + OBJVAR( VEC3, pos ); + OBJVAR( VEC3, rot ); + OBJVAR( I32, idx ); +}; + +inline const char* obj_classid_to_name( U32 classid ) { + if( classid > (sizeof(OBJCLASS_NAMES) / sizeof(OBJCLASS_NAMES[0])) ) { + dlog( "obj_classid_to_name() : invalid classid passed (%d)", classid ); + return "INVALIDCLASSID"; + } + + return OBJCLASS_NAMES[classid]; +} diff --git a/src/game/objlist.cpp b/src/game/objlist.cpp new file mode 100644 index 0000000..01ce49b --- /dev/null +++ b/src/game/objlist.cpp @@ -0,0 +1,44 @@ +#include "objlist.h" +#include "world/world.h" +#include "world/map.h" + +OBJECT_LIST* objl; + +OBJECT_LIST* objl_init() { + OBJECT_LIST* ret = new OBJECT_LIST; + ret->world = 0; + ret->pl = 0; + + return ret; +} + +void objl_clear( U8 delete_player ) { + objl->objects.each( fn( OBJECT** ptr ) { + OBJECT* obj = *ptr; + + if( !delete_player && obj == objl->pl ) + return; + + if( obj->deinit_fn ) + obj->deinit_fn( obj ); + delete obj; + } ); + + objl->objects.clear(); + objl->world = 0; + if( delete_player ) + objl->pl = 0; +} + +void objl_load_world( GAME_DATA* game, WORLD_MAP* map ) { + objl_clear(); + + objl->world = world_create( map ); + if( !objl->pl ) + objl->pl = player_create( map->startpos, map->startang ); + + if( !OK( world_populate_entities( objl->world ) ) ) { + objl_clear(); + dlog( "objl_load_world() : error populating entities\n" ); + } +} diff --git a/src/game/objlist.h b/src/game/objlist.h new file mode 100644 index 0000000..f8a46d7 --- /dev/null +++ b/src/game/objlist.h @@ -0,0 +1,122 @@ +#pragma once + +#include "player.h" +#include "world/world.h" + +const U32 OBJ_LIST_MAX = 8192; + +struct OBJECT_LIST { + template <typename T> + T* add( const char* name ) { + if( objects.size >= OBJ_LIST_MAX ) { + dlog( "OBJECT_LIST::add() : object list is full" ); + return 0; + } + + for( I32 i = 0; i < objects.size; ++i ) { + if( !strcmp( name, objects[i]->name ) ) { + dlog( "OBJECT_LIST::add() : object %s already exists", name ); + return 0; + } + } + + T* newt = new T; + newt->classid = T::CLASSID; + strcpy( newt->name, name ); + objects.push( newt ); + return newt; + } + + template < typename T > + T* add( const T& obj ) { + for( I32 i = 0; i < objects.size; ++i ) { + if( !strcmp( obj.name, objects[i]->name ) ) { + dlog( "OBJECT_LIST::add() : object %s already exists", obj.name ); + return 0; + } + } + + T* newt = new T( obj ); + newt->classid = T::CLASSID; + objects.push( newt ); + return newt; + } + + void remove( OBJECT* obj ) { + I32 idx = objects.idx_of( obj ); + if( idx != -1 ) + free( objects[idx] ); + } + + void remove( const char* obj ) { + for( I32 i = 0; i < objects.size; ++i ) { + if( !strcmp( obj, objects[i]->name ) ) { + free( objects[i] ); + return; + } + } + } + + template < typename T = OBJECT > + T* find( const char* name ) { + for( I32 i = 0; i < objects.size; ++i ) { + if( !strcmp( name, objects[i]->name ) ) + return objects[i]; + } + + return 0; + } + + void each( LIST<OBJECT*>::ON_EACH_FN func ) { + objects.each( func ); + } + + WORLD* world; + PLAYER* pl; + LIST<OBJECT*> objects; +}; + +extern OBJECT_LIST* objl; + +extern OBJECT_LIST* objl_init(); +extern void objl_clear( U8 delete_player = 0 ); +extern void objl_load_world( GAME_DATA* game, WORLD_MAP* map ); + + +template < typename T > +inline T* obj_add( const char* name ) { + if( !objl ) { + dlog( "obj_add() : objl null\n" ); + return 0; + } + + return objl->add<T>( name ); +} + +template < typename T > +inline T* obj_add( T obj ) { + T* newt = new T( obj ); + newt = objl->add( newt ); + return newt; +} + +template < typename T > +inline T* obj_add( OBJECT* obj, const char* name ) { + T* newt = objl->add<T>( name ); + if( !newt ) + return 0; + + newt->parent = obj; + obj->children.push( newt ); + return newt; +} + +template < typename T > +inline T* obj_add( OBJECT* obj, T newobj ) { + T* newt = objl->add( newobj ); + if( !newt ) + return 0; + + newt->parent = obj; + obj->children.push( newt ); +} diff --git a/src/game/physics/movement.h b/src/game/physics/movement.h new file mode 100644 index 0000000..e69de29 --- /dev/null +++ b/src/game/physics/movement.h diff --git a/src/game/player.cpp b/src/game/player.cpp new file mode 100644 index 0000000..87786df --- /dev/null +++ b/src/game/player.cpp @@ -0,0 +1,82 @@ +#include "../game.h" +#include "SDL_scancode.h" +#include "objlist.h" +#include "world/trace.h" +#include "player.h" + +F32 PLAYER_HULL_HEIGHT = 50.f; +F32 PLAYER_HULL_WIDTH = 32.f; + +PLAYER* player_create( VEC3 origin, F32 yaw ) { + PLAYER* p = obj_add<PLAYER>( "localplayer" ); + p->pos = origin; + p->rot = { 0, yaw, 0 }; + p->health = 100; + p->keeponlevel = 1; + p->velocity = {}; + + p->mins = { + -PLAYER_HULL_WIDTH * .5f, + -PLAYER_HULL_WIDTH * .5f, + 0 + }; + + p->maxs = { + PLAYER_HULL_WIDTH * .5f, + PLAYER_HULL_WIDTH * .5f, + PLAYER_HULL_HEIGHT + }; + + return p; +} + +void player_input( GAME_DATA* game, PLAYER* p ) { + F32* yaw = &p->rot.y; + VEC3* pos = &p->pos; + + if( input.keys[(U8)'a'] ) + *yaw -= 360.f * TICK_INTERVAL * 0.5f; + if( input.keys[(U8)'d'] ) + *yaw += 360.f * TICK_INTERVAL * 0.5f; + + if( input.keys[SDL_SCANCODE_UP] ) { + if( p->rot.x < 89.f ) + p->rot.x += 360.f * TICK_INTERVAL * 0.5f; + } + if( input.keys[SDL_SCANCODE_DOWN] ) { + if( p->rot.x > -89.f ) + p->rot.x -= 360.f * TICK_INTERVAL * 0.5f; + } + + *yaw = remainderf( *yaw, 360.f ); + + + VEC3 wishmove; + VEC3 dir = m_radial_offset( *yaw, 70.f ); + if( input.keys[(U8)'w'] ) { + wishmove = dir * TICK_INTERVAL; + } + if( input.keys[(U8)'s'] ) { + wishmove = dir * -TICK_INTERVAL; + } + + // todo : should never be false + if( vec_len( wishmove ) > BSP_TRACE_EPSILON && game->state.map->bsp ) { + BSP_TRACE tr{}; + AABB aabb{}; + tr.in_start = *pos; + tr.in_end = *pos + wishmove * 2.f; + aabb.min = p->mins; + aabb.max = p->maxs; + + bsp_trace( &tr, game->state.map->bsp, aabb ); + if( !tr.hit ) + *pos += wishmove; + else + *pos += wishmove * tr.frac; + } +} + +VEC3 player_get_view_pos( PLAYER* p ) { + return VEC3( p->pos.x, p->pos.y, p->pos.z + p->eyeoffset ); +} diff --git a/src/game/player.h b/src/game/player.h new file mode 100644 index 0000000..71b1c02 --- /dev/null +++ b/src/game/player.h @@ -0,0 +1,22 @@ +#pragma once +#include "object.h" + +struct PLAYER_INPUT { + +}; + +struct PLAYER : OBJECT { + static const U32 CLASSID = OBJCLASS_PLAYER; + U32 health; + F32 fov{72.f}; + F32 eyeoffset{40.f}; + + VEC3 mins; + VEC3 maxs; + + VEC3 velocity; +}; + +extern PLAYER* player_create( VEC3 origin, F32 yaw ); +extern void player_input( struct GAME_DATA* game, PLAYER* player ); +extern VEC3 player_get_view_pos( PLAYER* player ); diff --git a/src/game/raycast.cpp b/src/game/raycast.cpp new file mode 100644 index 0000000..c82a3ed --- /dev/null +++ b/src/game/raycast.cpp @@ -0,0 +1,162 @@ +#include <cmath> +#include <math.h> + +#include "raycast.h" +#include "../game.h" + +#include "../render/gl_2d.h" + +#include "objlist.h" +#include "player.h" +#include "world/world.h" +#include "world/map.h" + +const U32 RAY_DEPTH_MAX = 8; + +U8 line_intersects( VEC3 ray_start, VEC2 ray_dir, MAP_WALL* s, VEC3* p, F32* t ) { + VEC2 v1 = VEC2{ ray_start.x, ray_start.y } - VEC2{ s->start.x, s->start.y }; + VEC2 v2 = VEC2{ s->end.x, s->end.y } - VEC2{ s->start.x, s->start.y }; + VEC2 v3 = { -ray_dir.y, ray_dir.x }; + + F32 dot = vec_dot( v2, v3 ); + if( fabsf(dot) < 0.001f ) + return 0; + + F32 t1 = vec_cross( v2, v1 ) / dot; + F32 t2 = vec_dot( v1, v3 ) / dot; + + if( t1 >= 0.0f && ( t2 >= 0.0f && t2 <= 1.0f ) ) { + *t = t1; + *p = ray_start + VEC3{ ray_dir } * (*t); + return 1; + } + + return 0; +} + +F32 ray_calc_wall_hit_dir( VEC3 start, VEC3 end, F32 ang ) { + VEC3 walld = start - end; + vec_normalize( &walld ); + F32 wallang = atan2f( walld.y, walld.x ); + wallang += ang; + if( wallang < 0.f ) wallang += 2.f * M_PI; + if( wallang > 2.f * M_PI ) wallang -= 2.f * M_PI; + return wallang; +} + +LIST<RAY_HITDATA> ray_trace_list( VEC3 start, F32 ang, U32 max_iter ) { + WORLD* w = objl->world; + if( !w ) return {}; + LIST<RAY_HITDATA> hits; + + F32 rayang = m_deg2rad( ang ); + VEC2 raydir = { cos( rayang ), sin( rayang ) }; + VEC3 minp; + + for( U32 i = 0; i < w->map->walls.size; ++i ) { + VEC3 p; + F32 dist; + if( line_intersects( start, raydir, &w->map->walls[i], &p, &dist ) ) { + dist = vec_dist( start, p ); + + VEC3 start = w->map->walls[i].start; + VEC3 end = w->map->walls[i].end; + F32 hitang = ray_calc_wall_hit_dir( start, end, ang ); + hits.push( RAY_HITDATA{ p, dist, hitang, TH_WALL, i } ); + if( hits.size > max_iter ) + break; + } + } + + return hits; +} + +U32 ray_trace( TRACE* tr, F32 ang ) { + F32 rayang = m_deg2rad( ang ); + VEC3 start = tr->startpos; + VEC3 raydir = { cos( rayang ), sin( rayang ), 0.f }; + raydir *= RAY_DEPTH_MAX; + + return ray_trace( tr, raydir ); +} + +U32 ray_trace( TRACE* tr, VEC3 end ) { + WORLD* w = objl->world; + if( !w ) return TRACE_HIT_NONE; + WORLD_MAP* map = w->map; + + VEC3 start = tr->startpos; + VEC3 diff = end - start; + F32 trdist = vec_lensq( diff ); + vec_normalize( &diff ); + VEC2 raydir = { diff.x, diff.y }; + F32 ang = atan2f( raydir.y, raydir.x ); + + F32 mindist = INFINITY; + VEC3 minp; + U32 hitwall = TRACE_HIT_NONE; + + for( U32 i = 0; i < map->walls.size; ++i ) { + VEC3 p; + F32 dist; + + if( line_intersects( start, raydir, &map->walls[i], &p, &dist ) ) { + if( dist <= trdist && dist < mindist ) { + mindist = dist; + minp = p; + hitwall = i; + } + } + } + + if( hitwall != -1 ) { + tr->endpos = { minp.x, minp.y, 0.0f }; + tr->hitwall = hitwall; + + VEC3 start = map->walls[hitwall].start; + VEC3 end = map->walls[hitwall].end; + tr->hitang = ray_calc_wall_hit_dir( start, end, ang ); + return hitwall; + } + + tr->hitwall = -1; + tr->fract = -1.f; + tr->endpos = { start.x + RAY_DEPTH_MAX * raydir.x, start.y + RAY_DEPTH_MAX * raydir.y, 0.0f }; + return TRACE_HIT_NONE; +} + +void draw_player( GAME_DATA* game, PLAYER* pl ) { + GL_PROGRAM* gl2d = game->shaders.gl2d; + VEC2 xy = { pl->pos.x, pl->pos.y }; + + gl_2d_frect( gl2d, xy - VEC2{ 5.f, 5.f }, { 10.f, 10.f }, CLR::RED() ); +} + +void ray_draw_world2d( GAME_DATA* game, WORLD* world ) { + for( U32 i = 0; i < world->map->polygons.size; ++i ) { + MAP_POLYGON* p = &world->map->polygons[i]; + + LIST<VERTEX> verts{}; + SURF_PROPS* props = polygon_get_props( world->map, p ); + p->vertices.each( fn( MAP_VERTEX* mv ) { + VERTEX v{}; + v.pos = VEC2{ mv->pos.x, mv->pos.y }; + v.uv = mv->uv; + v.clr = props->clr; + verts.push( v ); + } ); + + if( props->tex ) + gl_textured_polygon( game->shaders.gl2d_texcoord, verts.data, verts.size, props->tex ); + else + gl_polygon( game->shaders.gl2d, verts.data, verts.size ); + } + + for( U32 i = 0; i < world->map->walls.size; ++i ) { + MAP_WALL* s = &world->map->walls[i]; + SURF_PROPS* p = &world->map->props[s->propid]; + gl_2d_line( game->shaders.gl2d, VEC2( s->start.x, s->start.y ), VEC2( s->end.x, s->end.y ), p->clr ); + } + + draw_player( game, objl->pl ); +} diff --git a/src/game/raycast.h b/src/game/raycast.h new file mode 100644 index 0000000..5920e09 --- /dev/null +++ b/src/game/raycast.h @@ -0,0 +1,65 @@ +#pragma once + +#include "../util.h" + +const static U32 RAY_ITER_MAX = 10; +const static U32 TRACE_HIT_NONE = (U32)-1; + +enum TraceHitType_t { + TH_NONE, + TH_WALL, + TH_CEILING, + TH_FLOOR +}; + +enum TraceFilter_t { + TF_NONE = 0x0, + TF_WALLS = 0x1, + TF_SPRITES = 0x2, + TF_ENTITIES = 0x4, + + TF_WORLD = TF_WALLS | TF_SPRITES, + TF_ALL = TF_WALLS | TF_SPRITES | TF_ENTITIES, +}; + +struct TRACE { + /* input */ + VEC3 startpos; + /* output*/ + VEC3 endpos; + /* output */ + F32 fract; + /* output */ + F32 hitang; + /* output */ + U32 hitwall; + /* input */ + U32 filter; +}; + +struct RAY_HITDATA { + VEC3 point; + F32 dist; + F32 ang; + + U8 type; + U32 obj_idx; +}; + +extern LIST<RAY_HITDATA> ray_trace_list( VEC3 start, F32 ang, U32 max_iter = RAY_ITER_MAX ); +extern LIST<RAY_HITDATA> ray_trace_list( VEC3 start, VEC3 end, U32 max_iter = RAY_ITER_MAX ); + +extern U32 ray_trace( TRACE* tr, F32 ang ); +extern U32 ray_trace( TRACE* tr, VEC3 end ); +extern U32 ray_trace( TRACE* tr, VEC3 start, F32 ang ); +extern U32 ray_trace( TRACE* tr, VEC3 start, VEC3 end ); + +extern VEC2 project_vertex( + VEC3 vertex_pos, + VEC3 player_pos, + F32 player_angle_deg, + F32 fov_deg, + VEC2 window, + VEC2 winsize, + U8* in_view = 0 +); diff --git a/src/game/vars.cpp b/src/game/vars.cpp new file mode 100644 index 0000000..d7b1f9b --- /dev/null +++ b/src/game/vars.cpp @@ -0,0 +1,115 @@ +#include "vars.h" +#include "../util.h" +#include <string.h> + +VAR_LIST* varl; + +VAR_LIST* vars_init() { + VAR_LIST* ret = new VAR_LIST; + return ret; +} + +CVAR* var_new( const char* name, F32 v ) { + if( !varl ) { + dlog( "var_new() : error: varl null\n" ); + return 0; + } + + CVAR* ret = new CVAR; + memset( ret, 0 , sizeof( CVAR ) ); + ret->type = CVAR_TYPE_FLOAT; + strcpy( ret->name, name ); + ret->fl_v = v; + + varl->vars.push( ret ); + return ret; +} + +CVAR* var_new( const char* name, I32 i ) { + if( !varl ) { + dlog( "var_new() : error: varl null\n" ); + return 0; + } + + CVAR* ret = new CVAR; + memset( ret, 0 , sizeof( CVAR ) ); + ret->type = CVAR_TYPE_INT; + strcpy( ret->name, name ); + ret->i_v = i; + + varl->vars.push( ret ); + return ret; +} + +CVAR* var_new( const char* name, CLR c ) { + if( !varl ) { + dlog( "var_new() : error: varl null\n" ); + return 0; + } + + CVAR* ret = new CVAR; + memset( ret, 0 , sizeof( CVAR ) ); + ret->type = CVAR_TYPE_COLOR; + strcpy( ret->name, name ); + ret->clr_v = c; + + varl->vars.push( ret ); + return ret; +} + +CVAR* var_new( const char* name, CVAR_FUNC func ) { + if( !varl ) { + dlog( "var_new() : error: varl null\n" ); + return 0; + } + + CVAR* ret = new CVAR; + memset( ret, 0 , sizeof( CVAR ) ); + ret->type = CVAR_TYPE_FN; + strcpy( ret->name, name ); + ret->fn_v = func; + + varl->vars.push( ret ); + return ret; +} + +CVAR* var_new( const char* name, const char* v ) { + if( !varl ) { + dlog( "var_new() : error: varl null\n" ); + return 0; + } + + CVAR* ret = new CVAR; + memset( ret, 0, sizeof( CVAR ) ); + ret->type = CVAR_TYPE_STRING; + strcpy( ret->name, name ); + + varl->vars.push( ret ); + return ret; +} + +CVAR* var_find( const char* name ) { + if( !varl ) { + dlog( "var_new() : error: varl null\n" ); + return 0; + } + + U32 len = strlen( name ); + for( U32 i = 0; i < varl->vars.size; ++i ) { + U32 len2 = strlen( varl->vars[i]->name ); + if( strncmp( varl->vars[i]->name, name, min( len, len2 ) ) == 0 ) { + return varl->vars[i]; + } + } + + return 0; +} + +I32 var_call( CVAR *cmdvar, const char *cmdline ) { + if( cmdvar->type != CVAR_TYPE_FN ) { + dlog( "var_call() : error: not a function\n" ); + return -1; + } + + return cmdvar->fn_v( cmdvar, cmdline ); +} diff --git a/src/game/vars.h b/src/game/vars.h new file mode 100644 index 0000000..395c470 --- /dev/null +++ b/src/game/vars.h @@ -0,0 +1,50 @@ +#pragma once +#include "../util.h" + +enum CvarTypes_t { + CVAR_TYPE_INT, + CVAR_TYPE_FLOAT, + CVAR_TYPE_COLOR, + CVAR_TYPE_STRING, + CVAR_TYPE_FN, + CVAR_TYPE_LAST +}; + +typedef I32(*CVAR_FUNC)( struct CVAR* self, const char* cmdline ); + +struct CVAR { + char name[100]; + U8 type; + union { + F32 fl_v; + I32 i_v; + CLR clr_v; + CVAR_FUNC fn_v; + char str_v[64]; + }; + + union { + F32 fl_v; + I32 i_v; + CLR clr_v; + CVAR_FUNC fn_v; + char str_v[64]; + } default_v; +}; + +struct VAR_LIST { + LIST<CVAR*> vars; +}; + +extern VAR_LIST* vars_init(); + +extern CVAR* var_new( const char* name, F32 v ); +extern CVAR* var_new( const char* name, I32 v ); +extern CVAR* var_new( const char* name, CLR v ); +extern CVAR* var_new( const char* name, const char* v ); +extern CVAR* var_new( const char* name, CVAR_FUNC fn ); + +extern CVAR* var_find( const char* name ); +extern I32 var_call( CVAR* cmdvar, const char* cmdline ); + +extern VAR_LIST* varl; diff --git a/src/game/world/bsp.cpp b/src/game/world/bsp.cpp new file mode 100644 index 0000000..9aa94d7 --- /dev/null +++ b/src/game/world/bsp.cpp @@ -0,0 +1,662 @@ +#include "bsp.h" +#include "../../util/math.h" +#include "map.h" +#include <climits> + +const F32 BSP_PORTAL_PLANE_SIZE = 64000.f; + +struct MAP_TRI { + MAP_VERTEX p0, p1, p2; +}; + +void bsp_face_gen_render_verts( BSP* bsp, BSP_FACE* f ) { + LIST<VERTEX3D> vertices; + vertices.reserve( f->verts.size ); + + SURF_PROPS* props = map_get_props( bsp->map, f->propid ); + + f->verts.each( fn( MAP_VERTEX* v ) { + VERTEX3D v3d; + v3d.pos = { v->pos.x, v->pos.z, v->pos.y }; + v3d.uv = v->uv; + v3d.clr = v->clr * props->clr; + v3d.sampler = 0; + vertices.push( v3d ); + } ); + + f->render_verts = triangle_fan_to_list( vertices.data, vertices.size ); +} + +void bsp_gen_render_vertices( BSP* bsp ) { + bsp->faces.each( fn( BSP_FACE* f ) { + bsp_face_gen_render_verts( bsp, f ); + } ); +} + +LIST<BSP_PLANE> bsp_map_get_wall_planes( WORLD_MAP* map ) { + LIST<BSP_PLANE> ret{}; + + auto calc_wall_plane = fn( MAP_WALL* w ) { + VEC3 dir = w->end - w->start; + VEC3 normal = vec_normalize( { -dir.y, dir.x, 0.0f } ); + F32 dist = vec_dot( normal, w->start ); + + BSP_PLANE p = { normal, dist }; + for( U32 i = 0; i < ret.size; ++i ) { + if( ret[i] == p ) + return; + } + + ret.push( p ); + }; + + map->walls.each( calc_wall_plane ); + for( U32 i = 0; i < 4; ++i ) + calc_wall_plane( &map->skybox.walls[i] ); + + return ret; +} + + +LIST<MAP_TRI> bsp_map_triangulate_planes( WORLD_MAP* map ) { + LIST<MAP_TRI> ret; + + auto triangulate_plane = fn( MAP_POLYGON* p ) { + MAP_VERTEX* start = &p->vertices[0]; + MAP_VERTEX* last = &p->vertices[1]; + for( U32 i = 2; i < p->vertices.size; ++i ) { + MAP_TRI tri = MAP_TRI{ + *start, + *last, + p->vertices[i] + }; + + last = &p->vertices[i]; + ret.push( tri ); + } + }; + + map->polygons.each( triangulate_plane ); + triangulate_plane( &map->skybox.polygons[0] ); + triangulate_plane( &map->skybox.polygons[1] ); + + return ret; +} + +LIST<BSP_PLANE> bsp_map_get_tri_planes( WORLD_MAP* map ) { + LIST<BSP_PLANE> ret; + LIST<MAP_TRI> tris = bsp_map_triangulate_planes( map ); + + tris.each( fn( MAP_TRI* tri ) { + VEC3 cross = vec_cross( tri->p1.pos - tri->p0.pos, tri->p2.pos - tri->p0.pos ); + VEC3 normal = vec_normalize( cross ); + F32 dist = vec_dot( normal, tri->p0.pos ); + + BSP_PLANE p = { normal, dist }; + for( U32 i = 0; i < ret.size; ++i ) { + if( ret[i] == p ) + return; + } + + ret.push( p ); + } ); + + return ret; +} + +LIST<BSP_PLANE> bsp_map_get_planes( WORLD_MAP* map ) { + LIST<BSP_PLANE> ret; + + LIST<BSP_PLANE> walls = bsp_map_get_wall_planes( map ); + LIST<BSP_PLANE> floors = bsp_map_get_tri_planes( map ); + + ret.push_list( walls ); + ret.push_list( floors ); + + // probably unnecessary, slow as fuck + // remove any dupes in case a polygon ends up the same as a wall + // unlikely but dont wanna run into the edge case + // this wont run that much anyway + // maybe remove later + for( U32 i = 0; i < ret.size; ++i ) { + BSP_PLANE* p = &ret[i]; + for( U32 i2 = 0; i2 < ret.size; ++i2 ) { + if( i == i2 ) + continue; + + if( ret.data[i2] == *p ) { + ret.erase( i ); + --i; + } + } + } + + return ret; +} + + +LIST<BSP_FACE> bsp_map_faces_from_walls( WORLD_MAP* map ) { + LIST<BSP_FACE> ret{}; + + auto triangulate_wall = fn( MAP_WALL* w ) { + SURF_PROPS* p = wall_get_props( map, w ); + VEC3 points[] = { + { w->start.x, w->start.y, w->start.z }, + { w->end.x, w->end.y, w->start.z }, + { w->end.x, w->end.y, w->start.z + w->end.z }, + { w->start.x, w->start.y, w->start.z + w->end.z }, + }; + + VEC2 uvs[] = { + { w->uvstart.x, 1.f - w->uvend.y }, + { 1.f - w->uvend.x, 1.f - w->uvend.y }, + { 1.f - w->uvend.x, w->uvstart.y }, + { w->uvstart.x, w->uvstart.y }, + }; + + MAP_VERTEX vertices[4]; + for( U32 i = 0; i < 4; ++i ) { + vertices[i] = { + .pos = points[i], + .uv = uvs[i], + .clr = p->clr, + }; + } + + BSP_FACE f1{ .propid = w->propid }; + f1.verts.push( vertices[2] ); + f1.verts.push( vertices[1] ); + f1.verts.push( vertices[0] ); + + BSP_FACE f2{ .propid = w->propid }; + f2.verts.push( vertices[3] ); + f2.verts.push( vertices[2] ); + f2.verts.push( vertices[0] ); + + ret.push( f1 ); + ret.push( f2 ); + }; + + map->walls.each( triangulate_wall ); + for( U32 i = 0; i < 4; ++i ) triangulate_wall( &map->skybox.walls[i] ); + + return ret; +} + +LIST<BSP_FACE> bsp_map_faces_from_planes( WORLD_MAP* map ) { + LIST<BSP_FACE> ret; + + auto triangulate_plane = fn( MAP_POLYGON* p ) { + MAP_VERTEX* start = &p->vertices[0]; + MAP_VERTEX* last = &p->vertices[1]; + for( U32 i = 2; i < p->vertices.size; ++i ) { + BSP_FACE f = { .propid = p->propid }; + f.verts.push( *start ); + f.verts.push( *last ); + f.verts.push( p->vertices[i] ); + + ret.push( f ); + last = &p->vertices[i]; + } + }; + + map->polygons.each( triangulate_plane ); + triangulate_plane( &map->skybox.polygons[0] ); + triangulate_plane( &map->skybox.polygons[1] ); + + return ret; +} + +LIST<BSP_FACE> bsp_map_get_faces( WORLD_MAP* map ) { + LIST<BSP_FACE> ret; + + LIST<BSP_FACE> walls = bsp_map_faces_from_walls( map ); + LIST<BSP_FACE> floors = bsp_map_faces_from_planes( map ); + + ret.emplace_list( walls ); + ret.emplace_list( floors ); + + return ret; +} + +inline I32 point_classify( const BSP_PLANE& plane, const VEC3& point ) { + F32 side = bsp_plane_dist( plane, point ); + if( side > BSP_NORM_EPSILON ) + return BSP_SIDE_FRONT; + else if( side < -BSP_NORM_EPSILON ) + return BSP_SIDE_BACK; + else + return BSP_SIDE_ON; +} + +inline I32 polygon_classify( + const BSP_PLANE& plane, + const BSP_FACE* face, + I32* out_fr = 0, + I32* out_bk = 0 +) { + I32 cf = 0, cb = 0; + for( U32 i = 0; i < face->verts.size; ++i ) { + I32 side = point_classify( plane, face->verts.data[i].pos ); + if( side & BSP_SIDE_FRONT ) + ++cf; + else if( side & BSP_SIDE_BACK ) + ++cb; + } + + if( out_fr ) + *out_fr = cf; + if( out_bk ) + *out_bk = cb; + + if( cf && cb ) return BSP_SIDE_SPAN; + if( cf ) return BSP_SIDE_FRONT; + if( cb ) return BSP_SIDE_BACK; + return BSP_SIDE_ON; +} + +STAT polygon_split( const BSP_PLANE& plane, const BSP_FACE* in, BSP_FACE* out_fr, BSP_FACE* out_bk ) { + LIST<MAP_VERTEX> front, back; + STAT stat = STAT_ERR; + + for( U32 i = 0; i < in->verts.size; ++i ) { + U32 i2 = (i+1) % in->verts.size; + MAP_VERTEX* a = &in->verts.data[i]; + MAP_VERTEX* c = &in->verts.data[i2]; + + F32 da = bsp_plane_dist( plane, a->pos ); + F32 dc = bsp_plane_dist( plane, c->pos ); + + U8 afront = da > BSP_NORM_EPSILON; + U8 aback = da < -BSP_NORM_EPSILON; + U8 cfront = dc > BSP_NORM_EPSILON; + U8 cback = dc < -BSP_NORM_EPSILON; + + if( !aback ) front.push( *a ); + if( !afront ) back.push( *a ); + + if( (afront && cback) || (aback && cfront) ) { + F32 t = m_clamp( da / (da - dc), 0.f, 1.f ); + VEC3 isec = a->pos + (c->pos - a->pos) * t; + VEC2 uv = a->uv + (c->uv - a->uv) * t; + CLR clr = a->clr + (c->clr - a->clr) * t; + MAP_VERTEX v = { .pos = isec, .uv = uv, .clr = clr }; + front.push( v ); + back.push( v ); + } + } + + if( front.size > 2 ) { + out_fr->propid = in->propid; + out_fr->verts = front; + stat = STAT_OK; + } + if( back.size > 2 ) { + out_bk->propid = in->propid; + out_bk->verts = back; + stat = STAT_OK; + } + + return stat; +} + +I32 bsp_splitter_cost( const BSP_PLANE& plane, const LIST<BSP_FACE>& faces ) { + I32 splits = 0, front = 0, back = 0; + for( U32 i = 0; i < faces.size; ++i ) { + I32 side = polygon_classify( plane, &faces.data[i], 0, 0 ); + if( side == BSP_SIDE_SPAN ) ++splits; + else if( side == BSP_SIDE_FRONT ) ++front; + else if( side == BSP_SIDE_BACK ) ++back; + else { ++front; ++back; } + } + + I32 diff = ( front > back ) ? front - back : back - front; + return splits * 8 + diff; +} + +I32 bsp_calc_splitter_index( const LIST<BSP_PLANE>& planes, const LIST<BSP_FACE>& faces ) { + I32 best = -1; + I32 bestc = INT_MAX; + + for( U32 i = 0; i < planes.size; ++i ) { + I32 cost = bsp_splitter_cost( planes.data[i], faces ); + if( cost < bestc ) { + best = i; + bestc = cost; + } + } + + return best; +} + +U8 bsp_is_convex_cluster( const LIST<BSP_FACE>& faces ) { + for( U32 i = 0; i < faces.size; ++i ) { + VEC3 e1 = faces.data[i].verts[1].pos - faces.data[i].verts[0].pos; + VEC3 e2 = faces.data[i].verts[2].pos - faces.data[i].verts[0].pos; + VEC3 norm = vec_normalize( vec_cross( e1, e2 ) ); + F32 dist = vec_dot( norm, faces.data[i].verts[0].pos ); + BSP_PLANE plane{ norm, dist }; + for( U32 i2 = 0; i2 < faces.size; ++i2 ) { + if( i == i2 ) + continue; + + if( polygon_classify( plane, &faces.data[i2], 0, 0 ) == BSP_SIDE_SPAN ) + return 0; + } + } + + return 1; +} + +I32 bsp_insert_leaf( BSP* bsp, LIST<BSP_FACE>& faces ) { + BSP_LEAF leaf{}; + leaf.first = bsp->faces.size; + leaf.count = faces.size; + leaf.cluster = -1; // todo: pvs + bsp->faces.push_list( faces ); + bsp->leaves.push( leaf ); + return ~( (I32)bsp->leaves.size - 1 ); +} + +I32 bsp_build_nodes( + BSP* bsp, + LIST<BSP_PLANE>& planes, + LIST<BSP_FACE>& faces, + I32 depth ) { + if( faces.size == 0 ) + return bsp_insert_leaf( bsp, faces ); + if( depth > BSP_DEPTH_MAX || bsp_is_convex_cluster( faces ) ) + return bsp_insert_leaf( bsp, faces ); + + I32 splitter_idx = bsp_calc_splitter_index( planes, faces ); + if( splitter_idx < 0 ) return bsp_insert_leaf( bsp, faces ); + + BSP_PLANE split = planes.data[splitter_idx]; + + LIST<BSP_FACE> front, back, coplane; + for( U32 i = 0; i < faces.size; ++i ) { + BSP_FACE* f = &faces.data[i]; + I32 frontc = 0, backc = 0; + I32 side = polygon_classify( split, f, &frontc, &backc ); + if( side == BSP_SIDE_FRONT ) + front.push( *f ); + else if( side == BSP_SIDE_BACK ) + back.push( *f ); + else if( side == BSP_SIDE_ON ) + coplane.push( *f ); + else { + BSP_FACE fr{}, bk{}; + if( OK( polygon_split( split, f, &fr, &bk ) ) ) { + if( fr.verts.size > 2 ) front.push( fr ); + if( bk.verts.size > 2 ) back.push( bk ); + } + } + } + + BSP_NODE node{}; + node.plane = split; + I32 nidx = bsp->nodes.size; + bsp->nodes.push( node ); + + LIST<BSP_PLANE> child_planes; + child_planes.reserve( planes.size > 0 ? planes.size - 1 : 0 ); + for( I32 i = 0; i < planes.size; ++i ) { + if( i != splitter_idx ) child_planes.push( planes.data[i] ); + } + + I32 front_child = bsp_build_nodes( bsp, child_planes, front, depth + 1 ); + if( coplane.size ) { + I32 coplane_leaf = bsp_insert_leaf( bsp, coplane ); + BSP_NODE glue{}; + glue.plane = split; + glue.front = front_child; + glue.back = coplane_leaf; + front_child = bsp->nodes.size; + bsp->nodes.push( glue ); + } + + I32 back_child = bsp_build_nodes( bsp, child_planes, back, depth + 1 ); + + bsp->nodes.data[nidx].front = front_child; + bsp->nodes.data[nidx].back = back_child; + return nidx; +} + +inline void bsp_plane_basis( const VEC3& norm, VEC3* u, VEC3* v ) { + VEC3 a = fabsf( norm.x ) > 0.9f ? VEC3{ 0, 1, 0 } : VEC3{ 1, 0, 0 }; + *u = vec_normalize( vec_cross( a, norm ) ); + *v = vec_normalize( vec_cross( norm, *v ) ); +} + +BSP_WINDING bsp_winding_create( BSP_PLANE* plane ) { + BSP_WINDING w{}; + VEC3 n = plane->normal, nu = n * -1.f; + VEC3 u, v; + bsp_plane_basis( n, &u, &v ); + + VEC3 c = n * plane->dist; + + F32 s = BSP_PORTAL_PLANE_SIZE; + w.points.reserve( 4 ); + w.points.push( c + ( u + v ) * s ); + w.points.push( c + ( nu + v ) * s ); + w.points.push( c + ( nu - v ) * s ); + w.points.push( c + ( u - v ) * s ); + + return w; +} + + +F32 signed_dist_point_to_plane( const BSP_PLANE& plane, const VEC3& point ) { + return vec_dot( plane.normal, point ) - plane.dist; +} + +U8 bsp_clip_winding( BSP_WINDING* w, const BSP_PLANE& plane, U8 keep_front ) { + if( w->points.size < 3 ) + return 0; + + LIST<VEC3> out; + out.reserve( w->points.size + 2 ); + + for( U32 i = 0; i < w->points.size; ++i ) { + VEC3* a = &w->points.data[i]; + VEC3* b = &w->points.data[(i + 1) % w->points.size]; + + F32 da = signed_dist_point_to_plane( plane, *a ); + F32 db = signed_dist_point_to_plane( plane, *b ); + + U8 ain = keep_front ? ( da >= -BSP_NORM_EPSILON ) : ( da <= BSP_NORM_EPSILON ); + U8 bin = keep_front ? ( db >= -BSP_NORM_EPSILON ) : ( db <= BSP_NORM_EPSILON ); + + if( ain ) + out.push( *a ); + + if( ain != bin ) { + F32 t = da / (da - db); + t = m_clamp( t, 0.f, 1.f ); + VEC3 p = *a + (*b - *a) * t; + out.push( p ); + } + } + + if( out.size < 3 ) { + w->points.clear(); + return 0; + } + + w->points = out; + return 1; +} + +U8 bsp_clip_winding_to_set( BSP_WINDING* w, LIST<BSP_PLANE>* planes ) { + for( U32 i = 0; i < planes->size; ++i ) { + if( !bsp_clip_winding( w, planes->data[i], 1 ) ) + return 0; + } + + return 1; +} + +U8 bsp_winding_intersect_plane( BSP_PLANE* plane, BSP_WINDING* wa, BSP_WINDING* wb, BSP_WINDING* out ) { + if( wa->points.size < 3 || wb->points.size < 3 ) + return 0; + + BSP_WINDING w = *wa; + + for( U32 i = 0; i < wb->points.size; ++i ) { + VEC3* b1 = &wb->points[i]; + VEC3* b2 = &wb->points[(i + 1) % wb->points.size]; + + VEC3 edge = *b2 - *b1; + + VEC3 inward_norm = vec_normalize( vec_cross( plane->normal, edge ) ); + F32 d = vec_dot( inward_norm, *b1 ); + + BSP_PLANE horip{ inward_norm, d }; + if( !bsp_clip_winding( &w, horip, 1 ) ) + return 0; + } + + *out = w; + return 1; +} + +void bsp_gather_volumes_for_node( + BSP* bsp, + I32 node_idx, + const LIST<BSP_PLANE>& parents, + U8 go_front, + LIST<BSP_PLANE>* out +) { + *out = parents; + BSP_NODE* n = &bsp->nodes.data[node_idx]; + BSP_PLANE newp = n->plane; + if( !go_front ) { + newp.normal = newp.normal * -1; + newp.dist = -newp.dist; + } + + out->push( newp ); +} + +void bsp_clip_portal_to_subtree( + BSP* bsp, + I32 idx, + BSP_WINDING w, + U8 in_front, + LIST<BSP_PORTAL_NODE>* nodes +) { + if( bsp_is_leaf( idx ) ) { + BSP_PORTAL_NODE newn{ .leaf = bsp_leaf_index( idx ), .w = w }; + nodes->push( newn ); + return; + } + + BSP_NODE* n = &bsp->nodes.data[idx]; + BSP_WINDING wf = w, wb = w; + + if( !bsp_clip_winding( in_front ? &wf : &wb, n->plane, in_front ) ) + return; + + if( in_front ) { + bsp_clip_portal_to_subtree( bsp, n->front, wf, 1, nodes ); + } else { + bsp_clip_portal_to_subtree( bsp, n->back, wb, 0, nodes ); + } +} + +void bsp_build_portals( BSP* bsp ) { + if( bsp->root < 0 ) + return; + + LIST<BSP_PORTAL_STACK_FRAME> stack; + stack.push( BSP_PORTAL_STACK_FRAME{ bsp->root, LIST<BSP_PLANE>{} } ); + + // todo: multithread this + while( stack.size ) { + BSP_PORTAL_STACK_FRAME frame = stack.data[stack.size - 1]; + stack.pop(); + + if( bsp_is_leaf( frame.idx ) ) + continue; + + BSP_NODE* n = &bsp->nodes.data[frame.idx]; + + BSP_WINDING wind = bsp_winding_create( &n->plane ); + BSP_WINDING bounded = wind; + if( bsp_clip_winding_to_set( &bounded, &frame.planes ) ) { + LIST<BSP_PORTAL_NODE> fr_nodes, bk_nodes; + + bsp_clip_portal_to_subtree( bsp, n->front, bounded, 1, &fr_nodes ); + bsp_clip_portal_to_subtree( bsp, n->back, bounded, 1, &bk_nodes ); + + for( U32 ifr = 0; ifr < fr_nodes.size; ++ifr ) { + for( U32 ibk = 0; ibk < bk_nodes.size; ++ibk ) { + BSP_WINDING intersect; + if( bsp_winding_intersect_plane( &n->plane, &fr_nodes.data[ifr].w, &bk_nodes.data[ibk].w, &intersect ) ) { + if( intersect.points.size > 3 ) { + BSP_PORTAL p; + p.w = intersect; + p.plane = n->plane; + p.front = bsp_leaf_index( fr_nodes.data[ifr].leaf ); + p.back = bsp_leaf_index( bk_nodes.data[ibk].leaf ); + bsp->portals.push( p ); + } + } + } + } + } + + LIST<BSP_PLANE> fr, bk; + bsp_gather_volumes_for_node( bsp, frame.idx, frame.planes, 1, &fr ); + bsp_gather_volumes_for_node( bsp, frame.idx, frame.planes, 0, &bk ); + + BSP_NODE* n1 = &bsp->nodes.data[frame.idx]; + if( !bsp_is_leaf( n1->front ) ) stack.push( BSP_PORTAL_STACK_FRAME{ n1->front, fr } ); + if( !bsp_is_leaf( n1->back ) ) stack.push( BSP_PORTAL_STACK_FRAME{ n1->back, bk } ); + } + + for( U32 i = 0; i < bsp->leaves.size; ++i ) { + bsp->leaves.data[i].cluster = (I32)i; + } +} + +void pvs_set( BSP_BITSET* pvs, U32 count, U32 cluster, U32 seen ) { + U32 wp = (count + 31) / 32; + U32 base = cluster * wp; + + U32 w = seen / 32, b = seen % 32; + while( base + w >= pvs->size ) pvs->push( 0 ); + pvs->data[base + w] |= (1u << b); +} + +U8 pvs_get( BSP_BITSET* pvs, U32 count, U32 cluster, U32 seen ) { + U32 wp = (count + 31) / 32; + U32 base = cluster * wp; + + U32 w = seen / 32, b = seen % 32; + if( base + w >= pvs->size ) return 0; + + return ( pvs->data[base + w] >> b) & 1u; +} + +BSP* bsp_build_map( WORLD_MAP* m ) { + BSP* bsp = new BSP; + bsp->map = m; + LIST<BSP_FACE> faces = bsp_map_get_faces( m ); + LIST<BSP_PLANE> planes = bsp_map_get_planes( m ); + + bsp->root = bsp_build_nodes( bsp, planes, faces, 0 ); + U32 vertc = 0; + m->polygons.each( fn( MAP_POLYGON* p ) { vertc += p->vertices.size; } ); + for( U32 i = 0; i < bsp->faces.size; ++i ) { + bsp->faces.data[i].id = i; + } + bsp_gen_render_vertices( bsp ); + return bsp; +} + +void bsp_free( BSP* bsp ) { + delete bsp; +} + diff --git a/src/game/world/bsp.h b/src/game/world/bsp.h new file mode 100644 index 0000000..439fce1 --- /dev/null +++ b/src/game/world/bsp.h @@ -0,0 +1,103 @@ +#pragma once + +#include "map.h" +#include "../../render/gl_3d.h" + +const F32 BSP_NORM_EPSILON = 0.0001f; +const F32 BSP_DIST_EPSILON = 0.01f; +const I32 BSP_DEPTH_MAX = 1024; +struct BSP_PLANE { + VEC3 normal; + F32 dist; + + const bool operator==( const BSP_PLANE& other ) const { + return fabsf( normal.x - other.normal.x ) < BSP_NORM_EPSILON && + fabsf( normal.y - other.normal.y ) < BSP_NORM_EPSILON && + fabsf( normal.z - other.normal.z ) < BSP_NORM_EPSILON && + fabsf( dist - other.dist ) < BSP_DIST_EPSILON; + } +}; + +enum BspSide_t { + BSP_SIDE_FRONT = 1, + BSP_SIDE_BACK = 2, + BSP_SIDE_ON = 4, + BSP_SIDE_SPAN = 8, +}; + +struct BSP_FACE { + I32 propid; + I32 id; + LIST<MAP_VERTEX> verts{}; + LIST<VERTEX3D> render_verts{}; +}; + +struct BSP_NODE { + BSP_PLANE plane; + I32 front; + I32 back; +}; + +struct BSP_LEAF { + U32 first; + U32 count; + I32 cluster; +}; + +struct BSP_WINDING { + LIST<VEC3> points; +}; + +struct BSP_PORTAL { + BSP_WINDING w; // winding + BSP_PLANE plane; + I32 front; + I32 back; +}; + +struct BSP_CLUSTER { + I32 first; + I32 count; + I32 pvs_off; +}; + +struct BSP_PORTAL_STACK_FRAME { + I32 idx; + LIST<BSP_PLANE> planes; +}; + +struct BSP_PORTAL_NODE { + I32 leaf; + BSP_WINDING w; +}; + +struct BSP_BITSET : public LIST<U32> {}; + +struct BSP { + LIST<BSP_NODE> nodes; + LIST<BSP_LEAF> leaves; + LIST<BSP_FACE> faces; + + // >= 0 = node, < 0 = leaf + I32 root; + WORLD_MAP* map; + + LIST<BSP_PORTAL> portals; + LIST<BSP_CLUSTER> clusters; + BSP_BITSET pvs_bits; +}; + +extern LIST<BSP_PLANE> bsp_map_get_planes( WORLD_MAP* map ); +extern LIST<BSP_FACE> bsp_map_get_faces( WORLD_MAP* map ); +extern BSP* bsp_build_map( WORLD_MAP* map ); +extern void bsp_free( BSP* bsp ); + + +inline U8 bsp_is_leaf( I32 child ) { return child < 0; } +inline I32 bsp_leaf_index( I32 child ) { return ~child; } +inline F32 bsp_plane_dist( const BSP_PLANE& p, const VEC3& v ) { return vec_dot( p.normal, v ) - p.dist; } +inline SURF_PROPS* bsp_face_get_props( WORLD_MAP* w, BSP_FACE* s ) { + return s->propid < 0 ? + map_props_get_special( w, s->propid ) : + &w->props[s->propid]; +} diff --git a/src/game/world/bsp_draw.cpp b/src/game/world/bsp_draw.cpp new file mode 100644 index 0000000..30e6731 --- /dev/null +++ b/src/game/world/bsp_draw.cpp @@ -0,0 +1,124 @@ +#include "bsp_draw.h" +#include "../../game.h" +#include "../../render/gl_3d.h" +#include "../objlist.h" +#include "map.h" +#include "trace.h" + + +void bsp_draw_leaf( BSP* bsp, GAME_DATA* game, I32 leafidx ) { + BSP_LEAF* leaf = &bsp->leaves.data[leafidx]; + // todo : handle transparency + + for( U32 i = 0; i < leaf->count; ++i ) { + BSP_FACE* f = &bsp->faces.data[leaf->first + i]; + U8 face_transparent = 0; + if( !face_transparent ) { + LIST<VERTEX3D> verts = f->render_verts; + SURF_PROPS* props = bsp_face_get_props( game->state.map, f ); + gl_3d_polygon( game->render.batch_3d, verts.data, verts.size, props->tex ); + } + } + + // todo : sort by depth + for( U32 i = 0; i < leaf->count; ++i ) { + BSP_FACE* f = &bsp->faces.data[leaf->first + i]; + U8 face_transparent = 0; + if( face_transparent ) { + LIST<VERTEX3D> verts = f->render_verts; + SURF_PROPS* props = bsp_face_get_props( game->state.map, f ); + gl_3d_polygon( game->render.batch_3d, verts.data, verts.size, props->tex ); + } + } +} + +U8 leaf_visible( I32 leaf_idx ) { + // todo : pvs + return 1; +} + +I32 bsp_point_leaf( BSP* bsp, const VEC3& p ) { + I32 n = bsp->root; + for( ;; ) { + if( bsp_is_leaf( n ) ) return bsp_leaf_index( n ); + BSP_NODE* node = &bsp->nodes.data[n]; + F32 dist = bsp_plane_dist( node->plane, p ); + n = (dist >= 0.f)? node->front : node->back; + } +} + +void bsp_traverse_draw( BSP* bsp, GAME_DATA* game, I32 node, VEC3 campos ) { + if( bsp_is_leaf( node ) ) { + I32 i = bsp_leaf_index( node ); + return bsp_draw_leaf( bsp, game, i ); + } + + BSP_NODE* n = &bsp->nodes.data[node]; + F32 dist = bsp_plane_dist( n->plane, campos ); + + I32 near = (dist >= 0.f)? n->front : n->back; + I32 far = (dist >= 0.f)? n->back : n->front; + + // todo : set up vertex buffer, draw all in one call + // todo todo : batch different shader calls separately + + bsp_traverse_draw( bsp, game, near, campos ); + bsp_traverse_draw( bsp, game, far, campos ); +} + +void bsp_draw_sprites( BSP* bsp, GAME_DATA* game ) { + LIST<MAP_SPRITE>* sprites = &game->state.map->sprites; + for( U32 i = 0; i < sprites->size; ++i ) { + MAP_SPRITE* sprite = &sprites->data[i]; + BSP_TRACE t; + t.in_start = player_get_view_pos( objl->pl ); + t.in_end = sprite->pos; + + bsp_trace( &t, bsp ); + if( !t.hit ) { + VEC3 angle = vec_angles( sprite->pos, t.in_start ); + gl_3d_plane( + game->shaders.gl3d, + sprite->pos, + sprite->size, + { -angle.x + 90.f, angle.y - 90.f }, + sprite->tex, + sprite->clr + ); + } + } +} + +void bsp_draw( BSP* bsp, GAME_DATA* game, VEC2 window, VEC2 winsize ) { + gl_set_viewport( game->gl, window, winsize ); + PLAYER* pl = objl->pl; + F32 fov = pl->fov; + gl_3d_projection_setup( + game->shaders.gl3d, + { pl->pos.x, pl->pos.y, pl->pos.z + pl->eyeoffset }, + fov, + pl->rot.y, + pl->rot.x, + 1.f, + 9999.f, + winsize + ); + + gl_set_clip( game->gl, window, winsize ); + gl_batch_empty( game->render.batch_3d ); + glEnable( GL_DEPTH_TEST ); + glEnable( GL_CULL_FACE ); + // ??? + glFrontFace( GL_CW ); + glDepthFunc( GL_LESS ); + bsp_traverse_draw( bsp, game, bsp->root, objl->pl->pos ); + gl_batch_draw( game->render.batch_3d ); + + glDisable( GL_DEPTH_TEST ); + glDisable( GL_CULL_FACE ); + bsp_draw_sprites( bsp, game ); + gl_reset_clip( game->gl ); + + VEC2 canvas = { (F32)game->gl->canvas_size[0], (F32)game->gl->canvas_size[1] }; + gl_set_viewport( game->gl, {}, canvas ); +} diff --git a/src/game/world/bsp_draw.h b/src/game/world/bsp_draw.h new file mode 100644 index 0000000..048fa61 --- /dev/null +++ b/src/game/world/bsp_draw.h @@ -0,0 +1,4 @@ +#include "bsp.h" + +extern void bsp_draw_leaf( BSP* bsp, I32 leafidx ); +extern void bsp_draw( BSP* bsp, GAME_DATA* game, VEC2 window, VEC2 winsize ); diff --git a/src/game/world/draw.cpp b/src/game/world/draw.cpp new file mode 100644 index 0000000..9269a21 --- /dev/null +++ b/src/game/world/draw.cpp @@ -0,0 +1,216 @@ +#include "draw.h" +#include "map.h" + +#include "../../game.h" +#include "../../render/gl_3d.h" + +#include "../objlist.h" +#include "../raycast.h" + +VEC2 world_project_pos( + GAME_DATA* game, + VEC3 pos +) { + GL_3D* gl3d = (GL_3D*)game->shaders.gl3d; + + VEC4 clip = { pos.x, pos.z, pos.y, 1.f }; + VEC4 ndc; + + mat4_mul_vec4( &gl3d->projmat, &clip, &ndc ); + + if( ndc.w <= 0 ) + return { -INFINITY, -INFINITY }; + + ndc.x /= ndc.w; + ndc.y /= ndc.w; + + VEC2 screen = { + ( ndc.x * 0.5f + 0.5f ) * gl3d->winsize.x, + ( -( ndc.y * 0.5f ) + 0.5f ) * gl3d->winsize.y + }; + + return screen; +} + +void world_draw_wall( + GAME_DATA* game, + WORLD* world, + VERTEX3D* verts, + MAP_WALL* s +) { + SURF_PROPS* p = wall_get_props( world->map, s ); + gl_3d_polygon( game->shaders.gl3d, verts, 4, p->tex ); +} + +void world_draw_walls( GAME_DATA* game, WORLD* world ) { + world->map->walls.each( fn( MAP_WALL* s ) { + SURF_PROPS* props = wall_get_props( world->map, s ); + VEC3 start = s->start; + VEC3 end = s->end; + + VERTEX3D verts[4]{}; + verts[0].pos = { start.x, start.z, start.y }; + verts[1].pos = { start.x, start.z + end.z, start.y }; + verts[2].pos = { end.x, start.z + end.z, end.y }; + verts[3].pos = { end.x, start.z, end.y }; + + verts[0].uv = { 0, 1 }; + verts[1].uv = { 0, 0 }; + verts[2].uv = { 1, 0 }; + verts[3].uv = { 1, 1 }; + + verts[0].clr = verts[1].clr = verts[2].clr = verts[3].clr = props->clr; + + world_draw_wall( game, world, verts, s ); + } ); +} + +void world_draw_polygon( + GAME_DATA* game, + WORLD* world, + MAP_POLYGON* p, + LIST<VERTEX3D>* verts +) { + SURF_PROPS* props = polygon_get_props( world->map, p ); + + gl_3d_polygon( + game->shaders.gl3d, + verts->data, + verts->size, + props->tex + ); +} + +void world_draw_polygons( GAME_DATA* game, WORLD* world ) { + for( U32 i = 0; i < world->map->polygons.size; ++i ) { + MAP_POLYGON* p = &world->map->polygons[i]; + SURF_PROPS* props = polygon_get_props( world->map, p ); + + LIST<VERTEX3D> proj_verts; + U8 visc = p->vertices.size; // todo: visc + p->vertices.each( fn( MAP_VERTEX* mv ) { + VERTEX3D v; + v.pos = { mv->pos.x, mv->pos.z, mv->pos.y }; + v.uv = mv->uv; + v.clr = mv->clr * props->clr; + v.sampler = 0; + proj_verts.push( v ); + } ); + + if( proj_verts.size > 2 && visc > 0 ) { + world_draw_polygon( game, world, p, &proj_verts ); + } + } +} + +U8 sprite_is_occluded( + VEC3 player_pos, + MAP_SPRITE* sprite, + WORLD* world +) { + VEC3 ray_dir = sprite->pos - player_pos; + F32 dist_to_sprite = vec_len2d( ray_dir ); + + vec_normalize( &ray_dir ); + + TRACE trace; + trace.startpos = player_pos; + U32 hit = ray_trace( &trace, sprite->pos ); + + F32 dist = vec_len( trace.endpos - player_pos ); + + return hit != TRACE_HIT_NONE && dist < dist_to_sprite; +} + +void world_draw_sprite( + GAME_DATA* game, + MAP_SPRITE* sprite, + VEC3 player_pos, + WORLD* world + ) { + if( sprite_is_occluded( player_pos, sprite, world ) ) + return; + + VEC3 angle = vec_angles( sprite->pos, player_pos ); + gl_3d_plane( + game->shaders.gl3d, + sprite->pos, + sprite->size, + { -angle.x + 90.f, angle.y - 90.f }, + sprite->tex, + sprite->clr + ); +} + +void world_draw_sprites( + GAME_DATA* game, + WORLD* world +) { + VEC3 player_pos = player_get_view_pos( objl->pl ); + for( U32 i = 0; i < world->map->sprites.size; ++i ) { + MAP_SPRITE* sprite = &world->map->sprites[i]; + world_draw_sprite( game, sprite, player_pos, world ); + } +} + +void world_draw( GAME_DATA* game, WORLD* world, VEC2 window, VEC2 winsize ) { + PLAYER* pl = objl->pl; + VEC2 start = { pl->pos.x, pl->pos.y }; + + F32 fov = pl->fov; + + gl_set_viewport( game->gl, window, winsize ); + gl_3d_projection_setup( + game->shaders.gl3d, + { pl->pos.x, pl->pos.y, pl->pos.z + pl->eyeoffset }, + fov, + pl->rot.y, + pl->rot.x, + 1.f, + 9999.f, + winsize + ); + + gl_set_clip( game->gl, window, winsize ); + glEnable( GL_DEPTH_TEST ); + glDepthFunc( GL_LESS ); + world_draw_polygons( game, world ); + world_draw_walls( game, world ); + glDisable( GL_DEPTH_TEST ); + world_draw_sprites( game, world ); + gl_reset_clip( game->gl ); + + VEC2 canvas = { (F32)game->gl->canvas_size[0], (F32)game->gl->canvas_size[1] }; + gl_set_viewport( game->gl, {}, canvas ); +} + +VEC2 world_draw_project_point( + VEC3 vertex_pos, + VEC3 player_pos, + F32 player_angle_deg, + F32 fov_deg, + VEC2 window, + VEC2 winsize, + U8* in_view +) { + F32 rx = vertex_pos.x; + F32 ry = vertex_pos.y; + + F32 dist = sqrtf( rx*rx + ry*ry ); + if( dist < 0.001f ) + return { -99999.0f, -99999.0f }; + + F32 ang = atan2f( ry, rx ); + F32 half_fov_rad = m_deg2rad( fov_deg * 0.5f ); + if( in_view ) *in_view = (ang >= -half_fov_rad && ang <= half_fov_rad); + + F32 cosang = cosf( ang ); + F32 normalized_x = (ang + half_fov_rad) / (2.f * half_fov_rad); + F32 screen_x = window.x + normalized_x * winsize.x; + + F32 screen_y_center = window.y + winsize.y * 0.5f; + F32 vz = ( vertex_pos.z - player_pos.z ); + F32 screen_y = screen_y_center + (vz * winsize.y) / (dist * -cosang); + + return { screen_x, screen_y }; +} diff --git a/src/game/world/draw.h b/src/game/world/draw.h new file mode 100644 index 0000000..6297a5e --- /dev/null +++ b/src/game/world/draw.h @@ -0,0 +1,19 @@ +#pragma once + +#include "../../util/vector.h" + +extern VEC2 world_project_pos( struct GAME_DATA* game, VEC3 pos ); + +extern void world_draw_walls( struct GAME_DATA* game, struct WORLD* world ); +extern void world_draw_sprites( struct GAME_DATA* game, struct WORLD* world ); +extern void world_draw_polygons( struct GAME_DATA* game, struct WORLD* world ); +extern void world_draw( struct GAME_DATA* game, struct WORLD* world, VEC2 window, VEC2 winsize ); +extern VEC2 world_draw_project_point( + VEC3 vertex_pos, + VEC3 player_pos, + F32 player_angle_deg, + F32 fov_deg, + VEC2 window, + VEC2 winsize, + U8* in_view = 0 +); diff --git a/src/game/world/map.cpp b/src/game/world/map.cpp new file mode 100644 index 0000000..aba283d --- /dev/null +++ b/src/game/world/map.cpp @@ -0,0 +1,640 @@ +#include <cfloat> + +#include "map.h" +#include "../../game.h" +#include "../assets.h" +#include "bsp.h" + +const F32 SKYBOX_OFFSET = 24.f; + +MAP_TEXTURE_ENTRY* map_load_texture( WORLD_MAP* m, GAME_DATA* game, const char* name ) { + FNV1A hash = fnv1a( name ); + MAP_TEXTURE_ENTRY** pentry = m->textures.where( fn( MAP_TEXTURE_ENTRY** t ) { return (*t)->hash == hash; } ); + if( pentry ) { + return *pentry; + } + + GL_TEX2D* tex = gl_texture_from_file( game->gl, name ); + if( !tex ) { + dlog( "map_load_texture() : failed to load texture %s\n", name ); + return 0; + } + + MAP_TEXTURE_ENTRY* entry = new MAP_TEXTURE_ENTRY; + entry->tex = tex; + entry->hash = fnv1a( name ); + strcpy( entry->name, name ); + + m->textures.push( entry ); + return entry; +} + +STAT map_add_texture_ref( WORLD_MAP* map, GL_TEX2D *tex ) { + const char* name = assets_abspath( tex->name ); + FNV1A hash = fnv1a( name ); + + I32 idx = map->textures.idx_where( fn( MAP_TEXTURE_ENTRY** e ) { + return (*e)->hash == hash; + } ); + + if( idx != -1 ) + return STAT_ALREADYEXISTS; + + MAP_TEXTURE_ENTRY* entry = new MAP_TEXTURE_ENTRY; + entry->tex = tex; + entry->hash = hash; + strcpy( entry->name, name ); + + return STAT_OK; +} + +GL_TEX2D* map_find_texture( WORLD_MAP* map, const char* name ) { + FNV1A hash = fnv1a( name ); + MAP_TEXTURE_ENTRY** entry = map->textures.where( fn( MAP_TEXTURE_ENTRY** e ) { + return (*e)->hash == hash; + } ); + + if( entry ) + return (*entry)->tex; + + return 0; +} + +void map_polygon_calc_vertex_normals( MAP_POLYGON* p ) { + for( U32 i = 0; i < p->vertices.size; i += 3 ) { + MAP_VERTEX* v1 = &p->vertices[i]; + MAP_VERTEX* v2 = &p->vertices[(i + 1) % p->vertices.size]; + MAP_VERTEX* v3 = &p->vertices[(i + 2) % p->vertices.size]; + + VEC3 d1 = v2->pos - v1->pos; + VEC3 d2 = v3->pos - v1->pos; + VEC3 n = vec_cross( d1, d2 ); + vec_normalize( &n ); + + v1->normal = v2->normal = v3->normal = n; + } +} + +STAT map_polygon_verts_from_section( WORLD_MAP* m, MAP_POLYGON* p, CFG_SECTION* vertices, U32 vertc ) { + char vertsec[256] = { 0 }; + for( U32 i = 0; i < vertc; ++i ) { + MAP_VERTEX vert; + + sprintf( vertsec, "%d", i ); + CFG_SECTION* v = cfg_section( vertices, vertsec ); + if( !v ) { + dlog( "map_polygon_verts_from_section() : missing vertex %d in polygon with propid %d in map %s\n", i, p->propid, m->name ); + continue; + } + + CFG_VEC3* pos = cfg_vec3( v, "pos" ); + CFG_VEC2* uv = cfg_vec2( v, "uv" ); + CFG_CLR* clr = cfg_clr( v, "clr" ); + + if( !pos ) { + dlog( "map_polygon_verts_from_section() : missing pos for vertex %d in polygon with propid %d in map %s\n", i, p->propid, m->name ); + return STAT_ERR; + } + + vert.pos = pos->value; + if( uv ) vert.uv = uv->value; + if( clr ) vert.clr = clr->value; + + p->vertices.push( vert ); + } + + if( !p->vertices.size ) { + dlog( "map_polygon_verts_from_section() : invalid vertex count for polygon with propid %d in map %s\n", p->propid, m->name ); + return STAT_ERR; + } + + map_polygon_calc_vertex_normals( p ); + return STAT_OK; +} + +void map_polygon_calc_bounds( MAP_POLYGON* p ) { + p->vertices.each( fn( MAP_VERTEX* v ) { + p->mins.x = min( v->pos.x, p->mins.x ); + p->mins.y = min( v->pos.y, p->mins.y ); + p->mins.z = min( v->pos.z, p->mins.z ); + + p->maxs.x = max( v->pos.x, p->maxs.x ); + p->maxs.y = max( v->pos.y, p->maxs.y ); + p->maxs.z = max( v->pos.z, p->maxs.z ); + } ); +} + +STAT map_polygons_from_section( WORLD_MAP* m, GAME_DATA* g, CFG_SECTION* polygons, U32 polyc ) { + if( !polyc ) { + dlog( "map_polygons_from_section() : no polygons in %s\n", m->name ); + return STAT_OK; + } + + U32 parsec = 0; + char polysec[256] = { 0 }; + for( U32 i = 0; i < polyc; ++i ) { + MAP_POLYGON poly; + + sprintf( polysec, "%d", i ); + CFG_SECTION* p = cfg_section( polygons, polysec ); + if( !p ) { + dlog( "map_polygons_from_section() : missing polygon %d in %s\n", i, m->name ); + continue; + } + + CFG_INT* propid = cfg_int( p, "propid" ); + if( !propid ) { + dlog( "map_polygons_from_section() : missing propid for polygon %d in map %s\n", i, m->name ); + continue; + } + + poly.propid = propid->value; + CFG_INT* polytype = cfg_int( p, "type" ); + if( !polytype ) { + dlog( "map_polygons_from_section() : missing polygon type for polygon %d in %s\n", i, m->name ); + continue; + } + + poly.type = (U8)polytype->value; + + CFG_INT* vertc = cfg_int( p, "vertcount" ); + if( !vertc || !vertc->value ) { + dlog( "map_polygons_from_section() : missing vertex count for polygon %d in %s\n", i, m->name ); + continue; + } + + CFG_SECTION* vertices = cfg_section( p, "vertices" ); + if( !vertices ) { + dlog( "map_polygons_from_section() : missing vertices for polygon %d in %s", i, m->name ); + continue; + } + + if( !OK( map_polygon_verts_from_section( m, &poly, vertices, vertc->value ) ) ) + continue; + + map_polygon_calc_bounds( &poly ); + m->polygons.push( poly ); + ++parsec; + } + + return parsec > 0 ? STAT_OK : STAT_ERR; +} + +STAT map_walls_from_section( WORLD_MAP* m, GAME_DATA* g, CFG_SECTION* walls, U32 wallc ) { + if( !wallc ) { + dlog( "map_walls_from_section() : no walls in %s\n", m->name ); + return STAT_OK; + } + + U32 parsec = 0; + char wallsec[256] = { 0 }; + for( U32 i = 0; i < wallc; i++ ) { + MAP_WALL seg; + + sprintf( wallsec, "%d", i ); + CFG_SECTION* w = cfg_section( walls, wallsec ); + if( !w ) { + dlog( "map_walls_from_section() : missing wall %d in %s\n", i, m->name ); + continue; + } + + CFG_VEC3* start = cfg_vec3( w, "start" ); + CFG_VEC3* end = cfg_vec3( w, "end" ); + CFG_INT* prop = cfg_int( w, "propid" ); + + if( !start || !end || !prop ) { + dlog( "map_walls_from_section() : bad wall definition in %s idx %d [%p %p %p]\n", m->name, i, start, end, prop ); + continue; + } + + seg.start = start->value; + seg.end = end->value; + seg.propid = prop->value; + m->walls.push( seg ); + + ++parsec; + } + + return parsec > 0 ? STAT_OK : STAT_ERR; +} + + +STAT map_props_from_section( WORLD_MAP* m, GAME_DATA* g, CFG_SECTION* props, U32 propc ) { + if( !propc ) { + dlog( "map_props_from_section() : no props in %s\n", m->name ); + return STAT_ERR; + } + + U32 parsec = 0; + char propsec[256] = { 0 }; + for( U32 i = 0; i < propc; ++i ) { + SURF_PROPS prop; + + sprintf( propsec, "%d", i ); + CFG_SECTION* p = cfg_section( props, propsec ); + if( !p ) { + dlog( "map_props_from_section() : missing prop %d in %s\n", i, m->name ); + continue; + } + + CFG_CLR* clr = cfg_clr( p, "clr" ); + CFG_STR* tex = cfg_str( p, "tex" ); + + CLR propclr = (!!clr)? clr->value : CLR{ 1.f, 1.f, 1.f, 1.f }; + GL_TEX2D* proptex = 0; + if( tex ) { + MAP_TEXTURE_ENTRY* entry = map_load_texture( m, g, tex->str ); + if( !entry ) continue; + + proptex = entry->tex; + } + + prop.clr = propclr; + prop.tex = proptex; + m->props.push( prop ); + + ++parsec; + } + + return parsec > 0 ? STAT_OK : STAT_ERR; +} + +STAT map_sprites_from_section( WORLD_MAP* m, GAME_DATA* g, CFG_SECTION* sprites, U32 spritec ) { + if( !spritec ) { + dlog( "map_sprites_from_section() : no sprites in %s\n", m->name ); + return STAT_OK; + } + + U32 parsec = 0; + char spritesec[256] = { 0 }; + for( U32 i = 0; i < spritec; ++i ) { + MAP_SPRITE sprite; + + sprintf( spritesec, "%d", i ); + CFG_SECTION* s = cfg_section( sprites, spritesec ); + if( !s ) { + dlog( "map_sprites_from_section() : missing sprite %d in %s\n", i, m->name ); + continue; + } + + CFG_CLR* clr = cfg_clr( s, "clr" ); + CFG_STR* tex = cfg_str( s, "tex" ); + CFG_VEC3* pos = cfg_vec3( s, "pos" ); + CFG_VEC2* size = cfg_vec2( s, "size" ); + + if( !tex || !pos || !size ) { + dlog( "map_sprites_from_section() : invalid sprite %d in %s\n", i, m->name ); + continue; + } + + CLR spriteclr = (!!clr)? clr->value : CLR{ 1.f, 1.f, 1.f, 1.f }; + MAP_TEXTURE_ENTRY* texentry = map_load_texture( m, g, tex->str ); + GL_TEX2D* spritetex = texentry->tex; + + sprite.tex = spritetex; + sprite.pos = pos->value; + sprite.size = size->value; + sprite.clr = spriteclr; + m->sprites.push( sprite ); + + ++parsec; + } + + return parsec > 0 ? STAT_OK : STAT_ERR; +} + +STAT map_skybox_from_section( CFG_SECTION* s, GAME_DATA* g, WORLD_MAP* m ) { + CFG_SECTION* props = cfg_section( s, "props" ); + if( !props ) + return STAT_ERR; + + CFG_CLR* clr = cfg_clr( props, "clr" ); + CFG_STR* tex = cfg_str( props, "tex" ); + + CLR skyclr; + GL_TEX2D* skytex; + if( !tex ) { + skyclr = (!!clr)? clr->value : CLR::BLACK(); + skytex = 0; + } + else { + MAP_TEXTURE_ENTRY* texentry = map_load_texture( m, g, tex->str ); + skyclr = (!!clr)? clr->value : CLR::WHITE(); + if( !texentry || !texentry->tex ) { + dlog( "map_skybox_from_section() : could not load skybox texture '%s'", tex->str ); + return STAT_ERR; + } + else { + skytex = texentry->tex; + } + } + + + m->skybox.props = SURF_PROPS{ .tex = skytex, .clr = skyclr }; + return STAT_OK; +} + +MAP_SKYBOX map_default_skybox() { + return { + .props = { .tex = 0, .clr = CLR::BLACK() } + }; +} + +// generates 4 walls and 2 polygons to enclose the map. +void map_gen_skybox( WORLD_MAP* m ) { + VEC3 mins = m->mins - VEC3( SKYBOX_OFFSET, SKYBOX_OFFSET, SKYBOX_OFFSET ); + VEC3 maxs = m->maxs + VEC3( SKYBOX_OFFSET, SKYBOX_OFFSET, SKYBOX_OFFSET ); + + m->skybox.walls[0] = { + .start = { mins.x, mins.y, mins.z }, + .end = { maxs.x, mins.y, maxs.z - mins.z }, + .propid = MAPPROP_SKYBOX + }; + m->skybox.walls[1] = { + .start = { maxs.x, mins.y, mins.z }, + .end = { maxs.x, maxs.y, maxs.z - mins.z }, + .propid = MAPPROP_SKYBOX + }; + m->skybox.walls[2] = { + .start = { maxs.x, maxs.y, mins.z }, + .end = { mins.x, maxs.y, maxs.z - mins.z }, + .propid = MAPPROP_SKYBOX + }; + m->skybox.walls[3] = { + .start = { mins.x, maxs.y, mins.z }, + .end = { mins.x, mins.y, maxs.z - mins.z }, + .propid = MAPPROP_SKYBOX + }; + + VEC2 floor[] = { + { mins.x, mins.y }, + { maxs.x, mins.y }, + { maxs.x, maxs.y }, + { mins.x, maxs.y } + }; + VEC2 floor_uv[] = { + { 0, 0 }, + { 1, 0 }, + { 1, 1 }, + { 0, 1 } + }; + + m->skybox.polygons[0] = m->skybox.polygons[1] = { + .propid = MAPPROP_SKYBOX + }; + + for( U32 i = 0; i < 4; ++i ) { + m->skybox.polygons[0].vertices.push( { + .pos = { floor[i].x, floor[i].y, mins.z }, + .uv = floor_uv[i], // todo: fix + .clr = CLR::WHITE() + } ); + + m->skybox.polygons[1].vertices.push( { + .pos = { floor[i].x, floor[i].y, maxs.z }, + .uv = floor_uv[i], // todo: fix + .clr = CLR::WHITE() + } ); + } +} + +void map_calc_bounds( WORLD_MAP* m ) { + m->mins = { FLT_MAX, FLT_MAX, FLT_MAX }; + m->maxs = { -FLT_MAX, -FLT_MAX, -FLT_MAX }; + m->walls.each( fn( MAP_WALL* l ) { + m->mins.x = min( min( l->start.x, l->end.x ), m->mins.x ); + m->mins.y = min( min( l->start.y, l->end.y ), m->mins.y ); + m->mins.z = min( min( l->start.z, l->start.x + l->end.z ), m->mins.z ); + + m->maxs.x = max( max( l->start.x, l->end.x ), m->maxs.x ); + m->maxs.y = max( max( l->start.y, l->end.y ), m->maxs.y ); + m->maxs.z = max( max( l->start.z, l->start.x + l->end.z ), m->maxs.z ); + } ); + + m->polygons.each( fn( MAP_POLYGON* p ) { + m->mins.x = min( p->mins.x, m->mins.x ); + m->mins.y = min( p->mins.y, m->mins.y ); + m->mins.z = min( p->mins.z, m->mins.z ); + + m->maxs.x = max( p->maxs.x, m->maxs.x ); + m->maxs.y = max( p->maxs.y, m->maxs.y ); + m->maxs.z = max( p->maxs.z, m->maxs.z ); + } ); +} + +void map_check_bounds( WORLD_MAP* m ) { + VEC3 mins = { FLT_MAX, FLT_MAX, FLT_MAX }; + VEC3 maxs = { -FLT_MAX, -FLT_MAX, -FLT_MAX }; + m->walls.each( fn( MAP_WALL* l ) { + mins.x = min( min( l->start.x, l->end.x ), mins.x ); + mins.y = min( min( l->start.y, l->end.y ), mins.y ); + mins.z = min( min( l->start.z, l->start.x + l->end.z ), mins.z ); + + maxs.x = max( max( l->start.x, l->end.x ), maxs.x ); + maxs.y = max( max( l->start.y, l->end.y ), maxs.y ); + maxs.z = max( max( l->start.z, l->start.x + l->end.z ), maxs.z ); + } ); + + m->polygons.each( fn( MAP_POLYGON* p ) { + mins.x = min( p->mins.x, mins.x ); + mins.y = min( p->mins.y, mins.y ); + mins.z = min( p->mins.z, mins.z ); + + maxs.x = max( p->maxs.x, maxs.x ); + maxs.y = max( p->maxs.y, maxs.y ); + maxs.z = max( p->maxs.z, maxs.z ); + } ); + + if( vec_dist( mins, m->mins ) > 0.01f || vec_dist( maxs, m->maxs ) > 0.01f ) { + m->mins = mins; + m->maxs = maxs; + map_gen_skybox( m ); + } +} + +// todo: too long -- extract parts into funcs +WORLD_MAP* map_from_file( GAME_DATA* game, const char* path ) { + CFG_SECTION* root = cfg_load( path ); + + ddef( const char* errstr = "map_from_file() : %s - missing %s section\n" ); + if( !root ) { dlog( "map_from_file() : error opening file %s\n", path ); return 0; } + defer( cfg_free( root ); ); + + WORLD_MAP* m = new WORLD_MAP; + const char* name = file_path_last_of( path ); + strcpy( m->name, name ); + + CFG_SECTION* s = cfg_section( root, "map" ); + if( !s ) { dlog( errstr, "map", path ); delete m; return 0; } + + CFG_SECTION* props = cfg_section( s, "props" ); + if( !props ) { dlog( errstr, path, "props" ); delete m; return 0; } + + CFG_SECTION* walls = cfg_section( s, "walls" ); + if( !walls ) { dlog( errstr, path, "walls" ); delete m; return 0; } + + CFG_SECTION* polys = cfg_section( s, "polygons" ); + if( !polys ) { dlog( errstr, path, "polygons" ); delete m; return 0; } + + CFG_INT* propc = cfg_int( s, "propcount" ); + if( !propc ) { dlog( errstr, path, "propcount" ); delete m; return 0; } + + CFG_INT* wallc = cfg_int( s, "wallcount" ); + if( !wallc ) { dlog( errstr, path, "wallcount" ); delete m; return 0; } + + CFG_INT* polyc = cfg_int( s, "polycount" ); + if( !polyc ) { dlog( errstr, path, "polycount" ); delete m; return 0; } + + if( !OK( map_props_from_section( m, game, props, propc->value ) ) ) { delete m; return 0; } + if( !OK( map_walls_from_section( m, game, walls, wallc->value ) ) ) { delete m; return 0; } + if( !OK( map_polygons_from_section( m, game, polys, polyc->value ) ) ) { delete m; return 0; } + + CFG_INT* spritec = cfg_int( s, "spritecount" ); + if( spritec ) { + CFG_SECTION* sprites = cfg_section( s, "sprites" ); + if( !sprites ) { dlog( errstr, path, "sprites" ); delete m; return 0; } + if( !OK( map_sprites_from_section( m, game, sprites, spritec->value ) ) ) { delete m; return 0; } + } + + CFG_VEC3* startpos = cfg_vec3( s, "startpos" ); + if( !startpos ) { + dlog( errstr, path, "startpos" ); + m->startpos = { 0, 0, 0 }; + } + else { + m->startpos = startpos->value; + } + + CFG_FLOAT* startang = cfg_float( s, "startang" ); + if( !startang ) { + dlog( errstr, path, "startang" ); + m->startang = 0; + } + else { + m->startang = startang->value; + } + + map_calc_bounds( m ); + CFG_SECTION* skybox = cfg_section( s, "skybox" ); + if( !skybox ) { + dlog( errstr, path, "skybox" ); + dlog( "using default skybox" ); + m->skybox = map_default_skybox(); + } + else { + if( !OK( map_skybox_from_section( skybox, game, m ) ) ) { + dlog( errstr, path, "skybox" ); + dlog( "using default skybox" ); + m->skybox = map_default_skybox(); + } + } + map_gen_skybox( m ); + return m; +} + +void map_serialize_info( CFG_SECTION* s, WORLD_MAP* map ) { + cfg_int( "wallcount", s, map->walls.size ); + cfg_int( "polycount", s, map->polygons.size ); + cfg_int( "propcount", s, map->props.size ); + cfg_int( "spritecount", s, map->sprites.size ); + cfg_vec3( "startpos", s, map->startpos ); + cfg_float( "startang", s, map->startang ); +} + +void map_serialize_walls( CFG_SECTION* s, WORLD_MAP* map ) { + CFG_SECTION* walls = cfg_section_new( "walls", s ); + char name[64]; + + U32 i = 0; + map->walls.each( fn( MAP_WALL* seg ) { + sprintf( name, "%d", i++ ); + CFG_SECTION* wallsec = cfg_section_new( name, walls ); + cfg_vec3( "start", wallsec, seg->start ); + cfg_vec3( "end", wallsec, seg->end ); + cfg_int( "propid", wallsec, seg->propid ); + } ); +} + +void map_serialize_polygons( CFG_SECTION* s, WORLD_MAP* map ) { + CFG_SECTION* polygons = cfg_section_new( "polygons", s ); + char name[64]; + + U32 i = 0; + map->polygons.each( fn( MAP_POLYGON* p ) { + sprintf( name, "%d", i++ ); + CFG_SECTION* polygonsec = cfg_section_new( name, polygons ); + cfg_int( "vertcount", polygonsec, p->vertices.size ); + cfg_int( "propid", polygonsec, p->propid ); + cfg_int( "type", polygonsec, p->type ); + + CFG_SECTION* vertices = cfg_section_new( "vertices", polygonsec ); + U32 i2 = 0; + p->vertices.each( fn( MAP_VERTEX* v ) { + sprintf( name, "%d", i2++ ); + CFG_SECTION* vertsec = cfg_section_new( name, vertices ); + cfg_vec3( "pos", vertsec, v->pos ); + cfg_vec2( "uv", vertsec, v->uv ); + cfg_clr( "clr", vertsec, v->clr ); + } ); + } ); +} + +void map_serialize_sprites( CFG_SECTION* s, WORLD_MAP* map ) { + CFG_SECTION* sprites = cfg_section_new( "sprites", s ); + char name[64]; + + U32 i = 0; + map->sprites.each( fn( MAP_SPRITE* s ) { + sprintf( name, "%d", i++ ); + CFG_SECTION* spritesec = cfg_section_new( name, sprites ); + cfg_vec3( "pos", spritesec, s->pos ); + cfg_vec2( "size", spritesec, s->size ); + cfg_clr( "clr", spritesec, s->clr ); + if( s->tex ) { + const char* endp = assets_abspath( s->tex->name ); + cfg_str( "tex", spritesec, endp, 256 ); + } + } ); +} + +void map_serialize_props( CFG_SECTION* s, WORLD_MAP* map ) { + CFG_SECTION* props = cfg_section_new( "props", s ); + char name[64]; + + U32 i = 0; + map->props.each( fn( SURF_PROPS* p ) { + sprintf( name, "%d", i++ ); + CFG_SECTION* propsec = cfg_section_new( name, props ); + cfg_clr( "clr", propsec, p->clr ); + if( p->tex ) { + const char* endp = assets_abspath( p->tex->name ); + cfg_str( "tex", propsec, endp, 256 ); + } + } ); +} + + +CFG_SECTION* map_serialize( WORLD_MAP* map ) { + CFG_SECTION* root = cfg_section_new( "root", 0 ); + CFG_SECTION* s = cfg_section_new( "map", root ); + + map_serialize_info( s, map ); + map_serialize_props( s, map ); + map_serialize_walls( s, map ); + map_serialize_polygons( s, map ); + map_serialize_sprites( s, map ); + + return s; +} + +void map_free( GAME_DATA* game, WORLD_MAP* m ) { + if( !m ) return; + if( m->bsp ) bsp_free( m->bsp ); + + m->textures.each( fn( MAP_TEXTURE_ENTRY** e ) { + gl_texture_destroy( game->gl, (*e)->tex ); + delete *e; + } ); + + delete m; +} diff --git a/src/game/world/map.h b/src/game/world/map.h new file mode 100644 index 0000000..29022db --- /dev/null +++ b/src/game/world/map.h @@ -0,0 +1,137 @@ +#pragma once + +#include "../../util/allocator.h" +#include "../../util/vector.h" +#include "../../util/color.h" +#include "../../util/fnv.h" + +#include <string.h> + +enum MapPropId_t { + MAPPROP_SKYBOX = -1, +}; + +struct SURF_PROPS { + struct GL_TEX2D* tex; + CLR clr; +}; + +struct MAP_VERTEX { + VEC3 pos; + VEC3 normal; + VEC2 uv; + CLR clr; +}; + +enum MapPolygonType_t { + MPT_FLOOR, + MPT_CEILING +}; + +struct MAP_POLYGON { + MAP_POLYGON& operator=( const MAP_POLYGON& other ) { + memcpy( this, &other, sizeof(*this) ); + vertices = other.vertices; + + return *this; + } + + LIST<MAP_VERTEX> vertices; + + VEC3 mins; + VEC3 maxs; + U8 type; + I32 propid; +}; + +struct MAP_WALL { + VEC3 start; + VEC3 end; + + VEC2 uvstart; + VEC2 uvend; + + I32 propid; +}; + +struct MAP_TEXTURE_ENTRY { + char name[256]; + struct GL_TEX2D* tex; + FNV1A hash; +}; + +struct MAP_SPRITE { + VEC3 pos; + VEC2 size; + CLR clr; + GL_TEX2D* tex; +}; + +struct MAP_ENTITY { + U32 classid; + LIST<struct OBJECT_PROP*> props; +}; + +struct MAP_SKYBOX { + SURF_PROPS props; + MAP_WALL walls[4]; + MAP_POLYGON polygons[2]; +}; + +struct WORLD_MAP { + LIST<MAP_WALL> walls; + LIST<MAP_POLYGON> polygons; + LIST<MAP_SPRITE> sprites; + LIST<SURF_PROPS> props; + MAP_SKYBOX skybox; + F32 w; + F32 h; + + VEC3 mins; + VEC3 maxs; + char name[256]; + + VEC3 startpos; + F32 startang; + + struct BSP* bsp{}; + + LIST<MAP_TEXTURE_ENTRY*> textures; +}; + +extern WORLD_MAP* map_from_file( struct GAME_DATA* game, const char* filename ); +extern struct CFG_SECTION* map_serialize( WORLD_MAP* map ); +extern STAT map_add_texture_ref( WORLD_MAP* map, GL_TEX2D* tex ); +extern GL_TEX2D* map_find_texture( WORLD_MAP* map, const char* name ); +extern void map_free( struct GAME_DATA* game, WORLD_MAP* map ); + +// checks mins/maxs and recreates skybox if needed +extern void map_check_bounds( WORLD_MAP* map ); +extern void map_calc_bounds( WORLD_MAP* map ); +extern void map_polygon_calc_bounds( MAP_POLYGON* p ); + + +inline SURF_PROPS* map_props_get_special( WORLD_MAP* w, I32 prop ) { + switch( prop ) { + case MAPPROP_SKYBOX: return &w->skybox.props; + }; + dlog( "map_props_get_special(): unknown special prop %d\n", prop ); + return 0; +} + +inline SURF_PROPS* map_get_props( WORLD_MAP* m, I32 idx ) { + return idx < 0 ? + map_props_get_special( m, idx ) : + &m->props[idx]; +} + +inline SURF_PROPS* wall_get_props( WORLD_MAP* w, MAP_WALL* s ) { + return s->propid < 0 ? + map_props_get_special( w, s->propid ) : + &w->props[s->propid]; +} +inline SURF_PROPS* polygon_get_props( WORLD_MAP* w, MAP_POLYGON* s ) { + return s->propid < 0 ? + map_props_get_special( w, s->propid ) : + &w->props[s->propid]; +} diff --git a/src/game/world/trace.cpp b/src/game/world/trace.cpp new file mode 100644 index 0000000..b3a0041 --- /dev/null +++ b/src/game/world/trace.cpp @@ -0,0 +1,393 @@ +#include <cfloat> + +#include "trace.h" +#include "../../util/math.h" +#include "bsp.h" + +// i have no idea how this works tbh i pasted this +// todo : backface culling (return 0 if d < BSP_TRACE_EPSILON) +U8 bsp_segment_intersects_tri( + const VEC3& va, const VEC3& vb, + const VEC3& v0, const VEC3& v1, const VEC3& v2, + F32* fract, + VEC2* out_uv = 0 +) { + VEC3 dir = vb - va; + VEC3 e1 = v1 - v0, + e2 = v2 - v0; + + VEC3 p = vec_cross( dir, e2 ); + F32 d = vec_dot( e1, p ); + + if( fabsf( d ) < BSP_TRACE_EPSILON ) { + // parallel + return 0; + } + + F32 inv = 1.0f / d; + VEC3 ivec = va - v0; + F32 u = vec_dot( ivec, p ) * inv; + if( u < -BSP_TRACE_EPSILON || u > 1.0f + BSP_TRACE_EPSILON ) { + // outside + return 0; + } + + VEC3 q = vec_cross( ivec, e1 ); + F32 v = vec_dot( dir, q ) * inv; + if( v < -BSP_TRACE_EPSILON || u + v > 1.0f + BSP_TRACE_EPSILON ) { + // outside + return 0; + } + + F32 t = vec_dot( e2, q ) * inv; + if( t < -BSP_TRACE_EPSILON || t > 1.0f + BSP_TRACE_EPSILON ) { + // outside + return 0; + } + + if( fract ) *fract = m_clamp( t, 0.f, 1.f ); + if( out_uv ) { + *out_uv = { u, v }; + } + + return 1; +} + +U8 bsp_trace_leaf_segment( BSP_TRACE* trace, BSP* bsp, I32 leafidx, const VEC3& va, const VEC3& vb ) { + BSP_LEAF* leaf = &bsp->leaves.data[leafidx]; + U8 didhit = 0; + F32 best = FLT_MAX; + + for( U32 i = 0; i < leaf->count; ++i ) { + BSP_FACE* face = &bsp->faces.data[leaf->first + i]; + + // shouldnt happen + if( face->verts.size < 3 ) + continue; + + for( U32 t0 = 1; t0 + 1 < face->verts.size; ++t0 ) { + VEC3 v0 = face->verts[0].pos; + VEC3 v1 = face->verts[t0].pos; + VEC3 v2 = face->verts[t0 + 1].pos; + + // todo : skip func for props + + F32 t; + VEC2 uv; + + if( !bsp_segment_intersects_tri( va, vb, v0, v1, v2, &t, &uv ) ) + continue; + + if( t < best ) { + best = t; + trace->hit = didhit = 1; + trace->frac = t; + trace->point = va + (vb - va) * t; + trace->normal = vec_normalize( vec_cross( v1 - v0, v2 - v0 ) ); + trace->propid = face->propid; + trace->leafid = leafidx; + trace->faceid = leaf->first + i; + } + } + } + + return didhit; +} + +U8 bsp_trace_segment( BSP_TRACE* trace, BSP* bsp, I32 nodeidx, const VEC3& va, const VEC3& vb ) { + if( bsp_is_leaf( nodeidx ) ) { + I32 i = bsp_leaf_index( nodeidx ); + return bsp_trace_leaf_segment( trace, bsp, i, va, vb ); + } + + BSP_NODE* node = &bsp->nodes.data[nodeidx]; + F32 da = bsp_plane_dist( node->plane, va ); + F32 db = bsp_plane_dist( node->plane, vb ); + + if( da > BSP_TRACE_EPSILON && db > BSP_TRACE_EPSILON ) + return bsp_trace_segment( trace, bsp, node->front, va, vb ); + if( da < -BSP_TRACE_EPSILON && db < -BSP_TRACE_EPSILON ) + return bsp_trace_segment( trace, bsp, node->back, va, vb ); + + VEC3 dir = vb - va; + F32 denom = vec_dot( node->plane.normal, dir ); + I32 near = (da >= 0.f) ? node->front : node->back; + I32 far = (da >= 0.f) ? node->back : node->front; + if( fabsf( denom ) < BSP_TRACE_EPSILON ) { + if( bsp_trace_segment( trace, bsp, near, va, vb ) ) + return 1; + return bsp_trace_segment( trace, bsp, far, va, vb ); + } + + F32 t = (node->plane.dist - vec_dot( node->plane.normal, va ) ) / denom; + t = m_clamp( t, 0.f, 1.f ); + + VEC3 iplane = va + dir * t; + if( bsp_trace_segment( trace, bsp, near, va, iplane ) ) + return 1; + + if( t <= BSP_TRACE_EPSILON || t >= 1.0f - BSP_TRACE_EPSILON ) { + return bsp_trace_segment( trace, bsp, far, iplane, vb ); + } + + return bsp_trace_segment( trace, bsp, far, iplane, vb ); +} + +// nudge the trace ends slightly to prevent exact-coplanar infinite recursion. +// quake does this every recursion divided up by the length of the trace +// should not matter in the end +void bsp_trace_nudge( BSP_TRACE* trace ) { + VEC3 dir = trace->in_end - trace->in_start; + VEC3 nudge = dir * BSP_TRACE_EPSILON; + trace->in_start += nudge; + trace->in_end -= nudge; +} + +U8 bsp_trace( BSP_TRACE* trace, BSP* bsp, VEC3 start, VEC3 end ) { + if( !trace ) { + dlog( "bsp_trace() : null trace\n" ); + abort(); + } + + trace->in_start = start; + trace->in_end = end; + + return bsp_trace( trace, bsp ); +} + +U8 bsp_trace( BSP_TRACE* trace, BSP* bsp ) { + if( !trace ) { + dlog( "bsp_trace() : null trace\n" ); + abort(); + } + + trace->hit = 0; + + bsp_trace_nudge( trace ); + VEC3 start = trace->in_start; + VEC3 end = trace->in_end; + + return bsp_trace_segment( trace, bsp, bsp->root, start, end ); +} + +inline VEC3 bsp_face_get_normal( const BSP_FACE* f ){ + const VEC3 v0 = f->verts.data[0].pos; + const VEC3 v1 = f->verts.data[1].pos; + const VEC3 v2 = f->verts.data[2].pos; + return vec_normalize( vec_cross( v1 - v0, v2 - v0 ) ); +} + +inline U8 point_in_inflated_poly( + const VEC3& point, + const BSP_FACE* face, + const AABB& hull, + const VEC3& norm +){ + VEC3 center{}; + U32 c = face->verts.size; + for( U32 i = 0; i < c; ++i ) + center = center + face->verts.data[i].pos; + if( c ) center = center * (1.0f / c); + + for( U32 i = 0; i < c; ++i ) { + VEC3 a = face->verts.data[i].pos; + VEC3 b = face->verts.data[(i+1)%c].pos; + VEC3 in = vec_cross( norm, b - a ); + F32 len = vec_len( in ); + if( len <= 0.00001f ) continue; + in *= (1.0f/ len); + if( vec_dot( center - a, in ) < 0.0f ) + in = in * -1.0f; + + F32 dist = vec_dot( point - a, in ); + VEC2 feet = aabb_extend_at_feet( hull, in ); + F32 pos_r = feet.x, neg_r = feet.y; + F32 expand = norm.z >= 0 ? neg_r : pos_r; + if( dist < -expand - BSP_TRACE_EPSILON ) + return 0; + } + return 1; +} + + +// big fat ass +U8 bsp_trace_sweep_aabb_to_face( + BSP_TRACE* trace, + const AABB& hull, + BSP_FACE* face, + const VEC3& start, + const VEC3& end, + U32 face_idx +) { + if( face->verts.size < 3 ) + return 0; + + VEC3 norm = bsp_face_get_normal( face ); + F32 d = vec_dot( norm, face->verts[0].pos ); + VEC2 feet = aabb_extend_at_feet( hull, norm ); + F32 pos_r = feet.x, neg_r = feet.y; + F32 radius = norm.z >= 0 ? neg_r : pos_r; + + VEC3 dir = end - start; + F32 s0 = vec_dot( norm, trace->in_start ) - d; + F32 ndir = vec_dot( norm, trace->in_end - trace->in_start ); + + if( fabsf( s0 ) <= radius + BSP_TRACE_EPSILON ) { + if( point_in_inflated_poly( start, face, hull, norm ) ) { + trace->hit = 1; + trace->frac = 0.f; + trace->point = start; + trace->normal = norm; + trace->propid = face->propid; + trace->faceid = face_idx; + return 1; + } + } + + if( fabsf( ndir ) < BSP_TRACE_EPSILON ) + return 0; + + F32 t_enter = (-radius - s0) / ndir; + F32 t_exit = ( radius - s0) / ndir; + if( t_enter > t_exit ) { F32 tmp=t_enter; t_enter=t_exit; t_exit=tmp; } + + if( t_exit < 0.f - BSP_TRACE_EPSILON || t_enter > 1.f + BSP_TRACE_EPSILON ) + return 0; + + F32 t = m_clamp( t_enter, 0.f, 1.f ); + VEC3 step = start + dir * t; + + if( !point_in_inflated_poly( step, face, hull, norm ) ) + return 0; + + trace->hit = 1; + trace->frac = t; + trace->point = step; + trace->normal = norm; + trace->propid = face->propid; + trace->faceid = face_idx; + return 1; +} + +U8 bsp_trace_sweep_aabb_to_leaf( + BSP* bsp, + BSP_TRACE* trace, + const AABB& hull, + const VEC3& start, + const VEC3& end, + I32 leaf_idx +) { + BSP_LEAF* leaf = &bsp->leaves.data[leaf_idx]; + U8 hit = 0; + F32 best = FLT_MAX; + + BSP_TRACE besthit = *trace; + + for( U32 i = 0; i < leaf->count; ++i ) { + U32 face_idx = leaf->first + i; + BSP_FACE* f = &bsp->faces.data[face_idx]; + + if( f->propid == MAPPROP_SKYBOX ) + continue; + + BSP_TRACE tr = *trace; + if( !bsp_trace_sweep_aabb_to_face( &tr, hull, f, start, end, face_idx ) ) + continue; + + if( tr.frac < best ) { + best = tr.frac; + besthit = tr; + besthit.leafid = leaf_idx; + hit = 1; + } + } + + if( hit ) { + *trace = besthit; + trace->hit = 1; + } + return hit; +} + +U8 bsp_trace_sweep_aabb( + BSP_TRACE* trace, + BSP* bsp, + const AABB& hull, + const VEC3& start, + const VEC3& end, + I32 node_idx +) { + if( bsp_is_leaf( node_idx ) ) { + I32 i = bsp_leaf_index( node_idx ); + return bsp_trace_sweep_aabb_to_leaf( bsp, trace, hull, start, end, i ); + } + + BSP_NODE* node = &bsp->nodes.data[node_idx]; + VEC3 n = node->plane.normal; + F32 d = node->plane.dist; + + F32 s_dist = vec_dot( n, start ) - d; + F32 e_dist = vec_dot( n, end ) - d; + F32 offset = aabb_support_radius( hull, n ); + + if( s_dist > offset && e_dist > offset ) + return bsp_trace_sweep_aabb( trace, bsp, hull, start, end, node->front ); + + if( s_dist < -offset && e_dist < -offset ) + return bsp_trace_sweep_aabb( trace, bsp, hull, start, end, node->back ); + + VEC3 dir = end - start; + F32 denom = ( e_dist - s_dist ); + + if( fabsf(denom) < BSP_TRACE_EPSILON ) { + I32 near = (s_dist >= 0.f) ? node->front : node->back; + I32 far = (s_dist >= 0.f) ? node->back : node->front; + if( bsp_trace_sweep_aabb( trace, bsp, hull, start, end, near ) ) + return 1; + return bsp_trace_sweep_aabb( trace, bsp, hull, start, end, far ); + } + + F32 tin = ( offset - s_dist) / denom; + F32 tout = (-offset - s_dist) / denom; + if( tin > tout ) { F32 tmp = tin; tin = tout; tout = tmp; } + + tin = m_clamp( tin, 0.f, 1.f ); + tout = m_clamp( tout , 0.f, 1.f ); + + VEC3 mid = start + dir * tin; + I32 near = (s_dist > 0.f) ? node->front : node->back; + I32 far = (s_dist > 0.f) ? node->back : node->front; + + if( bsp_trace_sweep_aabb( trace, bsp, hull, start, mid, near ) ) + return 1; + + if( tin <= BSP_TRACE_EPSILON || tin >= 1.f - BSP_TRACE_EPSILON ) { + F32 nudge = (denom >= 0.f ? 1.f : -1.f) * BSP_TRACE_EPSILON * 4.f; + VEC3 nudged = mid + dir * nudge; + return bsp_trace_sweep_aabb( trace, bsp, hull, nudged, end, far ); + } + + return bsp_trace_sweep_aabb(trace, bsp, hull, mid, end, far ); +} + +U8 bsp_trace( BSP_TRACE* trace, BSP* bsp, AABB hull ) { + if( !trace || !bsp ) + return 0; + + trace->hit = 0; + bsp_trace_nudge( trace ); + VEC3 start = trace->in_start; + VEC3 end = trace->in_end; + + U8 ret = bsp_trace_sweep_aabb( trace, bsp, hull, start, end, bsp->root ); + return ret; +} + +U8 bsp_trace( BSP_TRACE* trace, BSP* bsp, AABB hull, VEC3 start, VEC3 end ) { + if( !trace || !bsp ) + return 0; + + trace->in_start = start; + trace->in_end = end; + return bsp_trace( trace, bsp, hull ); +} diff --git a/src/game/world/trace.h b/src/game/world/trace.h new file mode 100644 index 0000000..8a0f4eb --- /dev/null +++ b/src/game/world/trace.h @@ -0,0 +1,29 @@ +#pragma once + +#include "bsp.h" +#include "../../util/aabb.h" + +// quake uses 1 / 32 +const F32 BSP_TRACE_EPSILON = 1.f / 64.f; + +struct BSP_TRACE { + VEC3 in_start; + VEC3 in_end; + + U8 hit; + F32 frac; + VEC3 point; + VEC3 normal; + U32 propid; + I32 leafid; + U32 faceid; +}; + +extern U8 bsp_trace( BSP_TRACE* trace, BSP* bsp ); +extern U8 bsp_trace( BSP_TRACE* trace, BSP* bsp, VEC3 start, VEC3 end ); + +extern U8 bsp_trace( BSP_TRACE* trace, BSP* bsp, AABB aabb ); +extern U8 bsp_trace( BSP_TRACE* trace, BSP* bsp, AABB aabb, VEC3 start, VEC3 end ); + +extern U8 bsp_trace( BSP_TRACE* trace ); +extern U8 bsp_trace( BSP_TRACE* trace, VEC3 start, VEC3 end ); diff --git a/src/game/world/world.cpp b/src/game/world/world.cpp new file mode 100644 index 0000000..337b496 --- /dev/null +++ b/src/game/world/world.cpp @@ -0,0 +1,14 @@ +#include "world.h" +#include "../objlist.h" + +WORLD* world_create( WORLD_MAP* map ) { + WORLD* w = obj_add<WORLD>( "root" ); + w->map = map; + + return w; +} + +STAT world_populate_entities( WORLD *world ) { + + return STAT_OK; +} diff --git a/src/game/world/world.h b/src/game/world/world.h new file mode 100644 index 0000000..229f235 --- /dev/null +++ b/src/game/world/world.h @@ -0,0 +1,11 @@ +#pragma once +#include "../object.h" + +struct WORLD : OBJECT { + static const U32 CLASSID = OBJCLASS_WORLD; + + struct WORLD_MAP* map; +}; + +extern WORLD* world_create( WORLD_MAP* map ); +extern STAT world_populate_entities( WORLD* world ); diff --git a/src/gl.h b/src/gl.h new file mode 100644 index 0000000..099788c --- /dev/null +++ b/src/gl.h @@ -0,0 +1,5 @@ +#pragma once + +#include "render/gl.h" +#include "render/gl_2d.h" +#include "render/gl_2d_font.h" diff --git a/src/gui.h b/src/gui.h new file mode 100644 index 0000000..8a620ff --- /dev/null +++ b/src/gui.h @@ -0,0 +1,13 @@ +#include "gui/base.h" + + +#ifdef IS_EDITOR +inline void gui_create( struct GAME_DATA* game ) { + gui_init( game ); + + //gui_window( 300, 200 ); + //gui_title( "hello, world!" ); + + gui_end(); +} +#endif diff --git a/src/gui/base.cpp b/src/gui/base.cpp new file mode 100644 index 0000000..5d345d8 --- /dev/null +++ b/src/gui/base.cpp @@ -0,0 +1,405 @@ +#include <string.h> + +#include "base.h" +#include "../game.h" + +// =========== [ impl ] =========== +#include "../util/input.h" +#include "../render/gl_2d.h" +#include "../render/gl_2d_font.h" +// ================================ + +__gui_internal _gui; + +void gui_load_font( GL_DATA* gl, __gui_internal::__font* fnt, const char* name ) { + // todo: this probably crashes when fonts r re-created + GL_FONT** fptr = gl->fonts.where( fn( GL_FONT** ptr ) { + GL_FONT* f = *ptr; + return strstr( f->name, name ) != 0; + } ); + + if( !fptr ) { + dlog( "gui_load_font() : error loading font %s\n", name ); + return; + } + + fnt->glfnt = *fptr; +} + +void gui_init( GAME_DATA* game ) { + GL_DATA* gl = game->gl; + gui_load_font( gl, &_gui.fonts.jpn12, "jpn12" ); + gui_load_font( gl, &_gui.fonts.jpn16, "jpn16" ); +} + +void gui_end() { + _gui.cur_window = 0; +} + +void gui_draw( GAME_DATA* game ) { + _gui.gl2d = game->shaders.gl2d; + _gui.gl2d_font = game->shaders.gl2d_texcoord; + + _gui.windows.each( fn( GUI_WINDOW** w ) { + GUI_WINDOW* wnd = *w; + if( !wnd->enabled ) return; + + wnd->draw_fn( wnd ); + } ); +} + +void gui_input( GAME_DATA* game ) { + for( I32 i = (I32)_gui.windows.size - 1; i >= 0; --i ) { + GUI_WINDOW* wnd = _gui.windows[i]; + if( !wnd->enabled ) continue; + + wnd->input_fn( wnd ); + break; + } +} + +void gui_onframe( GAME_DATA* game ) { + gui_run_callbacks(); + + gui_input( game ); + gui_draw( game ); +} + +void gui_free( GUI_BASE* node ) { + node->children.each( fn( GUI_BASE** ptr ) { + GUI_BASE* child = *ptr; + gui_free( child ); + } ); + + node->children.clear(); + + if( !node->parent ) { + I32 idx = _gui.windows.idx_of( (GUI_WINDOW*)node ); + if( idx != -1 ) + _gui.windows.erase( idx ); + + if( _gui.cur_window == node ) { + if( _gui.windows.size > 0 ) { + idx = idx > _gui.windows.size - 1 ? _gui.windows.size - 1 : 0; + _gui.cur_window = _gui.windows[idx]; + } + } + } + else if( _gui.cur_view == node ) { + _gui.cur_view = 0; + for( U32 i = 0; i < node->parent->children.size; ++i ) { + GUI_BASE* n = node->parent->children.data[i]; + if( n == node ) + continue; + + if( !strcmp( n->name, "BASE_VIEW" ) ) { + _gui.cur_view = (GUI_VIEW*)n; + break; + } + + n = gui_find_node( n, "BASE_VIEW" ); + if( n != node ) { + _gui.cur_view = (GUI_VIEW*)n; + break; + } + } + } + + delete node; +} + +I32 gui_relx( GUI_BASE* node ) { + I32 x = node->x; + for( GUI_BASE* p = node->parent; !!p; p = p->parent ) { + x += p->x; + } + return x; +} + +I32 gui_rely( GUI_BASE* node ) { + I32 y = node->y; + for( GUI_BASE* p = node->parent; !!p; p = p->parent ) { + y += p->y; + } + return y; +} + +GUI_BASE* gui_find_node( GUI_BASE* root, const char* name ) { + if( !root ) + return 0; + + if( !strcmp( root->name, name ) ) + return root; + + for( U32 i = 0; i < root->children.size; ++i ) { + GUI_BASE* child = root->children.data[i]; + GUI_BASE* node = gui_find_node( child, name ); + if( node ) + return node; + } + + return 0; +} + +GUI_WINDOW* gui_get_parent_wnd( GUI_BASE* node ) { + if( !node->parent ) + return (GUI_WINDOW*)node; + + for( GUI_BASE* b = node; !!b; b = b->parent ) { + if( !b->parent ) + return (GUI_WINDOW*)b; + } + + return (GUI_WINDOW*)node; +} + +void gui_empty_children( GUI_BASE* node ) { + node->children.each( fn( GUI_BASE** ptr ) { + gui_free( *ptr ); + } ); + + node->children.clear(); +} + +GUI_VIEW* gui_get_view() { + return _gui.cur_view; +} + +void gui_set_view( struct GUI_VIEW* view ) { + _gui.cur_view = view; +} + +GUI_WINDOW* gui_get_window() { + return _gui.cur_window; +} + +void gui_set_window( GUI_WINDOW* window ) { + _gui.cur_window = window; +} + +void gui_bring_to_top( GUI_WINDOW* window ) { + for( U32 i = 0; i < _gui.windows.size; ++i ) { + if( _gui.windows[i] == window ) { + _gui.windows.erase( i ); + break; + } + } + + _gui.windows.push( window ); +} + +U8 gui_is_fg_window( GUI_BASE* node ) { + GUI_WINDOW* top_wnd; + for( I32 i = _gui.windows.size - 1; i >= 0; --i ) { + GUI_WINDOW* wnd = _gui.windows[i]; + if( !wnd->enabled ) + continue; + + top_wnd = wnd; + break; + } + + if( !node->parent ) { + return top_wnd == node; + } + + return gui_get_parent_wnd( node ) == top_wnd; +} + +U8 gui_check_target() { + if( !_gui.cur_view ) { + dlog( "gui_check_target() : no view" ); + return 0; + } + + return 1; +} + +void gui_run_callbacks() { + LIST<__gui_internal::callback_entry> cb_copy = _gui.callbacks; + _gui.callbacks.clear(); + + for( I32 i = cb_copy.size - 1; i >= 0; --i ) { + __gui_internal::callback_entry* e = &cb_copy[i]; + if( e->callback ) + e->callback( e->data ); + } +} + +void gui_push_callback( GUI_CALLBACK cb ) { + __gui_internal::callback_entry e; + e.data = 0; + e.callback = cb; + + _gui.callbacks.push( e ); +} + +void gui_push_callback( void* data, GUI_CALLBACK cb ) { + __gui_internal::callback_entry e; + e.data = data; + e.callback = cb; + + _gui.callbacks.push( e ); +} + +void gui_base_input_fn( void* ptr ) { + GUI_BASE* e = (GUI_BASE*)ptr; + if( !e ) return; + + e->children.each( fn( GUI_BASE** ptr ) { + GUI_BASE* child = *ptr; + if( child->enabled && child->input_fn ) + child->input_fn( child ); + } ); +} + +void gui_draw_frect( I32 x, I32 y, I32 w, I32 h, CLR col ) { + gl_2d_frect( _gui.gl2d, { (F32)x, (F32)y }, { (F32)w, (F32)h }, col ); +} + +void gui_draw_rect( I32 x, I32 y, I32 w, I32 h, CLR col ) { + gl_2d_rect( _gui.gl2d, { (F32)x, (F32)y }, { (F32)w, (F32)h }, col ); +} + +void gui_draw_line( I32 x0, I32 y0, I32 x1, I32 y1, CLR col ) { + gl_2d_line( _gui.gl2d, { (F32)x0, (F32)y0 }, { (F32)x1, (F32)y1 }, col ); +} + +GL_FONT* gui_font_from_idx( U32 fnt ) { + if( fnt > FNT_LAST ) { + dlog( "gui_font_from_idx() : invalid font index %d", fnt ); + return 0; + } + + __gui_internal::__font* pfnt; + switch( fnt ) { + case FNT_JPN12: pfnt = &_gui.fonts.jpn12; break; + case FNT_JPN16: pfnt = &_gui.fonts.jpn16; break; + default: return 0; + }; + + return pfnt->glfnt; +} + +void gui_draw_str_internal( I32 x, I32 y, U8 align, U32 fnt, CLR col, const char* str ) { + GL_FONT* pfnt = gui_font_from_idx( fnt ); + if( !pfnt ) return; + + I32 w; + gui_draw_get_str_bounds( &w, 0, fnt, str ); + switch( align ) { + case ALIGN_R: x -= w; break; + case ALIGN_C: x -= (I32)( w * 0.5f ); break; + } + + gl_font_draw( pfnt, _gui.gl2d_font, { (F32)x, (F32)y }, str, col ); +} + +void gui_draw_str( I32 x, I32 y, U8 align, U32 fnt, CLR col, const char* fmt, ... ) { + va_list args; + va_start( args, fmt ); + char buf[GUI_STR_BUF_MAX]; + vsprintf( buf, fmt, args ); + va_end( args ); + + gui_draw_str_internal( x, y, align, fnt, col, buf ); +} + +void gui_draw_get_str_bounds_internal( I32* w, I32* h, U32 fnt, const char* buf ) { + GL_FONT* pfnt = gui_font_from_idx( fnt ); + if( !pfnt ) return; + + VEC2 dim = gl_font_dim( pfnt, buf ); + if( w ) *w = dim.x; + if( h ) *h = dim.y; +} + +void gui_draw_get_str_bounds( I32* w, I32* h, U32 fnt, const char *fmt, ... ) { + va_list args; + va_start( args, fmt ); + char buf[GUI_STR_BUF_MAX]; + vsprintf( buf, fmt, args ); + va_end( args ); + + gui_draw_get_str_bounds_internal( w, h, fnt, buf ); +} + +void gui_draw_get_clip( I32 *x, I32 *y, I32 *w, I32 *h ) { + VEC2 start, dim; + gl_get_clip( _gui.gl2d->gl, &start, &dim ); + + if( x ) *x = (I32)start.x; + if( y ) *y = (I32)start.y; + if( w ) *w = (I32)dim.x; + if( h ) *h = (I32)dim.y; +} + +void gui_draw_set_clip( I32 x, I32 y, I32 w, I32 h ) { + VEC2 start, dim; + start.x = (F32)x; + start.y = (F32)y; + dim.x = (F32)w; + dim.y = (F32)h; + + gl_set_clip( _gui.gl2d->gl, start, dim ); +} + +void gui_draw_reset_clip() { + gl_reset_clip( _gui.gl2d->gl ); +} + +void gui_draw_push_clip( I32 x, I32 y, I32 w, I32 h ) { + I32 cx,cy,cw,ch; + gui_draw_get_clip( &cx, &cy, &cw, &ch ); + + __gui_internal::clip_rect rect{ cx, cy, cw, ch }; + _gui.clip_rects.push( rect ); + + gui_draw_set_clip( x, y, w, h ); +} + +void gui_draw_pop_clip() { + if( !_gui.clip_rects.size ) { + dlog( "gui_draw_pop_clip() : clip stack empty\n" ); + return gui_draw_reset_clip(); + } + + __gui_internal::clip_rect rect = _gui.clip_rects.pop(); + gui_draw_set_clip( rect.x, rect.y, rect.w, rect.h ); +} + +void gui_cursor_pos( I32 *x, I32 *y ) { + if( x ) *x = input.mouse.pos.x; + if( y ) *y = input.mouse.pos.y; +} + +U8 gui_mbutton_down( U8 button ) { + switch( button ) { + case GUI_MBTNLEFT: + return input.mouse.left; + case GUI_MBTNRIGHT: + return input.mouse.right; + case GUI_MBTNMIDDLE: + return input.mouse.middle; + case GUI_MBTNSCROLL: + return input.mouse.wheel; + }; + + return 0; +} + +void gui_capture_scroll() { + input.mouse.wheel = 0; +} + +U8 gui_key_down( U8 key ) { + return input.keys[key]; +} + +F32 gui_frametime() { + return _gui.gl2d->gl->frametime; +} + +F32 gui_time() { + return u_time(); +} diff --git a/src/gui/base.h b/src/gui/base.h new file mode 100644 index 0000000..96e0fa5 --- /dev/null +++ b/src/gui/base.h @@ -0,0 +1,314 @@ +#pragma once +#include "../util/color.h" +#include "../util/allocator.h" + +// ======================================= [ colorscheme ] ======================================== + +static const struct { + CLR txt = CLR::WHITE(); + CLR txt_inactive = CLR::blend( CLR::WHITE(), CLR::BLACK(), 0.3f ); + CLR border = CLR::WHITE(); + CLR border_inactive = CLR::blend( CLR::WHITE(), CLR::BLACK(), 0.55f ); + CLR bg = CLR::blend( CLR::WHITE(), CLR::BLACK(), 0.92f ); + CLR bg_sec = CLR::blend( CLR::WHITE(), CLR::BLACK(), 0.97f ); + CLR bg_alt = CLR::blend( CLR::WHITE(), CLR::BLACK(), 0.985f ); +} ui_clr; + +// ========================================== [ base ] ============================================ + +const U32 GUI_STR_BUF_MAX = 16384; +const U32 GUI_TEXTBOX_MAX = 1024; +const U32 GUI_NAME_LEN = 512; + +/* impl */ +extern void gui_init( struct GAME_DATA* game ); +extern void gui_draw( struct GAME_DATA* game ); +extern void gui_input( struct GAME_DATA* game ); +extern void gui_onframe( struct GAME_DATA* game ); +extern void gui_end(); +/* */ + +extern I32 gui_relx( struct GUI_BASE* node ); //relative x pos for node +extern I32 gui_rely( struct GUI_BASE* node ); //relative y pos for node + +extern struct GUI_BASE* gui_find_node( struct GUI_BASE* root, const char* name ); +extern struct GUI_WINDOW* gui_get_parent_wnd( struct GUI_BASE* node ); +extern void gui_empty_children( struct GUI_BASE* node ); + +extern void gui_free( struct GUI_BASE* node ); + +extern struct GUI_VIEW* gui_get_view(); // gets currently edited view +extern void gui_set_view( struct GUI_VIEW* view ); // sets currently edited view +extern struct GUI_WINDOW* gui_get_window(); // gets currently edited window +extern void gui_set_window( struct GUI_WINDOW* wnd ); // sets currently edited window +extern void gui_bring_to_top( struct GUI_WINDOW* wnd ); // pushes window to be rendered last +extern U8 gui_is_fg_window( struct GUI_BASE* node ); // 1 if parent window (or window itself) is rendered last + +// ======================================= [ components ] ========================================= + +struct GUI_LIST_ENTRY { + I32 val; + char title[256]; +}; + +typedef void( *GUI_CALLBACK )( void* ptr ); + +extern struct GUI_VIEW* gui_view( I32 x, I32 y, I32 w, I32 h ); // sets current view +extern struct GUI_WINDOW* gui_window( I32 w, I32 h ); // sets current view and window +extern struct GUI_WINDOW* gui_window( I32 x, I32 y, I32 w, I32 h ); // sets current view and window +extern struct GUI_TITLE* gui_title( const char* title ); + +extern struct GUI_LABEL* gui_label( I32 x, I32 y, const char* title, ... ); +extern struct GUI_BUTTON* gui_button( I32 x, I32 y, I32 w, I32 h, const char* title, GUI_CALLBACK cb ); +extern struct GUI_TEXTBOX* gui_textbox( I32 x, I32 y, I32 w, I32 h, const char* title, U32 maxlen, U8 allow_newl = 0 ); +extern struct GUI_CHECKBOX* gui_checkbox( I32 x, I32 y, const char* title, U8* pval ); + +extern struct GUI_LIST* gui_list( I32 x, I32 y, I32 w, I32 h, const char* title, LIST<GUI_LIST_ENTRY>* list, I32* pval ); +extern struct GUI_LIST_ENTRY* gui_list_get_selected( struct GUI_LIST* list ); + +extern struct GUI_FLOATINPUT* gui_floatinput( + I32 x, I32 y, I32 w, + const char* title, + F32* pval, + F32 min, + F32 max, + F32 step = 1.f, + const char* printfmt = "%.2f" +); + +extern struct GUI_VECTORINPUT* gui_vectorinput( + I32 x, I32 y, I32 w, + const char* title, + F32* pval, + U32 valc, + F32 min, + F32 max, + F32 step = 1.f, + const char* letters = "xyzw", + const char* printfmt = "%.2f" +); + +extern struct GUI_COLORINPUT* gui_colorinput( + I32 x, + I32 y, + I32 w, + const char* title, + CLR* pval, + U8 showalpha = 1 +); + +// ======================================= [ definitions ] ======================================== + +enum GuiTxtAlign_t { + ALIGN_L, + ALIGN_R, + ALIGN_C +}; + +enum GuiFont_t { + FNT_JPN12, + FNT_JPN16, + FNT_LAST +}; + +typedef void( *GUI_DRAW_FN )( void* el ); +typedef void( *GUI_INPUT_FN )( void* el ); + +extern void gui_base_input_fn( void* ); + +struct GUI_BASE { + I32 x{}, y{}; + I32 w{}, h{}; + U8 enabled{1}; + U8 canvas{}; + + char name[GUI_NAME_LEN]; + + I32 xbound{}, ybound{}; + I32 xoff{}, yoff{}; + + GUI_BASE* parent{}; + LIST<GUI_BASE*> children{}; + + GUI_DRAW_FN draw_fn{}; + GUI_INPUT_FN input_fn = gui_base_input_fn; +}; + +struct GUI_VIEW : GUI_BASE { + U8 initheld; +}; +struct GUI_WINDOW : GUI_VIEW { + U8 locked{}; + U8 ontop{}; +}; + +struct GUI_LABEL : GUI_BASE {}; + +struct GUI_TITLE : GUI_BASE { + I32 xoff, yoff; + U8 held{}; +}; + +struct GUI_LIST : GUI_BASE { + LIST<GUI_LIST_ENTRY>* plist; + I32* pval; + U8 held; + + GUI_CALLBACK cb; +}; + +struct GUI_BUTTON : GUI_BASE { + GUI_CALLBACK cb; + U8 held; + void* extra; +}; + +struct GUI_CHECKBOX : GUI_BASE { + U8 held; + + U8* pval; + GUI_CALLBACK cb; +}; + +struct GUI_TEXTBOX : GUI_BASE { + char value[GUI_TEXTBOX_MAX]; + + U32 len; + U32 maxlen; + + U8 keys[0xff]; + U8 prevkeys[0xff]; + + U8 heldkey; + F32 heldtime; + F32 repeattime; + + F32 blinktime; + U8 blink; + + U8 allow_newl; + U8 active; + + U8 m1held; + + GUI_CALLBACK cb; +}; + +struct GUI_FLOATINPUT : GUI_BASE { + const char* valfmt; + F32* pval; + + F32 min; + F32 max; + F32 step; + + I32 lastmx; + U8 heldoutbounds; + U8 held; + + // for color sliders + CLR bgcol; + U8 customclr; + + U8 wraparound; + GUI_CALLBACK cb; +}; + +struct GUI_VECTORINPUT : GUI_VIEW { + const char* valfmt; + F32* pval; + U32 valc; + + const char* letters; + F32 min; + F32 max; + F32 step; + + GUI_VIEW* inputview; + LIST<GUI_FLOATINPUT*> inputs; + + GUI_CALLBACK cb; +}; + +struct GUI_COLORINPUT : GUI_VECTORINPUT {}; + +// ======================================== [ internal ] ========================================== + +extern void gui_push_callback( GUI_CALLBACK cb ); +extern void gui_push_callback( void* data, GUI_CALLBACK cb ); +extern void gui_run_callbacks(); + +extern U8 gui_check_target(); + +extern void gui_draw_line( I32 x0, I32 y0, I32 x1, I32 y1, CLR col ); +extern void gui_draw_rect( I32 x, I32 y, I32 w, I32 h, CLR col ); +extern void gui_draw_frect( I32 x, I32 y, I32 w, I32 h, CLR col ); +extern void gui_draw_circle( I32 x, I32 y, I32 r, CLR col ); + +extern void gui_draw_str( I32 x, I32 y, U8 align, U32 fnt, CLR col, const char* fmt, ... ); +extern void gui_draw_get_str_bounds( I32* w, I32* h, U32 fnt, const char* fmt, ... ); + +extern void gui_draw_get_clip( I32* x, I32* y, I32* w, I32* h ); +extern void gui_draw_set_clip( I32 x, I32 y, I32 w, I32 h ); +extern void gui_draw_reset_clip(); +extern void gui_draw_push_clip( I32 x, I32 y, I32 w, I32 h ); +extern void gui_draw_pop_clip(); + +extern F32 gui_frametime(); +extern F32 gui_time(); +extern void gui_cursor_pos( I32* x, I32* y ); +extern U8 gui_key_down( U8 key ); +extern U8 gui_mbutton_down( U8 button ); +extern void gui_capture_scroll(); + +enum __gui_internal_mouse_button { + GUI_MBTNLEFT, + GUI_MBTNRIGHT, + GUI_MBTNMIDDLE, + GUI_MBTNSCROLL +}; + +struct __gui_internal { + GUI_VIEW* cur_view; + GUI_WINDOW* cur_window; + LIST<GUI_WINDOW*> windows; + struct clip_rect { + I32 x,y,w,h; + }; + + LIST<clip_rect> clip_rects; + struct callback_entry { + GUI_CALLBACK callback; + void* data; + }; + + LIST<callback_entry> callbacks; + + struct __font { + struct GL_FONT* glfnt; + }; + + struct { + __font jpn12; + __font jpn16; + } fonts; + + // draw stuff below + struct GL_SHADER_PROGRAM* gl2d; + struct GL_SHADER_PROGRAM* gl2d_font; +}; + +void __gui_internal_vectorinput_init( + GUI_VECTORINPUT* input, + I32 x, I32 y, I32 w, + const char* title, + F32* pval, + U32 valc, + F32 min, + F32 max, + F32 step, + const char* letters, + const char* printfmt +); + +extern __gui_internal _gui; diff --git a/src/gui/button.cpp b/src/gui/button.cpp new file mode 100644 index 0000000..2b57772 --- /dev/null +++ b/src/gui/button.cpp @@ -0,0 +1,66 @@ +#include "base.h" + +void gui_button_draw_fn( void* ptr ) { + GUI_BUTTON* btn = (GUI_BUTTON*)ptr; + + I32 x = gui_relx( btn ); + I32 y = gui_rely( btn ); + + CLR col = gui_is_fg_window( btn )? ui_clr.border : ui_clr.border_inactive; + gui_draw_frect( x, y, btn->w, btn->h, col ); + gui_draw_frect( x+1, y+1, btn->w-2, btn->h-2, ui_clr.bg_sec ); + + I32 middle = x + btn->w/2; + I32 middle_y = y + btn->h/2 - 7; + + gui_draw_str( middle, middle_y, ALIGN_C, FNT_JPN12, ui_clr.txt, btn->name ); +} + +void gui_button_input_fn( void* ptr ) { + GUI_BUTTON* btn = (GUI_BUTTON*)ptr; + + I32 x = gui_relx( btn ); + I32 y = gui_rely( btn ); + + U8 m1 = gui_mbutton_down( 0 ); + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + U8 inbounds = mx >= x && mx <= x + btn->w && my >= y && my <= y + btn->h; + + if( !m1 ) { + // button could be destroyed by callback + U8 was_held = btn->held; + btn->held = 0; + if( inbounds && was_held ) + btn->cb( btn ); + + return; + } + + if( inbounds ) { + btn->held = 1; + } +} + +GUI_BUTTON* gui_button( I32 x, I32 y, I32 w, I32 h, const char* title, GUI_CALLBACK cb ) { + if( !gui_check_target() ) return 0; + + GUI_BUTTON* btn = new GUI_BUTTON; + btn->x = x; + btn->y = y; + btn->xbound = btn->w = w; + btn->ybound = btn->h = h; + btn->cb = cb; + btn->draw_fn = gui_button_draw_fn; + btn->input_fn = gui_button_input_fn; + + btn->parent = _gui.cur_view; + strcpy( btn->name, title ); + + btn->extra = 0; + btn->held = 0; + + gui_get_view()->children.push( btn ); + return btn; +} diff --git a/src/gui/checkbox.cpp b/src/gui/checkbox.cpp new file mode 100644 index 0000000..02cadc6 --- /dev/null +++ b/src/gui/checkbox.cpp @@ -0,0 +1,78 @@ +#include "base.h" + +const I32 CHECKBOX_SIZE = 14; + +void gui_checkbox_draw_fn( void* ptr ) { + GUI_CHECKBOX* check = (GUI_CHECKBOX*)ptr; + + I32 x = gui_relx( check ); + I32 y = gui_rely( check ); + + I32 half = CHECKBOX_SIZE / 2; + + CLR col = gui_is_fg_window( check )? ui_clr.border : ui_clr.border_inactive; + gui_draw_frect( x, y+2, CHECKBOX_SIZE, CHECKBOX_SIZE, col ); + gui_draw_frect( x+1, y+3, CHECKBOX_SIZE-2, CHECKBOX_SIZE-2, ui_clr.bg_sec ); + + if( *check->pval ) + gui_draw_str( x + half, y, ALIGN_C, FNT_JPN12, ui_clr.txt, "x" ); + + gui_draw_str( x + CHECKBOX_SIZE + 2, y, ALIGN_L, FNT_JPN12, ui_clr.txt, check->name ); +} + +void gui_checkbox_input_fn( void* ptr ) { + GUI_CHECKBOX* check = (GUI_CHECKBOX*)ptr; + + I32 x = gui_relx( check ); + I32 y = gui_rely( check ); + + U8 m1 = gui_mbutton_down( 0 ); + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + gui_draw_get_str_bounds( &check->w, 0, FNT_JPN12, check->name ); + check->w += CHECKBOX_SIZE + 2; + + U8 inbounds = mx >= x && mx <= x + check->w && my >= y && my <= y + check->h; + + if( !m1 ) { + // checkbox could be destroyed by callback + U8 was_held = check->held; + check->held = 0; + if( inbounds && was_held ) { + *check->pval = !*check->pval; + if( check->cb ) + check->cb( check ); + } + + return; + } + + if( inbounds ) + check->held = 1; +} + +GUI_CHECKBOX* gui_checkbox( I32 x, I32 y, const char* title, U8* pval ) { + if( !gui_check_target() ) return 0; + + GUI_CHECKBOX* check = new GUI_CHECKBOX; + check->x = x; + check->y = y; + check->cb = 0; + check->ybound = check->h = 16; + check->draw_fn = gui_checkbox_draw_fn; + check->input_fn = gui_checkbox_input_fn; + + check->pval = pval; + + check->parent = _gui.cur_view; + strcpy( check->name, title ); + + gui_draw_get_str_bounds( &check->w, 0, FNT_JPN12, check->name ); + check->w += CHECKBOX_SIZE + 2; + check->xbound = check->w; + check->ybound = check->h; + + gui_get_view()->children.push( check ); + return check; +} diff --git a/src/gui/colorinput.cpp b/src/gui/colorinput.cpp new file mode 100644 index 0000000..1d649fb --- /dev/null +++ b/src/gui/colorinput.cpp @@ -0,0 +1,59 @@ +#include "base.h" +#include <stdio.h> + +void gui_colorinput_draw_fn( void* ptr ) { + GUI_COLORINPUT* input = (GUI_COLORINPUT*)ptr; + + I32 x = gui_relx( input ); + I32 y = gui_rely( input ); + + gui_draw_str( x, y, ALIGN_L, FNT_JPN12, ui_clr.txt, input->name ); + + CLR val = *(CLR*)input->pval; + char hex[16]; + sprintf( hex, "#%02x%02x%02x", + (I32)(val.r * 255.f), + (I32)(val.g * 255.f), + (I32)(val.b * 255.f) + ); + + gui_draw_str( x + input->w, y, ALIGN_R, FNT_JPN12, ui_clr.txt, hex ); + + CLR border = { + 1.f - val.r, + 1.f - val.g, + 1.f - val.b, + 1.f + }; + + gui_draw_frect( x + input->w - 21, y + 14, 22, 22, border ); + gui_draw_frect( x + input->w - 19, y + 16, 18, 18, CLR::BLACK() ); + gui_draw_frect( x + input->w - 19, y + 16, 18, 18, *(CLR*)(input->pval) ); + + GUI_VIEW* inputview = input->inputview; + inputview->draw_fn( inputview ); +} + +GUI_COLORINPUT* gui_colorinput( I32 x, I32 y, I32 w, const char* title, CLR* pval, U8 showalpha ) { + if( !gui_check_target() ) return 0; + + GUI_COLORINPUT* input = new GUI_COLORINPUT; + + __gui_internal_vectorinput_init( + input, + x, y, + w - 26, + title, + (F32*)pval, + showalpha? 4 : 3, + 0.f, + 1.f, + 1.f / 255.f, + "rgba", + "%.02f" + ); + + input->xbound = input->w = w; + input->draw_fn = gui_colorinput_draw_fn; + return input; +} diff --git a/src/gui/console.h b/src/gui/console.h new file mode 100644 index 0000000..e69de29 --- /dev/null +++ b/src/gui/console.h diff --git a/src/gui/floatinput.cpp b/src/gui/floatinput.cpp new file mode 100644 index 0000000..9f7fdfe --- /dev/null +++ b/src/gui/floatinput.cpp @@ -0,0 +1,257 @@ +#include "base.h" +#include <math.h> + +U8 gui_floatinput_is_bound_val( GUI_FLOATINPUT* input ) { + if( input->wraparound ) + return 0; + if( !isfinite( input->min ) || isnan( input->min ) ) + return 0; + if( !isfinite( input->max ) || isnan( input->max ) ) + return 0; + if( input->min > input->max ) + return 0; + return 1; +} + +CLR gui_floatinput_get_progress_clr( GUI_FLOATINPUT* input ) { + if( input->customclr ) { + if( gui_is_fg_window( input ) ) return input->bgcol; + else return CLR::blend( input->bgcol, CLR::BLACK(), 0.333f ); + } + + CLR clr_border = gui_is_fg_window( input )? ui_clr.border : ui_clr.border_inactive; + return CLR::blend( clr_border, CLR::BLACK(), 0.5f ); +} + +void gui_floatinput_draw_bound( GUI_FLOATINPUT* input ) { + I32 x = gui_relx( input ); + I32 y = gui_rely( input ); + I32 w = input->w; + I32 h = input->h; + + F32 min = input->min; + F32 max = input->max; + + F32 val = *input->pval; + F32 percent = (val - min) / (max - min); + if( percent > 1.f ) percent = 1.f; + if( percent < 0.f ) percent = 0.f; + + CLR clr_progress = gui_floatinput_get_progress_clr( input ); + gui_draw_frect( x+1, y+1, (I32)( (w-2) * percent ), h-2, clr_progress ); + + gui_draw_push_clip( (x+1) + (I32)( (w-2) * percent ), (y+1), (I32)( (w-2) * (1.f - percent) ), h-2 ); + gui_draw_str( x + 2, y + 2, ALIGN_L, FNT_JPN12, ui_clr.txt, input->name ); + gui_draw_str( x + w - 2, y + 2, ALIGN_R, FNT_JPN12, ui_clr.txt, input->valfmt, val ); + gui_draw_set_clip( (x+1), (y+1), (I32)( (w-2) * percent ), h-2 ); + gui_draw_str( x + 2, y + 2, ALIGN_L, FNT_JPN12, ui_clr.bg_alt, input->name ); + gui_draw_str( x + w - 2, y + 2, ALIGN_R, FNT_JPN12, ui_clr.bg_alt, input->valfmt, val ); + gui_draw_pop_clip(); +} + +void gui_floatinput_draw_unbound( GUI_FLOATINPUT* input ) { + I32 x = gui_relx( input ); + I32 y = gui_rely( input ); + I32 w = input->w; + + F32 val = *input->pval; + + gui_draw_str( x + 2, y + 2, ALIGN_L, FNT_JPN12, ui_clr.txt, input->name ); + gui_draw_str( x + w - 2, y + 2, ALIGN_R, FNT_JPN12, ui_clr.txt, input->valfmt, val ); + + I32 t1w, t2w, t3w; + gui_draw_get_str_bounds( &t1w, 0, FNT_JPN12, input->name ); + gui_draw_get_str_bounds( &t2w, 0, FNT_JPN12, input->valfmt, val ); + gui_draw_get_str_bounds( &t3w, 0, FNT_JPN12, "<->" ); + + I32 stw = t2w + t3w - 2; + I32 pos = w/2; + if( stw > pos ) + pos = stw; + + CLR handleclr = CLR::blend( ui_clr.txt, CLR::BLACK(), .7f ); + + if( t1w < w / 2 && t1w + t2w + t3w + 8 < w ) { + gui_draw_str( + x + w - pos, + y + 2, + ALIGN_C, + FNT_JPN12, + handleclr, + "<->" + ); + } + else { + gui_draw_str( x + w - pos + 8, y, ALIGN_C, FNT_JPN12, handleclr, "-" ); + gui_draw_str( x + w - pos + 8, y + 5, ALIGN_C, FNT_JPN12, handleclr, "+" ); + } +} + +void gui_floatinput_draw_fn( void* ptr ) { + GUI_FLOATINPUT* input = (GUI_FLOATINPUT*)ptr; + + I32 x = gui_relx( input ); + I32 y = gui_rely( input ); + I32 w = input->w; + I32 h = input->h; + + CLR clr_border = gui_is_fg_window( input )? ui_clr.border : ui_clr.border_inactive; + + gui_draw_frect( x, y, w, h, clr_border ); + gui_draw_frect( x+1, y+1, w-2, h-2, ui_clr.bg_sec ); + + if( !gui_floatinput_is_bound_val( input ) ) + gui_floatinput_draw_unbound( input ); + else + gui_floatinput_draw_bound( input ); +} + +void gui_floatinput_input_bound( GUI_FLOATINPUT* input ) { + if( !gui_mbutton_down( GUI_MBTNLEFT ) ) + return; + + I32 x = gui_relx( input ); + I32 w = input->w; + + F32 min = input->min; + F32 max = input->max; + + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + F32 progress = (F32)(mx - x) / w; + if( progress < 0.f ) progress = 0.f; + if( progress > 1.f ) progress = 1.f; + + F32 nval = min + (max - min) * progress; + F32 rmn = remainderf( nval, input->step ); + *input->pval = nval - rmn; +} + +void gui_floatinput_input_unbound( GUI_FLOATINPUT* input ) { + if( !gui_mbutton_down( GUI_MBTNLEFT ) ) + return; + + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + I32 dx = mx - input->lastmx; + if( dx ) + *input->pval += dx * input->step; + + F32 min = input->min; + F32 max = input->max; + + if( isfinite( min ) && *input->pval < min ) + *input->pval = input->wraparound? max : min; + if( isfinite( max ) && *input->pval > max ) + *input->pval = input->wraparound? min : max; + + F32 rmn = remainderf( *input->pval, input->step ); + *input->pval -= rmn; +} + +void gui_floatinput_input_scroll( GUI_FLOATINPUT* input ) { + U8 scroll = gui_mbutton_down( GUI_MBTNSCROLL ); + gui_capture_scroll(); + F32 oldval = *input->pval; + F32 nval = oldval; + + if( !scroll ) + return; + + if( scroll == 1 ) + nval += input->step; + else if( scroll == (U8)-1 ) + nval -= input->step; + + if( isfinite( input->min ) && nval < input->min ) + nval = input->wraparound? input->max : input->min; + if( isfinite( input->max ) && nval > input->max ) + nval = input->wraparound? input->min : input->max; + + F32 rmn = remainderf( nval, input->step ); + *input->pval = nval - rmn; + + if( input->cb ) + input->cb( input ); +} + +void gui_floatinput_input_fn( void* ptr ) { + GUI_FLOATINPUT* input = (GUI_FLOATINPUT*)ptr; + + I32 m1 = gui_mbutton_down( 0 ); + + I32 x = gui_relx( input ); + I32 y = gui_rely( input ); + I32 w = input->w; + I32 h = input->h; + + I32 mx, my; + gui_cursor_pos( &mx, &my ); + U8 inbounds = mx >= x && mx <= x + w && my >= y && my <= y + h; + + if( inbounds ) + gui_floatinput_input_scroll( input ); + + if( !input->held && m1 && !inbounds ) + input->heldoutbounds = 1; + if( !input->heldoutbounds && m1 && inbounds ) { + if( !input->held ) { + input->lastmx = mx; + input->held = 1; + return; + } + } + + if( !m1 ) { + input->heldoutbounds = 0; + input->held = 0; + return; + } + + if( input->heldoutbounds ) + return; + + F32 oldv = *input->pval; + if( !gui_floatinput_is_bound_val( input ) ) + gui_floatinput_input_unbound( input ); + else + gui_floatinput_input_bound( input ); + + if( input->cb && oldv != *input->pval ) { + input->cb( input ); + } + + input->lastmx = mx; +} + +struct GUI_FLOATINPUT* gui_floatinput( I32 x, I32 y, I32 w, const char* title, F32* pval, F32 min, F32 max, F32 step, const char* valfmt ) { + if( !gui_check_target() ) return 0; + + GUI_FLOATINPUT* input = new GUI_FLOATINPUT; + input->x = x; + input->y = y; + input->xbound = input->w = w; + input->ybound = input->h = 20; + strcpy( input->name, title ); + input->input_fn = gui_floatinput_input_fn; + input->draw_fn = gui_floatinput_draw_fn; + + input->cb = 0; + input->pval = pval; + input->min = min; + input->max = max; + input->step = step; + input->valfmt = valfmt; + + input->wraparound = 0; + input->customclr = 0; + input->held = 0; + + GUI_VIEW* parent = gui_get_view(); + parent->children.push( input ); + input->parent = parent; + + return input; +} diff --git a/src/gui/label.cpp b/src/gui/label.cpp new file mode 100644 index 0000000..b5d6290 --- /dev/null +++ b/src/gui/label.cpp @@ -0,0 +1,40 @@ +#include "base.h" +#include <stdarg.h> +#include <stdio.h> + +void gui_label_draw_fn( void* ptr ) { + GUI_LABEL* label = (GUI_LABEL*)ptr; + + I32 x = gui_relx( label ); + I32 y = gui_rely( label ); + + gui_draw_get_str_bounds( &label->w, &label->h, FNT_JPN12, label->name ); + label->xbound = label->w; + label->ybound = label->h; + gui_draw_str( x, y, ALIGN_L, FNT_JPN12, ui_clr.txt, label->name ); +} + +GUI_LABEL* gui_label( I32 x, I32 y, const char* title, ... ) { + if( !gui_check_target() ) return 0; + char buf[GUI_NAME_LEN]; + + va_list args; + va_start( args, title ); + vsprintf( buf, title, args ); + va_end( args ); + + GUI_LABEL* label = new GUI_LABEL; + label->x = x; + label->y = y; + label->draw_fn = gui_label_draw_fn; + strcpy( label->name, buf ); + + gui_draw_get_str_bounds( &label->w, &label->h, FNT_JPN12, label->name ); + label->xbound = label->w; + label->ybound = label->h; + + label->parent = gui_get_view(); + label->parent->children.push( label ); + + return label; +} diff --git a/src/gui/list.cpp b/src/gui/list.cpp new file mode 100644 index 0000000..90f2f49 --- /dev/null +++ b/src/gui/list.cpp @@ -0,0 +1,98 @@ +#include "base.h" + +const I32 LIST_TITLE_OFFSET = 15; +const I32 LIST_ITEM_HEIGHT = 18; + +void gui_list_draw_fn( void* ptr ) { + GUI_LIST* list = (GUI_LIST*)ptr; + + I32 x = gui_relx( list ); + I32 y = gui_rely( list ); + + gui_draw_str( x, y, ALIGN_L, FNT_JPN12, ui_clr.txt, list->name ); + y += LIST_TITLE_OFFSET; + + CLR col = gui_is_fg_window( list )? ui_clr.border : ui_clr.border_inactive; + gui_draw_frect( x, y, list->w, list->h, col ); + gui_draw_frect( x+1, y+1, list->w-2, list->h-2, ui_clr.bg_sec ); + + I32 yoff = 0; + list->plist->each( fn( GUI_LIST_ENTRY* e ) { + U8 selected = e->val == *list->pval; + CLR col = selected? ui_clr.txt : ui_clr.txt_inactive; + + gui_draw_str( x + 4, y + yoff + 6, ALIGN_L, FNT_JPN12, col, e->title ); + yoff += LIST_ITEM_HEIGHT; + } ); +} + +void gui_list_input_fn( void* ptr ) { + GUI_LIST* list = (GUI_LIST*)ptr; + + I32 x = gui_relx( list ); + I32 y = gui_rely( list ); + I32 w = list->w; + I32 h = list->h; + + U8 m1 = gui_mbutton_down( 0 ); + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + U8 inbounds = mx >= x && mx <= x + w && my >= y && my <= y + h; + if( !inbounds ) + return; + + I32 diff = my - y; + I32 idx = diff / LIST_ITEM_HEIGHT - 1; + if( idx >= list->plist->size ) idx = list->plist->size - 1; + if( idx < 0 ) idx = 0; + + if( m1 ) { + if( !list->held ) { + GUI_LIST_ENTRY* e = &list->plist->data[idx]; + I32 prevval = *list->pval; + *list->pval = e->val; + + if( list->cb && prevval != *list->pval ) + list->cb( list ); + } + list->held = 1; + } else { + list->held = 0; + } +} + +GUI_LIST* gui_list( I32 x, I32 y, I32 w, I32 h, const char* title, LIST<GUI_LIST_ENTRY>* plist, I32* pval ) { + if( !gui_check_target() ) return 0; + + GUI_LIST* list = new GUI_LIST; + list->x = x; + list->y = y; + list->w = w; + list->h = h; + + list->xbound = w; + list->ybound = h + LIST_TITLE_OFFSET; + + list->cb = 0; + list->pval = pval; + list->plist = plist; + list->draw_fn = gui_list_draw_fn; + list->input_fn = gui_list_input_fn; + strcpy( list->name, title ); + + list->parent = gui_get_view(); + + gui_get_view()->children.push( list ); + return list; +} + +GUI_LIST_ENTRY* gui_list_get_selected( GUI_LIST* list ) { + for( U32 i = 0; i < list->plist->size; ++i ) { + GUI_LIST_ENTRY* e = &list->plist->data[i]; + if( e->val == *list->pval ) + return e; + } + + return 0; +} diff --git a/src/gui/textbox.cpp b/src/gui/textbox.cpp new file mode 100644 index 0000000..688450b --- /dev/null +++ b/src/gui/textbox.cpp @@ -0,0 +1,246 @@ +#include "base.h" +#include "../util/input.h" + +#include <SDL_keycode.h> + +const U32 TEXTBOX_TITLE_OFFSET = 15; + +U8 is_printable_char( U8 key ) { + return key >= 32 && key < 127; +} + +void gui_textbox_draw_newline_str( GUI_TEXTBOX* tb, I32 x, I32 y, CLR clr, I32* endposx, I32* endposy ) { + char linebuf[GUI_TEXTBOX_MAX]; + char* start = tb->value; + I32 tw, th; + + for( char* c = start; !!*c; ++c ) { + if( *c == '\n' ) { + U32 len = (U32)( c - start ); + memcpy( linebuf, start, len ); + linebuf[len] = 0; + + gui_draw_str( x, y, ALIGN_L, FNT_JPN12, clr, linebuf ); + gui_draw_get_str_bounds( &tw, &th, FNT_JPN12, linebuf ); + y += th + 1; + + *endposx = x + tw; + *endposy = y + th; + + if( *(c+1) == 0 ) + break; + + start = ++c; + } + else if( (*c+1) == 0 ) { + U32 len = (U32)( c - start ); + memcpy( linebuf, start, len ); + linebuf[len] = 0; + + gui_draw_str( x, y, ALIGN_L, FNT_JPN12, clr, linebuf ); + gui_draw_get_str_bounds( &tw, &th, FNT_JPN12, linebuf ); + y += th + 1; + *endposx = x + tw; + *endposy = y + th; + + break; + } + } +} + +void gui_textbox_draw_blinker( GUI_TEXTBOX* tb, I32 x, I32 y, CLR clr ) { + if( !tb->active ) + return; + if( gui_time() - tb->blinktime > 1.0f ) { + tb->blink = !tb->blink; + tb->blinktime = gui_time(); + } + + if( tb->blink ) + gui_draw_str( x, y, ALIGN_L, FNT_JPN12, clr, "|" ); +} + +void gui_textbox_draw_fn( void* ptr ) { + GUI_TEXTBOX* tb = (GUI_TEXTBOX*)ptr; + + I32 x = gui_relx( tb ); + I32 y = gui_rely( tb ); + + gui_draw_str( x, y, ALIGN_L, FNT_JPN12, ui_clr.txt, tb->name ); + y += TEXTBOX_TITLE_OFFSET; + + CLR col = gui_is_fg_window( tb )? ui_clr.border : ui_clr.border_inactive; + + gui_draw_frect( x, y, tb->w, tb->h, col ); + gui_draw_frect( x+1, y+1, tb->w-2, tb->h-2, ui_clr.bg_sec ); + + CLR clr = tb->active? ui_clr.txt : ui_clr.txt_inactive; + I32 ypos = y + 3; + I32 xpos = x + 2; + + I32 endposx, endposy; + + if( tb->allow_newl ) { + gui_textbox_draw_newline_str( tb, xpos, ypos, clr, &endposx, &endposy ); + } else { + gui_draw_str( xpos, ypos, ALIGN_L, FNT_JPN12, clr, tb->value ); + + I32 tw, th; + gui_draw_get_str_bounds( &tw, &th, FNT_JPN12, tb->value ); + endposx = xpos + tw; + endposy = ypos; + } + + gui_textbox_draw_blinker( tb, endposx, endposy, clr ); +} + +void gui_textbox_handle_mouse( GUI_TEXTBOX* tb ) { + I32 x = gui_relx( tb ); + I32 y = gui_rely( tb ) + TEXTBOX_TITLE_OFFSET; + I32 w = tb->w; + I32 h = tb->h; + + U8 m1 = gui_mbutton_down( 0 ); + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + U8 inbounds = mx >= x && mx <= x + w && my >= y && my <= y + h; + + if( m1 ) { + if( inbounds ) + tb->m1held = 1; + else + tb->active = 0; + } else { + if( inbounds && tb->m1held ) { + tb->active = 1; + tb->blink = 1; + tb->blinktime = gui_time(); + } + tb->m1held = 0; + } +} + +void gui_textbox_loop_keys( GUI_TEXTBOX* tb ) { + U32 len = strlen( tb->value ); + + for( U32 i = 0; i < 0xff; ++i ) { + U8 key = tb->keys[i]; + U8 prev = tb->prevkeys[i]; + + if( !key && prev && i == tb->heldkey ) + tb->heldkey = 0; + else if( key && !prev ) { + if( is_printable_char( i ) && len < tb->maxlen - 1 ) { + if( tb->keys[SDLK_LSHIFT & 0xff] || tb->keys[SDLK_RSHIFT & 0xff] ) + tb->value[len] = (char)i + ('a' - 'A'); + else + tb->value[len] = (char)i; + tb->value[++len] = 0; + } + else if( i == '\b' ) { + if( len > 0 ) { + tb->value[--len] = 0; + } + } + else if( i == '\r' ) { + if( tb->allow_newl ) { + tb->value[len] = '\n'; + tb->value[++len] = 0; + } + else { + tb->active = 0; + } + } else { + if( i == '\x1b' ) + tb->active = 0; + if( tb->cb ) + tb->cb( tb ); + continue; + } + + tb->heldkey = i; + tb->heldtime = gui_time(); + if( tb->cb ) + tb->cb( tb ); + } + } +} + +void gui_textbox_handle_repeat( GUI_TEXTBOX* tb ) { + if( !tb->heldkey ) + return; + + F32 delta = gui_time() - tb->heldtime; + if( delta <= 0.75f ) + return; + + if( gui_time() - tb->repeattime > 0.05f ) { + U32 len = strlen( tb->value ); + if( is_printable_char( tb->heldkey ) && len < (tb->maxlen - 1) ) { + if( tb->keys[SDLK_LSHIFT & 0xff] || tb->keys[SDLK_RSHIFT & 0xff] ) + tb->value[len] = tb->heldkey + ('a' - 'A'); + else + tb->value[len] = tb->heldkey; + tb->value[++len] = 0; + if( tb->cb ) + tb->cb( tb ); + } + else if( tb->heldkey == '\b' && len > 0 ) { + tb->value[--len] = 0; + if( tb->cb ) + tb->cb( tb ); + } + + tb->repeattime = gui_time(); + } +} + +void gui_textbox_handle_keyboard( GUI_TEXTBOX* tb ) { + if( !tb->active ) + return; + + memcpy( tb->prevkeys, tb->keys, 0xff ); + memcpy( tb->keys, input.keys, 0xff ); + + gui_textbox_loop_keys( tb ); + gui_textbox_handle_repeat( tb ); +} + +void gui_textbox_input_fn( void* ptr ) { + GUI_TEXTBOX* tb = (GUI_TEXTBOX*)ptr; + + gui_textbox_handle_mouse( tb ); + gui_textbox_handle_keyboard( tb ); +} + +GUI_TEXTBOX* gui_textbox( I32 x, I32 y, I32 w, I32 h, const char* title, U32 maxlen, U8 allow_newl ) { + GUI_TEXTBOX* tb = new GUI_TEXTBOX; + + tb->x = x; + tb->y = y; + tb->w = w; + tb->h = h; + tb->xbound = tb->w; + tb->ybound = tb->h + TEXTBOX_TITLE_OFFSET; + tb->parent = _gui.cur_view; + tb->draw_fn = gui_textbox_draw_fn; + tb->input_fn = gui_textbox_input_fn; + strcpy( tb->name, title ); + + tb->cb = 0; + tb->active = 0; + tb->m1held = 0; + tb->heldkey = 0; + tb->heldtime = 0; + tb->repeattime = 0; + tb->maxlen = maxlen; + tb->allow_newl = allow_newl; + memset( tb->value, 0, GUI_TEXTBOX_MAX ); + memset( tb->keys, 0, 0xff ); + memset( tb->prevkeys, 0, 0xff ); + + gui_get_view()->children.push( tb ); + + return tb; +} diff --git a/src/gui/vectorinput.cpp b/src/gui/vectorinput.cpp new file mode 100644 index 0000000..9ebb0fa --- /dev/null +++ b/src/gui/vectorinput.cpp @@ -0,0 +1,120 @@ +#include "base.h" + +const I32 VECTORINPUT_TITLE_OFFSET = 15; + +void gui_vectorinput_draw_fn( void* ptr ) { + GUI_VECTORINPUT* input = (GUI_VECTORINPUT*)ptr; + + I32 x = gui_relx( input ); + I32 y = gui_rely( input ); + + gui_draw_str( x, y, ALIGN_L, FNT_JPN12, ui_clr.txt, input->name ); + + GUI_VIEW* inputview = input->inputview; + inputview->draw_fn( inputview ); +} + +void gui_vectorinput_input_fn( void* ptr ) { + GUI_VECTORINPUT* input = (GUI_VECTORINPUT*)ptr; + + GUI_VIEW* inputview = input->inputview; + inputview->input_fn( inputview ); +} + +void gui_vectorinput_child_cb( void* ptr ) { + GUI_FLOATINPUT* child = (GUI_FLOATINPUT*)ptr; + // slider -> child view -> vectorinput + GUI_VECTORINPUT* parent = (GUI_VECTORINPUT*)child->parent->parent; + if( parent->cb ) + parent->cb( parent ); +} + +void __gui_internal_vectorinput_init( + GUI_VECTORINPUT *input, + I32 x, I32 y, I32 w, + const char *title, + F32 *pval, + U32 valc, + F32 min, + F32 max, + F32 step, + const char *letters, + const char *printfmt +) { + input->x = x; + input->y = y; + input->w = w; + input->h = 20; + strcpy( input->name, title ); + input->xbound = input->w; + input->ybound = input->h + VECTORINPUT_TITLE_OFFSET; + input->draw_fn = gui_vectorinput_draw_fn; + input->input_fn = gui_vectorinput_input_fn; + + GUI_VIEW* parent = gui_get_view(); + parent->children.push( input ); + input->parent = parent; + + gui_set_view( input ); + + input->cb = 0; + input->pval = pval; + input->letters = letters; + input->inputview = gui_view( 0, VECTORINPUT_TITLE_OFFSET, w, input->h ); + input->inputview->parent = input; + I32 wdiv = (input->w - 5 * (valc - 1)) / valc; + for( I32 i = 0; i < valc; ++i ) { + char letter[2] = { input->letters[i], 0 }; + GUI_FLOATINPUT* slider = gui_floatinput( + (wdiv + 5) * i, + 0, + wdiv, + letter, + &pval[i], + min, + max, + step, + printfmt + ); + + slider->cb = gui_vectorinput_child_cb; + } + + gui_set_view( parent ); +} + +GUI_VECTORINPUT* gui_vectorinput( + I32 x, + I32 y, + I32 w, + const char* title, + F32* pval, + U32 valc, + F32 min, + F32 max, + F32 step, + const char* letters, + const char* printfmt +) { + if( !gui_check_target() ) return 0; + if( valc > 25 ) { + dlog( "gui_vectorinput() : have you lost your mind?" ); + return 0; + } + + GUI_VECTORINPUT* input = new GUI_VECTORINPUT; + __gui_internal_vectorinput_init( + input, + x, y, w, + title, + pval, + valc, + min, + max, + step, + letters, + printfmt + ); + + return input; +} diff --git a/src/gui/view.cpp b/src/gui/view.cpp new file mode 100644 index 0000000..da470e7 --- /dev/null +++ b/src/gui/view.cpp @@ -0,0 +1,59 @@ +#include "base.h" + +void gui_view_draw_fn( void* ptr ) { + GUI_VIEW* view = (GUI_VIEW*)ptr; + + I32 x = gui_relx( view ); + I32 y = gui_rely( view ); + + gui_draw_push_clip( x, y, view->w, view->h ); + view->children.each( fn( GUI_BASE** childptr ) { + GUI_BASE* child = *childptr; + if( !child->enabled ) return; + + if( child->draw_fn ) child->draw_fn( child ); + else dlog( "gui_view_draw_fn(): child %p no draw_fn\n", child ); + } ); + gui_draw_pop_clip(); +} + +void gui_view_input_fn( void* ptr ) { + GUI_VIEW* view = (GUI_VIEW*)ptr; + + if( view->initheld ) { + U8 m1 = gui_mbutton_down( 0 ); + if( m1 ) + return; + + view->initheld = 0; + } + + gui_base_input_fn( view ); +} + +GUI_VIEW* gui_view( I32 x, I32 y, I32 w, I32 h ) { + GUI_VIEW* view = new GUI_VIEW; + view->x = x; + view->y = y; + view->xbound = view->w = w; + view->ybound = view->h = h; + strcpy( view->name, "BASE_VIEW" ); + + view->draw_fn = gui_view_draw_fn; + view->input_fn = gui_view_input_fn; + + view->initheld = 1; + + GUI_VIEW* curview = gui_get_view(); + if( !curview ) { + view->parent = gui_get_window(); + gui_get_window()->children.push( view ); + } + else { + view->parent = curview; + curview->children.push( view ); + } + + gui_set_view( view ); + return view; +} diff --git a/src/gui/window.cpp b/src/gui/window.cpp new file mode 100644 index 0000000..0183d8c --- /dev/null +++ b/src/gui/window.cpp @@ -0,0 +1,113 @@ +#include "base.h" + +#include "../render/gl.h" + +void gui_window_draw_fn( void* ptr ) { + GUI_WINDOW* wnd = (GUI_WINDOW*)ptr; + + CLR clr = gui_is_fg_window( wnd )? ui_clr.border : ui_clr.border_inactive; + gui_draw_frect( wnd->x, wnd->y, wnd->w, wnd->h, clr ); + gui_draw_frect( wnd->x + 1, wnd->y + 1, wnd->w - 2, wnd->h - 2, ui_clr.bg ); + + wnd->children.each( fn( GUI_BASE** ptr ) { + GUI_BASE* it = *ptr; + if( it->draw_fn ) it->draw_fn( it ); + else dlog( "window_draw_fn() : child %p no draw_fn", it ); + } ); +} + +GUI_WINDOW* gui_window( I32 x, I32 y, I32 w, I32 h ) { + GUI_WINDOW* wnd = new GUI_WINDOW; + wnd->x = x; + wnd->y = y; + wnd->xbound = wnd->w = w; + wnd->ybound = wnd->h = h; + wnd->draw_fn = gui_window_draw_fn; + strcpy( wnd->name, "BASE_WINDOW" ); + + _gui.windows.push( wnd ); + gui_set_window( wnd ); + gui_set_view( 0 ); + + gui_view( 1, 1, w - 2, h - 2 ); + return wnd; +} + +GUI_WINDOW* gui_window( I32 w, I32 h ) { + return gui_window( 0, 0, w, h ); +} + +void gui_title_draw_fn( void* ptr ) { + GUI_TITLE* t = (GUI_TITLE*)ptr; + + I32 relx = gui_relx( t ); + I32 rely = gui_rely( t ); + gui_draw_str( relx + 5, rely + 3, 0, FNT_JPN12, ui_clr.txt, t->name ); +} + +void gui_title_input_fn( void* ptr ) { + GUI_TITLE* t = (GUI_TITLE*)ptr; + GUI_WINDOW* w = gui_get_parent_wnd( t ); + + if( w->locked ) + return; + + I32 x = gui_relx( t ); + I32 y = gui_rely( t ); + + U8 m1 = gui_mbutton_down( 0 ); + I32 mx, my; + gui_cursor_pos( &mx, &my ); + + if( !m1 ) { + t->held = 0; + return; + } + + if( !t->held && + mx >= x && mx <= x + t->w && + my >= y && my <= y + t->h + ) { + I32 wx = w->x; + I32 wy = w->y; + + I32 moffx = mx - wx; + I32 moffy = my - wy; + + t->xoff = moffx; + t->yoff = moffy; + + t->held = 1; + } + else if( t->held ) { + w->x = mx - t->xoff; + w->y = my - t->yoff; + + if( w->x > _gui.gl2d->gl->canvas_size[0] - 5 ) + w->x = _gui.gl2d->gl->canvas_size[0] - 5; + if( w->x + w->w < 5 ) + w->x = 5 - w->w; + if( w->y > _gui.gl2d->gl->canvas_size[1] - 5 ) + w->y = _gui.gl2d->gl->canvas_size[1] - 5; + if( w->y + w->h < 5 ) + w->y = 5 - w->h; + } +} + +GUI_TITLE* gui_title( const char* title ) { + if( !gui_check_target() ) return 0; + + GUI_TITLE* t = new GUI_TITLE; + t->parent = gui_get_view(); + t->x = 0; + t->y = 0; + t->h = 25; + t->w = t->parent->w; + t->draw_fn = gui_title_draw_fn; + t->input_fn = gui_title_input_fn; + strcpy( t->name, title ); + + gui_get_view()->children.push( t ); + + return t; +} diff --git a/src/main.cpp b/src/main.cpp new file mode 100644 index 0000000..28886fa --- /dev/null +++ b/src/main.cpp @@ -0,0 +1,49 @@ +#include "game.h" +#include "game/assets.h" + +// disable new_delete_type_mismatch asan warning +#ifndef __has_feature + #define __has_feature(feature) 0 +#endif +#if __has_feature(address_sanitizer) || defined(__SANITIZE_ADDRESS__) + #ifdef __cplusplus + extern "C" + #endif + const char *__asan_default_options() { return "new_delete_type_mismatch=0"; } +#endif + +I32* canvas; +GAME_DATA* game; +GL_DATA* gl; + +void loop() { + static U8 init = 0; + + if( !init || !gl ) { + gl = gl_create( canvas ); + memcpy( gl->canvas_size, canvas, sizeof(I32) * 2 ); + + gl_gen_buffers( gl ); + game = game_init( gl ); + + init = 1; + return; + } + + game_main_loop( game ); +} + +int main() { + if( !canvas ) + canvas = (I32*)malloc( sizeof(I32) * 2 ); + + canvas[0] = 800; + canvas[1] = 600; + + while( true ) { + loop(); + } + + game_destroy( game ); + return 0; +} diff --git a/src/render/gl.cpp b/src/render/gl.cpp new file mode 100644 index 0000000..915722e --- /dev/null +++ b/src/render/gl.cpp @@ -0,0 +1,394 @@ +#include "SDL_video.h" +#include <unistd.h> +#define STB_IMAGE_IMPLEMENTATION +#include "../util/stb_image.h" + +#include "gl.h" +#include "gl_2d_font.h" +#include "../util.h" + +I32 SAMPLER_INDICES[255]; + +GL_DATA* gl_inst; + +GL_DATA* gl_instance() { + return gl_inst; +} + +GL_DATA* gl_create( I32* _canvas ) { + if( gl_inst ) { + dlog( "gl_create() : fatal: gl instance already exists\n" ); + abort(); + } + + if( !font_mutex.align ) + thread_mutex_init( &font_mutex ); + + if( !!SDL_Init( SDL_INIT_VIDEO | SDL_VIDEO_OPENGL ) ) { + dlog( "gl_create() could not init SDL: %s\n", SDL_GetError() ); + return 0; + } + + SDL_GL_SetAttribute( SDL_GL_CONTEXT_PROFILE_MASK, SDL_GL_CONTEXT_PROFILE_ES ); + SDL_GL_SetAttribute( SDL_GL_CONTEXT_MAJOR_VERSION, 3 ); + SDL_GL_SetAttribute( SDL_GL_CONTEXT_MINOR_VERSION, 2 ); + SDL_GL_SetAttribute( SDL_GL_DOUBLEBUFFER, 1 ); + SDL_GL_SetAttribute( SDL_GL_ACCELERATED_VISUAL, 1 ); + SDL_GL_SetAttribute( SDL_GL_DEPTH_SIZE, 24 ); + SDL_GL_SetAttribute( SDL_GL_RED_SIZE, 8 ); + SDL_GL_SetAttribute( SDL_GL_GREEN_SIZE, 8 ); + SDL_GL_SetAttribute( SDL_GL_BLUE_SIZE, 8 ); + SDL_GL_SetAttribute( SDL_GL_ALPHA_SIZE, 8 ); + + GL_DATA* gl = new GL_DATA; + gl->window = SDL_CreateWindow( + "game", + 0, + 0, + _canvas[0], + _canvas[1], + SDL_WINDOW_OPENGL | SDL_WINDOW_SHOWN + ); + + if( !gl->window ) { + dlog( "gl_init() could not create window: %s\n", SDL_GetError() ); + return 0; + } + + U32 renderer_flags = SDL_RENDERER_ACCELERATED | SDL_RENDERER_TARGETTEXTURE; + gl->renderer = SDL_CreateRenderer( gl->window, -1, renderer_flags ); + if( !gl->renderer ) { + dlog( "gl_init() could not create renderer: %s\n", SDL_GetError() ); + return 0; + } + + gl->ctx = SDL_GL_CreateContext( gl->window ); + if( !gl->ctx ) { + dlog( "gl_init() could not create context: %s\n", SDL_GetError() ); + return 0; + } + + glGetIntegerv( GL_MAX_TEXTURE_IMAGE_UNITS, &gl->shader_texture_limit ); + if( gl->shader_texture_limit > 255 ) + gl->shader_texture_limit = 255; + for( U32 i = 0; i < 255; ++i ) + SAMPLER_INDICES[i] = i; + + gl->programs.clear(); + memcpy( gl->canvas_size, _canvas, sizeof(I32) * 2 ); + gl->clip_start = { 0, 0 }; + gl->clip_dim = { (F32)gl->canvas_size[0], (F32)gl->canvas_size[1] }; + + gl_inst = gl; + return gl; +} + +void gl_gen_buffers( GL_DATA* gl ) { + glGenBuffers( 1, &gl->vbuffer ); + + glBindBuffer( GL_ARRAY_BUFFER, gl->vbuffer ); + glBufferData( GL_ARRAY_BUFFER, 8192, 0, GL_STATIC_DRAW ); + + glBindBuffer( GL_ARRAY_BUFFER, 0 ); +} + +void gl_destroy( GL_DATA *gl ) { + gl->programs.each( fn( GL_SHADER_PROGRAM** it ) { gl_program_destroy( gl, *it ); } ); + gl->fonts.each( fn( GL_FONT** it ) { gl_font_destroy( gl, *it ); } ); + gl->textures.each( fn( GL_TEX2D** it ) { gl_texture_destroy( gl, *it ); } ); + + if( gl->ctx ) + SDL_GL_DeleteContext( gl->ctx ); + if( gl->window ) + SDL_DestroyWindow( gl->window ); + + free( gl ); +} + +void gl_update_window( GL_DATA* gl, I32* size ) { + if( !gl->window ) + return; + SDL_SetWindowSize( gl->window, size[0], size[1] ); + SDL_SetWindowPosition( gl->window, SDL_WINDOWPOS_CENTERED, SDL_WINDOWPOS_CENTERED ); + + gl->canvas_size[0] = size[0]; + gl->canvas_size[1] = size[1]; + + gl->programs.each( fn( GL_SHADER_PROGRAM** it ) { + F32 screen_ratio[] = { + 2.f / (F32)gl->canvas_size[0], + 2.f / (F32)gl->canvas_size[1], + 1.f, + 1.f + }; + + glUseProgram( (*it)->id ); + I32 ratio_location = glGetUniformLocation( (*it)->id, "g_screenratio" ); + glUniform4fv( ratio_location, 1, &screen_ratio[0] ); + } ); + + glViewport( 0, 0, gl->canvas_size[0], gl->canvas_size[1] ); + glUseProgram( 0 ); +} + +STAT gl_shader_compile( GL_DATA* gl, GL_SHADER* shader ) { + static char* log_buf = 0; + if( !log_buf ) + log_buf = (char*)malloc( 8192 ); + + I32 res; + shader->id = glCreateShader( (GLenum)shader->type ); + glShaderSource( shader->id, 1, &shader->code, 0 ); + glCompileShader( shader->id ); + glGetShaderiv( shader->id, GL_COMPILE_STATUS, &res ); + + if( !res ) { + glGetShaderInfoLog( shader->id, 8192, 0, log_buf ); + dlog( "gl_shader_compile() : error compiling shader %s. log: \n%s\n%s", shader->name, log_buf, shader->code ); + + glDeleteShader( shader->id ); + return STAT_ERR; + } + + shader->compiled = 1; + return STAT_OK; +} + +void gl_shader_destroy( GL_DATA* gl, GL_SHADER* shader ) { + if( shader->code ) + free( (void*)shader->code ); +} + +GL_SHADER_PROGRAM* gl_program_create( GL_DATA* gl, const char* name, U32 size ) { + GL_SHADER_PROGRAM* program = (GL_SHADER_PROGRAM*)malloc( size ); + char shader_string[256]; + char* shader_code; + + sprintf( shader_string, "../assets/shaders/%s.vsh", name ); + shader_code = (char*)file_read( shader_string ); + if( !shader_code ) { + dlog( "gl_program_create() : could not read shader file %s\n", shader_string ); + return 0; + } + + program->vsh.name = name; + program->vsh.type = GL_VERTEX_SHADER; + program->vsh.code = shader_code; + + sprintf( shader_string, "../assets/shaders/%s.fsh", name ); + shader_code = (char*)file_read( shader_string ); + if( !shader_code ) { + dlog( "gl_program_create() : could not read shader file %s\n", shader_string ); + return 0; + } + + program->fsh.name = name; + program->fsh.type = GL_FRAGMENT_SHADER; + program->fsh.code = shader_code; + + program->name = name; + program->gl = gl; + + gl->programs.push( program ); + return program; +} + +STAT gl_program_compile( GL_DATA* gl, GL_SHADER_PROGRAM* program ) { + static char* log_buf = 0; + if( !log_buf ) + log_buf = (char*)malloc( 8192 ); + + program->id = glCreateProgram(); + if( !OK( gl_shader_compile( gl, &program->fsh ) ) + || !OK( gl_shader_compile( gl, &program->vsh ) ) ) + return STAT_ERR; + + glAttachShader( program->id, program->fsh.id ); + glAttachShader( program->id, program->vsh.id ); + + I32 status; + glLinkProgram( program->id ); + glGetProgramiv( program->id, GL_LINK_STATUS, &status ); + if( !status ) { + glGetProgramInfoLog( program->id, 8192, 0, log_buf ); + dlog( "gl_program_compile() : error compiling program %s. log: \n%s\n", + program->name, + log_buf + ); + + return STAT_ERR; + } + + glGenBuffers( 1, &program->vbuffer ); + return STAT_OK; +} + +void gl_program_destroy( GL_DATA* gl, GL_SHADER_PROGRAM* program ) { + if( program->fsh.compiled ) + gl_shader_destroy( gl, &program->fsh ); + if( program->vsh.compiled ) + gl_shader_destroy( gl, &program->vsh ); + + I32 idx = gl->programs.idx_of( program ); + if( idx != -1 ) + gl->programs.erase( idx ); +} + +GL_TEX2D* gl_texture_from_file( GL_DATA* gl, const char* name ) { + char filename[256]; + sprintf( filename, "../assets/%s", name ); + + I32 w, h, n; + U8* data = stbi_load( filename, &w, &h, &n, STBI_rgb_alpha ); + if( !data ) { + dlog( "gl_texture_create() : could not load image %s\n", filename ); + return 0; + } + + GL_TEX2D* tex = gl_texture_create( gl, filename ); + glBindTexture( GL_TEXTURE_2D, tex->id ); + glTexParameteri( GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_REPEAT ); + glTexParameteri( GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_REPEAT ); + glTexParameteri( GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR ); + glTexParameteri( GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR ); + + glTexImage2D( GL_TEXTURE_2D, 0, GL_RGBA, w, h, 0, GL_RGBA, GL_UNSIGNED_BYTE, data ); + glGenerateMipmap( GL_TEXTURE_2D ); + glBindTexture( GL_TEXTURE_2D, 0 ); + + tex->width = w; + tex->height = h; + + stbi_image_free( data ); + + return tex; +} + +GL_TEX2D* gl_texture_from_bitmap( GL_DATA* gl, const char* name, U8* bitmap, U32 width, U32 height ) { + GL_TEX2D* tex = gl_texture_create( gl, name ); + glBindTexture( GL_TEXTURE_2D, tex->id ); + glTexParameteri( GL_TEXTURE_2D, GL_TEXTURE_WRAP_S, GL_REPEAT ); + glTexParameteri( GL_TEXTURE_2D, GL_TEXTURE_WRAP_T, GL_REPEAT ); + glTexParameteri( GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR ); + glTexParameteri( GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR ); + + tex->width = width; + tex->height = height; + + glTexImage2D( + GL_TEXTURE_2D, + 0, GL_RGBA, + (I32)width, (I32)height, 0, + GL_RGBA, GL_UNSIGNED_BYTE, bitmap + ); + + glBindTexture( GL_TEXTURE_2D, 0 ); + + return tex; +} + + +GL_TEX2D* gl_texture_create( GL_DATA* gl, const char* name ) { + GL_TEX2D* tex = (GL_TEX2D*)malloc( sizeof(GL_TEX2D) ); + strcpy( tex->name, name ); + glGenTextures( 1, &tex->id ); + + gl->textures.push( tex ); + return tex; +} + +void gl_texture_destroy( GL_DATA* gl, GL_TEX2D* tex ) { + glDeleteTextures( 1, &tex->id ); + I32 idx = gl->textures.idx_of( tex ); + if( idx != -1 ) + gl->textures.erase( idx ); + + free( tex ); +} + +STAT gl_beginframe( GL_DATA* gl ) { + SDL_SetRenderTarget( gl->renderer, 0 ); + SDL_SetRenderDrawColor( gl->renderer, 0, 0, 0, 255 ); + SDL_RenderClear( gl->renderer ); + + glClear( GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT ); + gl->last_tick = u_tick(); + return STAT_OK; +} + +STAT gl_endframe( GL_DATA* gl ) { + SDL_GL_SwapWindow( gl->window ); + + // 1000 fps max be real + while( true ) { + U64 diff = u_tick() - gl->last_tick; + if( diff < 10 ) + usleep( 10 ); + else break; + } + + gl->frametime = (F32)(u_tick() - gl->last_tick) / 10000.0f; + gl->fps = 1.0f / gl->frametime; + + input_frame_end(); + + SDL_Event e; + while( SDL_PollEvent( &e ) ) { + input_on_event( &e ); + + switch( e.type ) { + case SDL_QUIT: + return STAT_BREAK; + case SDL_KEYDOWN: + return e.key.keysym.sym == SDLK_ESCAPE ? STAT_BREAK : STAT_OK; + default: break; + } + } + + return STAT_OK; +} + +void gl_set_clip( GL_DATA* gl, VEC2 start, VEC2 dim ) { + gl->clip_start = start; + gl->clip_dim = dim; + + glEnable( GL_SCISSOR_TEST ); + glScissor( start.x, gl->canvas_size[1] - (I32)(dim.y + start.y), dim.x, dim.y ); +} + +void gl_get_clip( GL_DATA* gl, VEC2* start, VEC2* dim ) { + *start = gl->clip_start; + *dim = gl->clip_dim; +} + +void gl_reset_clip( GL_DATA* gl ) { + glDisable( GL_SCISSOR_TEST ); + gl->clip_start = {}; + gl->clip_dim = { (F32)gl->canvas_size[0], (F32)gl->canvas_size[1] }; +} + +void gl_set_viewport( GL_DATA* gl, VEC2 start, VEC2 dim ) { + I32 vpy = (I32)gl->canvas_size[1] - start.y - dim.y; + glViewport( (I32)start.x, vpy, (I32)dim.x, (I32)dim.y ); + + gl->programs.each( fn( GL_SHADER_PROGRAM** it ) { + F32 screen_ratio[] = { + 2.f / (F32)dim.x, + 2.f / (F32)dim.y, + 1.f, + 1.f + }; + + glUseProgram( (*it)->id ); + I32 ratio_location = glGetUniformLocation( (*it)->id, "g_screenratio" ); + glUniform4fv( ratio_location, 1, &screen_ratio[0] ); + } ); + + gl->viewport_start = start; + gl->viewport_dim = dim; +} + +void gl_get_viewport( GL_DATA* gl, VEC2* start, VEC2* dim ) { + *start = gl->viewport_start; + *dim = gl->viewport_dim; +} diff --git a/src/render/gl.h b/src/render/gl.h new file mode 100644 index 0000000..cf2b5dd --- /dev/null +++ b/src/render/gl.h @@ -0,0 +1,117 @@ +#pragma once + +#include <SDL.h> +#include <GLES2/gl2.h> + +#include "../util.h" +#include "../util/matrix.h" + +typedef struct GL_FONT *PGL_FONT; +typedef struct GL_DATA *PGL_DATA; + +struct VERTEX { + VEC2 pos; + VEC2 uv; + CLR clr; + U8 sampler; +}; + +struct GL_SHADER_DEF { + const char* name; + const char* code; + I32 type; + U32 id; + U8 compiled; +}; + +struct GL_SHADER_PROGRAM { + const char* name; + GL_SHADER_DEF fsh; + GL_SHADER_DEF vsh; + U32 id; + U32 vbuffer; + + GL_DATA* gl; +}; + +struct GL_TEX2D { + GLuint id; + char name[256]; + U32 width; + U32 height; + U32 channels; + U8* data; +}; + +typedef struct GL_DATA { + SDL_Window* window; + SDL_GLContext ctx; + SDL_Renderer* renderer; + + I32 canvas_size[2]; + + LIST<GL_SHADER_PROGRAM*> programs; + LIST<GL_TEX2D*> textures; + LIST<GL_FONT*> fonts; + + I32 shader_texture_limit; + U64 last_tick; + F32 frametime; + F32 fps; + + GLuint vbuffer; + + VEC2 clip_start; + VEC2 clip_dim; + + VEC2 viewport_start; + VEC2 viewport_dim; + + MAT4* proj_matrix; +} *PGL_DATA; + +GL_DATA* gl_instance(); +GL_DATA* gl_create( I32* _canvas ); +void gl_destroy( GL_DATA* gl ); +void gl_gen_buffers( GL_DATA* gl ); +STAT gl_beginframe( GL_DATA* gl ); +STAT gl_endframe( GL_DATA* gl ); +STAT gl_shader_compile( GL_DATA* gl, GL_SHADER_DEF* shader ); +void gl_shader_destroy( GL_DATA* gl, GL_SHADER_DEF* shader ); +GL_SHADER_PROGRAM* gl_program_create( GL_DATA* gl, const char* name, U32 size = sizeof(GL_SHADER_PROGRAM) ); +STAT gl_program_compile( GL_DATA* gl, GL_SHADER_PROGRAM* program ); +void gl_program_destroy( GL_DATA* gl, GL_SHADER_PROGRAM* program ); +GL_TEX2D* gl_texture_create( GL_DATA* gl, const char* name ); +GL_TEX2D* gl_texture_from_file( GL_DATA* gl, const char* name ); +GL_TEX2D* gl_texture_from_bitmap( GL_DATA* gl, const char* name, U8* bitmap, U32 width, U32 height ); +void gl_texture_destroy( GL_DATA* gl, GL_TEX2D* tex ); +void gl_update_window( GL_DATA* gl, I32* size ); +void gl_set_clip( GL_DATA* gl, VEC2 start, VEC2 dim ); +void gl_get_clip( GL_DATA* gl, VEC2* start, VEC2* dim ); +void gl_set_viewport( GL_DATA* gl, VEC2 start, VEC2 dim ); +void gl_get_viewport( GL_DATA* gl, VEC2* start, VEC2* dim ); +void gl_reset_clip( GL_DATA* gl ); + +// special sampler id for no texture +const U8 SAMPLER_ID_NONE = 255; +extern I32 SAMPLER_INDICES[255]; + +template <typename VERTEX> +LIST<VERTEX> triangle_fan_to_list( VERTEX* vertices, U32 vert_count ) { + LIST<VERTEX> ret{}; + if( vert_count < 3 ) { + return ret; + } + + ret.reserve( (vert_count - 2) * 3 ); + VERTEX* start = &vertices[0]; + VERTEX* last = &vertices[1]; + for( U32 i = 2; i < vert_count; ++i ) { + ret.push( *start ); + ret.push( *last ); + ret.push( vertices[i] ); + last = &vertices[i]; + } + + return ret; +} diff --git a/src/render/gl_2d.cpp b/src/render/gl_2d.cpp new file mode 100644 index 0000000..44e7e00 --- /dev/null +++ b/src/render/gl_2d.cpp @@ -0,0 +1,314 @@ +#include "gl_2d.h" +#include "../util.h" + +GL_PROGRAM* gl_2d_init( GL_DATA* gl, VEC2 screensize, const char* shadername ) { + GL_PROGRAM* program = gl_program_create( gl, shadername ); + if( !OK( gl_program_compile( gl, program ) ) ) + dlog( "gl_2d_init() : error compiling shader %s\n", shadername ); + + F32 screen_ratio[] = { + 2.f / screensize.x, + 2.f / screensize.y, + 1.f, + 1.f + }; + + glUseProgram( program->id ); + glEnable( GL_BLEND ); + glBlendFunc( GL_SRC_ALPHA, GL_ONE_MINUS_SRC_ALPHA ); + + I32 ratio_location = glGetUniformLocation( program->id, "g_screenratio" ); + glUniform4fv( ratio_location, 1, &screen_ratio[0] ); + + I32 samplers_location = glGetUniformLocation( program->id, "g_samplers" ); + glUniform1iv( samplers_location, 255, SAMPLER_INDICES ); + + return program; +} + +void gl_2d_line( GL_PROGRAM* gl2d, VEC2 start, VEC2 end, CLR col ) { + static const U16 order[] = { 0, 1 }; + + glUseProgram( gl2d->id ); + + VERTEX vertices[] = { + { .pos = { start.x, start.y }, .clr = col }, + { .pos = { end.x, end.y }, .clr = col }, + }; + + glBindBuffer( GL_ARRAY_BUFFER, gl2d->gl->vbuffer ); + glBufferData( GL_ARRAY_BUFFER, sizeof(vertices), &vertices[0], GL_DYNAMIC_DRAW ); + I32 position = glGetAttribLocation( gl2d->id, "in_pos" ); + glEnableVertexAttribArray( position ); + glVertexAttribPointer( position, 2, GL_FLOAT, 0, sizeof(VERTEX), 0 ); + I32 color = glGetAttribLocation( gl2d->id, "in_col" ); + glEnableVertexAttribArray( color ); + glVertexAttribPointer( color, 4, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->clr ); + + glBindBuffer( GL_ARRAY_BUFFER, 0 ); + glDrawElements( GL_LINES, 2, GL_UNSIGNED_SHORT, order ); +} + +void gl_2d_rect( GL_PROGRAM* gl2d, VEC2 origin, VEC2 dim, CLR col ) { + static const U16 order[] = { 0, 1, 2, 3, 4 }; + + glUseProgram( gl2d->id ); + + VERTEX vertices[] = { + { .pos = { origin.x , origin.y }, .clr = col }, + { .pos = { origin.x + dim.x, origin.y }, .clr = col }, + { .pos = { origin.x + dim.x, origin.y + dim.y }, .clr = col }, + { .pos = { origin.x , origin.y + dim.y }, .clr = col }, + { .pos = { origin.x , origin.y }, .clr = col }, + }; + + glBindBuffer( GL_ARRAY_BUFFER, gl2d->gl->vbuffer ); + glBufferData( GL_ARRAY_BUFFER, sizeof(vertices), &vertices[0], GL_DYNAMIC_DRAW ); + I32 position = glGetAttribLocation( gl2d->id, "in_pos" ); + glEnableVertexAttribArray( position ); + glVertexAttribPointer( position, 2, GL_FLOAT, 0, sizeof(VERTEX), 0 ); + I32 color = glGetAttribLocation( gl2d->id, "in_col" ); + glEnableVertexAttribArray( color ); + glVertexAttribPointer( color, 4, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->clr ); + + glBindBuffer( GL_ARRAY_BUFFER, 0 ); + glDrawElements( GL_LINE_STRIP, 5, GL_UNSIGNED_SHORT, order ); +} + +void gl_2d_textured_frect( GL_PROGRAM* gl2d, VEC2 origin, VEC2 dim, GL_TEX2D* texture, CLR col, VEC2* uv, F32 rotation ) { + static const U16 order[] = { 0, 1, 2, 3 }; + + glUseProgram( gl2d->id ); + VERTEX vertices[] = { + { { origin.x, origin.y }, uv? uv[0] : VEC2{ 0.f, 0.f }, col, 0 }, + { { origin.x + dim.x, origin.y }, uv? uv[1] : VEC2{ 1.f, 0.f }, col, 0 }, + { { origin.x, origin.y + dim.y }, uv? uv[2] : VEC2{ 0.f, 1.f }, col, 0 }, + { { origin.x + dim.x, origin.y + dim.y }, uv? uv[3] : VEC2{ 1.f, 1.f }, col, 0 } + }; + + rotation = remainderf( rotation, 360.f ); + if( rotation != 0.f ) { + F32 rad2dg = rotation * (M_PI / 180.f); + + // rotate texture coordinates + for( U32 i = 0; i < 4; ++i ) { + F32 x = vertices[i].uv.x - 0.5f; + F32 y = vertices[i].uv.y - 0.5f; + vertices[i].uv.x = x * cosf( rad2dg ) - y * sinf( rad2dg ) + 0.5f; + vertices[i].uv.y = x * sinf( rad2dg ) + y * cosf( rad2dg ) + 0.5f; + if( vertices[i].uv.x < 0.f ) vertices[i].uv.x = 0.f; + if( vertices[i].uv.x > 1.f ) vertices[i].uv.x = 1.f; + } + } + + glBindBuffer( GL_ARRAY_BUFFER, gl2d->gl->vbuffer ); + glBufferData( GL_ARRAY_BUFFER, sizeof(vertices), &vertices[0], GL_DYNAMIC_DRAW ); + I32 position = glGetAttribLocation( gl2d->id, "in_pos" ); + glEnableVertexAttribArray( position ); + glVertexAttribPointer( position, 2, GL_FLOAT, 0, sizeof(VERTEX), 0 ); + I32 color = glGetAttribLocation( gl2d->id, "in_col" ); + glEnableVertexAttribArray( color ); + glVertexAttribPointer( color, 4, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->clr ); + I32 texcoord = glGetAttribLocation( gl2d->id, "in_texcoord" ); + glEnableVertexAttribArray( texcoord ); + glVertexAttribPointer( texcoord, 2, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->uv ); + I32 sampler = glGetAttribLocation( gl2d->id, "in_sampler" ); + glEnableVertexAttribArray( sampler ); + glVertexAttribPointer( sampler, 1, GL_UNSIGNED_BYTE, 1, sizeof(VERTEX), &( (VERTEX*)nullptr)->sampler ); + + glBindBuffer( GL_ARRAY_BUFFER, 0 ); + + glActiveTexture( GL_TEXTURE0 ); + glBindTexture( GL_TEXTURE_2D, texture->id ); + + glUniform1f( glGetUniformLocation( gl2d->id, "in_time" ), u_time() ); + + glDrawElements( GL_TRIANGLE_STRIP, 4, GL_UNSIGNED_SHORT, order ); + glBindTexture( GL_TEXTURE_2D, 0 ); +} + +void gl_2d_frect( GL_PROGRAM* gl2d, VEC2 origin, VEC2 dim, CLR col ) { + static const U16 order[] = { 0, 1, 2, 3 }; + + glUseProgram( gl2d->id ); + VERTEX vertices[] = { + { .pos = { origin.x, origin.y }, .uv = { 0.f, 0.f }, .clr = col }, + { .pos = { origin.x + dim.x, origin.y }, .uv = { 1.f, 0.f }, .clr = col }, + { .pos = { origin.x, origin.y + dim.y }, .uv = { 0.f, 1.f }, .clr = col }, + { .pos = { origin.x + dim.x, origin.y + dim.y }, .uv = { 1.f, 1.f }, .clr = col } + }; + + glBindBuffer( GL_ARRAY_BUFFER, gl2d->gl->vbuffer ); + glBufferData( GL_ARRAY_BUFFER, sizeof(vertices), &vertices[0], GL_DYNAMIC_DRAW ); + I32 position = glGetAttribLocation( gl2d->id, "in_pos" ); + glEnableVertexAttribArray( position ); + glVertexAttribPointer( position, 2, GL_FLOAT, 0, sizeof(VERTEX), 0 ); + I32 color = glGetAttribLocation( gl2d->id, "in_col" ); + glEnableVertexAttribArray( color ); + glVertexAttribPointer( color, 4, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->clr ); + I32 texcoord = glGetAttribLocation( gl2d->id, "in_texcoord" ); + glEnableVertexAttribArray( texcoord ); + glVertexAttribPointer( texcoord, 2, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->uv ); + + glBindBuffer( GL_ARRAY_BUFFER, 0 ); + glDrawElements( GL_TRIANGLE_STRIP, 4, GL_UNSIGNED_SHORT, order ); +} + +void gl_2d_circle( GL_PROGRAM* gl2d, VEC2 origin, F32 radius, CLR col, U32 res ) { + static U16* order = 0; + const F32 step = 360.f / (F32)res; + + glUseProgram( gl2d->id ); + + if( !order ) { + order = (U16*)malloc( sizeof( U16 ) * (res + 1) ); + for( U32 i = 0; i < res + 1; ++i ) + order[i] = i; + } + + VERTEX* vertices = (VERTEX*)malloc( sizeof( VERTEX ) * (res + 1) ); + for( U32 i = 0; i < res + 1; ++i ) { + VEC2 offset = m_radial_offset( step * ( i == res? 0 : i ), radius ); + + vertices[i] = (VERTEX){ + .pos = { + origin.x + offset.x, + origin.y + offset.y, + }, + .uv = {}, + .clr = col + }; + } + + glBindBuffer( GL_ARRAY_BUFFER, gl2d->gl->vbuffer ); + glBufferData( GL_ARRAY_BUFFER, sizeof(VERTEX) * (res + 1), &vertices[0], GL_STATIC_DRAW ); + I32 position = glGetAttribLocation( gl2d->id, "in_pos" ); + glEnableVertexAttribArray( position ); + glVertexAttribPointer( position, 2, GL_FLOAT, 0, sizeof(VERTEX), 0 ); + I32 color = glGetAttribLocation( gl2d->id, "in_col" ); + glEnableVertexAttribArray( color ); + glVertexAttribPointer( color, 4, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->clr ); + + glBindBuffer( GL_ARRAY_BUFFER, 0 ); + + glDrawElements( GL_LINE_STRIP, res + 1, GL_UNSIGNED_SHORT, order ); + free( vertices ); +} + +void gl_2d_fcircle( GL_PROGRAM* gl2d, VEC2 origin, F32 radius, CLR col, U32 res ) { + static U16* order = 0; + const F32 step = 360.f / (F32)res; + + glUseProgram( gl2d->id ); + + if( !order ) { + order = (U16*)malloc( sizeof( U16 ) * (res * 2) ); + for( U32 i = 0; i < res * 2; ++i ) + order[i] = i; + } + + VERTEX* vertices = (VERTEX*)malloc( sizeof(VERTEX) * (res * 2) ); + for( U32 i = 0; i < res * 2; i += 2 ) { + VEC2 offset = m_radial_offset( step * i, radius ); + vertices[i].pos = (VEC2){ + origin.x, + origin.y, + }; + + vertices[i + 1].pos = (VEC2){ + origin.x + offset.x, + origin.y + offset.y, + }; + + vertices[i].uv = (VEC2){ 0.5f, 0.5f }; + vertices[i + 1].uv = (VEC2){ + 0.5f + ( offset.x * 0.5f ) / radius, + 0.5f + ( offset.y * 0.5f ) / radius + }; + + vertices[i].clr = col; + vertices[i + 1].clr = col; + }; + + glBindBuffer( GL_ARRAY_BUFFER, gl2d->gl->vbuffer ); + glBufferData( GL_ARRAY_BUFFER, sizeof(VERTEX) * res * 2, &vertices[0], GL_STATIC_DRAW ); + I32 position = glGetAttribLocation( gl2d->id, "in_pos" ); + glEnableVertexAttribArray( position ); + glVertexAttribPointer( position, 2, GL_FLOAT, 0, sizeof(VERTEX), 0 ); + I32 color = glGetAttribLocation( gl2d->id, "in_col" ); + glEnableVertexAttribArray( color ); + glVertexAttribPointer( color, 4, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->clr ); + I32 texcoord = glGetAttribLocation( gl2d->id, "in_texcoord" ); + glEnableVertexAttribArray( texcoord ); + glVertexAttribPointer( texcoord, 2, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->uv ); + + glBindBuffer( GL_ARRAY_BUFFER, 0 ); + + glDrawElements( GL_TRIANGLE_STRIP, res + 2, GL_UNSIGNED_SHORT, order ); + free( vertices ); +} + +void gl_polygon( GL_PROGRAM* gl2d, VERTEX* vertices, U32 vertices_count ) { + glUseProgram( gl2d->id ); + + glBindBuffer( GL_ARRAY_BUFFER, gl2d->gl->vbuffer ); + glBufferData( GL_ARRAY_BUFFER, sizeof(VERTEX) * vertices_count, vertices, GL_STATIC_DRAW ); + + I32 position = glGetAttribLocation( gl2d->id, "in_pos" ); + glEnableVertexAttribArray( position ); + glVertexAttribPointer( position, 2, GL_FLOAT, 0, sizeof(VERTEX), 0 ); + I32 color = glGetAttribLocation( gl2d->id, "in_col" ); + glEnableVertexAttribArray( color ); + glVertexAttribPointer( color, 4, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->clr ); + I32 texcoord = glGetAttribLocation( gl2d->id, "in_texcoord" ); + glEnableVertexAttribArray( texcoord ); + glVertexAttribPointer( texcoord, 2, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->uv ); + + glBindBuffer( GL_ARRAY_BUFFER, 0 ); + + U16* order = (U16*)alloca( sizeof(U16) * vertices_count ); + for( U32 i = 0; i < vertices_count; i++ ) + order[i] = i; + + glDrawElements( GL_TRIANGLE_FAN, vertices_count, GL_UNSIGNED_SHORT, order ); +} + +void gl_textured_polygon( + GL_PROGRAM *gl2d, + VERTEX *vertices, + U32 vertices_count, + GL_TEX2D *tex +) { + glUseProgram( gl2d->id ); + + for( U32 i = 0; i < vertices_count; ++i ) + vertices[i].sampler = 0; + + glBindBuffer( GL_ARRAY_BUFFER, gl2d->gl->vbuffer ); + glBufferData( GL_ARRAY_BUFFER, sizeof(VERTEX) * vertices_count, vertices, GL_STATIC_DRAW ); + I32 position = glGetAttribLocation( gl2d->id, "in_pos" ); + glEnableVertexAttribArray( position ); + glVertexAttribPointer( position, 2, GL_FLOAT, 0, sizeof(VERTEX), 0 ); + I32 color = glGetAttribLocation( gl2d->id, "in_col" ); + glEnableVertexAttribArray( color ); + glVertexAttribPointer( color, 4, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->clr ); + I32 texcoord = glGetAttribLocation( gl2d->id, "in_texcoord" ); + glEnableVertexAttribArray( texcoord ); + glVertexAttribPointer( texcoord, 2, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->uv ); + I32 sampler = glGetAttribLocation( gl2d->id, "in_sampler" ); + glEnableVertexAttribArray( sampler ); + glVertexAttribPointer( sampler, 1, GL_UNSIGNED_BYTE, 1, sizeof(VERTEX), &( (VERTEX*)nullptr)->sampler ); + + glBindBuffer( GL_ARRAY_BUFFER, 0 ); + + glActiveTexture( GL_TEXTURE0 ); + glBindTexture( GL_TEXTURE_2D, tex->id ); + + glUniform1f( glGetUniformLocation( gl2d->id, "in_time" ), u_time() ); + + U16* order = (U16*)alloca( sizeof(U16) * vertices_count ); + for( U32 i = 0; i < vertices_count; i++ ) + order[i] = i; + + glDrawElements( GL_TRIANGLE_FAN, vertices_count, GL_UNSIGNED_SHORT, order ); +} diff --git a/src/render/gl_2d.h b/src/render/gl_2d.h new file mode 100644 index 0000000..9830516 --- /dev/null +++ b/src/render/gl_2d.h @@ -0,0 +1,27 @@ +#pragma once +#include "gl.h" + +extern GL_SHADER_PROGRAM* gl_2d_init( GL_DATA* gl, VEC2 screensize, const char* shadername ); + +void gl_polygon( GL_SHADER_PROGRAM* gl2d, VERTEX* vertices, U32 vertices_count ); +void gl_textured_polygon( + GL_SHADER_PROGRAM* gl2d, + VERTEX* vertices, + U32 vertices_count, + GL_TEX2D* tex +); + +extern void gl_2d_line( GL_SHADER_PROGRAM* gl2d, VEC2 start, VEC2 end, CLR col ); +extern void gl_2d_rect( GL_SHADER_PROGRAM* gl2d, VEC2 origin, VEC2 dimensions, CLR col ); +extern void gl_2d_frect( GL_SHADER_PROGRAM* gl2d, VEC2 origin, VEC2 dimensions, CLR col ); +extern void gl_2d_circle( GL_SHADER_PROGRAM* gl2d, VEC2 origin, F32 radius, CLR col, U32 res = 48 ); +extern void gl_2d_fcircle( GL_SHADER_PROGRAM* gl2d, VEC2 origin, F32 radius, CLR col, U32 res = 48 ); +extern void gl_2d_textured_frect( + GL_SHADER_PROGRAM* gl2d, + VEC2 origin, + VEC2 dim, + GL_TEX2D* texture, + CLR col = { 1.f, 1.f, 1.f, 1.f }, + VEC2* uv = 0, + F32 rotation = 0.F +); diff --git a/src/render/gl_2d_font.cpp b/src/render/gl_2d_font.cpp new file mode 100644 index 0000000..1dc498b --- /dev/null +++ b/src/render/gl_2d_font.cpp @@ -0,0 +1,366 @@ +#include "gl_2d_font.h" +#include "gl.h" + +FT_Library libft = NULL; + +THREAD_MUTEX font_mutex; + +void freetype_init() { + U32 err; + + if( !libft ) { + err = FT_Init_FreeType( &libft ); + + if( err ) + dlog( "gl_font_create() : FT_Init_FreeType() failed\n" ); + } +} + +void gl_font_create_bitmap( GL_FONT* font ) { + U32 char_width = (font->face->size->metrics.max_advance >> 6) + 1; + U32 char_height = (font->face->size->metrics.height >> 6) + 1; + U32 max_width = char_width * 16; + U32 max_height = char_height * 16; + + if( !max_width || !max_height ) { + dlog( "gl_font_create() : invalid font size\n" ); + return; + } + + font->bitmap = (U32*)malloc( sizeof(U32) * max_width * max_height ); + memset( font->bitmap, 0, sizeof(U32) * max_width * max_height ); + font->bitmap_width = max_width; + font->bitmap_height = max_height; + font->char_width = char_width; + font->char_height = char_height; + + U32 cur_x = 0, cur_y = 1; + for( I32 c = 0; c < GL_FONT_MAX_GLYPHS; ++c ) { + U32 err = FT_Load_Char( font->face, c, FT_LOAD_RENDER ); + if( !err ) { + FT_Bitmap* bitmap = &font->face->glyph->bitmap; + for( U32 y = 0; y < bitmap->rows; ++y ) { + for( U32 x = 0; x < bitmap->width; ++x ) { + U32 dest_x = x + cur_x; + U32 dest_y = y + cur_y; + + U8 pixel = bitmap->buffer[x + bitmap->width * y]; + U32 out_pixel = 0x00FFFFFF; + out_pixel |= ( pixel << 24 ); + font->bitmap[dest_x + max_width * dest_y] = out_pixel; + } + } + + font->glyphs[c].offset_x = (F32)cur_x; + font->glyphs[c].offset_y = (F32)cur_y; + font->glyphs[c].width = bitmap->width; + font->glyphs[c].height = bitmap->rows; + font->glyphs[c].advance_x = (F32)(font->face->glyph->metrics.horiAdvance >> 6); + font->glyphs[c].advance_y = (F32)(char_height - (font->face->glyph->metrics.horiBearingY >> 6)); + font->glyphs[c].bearing = (F32)(font->face->glyph->metrics.horiBearingX >> 6); + } + else { + dlog( "gl_font_create() : FT_Load_Char() failed (%s: %c) (%x)\n", font->name, c, err ); + } + + if( cur_x + char_width >= max_width ) { + cur_x = 0; + cur_y += char_height; + } else { + cur_x += char_width; + } + } +} + +void gl_font_create_atlas( GL_DATA* gl, GL_FONT* font ) { + if( !font->bitmap ) { + dlog( "gl_font_create_atlas() : no bitmap\n" ); + return; + } + + char texture_name[256]; + sprintf( texture_name, "%s_%d_atlas", font->name, font->size ); + + font->atlas = gl_texture_from_bitmap( gl, + texture_name, + (U8*)font->bitmap, + font->bitmap_width, + font->bitmap_height + ); + + free( font->bitmap ); +} + +GL_FONT* gl_font_create( GL_DATA* gl, const char* path, I32 size ) { + freetype_init(); + + if( size <= 2 ) { + dlog( "gl_font_create() : size too small.\n" ); + return 0; + } + + char full_path[256]; + GL_FONT* font = (GL_FONT*)malloc( sizeof( GL_FONT ) ); + font->size = size; + font->name = path; + + sprintf( full_path, "../assets/fonts/%s", path ); + U32 err = FT_New_Face( libft, full_path, 0, &font->face ); + + if( err ) { + dlog( "gl_font_create() : FT_New_Face() failed (%x)\n", err ); + free( font ); + return 0; + } + + err = FT_Set_Pixel_Sizes( font->face, 0, size ); + if( err ) { + dlog( "gl_font_create() : FT_Set_Char_Size() failed\n" ); + free( font ); + return 0; + } + + gl_font_create_bitmap( font ); + FT_Done_Face( font->face ); + + gl_font_create_atlas( gl, font ); + gl->fonts.push( font ); + return font; +} + +void gl_font_destroy( GL_DATA* gl, GL_FONT* font ) { + gl_texture_destroy( gl, font->atlas ); + I32 idx = gl->fonts.idx_of( font ); + if( idx != -1 ) + gl->fonts.erase( idx ); + + free( font ); +} + + +void gl_font_calc_vertices_uvs( + GL_FONT* font, + VEC2 origin, + const char* text, + F32 _scale, + VERTEX* vertices, + U16* indices, + VEC2* coords, + CLR clr +) { + U32 len = (U32)strlen( text ); + + F32 cur_x = origin.x; + F32 cur_y = origin.y; + + for( U32 i = 0; i < len; ++i ) { + VERTEX* v = &vertices[i * 6]; + U16* idx = &indices[i * 6]; + + if( text[i] == '\n' ) { + v[0] = v[1] = v[2] = v[3] = v[4] = v[5] = { { cur_x, cur_y }, {} }; + idx[0] = idx[1] = idx[2] = idx[3] = idx[4] = idx[5] = i * 6; + + cur_x = origin.x; + cur_y += (F32)font->char_height * _scale; + continue; + } + + U32 c = (U8)text[i]; + FONT_GLYPH* glyph = &font->glyphs[c]; + F32 final_y = cur_y + glyph->advance_y * _scale; + F32 final_x = cur_x + (F32)glyph->bearing * _scale; + + v[0].pos = { final_x, final_y }; + v[1].pos = { final_x + glyph->width * _scale, final_y }; + v[2].pos = { final_x, final_y + glyph->height * _scale }; + v[3].pos = { final_x + glyph->width * _scale, final_y + glyph->height * _scale }; + v[4].pos = { final_x, final_y + glyph->height * _scale }; + v[5].pos = { final_x + glyph->width * _scale, final_y }; + + idx[0] = i * 6; + idx[1] = i * 6 + 1; + idx[2] = i * 6 + 2; + idx[3] = i * 6 + 3; + idx[4] = i * 6 + 4; + idx[5] = i * 6 + 5; + + v[0].uv = { glyph->offset_x / (F32)font->bitmap_width, glyph->offset_y / font->bitmap_height }; + v[1].uv = { (glyph->offset_x + glyph->width) / (F32)font->bitmap_width, glyph->offset_y / font->bitmap_height }; + v[2].uv = { glyph->offset_x / (F32)font->bitmap_width, (glyph->offset_y + glyph->height) / font->bitmap_height }; + v[3].uv = { (glyph->offset_x + glyph->width) / (F32)font->bitmap_width, (glyph->offset_y + glyph->height) / font->bitmap_height }; + v[4].uv = { glyph->offset_x / (F32)font->bitmap_width, (glyph->offset_y + glyph->height) / font->bitmap_height }; + v[5].uv = { (glyph->offset_x + glyph->width) / (F32)font->bitmap_width, glyph->offset_y / font->bitmap_height }; + + v[0].clr = v[1].clr = v[2].clr = v[3].clr = v[4].clr = v[5].clr = clr; + v[0].sampler = v[1].sampler = v[2].sampler = v[3].sampler = v[4].sampler = v[5].sampler = 0; + + cur_x += glyph->advance_x * _scale; + } +} + +void gl_font_draw( GL_FONT* font, GL_SHADER_PROGRAM* shader, VEC2 origin, const char* text, CLR clr, F32 _scale ) { + U32 len = strlen( text ); + VERTEX* vertices = (VERTEX*)malloc( sizeof(VERTEX) * 6 * len ); + U16* indices = (U16*)malloc( sizeof(U16) * 6 * len ); + VEC2* coords = (VEC2*)malloc( sizeof(VEC2) * 6 * len ); + + gl_font_calc_vertices_uvs( font, origin, text, _scale, vertices, indices, coords, clr ); + glUseProgram( shader->id ); + + glBindBuffer( GL_ARRAY_BUFFER, shader->gl->vbuffer ); + glBufferData( GL_ARRAY_BUFFER, sizeof(VERTEX) * 6 * len, vertices, GL_STATIC_DRAW ); + I32 position = glGetAttribLocation( shader->id, "in_pos" ); + glEnableVertexAttribArray( position ); + glVertexAttribPointer( position, 2, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->pos ); + + I32 color = glGetAttribLocation( shader->id, "in_clr" ); + glEnableVertexAttribArray( color ); + glVertexAttribPointer( color, 4, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->clr ); + + I32 texcoord = glGetAttribLocation( shader->id, "in_texcoord" ); + glEnableVertexAttribArray( texcoord ); + glVertexAttribPointer( texcoord, 2, GL_FLOAT, 0, sizeof(VERTEX), &( (VERTEX*)nullptr)->uv ); + + I32 sampler = glGetAttribLocation( shader->id, "in_sampler" ); + glEnableVertexAttribArray( sampler ); + glVertexAttribPointer( sampler, 1, GL_UNSIGNED_BYTE, 1, sizeof(VERTEX), &( (VERTEX*)nullptr)->sampler ); + + glBindBuffer( GL_ARRAY_BUFFER, 0 ); + + glActiveTexture( GL_TEXTURE0 ); + + glBindTexture( GL_TEXTURE_2D, font->atlas->id ); + + glDrawElements( GL_TRIANGLES, len * 6, GL_UNSIGNED_SHORT, indices ); + glBindTexture( GL_TEXTURE_2D, 0 ); + + free( vertices ); + free( indices ); + free( coords ); +} + +void gl_font_textured( + GL_FONT* font, + GL_SHADER_PROGRAM* shader, + VEC2 origin, + const char* text, + GL_TEX2D* tex, + CLR clr, + F32 _scale +) { + U32 len = strlen( text ); + struct FONT_CUSTOM_VERTEX { + VERTEX v; + VEC2 uv2; + }; + + FONT_CUSTOM_VERTEX* custom_vertices = (FONT_CUSTOM_VERTEX*)malloc( sizeof(FONT_CUSTOM_VERTEX) * 6 * len ); + VERTEX* vertices = (VERTEX*)malloc( sizeof(VERTEX) * 6 * len ); + U16* indices = (U16*)malloc( sizeof(U16) * 6 * len ); + VEC2* coords = (VEC2*)malloc( sizeof(VEC2) * 6 * len ); + + gl_font_calc_vertices_uvs( font, origin, text, _scale, vertices, indices, coords, clr ); + VEC2 dim = gl_font_dim( font, text, _scale ); + + for( U32 i = 0; i < len * 6; ++i ) { + custom_vertices[i].v = vertices[i]; + custom_vertices[i].uv2 = { + (vertices[i].pos.x - origin.x) / dim.x, + (vertices[i].pos.y - origin.y - vertices[0].pos.y) / dim.y + }; + } + + glUseProgram( shader->id ); + + glBindBuffer( GL_ARRAY_BUFFER, shader->gl->vbuffer ); + glBufferData( GL_ARRAY_BUFFER, sizeof(FONT_CUSTOM_VERTEX) * 6 * len, custom_vertices, GL_STATIC_DRAW ); + I32 position = glGetAttribLocation( shader->id, "in_pos" ); + glEnableVertexAttribArray( position ); + glVertexAttribPointer( position, 2, GL_FLOAT, 0, sizeof(FONT_CUSTOM_VERTEX), 0 ); + + I32 texcoord = glGetAttribLocation( shader->id, "in_texcoord" ); + glEnableVertexAttribArray( texcoord ); + glVertexAttribPointer( texcoord, 2, GL_FLOAT, 0, sizeof(FONT_CUSTOM_VERTEX), (void*)(sizeof(VEC2)) ); + + I32 texcoord2 = glGetAttribLocation( shader->id, "in_texcoord2" ); + glEnableVertexAttribArray( texcoord2 ); + glVertexAttribPointer( texcoord2, 2, GL_FLOAT, 0, sizeof(FONT_CUSTOM_VERTEX), (void*)(sizeof(VERTEX)) ); + + glBindBuffer( GL_ARRAY_BUFFER, 0 ); + + I32 color = glGetUniformLocation( shader->id, "in_color" ); + glUniform4fv( color, 1, (F32*)&clr ); + + glActiveTexture( GL_TEXTURE0 ); + glUniform1iv( glGetUniformLocation( shader->id, "in_sampler" ), 0, 0 ); + glBindTexture( GL_TEXTURE_2D, font->atlas->id ); + + glDrawElements( GL_TRIANGLES, len * 6, GL_UNSIGNED_SHORT, indices ); + glBindTexture( GL_TEXTURE_2D, 0 ); + + free( vertices ); + free( indices ); + free( coords ); + free( custom_vertices ); +} + + +VEC2 gl_font_dim( GL_FONT* font, const char* text, F32 _scale ) { + U32 len = strlen( text ); + + F32 cur_x = 0.f; + F32 cur_y = font->char_height * _scale; + F32 max_x = 0.f; + + for( U32 i = 0; i < len; ++i ) { + if( text[i] == '\n' ) { + cur_x = 0; + cur_y += font->char_height * _scale; + continue; + } + + U32 c = (U8)text[i]; + FONT_GLYPH* glyph = &font->glyphs[c]; + cur_x += glyph->advance_x * _scale; + if( cur_x > max_x ) + max_x = cur_x; + } + + return { max_x, cur_y }; +} + +void gl_font_fit_into_box( GL_FONT* font, const char* source, char* out, F32 box_width, F32 box_height, F32 scale ) { + U32 len = strlen( source ); + U32 last_space = 0; + + F32 cur_x = 0.f; + for( U32 i = 0; i < len; ++i ) { + if( source[i] == ' ' ) + last_space = i; + + if( source[i] == '\n' ) { + cur_x = 0; + continue; + } + + FONT_GLYPH* glyph = &font->glyphs[(U8)source[i]]; + cur_x += glyph->advance_x * scale; + + out[i] = source[i]; + if( cur_x > box_width ) { + if( last_space == 0 ) { + out[i] = '\n'; + cur_x = 0; + } else { + out[last_space] = '\n'; + cur_x = 0; + i = last_space; + last_space = 0; + } + } + + } + + out[len] = '\0'; +} diff --git a/src/render/gl_2d_font.h b/src/render/gl_2d_font.h new file mode 100644 index 0000000..71154c0 --- /dev/null +++ b/src/render/gl_2d_font.h @@ -0,0 +1,52 @@ +#pragma once +#include <ft2build.h> +#include FT_FREETYPE_H + +#include "gl.h" +#include "../util/thread.h" + +extern THREAD_MUTEX font_mutex; + +typedef struct { + F32 offset_x; + F32 offset_y; + U32 width; + U32 height; + F32 advance_x; + F32 advance_y; + U32 bearing; +} FONT_GLYPH; + +const I32 GL_FONT_MAX_GLYPHS = 256; +typedef struct GL_FONT { + FT_Face face; + I32 size; + + U32* bitmap; + U32 bitmap_width; + U32 bitmap_height; + U32 char_width; + U32 char_height; + + GL_TEX2D* atlas; + const char* name; + + FONT_GLYPH glyphs[GL_FONT_MAX_GLYPHS]; +} *PGL_FONT; + +GL_FONT* gl_font_create( GL_DATA* gl, const char* path, I32 size ); +void gl_font_destroy( GL_DATA* gl, GL_FONT* font ); + +void gl_font_draw( GL_FONT* font, GL_SHADER_PROGRAM* shader, VEC2 pos, const char* text, CLR color, F32 scale = 1.0f ); +void gl_font_textured( + GL_FONT* font, + GL_SHADER_PROGRAM* shader, + VEC2 pos, + const char* text, + GL_TEX2D* tex, + CLR color = (CLR){1.f, 1.f, 1.f, 1.f}, + F32 scale = 1.0f +); + +VEC2 gl_font_dim( GL_FONT* font, const char* text, F32 scale = 1.0f ); +void gl_font_fit_into_box( GL_FONT* font, const char* source, char* out, F32 box_width, F32 box_height, F32 scale ); diff --git a/src/render/gl_3d.cpp b/src/render/gl_3d.cpp new file mode 100644 index 0000000..78e7f5e --- /dev/null +++ b/src/render/gl_3d.cpp @@ -0,0 +1,204 @@ +#include "gl_3d.h" +#include "../util/matrix.h" +#include "gl.h" +#include "gl_batch.h" + +GL_3D* gl_3d_init( GL_DATA* gl, VEC2 screensize, const char* shadername ) { + GL_3D *gl3d = (GL_3D*)gl_program_create( gl, shadername, sizeof(GL_3D) ); + if( !OK( gl_program_compile( gl, gl3d ) ) ) + dlog( "gl_2d_init() : error compiling shader %s\n", shadername ); + + gl3d->winsize = screensize; + gl3d->aspect = screensize.x / screensize.y; + + glUseProgram( gl3d->id ); + I32 samplers_location = glGetUniformLocation( gl3d->id, "g_samplers" ); + glUniform1iv( samplers_location, gl->shader_texture_limit, SAMPLER_INDICES ); + return gl3d; +} + +void gl_3d_projection_setup( GL_PROGRAM *_gl3d, VEC3 pos, F32 fov_deg, F32 yaw, F32 pitch, F32 near, F32 far, VEC2 winsize ) { + GL_3D* gl3d = (GL_3D*)_gl3d; + MAT4 proj, yaw_rot, pitch_rot, view, transl, tmp; + + F32 fov_rad = m_deg2rad( fov_deg ); + F32 yaw_rad = m_deg2rad( remainderf( yaw + 90.f, 360.f ) ); + F32 pitch_rad = m_deg2rad( remainderf( pitch , 360.f ) ); + gl3d->aspect = winsize.x / winsize.y; + gl3d->winsize = winsize; + + mat4_perspective( &proj, fov_rad, gl3d->aspect, near, far ); + mat4_rotation_y( &yaw_rot, -yaw_rad ); + mat4_rotation_x( &pitch_rot, -pitch_rad ); + mat4_translation( &transl, -pos.x, -pos.z, -pos.y ); + + mat4_mul( &yaw_rot, &transl, &tmp ); + mat4_mul( &pitch_rot, &tmp, &view ); + mat4_mul( &proj, &view, &gl3d->projmat ); + gl3d->gl->proj_matrix = &gl3d->projmat; + + glUseProgram( gl3d->id ); + I32 location = glGetUniformLocation( gl3d->id, "in_proj_matrix" ); + glUniformMatrix4fv( location, 1, 1, (GLfloat*)&gl3d->projmat ); + + glUniform1f( glGetUniformLocation( gl3d->id, "in_time" ), u_time() ); +} + +void gl_3d_batch_setup( GL_BATCH3D *batch ) { + glUseProgram( batch->shader->id ); + glBindBuffer( GL_ARRAY_BUFFER, batch->vbuffer ); + I32 position = glGetAttribLocation( batch->shader->id, "in_pos" ); + glEnableVertexAttribArray( position ); + glVertexAttribPointer( position, 3, GL_FLOAT, 0, sizeof(VERTEX3D), 0 ); + I32 color = glGetAttribLocation( batch->shader->id, "in_clr" ); + glEnableVertexAttribArray( color ); + glVertexAttribPointer( color, 3, GL_FLOAT, 0, sizeof(VERTEX3D), &( (VERTEX3D*)nullptr )->clr ); + I32 texcoord = glGetAttribLocation( batch->shader->id, "in_texcoord" ); + glEnableVertexAttribArray( texcoord ); + glVertexAttribPointer( texcoord, 2, GL_FLOAT, 0, sizeof(VERTEX3D), &( (VERTEX3D*)nullptr )->uv ); + I32 sampler = glGetAttribLocation( batch->shader->id, "in_sampler" ); + glEnableVertexAttribArray( sampler ); + glVertexAttribPointer( sampler, 1, GL_UNSIGNED_BYTE, 1, sizeof(VERTEX3D), &( (VERTEX3D*)nullptr )->sampler ); + glBindBuffer( GL_ARRAY_BUFFER, 0 ); +} + +void gl_3d_polygon( + GL_PROGRAM* gl3d, + VERTEX3D* vertices, + U32 vertices_count, + GL_TEX2D* tex +) { + glUseProgram( gl3d->id ); + + if( !!tex ) { + glActiveTexture( GL_TEXTURE0 ); + glBindTexture( GL_TEXTURE_2D, tex->id ); + } else { + for( U32 i = 0; i < vertices_count; ++i ) + vertices[i].sampler = SAMPLER_ID_NONE; + } + + glBindBuffer( GL_ARRAY_BUFFER, gl3d->gl->vbuffer ); + glBufferData( GL_ARRAY_BUFFER, sizeof(VERTEX3D) * vertices_count, vertices, GL_STATIC_DRAW ); + I32 position = glGetAttribLocation( gl3d->id, "in_pos" ); + glEnableVertexAttribArray( position ); + glVertexAttribPointer( position, 3, GL_FLOAT, 0, sizeof(VERTEX3D), 0 ); + I32 color = glGetAttribLocation( gl3d->id, "in_clr" ); + glEnableVertexAttribArray( color ); + glVertexAttribPointer( color, 3, GL_FLOAT, 0, sizeof(VERTEX3D), &( (VERTEX3D*)nullptr )->clr ); + I32 texcoord = glGetAttribLocation( gl3d->id, "in_texcoord" ); + glEnableVertexAttribArray( texcoord ); + glVertexAttribPointer( texcoord, 2, GL_FLOAT, 0, sizeof(VERTEX3D), &( (VERTEX3D*)nullptr )->uv ); + I32 sampler = glGetAttribLocation( gl3d->id, "in_sampler" ); + glEnableVertexAttribArray( sampler ); + glVertexAttribPointer( sampler, 1, GL_UNSIGNED_BYTE, 1, sizeof(VERTEX3D), &( (VERTEX3D*)nullptr )->sampler ); + + U16* order = (U16*)alloca( sizeof(U16) * vertices_count ); + for( U32 i = 0; i < vertices_count; i++ ) + order[i] = i; + + glBindBuffer( GL_ARRAY_BUFFER, 0 ); + glDrawElements( GL_TRIANGLE_FAN, vertices_count, GL_UNSIGNED_SHORT, order ); +} + +void gl_3d_polygon_fan( + GL_BATCH3D* batch, + VERTEX3D* vertices, + U32 vertices_count, + GL_TEX2D* tex +) { + if( vertices_count < 3 ) + return; + + LIST<VERTEX3D> list = triangle_fan_to_list( vertices, vertices_count ); + gl_batch_insert( + batch, + list.data, + list.size, + tex + ); +} + +void gl_3d_polygon( + GL_BATCH3D* batch, + VERTEX3D* vertices, + U32 vertices_count, + GL_TEX2D* tex +) { + gl_batch_insert( + batch, + vertices, + vertices_count, + tex + ); +} + +void build_plane( VERTEX3D* v, VEC3 pos, VEC2 size, VEC2 rot, CLR col ) { + F32 half_width = size.x * 0.5f; + F32 half_height = size.y * 0.5f; + + VEC3 vertices[4] = { + { -half_width, -half_height, 0 }, + { half_width, -half_height, 0 }, + { half_width, half_height, 0 }, + { -half_width, half_height, 0 } + }; + + F32 sin_pitch = sinf( m_deg2rad( rot.x ) ); + F32 cos_pitch = cosf( m_deg2rad( rot.x ) ); + F32 sin_yaw = sinf( m_deg2rad( rot.y ) ); + F32 cos_yaw = cosf( m_deg2rad( rot.y ) ); + + for( I32 i = 0; i < 4; i++) { + F32 x = vertices[i].x; + F32 y = vertices[i].y; + F32 z = vertices[i].z; + vertices[i].y = y * cos_pitch - z * sin_pitch; + vertices[i].z = y * sin_pitch + z * cos_pitch; + y = vertices[i].y; + vertices[i].x = x * cos_yaw - y * sin_yaw; + vertices[i].y = x * sin_yaw + y * cos_yaw; + + vertices[i].x += pos.x; + vertices[i].y += pos.y; + vertices[i].z += pos.z; + } + + VEC2 uvs[4] = { + { 0.0f, 1.0f }, + { 1.0f, 1.0f }, + { 1.0f, 0.0f }, + { 0.0f, 0.0f } + }; + + v[0] = { .pos = { vertices[0].x, vertices[0].z, vertices[0].y }, .uv = uvs[0], .clr = col, .sampler = 0 }; + v[1] = { .pos = { vertices[1].x, vertices[1].z, vertices[1].y }, .uv = uvs[1], .clr = col, .sampler = 0 }; + v[2] = { .pos = { vertices[2].x, vertices[2].z, vertices[2].y }, .uv = uvs[2], .clr = col, .sampler = 0 }; + v[3] = { .pos = { vertices[3].x, vertices[3].z, vertices[3].y }, .uv = uvs[3], .clr = col, .sampler = 0 }; +} + +void gl_3d_plane( + GL_SHADER_PROGRAM* gl3d, + VEC3 pos, + VEC2 size, + VEC2 rot, + GL_TEX2D* tex, + CLR col +) { + VERTEX3D v[4]; + build_plane( v, pos, size, rot, col ); + gl_3d_polygon( gl3d, v, 4, tex ); +} + +void gl_3d_plane( + GL_BATCH3D* batch, + VEC3 pos, + VEC2 size, + VEC2 rot, + GL_TEX2D* tex, + CLR col +) { + VERTEX3D v[4]; + build_plane( v, pos, size, rot, col ); + gl_3d_polygon_fan( batch, v, 4, tex ); +} diff --git a/src/render/gl_3d.h b/src/render/gl_3d.h new file mode 100644 index 0000000..531c3a3 --- /dev/null +++ b/src/render/gl_3d.h @@ -0,0 +1,69 @@ +#pragma once +#include "gl_batch.h" +#include "../util/matrix.h" + +// magic number sue me. yes this float is representable exactly. +const F32 SCREEN_INVALID_POS = -2137000.f; + +struct VERTEX3D { + VEC3 pos; + VEC2 uv; + CLR clr; + U8 sampler; +}; + +using GL_BATCH3D = GL_BATCH<VERTEX3D>; +struct GL_3D : GL_SHADER_PROGRAM { + F32 aspect; + MAT4 projmat; + + VEC2 winsize; +}; + +// shader init ----------------------------------------------------------- +extern GL_3D* gl_3d_init( GL_DATA* gl, VEC2 screensize, const char* shadername ); +// sets up projection matrix and sends it to the gpu ------------------------------------------------------------ +extern void gl_3d_projection_setup( GL_SHADER_PROGRAM* gl3d, VEC3 pos, F32 fov_deg, F32 yaw, F32 pitch, F32 near, F32 far, VEC2 winsize ); +extern void gl_3d_batch_setup( GL_BATCH3D* batch ); + +// takes a triangle fan not set of triangles +extern void gl_3d_polygon( + GL_SHADER_PROGRAM* gl3d, + VERTEX3D* vertices, + U32 vertices_count, + GL_TEX2D* tex +); + +// calculates triangle set from triangle fan before sending to batch +extern void gl_3d_polygon_fan( + GL_BATCH3D* batch, + VERTEX3D* vertices, + U32 vertices_count, + GL_TEX2D* tex +); + +// draws a set of triangles +extern void gl_3d_polygon( + GL_BATCH3D* batch, + VERTEX3D* vertices, + U32 vertices_count, + GL_TEX2D* tex +); + +extern void gl_3d_plane( + GL_SHADER_PROGRAM* gl3d, + VEC3 pos, + VEC2 size, + VEC2 rot, + GL_TEX2D* tex, + CLR col = CLR::WHITE() +); + +extern void gl_3d_plane( + GL_BATCH3D* batch, + VEC3 pos, + VEC2 size, + VEC2 rot, + GL_TEX2D* tex, + CLR col = CLR::WHITE() +); diff --git a/src/render/gl_batch.h b/src/render/gl_batch.h new file mode 100644 index 0000000..944521f --- /dev/null +++ b/src/render/gl_batch.h @@ -0,0 +1,193 @@ +#pragma once +#include "gl.h" + +// this is maybe a bit too much c++ voodoo for my liking + +const U32 BATCH_VERTICES_MAX = 16384; + +template <typename VERTEX> +struct GL_BATCH; + +// func for setting sampler idx inside of vertex data +template <typename VERTEX> +using GL_BATCH_VERTEX_SAMPLER_IDX_FN = void (*)(VERTEX* vertex, U8 idx); + +template <typename VERTEX> +using GL_BATCH_SETUP_FN = void (*)(GL_BATCH<VERTEX>* batch); + +template <typename VERTEX> +struct GL_BATCH_CALL { + LIST<VERTEX> vertices{}; + VEC2 clip_start; + VEC2 clip_dim; + + VEC2 viewport_start; + VEC2 viewport_dim; + + U16 primitive; + LIST<GL_TEX2D*> textures{}; +}; + +template <typename VERTEX> +struct GL_BATCH { + GL_DATA* gl; + GL_SHADER_PROGRAM* shader; + + VEC2 clip_start; + VEC2 clip_dim; + + VEC2 viewport_start; + VEC2 viewport_dim; + + GLuint vbuffer; + + LIST<GL_BATCH_CALL<VERTEX>> calls{}; + + // func for setting up vertex attribs + GL_BATCH_SETUP_FN<VERTEX> setup_cb; +}; + +/* + * takes in a callback func that gets called before the batch is drawn + * + * should set up vertex attribs correctly according to vertex fmt + */ +template <typename VERTEX> +inline GL_BATCH<VERTEX>* gl_batch_create( + GL_DATA* gl, + GL_SHADER_PROGRAM* shader, + GL_BATCH_SETUP_FN<VERTEX> setup_cb +) { + GL_BATCH<VERTEX>* batch = new GL_BATCH<VERTEX>; + batch->gl = gl; + batch->shader = shader; + + batch->clip_start = gl->clip_start; + batch->clip_dim = gl->clip_dim; + + batch->viewport_start = gl->viewport_start; + batch->viewport_dim = gl->viewport_dim; + + batch->setup_cb = setup_cb; + + glGenBuffers( 1, &batch->vbuffer ); + glBindBuffer( GL_ARRAY_BUFFER, batch->vbuffer ); + // make sure we have enough vram + glBufferData( GL_ARRAY_BUFFER, sizeof( VERTEX ) * BATCH_VERTICES_MAX, 0, GL_DYNAMIC_DRAW ); + return batch; +} + +template <typename VERTEX> +inline void gl_batch_destroy( GL_BATCH<VERTEX>* batch ) { + glDeleteBuffers( &batch->vbuffer ); + delete batch; +} + +template <typename VERTEX> +inline void gl_batch_empty( GL_BATCH<VERTEX>* batch ) { + batch->calls.clear(); +} + +template <typename VERTEX> +inline void gl_batch_draw( GL_BATCH<VERTEX>* batch ) { + glUseProgram( batch->shader->id ); + + VEC2 vp_start, vp_dim, clip_start, clip_dim; + gl_get_viewport( batch->gl, &vp_start, &vp_dim ); + gl_get_clip( batch->gl, &clip_start, &clip_dim ); + + batch->setup_cb( batch ); + batch->calls.each( fn( GL_BATCH_CALL<VERTEX>* call ) { + for( U32 i = 0; i < call->textures.size; ++i ) { + glActiveTexture( GL_TEXTURE0 + i ); + glBindTexture( GL_TEXTURE_2D, call->textures.data[i] ? call->textures.data[i]->id : 0 ); + } + + glBindBuffer( GL_ARRAY_BUFFER, batch->vbuffer ); + glBufferData( + GL_ARRAY_BUFFER, + sizeof(VERTEX) * call->vertices.size, + call->vertices.data, + GL_DYNAMIC_DRAW + ); + + glBindBuffer( GL_ARRAY_BUFFER, 0 ); + glDrawArrays( call->primitive, 0, call->vertices.size ); + } ); + + gl_set_viewport( batch->gl, vp_start, vp_dim ); + gl_set_clip( batch->gl, clip_start, clip_dim ); +} + +template <typename VERTEX> +inline GL_BATCH_CALL<VERTEX>* gl_batch_get_call( GL_BATCH<VERTEX>* batch, GL_TEX2D* tex, U16 primitive, I32* texid ) { + U32 n = batch->calls.size; + VEC2 vp_start = batch->viewport_start; + VEC2 vp_dim = batch->viewport_dim; + VEC2 clip_start = batch->clip_start; + VEC2 clip_dim = batch->clip_dim; + + if( n ) { + GL_BATCH_CALL<VERTEX>* last = &batch->calls.data[n - 1]; + if( last->primitive == primitive + && vp_start == last->clip_start && vp_dim == last->viewport_dim + && clip_start == last->clip_start && clip_dim == last->clip_dim + ) { + I32 idx = last->textures.idx_of( tex ); + if( idx != -1 ) { + *texid = idx; + return last; + } + else if( last->textures.size < batch->gl->shader_texture_limit ) { + *texid = last->textures.size; + last->textures.push( tex ); + return last; + } + } + } + + + GL_BATCH_CALL<VERTEX> call{ + .vertices = {}, + .clip_start = batch->clip_start, + .clip_dim = batch->clip_dim, + .viewport_start = batch->viewport_start, + .viewport_dim = batch->viewport_dim, + .primitive = primitive, + .textures = {}, + }; + + *texid = 0; + call.textures.push( tex ); + return batch->calls.push( call ); +} + +template <typename VERTEX> +inline void gl_batch_insert( + GL_BATCH<VERTEX>* batch, + VERTEX* vertices, + U32 count, + GL_TEX2D* tex, + U16 primitive = GL_TRIANGLES +) { + if( !count ) + return; + + GL_BATCH_CALL<VERTEX>* call; + I32 texid; + VEC2 vp_start, vp_dim; + VEC2 clip_start, clip_dim; + + gl_get_viewport( batch->gl, &vp_start, &vp_dim ); + gl_get_clip( batch->gl, &clip_start, &clip_dim ); + + batch->viewport_start = vp_start; batch->viewport_dim = vp_dim; + batch->clip_start = clip_start; batch->clip_dim = clip_dim; + + call = gl_batch_get_call( batch, tex, primitive, &texid ); + + for( U32 i = 0; i < count; ++i ) { + vertices[i].sampler = (!!tex)? texid : SAMPLER_ID_NONE; + call->vertices.push( vertices[i] ); + } +} diff --git a/src/util.h b/src/util.h new file mode 100644 index 0000000..32fdb7e --- /dev/null +++ b/src/util.h @@ -0,0 +1,31 @@ +#pragma once +#include <SDL.h> + +#include "SDL_timer.h" +#include "util/color.h" +#include "util/allocator.h" +#include "util/vector.h" +#include "util/math.h" +#include "util/config.h" +#include "util/screen.h" +#include "util/input.h" +#include "util/file.h" +#include "util/thread.h" + +inline U64 u_tick() { + return (F64)SDL_GetPerformanceCounter() * 10000 / ((F64)SDL_GetPerformanceFrequency() ); +} + +inline F32 u_time() { + return SDL_GetTicks64() / 1000.f; +} + +template < typename T > +T min( T a, T b ) { + return a < b? a : b; +} + +template < typename T > +T max( T a, T b ) { + return a > b? a : b; +} diff --git a/src/util/aabb.h b/src/util/aabb.h new file mode 100644 index 0000000..f2fd251 --- /dev/null +++ b/src/util/aabb.h @@ -0,0 +1,30 @@ +#pragma once + +#include "vector.h" + +struct AABB { + VEC3 min; + VEC3 max; +}; + +inline VEC3 aabb_center( const AABB& aabb ) { return (aabb.min + aabb.max) * 0.5f; } +inline VEC3 aabb_half( const AABB& aabb ) { return (aabb.max - aabb.min) * 0.5f; } +inline F32 aabb_support_radius( const AABB& aabb, const VEC3& dir ) { + VEC3 half = aabb_half( aabb ); + return fabsf(dir.x) * half.x + fabsf(dir.y) * half.y + fabsf(dir.z) * half.z; +} + +inline VEC2 aabb_extend_at_feet( const AABB& aabb, const VEC3& dir ) { + VEC3 half = aabb_half( aabb ); + + F32 hx = half.x; + F32 hy = half.y; + F32 hz = half.z * 2; + + F32 lateral = fabsf( dir.x ) * hx + fabsf( dir.y ) * hy; + + F32 pos_z = dir.z > 0 ? dir.z * hz : 0.f; + F32 neg_z = dir.z < 0 ? (-dir.z) * hz : 0.f; + + return { lateral + pos_z, lateral + neg_z }; +} diff --git a/src/util/allocator.h b/src/util/allocator.h new file mode 100644 index 0000000..ee0ddd8 --- /dev/null +++ b/src/util/allocator.h @@ -0,0 +1,297 @@ +#pragma once + +#include <stdlib.h> +#include <string.h> +#include <functional> +#include "typedef.h" + +template < typename T > +using QSORT_FN = std::function< U8( T*, T* ) >; + +template < typename T > +static U8 qsort_basic_sort( T* t1, T* t2 ) { + return (*t1 > *t2); +} + +// sort fn should return 0 if t1 < t2, else 1 +template < typename T > +static void _qsort( T* arr, I32 low, I32 high, QSORT_FN<T> sortfn = qsort_basic_sort<T> ) { + if( low > high ) + return; + + T pivot = arr[high]; + I32 i = ( low - 1 ); + + for( I32 j = low; j <= high - 1; j++ ) { + if( sortfn( &arr[j], &pivot ) ) { + i++; + T temp = arr[i]; + arr[i] = arr[j]; + arr[j] = temp; + } + } + T temp = arr[i + 1]; + arr[i + 1] = arr[high]; + arr[high] = temp; + I32 pi = i + 1; + + _qsort( arr, low, pi - 1, sortfn ); + _qsort( arr, pi + 1, high, sortfn ); +} + +template < typename T > +struct LIST { + T* data; + U32 capacity; + U32 size; + + using ON_EACH_FN = std::function< void(T*) >; + using ON_SEARCH_FN = std::function< bool(T*) >; + + LIST() { + data = (T*)malloc( sizeof( T ) ); + memset( data, 0, sizeof( T ) ); + capacity = 1; + size = 0; + } + + LIST( U32 _size ) { + if( !_size ) _size = 1; + data = (T*)malloc( sizeof( T ) * _size ); + memset( data, 0, _size * sizeof( T ) ); + capacity = _size; + size = 0; + } + + LIST( const LIST<T>& other ) { + if( !other.capacity || !other.size ) { + capacity = 1; + size = 0; + data = (T*)malloc( sizeof( T ) ); + memset( data, 0, sizeof( T ) ); + return; + } + + data = (T*)malloc( sizeof( T ) * other.capacity ); + memcpy( data, other.data, other.size * sizeof( T ) ); + size = other.size; + capacity = other.capacity; + } + + LIST<T>& operator=( const LIST<T>& other ) { + if( this == &other ) + return *this; + + if( data && data != other.data ) + free( data ); + + data = 0; + + if( !other.capacity || !other.size ) { + capacity = 1; + size = 0; + data = (T*)malloc( sizeof( T ) ); + return *this; + } + + data = (T*)malloc( other.capacity * sizeof( T ) ); + memcpy( data, other.data, other.size * sizeof( T ) ); + size = other.size; + capacity = other.capacity; + + return *this; + } + + ~LIST() { + free( data ); + } + + void reserve( U32 count ) { + if( capacity >= count ) + return; + + T* newblock = (T*)malloc( count * sizeof( T ) ); + memset( newblock, 0, capacity * sizeof( T ) ); + if( data ) { + memcpy( newblock, data, size * sizeof( T ) ); + free( data ); + } + data = newblock; + capacity = count; + } + + void grow() { + capacity = 2 * capacity; + T* newblock = (T*)malloc( capacity * sizeof( T ) ); + memset( newblock, 0, capacity * sizeof( T ) ); + if( data ) { + memcpy( newblock, data, size * sizeof( T ) ); + free( data ); + } + data = newblock; + } + + void shrink() { + if( capacity == 0 ) + return; + + capacity /= 2; + T* newblock = (T*)malloc( capacity * sizeof( T ) ); + if( data ) { + memcpy( newblock, data, capacity * sizeof( T ) ); + free( data ); + } + data = newblock; + } + + // does not call copy funcs or constructors + T* emplace( const T& item ) { + if( capacity == size ) + grow(); + memcpy( &data[size++], &item, sizeof( T ) ); + + return &data[size - 1]; + } + + T* push( const T& item ) { + if( capacity == size ) + grow(); + data[size++] = item; + + return &data[size - 1]; + } + + // does not call copy constructors, raw memcpy + void emplace_list( const LIST<T>& list ) { + if( !list.size ) + return; + capacity += list.size; + T* newblock = (T*)malloc( capacity * sizeof(T) ); + memset( newblock, 0, sizeof(T) * capacity ); + if( data ) { + memcpy( newblock, data, sizeof(T) * size ); + free( data ); + } + + memcpy( &newblock[size], list.data, sizeof(T) * list.size ); + data = newblock; + size += list.size; + } + + // slow - pushes every item individually, calls copy constructors + void push_list( LIST<T>& list ) { + list.each( fn( T* item ) { + push( *item ); + } ); + } + + T pop() { + if( size == 0 ) { + dlog( "LIST::pop_front() : called on empty list\n" ); + abort(); + return {}; + } + + T ret = data[--size]; + if( size < capacity / 4 ) + shrink(); + + return ret; + } + + T pop_front() { + if( size == 0 ) { + dlog( "LIST::pop_front() : called on empty list\n" ); + abort(); + return {}; + } + + T ret = data[0]; + erase( 0 ); + return ret; + } + + void erase( U32 idx ) { + if( idx >= size ) + return; + + for( U32 i = idx; i < size - 1; i++ ) + data[i] = data[i + 1]; + + if( --size < capacity / 4 ) + shrink(); + } + + T& operator[]( U32 idx ) { + return data[idx]; + } + + void clear() { + size = 0; + free( data ); + data = (T*)malloc( sizeof( T ) ); + memset( data, 0, sizeof( T ) ); + capacity = 1; + } + + void each( ON_EACH_FN func ) { + for( U32 i = 0; i < size; ++i ) { + func( &data[i] ); + } + } + + I32 idx_of( const T& what ) { + for( U32 i = 0; i < size; ++i ) { + if( data[i] == what ) { + return i; + } + } + + return -1; + } + + I32 idx_where( ON_SEARCH_FN what ) { + for( U32 i = 0; i < size; ++i ) { + if( what( &data[i] ) ) { + return i; + } + } + + return -1; + } + + T* find( const T& what ) { + for( U32 i = 0; i < size; ++i ) { + if( data[i] == what ) { + return &data[i]; + } + } + return 0; + } + + T* where( ON_SEARCH_FN what ) { + for( U32 i = 0; i < size; ++i ) { + if( what( &data[i] ) ) { + return &data[i]; + } + } + return 0; + } + + // does not copy, sorts in place + // + // sort fn should return 0 if t1 < t2, else 1 + LIST<T>& sort( QSORT_FN<T> fn = qsort_basic_sort<T> ) { + _qsort<T>( data, 0, size - 1, fn ); + return *this; + } + + // creates a sorted copy + // + // sort fn should return 0 if t1 < t2, else 1 + LIST<T> sorted( QSORT_FN<T> fn = qsort_basic_sort<T> ) { + LIST<T> ret = LIST<T>( *this ); + ret.sort( fn ); + return ret; + } +}; diff --git a/src/util/anim.h b/src/util/anim.h new file mode 100644 index 0000000..dbc2c11 --- /dev/null +++ b/src/util/anim.h @@ -0,0 +1,27 @@ +#include "typedef.h" + +#include <SDL.h> + +template < typename T > +inline T lerp( T a, T b, F32 t ) { + return a + ( b - a ) * t; +} + +template < typename t > +inline t clamp( t a, t min, t max ) { + if ( a < min ) return min; + if ( a > max ) return max; + return a; +} + +inline F32 ease_in( F32 t, F32 pow = 2.f ) { + return powf( t, pow ); +} + +inline F32 ease_out( F32 t, F32 pow = 2.f ) { + return 1.f - powf( 1.f - t, pow ); +} + +inline F32 ease_in_out( F32 t, F32 pow = 2.f ) { + return t < 0.5f ? ease_in( t * 2.f, pow ) / 2.f : ease_out( t * 2.f - 1.f, pow ) / 2.f + 0.5f; +} diff --git a/src/util/color.h b/src/util/color.h new file mode 100644 index 0000000..526cddc --- /dev/null +++ b/src/util/color.h @@ -0,0 +1,168 @@ +#pragma once +#include "typedef.h" +#include <math.h> + +struct CLR { + F32 r, g, b, a; + + static const CLR WHITE( F32 a_ = 1.f ) { return { 1.f, 1.f, 1.f, a_ }; } + static const CLR BLACK( F32 a_ = 1.f ) { return { 0.f, 0.f, 0.f, a_ }; } + static const CLR RED( F32 a_ = 1.f ) { return { 1.f, 0.f, 0.f, a_ }; } + static const CLR GREEN( F32 a_ = 1.f ) { return { 0.f, 1.f, 0.f, a_ }; } + static const CLR BLUE( F32 a_ = 1.f ) { return { 0.f, 0.f, 1.f, a_ }; } + static const CLR YELLOW( F32 a_ = 1.f ) { return { 1.f, 1.f, 0.f, a_ }; } + static const CLR MAGENTA( F32 a_ = 1.f ) { return { 1.f, 0.f, 1.f, a_ }; } + static const CLR CYAN( F32 a_ = 1.f ) { return { 0.f, 1.f, 1.f, a_ }; } + + // slow accurate blend, looks prettier but cache maybe if possible + static CLR blend( const CLR& c1, const CLR& c2, float t = 0.5f ) { + F32 c1r = c1.r * c1.r; + F32 c1g = c1.g * c1.g; + F32 c1b = c1.b * c1.b; + F32 c1a = c1.a * c1.a; + F32 c2r = c2.r * c2.r; + F32 c2g = c2.g * c2.g; + F32 c2b = c2.b * c2.b; + F32 c2a = c2.a * c2.a; + return { + sqrtf( ( c2r * t ) + ( c1r * ( 1.f - t ) ) ), + sqrtf( ( c2g * t ) + ( c1g * ( 1.f - t ) ) ), + sqrtf( ( c2b * t ) + ( c1b * ( 1.f - t ) ) ), + sqrtf( ( c2a * t ) + ( c1a * ( 1.f - t ) ) ) + }; + } + + // fast inaccurate blend + static CLR blend_root( const CLR& c1, const CLR& c2, float t = 0.5f ) { + return { + ( c2.r * t ) + ( c1.r * ( 1.f - t ) ), + ( c2.g * t ) + ( c1.g * ( 1.f - t ) ), + ( c2.b * t ) + ( c1.b * ( 1.f - t ) ), + ( c2.a * t ) + ( c1.a * ( 1.f - t ) ) + }; + } + + static CLR from_hsb( F32 hue, F32 saturation, F32 brightness ) { + CLR c; + if( saturation <= 0.0f ) { + c.r = c.g = c.b = brightness; + return c; + } + + I32 h = (I32)( hue * 6 ); + F32 f = hue * 6 - h; + F32 p = brightness * ( 1 - saturation ); + F32 q = brightness * ( 1 - f * saturation ); + F32 t = brightness * ( 1 - (1 - f) * saturation ); + + switch( h % 6 ) { + case 0: + c.r = brightness; c.g = t; c.b = p; + return c; + case 1: + c.r = q; c.g = brightness; c.b = p; + return c; + case 2: + c.r = p; c.g = brightness; c.b = t; + return c; + case 3: + c.r = p; c.g = q; c.b = brightness; + return c; + case 4: + c.r = t; c.g = p; c.b = brightness; + return c; + case 5: + c.r = brightness; c.g = p; c.b = q; + return c; + } + + return c; + } + + CLR operator+( const CLR& c2 ) { + return { this->r + c2.r, this->g + c2.g, this->b + c2.b, this->a + c2.a }; + } + CLR operator-( const CLR& c2 ) { + return { this->r - c2.r, this->g - c2.g, this->b - c2.b, this->a - c2.a }; + } + CLR operator*( const CLR& c2 ) { + return { this->r * c2.r, this->g * c2.g, this->b * c2.b, this->a * c2.a }; + } + CLR operator/( const CLR& c2 ) { + F32 r = (c2.r != 0.0f) ? this->r / c2.r : 0.0f; + F32 g = (c2.g != 0.0f) ? this->g / c2.g : 0.0f; + F32 b = (c2.b != 0.0f) ? this->b / c2.b : 0.0f; + F32 a = (c2.a != 0.0f) ? this->a / c2.a : 0.0f; + + return { r, g, b, a }; + } + + CLR& operator+=( const CLR& c2 ) { + r += c2.r; + g += c2.g; + b += c2.b; + a += c2.a; + return *this; + } + + CLR& operator-=( const CLR& c2 ) { + r -= c2.r; + g -= c2.g; + b -= c2.b; + a -= c2.a; + return *this; + } + + CLR& operator*=( const CLR& c2 ) { + r *= c2.r; + g *= c2.g; + b *= c2.b; + a *= c2.a; + return *this; + } + + CLR& operator/=( const CLR& c2 ) { + if( c2.r != 0.0f ) { r /= c2.r; } + else { r = 0.0f; } + + if( c2.g != 0.0f ) { g /= c2.g; } + else { g = 0.0f; } + + if( c2.b != 0.0f ) { b /= c2.b; } + else { b = 0.0f; } + + if( c2.a != 0.0f ) { a /= c2.a; } + else { a = 0.0f; } + + return *this; + } + + CLR operator*( F32 f ) { + return { r * f, g * f, b * f, a * f }; + } + + CLR operator/( F32 f ) { + if( f != 0.0f ) + return { this->r / f, this->g / f, this->b / f, this->a / f }; + else + return { 0.0f, 0.0f, 0.0f, 0.0f }; + } + + CLR& operator*=( F32 f ) { + r *= f; g *= f; b *= f; + return *this; + } + + CLR& operator/=( F32 f ) { + if( f != 0.0f ) { + r /= f; + g /= f; + b /= f; + } + else { + r = g = b = a = 0.f; + } + + return *this; + } +}; diff --git a/src/util/config.h b/src/util/config.h new file mode 100644 index 0000000..7973db0 --- /dev/null +++ b/src/util/config.h @@ -0,0 +1,166 @@ +#pragma once +#include <stdio.h> + +#include "allocator.h" +#include "vector.h" +#include "color.h" + +// config parsers and serializers go here + +struct CFG_NODE; +struct CFG_SECTION; + +struct CFG_PARSER; +struct CFG_SERIALIZER; + +enum CfgNodeType_t { + CFGT_SECTION, + CFGT_BYTES, + CFGT_STR, + CFGT_FLOAT, + CFGT_INT, + CFGT_VEC2, + CFGT_VEC3, + CFGT_CLR +}; + +// ===================================== [ definitions ] =========================================== + +using CFG_PARSEFN = std::function<void( CFG_PARSER*, CFG_SECTION*, char* )>; +using CFG_SERIALIZEFN = std::function<void( CFG_SERIALIZER*, CFG_NODE*, char* )>; + +struct CFG_TYPE { + U8 type; + const char* def; + CFG_PARSEFN parser; + CFG_SERIALIZEFN serializer; +}; + +struct CFG_NODE { + char name[64]; + CFG_NODE* parent; + U8 type; +}; + +struct CFG_SECTION : CFG_NODE { + LIST<CFG_NODE*> children; +}; + +struct CFG_BYTES : CFG_NODE { U8* bytes; U32 size; }; +struct CFG_STR : CFG_NODE { char* str; U32 len; }; +struct CFG_INT : CFG_NODE { I32 value; }; +struct CFG_FLOAT : CFG_NODE { F32 value; }; +struct CFG_VEC2 : CFG_NODE { VEC2 value; }; +struct CFG_VEC3 : CFG_NODE { VEC3 value; }; +struct CFG_CLR : CFG_NODE { CLR value; }; + +// ======================================= [ parsers ] ============================================= + +extern STAT cfg_parser_bytes( CFG_PARSER* parser, CFG_SECTION* section, char* name ); +extern STAT cfg_parser_str( CFG_PARSER* parser, CFG_SECTION* section, char* name ); +extern STAT cfg_parser_int( CFG_PARSER* parser, CFG_SECTION* section, char* name ); +extern STAT cfg_parser_float( CFG_PARSER* parser, CFG_SECTION* section, char* name ); +extern STAT cfg_parser_vec2( CFG_PARSER* parser, CFG_SECTION* section, char* name ); +extern STAT cfg_parser_vec3( CFG_PARSER* parser, CFG_SECTION* section, char* name ); +extern STAT cfg_parser_clr( CFG_PARSER* parser, CFG_SECTION* section, char* name ); + +// ====================================== [ serializers ] ========================================== + +extern void cfg_serialize_section( CFG_SERIALIZER* serializer, CFG_NODE* section, char* buf ); +extern void cfg_serialize_bytes( CFG_SERIALIZER* serializer, CFG_NODE* node, char* buf ); +extern void cfg_serialize_str( CFG_SERIALIZER* serializer, CFG_NODE* node, char* buf ); +extern void cfg_serialize_float( CFG_SERIALIZER* serializer, CFG_NODE* node, char* buf ); +extern void cfg_serialize_int( CFG_SERIALIZER* serializer, CFG_NODE* node, char* buf ); +extern void cfg_serialize_vec2( CFG_SERIALIZER* serializer, CFG_NODE* node, char* buf ); +extern void cfg_serialize_vec3( CFG_SERIALIZER* serializer, CFG_NODE* node, char* buf ); +extern void cfg_serialize_clr( CFG_SERIALIZER* serializer, CFG_NODE* node, char* buf ); + +// ========================================= [ config ] ============================================ + +static const CFG_TYPE cfg_types[] = { + { CFGT_SECTION, "DEF" , 0, &cfg_serialize_section }, // section + { CFGT_BYTES , "U8" , &cfg_parser_bytes, &cfg_serialize_bytes }, + { CFGT_STR , "STR" , &cfg_parser_str , &cfg_serialize_str }, + { CFGT_FLOAT , "F32" , &cfg_parser_float, &cfg_serialize_float }, + { CFGT_INT , "I32" , &cfg_parser_int , &cfg_serialize_int }, + { CFGT_VEC2 , "VEC2", &cfg_parser_vec2 , &cfg_serialize_vec2 }, + { CFGT_VEC3 , "VEC3", &cfg_parser_vec3 , &cfg_serialize_vec3 }, + { CFGT_CLR , "CLR" , &cfg_parser_clr , &cfg_serialize_clr }, +}; +// ========================================= [ io ] ================================================ + +extern CFG_SECTION* cfg_load( const char* path ); +extern STAT cfg_save( CFG_SECTION* root, const char* path ); +static void cfg_free( CFG_NODE* n ); + +// =================================== [ getters/setters ] ========================================= + +// cfg_section_get( CFG_SECTION*, const char* ) -- retrieve function +extern CFG_SECTION* cfg_section( CFG_SECTION* parent, const char* name ); +// cfg_str_get( CFG_SECTION*, const char* ) -- retrieve function +extern CFG_STR* cfg_str( CFG_SECTION* section, const char* name ); +// cfg_bytes_get( CFG_SECTION*, const char* ) -- retrieve function +extern CFG_BYTES* cfg_bytes( CFG_SECTION* section, const char* name ); +// cfg_float_get( CFG_SECTION*, const char* ) -- retrieve function +extern CFG_FLOAT* cfg_float( CFG_SECTION* section, const char* name ); +// cfg_int_get( CFG_SECTION*, const char* ) -- retrieve function +extern CFG_INT* cfg_int( CFG_SECTION* section, const char* name ); +// cfg_vec2_get( CFG_SECTION*, const char* ) -- retrieve function +extern CFG_VEC2* cfg_vec2( CFG_SECTION* section, const char* name ); +// cfg_vec3_get( CFG_SECTION*, const char* ) -- retrieve function +extern CFG_VEC3* cfg_vec3( CFG_SECTION* section, const char* name ); +// cfg_clr_get( CFG_SECTION*, const char* ) -- retrieve function for +extern CFG_CLR* cfg_clr( CFG_SECTION* section, const char* name ); + +// cfg_section_new( const char*, CFG_NODE*, I32 ) -- create function +extern CFG_SECTION* cfg_section_new( const char* name, CFG_NODE* parent ); +// cfg_str_new( const char*, CFG_NODE*, const char*, U32 ) -- create function +extern CFG_STR* cfg_str( const char* name, CFG_NODE* section, const char* value, U32 len ); +// cfg_bytes_new( const char*, CFG_NODE*, U8*, U32 ) -- create function +extern CFG_BYTES* cfg_bytes( const char* name, CFG_NODE* section, U8* value, U32 size ); +// cfg_float_new( const char*, CFG_NODE*, F32 ) -- create function +extern CFG_FLOAT* cfg_float( const char* name, CFG_NODE* section, F32 value ); +// cfg_int_new( const char*, CFG_NODE*, I32 ) -- create function +extern CFG_INT* cfg_int( const char* name, CFG_NODE* section, I32 value ); +// cfg_vec2_new( const char*, CFG_NODE*, VEC2 ) -- create function +extern CFG_VEC2* cfg_vec2( const char* name, CFG_NODE* section, VEC2 value ); +// cfg_vec3_new( const char*, CFG_NODE*, VEC3 ) -- create function +extern CFG_VEC3* cfg_vec3( const char* name, CFG_NODE* section, VEC3 value ); +// cfg_clr_new( const char*, CFG_NODE*, CLR ) -- create function +extern CFG_CLR* cfg_clr( const char* name, CFG_NODE* section, CLR value ); + +// ================================================================================================= + +struct CFG_PARSER { + FILE* file; + CFG_SECTION* root; + char err[256]; + U32 linen; + U8 iserr; +}; + +struct CFG_SERIALIZER { + FILE* f; + U8 iserr; + char err[256]; + U32 tabc; +}; + +static void cfg_free( CFG_NODE* n ) { + if( n->type == CFGT_STR ) { + CFG_STR* str = (CFG_STR*)n; + free( str->str ); + } + else if( n->type == CFGT_BYTES ) { + CFG_BYTES* bytes = (CFG_BYTES*)n; + free( bytes->bytes ); + } + else if( n->type == CFGT_SECTION ) { + CFG_SECTION* s = (CFG_SECTION*)n; + s->children.each( fn( CFG_NODE** child ) { cfg_free( *child ); } ); + } + + delete n; +} + +extern void cfg_seterr( CFG_PARSER* p, const char* fmt, ... ); diff --git a/src/util/config/config.cpp b/src/util/config/config.cpp new file mode 100644 index 0000000..1601afe --- /dev/null +++ b/src/util/config/config.cpp @@ -0,0 +1,263 @@ +#include <cstdarg> +#include <cstdio> + +#include "../config.h" + +void cfg_seterr( CFG_PARSER* p, const char* fmt, ... ) { + va_list args; + va_start( args, fmt ); + vsnprintf( p->err, sizeof(p->err), fmt, args ); + p->iserr = 1; + va_end( args ); +} + +inline U8 is_whitespace( char c ) { + return c == ' ' || c == '\t' || c == '\n'; +} + +inline void trim_whitespace( char* buf ) { + U32 i; + for( i = 0; !!buf[i]; ++i ) + if( !is_whitespace( buf[i] ) ) break; + for( U32 i2 = i; !!buf[i2]; ++i2 ) { + if( is_whitespace( buf[i2] ) ) { + buf[i2 - i] = 0; + return; + } + buf[i2 - i] = buf[i2]; + } +} + +inline void init_cfg_node( CFG_NODE* node, const char* name, CFG_NODE* parent, U8 type ) { + memset( node->name, 0, sizeof(node->name) ); + strcpy( node->name, name ); + node->name[sizeof(node->name) - 1] = '\0'; + node->parent = parent; + if( parent ) + ( (CFG_SECTION*)parent )->children.push( node ); + node->type = type; +} + +CFG_SECTION* cfg_section_new( const char* name, CFG_NODE* parent ) { + CFG_SECTION* section = new CFG_SECTION; + init_cfg_node( (CFG_NODE*)section, name, parent, CFGT_SECTION ); + return section; +} + +CFG_BYTES* cfg_bytes( const char* name, CFG_NODE* parent, U8* bytes, U32 size ) { + CFG_BYTES* cfg_bytes = new CFG_BYTES; + init_cfg_node( (CFG_NODE*)cfg_bytes, name, parent, CFGT_BYTES ); + cfg_bytes->bytes = bytes; + cfg_bytes->size = size; + return cfg_bytes; +} + +CFG_STR* cfg_str( const char* name, CFG_NODE* parent, const char* str, U32 len ) { + CFG_STR* cfg_str = new CFG_STR; + init_cfg_node( (CFG_NODE*)cfg_str, name, parent, CFGT_STR ); + cfg_str->str = (char*)malloc( len + 1 ); + strcpy( cfg_str->str, str ); + cfg_str->str[len] = '\0'; + cfg_str->len = len; + return cfg_str; +} + +CFG_INT* cfg_int( const char* name, CFG_NODE* parent, I32 value ) { + CFG_INT* cfg_int = new CFG_INT; + init_cfg_node( (CFG_NODE*)cfg_int, name, parent, CFGT_INT ); + cfg_int->value = value; + return cfg_int; +} + +CFG_FLOAT* cfg_float( const char* name, CFG_NODE* parent, F32 value) { + CFG_FLOAT* cfg_float = new CFG_FLOAT; + init_cfg_node( (CFG_NODE*)cfg_float, name, parent, CFGT_FLOAT ); + cfg_float->value = value; + return cfg_float; +} + +CFG_VEC2* cfg_vec2( const char* name, CFG_NODE* parent, VEC2 value ) { + CFG_VEC2* cfg_vec2 = new CFG_VEC2; + init_cfg_node( (CFG_NODE*)cfg_vec2, name, parent, CFGT_VEC2 ); + cfg_vec2->value = value; + return cfg_vec2; +} + +CFG_VEC3* cfg_vec3( const char* name, CFG_NODE* parent, VEC3 value ) { + CFG_VEC3* cfg_vec3 = new CFG_VEC3; + init_cfg_node( (CFG_NODE*)cfg_vec3, name, parent, CFGT_VEC3 ); + cfg_vec3->value = value; + return cfg_vec3; +} + +CFG_CLR* cfg_clr( const char* name, CFG_NODE* parent, CLR value ) { + CFG_CLR* cfg_clr = new CFG_CLR; + init_cfg_node( (CFG_NODE*)cfg_clr, name, parent, CFGT_CLR ); + cfg_clr->value = value; + return cfg_clr; +} + +CFG_SECTION* cfg_section( CFG_SECTION* parent, const char* name ) { + for( I32 i = 0; i < parent->children.size; ++i ) { + CFG_NODE* n = parent->children[i]; + if( n->type == CFGT_SECTION && strcmp( n->name, name ) == 0 ) + return (CFG_SECTION*)n; + } + + return 0; +} + +CFG_BYTES* cfg_bytes( CFG_SECTION* parent, const char* name ) { + for( I32 i = 0; i < parent->children.size; ++i ) { + CFG_NODE* n = parent->children[i]; + if( n->type == CFGT_BYTES && strcmp( n->name, name ) == 0 ) + return (CFG_BYTES*)n; + } + + return 0; +} + +CFG_STR* cfg_str( CFG_SECTION* parent, const char* name ) { + for( I32 i = 0; i < parent->children.size; ++i ) { + CFG_NODE* n = parent->children[i]; + if( n->type == CFGT_STR && strcmp( n->name, name ) == 0 ) + return (CFG_STR*)n; + } + + return 0; +} + +CFG_INT* cfg_int( CFG_SECTION* parent, const char* name ) { + for( I32 i = 0; i < parent->children.size; ++i ) { + CFG_NODE* n = parent->children[i]; + if( n->type == CFGT_INT && strcmp( n->name, name ) == 0 ) + return (CFG_INT*)n; + } + + return 0; +} + +CFG_FLOAT* cfg_float( CFG_SECTION* parent, const char* name ) { + for( I32 i = 0; i < parent->children.size; ++i ) { + CFG_NODE* n = parent->children[i]; + if( n->type == CFGT_FLOAT && strcmp( n->name, name ) == 0 ) + return (CFG_FLOAT*)n; + } + return 0; +} + +CFG_VEC2* cfg_vec2( CFG_SECTION* parent, const char* name ) { + for( I32 i = 0; i < parent->children.size; ++i ) { + CFG_NODE* n = parent->children[i]; + if( n->type == CFGT_VEC2 && strcmp( n->name, name ) == 0 ) + return (CFG_VEC2*)n; + } + return 0; +} + +CFG_VEC3* cfg_vec3( CFG_SECTION* parent, const char* name ) { + for( I32 i = 0; i < parent->children.size; ++i ) { + CFG_NODE* n = parent->children[i]; + if( n->type == CFGT_VEC3 && strcmp( n->name, name ) == 0 ) + return (CFG_VEC3*)n; + } + return 0; +} + +CFG_CLR* cfg_clr( CFG_SECTION* parent, const char* name ) { + for( I32 i = 0; i < parent->children.size; ++i ) { + CFG_NODE* n = parent->children[i]; + if( n->type == CFGT_CLR && strcmp( n->name, name ) == 0 ) + return (CFG_CLR*)n; + } + return 0; +} + +void parse_section( CFG_PARSER* parser, CFG_SECTION* current_section ) { + char line[8192]; + char* token; + char* next_token; + + while( fgets( line, sizeof(line), parser->file ) ) { + token = strtok( line, " \t\n" ); + if( !token ) continue; + if( strcmp( token, "{" ) == 0 ) + continue; + else if( strcmp( token, "}" ) == 0 ) { + return; + } else if( strcmp( token, cfg_types[CFGT_SECTION].def ) == 0 ) { + next_token = strtok( NULL, " \t\n" ); + char sectname[64]; + strcpy( sectname, next_token ); + trim_whitespace( sectname ); + + CFG_SECTION* new_section = cfg_section_new( sectname, (CFG_NODE*)current_section ); + strtok( NULL, " \t\n" ); + parse_section( parser, new_section ); + } else { + char name[64]; + strcpy( name, token ); + trim_whitespace( name ); + + token = strtok( NULL, "=[" ); + if( !token ) + continue; + + char varname[64]; + strcpy( varname, token ); + trim_whitespace( varname ); + + for( I32 i = 0; i < sizeof(cfg_types) / sizeof(CFG_TYPE); ++i ) { + const CFG_TYPE* fn = &cfg_types[i]; + if( strncmp( name, fn->def, strlen( fn->def ) ) == 0 ) { + fn->parser( parser, current_section, varname ); + break; + } + } + + if( parser->iserr ) { + dlog( "parse_section() : %s parse error:\n - %s\n", name, parser->err ); + return; + } + } + + parser->linen++; + } +} + +CFG_SECTION* cfg_load( const char* path ) { + FILE* f = fopen( path, "rb" ); + if( !f ) + return 0; + + CFG_PARSER p; + p.iserr = 0; + p.file = f; + p.linen = 0; + p.root = cfg_section_new( "root", 0 ); + parse_section( &p, p.root ); + + fclose( f ); + return p.root; +} + +STAT cfg_save( CFG_SECTION* root, const char* path ) { + FILE* f = fopen( path, "w" ); + if( !f ) + return STAT_ERR; + + char* buf = (char*)malloc( 999999 ); + buf[0] = 0; + CFG_SERIALIZER s; + s.tabc = 0; + s.f = f; + + cfg_serialize_section( &s, (CFG_NODE*)root, buf ); + U32 len = strlen( buf ); + + fwrite( buf, 1, len, f ); + free( buf ); + + fclose( f ); + return STAT_OK; +} diff --git a/src/util/config/parsers.cpp b/src/util/config/parsers.cpp new file mode 100644 index 0000000..1dbfc07 --- /dev/null +++ b/src/util/config/parsers.cpp @@ -0,0 +1,156 @@ +#include <cstdarg> +#include <cstdio> + +#include "../config.h" + +U8* parse_hex_string( CFG_PARSER* p, const char* hex_str, U8* bytes, U32* size ) { + *size = strlen( hex_str ) / 2; + for( U32 i = 0; i < *size; ++i ) { + U32 buf; + I32 c = sscanf( hex_str + 2 * i, "%02x", &buf ); + if( c != 1 ) { + cfg_seterr( p, "Invalid hex string: %s", hex_str ); + free( bytes ); + return NULL; + } + + bytes[i] = ( buf & 0xff ); + } + + return bytes; +} + +STAT cfg_parser_bytes( CFG_PARSER* p, CFG_SECTION* s, char* name ) { + char* sizep = strtok( NULL, "]" ); + if( !sizep ) { + cfg_seterr( p, "cfg_parser_bytes() : %s: invalid size [L:%d]", name, p->linen ); + return STAT_ERR; + } + + U32 size = atoi( sizep ); + strtok( NULL, "=" ); + strtok( NULL, "\"" ); + char* hex_value = strtok( NULL, "\"" ); + if( !hex_value ) { + cfg_seterr( p, "cfg_parser_bytes() : %s: invalid value (size: %d) [L:%d]", name, size, p->linen ); + return STAT_ERR; + } + + U32 totals; + U8* bytes = (U8*)malloc( size ); + parse_hex_string( p, hex_value, bytes, &totals ); + cfg_bytes( name, (CFG_NODE*)s, bytes, size ); + + return STAT_OK; +} + +STAT cfg_parser_str( CFG_PARSER* p, CFG_SECTION* s, char* name ) { + char* sizep = strtok( NULL, "]" ); + if( !sizep ) { + cfg_seterr( p, "cfg_parser_str() : %s: invalid size [L:%d]", name, p->linen ); + return STAT_ERR; + } + + U32 size = atoi( sizep ); + strtok( NULL, "=" ); + strtok( NULL, "\"" ); + char* v = strtok( NULL, "\"" ); + if( !v ) { + cfg_seterr( p, "cfg_parser_str() : %s: incorrect value [L:%d]", name, p->linen ); + return STAT_ERR; + } + + cfg_str( name, (CFG_NODE*)s, v, size ); + + return STAT_OK; +} + +STAT cfg_parser_int( CFG_PARSER* p, CFG_SECTION* s, char* name ) { + char* vp = strtok( NULL, ";" ); + if( !vp ) { + cfg_seterr( p, "cfg_parser_int() : %s: incorrect value [L:%d]", name, p->linen ); + return STAT_ERR; + } + + I32 value = atoi( vp ); + cfg_int( name, (CFG_NODE*)s, value ); + + return STAT_OK; +} + +STAT cfg_parser_float( CFG_PARSER* p, CFG_SECTION* s, char* name ) { + char* vp = strtok( NULL, ";" ); + if( !vp ) { + cfg_seterr( p, "cfg_parser_float() : %s: incorrect value [L:%d]", name, p->linen ); + return STAT_ERR; + } + + F32 value = atof( vp ); + cfg_float( name, (CFG_NODE*)s, value ); + + return STAT_OK; +} + +STAT cfg_parser_vec2( CFG_PARSER *p, CFG_SECTION* s, char* name ) { + char* vp = strtok( NULL, "{" ); + const char* errstr = "cfg_parser_vec2() : %s: incorrect value [L:%d]"; + + if( !vp ) { cfg_seterr( p, errstr, name, p->linen ); return STAT_ERR; } + vp = strtok( NULL, "," ); + if( !vp ) { cfg_seterr( p, errstr, name, p->linen ); return STAT_ERR; } + F32 v1 = atof( vp ); + vp = strtok( NULL, "}" ); + if( !vp ) { cfg_seterr( p, errstr, name, p->linen ); return STAT_ERR; } + F32 v2 = atof( vp ); + + VEC2 v = { v1, v2 }; + cfg_vec2( name, (CFG_NODE*)s, v ); + + return STAT_OK; +} + +STAT cfg_parser_vec3( CFG_PARSER *parser, CFG_SECTION *section, char *name ) { + char *vp = strtok( NULL, "{" ); + const char* errstr = "cfg_parser_vec3() : %s: incorrect value [L:%d]"; + + if( !vp ) { cfg_seterr( parser, errstr, name, parser->linen ); return STAT_ERR; } + vp = strtok( NULL, "," ); + if( !vp ) { cfg_seterr( parser, errstr, name, parser->linen ); return STAT_ERR; } + F32 x = atof( vp ); + vp = strtok( NULL, "," ); + if( !vp ) { cfg_seterr( parser, errstr, name, parser->linen ); return STAT_ERR; } + F32 y = atof( vp ); + vp = strtok( NULL, "}" ); + if( !vp ) { cfg_seterr( parser, errstr, name, parser->linen ); return STAT_ERR; } + F32 z = atof( vp ); + VEC3 v = { x, y, z }; + cfg_vec3( name, (CFG_NODE*)section, v ); + + return STAT_OK; +} + +STAT cfg_parser_clr( CFG_PARSER* parser, CFG_SECTION* section, char* name ) { + const char* vp = strtok( NULL, "{" ); + const char* errstr = "cfg_parser_clr() : %s: incorrect value [L:%d]"; + if( !vp ) { + cfg_seterr( parser, errstr, name, parser->linen ); + return STAT_ERR; + } + + vp = strtok( NULL, "," ); + if( !vp ) { cfg_seterr( parser, errstr, name, parser->linen ); return STAT_ERR; } + F32 r = atof( vp ); + vp = strtok( NULL, "," ); + if( !vp ) { cfg_seterr( parser, errstr, name, parser->linen ); return STAT_ERR; } + F32 g = atof( vp ); + vp = strtok( NULL, "," ); + if( !vp ) { cfg_seterr( parser, errstr, name, parser->linen ); return STAT_ERR; } + F32 b = atof( vp ); + vp = strtok( NULL, "}" ); + if( !vp ) { cfg_seterr( parser, errstr, name, parser->linen ); return STAT_ERR; } + F32 a = atof( vp ); + + CLR c = { r, g, b, a }; + cfg_clr( name, (CFG_NODE*)section, c ); + return STAT_OK; +} diff --git a/src/util/config/serializers.cpp b/src/util/config/serializers.cpp new file mode 100644 index 0000000..01f416c --- /dev/null +++ b/src/util/config/serializers.cpp @@ -0,0 +1,109 @@ +#include <cstdarg> +#include <cstdio> + +#include "../config.h" + +void serialize_node( CFG_SERIALIZER* s, CFG_NODE* n, char* buf ) { + U32 len = strlen( buf ); + for( U32 i = 0; i < s->tabc * 2; ++i ) buf[i + len] = ' '; + buf[len + s->tabc * 2] = 0; + sprintf( buf, "%s%s %s", buf, cfg_types[n->type].def, n->name ); +} + +void cfg_serialize_section( CFG_SERIALIZER* s, CFG_NODE *n, char *buf ) { + CFG_SECTION* sec = (CFG_SECTION*)n; + if( sec->parent ) { + serialize_node( s, n, buf ); + strcat( buf, " {\n" ); + s->tabc++; + } + + char line[8192]; + for( U32 i = 0; i < sec->children.size; ++i ) { + line[0] = 0; + CFG_NODE* c = sec->children[i]; + if( c->type == CFGT_SECTION ) { + cfg_serialize_section( s, c, buf ); + continue; + } + + cfg_types[c->type].serializer( s, c, line ); + strcat( buf, line ); + strcat( buf, "\n" ); + } + + if( sec->parent ) { + s->tabc--; + char tabbuf[512] = { 0 }; + for( U32 i = 0 ; i < s->tabc * 2; ++i ) + tabbuf[i] = ' '; + tabbuf[s->tabc * 2] = 0; + strcat( buf, tabbuf ); + strcat( buf, "}" ); + if( sec->parent ) + strcat( buf, "\n" ); + } +} + +void cfg_serialize_bytes( CFG_SERIALIZER* s, CFG_NODE* n, char* buf ) { + CFG_BYTES* b = (CFG_BYTES*)n; + U32 size = b->size; + U8* bytes = b->bytes; + + serialize_node( s, n, buf ); + sprintf( buf, "%s[%d] = \"", buf, size ); + U32 len = strlen( buf ); + for( U32 i = 0; i < size; ++i ) { + sprintf( buf + len + i * 2, "%02X", bytes[i] ); + } + + sprintf( buf, "%s%s", buf, "\";" ); +} + +void cfg_serialize_str( CFG_SERIALIZER* s, CFG_NODE* n, char *buf ) { + CFG_STR* sn = (CFG_STR*)n; + char* str = sn->str; + + serialize_node( s, n, buf ); + sprintf( buf, "%s[%d] = \"%s\";", buf, sn->len, str ); +} + +void cfg_serialize_int( CFG_SERIALIZER* s, CFG_NODE* n, char *buf ) { + CFG_INT* i = (CFG_INT*)n; + I32 ival = i->value; + + serialize_node( s, n, buf ); + sprintf( buf, "%s = %d;", buf, ival ); +} + +void cfg_serialize_float( CFG_SERIALIZER* s, CFG_NODE* n, char *buf ) { + CFG_FLOAT* f = (CFG_FLOAT*)n; + F32 fval = f->value; + + serialize_node( s, n, buf ); + sprintf( buf, "%s = %g;", buf, fval ); +} + +void cfg_serialize_vec2( CFG_SERIALIZER* s, CFG_NODE* n, char *buf ) { + CFG_VEC2* v = (CFG_VEC2*)n; + VEC2 val = v->value; + + serialize_node( s, n, buf ); + sprintf( buf, "%s = { %g, %g };", buf, val.x, val.y ); +} + +void cfg_serialize_vec3( CFG_SERIALIZER* s, CFG_NODE* n, char *buf ) { + CFG_VEC3* v = (CFG_VEC3*)n; + VEC3 val = v->value; + + serialize_node( s, n, buf ); + sprintf( buf, "%s = { %g, %g, %g };", buf, val.x, val.y, val.z ); +} + +void cfg_serialize_clr( CFG_SERIALIZER* s, CFG_NODE* n, char *buf ) { + CFG_CLR* v = (CFG_CLR*)n; + CLR val = v->value; + + serialize_node( s, n, buf ); + sprintf( buf, "%s = { %g, %g, %g, %g };", buf, val.r, val.g, val.b, val.a ); +} diff --git a/src/util/file.h b/src/util/file.h new file mode 100644 index 0000000..e19a390 --- /dev/null +++ b/src/util/file.h @@ -0,0 +1,78 @@ +#pragma once + +#include <cstdlib> +#include <cstring> +#include <stdio.h> +#include <dirent.h> +#include <sys/stat.h> +#include "typedef.h" +#include "allocator.h" + +inline void* file_read( const char* file ) { + FILE* f = fopen( file, "rb" ); + if( !f ) + return 0; + + defer( fclose( f ) ); + + fseek( f, 0, SEEK_END ); + U64 size = ftell( f ); + rewind( f ); + if( !size ) + return 0; + + void* block = malloc( size + 1 ); + fread( block, size, 1, f ); + + ( (U8*)block )[size] = 0; + return block; +} + +inline void file_write( const char* file, U8* data, U32 size ) { + FILE* f = fopen( file, "wb" ); + if( !f ) + return; + + fwrite( data, size, 1, f ); + fclose( f ); +} + +inline const char* file_path_last_of( const char* path ) { + const char* last_slash = strrchr( path, '/' ); + return last_slash? last_slash + 1 : path; +} + +struct FILE_ENTRY { + char name[256]; + U8 dir; +}; + +inline LIST<FILE_ENTRY> dir_get_entries( const char* path ) { + LIST<FILE_ENTRY> entries; + + DIR* dir = opendir( path ); + if( !dir ) + return entries; + + struct dirent* entry; + struct stat file_stat; + char full_path[512]; + + while( !!( entry = readdir( dir ) ) ) { + if( !strcmp( entry->d_name, "." ) || !strcmp( entry->d_name, ".." ) ) + continue; + + snprintf( full_path, sizeof(full_path), "%s/%s", path, entry->d_name); + if( stat( full_path, &file_stat ) ) + continue; + + FILE_ENTRY file_entry; + strncpy( file_entry.name, entry->d_name, sizeof(file_entry.name) - 1 ); + file_entry.name[sizeof(file_entry.name) - 1] = '\0'; + file_entry.dir = S_ISDIR( file_stat.st_mode ) ? 1 : 0; + entries.push( file_entry ); + } + + closedir( dir ); + return entries; +} diff --git a/src/util/fnv.h b/src/util/fnv.h new file mode 100644 index 0000000..534063d --- /dev/null +++ b/src/util/fnv.h @@ -0,0 +1,31 @@ +#pragma once +#include "typedef.h" + +typedef U32 FNV1A; + +enum : FNV1A { + FNV1A_PRIME = 0x1000193, + FNV1A_BASIS = 0x811C9DC5 +}; + +inline constexpr U32 strlen_ct( const char* str ) { + U32 s = 0; + for( ; !!str[s]; ++s ); + + return s; +} + +inline constexpr FNV1A fnv1a( const U8* data, const U32 size ) { + FNV1A out = FNV1A_BASIS; + + for( U32 i = 0; i < size; ++i ) + out = ( out ^ data[i] ) * FNV1A_PRIME; + + return out; +} + +inline constexpr FNV1A fnv1a( const char* str ) { + U32 len = strlen_ct( str ); + + return fnv1a( (const U8*)str, len ); +} diff --git a/src/util/input.cpp b/src/util/input.cpp new file mode 100644 index 0000000..4a786aa --- /dev/null +++ b/src/util/input.cpp @@ -0,0 +1,74 @@ +#include "input.h" +#include "SDL_events.h" + +INPUT_DATA input; + +void input_on_event( SDL_Event* e ) { + input.on_input.each( fn( ON_INPUT_FN* func ) { (*func)( e ); } ); + + switch( e->type ) { + case SDL_MOUSEBUTTONDOWN: { + switch( e->button.button ) { + case 1: input.mouse.left = 1; break; + case 2: input.mouse.middle = 1; break; + case 3: input.mouse.right = 1; break; + } + } break; + case SDL_MOUSEBUTTONUP: { + switch( e->button.button ) { + case 1: input.mouse.left = 0; break; + case 2: input.mouse.middle = 0; break; + case 3: input.mouse.right = 0; break; + } + } break; + case SDL_MOUSEWHEEL: { + input.mouse.wheel = e->wheel.y; + } break; + case SDL_MOUSEMOTION: { + input.mouse.pos.x = (F32)e->motion.x; + input.mouse.pos.y = (F32)e->motion.y; + } break; + case SDL_KEYDOWN: { + input.keys[e->key.keysym.sym & 0xff] = 1; + } break; + case SDL_KEYUP: { + input.keys[e->key.keysym.sym & 0xff] = 0; + } break; + } +} + + +void input_on_mouse( I32 type, I32 x, I32 y ) { + if( type == MOUSEEV_MOVE ) { + input.mouse.pos.x = (F32)x; + input.mouse.pos.y = (F32)y; + } + + if( type == MOUSEEV_DOWN ) { + I32 button = x; + if( button == MOUSE_LEFT ) + input.mouse.left = 1; + if( button == MOUSE_RIGHT ) + input.mouse.right = 1; + if( button == MOUSE_MIDDLE ) + input.mouse.middle = 1; + if( button == MOUSE_WHEEL ) + input.mouse.wheel = -1; + } + + if( type == MOUSEEV_UP ) { + I32 button = x; + if( button == MOUSE_LEFT ) + input.mouse.left = 0; + if( button == MOUSE_RIGHT ) + input.mouse.right = 0; + if( button == MOUSE_MIDDLE ) + input.mouse.middle = 0; + if( button == MOUSE_WHEEL ) + input.mouse.wheel = 1; + } +} + +void input_frame_end() { + input.mouse.wheel = 0; +} diff --git a/src/util/input.h b/src/util/input.h new file mode 100644 index 0000000..e174fe6 --- /dev/null +++ b/src/util/input.h @@ -0,0 +1,41 @@ +#pragma once + +#include <SDL_events.h> +#include <functional> +#include "vector.h" +#include "allocator.h" + +const U32 MOUSEEV_MOVE = 0x0; +const U32 MOUSEEV_DOWN = 0x1; +const U32 MOUSEEV_UP = 0x2; + +const U32 MOUSE_LEFT = 0x1; +const U32 MOUSE_RIGHT = 0x2; +const U32 MOUSE_MIDDLE = 0x3; +const U32 MOUSE_WHEEL = 0x4; + +struct MOUSE_DATA { + VEC2 pos; + U8 left; + U8 right; + U8 middle; + U8 wheel; +}; + +using ON_INPUT_FN = std::function<void( SDL_Event* )>; + +struct INPUT_DATA { + MOUSE_DATA mouse; + U8 keys[0xff]; + + LIST<ON_INPUT_FN> on_input; +}; + +extern INPUT_DATA input; + +extern void input_frame_end(); +extern void input_on_event( SDL_Event* e ); +extern void input_on_mouse( I32 type, I32 x, I32 y ); +extern void input_is_key_down( U32 key ); + +#define kb_down( key ) input_is_key_down( key ) diff --git a/src/util/math.cpp b/src/util/math.cpp new file mode 100644 index 0000000..4ab3f2b --- /dev/null +++ b/src/util/math.cpp @@ -0,0 +1,26 @@ +#include "math.h" + +#include "../render/gl_3d.h" + +VEC2 m_screen_transform( const VEC3 &world ) { + GL_DATA* gl = gl_instance(); + MAT4* gl_proj_matrix = gl->proj_matrix; + VEC2 screen = gl->viewport_start; + VEC3 clip; + + clip.x = world.x * gl_proj_matrix->m[0][0] + world.y * gl_proj_matrix->m[1][0] + world.z * gl_proj_matrix->m[2][0] + gl_proj_matrix->m[3][0]; + clip.y = world.x * gl_proj_matrix->m[0][1] + world.y * gl_proj_matrix->m[1][1] + world.z * gl_proj_matrix->m[2][1] + gl_proj_matrix->m[3][1]; + clip.z = world.x * gl_proj_matrix->m[0][3] + world.y * gl_proj_matrix->m[1][3] + world.z * gl_proj_matrix->m[2][3] + gl_proj_matrix->m[3][3]; + + if( clip.z <= 0.f ) { + screen.x = screen.y = SCREEN_INVALID_POS; + } + + F32 ndcx = clip.x / clip.z; + F32 ndcy = clip.y / clip.z; + + screen.x += ( ndcx * .5f + .5f ) * (F32)gl->viewport_dim.x; + screen.y += ( 1.f - ( ndcy * .5f + .5f ) ) * (F32)gl->viewport_dim.y; + + return screen; +} diff --git a/src/util/math.h b/src/util/math.h new file mode 100644 index 0000000..add705a --- /dev/null +++ b/src/util/math.h @@ -0,0 +1,74 @@ +#pragma once + +#include <math.h> +#include "vector.h" + +inline F32 m_deg2rad( F32 deg ) { return deg * PIRAD; } +inline F32 m_rad2deg( F32 rad ) { return rad * RADPI; } + +inline VEC2 m_radial_offset( F32 degrees, F32 distance ) { + F32 rad = m_deg2rad( degrees ); + F32 x = cos( rad ); + F32 y = sin( rad ); + + VEC2 ret = { x * distance, y * distance }; + return ret; +} + +inline F32 m_dist_line_to_point( VEC2 p1, VEC2 p2, VEC2 p ) { + if( vec_dist( p1, p2 ) < 0.001f ) + return vec_dist( p1, p ); + + F32 a = p2.y - p1.y; + F32 b = p1.x - p2.x; + F32 c = p2.x * p1.y - p1.x * p2.y; + + F32 d = a * p.x + b * p.y + c; + F32 n = sqrtf( a * a + b * b ); + + F32 dot1 = (p.x - p1.x) * (p2.x - p1.x) + (p.y - p1.y) * (p2.y - p1.y); + F32 dot2 = (p.x - p2.x) * (p1.x - p2.x) + (p.y - p2.y) * (p1.y - p2.y); + + if( dot1 < 0 ) return vec_dist( p, p1 ); + if( dot2 < 0 ) return vec_dist( p, p2 ); + + return fabsf( d ) / n; +} + +inline U8 m_point_in_polygon( VEC2 p, VEC2* points, U32 pointc, U32 point_stride = sizeof(VEC2) ) { + U8 in = 0; + for( U32 i = 0, j = pointc - 1; i < pointc; j = i++ ) { + U8* pointarr = (U8*)points; + + VEC2* pi = (VEC2*)( pointarr + i * point_stride ); + VEC2* pj = (VEC2*)( pointarr + j * point_stride ); + + if( ( pi->y > p.y ) != ( pj->y > p.y ) ) { + if( ( p.x < ( pj->x - pi->x ) * ( p.y - pi->y ) / ( pj->y - pi->y ) + pi->x ) ) + in = !in; + } + } + + return in; +} + +inline F32 m_snap_to_grid( F32 x, F32 grid ) { + return x - copysignf( fmodf( fabsf( x ), grid ), x ); +} + +template <typename T> +inline T m_min( T a, T b ) { + return a < b ? a : b; +} + +template <typename T> +inline T m_max( T a, T b ) { + return a > b ? a : b; +} + +template <typename T> +inline T m_clamp( T x, T a, T b ) { + return m_min( m_max( x, a ), b ); +} + +extern VEC2 m_screen_transform( const VEC3& world ); diff --git a/src/util/matrix.h b/src/util/matrix.h new file mode 100644 index 0000000..6657323 --- /dev/null +++ b/src/util/matrix.h @@ -0,0 +1,81 @@ +#pragma once +#include "typedef.h" +#include "vector.h" + +typedef struct { + F32 m[4][4]; +} MAT4; + +inline void mat4_identity( MAT4* out ) { + for( I32 i = 0; i < 16; i++ ) { + out->m[0][i] = (i % 5 == 0) ? 1.0f : 0.0f; + } +} + +inline void mat4_mul( MAT4* a, MAT4* b, MAT4* out ) { + for( I32 row = 0; row < 4; row++ ) { + for( I32 col = 0; col < 4; col++ ) { + out->m[row][col] = 0.0f; + for( I32 k = 0; k < 4; k++ ) { + out->m[row][col] += a->m[row][k] * b->m[k][col]; + } + } + } +} + +inline void mat4_mul_vec4( MAT4 *m, VEC4 *in, VEC4 *out ) { + out->x = m->m[0][0] * in->x + m->m[0][1] * in->y + m->m[0][2] * in->z + m->m[0][3] * in->w; + out->y = m->m[1][0] * in->x + m->m[1][1] * in->y + m->m[1][2] * in->z + m->m[1][3] * in->w; + out->z = m->m[2][0] * in->x + m->m[2][1] * in->y + m->m[2][2] * in->z + m->m[2][3] * in->w; + out->w = m->m[3][0] * in->x + m->m[3][1] * in->y + m->m[3][2] * in->z + m->m[3][3] * in->w; +} + +inline void mat4_perspective( MAT4* out, F32 fovy_radians, F32 aspect, F32 znear, F32 zfar ) { + mat4_identity( out ); + F32 f = 1.f / tanf( fovy_radians * .5f ); + + out->m[0][0] = f / aspect; + out->m[1][1] = f; + out->m[2][2] = (zfar + znear) / (znear - zfar); + out->m[2][3] = (2.f * zfar * znear) / (znear - zfar); + out->m[3][2] = -1.f; + out->m[3][3] = 0.f; +} + +inline void mat4_translation( MAT4* out, F32 tx, F32 ty, F32 tz ) { + mat4_identity( out ); + out->m[0][3] = tx; + out->m[1][3] = ty; + out->m[2][3] = tz; +} + +inline void mat4_rotation_x( MAT4* out, F32 angle ) { + mat4_identity( out ); + F32 c = cosf( angle ); + F32 s = sinf( angle ); + out->m[1][1] = c; + out->m[1][2] = -s; + out->m[2][1] = s; + out->m[2][2] = c; +} + +inline void mat4_rotation_y( MAT4* out, F32 angle ) { + mat4_identity( out ); + F32 c = cosf( angle ); + F32 s = sinf( angle ); + out->m[0][0] = c; + out->m[0][2] = -s; + out->m[2][0] = s; + out->m[2][2] = c; +} + +inline void mat4_rotation_z( MAT4* out, F32 angle ) { + mat4_identity( out ); + F32 c = cosf( angle ); + F32 s = sinf( angle ); + out->m[0][0] = c; + out->m[0][1] = -s; + out->m[1][0] = s; + out->m[1][1] = c; +} + diff --git a/src/util/screen.h b/src/util/screen.h new file mode 100644 index 0000000..34effb3 --- /dev/null +++ b/src/util/screen.h @@ -0,0 +1,41 @@ +#pragma once +#include "typedef.h" +#include "vector.h" + +extern I32* canvas; + +inline VEC2 s_tl( VEC2 offset = { 0.f, 0.f } ) { + return { offset.x, offset.y }; +} + +inline VEC2 s_tr( VEC2 offset = { 0.f, 0.f } ) { + return { offset.x + (F32)canvas[0], offset.y }; +} + +inline VEC2 s_bl( VEC2 offset = { 0.f, 0.f } ) { + return { offset.x, offset.y + (F32)canvas[1] }; +} + +inline VEC2 s_br( VEC2 offset = { 0.f, 0.f } ) { + return { offset.x + (F32)canvas[0], offset.y + (F32)canvas[1] }; +} + +inline VEC2 s_c( VEC2 offset = { 0.f, 0.f } ) { + return { offset.x + (F32)canvas[0] / 2, offset.y + (F32)canvas[1] / 2 }; +} + +inline VEC2 s_tc( VEC2 offset = { 0.f, 0.f } ) { + return { offset.x + (F32)canvas[0] / 2, 0 }; +} + +inline VEC2 s_bc( VEC2 offset = { 0.f, 0.f } ) { + return { offset.x + (F32)canvas[0] / 2, (F32)canvas[1] }; +} + +inline VEC2 s_cl( VEC2 offset = { 0.f, 0.f } ) { + return { 0, offset.y + (F32)canvas[1] / 2 }; +} + +inline VEC2 s_cr( VEC2 offset = { 0.f, 0.f } ) { + return { (F32)canvas[0], offset.y + (F32)canvas[1] / 2 }; +} diff --git a/src/util/stb_image.h b/src/util/stb_image.h new file mode 100644 index 0000000..5e807a0 --- /dev/null +++ b/src/util/stb_image.h @@ -0,0 +1,7987 @@ +/* stb_image - v2.28 - 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: + + 2.28 (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 signed shorts is valid, 0 on overflow. +static int stbi__mul2shorts_valid(short a, short 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 int 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)) { + int x = stbi__get8(j->s); + while (x == 255) { // 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; + 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)) { + return -1; /* report error for unexpected end of data. */ + } + 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; + 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 (zout + len > a->zout_end) { + if (!stbi__zexpand(a, zout, len)) return 0; + zout = a->zout; + } + p = (stbi_uc *) (zout - dist); + if (dist == 1) { // run of one byte; common in images. + stbi_uc v = *p; + if (len) { do *zout++ = v; while (--len); } + } else { + if (len) { do *zout++ = *p++; while (--len); } + } + } + } +} + +static int stbi__compute_huffman_codes(stbi__zbuf *a) +{ + static const stbi_uc length_dezigzag[19] = { 16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15 }; + stbi__zhuffman z_codelength; + stbi_uc lencodes[286+32+137];//padding for maximum single op + stbi_uc codelength_sizes[19]; + int i,n; + + int hlit = stbi__zreceive(a,5) + 257; + int hdist = stbi__zreceive(a,5) + 1; + int hclen = stbi__zreceive(a,4) + 4; + int ntot = hlit + hdist; + + memset(codelength_sizes, 0, sizeof(codelength_sizes)); + for (i=0; i < hclen; ++i) { + int s = stbi__zreceive(a,3); + codelength_sizes[length_dezigzag[i]] = (stbi_uc) s; + } + if (!stbi__zbuild_huffman(&z_codelength, codelength_sizes, 19)) return 0; + + n = 0; + while (n < ntot) { + int c = stbi__zhuffman_decode(a, &z_codelength); + if (c < 0 || c >= 19) return stbi__err("bad codelengths", "Corrupt PNG"); + if (c < 16) + lencodes[n++] = (stbi_uc) c; + else { + stbi_uc fill = 0; + if (c == 16) { + c = stbi__zreceive(a,2)+3; + if (n == 0) return stbi__err("bad codelengths", "Corrupt PNG"); + fill = lencodes[n-1]; + } else if (c == 17) { + c = stbi__zreceive(a,3)+3; + } else 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; + 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 filters used for first scanline to avoid needing a dummy row of 0s + STBI__F_avg_first, + STBI__F_paeth_first +}; + +static stbi_uc first_row_filter[5] = +{ + STBI__F_none, + STBI__F_sub, + STBI__F_none, + STBI__F_avg_first, + STBI__F_paeth_first +}; + +static int stbi__paeth(int a, int b, int c) +{ + int p = a + b - c; + int pa = abs(p-a); + int pb = abs(p-b); + int pc = abs(p-c); + if (pa <= pb && pa <= pc) return a; + if (pb <= pc) return b; + return c; +} + +static const stbi_uc stbi__depth_scale_table[9] = { 0, 0xff, 0x55, 0, 0x11, 0,0,0, 0x01 }; + +// create the png data from post-deflated data +static int stbi__create_png_image_raw(stbi__png *a, stbi_uc *raw, stbi__uint32 raw_len, int out_n, stbi__uint32 x, stbi__uint32 y, int depth, int color) +{ + int bytes = (depth == 16? 2 : 1); + stbi__context *s = a->s; + stbi__uint32 i,j,stride = x*out_n*bytes; + stbi__uint32 img_len, img_width_bytes; + int k; + int img_n = s->img_n; // copy it into a local for later + + int output_bytes = out_n*bytes; + int filter_bytes = img_n*bytes; + int width = x; + + STBI_ASSERT(out_n == s->img_n || out_n == s->img_n+1); + a->out = (stbi_uc *) stbi__malloc_mad3(x, y, output_bytes, 0); // extra bytes to write off the end into + if (!a->out) return stbi__err("outofmem", "Out of memory"); + + if (!stbi__mad3sizes_valid(img_n, x, depth, 7)) return stbi__err("too large", "Corrupt PNG"); + img_width_bytes = (((img_n * x * depth) + 7) >> 3); + img_len = (img_width_bytes + 1) * y; + + // we used to check for exact match between raw_len and img_len on non-interlaced PNGs, + // but issue #276 reported a PNG in the wild that had extra data at the end (all zeros), + // so just check for raw_len < img_len always. + if (raw_len < img_len) return stbi__err("not enough pixels","Corrupt PNG"); + + for (j=0; j < y; ++j) { + stbi_uc *cur = a->out + stride*j; + stbi_uc *prior; + int filter = *raw++; + + if (filter > 4) + return stbi__err("invalid filter","Corrupt PNG"); + + if (depth < 8) { + if (img_width_bytes > x) return stbi__err("invalid width","Corrupt PNG"); + cur += x*out_n - img_width_bytes; // store output to the rightmost img_len bytes, so we can decode in place + filter_bytes = 1; + width = img_width_bytes; + } + prior = cur - stride; // bugfix: need to compute this after 'cur +=' computation above + + // if first row, use special filter that doesn't sample previous row + if (j == 0) filter = first_row_filter[filter]; + + // handle first byte explicitly + for (k=0; k < filter_bytes; ++k) { + switch (filter) { + case STBI__F_none : cur[k] = raw[k]; break; + case STBI__F_sub : cur[k] = raw[k]; break; + case STBI__F_up : cur[k] = STBI__BYTECAST(raw[k] + prior[k]); break; + case STBI__F_avg : cur[k] = STBI__BYTECAST(raw[k] + (prior[k]>>1)); break; + case STBI__F_paeth : cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(0,prior[k],0)); break; + case STBI__F_avg_first : cur[k] = raw[k]; break; + case STBI__F_paeth_first: cur[k] = raw[k]; break; + } + } + + if (depth == 8) { + if (img_n != out_n) + cur[img_n] = 255; // first pixel + raw += img_n; + cur += out_n; + prior += out_n; + } else if (depth == 16) { + if (img_n != out_n) { + cur[filter_bytes] = 255; // first pixel top byte + cur[filter_bytes+1] = 255; // first pixel bottom byte + } + raw += filter_bytes; + cur += output_bytes; + prior += output_bytes; + } else { + raw += 1; + cur += 1; + prior += 1; + } + + // this is a little gross, so that we don't switch per-pixel or per-component + if (depth < 8 || img_n == out_n) { + int nk = (width - 1)*filter_bytes; + #define STBI__CASE(f) \ + case f: \ + for (k=0; k < nk; ++k) + switch (filter) { + // "none" filter turns into a memcpy here; make that explicit. + case STBI__F_none: memcpy(cur, raw, nk); break; + STBI__CASE(STBI__F_sub) { cur[k] = STBI__BYTECAST(raw[k] + cur[k-filter_bytes]); } break; + STBI__CASE(STBI__F_up) { cur[k] = STBI__BYTECAST(raw[k] + prior[k]); } break; + STBI__CASE(STBI__F_avg) { cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k-filter_bytes])>>1)); } break; + STBI__CASE(STBI__F_paeth) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-filter_bytes],prior[k],prior[k-filter_bytes])); } break; + STBI__CASE(STBI__F_avg_first) { cur[k] = STBI__BYTECAST(raw[k] + (cur[k-filter_bytes] >> 1)); } break; + STBI__CASE(STBI__F_paeth_first) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k-filter_bytes],0,0)); } break; + } + #undef STBI__CASE + raw += nk; + } else { + STBI_ASSERT(img_n+1 == out_n); + #define STBI__CASE(f) \ + case f: \ + for (i=x-1; i >= 1; --i, cur[filter_bytes]=255,raw+=filter_bytes,cur+=output_bytes,prior+=output_bytes) \ + for (k=0; k < filter_bytes; ++k) + switch (filter) { + STBI__CASE(STBI__F_none) { cur[k] = raw[k]; } break; + STBI__CASE(STBI__F_sub) { cur[k] = STBI__BYTECAST(raw[k] + cur[k- output_bytes]); } break; + STBI__CASE(STBI__F_up) { cur[k] = STBI__BYTECAST(raw[k] + prior[k]); } break; + STBI__CASE(STBI__F_avg) { cur[k] = STBI__BYTECAST(raw[k] + ((prior[k] + cur[k- output_bytes])>>1)); } break; + STBI__CASE(STBI__F_paeth) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k- output_bytes],prior[k],prior[k- output_bytes])); } break; + STBI__CASE(STBI__F_avg_first) { cur[k] = STBI__BYTECAST(raw[k] + (cur[k- output_bytes] >> 1)); } break; + STBI__CASE(STBI__F_paeth_first) { cur[k] = STBI__BYTECAST(raw[k] + stbi__paeth(cur[k- output_bytes],0,0)); } break; + } + #undef STBI__CASE + + // the loop above sets the high byte of the pixels' alpha, but for + // 16 bit png files we also need the low byte set. we'll do that here. + if (depth == 16) { + cur = a->out + stride*j; // start at the beginning of the row again + for (i=0; i < x; ++i,cur+=output_bytes) { + cur[filter_bytes+1] = 255; + } + } + } + } + + // we make a separate pass to expand bits to pixels; for performance, + // this could run two scanlines behind the above code, so it won't + // intefere with filtering but will still be in the cache. + if (depth < 8) { + for (j=0; j < y; ++j) { + stbi_uc *cur = a->out + stride*j; + stbi_uc *in = a->out + stride*j + x*out_n - img_width_bytes; + // unpack 1/2/4-bit into a 8-bit buffer. allows us to keep the common 8-bit path optimal at minimal cost for 1/2/4-bit + // png guarante byte alignment, if width is not multiple of 8/4/2 we'll decode dummy trailing data that will be skipped in the later loop + stbi_uc scale = (color == 0) ? stbi__depth_scale_table[depth] : 1; // scale grayscale values to 0..255 range + + // note that the final byte might overshoot and write more data than desired. + // we can allocate enough data that this never writes out of memory, but it + // could also overwrite the next scanline. can it overwrite non-empty data + // on the next scanline? yes, consider 1-pixel-wide scanlines with 1-bit-per-pixel. + // so we need to explicitly clamp the final ones + + if (depth == 4) { + for (k=x*img_n; k >= 2; k-=2, ++in) { + *cur++ = scale * ((*in >> 4) ); + *cur++ = scale * ((*in ) & 0x0f); + } + if (k > 0) *cur++ = scale * ((*in >> 4) ); + } else if (depth == 2) { + for (k=x*img_n; k >= 4; k-=4, ++in) { + *cur++ = scale * ((*in >> 6) ); + *cur++ = scale * ((*in >> 4) & 0x03); + *cur++ = scale * ((*in >> 2) & 0x03); + *cur++ = scale * ((*in ) & 0x03); + } + if (k > 0) *cur++ = scale * ((*in >> 6) ); + if (k > 1) *cur++ = scale * ((*in >> 4) & 0x03); + if (k > 2) *cur++ = scale * ((*in >> 2) & 0x03); + } else if (depth == 1) { + for (k=x*img_n; k >= 8; k-=8, ++in) { + *cur++ = scale * ((*in >> 7) ); + *cur++ = scale * ((*in >> 6) & 0x01); + *cur++ = scale * ((*in >> 5) & 0x01); + *cur++ = scale * ((*in >> 4) & 0x01); + *cur++ = scale * ((*in >> 3) & 0x01); + *cur++ = scale * ((*in >> 2) & 0x01); + *cur++ = scale * ((*in >> 1) & 0x01); + *cur++ = scale * ((*in ) & 0x01); + } + if (k > 0) *cur++ = scale * ((*in >> 7) ); + if (k > 1) *cur++ = scale * ((*in >> 6) & 0x01); + if (k > 2) *cur++ = scale * ((*in >> 5) & 0x01); + if (k > 3) *cur++ = scale * ((*in >> 4) & 0x01); + if (k > 4) *cur++ = scale * ((*in >> 3) & 0x01); + if (k > 5) *cur++ = scale * ((*in >> 2) & 0x01); + if (k > 6) *cur++ = scale * ((*in >> 1) & 0x01); + } + if (img_n != out_n) { + int q; + // insert alpha = 255 + cur = a->out + stride*j; + if (img_n == 1) { + for (q=x-1; q >= 0; --q) { + cur[q*2+1] = 255; + cur[q*2+0] = cur[q]; + } + } else { + STBI_ASSERT(img_n == 3); + for (q=x-1; q >= 0; --q) { + cur[q*4+3] = 255; + cur[q*4+2] = cur[q*3+2]; + cur[q*4+1] = cur[q*3+1]; + cur[q*4+0] = cur[q*3+0]; + } + } + } + } + } else if (depth == 16) { + // force the image data from big-endian to platform-native. + // this is done in a separate pass due to the decoding relying + // on the data being untouched, but could probably be done + // per-line during decode if care is taken. + stbi_uc *cur = a->out; + stbi__uint16 *cur16 = (stbi__uint16*)cur; + + for(i=0; i < x*y*out_n; ++i,cur16++,cur+=2) { + *cur16 = (cur[0] << 8) | cur[1]; + } + } + + return 1; +} + +static int stbi__create_png_image(stbi__png *a, stbi_uc *image_data, stbi__uint32 image_data_len, int out_n, int depth, int color, int interlaced) +{ + int bytes = (depth == 16 ? 2 : 1); + int out_bytes = out_n * bytes; + stbi_uc *final; + int p; + if (!interlaced) + return stbi__create_png_image_raw(a, image_data, image_data_len, out_n, a->s->img_x, a->s->img_y, depth, color); + + // de-interlacing + final = (stbi_uc *) stbi__malloc_mad3(a->s->img_x, a->s->img_y, out_bytes, 0); + 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) tc16[k] = (stbi__uint16)stbi__get16be(s); // copy the values as-is + } else { + for (k = 0; k < s->img_n; ++k) tc[k] = (stbi_uc)(stbi__get16be(s) & 255) * stbi__depth_scale_table[z->depth]; // non 8-bit images will be larger + } + } + break; + } + + case STBI__PNG_TYPE('I','D','A','T'): { + if (first) return stbi__err("first not IHDR", "Corrupt PNG"); + if (pal_img_n && !pal_len) return stbi__err("no PLTE","Corrupt PNG"); + if (scan == STBI__SCAN_header) { + // 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/util/string.h b/src/util/string.h new file mode 100644 index 0000000..b261e76 --- /dev/null +++ b/src/util/string.h @@ -0,0 +1,23 @@ +#pragma once +#include <string.h> + +#include "typedef.h" + +template <U32 N> +struct STR { + char data[N]; + enum { + size = N + }; + STR() { + memset( data, 0, N ); + } + STR( const char* str ) { + memcpy( data, str, N ); + } + STR( const STR<N>& str ) { + memcpy( data, str.data, N ); + } + + operator char*() { return data; } +}; diff --git a/src/util/thread.cpp b/src/util/thread.cpp new file mode 100644 index 0000000..74b01fc --- /dev/null +++ b/src/util/thread.cpp @@ -0,0 +1,2 @@ +#define THREAD_IMPLEMENTATION +#include "thread.h" diff --git a/src/util/thread.h b/src/util/thread.h new file mode 100644 index 0000000..72ef5fb --- /dev/null +++ b/src/util/thread.h @@ -0,0 +1,1518 @@ +/* +------------------------------------------------------------------------------ + Licensing information can be found at the end of the file. +------------------------------------------------------------------------------ + +thread.h - v0.3 - Cross platform threading functions for C/C++. + +Do this: + #define THREAD_IMPLEMENTATION +before you include this file in *one* C/C++ file to create the implementation. +*/ + +#ifndef thread_h +#define thread_h + +#ifndef THREAD_U64 + #define THREAD_U64 unsigned long long +#endif + +#define THREAD_STACK_SIZE_DEFAULT ( 0 ) +#define THREAD_SIGNAL_WAIT_INFINITE ( -1 ) +#define THREAD_QUEUE_WAIT_INFINITE ( -1 ) + +typedef void* thread_id_t; +typedef thread_id_t THREAD_IT; +thread_id_t thread_current_thread_id( void ); +void thread_yield( void ); +void thread_set_high_priority( void ); +void thread_exit( int return_code ); + +typedef void* thread_ptr_t; +typedef thread_ptr_t THREAD_PTR; +thread_ptr_t thread_create( int (*thread_proc)( void* ), void* user_data, int stack_size ); +void thread_destroy( thread_ptr_t thread ); +int thread_join( thread_ptr_t thread ); +int thread_detach( thread_ptr_t thread ); + +typedef union thread_mutex_t thread_mutex_t; +typedef thread_mutex_t THREAD_MUTEX; +void thread_mutex_init( thread_mutex_t* mutex ); +void thread_mutex_term( thread_mutex_t* mutex ); +void thread_mutex_lock( thread_mutex_t* mutex ); +void thread_mutex_unlock( thread_mutex_t* mutex ); + +typedef union thread_signal_t thread_signal_t; +typedef thread_signal_t THREAD_SIGNAL; +void thread_signal_init( thread_signal_t* signal ); +void thread_signal_term( thread_signal_t* signal ); +void thread_signal_raise( thread_signal_t* signal ); +int thread_signal_wait( thread_signal_t* signal, int timeout_ms ); + +typedef union thread_atomic_int_t thread_atomic_int_t; +typedef thread_atomic_int_t THREAD_ATOMIC_INT; +int thread_atomic_int_load( thread_atomic_int_t* atomic ); +void thread_atomic_int_store( thread_atomic_int_t* atomic, int desired ); +int thread_atomic_int_inc( thread_atomic_int_t* atomic ); +int thread_atomic_int_dec( thread_atomic_int_t* atomic ); +int thread_atomic_int_add( thread_atomic_int_t* atomic, int value ); +int thread_atomic_int_sub( thread_atomic_int_t* atomic, int value ); +int thread_atomic_int_swap( thread_atomic_int_t* atomic, int desired ); +int thread_atomic_int_compare_and_swap( thread_atomic_int_t* atomic, int expected, int desired ); + +typedef union thread_atomic_ptr_t thread_atomic_ptr_t; +typedef thread_atomic_ptr_t THREAD_ATOMIC_PTR; +void* thread_atomic_ptr_load( thread_atomic_ptr_t* atomic ); +void thread_atomic_ptr_store( thread_atomic_ptr_t* atomic, void* desired ); +void* thread_atomic_ptr_swap( thread_atomic_ptr_t* atomic, void* desired ); +void* thread_atomic_ptr_compare_and_swap( thread_atomic_ptr_t* atomic, void* expected, void* desired ); + +typedef union thread_timer_t thread_timer_t; +typedef thread_timer_t THREAD_TIMER; +void thread_timer_init( thread_timer_t* timer ); +void thread_timer_term( thread_timer_t* timer ); +void thread_timer_wait( thread_timer_t* timer, THREAD_U64 nanoseconds ); + +typedef void* thread_tls_t; +typedef thread_tls_t THREAD_TLS; +thread_tls_t thread_tls_create( void ); +void thread_tls_destroy( thread_tls_t tls ); +void thread_tls_set( thread_tls_t tls, void* value ); +void* thread_tls_get( thread_tls_t tls ); + +typedef struct thread_queue_t thread_queue_t; +typedef thread_queue_t THREAD_QUEUE; +void thread_queue_init( thread_queue_t* queue, int size, void** values, int count ); +void thread_queue_term( thread_queue_t* queue ); +int thread_queue_produce( thread_queue_t* queue, void* value, int timeout_ms ); +void* thread_queue_consume( thread_queue_t* queue, int timeout_ms ); +int thread_queue_count( thread_queue_t* queue ); + +#endif /* thread_h */ + + +/** + +thread.h +======== + +Cross platform threading functions for C/C++. + +Example +------- + +Here's a basic sample program which starts a second thread which just waits and prints a message. + + #define THREAD_IMPLEMENTATION + #include "thread.h" + + #include <stdio.h> // for printf + + int thread_proc( void* user_data) { + thread_timer_t timer; + thread_timer_init( &timer ); + + int count = 0; + thread_atomic_int_t* exit_flag = (thread_atomic_int_t*) user_data; + while( thread_atomic_int_load( exit_flag ) == 0 ) { + printf( "Thread... " ); + thread_timer_wait( &timer, 1000000000 ); // sleep for a second + ++count; + } + + thread_timer_term( &timer ); + printf( "Done\n" ); + return count; + } + + int main( int argc, char** argv ) { + thread_atomic_int_t exit_flag; + thread_atomic_int_store( &exit_flag, 0 ); + + thread_ptr_t thread = thread_create( thread_proc, &exit_flag, "Example thread", THREAD_STACK_SIZE_DEFAULT ); + + thread_timer_t timer; + thread_timer_init( &timer ); + for( int i = 0; i < 5; ++i ) { + printf( "Main... " ); + thread_timer_wait( &timer, 2000000000 ); // sleep for two seconds + } + thread_timer_term( &timer ); + + thread_atomic_int_store( &exit_flag, 1 ); // signal thread to exit + int retval = thread_join( thread ); + + printf( "Count: %d\n", retval ); + + thread_destroy( thread ); + return retval; + } + + +API Documentation +----------------- + +thread.h is a single-header library, and does not need any .lib files or other binaries, or any build scripts. To use it, +you just include thread.h to get the API declarations. To get the definitions, you must include thread.h from *one* +single C or C++ file, and #define the symbol `THREAD_IMPLEMENTATION` before you do. + + +### Customization + +thread.h allows for specifying the exact type of 64-bit unsigned integer to be used in its API. By default, it is +defined as `unsigned long long`, but as this is not a standard type on all compilers, you can redefine it by #defining +THREAD_U64 before including thread.h. This is useful if you, for example, use the types from `<stdint.h>` in the rest of +your program, and you want thread.h to use compatible types. In this case, you would include thread.h using the +following code: + + #define THREAD_U64 uint64_t + #include "thread.h" + +Note that when customizing this data type, you need to use the same definition in every place where you include +thread.h, as it affect the declarations as well as the definitions. + + +thread_current_thread_id +------------------------ + + thread_id_t thread_current_thread_id( void ) + +Returns a unique identifier for the calling thread. After the thread terminates, the id might be reused for new threads. + + +thread_yield +------------ + + void thread_yield( void ) + +Makes the calling thread yield execution to another thread. The operating system controls which thread is switched to. + + +thread_set_high_priority +------------------------ + + void thread_set_high_priority( void ) + +When created, threads are set to run at normal priority. In some rare cases, such as a sound buffer update loop, it can +be necessary to have one thread of your application run on a higher priority than the rest. Calling +`thread_set_high_priority` will raise the priority of the calling thread, giving it a chance to be run more often. +Do not increase the priority of a thread unless you absolutely have to, as it can negatively affect performance if used +without care. + + +thread_exit +----------- + + void thread_exit( int return_code ) + +Exits the calling thread, as if you had done `return return_code;` from the main body of the thread function. + + +thread_create +------------- + + thread_ptr_t thread_create( int (*thread_proc)( void* ), void* user_data, int stack_size ) + +Creates a new thread running the `thread_proc` function, passing the `user_data` through to it. The thread will have +the stack size specified in the `stack_size` parameter. To get the operating system default stack size, use the +defined constant `THREAD_STACK_SIZE_DEFAULT`. When returning from the thread_proc function, the value you return can +be received in another thread by calling thread_join. `thread_create` returns a pointer to the thread instance, which +can be used as a parameter to the functions `thread_destroy` and `thread_join`. + + +thread_destroy +-------------- + + void thread_destroy( thread_ptr_t thread ) + +Destroys a thread that was created by calling `thread_create`. Make sure the thread has exited before you attempt to +destroy it. This can be accomplished by calling `thread_join`. It is not possible for force termination of a thread by +calling `thread_destroy`. + + +thread_join +----------- + + int thread_join( thread_ptr_t thread ) + +Waits for the specified thread to exit. Returns the value which the thread returned when exiting. + + +thread_detach +------------- + + int thread_detach( thread_ptr_t thread ) + +Marks the thread as detached. When a detached thread terminates, its resources are automatically released back to the +system without the need for another thread to join with the terminated thread. + + +thread_mutex_init +----------------- + + void thread_mutex_init( thread_mutex_t* mutex ) + +Initializes the specified mutex instance, preparing it for use. A mutex can be used to lock sections of code, such that +it can only be run by one thread at a time. + + +thread_mutex_term +----------------- + + void thread_mutex_term( thread_mutex_t* mutex ) + +Terminates the specified mutex instance, releasing any system resources held by it. + + +thread_mutex_lock +----------------- + + void thread_mutex_lock( thread_mutex_t* mutex ) + +Takes an exclusive lock on a mutex. If the lock is already taken by another thread, `thread_mutex_lock` will yield the +calling thread and wait for the lock to become available before returning. The mutex must be initialized by calling +`thread_mutex_init` before it can be locked. + + +thread_mutex_unlock +------------------- + + void thread_mutex_unlock( thread_mutex_t* mutex ) + +Releases a lock taken by calling `thread_mutex_lock`. + + +thread_signal_init +------------------ + + void thread_signal_init( thread_signal_t* signal ) + +Initializes the specified signal instance, preparing it for use. A signal works like a flag, which can be waited on by +one thread, until it is raised from another thread. + + +thread_signal_term +------------------ + + void thread_signal_term( thread_signal_t* signal ) + +Terminates the specified signal instance, releasing any system resources held by it. + + +thread_signal_raise +------------------- + + void thread_signal_raise( thread_signal_t* signal ) + +Raise the specified signal. Other threads waiting for the signal will proceed. + + +thread_signal_wait +------------------ + + int thread_signal_wait( thread_signal_t* signal, int timeout_ms ) + +Waits for a signal to be raised, or until `timeout_ms` milliseconds have passed. If the wait timed out, a value of 0 is +returned, otherwise a non-zero value is returned. If the `timeout_ms` parameter is THREAD_SIGNAL_WAIT_INFINITE, +`thread_signal_wait` waits indefinitely. + + +thread_atomic_int_load +---------------------- + + int thread_atomic_int_load( thread_atomic_int_t* atomic ) + +Returns the value of `atomic` as an atomic operation. + + +thread_atomic_int_store +----------------------- + + void thread_atomic_int_store( thread_atomic_int_t* atomic, int desired ) + +Sets the value of `atomic` as an atomic operation. + + +thread_atomic_int_inc +--------------------- + + int thread_atomic_int_inc( thread_atomic_int_t* atomic ) + +Increments the value of `atomic` by one, as an atomic operation. Returns the value `atomic` had before the operation. + + +thread_atomic_int_dec +--------------------- + + int thread_atomic_int_dec( thread_atomic_int_t* atomic ) + +Decrements the value of `atomic` by one, as an atomic operation. Returns the value `atomic` had before the operation. + + +thread_atomic_int_add +--------------------- + + int thread_atomic_int_add( thread_atomic_int_t* atomic, int value ) + +Adds the specified value to `atomic`, as an atomic operation. Returns the value `atomic` had before the operation. + + +thread_atomic_int_sub +--------------------- + + int thread_atomic_int_sub( thread_atomic_int_t* atomic, int value ) + +Subtracts the specified value to `atomic`, as an atomic operation. Returns the value `atomic` had before the operation. + + +thread_atomic_int_swap +---------------------- + + int thread_atomic_int_swap( thread_atomic_int_t* atomic, int desired ) + +Sets the value of `atomic` as an atomic operation. Returns the value `atomic` had before the operation. + + +thread_atomic_int_compare_and_swap +---------------------------------- + + int thread_atomic_int_compare_and_swap( thread_atomic_int_t* atomic, int expected, int desired ) + +Compares the value of `atomic` to the value of `expected`, and if they match, sets the vale of `atomic` to `desired`, +all as an atomic operation. Returns the value `atomic` had before the operation. + + +thread_atomic_ptr_load +---------------------- + + void* thread_atomic_ptr_load( thread_atomic_ptr_t* atomic ) + +Returns the value of `atomic` as an atomic operation. + + +thread_atomic_ptr_store +----------------------- + + void thread_atomic_ptr_store( thread_atomic_ptr_t* atomic, void* desired ) + +Sets the value of `atomic` as an atomic operation. + + +thread_atomic_ptr_swap +---------------------- + + void* thread_atomic_ptr_swap( thread_atomic_ptr_t* atomic, void* desired ) + +Sets the value of `atomic` as an atomic operation. Returns the value `atomic` had before the operation. + + +thread_atomic_ptr_compare_and_swap +---------------------------------- + + void* thread_atomic_ptr_compare_and_swap( thread_atomic_ptr_t* atomic, void* expected, void* desired ) + +Compares the value of `atomic` to the value of `expected`, and if they match, sets the vale of `atomic` to `desired`, +all as an atomic operation. Returns the value `atomic` had before the operation. + + +thread_timer_init +----------------- + + void thread_timer_init( thread_timer_t* timer ) + +Initializes the specified timer instance, preparing it for use. A timer can be used to sleep a thread for a high +precision duration. + + +thread_timer_term +----------------- + + void thread_timer_term( thread_timer_t* timer ) + +Terminates the specified timer instance, releasing any system resources held by it. + + +thread_timer_wait +----------------- + + void thread_timer_wait( thread_timer_t* timer, THREAD_U64 nanoseconds ) + +Waits until `nanoseconds` amount of time have passed, before returning. + + +thread_tls_create +----------------- + + thread_tls_t thread_tls_create( void ) + +Creates a thread local storage (TLS) index. Once created, each thread has its own value for that TLS index, which can +be set or retrieved individually. + + +thread_tls_destroy +------------------ + + void thread_tls_destroy( thread_tls_t tls ) + +Destroys the specified TLS index. No further calls to `thread_tls_set` or `thread_tls_get` are valid after this. + + +thread_tls_set +-------------- + + void thread_tls_set( thread_tls_t tls, void* value ) + +Stores a value in the calling thread's slot for the specified TLS index. Each thread has its own value for each TLS +index. + + +thread_tls_get +-------------- + + void* thread_tls_get( thread_tls_t tls ) + +Retrieves the value from the calling thread's slot for the specified TLS index. Each thread has its own value for each +TLS index. + + +thread_queue_init +----------------- + + void thread_queue_init( thread_queue_t* queue, int size, void** values, int count ) + +Initializes the specified queue instance, preparing it for use. The queue is a lock-free (but not wait-free) +single-producer/single-consumer queue - it will not acquire any locks as long as there is space for adding or items to +be consume, but will lock and wait when there is not. The `size` parameter specifies the number of elements in the +queue. The `values` parameter is an array of queue slots (`size` elements in length), each being of type `void*`. If +the queue is initially empty, the `count` parameter should be 0, otherwise it indicates the number of entires, from the +start of the `values` array, that the queue is initialized with. The `values` array is not copied, and must remain valid +until `thread_queue_term` is called. + + +thread_queue_term +----------------- + + void thread_queue_term( thread_queue_t* queue ) + +Terminates the specified queue instance, releasing any system resources held by it. + + +thread_queue_produce +-------------------- + + int thread_queue_produce( thread_queue_t* queue, void* value, int timeout_ms ) + +Adds an element to a single-producer/single-consumer queue. If there is space in the queue to add another element, no +lock will be taken. If the queue is full, calling thread will sleep until an element is consumed from another thread, +before adding the element, or until `timeout_ms` milliseconds have passed. If the wait timed out, a value of 0 is +returned, otherwise a non-zero value is returned. If the `timeout_ms` parameter is THREAD_QUEUE_WAIT_INFINITE, +`thread_queue_produce` waits indefinitely. + + +thread_queue_consume +-------------------- + + void* thread_queue_consume( thread_queue_t* queue, int timeout_ms ) + +Removes an element from a single-producer/single-consumer queue. If the queue contains at least one element, no lock +will be taken. If the queue is empty, the calling thread will sleep until an element is added from another thread, or +until `timeout_ms` milliseconds have passed. If the wait timed out, a value of NULL is returned, otherwise +`thread_queue_consume` returns the value that was removed from the queue. If the `timeout_ms` parameter is +THREAD_QUEUE_WAIT_INFINITE, `thread_queue_consume` waits indefinitely. + + +thread_queue_count +------------------ + + int thread_queue_count( thread_queue_t* queue ) + +Returns the number of elements currently held in a single-producer/single-consumer queue. Be aware that by the time you +get the count, it might have changed by another thread calling consume or produce, so use with care. + +*/ + + +/* +---------------------- + IMPLEMENTATION +---------------------- +*/ + +#ifndef thread_impl +#define thread_impl + +union thread_mutex_t + { + void* align; + char data[ 64 ]; + }; + +union thread_signal_t + { + void* align; + char data[ 116 ]; + }; + +union thread_atomic_int_t + { + void* align; + long i; + }; + +union thread_atomic_ptr_t + { + void* ptr; + }; + +union thread_timer_t + { + void* data; + char d[ 8 ]; + }; + +struct thread_queue_t + { + thread_signal_t data_ready; + thread_signal_t space_open; + thread_atomic_int_t count; + thread_atomic_int_t head; + thread_atomic_int_t tail; + void** values; + int size; + #ifndef NDEBUG + thread_atomic_int_t id_produce_is_set; + thread_id_t id_produce; + thread_atomic_int_t id_consume_is_set; + thread_id_t id_consume; + #endif + }; + +#endif /* thread_impl */ + + + +#ifdef THREAD_IMPLEMENTATION +#undef THREAD_IMPLEMENTATION + +#ifndef THREAD_ASSERT + #undef _CRT_NONSTDC_NO_DEPRECATE + #define _CRT_NONSTDC_NO_DEPRECATE + #undef _CRT_SECURE_NO_WARNINGS + #define _CRT_SECURE_NO_WARNINGS + #include <assert.h> + #define THREAD_ASSERT( expression, message ) assert( ( expression ) && ( message ) ) +#endif + + +#if defined( _WIN32 ) + + #pragma comment( lib, "winmm.lib" ) + + #define _CRT_NONSTDC_NO_DEPRECATE + #define _CRT_SECURE_NO_WARNINGS + + #if !defined( _WIN32_WINNT ) || _WIN32_WINNT < 0x0501 + #undef _WIN32_WINNT + #define _WIN32_WINNT 0x501// requires Windows XP minimum + #endif + + #define _WINSOCKAPI_ + #pragma warning( push ) + #pragma warning( disable: 4619 ) + #pragma warning( disable: 4668 ) // 'symbol' is not defined as a preprocessor macro, replacing with '0' for 'directives' + #pragma warning( disable: 4768 ) // __declspec attributes before linkage specification are ignored + #pragma warning( disable: 4255 ) // 'function' : no function prototype given: converting '()' to '(void)' + #include <windows.h> + #pragma warning( pop ) + + +#elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + #include <pthread.h> + #include <errno.h> + #include <string.h> + #include <sys/time.h> + #include <stdint.h> + +#else + #error Unknown platform. +#endif + + + +thread_id_t thread_current_thread_id( void ) + { + #if defined( _WIN32 ) + + return (void*) (uintptr_t)GetCurrentThreadId(); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + return (void*) pthread_self(); + + #else + #error Unknown platform. + #endif + } + + +void thread_yield( void ) + { + #if defined( _WIN32 ) + + SwitchToThread(); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + sched_yield(); + + #else + #error Unknown platform. + #endif + } + + +void thread_exit( int return_code ) + { + #if defined( _WIN32 ) + + ExitThread( (DWORD) return_code ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + pthread_exit( (void*)(uintptr_t) return_code ); + + #else + #error Unknown platform. + #endif + } + + +thread_ptr_t thread_create( int (*thread_proc)( void* ), void* user_data, int stack_size ) + { + #if defined( _WIN32 ) + + DWORD thread_id; + HANDLE handle = CreateThread( NULL, stack_size > 0 ? (size_t)stack_size : 0U, + (LPTHREAD_START_ROUTINE)(uintptr_t) thread_proc, user_data, 0, &thread_id ); + if( !handle ) return NULL; + + return (thread_ptr_t) handle; + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + pthread_t thread; + if( 0 != pthread_create( &thread, NULL, ( void* (*)( void * ) ) thread_proc, user_data ) ) + return NULL; + + return (thread_ptr_t) thread; + + #else + #error Unknown platform. + #endif + } + + +void thread_destroy( thread_ptr_t thread ) + { + #if defined( _WIN32 ) + + WaitForSingleObject( (HANDLE) thread, INFINITE ); + CloseHandle( (HANDLE) thread ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + pthread_join( (pthread_t) thread, NULL ); + + #else + #error Unknown platform. + #endif + } + + +int thread_join( thread_ptr_t thread ) + { + #if defined( _WIN32 ) + + WaitForSingleObject( (HANDLE) thread, INFINITE ); + DWORD retval; + GetExitCodeThread( (HANDLE) thread, &retval ); + return (int) retval; + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + void* retval; + pthread_join( (pthread_t) thread, &retval ); + return (int)(uintptr_t) retval; + + #else + #error Unknown platform. + #endif + } + + +int thread_detach( thread_ptr_t thread ) + { + #if defined( _WIN32 ) + + return CloseHandle( (HANDLE) thread ) != 0; + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + return pthread_detach( (pthread_t) thread ) == 0; + + #else + #error Unknown platform. + #endif + } + + +void thread_set_high_priority( void ) + { + #if defined( _WIN32 ) + + SetThreadPriority( GetCurrentThread(), THREAD_PRIORITY_HIGHEST ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + struct sched_param sp; + memset( &sp, 0, sizeof( sp ) ); + sp.sched_priority = sched_get_priority_min( SCHED_RR ); + pthread_setschedparam( pthread_self(), SCHED_RR, &sp); + + #else + #error Unknown platform. + #endif + } + + +void thread_mutex_init( thread_mutex_t* mutex ) + { + #if defined( _WIN32 ) + + // Compile-time size check + #pragma warning( push ) + #pragma warning( disable: 4214 ) // nonstandard extension used: bit field types other than int + struct x { char thread_mutex_type_too_small : ( sizeof( thread_mutex_t ) < sizeof( CRITICAL_SECTION ) ? 0 : 1 ); }; + #pragma warning( pop ) + + InitializeCriticalSectionAndSpinCount( (CRITICAL_SECTION*) mutex, 32 ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + // Compile-time size check + struct x { char thread_mutex_type_too_small : ( sizeof( thread_mutex_t ) < sizeof( pthread_mutex_t ) ? 0 : 1 ); }; + + pthread_mutex_init( (pthread_mutex_t*) mutex, NULL ); + + #else + #error Unknown platform. + #endif + } + + +void thread_mutex_term( thread_mutex_t* mutex ) + { + #if defined( _WIN32 ) + + DeleteCriticalSection( (CRITICAL_SECTION*) mutex ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + pthread_mutex_destroy( (pthread_mutex_t*) mutex ); + + #else + #error Unknown platform. + #endif + } + + +void thread_mutex_lock( thread_mutex_t* mutex ) + { + #if defined( _WIN32 ) + + EnterCriticalSection( (CRITICAL_SECTION*) mutex ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + pthread_mutex_lock( (pthread_mutex_t*) mutex ); + + #else + #error Unknown platform. + #endif + } + + +void thread_mutex_unlock( thread_mutex_t* mutex ) + { + #if defined( _WIN32 ) + + LeaveCriticalSection( (CRITICAL_SECTION*) mutex ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + pthread_mutex_unlock( (pthread_mutex_t*) mutex ); + + #else + #error Unknown platform. + #endif + } + + +struct thread_internal_signal_t + { + #if defined( _WIN32 ) + + #if _WIN32_WINNT >= 0x0600 + CRITICAL_SECTION mutex; + CONDITION_VARIABLE condition; + int value; + #else + #pragma message( "Warning: _WIN32_WINNT < 0x0600 - condition variables not available" ) + HANDLE event; + #endif + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + pthread_mutex_t mutex; + pthread_cond_t condition; + int value; + + #else + #error Unknown platform. + #endif + }; + + +void thread_signal_init( thread_signal_t* signal ) + { + // Compile-time size check + #if defined( _MSC_VER ) + #pragma warning( push ) + #pragma warning( disable: 4214 ) // nonstandard extension used: bit field types other than int + #endif + struct x { char thread_signal_type_too_small : ( sizeof( thread_signal_t ) < sizeof( struct thread_internal_signal_t ) ? 0 : 1 ); }; + #if defined( _MSC_VER ) + #pragma warning( pop ) + #endif + + struct thread_internal_signal_t* internal = (struct thread_internal_signal_t*) signal; + + #if defined( _WIN32 ) + + #if _WIN32_WINNT >= 0x0600 + InitializeCriticalSectionAndSpinCount( &internal->mutex, 32 ); + InitializeConditionVariable( &internal->condition ); + internal->value = 0; + #else + internal->event = CreateEvent( NULL, FALSE, FALSE, NULL ); + #endif + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + pthread_mutex_init( &internal->mutex, NULL ); + pthread_cond_init( &internal->condition, NULL ); + internal->value = 0; + + #else + #error Unknown platform. + #endif + } + + + void thread_signal_term( thread_signal_t* signal ) + { + struct thread_internal_signal_t* internal = (struct thread_internal_signal_t*) signal; + + #if defined( _WIN32 ) + + #if _WIN32_WINNT >= 0x0600 + DeleteCriticalSection( &internal->mutex ); + #else + CloseHandle( internal->event ); + #endif + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + pthread_mutex_destroy( &internal->mutex ); + pthread_cond_destroy( &internal->condition ); + + #else + #error Unknown platform. + #endif + } + + +void thread_signal_raise( thread_signal_t* signal ) + { + struct thread_internal_signal_t* internal = (struct thread_internal_signal_t*) signal; + + #if defined( _WIN32 ) + + #if _WIN32_WINNT >= 0x0600 + EnterCriticalSection( &internal->mutex ); + internal->value = 1; + LeaveCriticalSection( &internal->mutex ); + WakeConditionVariable( &internal->condition ); + #else + SetEvent( internal->event ); + #endif + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + pthread_mutex_lock( &internal->mutex ); + internal->value = 1; + pthread_mutex_unlock( &internal->mutex ); + pthread_cond_signal( &internal->condition ); + + #else + #error Unknown platform. + #endif + } + + +int thread_signal_wait( thread_signal_t* signal, int timeout_ms ) + { + struct thread_internal_signal_t* internal = (struct thread_internal_signal_t*) signal; + + #if defined( _WIN32 ) + + #if _WIN32_WINNT >= 0x0600 + int timed_out = 0; + EnterCriticalSection( &internal->mutex ); + while( internal->value == 0 ) + { + BOOL res = SleepConditionVariableCS( &internal->condition, &internal->mutex, timeout_ms < 0 ? INFINITE : timeout_ms ); + if( !res && GetLastError() == ERROR_TIMEOUT ) { timed_out = 1; break; } + } + internal->value = 0; + LeaveCriticalSection( &internal->mutex ); + return !timed_out; + #else + int failed = WAIT_OBJECT_0 != WaitForSingleObject( internal->event, timeout_ms < 0 ? INFINITE : timeout_ms ); + return !failed; + #endif + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + struct timespec ts; + if( timeout_ms >= 0 ) + { + struct timeval tv; + gettimeofday( &tv, NULL ); + ts.tv_sec = time( NULL ) + timeout_ms / 1000; + ts.tv_nsec = tv.tv_usec * 1000 + 1000 * 1000 * ( timeout_ms % 1000 ); + ts.tv_sec += ts.tv_nsec / ( 1000 * 1000 * 1000 ); + ts.tv_nsec %= ( 1000 * 1000 * 1000 ); + } + + int timed_out = 0; + pthread_mutex_lock( &internal->mutex ); + while( internal->value == 0 ) + { + if( timeout_ms < 0 ) + pthread_cond_wait( &internal->condition, &internal->mutex ); + else if( pthread_cond_timedwait( &internal->condition, &internal->mutex, &ts ) == ETIMEDOUT ) + { + timed_out = 1; + break; + } + + } + if( !timed_out ) internal->value = 0; + pthread_mutex_unlock( &internal->mutex ); + return !timed_out; + + #else + #error Unknown platform. + #endif + } + + +int thread_atomic_int_load( thread_atomic_int_t* atomic ) + { + #if defined( _WIN32 ) + + return InterlockedCompareExchange( &atomic->i, 0, 0 ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + return (int)__sync_fetch_and_add( &atomic->i, 0 ); + + #else + #error Unknown platform. + #endif + } + + +void thread_atomic_int_store( thread_atomic_int_t* atomic, int desired ) + { + #if defined( _WIN32 ) + + InterlockedExchange( &atomic->i, desired ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + __sync_fetch_and_and( &atomic->i, 0 ); + __sync_fetch_and_or( &atomic->i, desired ); + + #else + #error Unknown platform. + #endif + } + + +int thread_atomic_int_inc( thread_atomic_int_t* atomic ) + { + #if defined( _WIN32 ) + + return InterlockedIncrement( &atomic->i ) - 1; + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + return (int)__sync_fetch_and_add( &atomic->i, 1 ); + + #else + #error Unknown platform. + #endif + } + + +int thread_atomic_int_dec( thread_atomic_int_t* atomic ) + { + #if defined( _WIN32 ) + + return InterlockedDecrement( &atomic->i ) + 1; + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + return (int)__sync_fetch_and_sub( &atomic->i, 1 ); + + #else + #error Unknown platform. + #endif + } + + +int thread_atomic_int_add( thread_atomic_int_t* atomic, int value ) + { + #if defined( _WIN32 ) + + return InterlockedExchangeAdd ( &atomic->i, value ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + return (int)__sync_fetch_and_add( &atomic->i, value ); + + #else + #error Unknown platform. + #endif + } + + +int thread_atomic_int_sub( thread_atomic_int_t* atomic, int value ) + { + #if defined( _WIN32 ) + + return InterlockedExchangeAdd( &atomic->i, -value ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + return (int)__sync_fetch_and_sub( &atomic->i, value ); + + #else + #error Unknown platform. + #endif + } + + +int thread_atomic_int_swap( thread_atomic_int_t* atomic, int desired ) + { + #if defined( _WIN32 ) + + return InterlockedExchange( &atomic->i, desired ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + int old = (int)__sync_lock_test_and_set( &atomic->i, desired ); + __sync_lock_release( &atomic->i ); + return old; + + #else + #error Unknown platform. + #endif + } + + +int thread_atomic_int_compare_and_swap( thread_atomic_int_t* atomic, int expected, int desired ) + { + #if defined( _WIN32 ) + + return InterlockedCompareExchange( &atomic->i, desired, expected ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + return (int)__sync_val_compare_and_swap( &atomic->i, expected, desired ); + + #else + #error Unknown platform. + #endif + } + + +void* thread_atomic_ptr_load( thread_atomic_ptr_t* atomic ) + { + #if defined( _WIN32 ) + + return InterlockedCompareExchangePointer( &atomic->ptr, 0, 0 ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + return __sync_fetch_and_add( &atomic->ptr, 0 ); + + #else + #error Unknown platform. + #endif + } + + +void thread_atomic_ptr_store( thread_atomic_ptr_t* atomic, void* desired ) + { + #if defined( _WIN32 ) + + #pragma warning( push ) + #pragma warning( disable: 4302 ) // 'type cast' : truncation from 'void *' to 'LONG' + #pragma warning( disable: 4311 ) // pointer truncation from 'void *' to 'LONG' + #pragma warning( disable: 4312 ) // conversion from 'LONG' to 'PVOID' of greater size + InterlockedExchangePointer( &atomic->ptr, desired ); + #pragma warning( pop ) + + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + __sync_lock_test_and_set( &atomic->ptr, desired ); + __sync_lock_release( &atomic->ptr ); + + #else + #error Unknown platform. + #endif + } + + +void* thread_atomic_ptr_swap( thread_atomic_ptr_t* atomic, void* desired ) + { + #if defined( _WIN32 ) + + #pragma warning( push ) + #pragma warning( disable: 4302 ) // 'type cast' : truncation from 'void *' to 'LONG' + #pragma warning( disable: 4311 ) // pointer truncation from 'void *' to 'LONG' + #pragma warning( disable: 4312 ) // conversion from 'LONG' to 'PVOID' of greater size + return InterlockedExchangePointer( &atomic->ptr, desired ); + #pragma warning( pop ) + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + void* old = __sync_lock_test_and_set( &atomic->ptr, desired ); + __sync_lock_release( &atomic->ptr ); + return old; + + #else + #error Unknown platform. + #endif + } + + +void* thread_atomic_ptr_compare_and_swap( thread_atomic_ptr_t* atomic, void* expected, void* desired ) + { + #if defined( _WIN32 ) + + return InterlockedCompareExchangePointer( &atomic->ptr, desired, expected ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + return __sync_val_compare_and_swap( &atomic->ptr, expected, desired ); + + #else + #error Unknown platform. + #endif + } + + +void thread_timer_init( thread_timer_t* timer ) + { + #if defined( _WIN32 ) + + // Compile-time size check + #pragma warning( push ) + #pragma warning( disable: 4214 ) // nonstandard extension used: bit field types other than int + struct x { char thread_timer_type_too_small : ( sizeof( thread_mutex_t ) < sizeof( HANDLE ) ? 0 : 1 ); }; + #pragma warning( pop ) + + TIMECAPS tc; + if( timeGetDevCaps( &tc, sizeof( TIMECAPS ) ) == TIMERR_NOERROR ) + timeBeginPeriod( tc.wPeriodMin ); + + *(HANDLE*)timer = CreateWaitableTimer( NULL, TRUE, NULL ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + // Nothing + + #else + #error Unknown platform. + #endif + } + + +void thread_timer_term( thread_timer_t* timer ) + { + #if defined( _WIN32 ) + + CloseHandle( *(HANDLE*)timer ); + + TIMECAPS tc; + if( timeGetDevCaps( &tc, sizeof( TIMECAPS ) ) == TIMERR_NOERROR ) + timeEndPeriod( tc.wPeriodMin ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + // Nothing + + #else + #error Unknown platform. + #endif + } + + +void thread_timer_wait( thread_timer_t* timer, THREAD_U64 nanoseconds ) + { + #if defined( _WIN32 ) + + LARGE_INTEGER due_time; + due_time.QuadPart = - (LONGLONG) ( nanoseconds / 100 ); + BOOL b = SetWaitableTimer( *(HANDLE*)timer, &due_time, 0, 0, 0, FALSE ); + (void) b; + WaitForSingleObject( *(HANDLE*)timer, INFINITE ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + struct timespec rem; + struct timespec req; + req.tv_sec = nanoseconds / 1000000000ULL; + req.tv_nsec = nanoseconds - req.tv_sec * 1000000000ULL; + while( nanosleep( &req, &rem ) ) + req = rem; + + #else + #error Unknown platform. + #endif + } + + +thread_tls_t thread_tls_create( void ) + { + #if defined( _WIN32 ) + + DWORD tls = TlsAlloc(); + if( tls == TLS_OUT_OF_INDEXES ) + return NULL; + else + return (thread_tls_t) (uintptr_t) tls; + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + pthread_key_t tls; + if( pthread_key_create( &tls, NULL ) == 0 ) + return (thread_tls_t) (uintptr_t) tls; + else + return NULL; + + #else + #error Unknown platform. + #endif + } + + +void thread_tls_destroy( thread_tls_t tls ) + { + #if defined( _WIN32 ) + + TlsFree( (DWORD) (uintptr_t) tls ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + pthread_key_delete( (pthread_key_t) (uintptr_t) tls ); + + #else + #error Unknown platform. + #endif + } + + +void thread_tls_set( thread_tls_t tls, void* value ) + { + #if defined( _WIN32 ) + + TlsSetValue( (DWORD) (uintptr_t) tls, value ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + pthread_setspecific( (pthread_key_t) (uintptr_t) tls, value ); + + #else + #error Unknown platform. + #endif + } + + +void* thread_tls_get( thread_tls_t tls ) + { + #if defined( _WIN32 ) + + return TlsGetValue( (DWORD) (uintptr_t) tls ); + + #elif defined( __linux__ ) || defined( __APPLE__ ) || defined( __ANDROID__ ) + + return pthread_getspecific( (pthread_key_t) (uintptr_t) tls ); + + #else + #error Unknown platform. + #endif + } + + +void thread_queue_init( thread_queue_t* queue, int size, void** values, int count ) + { + queue->values = values; + thread_signal_init( &queue->data_ready ); + thread_signal_init( &queue->space_open ); + thread_atomic_int_store( &queue->head, 0 ); + thread_atomic_int_store( &queue->tail, count > size ? size : count ); + thread_atomic_int_store( &queue->count, count > size ? size : count ); + queue->size = size; + #ifndef NDEBUG + thread_atomic_int_store( &queue->id_produce_is_set, 0 ); + thread_atomic_int_store( &queue->id_consume_is_set, 0 ); + #endif + } + + +void thread_queue_term( thread_queue_t* queue ) + { + thread_signal_term( &queue->space_open ); + thread_signal_term( &queue->data_ready ); + } + + +int thread_queue_produce( thread_queue_t* queue, void* value, int timeout_ms ) + { + #ifndef NDEBUG + if( thread_atomic_int_compare_and_swap( &queue->id_produce_is_set, 0, 1 ) == 0 ) + queue->id_produce = thread_current_thread_id(); + THREAD_ASSERT( thread_current_thread_id() == queue->id_produce, "thread_queue_produce called from multiple threads" ); + #endif + while( thread_atomic_int_load( &queue->count ) == queue->size ) // TODO: fix signal so that this can be an "if" instead of "while" + { + if( timeout_ms == 0 ) return 0; + if( thread_signal_wait( &queue->space_open, timeout_ms == THREAD_QUEUE_WAIT_INFINITE ? THREAD_SIGNAL_WAIT_INFINITE : timeout_ms ) == 0 ) + return 0; + } + int tail = thread_atomic_int_inc( &queue->tail ); + queue->values[ tail % queue->size ] = value; + if( thread_atomic_int_inc( &queue->count ) == 0 ) + thread_signal_raise( &queue->data_ready ); + return 1; + } + + +void* thread_queue_consume( thread_queue_t* queue, int timeout_ms ) + { + #ifndef NDEBUG + if( thread_atomic_int_compare_and_swap( &queue->id_consume_is_set, 0, 1 ) == 0 ) + queue->id_consume = thread_current_thread_id(); + THREAD_ASSERT( thread_current_thread_id() == queue->id_consume, "thread_queue_consume called from multiple threads" ); + #endif + while( thread_atomic_int_load( &queue->count ) == 0 ) // TODO: fix signal so that this can be an "if" instead of "while" + { + if( timeout_ms == 0 ) return NULL; + if( thread_signal_wait( &queue->data_ready, timeout_ms == THREAD_QUEUE_WAIT_INFINITE ? THREAD_SIGNAL_WAIT_INFINITE : timeout_ms ) == 0 ) + return NULL; + } + int head = thread_atomic_int_inc( &queue->head ); + void* retval = queue->values[ head % queue->size ]; + if( thread_atomic_int_dec( &queue->count ) == queue->size ) + thread_signal_raise( &queue->space_open ); + return retval; + } + + +int thread_queue_count( thread_queue_t* queue ) + { + return thread_atomic_int_load( &queue->count ); + } + + +#endif /* THREAD_IMPLEMENTATION */ + +/* +revision history: + 0.3 set_high_priority API change. Fixed spurious wakeup bug in signal. Added + timeout param to queue produce/consume. Various cleanup and trivial fixes. + 0.2 first publicly released version +*/ + +/* +------------------------------------------------------------------------------ + +This software is available under 2 licenses - you may choose the one you like. + +------------------------------------------------------------------------------ + +ALTERNATIVE A - MIT License + +Copyright (c) 2015 Mattias Gustavsson + +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/util/typedef.h b/src/util/typedef.h new file mode 100644 index 0000000..53c4010 --- /dev/null +++ b/src/util/typedef.h @@ -0,0 +1,58 @@ +#pragma once +#ifdef DEBUG +#include <stdio.h> +#endif + +enum STAT { + STAT_OK, + STAT_ERR, + STAT_BREAK, + STAT_ALREADYEXISTS, +}; + +#define OK( x ) ( (STAT)( x ) == STAT_OK ) + +typedef char I8; +typedef short I16; +typedef int I32; +typedef long long I64; + +typedef unsigned char U8; +typedef unsigned short U16; +typedef unsigned int U32; +typedef unsigned long long U64; + +typedef unsigned long ULONG; + +typedef float F32; +typedef double F64; + +typedef unsigned long PTR; + +#define fn( ... ) [&]( __VA_ARGS__ ) +#define pfn( ... ) []( __VA_ARGS__ ) + +template <typename T> +struct __defer_t { + T f; + __defer_t( T f ) : f( f ) {}; + ~__defer_t() { f(); } +}; + +template <typename T> +__defer_t<T> __defer_func( T f ) { + return __defer_t<T>( f ); +} + +#define DEFER_1( x, y ) x##y +#define DEFER_2( x, y ) DEFER_1( x, y ) +#define DEFER_3( x ) DEFER_2( x, __COUNTER__ ) +#define defer( code ) auto DEFER_3( _defer_ ) = __defer_func( fn() { code; } ) + +#ifdef DEBUG +#define dlog( ... ) (void)printf( __VA_ARGS__ ) +#define ddef( x ) x +#else +#define dlog( ... ) {} +#define ddef( x ) +#endif diff --git a/src/util/vector.h b/src/util/vector.h new file mode 100644 index 0000000..33f6551 --- /dev/null +++ b/src/util/vector.h @@ -0,0 +1,192 @@ +#pragma once + +#include <math.h> +#include "typedef.h" + +static const F32 PI = 3.14159265359f; +static const F32 PIRAD = 0.01745329251f; +static const F32 RADPI = 57.2957795131f; + +struct VEC2 { + F32 x; + F32 y; + + VEC2() { x = y = 0.0f; } + VEC2( F32 X, F32 Y ) { x = X; y = Y; } + VEC2( const F32* v ) { x = v[0]; y = v[1]; } + VEC2( const VEC2& v ) { x = v.x; y = v.y; } + + bool operator==( const VEC2& v ) const { return ( x == v.x && y == v.y ); } + VEC2& operator=( const VEC2& v ) { x = v.x; y = v.y; return *this; } + F32& operator[]( I32 i ) { return ( (F32*)this )[i]; } + F32 operator[]( I32 i ) const { return ( (const F32*)this )[i]; } + + VEC2& operator+=( const VEC2& v ) { x += v.x; y += v.y; return *this; } + VEC2& operator-=( const VEC2& v ) { x -= v.x; y -= v.y; return *this; } + VEC2& operator*=( const VEC2& v ) { x *= v.x; y *= v.y; return *this; } + VEC2& operator/=( const VEC2& v ) { x /= v.x; y /= v.y; return *this; } + + VEC2 operator+( const VEC2& v ) const { return VEC2( x + v.x, y + v.y ); } + VEC2 operator-( const VEC2& v ) const { return VEC2( x - v.x, y - v.y ); } + VEC2 operator*( const VEC2& v ) const { return VEC2( x * v.x, y * v.y ); } + VEC2 operator/( const VEC2& v ) const { return VEC2( x / v.x, y / v.y ); } + + VEC2& operator+=( const F32& v ) { x += v; y += v; return *this; } + VEC2& operator-=( const F32& v ) { x -= v; y -= v; return *this; } + VEC2& operator*=( const F32& v ) { x *= v; y *= v; return *this; } + VEC2& operator/=( const F32& v ) { x /= v; y /= v; return *this; } + + VEC2 operator/( const F32& v ) const { return VEC2( x / v, y / v ); } + VEC2 operator*( const F32& v ) const { return VEC2( x * v, y * v ); } + VEC2 operator+( const F32& v ) const { return VEC2( x + v, y + v ); } + VEC2 operator-( const F32& v ) const { return VEC2( x - v, y - v ); } +}; + +struct VEC3 { + F32 x; + F32 y; + F32 z; + + VEC3() { x = y = z = 0.0f; } + VEC3( F32 X, F32 Y, F32 Z ) { x = X; y = Y; z = Z; } + VEC3( const F32* v ) { x = v[0]; y = v[1]; z = v[2]; } + VEC3( const VEC3& v ) { x = v.x; y = v.y; z = v.z; } + + VEC3( const VEC2& v ) { x = v.x; y = v.y; z = 0.f; } + + VEC3& operator=( const VEC3& v ) { x = v.x; y = v.y; z = v.z; return *this; } + F32& operator[]( I32 i ) { return ( (F32*)this )[i]; } + F32 operator[]( I32 i ) const { return ( (F32*)this )[i]; } + + VEC3& operator+=( const VEC3& v ) { x += v.x; y += v.y; z += v.z; return *this; } + VEC3& operator-=( const VEC3& v ) { x -= v.x; y -= v.y; z -= v.z; return *this; } + VEC3& operator*=( const VEC3& v ) { x *= v.x; y *= v.y; z *= v.z; return *this; } + VEC3& operator/=( const VEC3& v ) { x /= v.x; y /= v.y; z /= v.z; return *this; } + + VEC3 operator+( const VEC3& v ) const { return VEC3( x + v.x, y + v.y, z + v.z ); } + VEC3 operator-( const VEC3& v ) const { return VEC3( x - v.x, y - v.y, z - v.z ); } + VEC3 operator*( const VEC3& v ) const { return VEC3( x * v.x, y * v.y, z * v.z ); } + VEC3 operator/( const VEC3& v ) const { return VEC3( x / v.x, y / v.y, z / v.z ); } + + VEC3& operator+=( const F32& v ) { x += v; y += v; z += v; return *this; } + VEC3& operator-=( const F32& v ) { x -= v; y -= v; z -= v; return *this; } + VEC3& operator*=( const F32& v ) { x *= v; y *= v; z *= v; return *this; } + VEC3& operator/=( const F32& v ) { x /= v; y /= v; z /= v; return *this; } + + VEC3 operator/( const F32& v ) const { return VEC3( x / v, y / v, z / v ); } + VEC3 operator*( const F32& v ) const { return VEC3( x * v, y * v, z * v ); } + VEC3 operator+( const F32& v ) const { return VEC3( x + v, y + v, z + v ); } + VEC3 operator-( const F32& v ) const { return VEC3( x - v, y - v, z - v ); } +}; + +struct VEC4 { + F32 x, y, z, w; + + VEC4() { x = y = z = w = 0; } + VEC4( F32 X, F32 Y, F32 Z, F32 W ) { x = X; y = Y; z = Z; w = W; } + VEC4( const VEC4& v ) { x = v.x; y = v.y; z = v.z; w = v.w; } + VEC4( const VEC3& v ) { x = v.x; y = v.y; z = v.z; w = 1.f; } + VEC4( const VEC2& v ) { x = v.x; y = v.y; z = 0.f; w = 1.f; } + + VEC4( const VEC2& v, F32 W ) { x = v.x; y = v.y; z = 0.f; w = W; } + VEC4( const VEC3& v, F32 W ) { x = v.x; y = v.y; z = v.z; w = W; } + + VEC4( F32* v ) { x = v[0]; y = v[1]; z = v[2]; w = v[3]; } + + VEC4 operator=( const VEC4& v ) { x = v.x; y = v.y; z = v.z; w = v.w; return *this; } + F32& operator[]( const int i ) { return *(&x + i); } + F32 operator[]( const int i ) const { return *(&x + i); } + + VEC4& operator+=( const VEC4& v ) { x += v.x; y += v.y; z += v.z; w += v.w; return *this; } + VEC4& operator-=( const VEC4& v ) { x -= v.x; y -= v.y; z -= v.z; w -= v.w; return *this; } + VEC4& operator*=( const VEC4& v ) { x *= v.x; y *= v.y; z *= v.z; w *= v.w; return *this; } + VEC4& operator/=( const VEC4& v ) { x /= v.x; y /= v.y; z /= v.z; w /= v.w; return *this; } + + VEC4 operator+( const VEC4& v ) { return VEC4( x + v.x, y + v.y, z + v.z, w + v.w ); } + VEC4 operator-( const VEC4& v ) { return VEC4( x - v.x, y - v.y, z - v.z, w - v.w ); } + VEC4 operator*( const VEC4& v ) { return VEC4( x * v.x, y * v.y, z * v.z, w * v.w ); } + VEC4 operator/( const VEC4& v ) { return VEC4( x / v.x, y / v.y, z / v.z, w / v.w ); } + + VEC4& operator+=( const F32& v ) { x += v; y += v; z += v; w += v; return *this; } + VEC4& operator-=( const F32& v ) { x -= v; y -= v; z -= v; w -= v; return *this; } + VEC4& operator*=( const F32& v ) { x *= v; y *= v; z *= v; w *= v; return *this; } + VEC4& operator/=( const F32& v ) { x /= v; y /= v; z /= v; w /= v; return *this; } + + VEC4 operator+( const F32& v ) { return VEC4( x + v, y + v, z + v, w + v ); } + VEC4 operator-( const F32& v ) { return VEC4( x - v, y - v, z - v, w - v ); } + VEC4 operator*( const F32& v ) { return VEC4( x * v, y * v, z * v, w * v ); } + VEC4 operator/( const F32& v ) { return VEC4( x / v, y / v, z / v, w / v ); } +}; + +inline U8 is_zero( VEC2 v ) { return v.x == 0.f && v.y == 0.f; } +inline U8 is_zero( VEC3 v ) { return v.x == 0.f && v.y == 0.f && v.z == 0.f; } +inline U8 is_zero( VEC4 v ) { return v.x == 0.f && v.y == 0.f && v.z == 0.f && v.w == 0.f; } + +inline F32 vec_len( VEC2 v ) { return sqrtf( v.x * v.x + v.y * v.y ); } +inline F32 vec_lensq( VEC2 v ) { return v.x * v.x + v.y * v.y; } + +inline F32 vec_len( VEC3 v ) { return sqrtf( v.x * v.x + v.y * v.y + v.z * v.z ); } +inline F32 vec_lensq( VEC3 v ) { return v.x * v.x + v.y * v.y + v.z * v.z; } +inline F32 vec_len2d( VEC3 v ) { return sqrtf( v.x * v.x + v.y * v.y ); } +inline F32 vec_len2dsq( VEC3 v ) { return v.x * v.x + v.y * v.y; } + +inline F32 vec_len( VEC4 v ) { return sqrtf( v.x * v.x + v.y * v.y + v.z * v.z + v.w * v.w ); } +inline F32 vec_lensq( VEC4 v ) { return v.x * v.x + v.y * v.y + v.z * v.z + v.w * v.w; } +inline F32 vec_len3d( VEC4 v ) { return sqrtf( v.x * v.x + v.y * v.y + v.z * v.z ); } +inline F32 vec_len3dsq( VEC4 v ) { return v.x * v.x + v.y * v.y + v.z * v.z; } + +inline F32 vec_distsq( VEC2 v1, VEC2 v2 ) { return vec_lensq( v1 - v2 ); } +inline F32 vec_dist( VEC2 v1, VEC2 v2 ) { return sqrtf( vec_distsq( v1, v2 ) ); } +inline F32 vec_dot( VEC2 v1, VEC2 v2 ) { return v1.x * v2.x + v1.y * v2.y; } +inline F32 vec_cross( VEC2 v1, VEC2 v2 ) { return v1.x * v2.y - v1.y * v2.x; } + +inline F32 vec_distsq( VEC3 v1, VEC3 v2 ) { return vec_lensq( v1 - v2 ); } +inline F32 vec_dist( VEC3 v1, VEC3 v2 ) { return sqrtf( vec_distsq( v1, v2 ) ); } +inline F32 vec_dot( VEC3 v1, VEC3 v2 ) { return v1.x * v2.x + v1.y * v2.y + v1.z * v2.z; } +inline VEC3 vec_cross( VEC3 v1, VEC3 v2 ) { return { v1.y * v2.z - v1.z * v2.y, v1.z * v2.x - v1.x * v2.z, v1.x * v2.y - v1.y * v2.x }; } + +inline void vec_normalize( VEC2* v ) { F32 l = vec_len( *v ); v->x /= l; v->y /= l; } +inline void vec_normalize( VEC3* v ) { F32 l = vec_len( *v ); v->x /= l; v->y /= l; v->z /= l; } + +inline VEC2 vec_normalize( VEC2 v ) { vec_normalize( &v ); return v; }; +inline VEC3 vec_normalize( VEC3 v ) { vec_normalize( &v ); return v; }; + +inline VEC3 vec_normangle( VEC3* v ) { + v->x = remainderf( v->x, 360.f ); + v->y = remainderf( v->y, 360.f ); + v->z = remainderf( v->z, 360.f ); + + v->x = fmaxf( fminf( v->x, 90.f ), -90.f ); + return *v; +} + +inline VEC3 vec_normangle( VEC3 v ) { + return vec_normangle( &v ); +} + +inline VEC2 vec_normangle( VEC2* v ) { + v->x = remainderf( v->x, 360.f ); + v->y = remainderf( v->y, 360.f ); + + v->x = fmaxf( fminf( v->x, 90.f ), -90.f ); + return *v; +} + +inline VEC2 vec_normangle( VEC2 v ) { + return vec_normangle( &v ); +} + +inline VEC3 vec_angles( VEC3 v1, VEC3 v2 ) { + VEC3 d = v2 - v1; + + F32 m = sqrtf( d.x * d.x + d.y * d.y ); + F32 p = atan2f( -d.z, m ) * RADPI; + F32 y = atan2f( d.y, d.x ) * RADPI; + + return vec_normangle( VEC3( p, y, 0.0f ) ); +} + +inline F32 vec_angle( VEC2 v1, VEC2 v2 ) { + VEC2 d = v2 - v1; + return remainderf( atan2f( d.y, d.x ) * RADPI, 360.f ); +} |