merge changes

.clang-format

@@ -0,0 +1,14 @@
+# We'll use defaults from the LLVM style, but with 4 columns indentation.
+BasedOnStyle: LLVM
+IndentWidth: 4
+
+
+Language: Cpp
+
+# Force pointers to the type for C++.
+DerivePointerAlignment: false
+PointerAlignment: Left
+
+LambdaBodyIndentation: OuterScope
+
+AlignConsecutiveAssignments: true

.gitignore

@@ -3,3 +3,4 @@
 compile_commands.json
 *.autosave
 build
+.cache

.vscode/launch.json

@@ -1,7 +0,0 @@
-{
-    // Use IntelliSense to learn about possible attributes.
-    // Hover to view descriptions of existing attributes.
-    // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
-    "version": "0.2.0",
-    "configurations": []
-}

.vscode/settings.json

@@ -1,22 +0,0 @@
-{
-    "C_Cpp.default.configurationProvider": "ms-vscode.cmake-tools",
-    "cmake.configureOnOpen": true,
-    "files.associations": {
-        "array": "cpp",
-        "cwchar": "cpp",
-        "unordered_map": "cpp",
-        "vector": "cpp",
-        "exception": "cpp",
-        "functional": "cpp",
-        "initializer_list": "cpp",
-        "iosfwd": "cpp",
-        "istream": "cpp",
-        "new": "cpp",
-        "ostream": "cpp",
-        "stdexcept": "cpp",
-        "streambuf": "cpp",
-        "type_traits": "cpp",
-        "tuple": "cpp"
-    },
-    "mesonbuild.configureOnOpen": false
-}

.vscode/tasks.json

@@ -1,28 +0,0 @@
-{
-    "tasks": [
-        {
-            "type": "cppbuild",
-            "label": "C/C++: g++ build active file",
-            "command": "/usr/bin/g++",
-            "args": [
-                "-fdiagnostics-color=always",
-                "-g",
-                "${file}",
-                "-o",
-                "${fileDirname}/${fileBasenameNoExtension}"
-            ],
-            "options": {
-                "cwd": "${fileDirname}"
-            },
-            "problemMatcher": [
-                "$gcc"
-            ],
-            "group": {
-                "kind": "build",
-                "isDefault": true
-            },
-            "detail": "Task generated by Debugger."
-        }
-    ],
-    "version": "2.0.0"
-}

CMakeLists.txt

@@ -1,17 +1,19 @@
 #
 # CMake build system for pixwerx
 #
-cmake_minimum_required(VERSION 3.8)
+cmake_minimum_required(VERSION 3.28)
 
 
 project(pixwerx)
 
 #
-# pixwerx ist C++20
+# pixwerx is C++23
 #
-set (CMAKE_CXX_STANDARD 20)
+set (CMAKE_CXX_STANDARD 23)
 
+#
 # internal cmake modules
+#
 set(CMAKE_MODULE_PATH ${CMAKE_CURRENT_SOURCE_DIR}/etc/cmake ${CMAKE_MODULE_PATH})
 
 # boilerplate locations for build
@@ -28,3 +30,4 @@ set(CMAKE_POSITION_INDEPENDENT_CODE ON)
 #
 add_subdirectory(src)
 add_subdirectory(share)
+add_subdirectory(tests)

LICENSE

@@ -1,6 +1,6 @@
 MIT License
 
-Copyright (c) 1999-2021 Hartmut Seichter
+Copyright (c) 1999-2024 Hartmut Seichter
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

README.md

@@ -1,34 +1,19 @@
-# pixwerx 
+# pixwerx
 
-pixwerx is an opinionated, academic approach to a 3D engine. It tries to mix 
-modern andproven methods to build a fast and portable engine beyond pure 3D. 
+pixwerx is a general purpose 3D engine written in C++23 and scriptable with Lua.
 
-## Design Principles
-
-Dogdooding: pixwerx is built on the principle of dogfooding. The engine is a 
-full-stack system that tries to implement all necessary functions. The engine 
-editor is just a UI build with the engine.
-
-Reasonable dependencies: like many engines pixwerx tries to include as much 3rd-
-party code as possible and implements some of the core systems itself.
-
-No premature optimization: pixwerx implements only very few systems with machine
-code. It tries to utilize the power of the compiler as much as possible.
-
-Computer graphics 101: pixwerx does implement graphics components and systems in 
-a  way that makes code portable and not over-parameterized. Usability is 
-achieved through layers on top of the core components. 
+## Principles
 
+-   datadriven design
+-   multiple platform capable through heavy usage of STL
+-   minimal set of dependencies
+-   efficiency through meta-meta programming techniques
+-   no CPU intrinsics
 
 ## License
 
-pixwerx is licenced under the terms of the MIT License. Please consult the 
-LICENSE file.
+pixwerx is licenced under the terms of the MIT License. Please consult the LICENSE file.
 
 ## Authors
 
-© 1999-2020 Hartmut Seichter
-
-
-
-
+© 1999-2024 Hartmut Seichter

share/CMakeLists.txt

@@ -1,4 +1,4 @@
-find_package(mkdocs 1.0)
+# find_package(mkdocs 1.0)
 
 # if(MKDOCS_FOUND)
 #	configure_file(mkdocs.yml "${CMAKE_CURRENT_BINARY_DIR}/mkdocs.yml" COPYONLY)

src/CMakeLists.txt

@@ -3,13 +3,13 @@ add_subdirectory(deps)
 
 # build internal core
 add_subdirectory(core)
-add_subdirectory(scene)
-add_subdirectory(system)
+# add_subdirectory(scene)
+# add_subdirectory(system)
 add_subdirectory(io)
 
 #add_subdirectory(ui)
-add_subdirectory(binding)
-add_subdirectory(visual)
-add_subdirectory(geometry)
+# add_subdirectory(binding)
+# add_subdirectory(visual)
+# add_subdirectory(geometry)
 
-add_subdirectory(runtime)
+# add_subdirectory(runtime)

src/binding/include/pw/scripting/script.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal

src/binding/src/script_core.cpp

@@ -1,39 +1,34 @@
-#include "pw/core/vector.hpp"
-#include "pw/core/quaternion.hpp"
 #include "pw/core/axisangle.hpp"
+#include "pw/core/color.hpp"
 #include "pw/core/debug.hpp"
-#include "pw/core/size.hpp"
-#include "pw/core/point.hpp"
-#include "pw/core/time.hpp"
-#include "pw/core/geometry.hpp"
 #include "pw/core/image.hpp"
 #include "pw/core/matrix_transform.hpp"
+#include "pw/core/point.hpp"
+#include "pw/core/primitives.hpp"
+#include "pw/core/quaternion.hpp"
 #include "pw/core/rectangle.hpp"
-#include "pw/core/color.hpp"
+#include "pw/core/size.hpp"
+#include "pw/core/time.hpp"
+#include "pw/core/vector.hpp"
 
 #include "runtime_lua.hpp"
 
-
 // seems CRTP magic doesnt work with SOL
 namespace sol {
-template <> struct is_automagical<pw::matrix4x4> : std::false_type {};
-template <> struct is_automagical<pw::vector3> : std::false_type {};
-template <> struct is_automagical<pw::vector2> : std::false_type {};
-template <> struct is_automagical<pw::vector4> : std::false_type {};
-template <> struct is_automagical<pw::quaternion> : std::false_type {};
-template <> struct is_automagical<pw::rectangle> : std::false_type {};
+// template <> struct is_automagical<pw::matrix4x4> : std::false_type {};
+// template <> struct is_automagical<pw::vector3> : std::false_type {};
+// template <> struct is_automagical<pw::vector2> : std::false_type {};
+// template <> struct is_automagical<pw::vector4> : std::false_type {};
+// template <> struct is_automagical<pw::quaternion> : std::false_type {};
+// template <> struct is_automagical<pw::rectangle> : std::false_type {};
 }
 
 namespace pw {
 
-vector3 table_to_vector3(sol::table t)
-{
-    return vector3(t[0],t[1],t[2]);
-}
+// vector3 table_to_vector3(sol::table t) { return vector3(t[0], t[1], t[2]); }
 
 template <typename elementType>
-std::vector<elementType> convert_sequence(sol::state& lua,sol::table t)
-{
+std::vector<elementType> convert_sequence(sol::state& lua, sol::table t) {
     const std::size_t sz = t.size();
     std::vector<elementType> res(sz);
     for (std::size_t i = 1; i <= sz; i++) {
@@ -42,45 +37,57 @@ std::vector<elementType> convert_sequence(sol::state& lua,sol::table t)
     return res;
 }
 
+void register_core_function(sol::state& lua, sol::table& ns) {
 
+    ns.set("pi", pw::pi<float>());
 
-void register_core_function(sol::state& lua,sol::table& ns)
-{
+    ns.new_usertype<color>(
+        "color", sol::call_constructor, sol::constructors<color()>(),  //
+        "rgba", &color::rgba,                                          //
+        "data", sol::property([](color& c) { return c.rgba.data(); }), //
+        "table",
+        sol::property([](const color& c) { //
+        return sol::as_table(std::array<vector4f::value_type, 4>{
+            c.rgba.x(), c.rgba.y(), c.rgba.z(), c.rgba.w()});
+    }, [](color& c, const sol::table& t) {
+        c = color{
+            .rgba = vector4f{t[0], t[1], t[2], t[3]}
+        };
+    }));
 
-    ns.set("pi",pw::pi<real_t>());
+#if 0
 
+    ns.new_usertype<matrix4x4>(
+        "matrix4x4", sol::call_constructor, sol::constructors<matrix4x4()>(),
+        "row", &matrix4x4::row, "column", &matrix4x4::column, "inverse",
+        sol::readonly_property(&matrix4x4::inverse), "identity",
+        sol::readonly_property(&matrix4x4::identity),
+        sol::meta_function::multiplication,
+        [](const matrix4x4& a, const vector4& b) { return vector4(a * b); });
 
+    ns.new_usertype<vector2f>(
+        "vector2", sol::call_constructor,
+        sol::constructors<vector2(), vector2(real_t, real_t)>(),
+        "x",
+            sol::property(
+            sol::resolve<const vector2::value_type&() const>(&vector2::x),
+            [](vector2& v, vector2::value_type val) { v.x() = val; }),
+        "y",
+        sol::property(
+            sol::resolve<const vector2::value_type&() const>(&vector2::y),
+            [](vector2& v, vector2::value_type val) { v.y() = val; }),
+        "dot", &vector2::dot, sol::meta_function::addition,
+        [](const vector2& a, const vector2& b) { return vector2(a + b); },
+        sol::meta_function::subtraction,
+        [](const vector2& a, const vector2& b) { return vector2(a - b); },
+        "data", sol::property([](vector2& s) { return std::ref(s.data); }),
+        "table", sol::property([](const vector2& s) {
+        return sol::as_table(std::array<vector2::value_type, 2>{s.x(), s.y()});
+    }, [](vector2& s, const sol::table& t) {
+        s.data[0] = t[1];
+        s.data[1] = t[2];
+    }));
 
-    ns.new_usertype<color>("color",
-                           sol::call_constructor,sol::constructors<color(),color(real_t,real_t,real_t,real_t)>(),
-                           "rgba",&color::rgba,
-                           "data",sol::property([](color& c) { return std::ref(c.rgba.data);} ),
-                           "table",sol::property([](const color& c){ return sol::as_table(std::array<vector4::value_type,4>{c.rgba.x(),c.rgba.y(),c.rgba.z(),c.rgba.w()}); },
-                                                 [](color& c,const sol::table& t) { c = color((real_t)t[0],t[1],t[2],t[3]);})
-
-                            );
-
-
-    ns.new_usertype<matrix4x4>("matrix4x4",
-                               sol::call_constructor,sol::constructors<matrix4x4()>(),
-                               "row",&matrix4x4::row,
-                               "column",&matrix4x4::column,
-                               "inverse", sol::readonly_property(&matrix4x4::inverse),
-                               "identity",sol::readonly_property(&matrix4x4::identity),
-                               sol::meta_function::multiplication,[](const matrix4x4& a,const vector4& b) { return vector4(a * b); }
-    );
-
-    ns.new_usertype<vector2>("vector2",
-                             sol::call_constructor,sol::constructors<vector2(),vector2(real_t,real_t)>(),
-                             "x", sol::property(sol::resolve<const vector2::value_type&() const>(&vector2::x), [](vector2& v,vector2::value_type val){ v.x() = val;}),
-                             "y", sol::property(sol::resolve<const vector2::value_type&() const>(&vector2::y), [](vector2& v,vector2::value_type val){ v.y() = val;}),
-                             "dot",&vector2::dot,
-                             sol::meta_function::addition,[](const vector2& a,const vector2& b){ return vector2(a + b); },
-                             sol::meta_function::subtraction,[](const vector2& a,const vector2& b){ return vector2(a - b); },
-                             "data",sol::property([](vector2& s) { return std::ref(s.data);} ),
-                             "table",sol::property([](const vector2& s){ return sol::as_table(std::array<vector2::value_type,2>{s.x(),s.y()}); },
-                                                   [](vector2& s,const sol::table& t) { s.data[0] = t[1]; s.data[1] = t[2];} )
-                             );
 
     ns.new_usertype<vector3>("vector3"
                             ,sol::call_constructor,sol::constructors<vector3(),vector3(real_t,real_t,real_t)>()
@@ -258,9 +265,9 @@ void register_core_function(sol::state& lua,sol::table& ns)
     ns.create_named("mathf")
             .set_function("ping_pong",ping_pong<real_t>);
 
+#endif
 }
 
-
 PW_REGISTER_LUA(core)
 
-}
+} // namespace pw

src/binding/src/script_system.cpp

@@ -1,9 +1,9 @@
-//#include "script_system.hpp"
+// #include "script_system.hpp"
 
-#include "pw/system/window.hpp"
-#include "pw/system/input.hpp"
 #include "pw/system/display.hpp"
+#include "pw/system/input.hpp"
 #include "pw/system/path.hpp"
+#include "pw/system/window.hpp"
 
 #include "runtime_lua.hpp"
 
@@ -13,49 +13,37 @@
 
 namespace pw {
 
-void register_system_function(sol::state&, sol::table &ns)
-{
-    ns.new_usertype<window>("window",
-                            "update",&window::update,
-                            "on_update",sol::writeonly_property(&window::set_on_update),
-                            "on_resize",sol::writeonly_property(&window::set_on_resize),
-                            "title",sol::writeonly_property(&window::set_title),
-                            "size",sol::property(&window::size,&window::set_size),
-                            "client_size",sol::readonly_property(&window::client_size),
-                            "position",sol::property(&window::position,&window::set_position),
-                            "fullscreen",sol::property(&window::fullscreen,&window::set_fullscreen),
-                            "visible",sol::property(&window::visible,&window::set_visible)
-                            );
+void register_system_function(sol::state&, sol::table& ns) {
+    ns.new_usertype<window>(
+        "window", "update", &window::update, "on_update",
+        sol::writeonly_property(&window::set_on_update), "on_resize",
+        sol::writeonly_property(&window::set_on_resize), "title",
+        sol::writeonly_property(&window::set_title), "size",
+        sol::property(&window::size, &window::set_size), "client_size",
+        sol::readonly_property(&window::client_size), "position",
+        sol::property(&window::position, &window::set_position), "fullscreen",
+        sol::property(&window::fullscreen, &window::set_fullscreen), "visible",
+        sol::property(&window::visible, &window::set_visible));
 
+    ns.new_usertype<input>(
+        "input", "new", sol::no_constructor, "get", &input::get,
+        "mouse_position", sol::readonly_property(&input::mouse_position),
+        "mouse_button", sol::readonly_property(&input::mouse_button),
+        "mouse_pressed", sol::readonly_property(&input::mouse_pressed),
+        "has_input", sol::readonly_property(&input::has_input), "input_string",
+        sol::readonly_property(&input::input_string));
 
-    ns.new_usertype<input>("input",
-                           "new", sol::no_constructor,
-                           "get",&input::get,
-                           "mouse_position",sol::readonly_property(&input::mouse_position),
-                           "mouse_button",sol::readonly_property(&input::mouse_button),
-                           "mouse_pressed",sol::readonly_property(&input::mouse_pressed),
-                           "has_input",sol::readonly_property(&input::has_input),
-                           "input_string",sol::readonly_property(&input::input_string)
-                           );
+    ns.new_usertype<display>("display", "all", &display::all, "name",
+                             sol::readonly_property(&display::name));
 
-    ns.new_usertype<display>("display",
-                             "all",&display::all,
-                             "name",sol::readonly_property(&display::name)
-                             );
-
-    ns.new_usertype<path>("path"
-                          ,"new",sol::no_constructor
-                          ,"get",&path::get
-                          ,"separator",sol::readonly_property(&path::separator)
-                          ,"executable_path",sol::readonly_property(&path::executable_path)
-                          ,"resource_paths",sol::readonly_property([](const path& p){return sol::as_table(p.resource_paths());})
-                          );
+    ns.new_usertype<path>(
+        "path", "new", sol::no_constructor, "get", &path::get, "separator",
+        sol::readonly_property(&path::separator), "executable_path",
+        sol::readonly_property(&path::executable_path), "resource_paths",
+        sol::readonly_property(
+            [](const path& p) { return sol::as_table(p.resource_paths()); }));
 }
 
-
-
-
 PW_REGISTER_LUA(system)
 
-
-}
+} // namespace pw

src/binding/src/script_visual.cpp

@@ -1,92 +1,76 @@
 
 #include "pw/core/debug.hpp"
+#include "pw/visual/context.hpp"
+#include "pw/visual/framebuffer.hpp"
 #include "pw/visual/pipeline.hpp"
 #include "pw/visual/shader.hpp"
-#include "pw/visual/framebuffer.hpp"
 #include "pw/visual/texture.hpp"
-#include "pw/visual/context.hpp"
 
 #include "pw/core/size.hpp"
 
-
 #include "runtime_lua.hpp"
 
 namespace pw {
 
-void register_visual_function(sol::state& lua,sol::table &ns)
-{
+void register_visual_function(sol::state& lua, sol::table& ns) {
 
-//    ns.new_usertype<pipeline>("pipeline"
-//							  ,"create",&pipeline::create
-//                              ,"draw",&pipeline::draw
-//                              );
+    //    ns.new_usertype<pipeline>("pipeline"
+    //							  ,"create",&pipeline::create
+    //                              ,"draw",&pipeline::draw
+    //                              );
 
-    ns.new_usertype<shader>("shader"
-                            ,sol::call_constructor,sol::constructors<shader()>()
-                            ,"ready",sol::readonly_property(&shader::ready)
-                            ,"use",&shader::use
-                            ,"build",&shader::build
-                            ,"source",&shader::source
-                            ,"set_source",&shader::set_source
-                            ,"set_uniforms",&shader::set_uniforms
-                            ,"set_uniform_float",&shader::set_uniform<float>
-                            ,"set_uniform_uint",&shader::set_uniform<uint32_t>
-                            ,"set_uniform_int",&shader::set_uniform<int32_t>
-                            ,"set_uniform_mat4",&shader::set_uniform<matrix4x4&>
-                            ,"set_uniform_vec4",&shader::set_uniform<vector4&>
-                            ,"set_uniform_texture",&shader::set_uniform<texture&>
+    ns.new_usertype<shader>(
+        "shader", sol::call_constructor, sol::constructors<shader()>(), "ready",
+        sol::readonly_property(&shader::ready), "use", &shader::use, "build",
+        &shader::build, "source", &shader::source, "set_source",
+        &shader::set_source, "set_uniforms", &shader::set_uniforms,
+        "set_uniform_float", &shader::set_uniform<float>, "set_uniform_uint",
+        &shader::set_uniform<uint32_t>, "set_uniform_int",
+        &shader::set_uniform<int32_t>, "set_uniform_mat4",
+        &shader::set_uniform<matrix4x4&>, "set_uniform_vec4",
+        &shader::set_uniform<vector4&>, "set_uniform_texture",
+        &shader::set_uniform<texture&>
 
-                            );
+    );
 
+    ns["shader_type"] = ns.create_named(
+        "shader_type", "fragment", shader::code_type::fragment, "vertex",
+        shader::code_type::vertex, "geometry", shader::code_type::geometry,
+        "compute", shader::code_type::compute);
 
-    ns["shader_type"] = ns.create_named("shader_type"
-                                       ,"fragment",shader::code_type::fragment
-                                        ,"vertex",shader::code_type::vertex
-                                        ,"geometry",shader::code_type::geometry
-                                        ,"compute",shader::code_type::compute);
+    //    new_enum<false>(
 
-//    new_enum<false>(
+    //    ns.new_usertype<render_pass>("render_pass"
+    //                                ,"submit",&render_pass::submit
+    //                                 );
 
+    ns.new_usertype<renderer>(
+        "renderer", sol::call_constructor,
+        sol::constructors<renderer(), renderer(const geometry&)>(), "update",
+        &renderer::update, "ready", sol::readonly_property(&renderer::ready),
+        "change_count", sol::readonly_property(&renderer::change_count),
+        "release", &renderer::release, "draw", &renderer::draw);
 
-//    ns.new_usertype<render_pass>("render_pass"
-//                                ,"submit",&render_pass::submit
-//                                 );
+    ns.new_usertype<framebuffer>(
+        "framebuffer", sol::call_constructor,
+        sol::constructors<framebuffer()>(), "create", &framebuffer::create,
+        "bind", &framebuffer::bind, "unbind", &framebuffer::unbind, "blit",
+        &framebuffer::blit);
 
+    ns.new_usertype<texture>("texture", sol::call_constructor,
+                             sol::constructors<texture()>(), "create",
+                             &texture::create, "update", &texture::update,
+                             "bind", &texture::bind, "unbind", &texture::unbind,
+                             "native_handle", &texture::native_handle);
 
-    ns.new_usertype<renderer>("renderer"
-                              ,sol::call_constructor,sol::constructors<renderer(),renderer(const geometry&)>()
-                              ,"update",&renderer::update
-                              ,"ready",sol::readonly_property(&renderer::ready)
-                              ,"change_count",sol::readonly_property(&renderer::change_count)
-                              ,"release",&renderer::release
-                              ,"draw",&renderer::draw
-                              );
-
-    ns.new_usertype<framebuffer>("framebuffer"
-                                 ,sol::call_constructor,sol::constructors<framebuffer()>()
-                                 ,"create",&framebuffer::create
-                                 ,"bind",&framebuffer::bind
-                                 ,"unbind",&framebuffer::unbind
-                                 ,"blit",&framebuffer::blit);
-
-    ns.new_usertype<texture>("texture"
-                             ,sol::call_constructor,sol::constructors<texture()>()
-                             ,"create",&texture::create
-                             ,"update",&texture::update
-                             ,"bind",&texture::bind
-                             ,"unbind",&texture::unbind
-                             ,"native_handle",&texture::native_handle
-                             );
-
-    ns.new_usertype<context>("context"
-                             ,sol::call_constructor,sol::constructors<context()>()
-                             ,"clear",&context::clear
-                             ,"clearcolor",sol::property(&context::clearcolor,&context::set_clearcolor)
-                             ,"set_viewport",&context::set_viewport
-                             ,"get_error",&context::get_error
-                             );
+    ns.new_usertype<context>(
+        "context", sol::call_constructor, sol::constructors<context()>(),
+        "clear", &context::clear, "clearcolor",
+        sol::property(&context::clearcolor, &context::set_clearcolor),
+        "set_viewport", &context::set_viewport, "get_error",
+        &context::get_error);
 }
 
 PW_REGISTER_LUA(visual)
 
-}
+} // namespace pw

src/core/CMakeLists.txt

@@ -8,7 +8,6 @@ set(hdrs
     include/pw/core/globals.hpp
     include/pw/core/material.hpp
     include/pw/core/math.hpp
-    include/pw/core/matrixbase.hpp
     include/pw/core/matrix.hpp
     include/pw/core/quaternion.hpp
     include/pw/core/image.hpp
@@ -18,7 +17,7 @@ set(hdrs
     include/pw/core/serialize.hpp
     include/pw/core/size.hpp
     include/pw/core/time.hpp
-    include/pw/core/geometry.hpp
+    include/pw/core/primitives.hpp
     include/pw/core/image.hpp
     include/pw/core/vector.hpp
     include/pw/core/matrix_transform.hpp
@@ -35,12 +34,12 @@ set(srcs
     src/core.cpp
     src/image.cpp
     src/debug.cpp
-    src/geometry.cpp
-    src/material.cpp
-    src/resource.cpp
-    src/serialize.cpp
-    src/time.cpp
-    src/image.cpp
+    # src/geometry.cpp
+    # src/material.cpp
+    # src/resource.cpp
+    # src/serialize.cpp
+    # src/time.cpp
+    # src/image.cpp
     )
 
 add_library(pwcore
@@ -59,4 +58,3 @@ target_include_directories(
 target_link_libraries(pwcore)
 
 add_subdirectory(tests)
-

src/core/include/pw/core/aabb.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -23,30 +23,47 @@
 #ifndef PW_CORE_AABB_HPP
 #define PW_CORE_AABB_HPP
 
+#include "primitives.hpp"
 #include <pw/core/vector.hpp>
 
+#include <numeric>
+#include <vector>
+
 namespace pw {
 
-struct aabb {
+template <typename Scalar, std::size_t N> struct aabb final {
 
-	vector3 min;
-	vector3 max;
+    using value_type = vector<Scalar, N>;
 
-	aabb(const vector3 min_vec,const vector3 max_vec)
-		: min(min_vec)
-		, max(max_vec)
-	{}
+    value_type min{};
+    value_type max{};
 
-	aabb() {
-		min.zero(); max.zero();
-	}
+    constexpr auto dimension() const noexcept { return max - min; }
 
-	vector3 dimension() const {
-		return max - min;
-	}
+    static constexpr auto
+    make(const std::vector<pw::vector<Scalar, N>>& vertices)
+        -> aabb<Scalar, N> {
+        return std::accumulate(
+            std::begin(vertices), std::end(vertices),
+            aabb{.min{vertices.front()}, .max{vertices.front()}},
+            [](const auto& prev, const auto& e) {
+            return aabb{.min{e.min(prev.min)}, .max{e.max(prev.max)}};
+        });
+    }
 
+    static constexpr auto make_from_indexed_vertices(
+        const std::vector<pw::vector<Scalar, N>>& vertices,
+        std::vector<size_t>& indices) -> aabb<Scalar, N> {
+        return std::accumulate(std::begin(indices), std::end(indices),
+                               aabb{.min{vertices[indices.front()]},
+                                    .max{vertices[indices.front()]}},
+                               [&vertices](const auto& prev, const auto& e) {
+            return aabb{.min{vertices[e].min(prev.min)},
+                        .max{vertices[e].max(prev.max)}};
+        });
+    }
 };
 
-}
+} // namespace pw
 
 #endif

src/core/include/pw/core/axisangle.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -23,95 +23,64 @@
 #ifndef PW_CORE_AXISANGLE_HPP
 #define PW_CORE_AXISANGLE_HPP
 
+#include <pw/core/matrix.hpp>
 #include <pw/core/vector.hpp>
 
 namespace pw {
 
-template <typename T>
-struct axisangle_ {
+template <std::floating_point Scalar> struct axisangle final {
 
-    using value_type = T;
-    using axis_type = vector3_<T>;
+    using value_type = Scalar;
+    using axis_type  = vector3<Scalar>;
 
-    axis_type axis;
-    T angle;
+    axis_type axis = axis_type::basis(2);
+    Scalar angle{};
 
-    axisangle_()
-        : axis(vector3_<T>::up()),
-          angle(0)
-    {}
+    constexpr static auto
+    from_matrix(const matrix<Scalar, 3, 3>& m) noexcept -> axisangle {
 
-    axisangle_(vector3_<T> axis,T angle)
-        : axis(std::move(axis))
-        , angle(std::move(angle))
-    {
+        return {.axis = axis_type{m[2][1] - m[1][2], // x
+                                  m[0][2] - m[2][0], // y
+                                  m[1][0] - m[0][1]} // z
+                            .normalized(),
+                .angle =
+                    std::acos((m[0][0] + m[1][1] + m[2][2] - 1) / Scalar{2})};
     }
 
-	static const axisangle_ from_matrix(const matrix_<4,4,T>& m)
-	{
-		using std::acos;
-		using std::sqrt;
+    constexpr auto to_matrix() const noexcept -> matrix<Scalar, 3, 3> {
 
-		axisangle_ aa_res;
-		aa_res.angle = acos((m(0,0) + m(1,1) + m(2,2) - 1) / T(2));
-		const T m2112 = m(2,1) - m(1,2);
-		const T m0220 = m(0,2) - m(2,0);
-		const T m1001 = m(1,0) - m(0,1);
-		// no singularity check here ...
-		const T mrot_denom = sqrt( m2112 * m2112 + m0220 * m0220 + m1001 * m1001 );
+        const auto axis_n = axis.normalized(); // always normalize
 
-		aa_res.axis.x() = m2112 / mrot_denom;
-		aa_res.axis.y() = m0220 / mrot_denom;
-		aa_res.axis.z() = m1001 / mrot_denom;
+        const auto cos_a   = std::cos(angle);
+        const auto sin_a   = std::sin(angle);
+        const auto cos_1_a = 1 - cos_a;
 
-		return aa_res;
-	}
+        const auto v1_a = axis_n.x() * axis_n.y() * cos_1_a;
+        const auto v2_a = axis_n.z() * sin_a;
 
-    matrix_<4,4,T> to_matrix() const
-    {
-        using std::cos;
-        using std::sin;
+        const auto v1_b = axis_n.x() * axis_n.z() * cos_1_a;
+        const auto v2_b = axis_n.y() * sin_a;
 
-        // result
-        matrix_<4,4,T> rot_mat; rot_mat.set_identity();
-		axis_type axis_n = axis.normalized(); // always normalize
+        const auto v1_c = axis_n.y() * axis_n.z() * cos_1_a;
+        const auto v2_c = axis_n.x() * sin_a;
 
-		const T cos_a = cos(angle);
-		const T sin_a = sin(angle);
-		const T cos_1_a = T(1) - cos_a;
-
-		rot_mat(0,0) = cos_a + axis_n.x() * axis_n.x() * cos_1_a;
-		rot_mat(1,1) = cos_a + axis_n.y() * axis_n.y() * cos_1_a;
-		rot_mat(2,2) = cos_a + axis_n.z() * axis_n.z() * cos_1_a;
-
-		T v1 = axis_n.x() * axis_n.y() * cos_1_a;
-		T v2 = axis_n.z() * sin_a;
-		rot_mat(1,0) = v1 + v2;
-		rot_mat(0,1) = v1 - v2;
-
-		v1 = axis_n.x() * axis_n.z() * cos_1_a;
-		v2 = axis_n.y() * sin_a;
-		rot_mat(2,0) = v1 - v2;
-		rot_mat(0,2) = v1 + v2;
-
-		v1 = axis_n.y() * axis_n.z() * cos_1_a;
-		v2 = axis_n.x() * sin_a;
-		rot_mat(2,1) = v1 + v2;
-		rot_mat(1,2) = v1 - v2;
-
-        return rot_mat;
+        return {
+            pw::vector{cos_a + axis_n.x() * axis_n.x() * cos_1_a, // [0,0]
+                       v1_a - v2_a,                               // [0,1]
+                       v1_b + v2_b},                              // [0,2]
+            pw::vector{v1_a + v2_a,                               // [1,0]
+                       cos_a + axis_n.y() * axis_n.y() * cos_1_a, // [1,1]
+                       v1_c - v2_c},                              // [1,2]
+            pw::vector{v1_b - v2_b,                               // [2,0]
+                       v1_c + v2_c,                               // [2,1]
+                       cos_a + axis_n.z() * axis_n.z() * cos_1_a} // [2,2]
+        };
     }
 };
 
+using axisanglef = axisangle<float>;
+using axisangled = axisangle<double>;
 
-using axisangle = axisangle_<real_t> ;
-using axisanglef = axisangle_<float> ;
-using axisangled = axisangle_<double> ;
-
-}
-
-
-
-
+} // namespace pw
 
 #endif

src/core/include/pw/core/buffer.hpp

@@ -1,11 +0,0 @@
-#ifndef PW_CORE_BUFFER_HPP
-#define PW_CORE_BUFFER_HPP
-
-#include <pw/core/globals.hpp>
-#include <vector>
-
-
-
-
-
-#endif // PW_CORE_BUFFER_HPP

src/core/include/pw/core/color.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -23,40 +23,34 @@
 #ifndef PW_CORE_COLOR_HPP
 #define PW_CORE_COLOR_HPP
 
+#include <pw/core/globals.hpp>
 #include <pw/core/vector.hpp>
-#include <limits>
+
 
 namespace pw {
 
-struct color {
+struct color final {
 
-    vector4 rgba {0, 0, 0, 1};
+    vector4f rgba{vector4f::all(1)};
 
-    color() = default;
+    static constexpr auto from_rgb8888(uint8_t r8, uint8_t g8, uint8_t b8,
+                                       uint8_t a8) -> color {
+        return {static_cast<float>(r8) / std::numeric_limits<uint8_t>::max(),
+                static_cast<float>(g8) / std::numeric_limits<uint8_t>::max(),
+                static_cast<float>(b8) / std::numeric_limits<uint8_t>::max(),
+                static_cast<float>(a8) / std::numeric_limits<uint8_t>::max()};
+    }
 
-	color(uint8_t r8,uint8_t g8,uint8_t b8,uint8_t a8)
-		: color(static_cast<real_t>(r8 / std::numeric_limits<uint8_t>::max()),
-				static_cast<real_t>(g8 / std::numeric_limits<uint8_t>::max()),
-				static_cast<real_t>(b8 / std::numeric_limits<uint8_t>::max()),
-				static_cast<real_t>(a8 / std::numeric_limits<uint8_t>::max()))
-	{
-	}
+    static constexpr auto from_rgb8888(uint32_t v) -> color {
+        return from_rgb8888((v & 0xff000000) >> 24, (v & 0x00ff0000) >> 16,
+                            (v & 0x0000ff00) >> 8, (v & 0x000000ff));
+    }
 
-	color(real_t r,real_t g,real_t b,real_t a)
-        : rgba({r,g,b,a})
-	{
-	}
-
-    color(const vector4& v) : rgba(v) { }
-
-    operator vector4() const { return rgba; }
-
-	uint32_t to_rgb8888() const {
-		return 0;
-	}
+    uint32_t to_rgb8888() const {
+        return 0;
+    }
 };
 
-}
-
+} // namespace pw
 
 #endif

src/core/include/pw/core/core.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal

src/core/include/pw/core/debug.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -26,72 +26,67 @@
 
 #include <pw/core/globals.hpp>
 
-#include <vector>
 #include <functional>
+#include <vector>
 
 namespace pw {
 
 /**
-  * @brief multipurpose logger used internally
-  */
+ * @brief multipurpose logger used internally
+ */
 class debug {
-public:
-
+  public:
     enum level {
-        none,			//!< nothing will be logged, even no errors
-        error,			//!< only errors will be logged
-        warning,		//!< log warnings (non-critical errors)
-        message,		//!< log messages (something to note but not an error)
-        notification,	//!< log some more information
-        info,			//!< log verbose information
-        all = 0xFF		//!< log absolutely everything
+        none,         //!< nothing will be logged, even no errors
+        error,        //!< only errors will be logged
+        warning,      //!< log warnings (non-critical errors)
+        message,      //!< log messages (something to note but not an error)
+        notification, //!< log some more information
+        info,         //!< log verbose information
+        all = 0xFF    //!< log absolutely everything
     };
 
     /**
      * @brief the streaming interface for the logger
      */
     class stream {
-    public:
-
-
+      public:
         stream(debug* log = nullptr);
         ~stream();
         stream(const stream& other);
 
+        stream& operator<<(const bool& value);
+        stream& operator<<(const char* value);
+        stream& operator<<(const std::string& value); ///! log a string
+        stream&
+        operator<<(const std::string_view& value); ///! log a string_view
 
-        stream& operator << (const bool &value);
-        stream& operator << (const char *value);
-        stream& operator << (const std::string& value);		///! log a string
-        stream& operator << (const std::string_view& value); ///! log a string_view
+        stream& operator<<(const float& value);  ///! log a float value
+        stream& operator<<(const double& value); ///! log a double value
 
-        stream& operator << (const float &value);			///! log a float value
-        stream& operator << (const double &value);			///! log a double value
+        stream& operator<<(const int& value);          ///! log a int value
+        stream& operator<<(const unsigned int& value); ///! log a int value
 
-        stream& operator << (const int &value);			///! log a int value
-        stream& operator << (const unsigned int &value);	///! log a int value
+        stream& operator<<(const long& value);          ///! log a long value
+        stream& operator<<(const unsigned long& value); ///! log a int value
 
-        stream& operator << (const long &value);			///! log a long value
-        stream& operator << (const unsigned long &value);	///! log a int value
-
-        stream& operator << (const void *value);			///! pointer
-
-
-    protected:
+        stream& operator<<(const void* value); ///! pointer
 
+      protected:
         debug* _log;
         std::string _line;
     };
 
     /** sets the logging level */
-    void set_level(level level) {_level = level;}
+    void set_level(level level) { _level = level; }
 
     /** gets the logging level */
     level level() const { return _level; }
 
     /**
-      * @brief get the stream interface of the logger
-      * @return return a temporary object that will write and flush the logger
-      */
+     * @brief get the stream interface of the logger
+     * @return return a temporary object that will write and flush the logger
+     */
     static stream s(enum level level = info);
 
     inline static stream d() { return s(debug::info); }
@@ -108,15 +103,14 @@ public:
      * @brief write a message to the log
      * @param message string
      */
-    void write(const std::string &message);
+    void write(const std::string& message);
 
     typedef std::function<void(const char*)> Callback;
     typedef std::vector<Callback> CallbackList;
 
     ~debug();
 
-protected:
-
+  protected:
     debug();
 
     CallbackList _callbacks;
@@ -126,7 +120,7 @@ protected:
 ///**
 //  * @brief helper for changing the log level in a scope
 //  */
-//struct ScopeLogLevel
+// struct ScopeLogLevel
 //{
 //    debug::level levelOutside;
 //    explicit ScopeLogLevel(debug::level level)
@@ -140,12 +134,12 @@ protected:
 //    }
 //};
 
-//template <debug::level level>
-//struct tpScopeLog {
-//    const char* info;
-//    explicit tpScopeLog(const char* i) : info(i) {
-//        debug::s(level) << info;
-//    }
+// template <debug::level level>
+// struct tpScopeLog {
+//     const char* info;
+//     explicit tpScopeLog(const char* i) : info(i) {
+//         debug::s(level) << info;
+//     }
 
 //    ~tpScopeLog() {
 //        debug::s(level) << info;
@@ -154,9 +148,8 @@ protected:
 
 // some macros
 
-//#define LOG_INFO() Log::s()
-//#define LOG_FUNC() LOG_INFO() << __FUNCTION__ << " " << __LINE__ << " "
-}
-
+// #define LOG_INFO() Log::s()
+// #define LOG_FUNC() LOG_INFO() << __FUNCTION__ << " " << __LINE__ << " "
+} // namespace pw
 
 #endif

src/core/include/pw/core/formatter.hpp

@@ -0,0 +1,69 @@
+/*
+ * Copyright (c) 1999-2024 Hartmut Seichter
+ *
+ * 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.
+ *
+ */
+
+#ifndef PW_CORE_FORMATTER_HPP
+#define PW_CORE_FORMATTER_HPP
+
+#include <pw/core/globals.hpp>
+#include <pw/core/matrix.hpp>
+#include <pw/core/vector.hpp>
+
+#include <format>
+#include <numeric>
+#include <string>
+
+template <typename Scalar, std::size_t N>
+struct std::formatter<pw::vector<Scalar, N>> {
+
+    constexpr auto parse(std::format_parse_context& ctx) { return ctx.begin(); }
+
+    constexpr auto format(const pw::vector<Scalar, N>& v,
+                          std::format_context& ctx) const noexcept {
+        return std::format_to(
+            ctx.out(), "{}",
+            std::accumulate(std::next(std::begin(v)), std::end(v),
+                            std::to_string(v[0]),
+                            [](const auto& prev, const auto& e) {
+            return prev + " " + std::to_string(e);
+        }));
+    }
+};
+
+template <typename Scalar, std::size_t Rows, std::size_t Cols>
+struct std::formatter<pw::matrix<Scalar, Rows, Cols>> {
+
+    constexpr auto parse(std::format_parse_context& ctx) { return ctx.begin(); }
+
+    constexpr auto format(const pw::matrix<Scalar, Rows, Cols>& m,
+                          std::format_context& ctx) const noexcept {
+        return std::format_to(
+            ctx.out(), "{}",
+            std::accumulate(std::next(std::begin(m)), std::end(m),
+                            std::format("{}", m[0]),
+                            [](const auto& prev, const auto& e) {
+            return prev + "\n" + std::format("{}", e);
+        }));
+    }
+};
+
+#endif

src/core/include/pw/core/frustum.hpp

@@ -0,0 +1,67 @@
+/*
+ * Copyright (c) 1999-2024 Hartmut Seichter
+ *
+ * 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.
+ *
+ */
+
+#ifndef PW_CORE_FRUSTUM_HPP
+#define PW_CORE_FRUSTUM_HPP
+
+#include <pw/core/globals.hpp>
+#include <pw/core/math.hpp>
+
+namespace pw {
+
+template <std::floating_point Scalar> struct frustum final {
+    Scalar left{-1}, right{1}, bottom{-1}, top{1}, z_near{-1}, z_far{1};
+
+    static constexpr auto make_perspective_symmetric(Scalar fov_h_deg,
+                                                     Scalar aspect_ratio,
+                                                     Scalar z_near,
+                                                     Scalar z_far) -> frustum {
+        const auto tangent_half{
+            std::tan(pw::deg_to_rad(fov_h_deg / Scalar{2}))};
+        const auto top{tangent_half * z_near};
+        const auto right{top * aspect_ratio};
+
+        return {.left   = -right,
+                .right  = right,
+                .bottom = -top,
+                .top    = top,
+                .z_near = z_near,
+                .z_far  = z_far};
+    }
+
+    static constexpr auto make_orthographic_symmetric(Scalar scale,
+                                                      Scalar aspect_ratio,
+                                                      Scalar z_near,
+                                                      Scalar z_far) -> frustum {
+        return {.left   = -scale,
+                .right  = scale,
+                .bottom = -scale * aspect_ratio,
+                .top    = scale * aspect_ratio,
+                .z_near = z_near,
+                .z_far  = z_far};
+    }
+};
+
+} // namespace pw
+
+#endif

src/core/include/pw/core/globals.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -24,6 +24,9 @@
 #define PW_CORE_GLOBALS_HPP
 
 #include <cstddef>
+#include <cstdint>
+#include <cstdlib>
+
 #include <memory>
 #include <string>
 #include <vector>
@@ -34,11 +37,9 @@ using std::shared_ptr;
 using std::unique_ptr;
 using std::weak_ptr;
 
-using std::make_unique;
 using std::make_shared;
+using std::make_unique;
 
-using real_t = float;
-
-}
+} // namespace pw
 
 #endif

src/core/include/pw/core/image.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -29,58 +29,53 @@
 namespace pw {
 
 class image {
-public:
+  public:
+    using size_type = pw::size<std::size_t>;
 
     image() = default;
 
-	using data_t = std::byte;
+    using data_t = std::byte;
 
-    enum pixel_layout {
-        RGB8,
-        RGBA8,
-		LUM,
-		DEPTH,
-		HDR
-    };
+    enum pixel_layout { RGB8, RGBA8, LUM, DEPTH, HDR };
 
+    image(const size_type& s, pixel_layout t, const data_t* ptr = nullptr);
 
-    image(const size& s, pixel_layout t, const data_t *ptr = nullptr);
+    bool create(const size_type& s, pixel_layout t,
+                const data_t* ptr = nullptr);
 
-    bool create(const size& s, pixel_layout t, const data_t *ptr = nullptr);
+    void release(bool release_memory = false);
 
-	void release(bool release_memory = false);
+    const data_t* data() const { return _data.data(); }
+    data_t* data() { return _data.data(); }
 
-	const data_t *data() const { return _data.data(); }
-	data_t *data() { return _data.data(); }
-
-	const float *data_float() const { return reinterpret_cast<const float*>(_data.data()); }
-	float *data_float() { return reinterpret_cast<float*>(_data.data());}
+    const float* data_float() const {
+        return reinterpret_cast<const float*>(_data.data());
+    }
+    float* data_float() { return reinterpret_cast<float*>(_data.data()); }
 
     pixel_layout layout() const;
-    void set_layout(const pixel_layout &layout);
+    void set_layout(const pixel_layout& layout);
 
     uint64_t change_count() const;
-    void set_change_count(const uint64_t &change_count);
+    void set_change_count(const uint64_t& change_count);
 
     static uint32_t bytes_per_pixel(pixel_layout t);
 
     static uint32_t components(pixel_layout t);
 
-    ::pw::size size() const;
+    size_type size() const { return _size; }
 
     void generate_noise();
     bool is_valid() const;
 
-protected:
-
-    ::pw::size _size { 0, 0 };
-    pixel_layout _layout { pixel_layout::RGB8 };
-    uint64_t _change_count { 0 };
-
-	std::vector<data_t> _data;
+  protected:
+    size_type _size{.width = 0, .height = 0};
+    pixel_layout _layout{pixel_layout::RGB8};
+    uint64_t _change_count{0};
 
+    std::vector<data_t> _data;
 };
 
-}
+} // namespace pw
 
 #endif

src/core/include/pw/core/material.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -58,13 +58,10 @@ public:
 
     using texture_t = std::tuple<texture_type,std::string,uint32_t>;
 
-
-
 protected:
+  vector4 _color = {{}, {1.0, 0.0, 1.0, 1.0}};
 
-    vector4 _color = vector4 { 1.0, 0.0, 1.0, 1.0 };
-
-    std::vector<texture_t> _textures;
+  std::vector<texture_t> _textures;
 
 };
 

src/core/include/pw/core/math.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -23,55 +23,60 @@
 #ifndef PW_CORE_MATH_HPP
 #define PW_CORE_MATH_HPP
 
-#include <cmath>
 #include <algorithm>
-//#include <pw/core/matrix.hpp>
-//#include <pw/core/quaternion.hpp>
+#include <cmath>
+#include <concepts>
+#include <numbers>
 
 namespace pw {
 
-constexpr double __PW_PI = 3.1415926535897932384626433832795028841971693993751058209;
-constexpr double __PW_PI_DOUBLE = 2.0 * __PW_PI;
+static constexpr double __PW_PI{
+    3.1415926535897932384626433832795028841971693993751058209};
+static constexpr double __PW_PI_DOUBLE{2 * __PW_PI};
 
-template <typename T>
-inline const T pi() { return static_cast<T>(__PW_PI); }
-
-template <typename T>
-inline const T one_over_pi() { return static_cast<T>(1 / __PW_PI); }
-
-template <typename T>
-inline T rad_to_deg(const T& angle_in_radian) {
-	return angle_in_radian * static_cast<T>(180) * one_over_pi<T>();
+template <std::floating_point T> inline static constexpr T pi() {
+    return static_cast<T>(__PW_PI);
 }
 
-template <typename T>
-inline T deg_to_rad(const T& angle_in_degree) {
-	return angle_in_degree * pi<T>() / static_cast<T>(180);
+template <std::floating_point T> inline const T one_over_pi() {
+    return static_cast<T>(1 / __PW_PI);
 }
 
-template <typename T>
-inline T repeat(const T& t, const T& length) {
-	return std::clamp(t - std::floor(t / length) * length, T(0), length);
+template <std::floating_point T> inline T rad_to_deg(const T& angle_in_radian) {
+    return angle_in_radian * static_cast<T>(180) * one_over_pi<T>();
 }
 
-template <typename T>
-inline T ping_pong(const T& t,const T& length) {
-	auto tn = repeat(t, length * T(2));
-	return length - std::abs(tn - length);
+template <std::floating_point T> inline T deg_to_rad(const T& angle_in_degree) {
+    return angle_in_degree * pi<T>() / static_cast<T>(180);
 }
 
-template <typename T>
-inline T wrap_angle(const T& angle_in_radian) {
-	using std::floor;
-	return angle_in_radian - __PW_PI_DOUBLE * floor( angle_in_radian / __PW_PI_DOUBLE );
+template <typename T> inline T repeat(const T& t, const T& length) {
+    return std::clamp(t - std::floor(t / length) * length, T{}, length);
 }
 
-//void extractRotation(const matrix &A, Quaterniond &q,const unsigned int maxIter)
+template <typename T> inline T ping_pong(const T& t, const T& length) {
+    auto tn = repeat(t, length * T{2});
+    return length - std::abs(tn - length);
+}
+
+template <std::floating_point T> inline T wrap_angle(const T& angle_in_radian) {
+    using std::floor;
+    return angle_in_radian -
+           __PW_PI_DOUBLE * floor(angle_in_radian / __PW_PI_DOUBLE);
+}
+
+// void extractRotation(const matrix &A, Quaterniond &q,const unsigned int
+// maxIter)
 //{
 //	for (auto iter = 0; iter < maxIter; iter++){
 //		auto R = q.matrix();
-//		Vector3d omega = (R.col(0).cross(A.col(0)) + R.col(1).cross(A.col(1)) + R.col(2).cross(A.col(2)))*(1.0 / fabs(R.col(0).dot(A.col(0)) + R.col(1).dot(A.col(1)) + R.col(2).dot(A.col(2))) +1.0e-9);double w = omega.norm();if (w < 1.0e-9)break;q = Quaterniond(AngleAxisd(w, (1.0 / w)*omega))*q;q.normalize();}}
+//		Vector3d omega = (R.col(0).cross(A.col(0)) +
+// R.col(1).cross(A.col(1)) + R.col(2).cross(A.col(2)))*(1.0 /
+// fabs(R.col(0).dot(A.col(0)) + R.col(1).dot(A.col(1)) +
+// R.col(2).dot(A.col(2))) +1.0e-9);double w = omega.norm();if (w
+//< 1.0e-9)break;q = Quaterniond(AngleAxisd(w, (1.0 /
+// w)*omega))*q;q.normalize();}}
 
-}
+} // namespace pw
 
 #endif

src/core/include/pw/core/matrix.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -25,255 +25,279 @@
 
 #include <pw/core/globals.hpp>
 #include <pw/core/math.hpp>
-#include <pw/core/matrixbase.hpp>
+#include <pw/core/vector.hpp>
 
-#include <numeric>
-#include <array>
+#include <type_traits>
+#include <utility>
 
-namespace pw
-{
+namespace pw {
 
-	template <std::size_t R, std::size_t C, typename T, bool RowMajor = false>
-	struct matrix_ : matrixbase_<T, matrix_<R, C, T>>
-	{
-		T data[R * C];
+template <typename Scalar, std::size_t Rows, std::size_t Cols>
+struct matrix final {
 
-		using matrixbase_<T, matrix_<R, C, T>>::matrixbase_;
+    static_assert(Rows > 0 && Cols > 0, "Undefined matrix type.");
 
-		static constexpr std::size_t rows{R};
-		static constexpr std::size_t cols{C};
+    static constexpr unsigned int diag_size{std::min(Rows, Cols)};
+    static constexpr unsigned int rows{Rows};
+    static constexpr unsigned int cols{Cols};
+    static constexpr bool is_square{Rows == Cols};
 
-		static constexpr std::size_t coefficients{R * C};
+    using size_type      = std::common_type_t<decltype(Rows), decltype(Cols)>;
+    using row_type       = vector<Scalar, Cols>;
+    using column_type    = vector<Scalar, Rows>;
+    using diag_type      = vector<Scalar, diag_size>;
+    using transpose_type = matrix<Scalar, Cols, Rows>;
+    using minor_type     = matrix<Scalar, Cols - 1, Rows - 1>;
 
-		using col_type = matrix_<R, 1, T>;
-		using row_type = matrix_<1, C, T>;
-		using transpose_type = matrix_<C, R, T>;
+    using reference       = row_type&;
+    using const_reference = const row_type&;
+    using pointer         = row_type*;
+    using const_pointer   = const row_type*;
 
-		matrix_<R, C, T> &operator=(const matrix_<R, C, T> &other)
-		{
-			for (size_t i = 0; i < other.size(); i++)
-				(*this)[i] = other[i];
-			return *this;
-		}
+    vector<row_type, Rows> m_{};
 
-		matrix_(std::initializer_list<T> args)
-		{
-			typename std::initializer_list<T>::iterator it = args.begin();
-			for (; it != args.end(); it++)
-				data[it - args.begin()] = *it;
-		}
+    auto&& data(this auto&& self) {
+        return std::forward<decltype(self)>(self).m_[0].data();
+    }
 
-		template <typename... Arguments>
-		matrix_ &set(Arguments... values)
-		{
-			static_assert(sizeof...(Arguments) == R * C, "Incorrect number of arguments");
-			data = {values...};
-			return *this;
-		}
+    auto&& operator[](this auto&& self, int r) {
+        return std::forward<decltype(self)>(self).m_[r];
+    }
 
-		inline size_t offset(size_t r, size_t c) const
-		{
-			return (RowMajor) ? r * C + c : c * R + r;
-		}
+    //
+    // part of tuple-protocol
+    //
+    template <size_type idx> auto get(this auto&& self) -> decltype(auto) {
+        static_assert(idx < Rows, "Out of bounds access to a member.");
+        // TODO: use forward_like when clang is catching up
+        return std::forward_like<decltype(self)>(self.m_[idx]);
+        // return std::forward<decltype(self)>(self).m_[idx];
+    }
 
-		inline T &operator()(std::size_t r, std::size_t c)
-		{
-			return data[offset(r, c)];
-		}
+    constexpr auto diagonal() const noexcept -> diag_type {
+        return [this]<std::size_t... Is>(std::index_sequence<Is...>) {
+            return diag_type{(*this)[Is][Is]...};
+        }(std::make_index_sequence<diag_size>{});
+    }
 
-		inline const T &operator()(std::size_t r, std::size_t c) const
-		{
-			return data[offset(r, c)];
-		}
+    constexpr auto trace() const noexcept -> Scalar {
+        static_assert(Rows == Cols,
+                      "trace is only defined for square matrices");
+        return [this, sum = Scalar{0}]<std::size_t... Is>(
+                   std::index_sequence<Is...>) mutable {
+            return ((*this)[Is][Is] + ... + sum);
+        }(std::make_index_sequence<diag_size>{});
+    }
 
-		inline const T *ptr() const { return &data[0]; }
+    constexpr auto column(const unsigned int& c) const noexcept -> column_type {
+        return [this, &c]<std::size_t... Rs>(std::index_sequence<Rs...>) {
+            return column_type{(*this)[Rs][c]...};
+        }(std::make_index_sequence<Cols>{});
+    }
 
-		//! set identity
-		inline matrix_ &set_identity()
-		{
-			for (std::size_t r = 0; r < rows; r++)
-				for (std::size_t c = 0; c < cols; c++)
-					(*this)(r, c) = (c == r) ? T(1) : T(0);
-			return *this;
-		}
+    constexpr auto transposed() const noexcept -> transpose_type {
+        return [this]<std::size_t... Cs>(std::index_sequence<Cs...>) {
+            return transpose_type{(*this).column(Cs)...};
+        }(std::make_index_sequence<Cols>{});
+    }
 
-		inline matrix_ &set_uniform(const T& v)
-		{
-			std::fill(std::begin(data),std::end(data),v);
-			return *this;
-		}
+    static constexpr auto all(const Scalar& value) {
+        return [&value]<std::size_t... Rs>(std::index_sequence<Rs...>) {
+            return matrix{vector<Scalar, Cols>::all((value) + Rs - Rs)...};
+        }(std::make_index_sequence<Rows>{});
+    }
 
-		template <std::size_t Rs, std::size_t Cs, bool RowMajorSlice = RowMajor>
-		auto slice(std::size_t r, std::size_t c) const
-		{
-			matrix_<Rs, Cs, T, RowMajorSlice> s;
-			for (std::size_t ri = 0; ri < Rs; ri++)
-				for (std::size_t ci = 0; ci < Cs; ci++)
-					s(ri, ci) = (*this)(ri + r, ci + c);
-			return s;
-		}
+    static constexpr auto identity() noexcept -> matrix {
+        static_assert(Rows == Cols,
+                      "identity only defined for square matrices");
+        return []<std::size_t... Rs>(std::index_sequence<Rs...>) {
+            return matrix{vector<Scalar, Cols>::basis(Rs)...};
+        }(std::make_index_sequence<Rows>{});
+    }
 
-		template <std::size_t Rs, std::size_t Cs, bool RowMajorSlice = RowMajor>
-		matrix_ &set_slice(const matrix_<Rs, Cs, T, RowMajorSlice> &s,
-						   std::size_t r, std::size_t c)
-		{
-			for (std::size_t ri = 0; ri < Rs; ri++)
-				for (std::size_t ci = 0; ci < Cs; ci++)
-					(*this)(ri + r, ci + c) = s(ri, ci);
-			return *this;
-		}
+    static constexpr auto from_diagonal(const diag_type& d) noexcept -> matrix {
+        static_assert(
+            Rows == Cols,
+            "creating from diagonal vector only defined for square matrices");
+        return [&d]<std::size_t... Rs>(std::index_sequence<Rs...>) {
+            return matrix{vector<Scalar, Cols>::basis(Rs) * d[Rs]...};
+        }(std::make_index_sequence<Rows>{});
+    }
 
-		template <std::size_t Rs, std::size_t Cs, bool RowMajorSlice = RowMajor>
-		auto minor(std::size_t r0, std::size_t c0) const
-		{
-			matrix_<Rs, Cs, T, RowMajorSlice> m;
-			size_t r = 0;
-			for (size_t ri = 0; ri < R; ri++)
-			{
-				size_t c = 0;
-				if (ri == r0)
-					continue;
-				for (size_t ci = 0; ci < C; ci++)
-				{
-					if (ci == c0)
-						continue;
-					m(r, c) = (*this)(ri, ci);
-					c++;
-				}
-				r++;
-			}
-			return m;
-		}
+    constexpr auto minor(std::unsigned_integral auto r0,
+                         std::unsigned_integral auto c0) const noexcept {
+        static_assert(Rows > 1 && Cols > 1, "cannot create minor matrix");
+        return [this, &r0, &c0]<std::size_t... Rs>(std::index_sequence<Rs...>) {
+            return matrix<Scalar, Rows - 1, Cols - 1>{
+                (*this)[(Rs < r0) ? Rs : Rs + 1].minor(c0)...};
+        }(std::make_index_sequence<Rows - 1>{});
+    }
 
-		T determinant() const
-		{
-			T det(0);
-			for (size_t c = 0; c < C; c++)
-				det += ((c % 2 == 0) ? (*this)(0, c) : -(*this)(0, c)) * this->minor<R - 1, C - 1, RowMajor>(0, c).determinant();
-			return det;
-		}
+    constexpr auto determinant() const noexcept -> Scalar {
+        static_assert(Rows == Cols,
+                      "determinant only defined for square matrices");
+        return [this]<std::size_t... Cs>(std::index_sequence<Cs...>) {
+            return ((((Cs % 2 == 0) ? (*this)[0][Cs] : -(*this)[0][Cs]) *
+                     (*this).minor(0u, Cs).determinant()) +
+                    ...);
+        }(std::make_index_sequence<Cols>{});
+    }
 
-		auto transposed() const
-		{
-			transpose_type res;
-			for (size_t r = rows; r-- > 0;)
-				for (size_t c = cols; c-- > 0;)
-					res(c, r) = (*this)(r, c);
-			return res;
-		}
+    constexpr auto inverse() const noexcept -> matrix {
+        const auto inv_det{Scalar(1) / this->determinant()};
+        matrix<Scalar, Rows, Cols> inv; // no initialization needed!
+        for (auto c = 0u; c < Cols; c++) {
+            for (auto r = 0u; r < Rows; r++) {
+                const auto min_det = this->minor(r, c).determinant();
+                const auto co_fact{((r + c) % 2 == 1) ? -min_det : min_det};
+                inv[r][c] = inv_det * co_fact;
+            }
+        }
+        return inv;
+    }
 
-		auto inverse() const
-		{
-			T invDet = T(1) / this->determinant();
-			matrix_<R, C, T, RowMajor> inv;
-			for (int j = 0; j < C; j++)
-				for (int i = 0; i < R; i++)
-				{
-					const T minorDet = this->minor<R - 1, C - 1, RowMajor>(j, i).determinant();
-					const T coFactor = ((i + j) % 2 == 1) ? -minorDet : minorDet;
-					inv(i, j) = invDet * coFactor;
-				}
-			return inv;
-		}
+    constexpr matrix operator*(const Scalar& v) const noexcept {
+        return [this, &v]<std::size_t... Rs>(std::index_sequence<Rs...>) {
+            return matrix{((*this)[Rs] * v)...};
+        }(std::make_index_sequence<Rows>{});
+    }
+    constexpr matrix operator/(const Scalar& v) const noexcept {
+        return operator*(Scalar{1} / v);
+    }
 
-		inline bool row_major() const
-		{
-			return RowMajor;
-		}
+    constexpr matrix operator+(const Scalar& v) const noexcept {
+        return [this, &v]<std::size_t... Rs>(std::index_sequence<Rs...>) {
+            return matrix{((*this)[Rs] + v)...};
+        }(std::make_index_sequence<Rows>{});
+    }
+    constexpr matrix operator-(const Scalar& v) const noexcept {
+        return operator+(-v);
+    }
 
-		inline bool square() const { return R == C; }
+    constexpr auto operator*=(const Scalar& v) noexcept {
+        [this, &v]<std::size_t... Rs>(std::index_sequence<Rs...>) {
+            (((*this)[Rs] *= v), ...);
+        }(std::make_index_sequence<Rows>{});
+        return *this;
+    }
 
-		inline const matrix_ operator+(const matrix_ &other) const
-		{
-			matrix_ res(*this);
-			for (size_t r = 0; r < R; r++)
-				for (size_t c = 0; c < C; c++)
-					res(r, c) += other(r, c);
-			return res;
-		}
+    constexpr auto operator/=(const Scalar& v) noexcept {
+        return operator*=(Scalar{1} / v);
+    }
 
-		inline const matrix_ operator-(const matrix_ &other) const
-		{
-			matrix_ res(*this);
-			for (size_t r = 0; r < R; r++)
-				for (size_t c = 0; c < C; c++)
-					res(r, c) -= other(r, c);
-			return res;
-		}
+    constexpr auto operator+=(const Scalar& v) noexcept {
+        [this, &v]<std::size_t... Rs>(std::index_sequence<Rs...>) {
+            (((*this)[Rs] += v), ...);
+        }(std::make_index_sequence<Rows>{});
+        return *this;
+    }
 
-		auto row(size_t row_) const
-		{
-			row_type r;
-			for (size_t i = 0; i < cols; i++)
-				r[i] = (*this)(row_, i);
-			return r;
-		}
+    constexpr auto operator-=(const Scalar& v) noexcept {
+        return operator+=(-v);
+    }
 
-		auto column(size_t col_) const
-		{
-			col_type c;
-			for (size_t i = 0; i < rows; i++)
-				c[i] = (*this)(i, col_);
-			return c;
-		}
+    template <std::size_t RowsSlice, std::size_t ColsSlice>
+    constexpr auto slice(std::size_t r = 0, std::size_t c = 0) const noexcept
+        -> matrix<Scalar, RowsSlice, ColsSlice> {
+        return [&r, &c, this]<std::size_t... Rs>(std::index_sequence<Rs...>) {
+            return matrix<Scalar, RowsSlice, ColsSlice>{(*this)[r + Rs].slice(
+                std::make_index_sequence<ColsSlice>{}, c)...};
+        }(std::make_index_sequence<RowsSlice>{}); // rowwise expansion
+    }
 
-		static constexpr auto identity()
-		{
-			matrix_ res;
-			for (std::size_t r = 0; r < rows; r++)
-				for (std::size_t c = 0; c < cols; c++)
-					res(r, c) = (c == r) ? T(1) : T(0);
-			return res;
-		}
-	};
+    constexpr auto unproject(const Scalar& w) const noexcept
+        -> matrix<Scalar, Rows + 1, Cols + 1> {
+        return [&w, this]<std::size_t... Rs>(std::index_sequence<Rs...>) {
+            return matrix<Scalar, Rows + 1, Cols + 1>{
+                (Rs < Rows) ? (*this)[Rs].unproject(0)
+                            : vector<Scalar, Cols + 1>::basis(Cols) * w...};
+        }(std::make_index_sequence<Rows + 1>{}); // rowwise expansion
+    }
 
-	template <>
-	inline float matrix_<1, 1, float>::determinant() const
-	{
-		return (*this)(0, 0);
-	}
+    //
+    // Iterators
+    //
+    constexpr const_pointer begin() const { return &m_[0]; }
 
-	template <>
-	inline double matrix_<1, 1, double>::determinant() const
-	{
-		return (*this)(0, 0);
-	}
+    constexpr const_pointer end() const { return &m_[Rows]; }
+};
 
-	template <typename T, std::size_t R, std::size_t Ca, std::size_t Cb>
-	auto operator*(const matrix_<R, Ca, T> &A,
-				   const matrix_<R, Cb, T> &B)
-	{
-		matrix_<R, Cb, T> result;
-		result.zero(); // zero the output
-		for (size_t r = 0; r < R; r++)
-			for (size_t c = 0; c < Cb; c++)
-				for (size_t iI = 0; iI < R; iI++)
-					result(r, c) += A(r, iI) * B(iI, c); // inner product
-		return result;
-	}
-
-	//
-	// common matrix types
-	//
-
-	template <typename T>
-	using matrix2x2_ = matrix_<2, 2, T>;
-	template <typename T>
-	using matrix3x3_ = matrix_<3, 3, T>;
-	template <typename T>
-	using matrix4x4_ = matrix_<4, 4, T>;
-
-	using matrix2x2f = matrix2x2_<float>;
-	using matrix2x2d = matrix2x2_<double>;
-	using matrix2x2 = matrix2x2_<real_t>;
-
-	using matrix3x3f = matrix3x3_<float>;
-	using matrix3x3d = matrix3x3_<double>;
-	using matrix3x3 = matrix3x3_<real_t>;
-
-	using matrix4x4f = matrix4x4_<float>;
-	using matrix4x4d = matrix4x4_<double>;
-	using matrix4x4 = matrix4x4_<real_t>;
+template <> constexpr float matrix<float, 1, 1>::determinant() const noexcept {
+    return (*this)[0][0];
 }
 
+template <>
+constexpr double matrix<double, 1, 1>::determinant() const noexcept {
+    return (*this)[0][0];
+}
+
+template <typename Scalar, std::size_t Ra, std::size_t CaRb, std::size_t Cb>
+constexpr auto operator*(const matrix<Scalar, Ra, CaRb>& A,
+                         const matrix<Scalar, CaRb, Cb>& B) noexcept {
+    matrix<Scalar, Ra, Cb> result{};
+    [&result, &A, &B]<std::size_t... Ia>(std::index_sequence<Ia...>) {
+        (
+            [&]<std::size_t... Ib>(auto&& Ai, std::index_sequence<Ib...>) {
+            (
+                [&]<std::size_t... iI>(auto&& Ai, auto&& Bi,
+                                       std::index_sequence<iI...>) {
+                ((result[Ai][Bi] += A[Ai][iI] * B[iI][Bi]),
+                 ...); // do the actual multiplication
+            }(Ai, Ib, std::make_index_sequence<CaRb>{}),
+                ...); // forward row of A, colum of B and unpack inner loop
+        }(Ia, std::make_index_sequence<Cb>{}),
+            ...); // forward current row, unpack A
+    }(std::make_index_sequence<Ra>{}); // rows of A as input
+    return result;
+}
+
+template <typename ScalarA, typename ScalarB, std::size_t Ra, std::size_t Ca>
+constexpr auto operator*(const matrix<ScalarA, Ra, Ca>& A,
+                         const vector<ScalarB, Ca>& B) noexcept {
+    // first step - should move to concepts to allow for std::array and
+    // std::span to be allowed
+    vector<std::common_type_t<ScalarA, ScalarB>, Ca> result{};
+    [&result, &A, &B]<std::size_t... Ia>(std::index_sequence<Ia...>) {
+        (
+            [&]<std::size_t... Ib>(auto&& Ai, std::index_sequence<Ib...>) {
+            ((result[Ib] += A[Ai][Ib] * B[Ib]), ...);
+        }(Ia, std::make_index_sequence<Ca>{}),
+            ...); // forward current row, unpack A
+    }(std::make_index_sequence<Ra>{}); // rows of A as input
+    return result;
+}
+
+//
+// deduction guides
+//
+
+//
+// type aliases
+//
+template <typename Scalar> using matrix2x2 = matrix<Scalar, 2, 2>;
+template <typename Scalar> using matrix3x3 = matrix<Scalar, 3, 3>;
+template <typename Scalar> using matrix4x4 = matrix<Scalar, 4, 4>;
+
+using matrix2x2f = matrix2x2<float>;
+using matrix3x3f = matrix3x3<float>;
+using matrix4x4f = matrix4x4<float>;
+
+using matrix2x2d = matrix2x2<double>;
+using matrix3x3d = matrix3x3<double>;
+using matrix4x4d = matrix4x4<double>;
+
+} // namespace pw
+
+//
+// tuple protocol
+//
+template <class Scalar, std::size_t R, std::size_t C>
+struct std::tuple_size<pw::matrix<Scalar, R, C>>
+    : std::integral_constant<std::size_t, R> {};
+
+template <std::size_t I, class Scalar, std::size_t R, std::size_t C>
+struct std::tuple_element<I, pw::matrix<Scalar, R, C>> {
+    using type = pw::matrix<Scalar, R, C>::row_type;
+};
+
 #endif

src/core/include/pw/core/matrix_transform.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -23,149 +23,159 @@
 #ifndef PW_CORE_MATRIX_TRANSFORM_HPP
 #define PW_CORE_MATRIX_TRANSFORM_HPP
 
+#include <pw/core/frustum.hpp>
+#include <pw/core/math.hpp>
+#include <pw/core/matrix.hpp>
 #include <pw/core/vector.hpp>
 
 namespace pw {
 
-template <typename T>
 struct matrix_transform {
 
-    inline static
-    matrix_<4,4,T> scale_matrix(const vector3_<T>& s)
-    {
-        matrix_<4,4,T> scale; scale.zero();
-        scale(0,0) = s[0]; scale(1,1) = s[1]; scale(2,2) = s[2]; scale(3,3) = T(1);
-        return scale;
+    template <typename Scalar>
+    constexpr static auto
+    scale_matrix(vector<Scalar, 4>&& diag) noexcept -> matrix<Scalar, 4, 4> {
+        return matrix<Scalar, 4, 4>::from_diagonal(
+            std::forward<vector<Scalar, 4>>(diag));
     }
 
-	inline static
-    matrix_<4,4,T> perspective_frustum_rh(const T &left,const T &right,
-									   const T &bottom,const T &top,
-									   const T &z_near,const T &z_far)
-	{
-		matrix_<4,4,T> frustum; frustum.zero();
+    template <typename Scalar>
+    constexpr static auto
+    frustum_matrix(const frustum<Scalar>& f) -> matrix<Scalar, 4, 4> {
 
-		frustum(0,0) =  T(2) * z_near / (right - left);
-		frustum(1,1) =  T(2) * z_near / (top - bottom);
+        const auto Sx{Scalar{2} * f.z_near / (f.right - f.left)};
+        const auto Sy{Scalar{2} * f.z_near / (f.top - f.bottom)};
 
-        frustum(0,2) =  (right+left)/(right-left);          //A
-        frustum(1,2) =  (top+bottom)/(top-bottom);          //B
-        frustum(2,2) = -(z_far+z_near)/(z_far-z_near);      //C
-        frustum(2,3) = -T(2) * z_far*z_near/(z_far-z_near); //D
+        const auto A{(f.right + f.left) / (f.right - f.left)};
+        const auto B{(f.top + f.bottom) / (f.top - f.bottom)};
+        const auto C{-(f.z_far + f.z_near) / (f.z_far - f.z_near)};
+        const auto D{-Scalar{2} * f.z_far * f.z_near / (f.z_far - f.z_near)};
 
-		frustum(3,2) = -T(1);
+        return {
+            vector{Sx, 0,  A,  0        }, //
+            vector{0,  Sy, B,  0        }, //
+            vector{0,  0,  C,  D        }, //
+            vector{0,  0,  -1, Scalar{1}}  //
+        };
+    }
 
-		return frustum;
-	}
+    template <typename Scalar>
+    constexpr static auto
+    look_at_matrix(vector3<Scalar> position, vector3<Scalar> target,
+                   vector3<Scalar> up) -> matrix<Scalar, 4, 4> {
 
+        const auto lofs = (target - position).normalized(); // line of sight
+        const auto side = lofs.cross(up).normalized();      // side vector
+        const auto upvc = side.cross(lofs).normalized();    // upvector
 
-//    /// creates a projection from a frustum planes with a reversed depth mapped to [0..1]
-//    pub fn make_projection_rh_from_frustum_reversed(
-//        left: f32,
-//        right: f32,
-//        bottom: f32,
-//        top: f32,
-//        z_near: f32,
-//        z_far: f32,
-//    ) -> Mat4 {
-//        assert!(z_near > 0.0 && z_far > 0.0);
+        return {side.unproject(0), upvc.unproject(0),
+                lofs.unproject(0) * Scalar{-1}, position.unproject(1)};
+    }
+};
 
-//        // info!("near {:?}", z_near);
+#if 0
 
-//        //
-//        // reversed z 0..1 projection
-//        //
-//        let a = (right + left) / (right - left); // position in frame horizontal
-//        let b = (top + bottom) / (top - bottom); // position in frame vertical
+    //    /// creates a projection from a frustum planes with a reversed depth
+    //    mapped to [0..1] pub fn make_projection_rh_from_frustum_reversed(
+    //        left: f32,
+    //        right: f32,
+    //        bottom: f32,
+    //        top: f32,
+    //        z_near: f32,
+    //        z_far: f32,
+    //    ) -> Mat4 {
+    //        assert!(z_near > 0.0 && z_far > 0.0);
 
-//        let c = z_near / (z_far - z_near); // lower bound
-//        let d = z_far * z_near / (z_far - z_near); // upper bound
+    //        // info!("near {:?}", z_near);
 
-//        let sx = 2.0 * z_near / (right - left); // scale x
-//        let sy = 2.0 * z_near / (top - bottom); // scale y
+    //        //
+    //        // reversed z 0..1 projection
+    //        //
+    //        let a = (right + left) / (right - left); // position in frame
+    //        horizontal let b = (top + bottom) / (top - bottom); // position in
+    //        frame vertical
 
-//        // reverse z 0..1
-//        // --------------
-//        // sx   0   a   0
-//        // 0    sy  b   0
-//        // 0    0   c   d
-//        // 0    0  -1   0
+    //        let c = z_near / (z_far - z_near); // lower bound
+    //        let d = z_far * z_near / (z_far - z_near); // upper bound
 
-//        Mat4::from_cols(
-//            Vec4::new(sx,   0.0,    0.0,    0.0),
-//            Vec4::new(0.0,  sy,     0.0,    0.0),
-//            Vec4::new(a,    b,      c,     -1.0),
-//            Vec4::new(0.0,  0.0,    d,      0.0),
-//        )
-//    }
-	inline static
-	matrix_<4,4,T> perspective_projection(const T &field_of_view,
-										  const T &aspect_ratio,
-										  const T &z_near,const T &z_far)
-	{
-		const auto tan_half = tan(field_of_view / T(2));
-		const auto right = tan_half * z_near / aspect_ratio;
-		const auto left = -right;
-		const auto top = right / aspect_ratio;
-		const auto bottom = -top;
+    //        let sx = 2.0 * z_near / (right - left); // scale x
+    //        let sy = 2.0 * z_near / (top - bottom); // scale y
 
-        return perspective_frustum_rh(left,right,
-								   bottom,top,
-								   z_near,z_far);
-	}
+    //        // reverse z 0..1
+    //        // --------------
+    //        // sx   0   a   0
+    //        // 0    sy  b   0
+    //        // 0    0   c   d
+    //        // 0    0  -1   0
 
+    //        Mat4::from_cols(
+    //            Vec4::new(sx,   0.0,    0.0,    0.0),
+    //            Vec4::new(0.0,  sy,     0.0,    0.0),
+    //            Vec4::new(a,    b,      c,     -1.0),
+    //            Vec4::new(0.0,  0.0,    d,      0.0),
+    //        )
+    //    }
+    inline static matrix_<4, 4, T>
+    perspective_projection(const T& field_of_view, const T& aspect_ratio,
+                           const T& z_near, const T& z_far) {
+        const auto tan_half = tan(field_of_view / T(2));
+        const auto right    = tan_half * z_near / aspect_ratio;
+        const auto left     = -right;
+        const auto top      = right / aspect_ratio;
+        const auto bottom   = -top;
 
-    inline static
-	matrix_<4,4,T>	orthographic_frustum(T left, T right,
-                                         T bottom,T top,
-                                         T z_near, T z_far)
-    {
+        return perspective_frustum_rh(left, right, bottom, top, z_near, z_far);
+    }
 
-		matrix_<4,4,T> ortho;  ortho.fill(0);
+    inline static matrix_<4, 4, T>
+    orthographic_frustum(T left, T right, T bottom, T top, T z_near, T z_far) {
 
-		ortho(0,0) =  static_cast<T>(2) / (right-left);
-		ortho(1,1) =  static_cast<T>(2) / (top-bottom);
-		ortho(2,2) = -static_cast<T>(2) / (z_far-z_near);
+        matrix_<4, 4, T> ortho;
+        ortho.fill(0);
 
-		ortho(3,0) = -(right+left)/(right-left);
-		ortho(3,1) = -(top+bottom)/(top-bottom);
-		ortho(3,2) = -(z_far+z_near)/(z_far-z_near);
+        ortho(0, 0) = static_cast<T>(2) / (right - left);
+        ortho(1, 1) = static_cast<T>(2) / (top - bottom);
+        ortho(2, 2) = -static_cast<T>(2) / (z_far - z_near);
 
-        ortho(3,3) = 1;
+        ortho(3, 0) = -(right + left) / (right - left);
+        ortho(3, 1) = -(top + bottom) / (top - bottom);
+        ortho(3, 2) = -(z_far + z_near) / (z_far - z_near);
+
+        ortho(3, 3) = 1;
 
         return ortho;
     }
 
-	inline static
-	matrix_<4,4,T>	orthographic_projection(T width,T height,T z_near, T z_far)
-	{
-		return orthographic_frustum(-width / 2, width / 2,
-									-height / 2, height / 2,
-									 z_near,z_far);
-	}
+    inline static matrix_<4, 4, T> orthographic_projection(T width, T height,
+                                                           T z_near, T z_far) {
+        return orthographic_frustum(-width / 2, width / 2, -height / 2,
+                                    height / 2, z_near, z_far);
+    }
 
+    inline static matrix_<4, 4, T> look_at(const vector3_<T>& position,
+                                           const vector3_<T>& target,
+                                           const vector3_<T>& up) {
+        matrix_<4, 4, T> view_matrix;
+        view_matrix.set_identity();
 
-	inline static
-	matrix_<4,4,T> look_at(const vector3_<T> &position,
-						   const vector3_<T> &target,
-						   const vector3_<T> &up)
-	{
-        matrix_<4,4,T> view_matrix; view_matrix.set_identity();
-
-        const vector3_<T> los = (target - position).normalized();           // line of sight
-        const vector3_<T> sid = vector3_<T>::cross(los,up).normalized();    // side vector
-        const vector3_<T> upd = vector3_<T>::cross(sid,los).normalized();   // upvector
+        const vector3_<T> los =
+            (target - position).normalized(); // line of sight
+        const vector3_<T> sid =
+            vector3_<T>::cross(los, up).normalized(); // side vector
+        const vector3_<T> upd =
+            vector3_<T>::cross(sid, los).normalized(); // upvector
 
         // set base vectors
-		view_matrix.set_slice(sid, 0, 0);
-		view_matrix.set_slice(upd, 0, 1);
-		view_matrix.set_slice(los * T(-1), 0, 2);
-		view_matrix.set_slice(position, 0, 3);
+        view_matrix.set_slice(sid, 0, 0);
+        view_matrix.set_slice(upd, 0, 1);
+        view_matrix.set_slice(los * T(-1), 0, 2);
+        view_matrix.set_slice(position, 0, 3);
 
-		return view_matrix;
+        return view_matrix;
     }
-};
-
-}
+
+#endif
+
+} // namespace pw
 
 #endif

src/core/include/pw/core/matrixbase.hpp

@@ -1,121 +0,0 @@
-/*
- * Copyright (c) 1999-2021 Hartmut Seichter
- *
- * 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.
- *
- */
-#ifndef PW_CORE_MATRIXBASE_HPP
-#define PW_CORE_MATRIXBASE_HPP
-
-#include <algorithm>
-#include <iostream>
-#include <iterator>
-#include <numeric>
-#include <type_traits>
-#include <utility>
-#include <initializer_list>
-#include <cmath>
-
-#include <iostream>
-
-namespace pw {
-
-template <typename T, typename Derived>
-struct matrixbase_ {
-
-    using value_type = T;
-
-	Derived& derived() { return static_cast<Derived&>(*this); }
-    const Derived& derived() const { return static_cast<const Derived&>(*this); }
-
-    std::size_t size() const {
-        return std::extent<decltype(Derived::data)>::value;
-    }
-
-    Derived& fill(const T& v) {
-        std::fill(std::begin(derived().data), std::end(derived().data), T(v));
-        return derived();
-    }
-
-    Derived& zero() {
-        return derived().fill(0);
-    }
-
-	inline T squared_norm() const {
-		return std::accumulate(std::begin(derived().data),std::end(derived().data), T(0),
-							   [&](const T& a,const T& b){
-			return a + b * b;
-		});
-	}
-
-	inline T norm() const {
-		return std::sqrt(squared_norm());
-	}
-
-    inline constexpr Derived normalized() const {
-		return *this / this->norm() ;
-	}
-
-	inline void normalize() {
-		*this /= this->norm();
-	}
-
-    using iterator = T*;
-    using const_iterator = const T*;
-    iterator begin() { return &derived().data[0]; }
-    iterator end() { return &derived().data[0] + size(); }
-    const_iterator begin() const { return &derived().data[0]; }
-    const_iterator end() const { return &derived().data[0] + size(); }
-
-    T& operator [] (std::size_t i) {
-        return derived().data[i];
-    }
-
-    const T& operator [] (std::size_t i) const {
-        return derived().data[i];
-    }
-
-    static constexpr T dot(const Derived &a,const Derived &b)
-    {
-        return std::inner_product(std::begin(a),std::end(a),std::begin(b),T(0));
-	}
-
-    static constexpr Derived lerp(const Derived &a,const Derived &b,const T& t)
-    {
-		return a + (b - a) * t;
-	}
-
-
-	inline void operator *= (const T& b) { for (auto & e : *this) e *= b; }
-	inline void operator /= (const T& b) { for (auto & e : *this) e /= b; }
-	inline void operator += (const T& b) { for (auto & e : *this) e += b; }
-	inline void operator -= (const T& b) { for (auto & e : *this) e -= b; }
-
-	inline const Derived operator * (const T& b) const { Derived r(derived()); for (auto & e : r) e *= b; return r; }
-	inline const Derived operator / (const T& b) const { Derived r(derived()); for (auto & e : r) e /= b; return r; }
-	inline const Derived operator + (const T& b) const { Derived r(derived()); for (auto & e : r) e += b; return r; }
-	inline const Derived operator - (const T& b) const { Derived r(derived()); for (auto & e : r) e -= b; return r; }
-
-
-};
-
-}
-
-
-#endif

src/core/include/pw/core/mesh.hpp

@@ -0,0 +1,65 @@
+/*
+ * Copyright (c) 1999-2024 Hartmut Seichter
+ *
+ * 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.
+ *
+ */
+#ifndef PW_CORE_MESH_HPP
+#define PW_CORE_MESH_HPP
+
+#include <pw/core/globals.hpp>
+#include <pw/core/primitives.hpp>
+
+#include <atomic>
+#include <variant>
+
+namespace pw {
+
+struct attribute final {
+    using mask_type    = std::vector<bool>;
+    using int8_type    = std::vector<int8_t>;
+    using int32_type   = std::vector<int32_t>;
+    using float_type   = std::vector<float>;
+    using vector2_type = std::vector<vector2<float>>;
+    using vector3_type = std::vector<vector3<float>>;
+    using vector4_type = std::vector<vector4<float>>;
+
+    using attribute_data =
+        std::variant<std::monostate, mask_type, int8_type, int32_type,
+                     float_type, vector2_type, vector3_type, vector4_type>;
+
+    enum attribute_type {
+        normals,
+        texture_coordinates,
+    };
+
+    attribute_type type{};
+    attribute_data data{};
+};
+
+struct mesh {
+    primitives geometry{};
+    std::vector<attribute> attributes;
+
+    std::atomic<uint64_t> change_count{0};
+};
+
+} // namespace pw
+
+#endif

src/core/include/pw/core/point.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -23,32 +23,14 @@
 #ifndef PW_CORE_POINT_HPP
 #define PW_CORE_POINT_HPP
 
+#include <cstddef>
 #include <pw/core/globals.hpp>
+#include <pw/core/vector.hpp>
 
 namespace pw {
 
-template <typename T_>
-struct point_ {
+template <typename Scalar> using point = vector<Scalar, 2>;
 
-    using value_type = T_;
-
-    T_ x {0}, y {0};
-
-    point_() = default;
-	point_(T_ x_,T_ y_) : x(x_), y(y_) {}
-
-    template <typename To_>
-    point_<To_> cast() const { return point_<To_>(static_cast<To_>(x),static_cast<To_>(y)); }
-
-};
-
-using point = point_<real_t>;
-
-using pointf = point_<float>;
-using pointd = point_<double>;
-using pointi = point_<int>;
-
-
-}
+} // namespace pw
 
 #endif

src/core/include/pw/core/primitives.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -20,26 +20,53 @@
  * SOFTWARE.
  *
  */
-#ifndef PW_CORE_GEOMETRY_HPP
-#define PW_CORE_GEOMETRY_HPP
+#ifndef PW_CORE_PRIMITIVES_HPP
+#define PW_CORE_PRIMITIVES_HPP
 
-#include <pw/core/globals.hpp>
-#include <pw/core/vector.hpp>
 #include <pw/core/aabb.hpp>
+#include <pw/core/globals.hpp>
 #include <pw/core/resource.hpp>
+#include <pw/core/vector.hpp>
+#include <vector>
 
 namespace pw {
 
+struct primitives final {
+
+    enum struct topology_type {
+        point_list,
+        line_list,
+        line_strip,
+        triangle_list,
+        triangle_strip,
+        triangle_fan,
+        line_list_with_adjacency,
+        line_strip_with_adjacency,
+        triangle_list_with_adjacency,
+        triangle_strip_with_adjacency,
+        patch_list
+    };
+
+    using vertex_type = vector3<float>;
+    using index_type  = std::size_t;
+
+    std::vector<vertex_type> vertices{};
+    std::vector<std::size_t> indices{};
+    topology_type topology{};
+};
+
+} // namespace pw
+
+#if 0
 /*
- * NOTE this needs to be rewritten to take into account for *any* kind of geometry
- * Some ideas are drafted down there to separate out the attribute buffers. Things to
- * consider: multiple UVs, triangle soup, per-vertex-color, texture transforms, weights,
- * etc. pp.
+ * NOTE this needs to be rewritten to take into account for *any* kind of
+ * geometry Some ideas are drafted down there to separate out the attribute
+ * buffers. Things to consider: multiple UVs, triangle soup, per-vertex-color,
+ * texture transforms, weights, etc. pp.
  */
 
 class geometry {
-public:
-
+  public:
     /**
      * @brief describes the topology for the primitives based on Vulkan
      */
@@ -52,14 +79,16 @@ public:
         triangle_fan
     };
 
-    using index_t = uint32_t;   //< needs to be compatible with GL_UNSIGNED_INT
+    using index_t   = uint32_t; //< needs to be compatible with GL_UNSIGNED_INT
     using indices_t = std::vector<index_t>;
 
     geometry() = default;
     geometry(primitive_topology_type t, vector3_array v, indices_t i);
     ~geometry() = default;
 
-    void set_primitive_topology(primitive_topology_type t) { _primitive_topology = t; }
+    void set_primitive_topology(primitive_topology_type t) {
+        _primitive_topology = t;
+    }
     primitive_topology_type primitive_topology() { return _primitive_topology; }
 
     void set_vertices(vector3_array v);
@@ -72,7 +101,9 @@ public:
     const vector3_array& normals() const;
 
     void set_texture_coordinates(std::vector<vector2_array> v);
-    const std::vector<vector2_array>& texture_coordinates() const { return _texture_coords;}
+    const std::vector<vector2_array>& texture_coordinates() const {
+        return _texture_coords;
+    }
 
     void transform(const matrix4x4& m);
 
@@ -85,13 +116,12 @@ public:
     uint64_t change_count() const { return _change_count; }
     void set_change_count(uint64_t n) { _change_count = n; }
 
+  protected:
+    primitive_topology_type _primitive_topology =
+        primitive_topology_type::point_list;
 
-protected:
-
-    primitive_topology_type _primitive_topology = primitive_topology_type::point_list;
-
-    vector3_array _vertices;   //!< vertex data
-    indices_t _indices;        //!< indices
+    vector3_array _vertices; //!< vertex data
+    indices_t _indices;      //!< indices
 
     vector3_array _normals;    //!< normal data
     vector3_array _tangents;   //!< tangent data
@@ -99,10 +129,13 @@ protected:
 
     std::vector<vector2_array> _texture_coords; //! texture coordinates
 
-    uint64_t _change_count { 0 };
+    uint64_t _change_count{0};
 };
 
-}
+struct attribute final {
+
+};
 
 #endif
 
+#endif

src/core/include/pw/core/quaternion.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -23,172 +23,313 @@
 #ifndef PW_CORE_QUATERNION_HPP
 #define PW_CORE_QUATERNION_HPP
 
-#include <pw/core/vector.hpp>
 #include <pw/core/axisangle.hpp>
+#include <pw/core/matrix.hpp>
+#include <pw/core/vector.hpp>
+
+#include <concepts>
 
 namespace pw {
 
 /**
  * simplified quaternion class
  */
-template <typename T>
-struct quaternion_ : vector4_<T> {
+template <std::floating_point Scalar> struct quaternion final {
 
-    typedef vector4_<T> base_type;
+    using value_type = vector<Scalar, 4>;
 
-	using typename base_type::value_type;
-    using base_type::base_type;
-    using base_type::x;
-    using base_type::y;
-    using base_type::z;
-    using base_type::w;
-    using base_type::lerp;
-	using base_type::xyz;
-//    using base_type::operator*;
-//    using base_type::operator/;
+    value_type q_{value_type::basis(3)}; // preset to identity
 
+    auto&& x(this auto&& self) {
+        return std::forward<decltype(self)>(self).q_.x();
+    }
+    auto&& y(this auto&& self) {
+        return std::forward<decltype(self)>(self).q_.y();
+    }
+    auto&& z(this auto&& self) {
+        return std::forward<decltype(self)>(self).q_.z();
+    }
+    auto&& w(this auto&& self) {
+        return std::forward<decltype(self)>(self).q_.w();
+    }
 
-	quaternion_(const base_type& other) : base_type(other) {}
+    constexpr auto
+    operator*(const quaternion& rhs) const noexcept -> quaternion {
+        return {rhs.w() * x() + rhs.x() * w() + rhs.y() * z() - rhs.z() * y(),
+                rhs.w() * y() - rhs.x() * z() + rhs.y() * w() + rhs.z() * x(),
+                rhs.w() * z() + rhs.x() * y() - rhs.y() * x() + rhs.z() * w(),
+                rhs.w() * w() - rhs.x() * x() - rhs.y() * y() - rhs.z() * z()};
+    }
 
-	inline const quaternion_ operator * (const quaternion_& rhs) const {
-		return quaternion_(
-					rhs.w()*x() + rhs.x()*w() + rhs.y()*z() - rhs.z()*y(),
-					rhs.w()*y() - rhs.x()*z() + rhs.y()*w() + rhs.z()*x(),
-					rhs.w()*z() + rhs.x()*y() - rhs.y()*x() + rhs.z()*w(),
-					rhs.w()*w() - rhs.x()*x() - rhs.y()*y() - rhs.z()*z()
-					);
-	}
+    constexpr auto conjugate() const noexcept -> quaternion {
+        return {-x(), -y(), -z(), w()};
+    }
 
-	inline auto operator / (const T& rhs) const {
-		return quaternion_( { x() / rhs, y() / rhs, z() / rhs, w() / rhs, } );
-	}
+    constexpr auto norm() const noexcept -> Scalar { return q_.norm(); }
 
-	//! conjugate
-	inline auto conjugate() const { return quaternion_( { -x(),-y(),-z(),w() } ); }
+    constexpr auto operator*(const Scalar& rhs) const noexcept -> quaternion {
+        return {.q_ = q_ * rhs};
+    }
 
-	//! compute inverse
-	inline auto inverse() const {
-		return conjugate() / this->norm();
-	}
+    constexpr auto operator/(const Scalar& rhs) const noexcept -> quaternion {
+        return operator*(Scalar{1} / rhs);
+    }
 
-	inline static auto identity() {
-		return quaternion_({0,0,0,1});
-	}
+    constexpr auto normalized() const noexcept -> quaternion {
+        return {.q_ = q_.normalized()};
+    }
 
-	const matrix4x4_<T> to_matrix() const {
+    constexpr auto inverse() const noexcept -> quaternion {
+        return conjugate() / this->norm();
+    }
 
-		matrix4x4_<T> m; m.set_identity();
+    constexpr auto dot(const quaternion& b) const noexcept -> Scalar {
+        return q_.dot(b.q_);
+    }
 
-		T xx      = x() * x();
-		T xy      = x() * y();
-		T xz      = x() * z();
-		T xw      = x() * w();
+    constexpr auto to_matrix() const noexcept -> matrix<Scalar, 3, 3> {
 
-		T yy      = y() * y();
-		T yz      = y() * z();
-		T yw      = y() * w();
+        const Scalar xx = x() * x();
+        const Scalar xy = x() * y();
+        const Scalar xz = x() * z();
+        const Scalar xw = x() * w();
 
-		T zz      = z() * z();
-		T zw      = z() * w();
+        const Scalar yy = y() * y();
+        const Scalar yz = y() * z();
+        const Scalar yw = y() * w();
 
-		m(0,0) = 1 - 2 * ( yy + zz );
-		m(0,1) =     2 * ( xy - zw );
-		m(0,2) =     2 * ( xz + yw );
+        const Scalar zz = z() * z();
+        const Scalar zw = z() * w();
 
-		m(1,0) =     2 * ( xy + zw );
-		m(1,1) = 1 - 2 * ( xx + zz );
-		m(1,2) =     2 * ( yz - xw );
+        return {
+            vector{1 - 2 * (yy + zz), 2 * (xy - zw), 2 * (xz + yw)},
+            vector{2 * (xy + zw), 1 - 2 * (xx + zz), 2 * (yz - xw)},
+            vector{2 * (xz - yw), 2 * (yz + xw), 1 - 2 * (xx + yy)},
+        };
+    }
 
-		m(2,0) =     2 * ( xz - yw );
-		m(2,1) =     2 * ( yz + xw );
-		m(2,2) = 1 - 2 * ( xx + yy );
+    static constexpr auto identity() noexcept -> quaternion { return {}; }
 
-		return m;
-	}
+    static constexpr auto pi_around_x() noexcept -> quaternion {
+        return {Scalar{1}, Scalar{0}, Scalar{0}, Scalar{0}};
+    }
 
-	static auto from_matrix(const matrix_<4,4,T> &m) {
-		using std::sqrt;
-		auto wtemp = sqrt(T(1) + m(0,0) + m(1,1) + m(2,2)) / T(2);
-		auto w4 = T(4.0) * wtemp;
+    static constexpr auto pi_around_y() noexcept -> quaternion {
+        return {Scalar{0}, Scalar{1}, Scalar{0}, Scalar{0}};
+    }
+
+    static constexpr auto pi_around_z() noexcept -> quaternion {
+        return {Scalar{0}, Scalar{0}, Scalar{1}, Scalar{0}};
+    }
+
+    static constexpr auto lerp(const quaternion& a, const quaternion& b,
+                               const Scalar& t) -> quaternion {
+        return {value_type::lerp(a.q_, b.q_, t).normalized()};
+    }
+
+    /**
+     * @note: a and b need to be normalized
+     */
+    static constexpr auto
+    slerp(const quaternion& a, const quaternion& b, const Scalar& t,
+          const Scalar& eps = Scalar{0.001}) -> quaternion {
+
+        // Calculate angle between them.
+        const Scalar cos_half_theta{a.dot(b)};
+
+        // if qa=qb or qa=-qb then theta = 0 and we can return a
+        if (std::fabs(cos_half_theta) >= Scalar{1}) {
+            return a;
+        }
+
+        // Calculate temporary values.
+        const Scalar half_theta{std::acos(cos_half_theta)};
+        const Scalar sin_half_theta{
+            std::sqrt(Scalar{1} - cos_half_theta * cos_half_theta)};
+
+        // if theta = 180 degrees then result is not fully defined
+        // we could rotate around any axis normal to a or b
+        const auto is_pi = std::fabs(sin_half_theta) < eps;
+
+        // now do the lerp either halfway or across unit sphere
+        const Scalar ratio_a{is_pi ? Scalar{0.5}
+                                   : std::sin((Scalar{1} - t) * half_theta) /
+                                         sin_half_theta};
+
+        const Scalar ratio_b{is_pi ? Scalar{0.5}
+                                   : std::sin(t * half_theta) / sin_half_theta};
+        return {
+            (a.x() * ratio_a + b.x() * ratio_b), // x
+            (a.y() * ratio_a + b.y() * ratio_b), // y
+            (a.z() * ratio_a + b.z() * ratio_b), // z
+            (a.w() * ratio_a + b.w() * ratio_b)  // w
+        };
+    }
+
+    static constexpr auto
+    from_axisangle(const axisangle<Scalar>& aa) -> quaternion {
+
+        const auto sin_half_angle{std::sin(aa.angle / Scalar{2})};
+
+        return {
+            aa.axis.x() * sin_half_angle,  // x
+            aa.axis.y() * sin_half_angle,  // y
+            aa.axis.z() * sin_half_angle,  // z
+            std::cos(aa.angle / Scalar{2}) // w
+        };
+    }
+
+    constexpr static auto
+    from_matrix(const matrix<Scalar, 3, 3>& m) -> quaternion {
+        const auto wtemp =
+            std::sqrt(Scalar{1} + m[0][0] + m[1][1] + m[2][2]) / Scalar{2};
+        const auto w4 = Scalar{4} * wtemp;
+        return {
+            (m[2][1] - m[1][2]) / w4, // x
+            (m[0][2] - m[2][0]) / w4, // y
+            (m[1][0] - m[0][1]) / w4, // z
+            wtemp                     // w
+        };
+    }
+};
+
+// deduction guide for quaternion
+template <std::floating_point... U, class CT = std::common_type_t<U...>>
+quaternion(U...) -> quaternion<CT>;
+
+using quaternionf = quaternion<float>;
+using quaterniond = quaternion<double>;
+
+#if 0
+    constexpr quaternion_ operator*(const quaternion_& rhs) const {
+        return quaternion_(
+            rhs.w() * x() + rhs.x() * w() + rhs.y() * z() - rhs.z() * y(),
+            rhs.w() * y() - rhs.x() * z() + rhs.y() * w() + rhs.z() * x(),
+            rhs.w() * z() + rhs.x() * y() - rhs.y() * x() + rhs.z() * w(),
+            rhs.w() * w() - rhs.x() * x() - rhs.y() * y() - rhs.z() * z());
+    }
+
+    inline auto operator/(const T& rhs) const {
         return quaternion_({
-					(m(2,1) - m(1,2)) / w4,
-					(m(0,2) - m(2,0)) / w4,
-					(m(1,0) - m(0,1)) / w4,
-                    wtemp});
-	}
+            x() / rhs,
+            y() / rhs,
+            z() / rhs,
+            w() / rhs,
+        });
+    }
 
-	static auto normalized_lerp(const quaternion_ &a,const quaternion_ &b,const T &t) {
-		return quaternion_(lerp(a,b,t).normalized());
-	}
+    //! conjugate
+    inline auto conjugate() const {
+        return quaternion_({-x(), -y(), -z(), w()});
+    }
 
-	static auto slerp(const quaternion_& qa,const quaternion_& qb,const T& t)
-	{
-		using std::abs;
-		using std::sqrt;
-		using std::acos;
+    //! compute inverse
+    inline auto inverse() const { return conjugate() / this->norm(); }
 
-		// quaternion to return
-		quaternion_ qm;
-		// Calculate angle between them.
-		T cosHalfTheta = qa.w() * qb.w() + qa.x() * qb.x() + qa.y() * qb.y() + qa.z() * qb.z();
-		// if qa=qb or qa=-qb then theta = 0 and we can return qa
-		if (abs(cosHalfTheta) >= T(1)) {
-			return qa;
-		}
+    inline static auto identity() { return quaternion_({0, 0, 0, 1}); }
 
-		// Calculate temporary values.
-		const T halfTheta = acos(cosHalfTheta);
-		const T sinHalfTheta = sqrt(1.0 - cosHalfTheta * cosHalfTheta);
-		// if theta = 180 degrees then result is not fully defined
-		// we could rotate around any axis normal to qa or qb
-		if (abs(sinHalfTheta) < 0.001){ // fabs is floating point absolute
-			qm.w() = (qa.w() * T(0.5) + qb.w() * T(0.5));
-			qm.x() = (qa.x() * T(0.5) + qb.x() * T(0.5));
-			qm.y() = (qa.y() * T(0.5) + qb.y() * T(0.5));
-			qm.z() = (qa.z() * T(0.5) + qb.z() * T(0.5));
-			return qm;
-		}
-		const T ratioA = sin((value_type(1) - t) * halfTheta) / sinHalfTheta;
-		const T ratioB = sin(t * halfTheta) / sinHalfTheta;
-		//calculate Quaternion.
-		qm.w() = (qa.w() * ratioA + qb.w() * ratioB);
-		qm.x() = (qa.x() * ratioA + qb.x() * ratioB);
-		qm.y() = (qa.y() * ratioA + qb.y() * ratioB);
-		qm.z() = (qa.z() * ratioA + qb.z() * ratioB);
+    const matrix4x4_<T> to_matrix() const {
 
-		return qm;
-	}
+        matrix4x4_<T> m;
+        m.set_identity();
 
+        T xx = x() * x();
+        T xy = x() * y();
+        T xz = x() * z();
+        T xw = x() * w();
 
-	static auto from_axisangle(const axisangle_<T> &aa) {
-		using std::sin;
-		using std::cos;
+        T yy = y() * y();
+        T yz = y() * z();
+        T yw = y() * w();
 
-        const T sinHalfAngle( sin(aa.angle * T(0.5) ));
+        T zz = z() * z();
+        T zw = z() * w();
 
-        return quaternion_<T>( { aa.axis.x() * sinHalfAngle, // x
-                                 aa.axis.y() * sinHalfAngle, // y
-                                 aa.axis.z() * sinHalfAngle, // z
-                                 cos(aa.angle * T(0.5))          // w
-							   }
-							   );
+        m(0, 0) = 1 - 2 * (yy + zz);
+        m(0, 1) = 2 * (xy - zw);
+        m(0, 2) = 2 * (xz + yw);
 
-	}
+        m(1, 0) = 2 * (xy + zw);
+        m(1, 1) = 1 - 2 * (xx + zz);
+        m(1, 2) = 2 * (yz - xw);
+
+        m(2, 0) = 2 * (xz - yw);
+        m(2, 1) = 2 * (yz + xw);
+        m(2, 2) = 1 - 2 * (xx + yy);
+
+        return m;
+    }
+
+    static auto from_matrix(const matrix_<4, 4, T>& m) {
+        using std::sqrt;
+        auto wtemp = sqrt(T(1) + m(0, 0) + m(1, 1) + m(2, 2)) / T(2);
+        auto w4    = T(4.0) * wtemp;
+        return quaternion_({(m(2, 1) - m(1, 2)) / w4, (m(0, 2) - m(2, 0)) / w4,
+                            (m(1, 0) - m(0, 1)) / w4, wtemp});
+    }
+
+    static auto normalized_lerp(const quaternion_& a, const quaternion_& b,
+                                const T& t) {
+        return quaternion_(lerp(a, b, t).normalized());
+    }
+
+    static auto slerp(const quaternion_& qa, const quaternion_& qb,
+                      const T& t) {
+        using std::abs;
+        using std::acos;
+        using std::sqrt;
+
+        // quaternion to return
+        quaternion_ qm;
+        // Calculate angle between them.
+        T cosHalfTheta = qa.w() * qb.w() + qa.x() * qb.x() + qa.y() * qb.y() +
+                         qa.z() * qb.z();
+        // if qa=qb or qa=-qb then theta = 0 and we can return qa
+        if (abs(cosHalfTheta) >= T(1)) {
+            return qa;
+        }
+
+        // Calculate temporary values.
+        const T halfTheta    = acos(cosHalfTheta);
+        const T sinHalfTheta = sqrt(1.0 - cosHalfTheta * cosHalfTheta);
+        // if theta = 180 degrees then result is not fully defined
+        // we could rotate around any axis normal to qa or qb
+        if (abs(sinHalfTheta) < 0.001) { // fabs is floating point absolute
+            qm.w() = (qa.w() * T(0.5) + qb.w() * T(0.5));
+            qm.x() = (qa.x() * T(0.5) + qb.x() * T(0.5));
+            qm.y() = (qa.y() * T(0.5) + qb.y() * T(0.5));
+            qm.z() = (qa.z() * T(0.5) + qb.z() * T(0.5));
+            return qm;
+        }
+        const T ratioA = sin((value_type(1) - t) * halfTheta) / sinHalfTheta;
+        const T ratioB = sin(t * halfTheta) / sinHalfTheta;
+        // calculate Quaternion.
+        qm.w() = (qa.w() * ratioA + qb.w() * ratioB);
+        qm.x() = (qa.x() * ratioA + qb.x() * ratioB);
+        qm.y() = (qa.y() * ratioA + qb.y() * ratioB);
+        qm.z() = (qa.z() * ratioA + qb.z() * ratioB);
+
+        return qm;
+    }
+
+    static auto from_axisangle(const axisangle_<T>& aa) {
+        using std::cos;
+        using std::sin;
+
+        const T sinHalfAngle(sin(aa.angle * T(0.5)));
+
+        return quaternion_<T>({
+            aa.axis.x() * sinHalfAngle, // x
+            aa.axis.y() * sinHalfAngle, // y
+            aa.axis.z() * sinHalfAngle, // z
+            cos(aa.angle * T(0.5))      // w
+        });
+    }
 
 };
-//
-//
-//
-using quaternion = quaternion_<real_t>;
-using quaternionf = quaternion_<float>;
-using quaterniond = quaternion_<double>;
-
-
-
-}
-
-
+#endif
 
+} // namespace pw
 
 #if 0
 /**
@@ -412,9 +553,6 @@ const quaternion_<T> quaternion_<T>::slerp(const quaternion_<T>& qa,const quater
 	return qm;
 }
 
-
-
 #endif
 
-
 #endif

src/core/include/pw/core/rectangle.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -23,53 +23,28 @@
 #ifndef PW_CORE_RECTANGLE_HPP
 #define PW_CORE_RECTANGLE_HPP
 
-#include <initializer_list>
-
 #include <pw/core/point.hpp>
 #include <pw/core/size.hpp>
 
 namespace pw {
 
-template <typename T_>
-struct rectangle_ {
+template <typename Scalar> struct rectangle final {
 
-    point_<T_> position;
-    size_<T_> size;
+    point<Scalar> position{};
+    size<Scalar> size{};
 
-    rectangle_() = default;
-
-    rectangle_(const T_ l[4])
-        : position(point_<T_>(l[0],l[1]))
-        , size(size_<T_>(l[2],l[3]))
-    {
-    }
-
-    rectangle_(const T_(&l)[4])
-        : position(point_<T_>(l[0],l[1]))
-        , size(size_<T_>(l[2],l[3]))
-    {
-    }
-
-    rectangle_(point_<T_> const & p,size_<T_> const & s) : size(s), position(p) {}
-
-    bool contains(const point_<T_>& p) const noexcept
-    {
+    constexpr bool contains(const point<Scalar>& p) const noexcept {
         return p.x >= position.x && p.x <= position.x + size.width &&
-                p.y >= position.y && p.y <= position.y + size.height;
+               p.y >= position.y && p.y <= position.y + size.height;
     }
 
-    template <typename To_>
-    rectangle_<To_> cast() const
-    {
-        return rectangle_<To_>(position.template cast<To_>(),size.template cast<To_>());
+    template <typename ScalarOut>
+    constexpr auto cast() const noexcept -> rectangle<ScalarOut> {
+        return {.position = position.template cast<ScalarOut>(),
+                .size     = size.template cast<ScalarOut>()};
     }
 };
 
-using rectangle  = rectangle_<real_t>;
-using rectanglei = rectangle_<int>;
-using rectanglef = rectangle_<float>;
-using rectangled = rectangle_<double>;
-
-}
+} // namespace pw
 
 #endif

src/core/include/pw/core/resource.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -28,18 +28,15 @@
 
 namespace pw {
 
-class resource {
-public:
-	using change_t = std::atomic_int_fast64_t;
+struct resource final {
 
-    resource() = default;
+    using change_t = std::atomic_int_fast64_t;
 
-    int64_t changecount() const { return _changecount; }
-	void dirty() { ++_changecount; };
+    constexpr int64_t changecount() const noexcept { return changecount_; }
+	void touch() { ++changecount_; };
 
 protected:
-
-	change_t _changecount;
+	change_t changecount_{};
 };
 
 }

src/core/include/pw/core/serialize.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -23,31 +23,52 @@
 #ifndef PW_CORE_SERIALIZE_HPP
 #define PW_CORE_SERIALIZE_HPP
 
+#include <pw/core/axisangle.hpp>
+#include <pw/core/color.hpp>
 #include <pw/core/matrix.hpp>
+#include <pw/core/quaternion.hpp>
+#include <pw/core/vector.hpp>
 
-
-#include <string>
+#include <numeric>
 #include <sstream>
+#include <string>
 
 namespace pw {
 
 struct serialize {
 
-    template <size_t R,size_t C,typename T>
-    inline static std::string matrix(const matrix_<R,C,T>& m) {
+    template <typename T, auto N>
+    constexpr static std::string to_string(const vector<T, N>& v) {
+        return std::accumulate(std::next(std::begin(v)), std::end(v),
+                               std::string(std::to_string(v[0])),
+                               [](const auto& str, const auto& e) {
+            return str + " " + std::to_string(e);
+        });
+    }
+
+    template <typename T, auto R, auto C>
+    constexpr static std::string to_string(const matrix<T, R, C>& m) {
         std::stringstream ss;
 
-        for (int r = 0; r < m.rows;r++) {
-            for (int c = 0; c < m.cols;c++) {
-                ss << m(r,c) << " ";
-            }
-            ss << std::endl;
+        for (int r = 0; r < R; r++) {
+            ss << to_string(m[r]) << '\n';
         }
 
         return ss.str();
     }
+
+    template <typename T>
+    constexpr static std::string to_string(const quaternion<T>& v) {
+        std::stringstream ss;
+        ss << to_string(v.q_);
+        return ss.str();
+    }
+
+    constexpr static std::string to_string(const color& v) {
+        return to_string(v.rgba);
+    }
 };
 
-}
+} // namespace pw
 
 #endif

src/core/include/pw/core/size.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -24,32 +24,41 @@
 #define PW_CORE_SIZE_HPP
 
 #include <pw/core/globals.hpp>
+#include <type_traits>
 
 namespace pw {
 
-template <typename T_>
-struct size_ {
+template <typename Scalar> struct size {
 
-    using value_type = T_;
+    using value_type = Scalar;
+    Scalar width{}, height{};
 
-	T_ width,height;
+    constexpr auto area() const noexcept -> Scalar {
+        if constexpr (std::is_unsigned_v<Scalar>) {
+            return width * height;
+        } else {
+            return std::abs(width * height);
+        }
+    }
 
-    size_() = default;
-
-	size_(T_ w,T_ h) : width(w), height(h) {}
-
-    T_ area() const { return std::abs(width * height); }
-
-    template <typename To_>
-    size_<To_> cast() const { return size_<To_>(static_cast<To_>(width),static_cast<To_>(height)); }
+    template <typename ScalarOut>
+    constexpr auto cast(this auto&& self) noexcept -> size<ScalarOut> {
+        return {.width  = ScalarOut(self.width),
+                .height = ScalarOut(self.height)};
+    }
 
+    template <typename Other>
+    constexpr auto operator()(this auto&& self) noexcept {
+        return Other{self.width, self.height};
+    }
 };
 
-typedef size_<int> size;
+//
+// deduction guide
+//
+template <class... U, class CT = std::common_type_t<U...>>
+size(U...) -> size<CT>;
 
-typedef size_<float> sizef;
-typedef size_<double> sized;
-
-}
+} // namespace pw
 
 #endif

src/core/include/pw/core/time.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -32,14 +32,13 @@ namespace pw {
 /**
  * @brief A simple high resolution timer
  */
-class time {
-public:
+struct time final {
 
-	using tick_t = std::chrono::time_point<std::chrono::high_resolution_clock> ;
+    using tick_t = std::chrono::time_point<std::chrono::high_resolution_clock>;
 
-    time() = default;
+    time()            = default;
     time(const time&) = default;
-    ~time() = default; /// d'tor
+    ~time()           = default; /// d'tor
 
     /**
      * @brief reset timer to current system time
@@ -58,12 +57,10 @@ public:
      */
     static double now();
 
-protected:
-
-    tick_t _start = std::chrono::high_resolution_clock::now();
+  protected:
+    tick_t _start{std::chrono::high_resolution_clock::now()};
 };
 
-}
-
+} // namespace pw
 
 #endif

src/core/include/pw/core/vector.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -23,146 +23,321 @@
 #ifndef PW_CORE_VECTOR_HPP
 #define PW_CORE_VECTOR_HPP
 
-#include <pw/core/matrix.hpp>
+#include <pw/core/globals.hpp>
+
+#include <cmath>
+#include <concepts>
+#include <functional>
+#include <type_traits>
+#include <utility>
 
 namespace pw {
 
-/**
- * Basic vector types used in pixwerx.
- */
+template <typename T, auto N>
+concept Vector2 = (N == 2);
 
-template <typename T>
-struct vector2_ : matrix_<2,1,T> {
+template <typename T, auto N>
+concept Vector3 = (N == 3);
 
-    typedef matrix_<2,1,T> base_type;
+template <typename T, auto N>
+concept Vector4 = (N == 4);
 
-    using base_type::base_type;
-    using base_type::operator = ;
+template <typename Scalar, std::size_t N> struct vector final {
 
-    vector2_(const base_type& m) : base_type(m) {}
-    vector2_(T x_,T y_) : base_type({x_,y_}) {}
+    static_assert(N > 0, "Undefined vector space.");
 
-	inline const T& x() const { return (*this)[0]; }
-	inline T& x() { return (*this)[0]; }
+    using value_type      = Scalar;
+    using size_type       = decltype(N);
+    using difference_type = std::ptrdiff_t;
+    using reference       = value_type&;
+    using const_reference = const value_type&;
+    using pointer         = value_type*;
+    using const_pointer   = const value_type*;
 
-	inline const T& y() const { return (*this)[1]; }
-	inline T& y() { return (*this)[1]; }
+    static constexpr size_type coefficients{N};
 
-    inline auto homogenous(T w = 1) const { return matrix_<3,1,T>( { x(),y(),w } ); }
+    Scalar v_[N]{};
 
-    static T angle_between(const vector2_ &a,const vector2_ &b) {
-        return std::acos( dot( a.normalized(), b.normalized() ) );
+    auto&& data(this auto&& self) {
+        return std::forward<decltype(self)>(self).v_;
     }
 
-    static constexpr auto zero() { return vector2_<T>(0,0); }
+    auto&& operator[](this auto&& self, size_type e) {
+        return std::forward<decltype(self)>(self).v_[e];
+    }
 
-};
+    template <typename ScalarOut>
+    constexpr auto cast() const noexcept -> vector<ScalarOut, N> {
+        return [this]<std::size_t... Ss>(std::index_sequence<Ss...>) {
+            return vector<ScalarOut, N>{ScalarOut((*this)[Ss])...};
+        }(std::make_index_sequence<N>{});
+    }
 
-template <typename T>
-struct vector3_ : matrix_<3,1,T>  {
+    static constexpr auto basis(const auto& d) noexcept {
+        return [&d]<std::size_t... Ss>(std::index_sequence<Ss...>) {
+            return vector{(d == Ss) ? Scalar(1) : Scalar(0)...};
+        }(std::make_index_sequence<N>{});
+    }
 
-    typedef matrix_<3,1,T> base_type;
+    template <typename... Args>
+    static constexpr auto
+    make(Args&&... values) noexcept -> vector<Scalar, sizeof...(Args)> {
+        static_assert(sizeof...(Args) == N, "Incorrect number of arguments.");
+        return {{Scalar(values)...}};
+    }
 
-    using base_type::base_type;
-    using base_type::operator = ;
+    static constexpr auto all(Scalar value) noexcept -> vector {
+        return [&value]<std::size_t... Is>(std::index_sequence<Is...>) {
+            return vector{value + Scalar{Is * 0}...};
+        }(std::make_index_sequence<N>{});
+    }
 
-    vector3_() : base_type() {}
-    vector3_(const base_type& m) : base_type(m) {}
-    vector3_(T x_,T y_,T z_) : base_type({x_,y_,z_}) {}
-	vector3_(const vector2_<T> &m, T w) : base_type({m(0),m(1),w}) {}
+    template <std::integral... Args>
+    constexpr auto swizzle(Args&&... indices) const noexcept
+        -> vector<Scalar, sizeof...(Args)> {
+        return {{Scalar{v_[indices]}...}};
+    }
 
-    inline const T& x() const { return (*this)[0]; }
-    inline T& x() { return (*this)[0]; }
-    inline vector3_& set_x(T val) { (*this)[0] = val; return *this;}
+    template <typename T, T... indices>
+    constexpr auto slice(std::integer_sequence<T, indices...>, T offset = T{0})
+        const noexcept -> vector<Scalar, sizeof...(indices)> {
+        return {{Scalar{v_[indices + offset]}...}};
+    }
 
-    inline const T& y() const { return (*this)[1]; }
-    inline T& y() { return (*this)[1]; }
-    inline vector3_& set_y(T val) { (*this)[1] = val; return *this;}
+    constexpr auto minor(std::unsigned_integral auto d0) const noexcept {
+        return [this, &d0]<std::size_t... Ss>(std::index_sequence<Ss...>) {
+            return vector<Scalar, N - 1>{
+                (*this).v_[(Ss < d0) ? Ss : Ss + 1]...};
+        }(std::make_index_sequence<N - 1>{});
+    }
 
-    inline const T& z() const { return (*this)[2]; }
-    inline T& z() { return (*this)[2]; }
-    inline vector3_& set_z(T val) { (*this)[2] = val; return *this;}
+    static constexpr auto sequence(Scalar factor = Scalar{1},
+                                   Scalar offset = Scalar{0}) noexcept {
+        return [&]<std::size_t... Ss>(std::index_sequence<Ss...>) {
+            return vector<Scalar, N>{{Scalar{Ss} * factor + offset...}};
+        }(std::make_index_sequence<N>{});
+    }
 
-    inline auto xy() const { return vector2_( { x(),y() } ); }
-    inline auto homogenous(T w = 1) const { return matrix_<4,1,T>( { x(),y(),z(),w } ); }
+    static constexpr auto lerp(const vector& A, const vector& B,
+                               const Scalar& t) -> vector {
+        return [&]<std::size_t... Ss>(std::index_sequence<Ss...>) {
+            return vector<Scalar, N>{{A[Ss] + t * (B[Ss] - A[Ss])...}};
+        }(std::make_index_sequence<N>{});
+    }
 
-    inline static constexpr vector3_ cross(const vector3_& lhs,
-                                           const vector3_& rhs)
+    constexpr Scalar dot(const auto& other) const {
+        return [this, &other]<std::size_t... Ss>(std::index_sequence<Ss...>) {
+            return (... + (other[Ss] * (*this)[Ss]));
+        }(std::make_index_sequence<N>{});
+    }
+
+    constexpr auto squared_norm() const noexcept { return dot(*this); }
+
+    constexpr auto norm() const noexcept {
+        return std::sqrt(this->squared_norm());
+    }
+
+    constexpr vector operator*(const Scalar& v) const noexcept {
+        return [this, &v]<std::size_t... Ss>(std::index_sequence<Ss...>) {
+            return vector{{(*this)[Ss] * v...}};
+        }(std::make_index_sequence<N>{});
+    }
+
+    constexpr vector operator/(const Scalar& v) const noexcept {
+        return (*this).operator*(Scalar{1} / v);
+    }
+
+    constexpr vector operator+(const vector& v) const noexcept {
+        return [this, &v]<std::size_t... Ss>(std::index_sequence<Ss...>) {
+            return vector{{(*this)[Ss] + v[Ss]...}};
+        }(std::make_index_sequence<N>{});
+    }
+
+    constexpr vector operator-(const vector& v) const noexcept {
+        return [this, &v]<std::size_t... Ss>(std::index_sequence<Ss...>) {
+            return vector{{(*this)[Ss] - v[Ss]...}};
+        }(std::make_index_sequence<N>{});
+    }
+
+    constexpr vector& operator*=(const Scalar& v) noexcept {
+        [this, &v]<std::size_t... Ss>(std::index_sequence<Ss...>) {
+            (((*this)[Ss] *= v), ...);
+        }(std::make_index_sequence<N>{});
+        return *this;
+    }
+
+    constexpr vector& operator/=(const Scalar& v) noexcept {
+        return operator*=(Scalar{1} / v);
+    }
+
+    constexpr vector& operator+=(const Scalar& v) noexcept {
+        [this, &v]<std::size_t... Ss>(std::index_sequence<Ss...>) {
+            (((*this)[Ss] += v), ...);
+        }(std::make_index_sequence<N>{});
+        return *this;
+    }
+
+    constexpr vector& operator-=(const Scalar& v) noexcept {
+        return operator+=(-v);
+    }
+
+    constexpr vector min(const vector& v) const noexcept {
+        return [this, &v]<std::size_t... Ss>(std::index_sequence<Ss...>) {
+            return pw::vector<Scalar, N>{(std::min(v[Ss], (*this)[Ss]))...};
+        }(std::make_index_sequence<N>{});
+    }
+
+    constexpr vector max(const vector& v) const noexcept {
+        return [this, &v]<std::size_t... Ss>(std::index_sequence<Ss...>) {
+            return pw::vector<Scalar, N>{(std::max(v[Ss], (*this)[Ss]))...};
+        }(std::make_index_sequence<N>{});
+    }
+
+    constexpr auto project() const noexcept -> vector<Scalar, N - 1> {
+        return [this]<std::size_t... Ss>(std::index_sequence<Ss...>) {
+            return vector<Scalar, N - 1>{{(*this)[Ss] / (*this)[N - 1]...}};
+        }(std::make_index_sequence<N - 1>{});
+    }
+
+    constexpr auto
+    unproject(const Scalar& w) const noexcept -> vector<Scalar, N + 1> {
+        return [this, &w]<std::size_t... Ss>(std::index_sequence<Ss...>) {
+            return vector<Scalar, N + 1>{{(Ss < N) ? (*this)[Ss] : w...}};
+        }(std::make_index_sequence<N + 1>{});
+    }
+
+    constexpr vector cross(const auto& rhs) const noexcept
+        requires(Vector3<Scalar, N>)
     {
-        return vector3_( {
-                             lhs[1] * rhs[2] - rhs[1] * lhs[2],
-                             lhs[2] * rhs[0] - rhs[2] * lhs[0],
-                             lhs[0] * rhs[1] - rhs[0] * lhs[1]
-                         }
-                         );
+        return {(*this)[1] * rhs[2] - rhs[1] * (*this)[2],
+                (*this)[2] * rhs[0] - rhs[2] * (*this)[0],
+                (*this)[0] * rhs[1] - rhs[0] * (*this)[1]};
     }
 
+    constexpr vector normalized() const noexcept { return (*this) / norm(); }
 
-    inline static constexpr vector3_<T> forward() { return vector3_<T> ( { T(0), T(0),-T(1) } ); }
-    inline static constexpr vector3_<T> backward() { return vector3_<T>( { T(0), T(0), T(1) } ); }
-    inline static constexpr vector3_<T> right() { return vector3_<T>   ( { T(1), T(0), T(0) } ); }
-    inline static constexpr vector3_<T> left() { return vector3_<T>    ( {-T(1), T(0), T(0) } ); }
-    inline static constexpr vector3_<T> up() { return vector3_<T>      ( { T(0), T(1), T(0) } ); }
-    inline static constexpr vector3_<T> down() { return vector3_<T>    ( { T(0),-T(1), T(0) } ); }
+    auto&& x(this auto&& self)
+        requires(Vector2<Scalar, N> || Vector3<Scalar, N> || Vector4<Scalar, N>)
+    {
+        return std::forward<decltype(self)>(self).v_[0];
+    }
 
-    inline static vector3_<T> x_axis() { return vector3_<T> ( { T(1), T(0), T(0) } ); }
-    inline static vector3_<T> y_axis() { return vector3_<T> ( { T(0), T(1), T(0) } ); }
-    inline static vector3_<T> z_axis() { return vector3_<T> ( { T(0), T(0), T(1) } ); }
+    auto&& y(this auto&& self)
+        requires(Vector2<Scalar, N> || Vector3<Scalar, N> || Vector4<Scalar, N>)
+    {
+        return std::forward<decltype(self)>(self).v_[1];
+    }
 
-    static constexpr auto zero() { return vector3_(0,0,0); }
-};
+    auto&& z(this auto&& self)
+        requires(Vector3<Scalar, N> || Vector4<Scalar, N>)
+    {
+        return std::forward<decltype(self)>(self).v_[2];
+    }
 
-template <typename T>
-struct vector4_ : matrix_<4,1,T>  {
+    auto&& w(this auto&& self)
+        requires(Vector4<Scalar, N>)
+    {
+        return std::forward<decltype(self)>(self).v_[3];
+    }
 
-    typedef matrix_<4,1,T> base_type;
+    static constexpr vector right() noexcept
+        requires(Vector3<Scalar, N>)
+    {
+        return vector{1, 0, 0};
+    };
 
-    using base_type::base_type;
-    using base_type::operator = ;
+    static constexpr vector up() noexcept
+        requires(Vector3<Scalar, N>)
+    {
+        return vector{0, 1, 0};
+    };
 
-    vector4_(T x_,T y_,T z_,T w_) : base_type( {x_,y_,z_,w_} ) {}
-    vector4_(const base_type& m) : base_type(m) {}
-	vector4_(const vector3_<T> &m, T w) : base_type({m(0),m(1),m(2),w}) {}
+    static constexpr vector forward() noexcept
+        requires(Vector3<Scalar, N>)
+    {
+        return vector{0, 0, -1};
+    };
 
-	inline const T& x() const { return (*this)[0]; }
-	inline T& x() { return (*this)[0]; }
+    static constexpr vector x_axis() noexcept
+        requires(Vector2<Scalar, N> || Vector3<Scalar, N> || Vector4<Scalar, N>)
+    {
+        return vector::basis(0);
+    };
 
-	inline const T& y() const { return (*this)[1]; }
-	inline T& y() { return (*this)[1]; }
+    static constexpr vector y_axis() noexcept
+        requires(Vector2<Scalar, N> || Vector3<Scalar, N> || Vector4<Scalar, N>)
+    {
+        return vector::basis(1);
+    };
 
-	inline const T& z() const { return (*this)[2]; }
-	inline T& z() { return (*this)[2]; }
+    static constexpr vector z_axis() noexcept
+        requires(Vector3<Scalar, N> || Vector4<Scalar, N>)
+    {
+        return vector::basis(2);
+    };
 
-	inline const T& w() const { return (*this)[3]; }
-	inline T& w() { return (*this)[3]; }
+    static constexpr vector w_axis() noexcept
+        requires(Vector4<Scalar, N>)
+    {
+        return vector::basis(3);
+    };
 
-	inline auto xyz() const { return vector3_<T>({ x(),y(),z() } ); }
+    //
+    // part of tuple-protocol
+    //
+    template <size_type idx> auto get(this auto&& self) -> decltype(auto) {
+        static_assert(idx < N, "Out of bounds access to a member.");
+        // TODO: use forward_like when clang is catching up
+        // return std::forward_like<decltype(self)>(self.v_[idx]);
+        return std::forward<decltype(self)>(self).v_[idx];
+    }
 
-	inline auto project() const { return vector3_<T>({ x()/w(),y()/w(),z()/w() } ); }
-
-    static constexpr auto zero() { return vector2_<T>(0,0,0,0); }
+    //
+    // Iterators
+    //
+    constexpr const_pointer begin() const { return &v_[0]; }
 
+    constexpr const_pointer end() const { return &v_[N]; }
 };
 
 //
+// deduction guide
 //
+template <class... Scalar>
+vector(Scalar&&... s)
+    -> vector<std::common_type_t<Scalar...>, sizeof...(Scalar)>;
+
 //
+// type aliases
+//
+template <typename Scalar> using vector2 = vector<Scalar, 2>;
+template <typename Scalar> using vector3 = vector<Scalar, 3>;
+template <typename Scalar> using vector4 = vector<Scalar, 4>;
 
-using vector2f = vector2_<float>;
-using vector2d = vector2_<double>;
-using vector2i = vector2_<int>;
-using vector2  = vector2_<real_t>;
-using vector2_array = std::vector<vector2>;
+using vector2f = vector2<float>;
+using vector2d = vector2<double>;
+using vector2i = vector2<int>;
 
-using vector3f = vector3_<float>;
-using vector3d = vector3_<double>;
-using vector3  = vector3_<real_t>;
-using vector3_array = std::vector<vector3>;
+using vector3f = vector3<float>;
+using vector3d = vector3<double>;
+using vector3i = vector3<int>;
 
-using vector4f = vector4_<float>;
-using vector4d = vector4_<double>;
-using vector4  = vector4_<real_t>;
+using vector4f = vector4<float>;
+using vector4d = vector4<double>;
+using vector4i = vector4<int>;
 
-}
+} // namespace pw
+
+//
+// tuple protocol
+//
+template <class Scalar, std::size_t N>
+struct std::tuple_size<pw::vector<Scalar, N>>
+    : std::integral_constant<std::size_t, N> {};
+
+template <std::size_t I, class Scalar, std::size_t N>
+struct std::tuple_element<I, pw::vector<Scalar, N>> {
+    using type = Scalar;
+};
 
 #endif

src/core/src/image.cpp

@@ -1,12 +1,14 @@
 #include "pw/core/image.hpp"
 #include "pw/core/debug.hpp"
 
+#include <cstddef>
 #include <limits>
 #include <random>
+#include <algorithm>
 
 namespace pw {
 
-image::image(const ::pw::size &s,
+image::image(const ::pw::size<std::size_t> &s,
              image::pixel_layout t,
              const data_t *ptr)
 {
@@ -15,11 +17,11 @@ image::image(const ::pw::size &s,
     }
 }
 
-bool image::create(const ::pw::size &s,
+bool image::create(const ::pw::size<std::size_t> &s,
                    image::pixel_layout t,
                    const data_t *ptr)
 {
-	size_t n = bytes_per_pixel(t) * s.area();
+	const auto n = bytes_per_pixel(t) * s.area();
     _size = s;
     _layout = t;
 
@@ -51,7 +53,7 @@ void image::generate_noise()
 void image::release(bool release_memory)
 {
     _data.clear();
-    _size = pw::size();
+    _size = size_type{};
 
     if (release_memory) _data.shrink_to_fit();
 }
@@ -94,10 +96,6 @@ uint32_t image::components(image::pixel_layout t)
     return 0;
 }
 
-::pw::size image::size() const
-{
-    return _size;
-}
 
 image::pixel_layout image::layout() const
 {

src/core/src/time.cpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -42,4 +42,3 @@ double time::now()
 }
 
 }
-

src/core/tests/CMakeLists.txt

@@ -1,15 +1,16 @@
 macro(make_test arg1)
-    add_executable(${arg1}
-        ${arg1}.cpp
-        )
-    target_link_libraries(${arg1}
-            pwcore)
+    add_executable(${arg1} ${arg1}.cpp)
+    target_link_libraries(${arg1} pwcore)
 endmacro()
 
-
-make_test(pwcore_test_matrix)
 make_test(pwcore_test_vector)
+make_test(pwcore_test_matrix)
 make_test(pwcore_test_axisangle)
 make_test(pwcore_test_quaternion)
+make_test(pwcore_test_color)
 make_test(pwcore_test_transform_tools)
+make_test(pwcore_test_aabb)
+make_test(pwcore_test_frustum)
 make_test(pwcore_test_mesh)
+make_test(pwcore_test_serialize)
+make_test(pwcore_test_image)

src/core/tests/pwcore_test_aabb.cpp

@@ -0,0 +1,37 @@
+#include <cstddef>
+#include <iterator>
+#include <pw/core/aabb.hpp>
+
+#include <print>
+#include <vector>
+
+auto main() -> int {
+
+    auto vertices =
+        std::vector<pw::vector<float, 3>>{{-1.f, 1.f, 0},
+                                          {2.f, -1.f, 3.f},
+                                          {0, 0, -1.f},
+                                          {-1'000.f, 2'000.f, 3'000.f}};
+
+    auto indices = std::vector<std::size_t>{0, 1, 2};
+
+    auto aabb_1 =
+        pw::aabb<decltype(vertices)::value_type::value_type,
+                 decltype(vertices)::value_type::coefficients>::make(vertices);
+
+    std::print("aabb::min -> {} {} {}\n", aabb_1.min.x(), aabb_1.min.y(),
+               aabb_1.min.z());
+
+    std::print("aabb::max -> {} {} {}\n", aabb_1.max.x(), aabb_1.max.y(),
+               aabb_1.max.z());
+
+    auto aabb_2 = pw::aabb<decltype(vertices)::value_type::value_type,
+                           decltype(vertices)::value_type::coefficients>::
+        make_from_indexed_vertices(vertices, indices);
+
+    std::print("aabb::min -> {} {} {}\n", aabb_2.min.x(), aabb_2.min.y(),
+               aabb_2.min.z());
+
+    std::print("aabb::max -> {} {} {}\n", aabb_2.max.x(), aabb_2.max.y(),
+               aabb_2.max.z());
+}

src/core/tests/pwcore_test_axisangle.cpp

@@ -1,53 +1,46 @@
 #include <pw/core/axisangle.hpp>
-#include <pw/core/quaternion.hpp>
 #include <pw/core/serialize.hpp>
 
 #include <iostream>
 
-int main(int argc,char **argv) {
+auto main() -> int {
 
-    pw::axisangle_<float> aa = pw::axisangle_<float>();
-
-//	pw::quaternionf qf = pw::quaternionf::from_axisangle(aa);
-//	std::cout << "aa as quaternion as matrix = " << pw::serialize::matrix(qf.to_matrix()) << std::endl;
-//	std::cout << "aa.matrix() = " << pw::serialize::matrix(qf.to_matrix()) << std::endl;
+    auto aa = pw::axisangle<float>();
 
     std::cout << "x-axis" << std::endl;
 
-    aa.axis = pw::vector3::x_axis();
+    aa.axis  = pw::vector3f::x_axis();
     aa.angle = pw::deg_to_rad(45.f);
 
-    std::cout << "aa.matrix() = " << std::endl << pw::serialize::matrix(aa.to_matrix()) << std::endl;
-
+    std::cout << "aa.matrix() = " << std::endl
+              << pw::serialize::to_string(aa.to_matrix()) << std::endl;
     std::cout << "y-axis" << std::endl;
 
-    aa.axis = pw::vector3::y_axis();
+    aa.axis  = pw::vector3f::y_axis();
     aa.angle = pw::deg_to_rad(45.f);
 
-    std::cout << "aa.matrix() = " << std::endl << pw::serialize::matrix(aa.to_matrix()) << std::endl;
+    std::cout << "aa.matrix() = " << std::endl
+              << pw::serialize::to_string(aa.to_matrix()) << std::endl;
 
+    std::cout << "z-axis" << std::endl;
 
+    aa.axis  = pw::vector3f::z_axis();
+    aa.angle = pw::deg_to_rad(45.f);
 
-	std::cout << "z-axis" << std::endl;
+    std::cout << "aa.matrix() = " << std::endl
+              << pw::serialize::to_string(aa.to_matrix()) << std::endl;
 
-	aa.axis = pw::vector3::z_axis();
-	aa.angle = pw::deg_to_rad(45.f);
+    std::cout << "from matrix" << std::endl;
 
-	std::cout << "aa.matrix() = " << std::endl << pw::serialize::matrix(aa.to_matrix()) << std::endl;
-
-
-	std::cout << "from matrix" << std::endl;
-
-	pw::matrix4x4f mrot; mrot.zero();
-	mrot(0,0) = 1;
-	mrot(2,1) = 1;
-	mrot(1,2) = -1;
-
-	pw::axisanglef aa_fm = pw::axisangle::from_matrix(mrot);
-
-	std::cout << pw::serialize::matrix(aa_fm.axis.transposed()) << " " << pw::rad_to_deg(aa_fm.angle) << "deg" << std::endl;
+    auto mrot  = pw::matrix<float, 3, 3>{};
+    mrot[0][0] = 1;
+    mrot[2][1] = 1;
+    mrot[1][2] = -1;
 
+    auto aa_fm = pw::axisanglef::from_matrix(mrot);
 
+    std::cout << pw::serialize::to_string(aa_fm.axis) << " "
+              << pw::rad_to_deg(aa_fm.angle) << "deg" << std::endl;
 
     return 0;
 }

src/core/tests/pwcore_test_color.cpp

@@ -0,0 +1,20 @@
+#include <cstdio>
+#include <pw/core/color.hpp>
+
+#include <pw/core/debug.hpp>
+#include <pw/core/serialize.hpp>
+
+auto main() -> int {
+
+    pw::debug::d() << "pixwerx.color.test\n";
+
+    auto col = pw::color::from_rgb8888(255,0,0,255);
+    pw::debug::d() << "col -> \n" << pw::serialize::to_string(col);
+
+    auto col_hx = pw::color::from_rgb8888(0xFF'EE'BB'AA);
+    pw::debug::d() << "col_hx(0xFFAABBAA) -> \n" << pw::serialize::to_string(col_hx);
+
+
+
+    return 0;
+}

src/core/tests/pwcore_test_frustum.cpp

@@ -0,0 +1,22 @@
+
+#include <pw/core/frustum.hpp>
+
+#include <print>
+#include <vector>
+
+auto main() -> int {
+
+    auto frustum_p = pw::frustum<float>::make_perspective_symmetric(
+        45.f, float{320} / 240, 0.1f, 1'000.f);
+
+    auto frustum_o = pw::frustum<float>::make_orthographic_symmetric(
+        1.5f, float{320} / 240, 0.1f, 1'000.f);
+
+    std::print("l:{} r:{} b:{} t:{} n:{} f:{}\n", frustum_p.left,
+               frustum_p.right, frustum_p.bottom, frustum_p.top, frustum_p.z_near,
+               frustum_p.z_far);
+
+    std::print("l:{} r:{} b:{} t:{} n:{} f:{}\n", frustum_p.left,
+               frustum_o.right, frustum_o.bottom, frustum_o.top, frustum_o.z_near,
+               frustum_o.z_far);
+}

src/core/tests/pwcore_test_image.cpp

@@ -0,0 +1,95 @@
+#include <cstddef>
+#include <cstdint>
+#include <map>
+#include <print>
+#include <tuple>
+#include <vector>
+
+#include <pw/core/image.hpp>
+
+namespace pw {
+
+struct image_layout {
+
+    enum struct pixel_layout { RGB8, RGBA8, LUM8, DEPTH };
+
+    static constexpr auto bits_per_channel(pixel_layout layout) {
+        constexpr static auto lut = std::array{
+            std::make_tuple(pixel_layout::RGB8, sizeof(std::uint8_t) << 3),
+            std::make_tuple(pixel_layout::RGBA8, sizeof(std::uint8_t) << 3),
+            std::make_tuple(pixel_layout::LUM8, sizeof(std::uint8_t) << 3),
+            std::make_tuple(pixel_layout::DEPTH, sizeof(float) << 3)};
+
+        const auto iter = std::find_if(
+            std::begin(lut), std::end(lut),
+            [&layout](const auto& v) { return std::get<0>(v) == layout; });
+
+        return iter != std::end(lut) ? std::get<1>(*iter) : 0u;
+    }
+
+    static constexpr auto channels(pixel_layout layout) {
+        constexpr static auto lut =
+            std::array{std::make_tuple(pixel_layout::RGB8, 3),
+                       std::make_tuple(pixel_layout::RGBA8, 4),
+                       std::make_tuple(pixel_layout::LUM8, 1),
+                       std::make_tuple(pixel_layout::DEPTH, 1)};
+
+        const auto iter = std::find_if(
+            std::begin(lut), std::end(lut),
+            [&layout](const auto& v) { return std::get<0>(v) == layout; });
+
+        return iter != std::end(lut) ? std::get<1>(*iter) : 0u;
+    }
+
+    size<int32_t> size_{};
+    pixel_layout layout_{pixel_layout::RGB8};
+};
+
+struct image_new {
+
+    using buffer_type = std::vector<uint8_t>;
+    using size_type   = pw::size<int32_t>;
+
+    image_layout layout;
+    buffer_type buffer{};
+
+    static constexpr auto make(const image_new::size_type& size_,
+                               image_layout::pixel_layout layout_)
+        -> image_new {
+
+        return {.layout{.layout_ = layout_},
+                .buffer{buffer_type{
+                    static_cast<buffer_type::value_type>(
+                        (image_layout::bits_per_channel(layout_) >> 3) *
+                        size_.area() * image_layout::channels(layout_)),
+                    buffer_type::value_type{0}}}};
+    }
+};
+
+} // namespace pw
+
+auto main() -> int {
+
+    std::print("RGB8 = {}bits per channel\n",
+               pw::image_layout::bits_per_channel(
+                   pw::image_layout::pixel_layout::RGB8));
+    std::print("DEPTH = {}bits per channel\n",
+               pw::image_layout::bits_per_channel(
+                   pw::image_layout::pixel_layout::DEPTH));
+    std::print("LUM8 = {}bits per channel\n",
+               pw::image_layout::bits_per_channel(
+                   pw::image_layout::pixel_layout::LUM8));
+
+    std::print(
+        "RGB8 = {}channels\n",
+        pw::image_layout::channels(pw::image_layout::pixel_layout::RGB8));
+    std::print(
+        "DEPTH = {}channels\n",
+        pw::image_layout::channels(pw::image_layout::pixel_layout::DEPTH));
+    std::print(
+        "LUM8 = {}channels\n",
+        pw::image_layout::channels(pw::image_layout::pixel_layout::LUM8));
+    // pw::image
+
+    pw::image_new::make(pw::iamge, image_layout::pixel_layout layout_)
+}

src/core/tests/pwcore_test_matrix.cpp

@@ -1,130 +1,77 @@
-
-
-#include <pw/core/matrix.hpp>
-#include <pw/core/vector.hpp>
-#include <pw/core/serialize.hpp>
-
 #include <pw/core/debug.hpp>
+#include <pw/core/formatter.hpp>
+#include <pw/core/matrix.hpp>
+#include <pw/core/serialize.hpp>
+#include <pw/core/vector.hpp>
 
-#include <iostream>
-#include <sstream>
+#include <print>
+#include <utility>
 
+auto main() -> int {
 
-int main(int argc,char **argv) {
+    pw::debug::d() << "pixwerx.matrix.test\n";
 
-    using namespace pw;
+    auto m22 = pw::matrix<float, 2, 2>{};
+    pw::debug::d() << "matrix<2,2>{} -> \n" << pw::serialize::to_string(m22);
 
-    matrix2x2f m22; m22.zero();
+    m22[0][0] = 1;
+    m22[0][1] = 2;
+    m22[1][0] = 3;
+    m22[1][1] = 4;
 
+    pw::debug::d() << "matrix<2,2>{1,2,3,4} -> \n"
+                   << pw::serialize::to_string(m22);
 
-    m22(0,0) = 1; m22(0,1) = 2;
-    m22(1,0) = 3; m22(1,1) = 4;
+    auto m22_inv = m22.inverse();
 
-	vector2f v2;
-	v2[0] = 1; v2[1] = 3;
+    pw::debug::d() << "matrix<2,2>{1,2,3,4}.inverse() -> \n"
+                   << pw::serialize::to_string(m22_inv);
 
-	vector2f v3( { 1.f,2.f } );
+    auto m22_inv_inv = m22_inv.inverse();
+    pw::debug::d() << "matrix<2,2>{1,2,3,4}.inverse().inverse() -> \n"
+                   << pw::serialize::to_string(m22_inv_inv);
 
+    auto row_1 = m22_inv[1];
+    pw::debug::d() << "matrix<2,2>{1,2,3,4}.inverse()[1] -> \n"
+                   << pw::serialize::to_string(row_1);
 
-	auto m22_inv = m22.inverse();
-	auto m22_id = m22_inv * m22;
+    auto m22_tp = m22.transposed();
+    pw::debug::d() << "matrix<2,2>{1,2,3,4}.transposed() -> \n"
+                   << pw::serialize::to_string(m22_tp);
 
-	auto v2_t = m22_id * v2;
-	auto v3_t = m22_id * v3;
+    auto m22_tp_col1 = m22_tp.column(1);
+    pw::debug::d() << "matrix<2,2>{1,2,3,4}.transposed().column(1) -> \n"
+                   << pw::serialize::to_string(m22_tp_col1);
 
-	auto v2_f = m22 * v2;
-	auto v2_b = m22_inv * v2_f;
+    auto& [r1, r2] = m22_inv;
 
-    vector2_<float> r_m22 = m22.row(0).transposed();
-    vector2_<float> c_m22 = m22.column(1);
+    auto& [r1_x, r1_y] = r1;
 
-    auto r_m22_h = r_m22.homogenous();
+    std::print("r1.x:{} r1.y:{}\n", r1_x, r1_y);
 
-    debug::d() << "offset(0,1) col-major " << m22.offset(0,1);
-    debug::d() << "det " << m22.determinant();
+    auto m33 = pw::matrix<float, 3, 3>::identity();
 
-	std::cout << "m22 " << pw::serialize::matrix(m22) << std::endl;
-	std::cout << "m22-1 " << pw::serialize::matrix(m22_inv) << std::endl;
-	std::cout << "m22-i " << pw::serialize::matrix(m22_id) << std::endl;
-	std::cout << "v22_t " << pw::serialize::matrix(v2_t) << std::endl;
-	std::cout << "v3_t " << pw::serialize::matrix(v3_t) << std::endl;
+    auto m22_slice = m33.slice<2, 2>(0, 0);
 
+    std::print("{}\n", typeid(m22_slice).name());
 
-	std::cout << "v2_f " << pw::serialize::matrix(v2_f) << std::endl;
-	std::cout << "v2_b " << pw::serialize::matrix(v2_b) << std::endl;
+    auto vvv = pw::vector{1.f, 2, 3, 4};
 
-	std::cout << "v2_b.norm " << v2_b.norm() << std::endl;
+    auto vvv_swizzle = vvv.slice(std::make_index_sequence<2>{});
 
-//	v2_b.normalize();
-	std::cout << "v2_b.normalized " << pw::serialize::matrix(v2_b.normalized()) << std::endl;
+    pw::debug::d() << "matrix<3,3>.slice() -> \n"
+                   << pw::serialize::to_string(m33) << "\n to \n"
+                   << pw::serialize::to_string(m22_slice);
 
-//	std::cout << "v2_b~v3_t " << rad_to_deg(vector2f::angle_between(v2,v3)) << std::endl;
+    // octave/matlab style output
+    std::print("\nm33=[{}]\n", m33);
 
+    auto m44 = m33.unproject(1);
 
+    std::print("\nm44=[{}]\n", m44);
 
-
-
-
-//    vector2f v2 = m22.slice<2,1>(0,0);
-//    m22.set_slice<2,1>(v2,0,0);
-
-//    v2 *= 2;
-//    v2 += 1;
-
-//    m22 *= 3;
-
-//    m22.determinant();
-
-//    debug::d() << sizeof(v2);
-
-//    auto m22_i = m22.inverse();
-
-
-
-#if 0
-
-    pw::matrix4x4d m;
-
-    m.set_identity();
-
-    std::cout << "m = " << pw::serialize::matrix(m) << std::endl;
-
-    std::cout << "row_stride() : " << m.row_stride() << std::endl;
-    std::cout << "col_stride() : " << m.col_stride() << std::endl;
-    std::cout << "rows()       : " << m.rows() << std::endl;
-    std::cout << "cols()       : " << m.cols() << std::endl;
-    std::cout << "data()       : " << m.data() << std::endl;
-    std::cout << "data()[0]    : " << m.data()[0] << std::endl;
-    std::cout << "at(0,0)      : " << m.at(0,0) << std::endl;
-    std::cout << "determinant(): " << m.determinant() << std::endl;
-
-
-    pw::matrix4x4d mi = m.get_inverse();
-
-    std::cout << "mi.at(0,0)   : " << mi.at(0,0) << std::endl;
-
-
-    pw::matrix4x4d mscale = m * 4.2;
-
-
-    std::cout << "mscale = " << pw::serialize::matrix(mscale) << std::endl;
-
-
-    pw::matrix4x4d a;
-
-    for (int r = 0; r < m.rows(); r++) {
-        for (int c = 0; c < m.cols(); c++) {
-            a.at(r,c) = r * m.cols() + c;
-        }
-    }
-    std::cout << "a = " << pw::serialize::matrix(a) << std::endl;
-
-    pw::matrix4x4d r = a * mscale;
-
-    std::cout << "a * mscale = " << pw::serialize::matrix(r) << std::endl;
-
-
-    return 0;
-
-#endif
+    // auto m_init_ded = pw::matrix{
+    //     {1, 2},
+    //     {3, 4}
+    // };
 }

src/core/tests/pwcore_test_mesh.cpp

@@ -1,66 +1,39 @@
-#include <pw/core/vector.hpp>
-#include <pw/core/serialize.hpp>
-#include <pw/core/matrix_transform.hpp>
-#include <pw/core/geometry.hpp>
+#include <pw/core/aabb.hpp>
 #include <pw/core/axisangle.hpp>
+#include <pw/core/formatter.hpp>
+#include <pw/core/matrix_transform.hpp>
+#include <pw/core/mesh.hpp>
+#include <pw/core/primitives.hpp>
+#include <pw/core/serialize.hpp>
+#include <pw/core/vector.hpp>
 
-#include <iostream>
+#include <print>
+#include <variant>
+#include <vector>
 
-int main(int argc,char **argv) {
+namespace pw {} // namespace pw
 
-    pw::geometry geo;
+auto main() -> int {
 
-    pw::vector3_array vs = {
-        { -1, 1, 0 },
-        { -1,-1, 0 },
-        {  1,-1, 0 },
-        {  1, 1, 0 }
-    };
+    auto geom = pw::primitives{
+        .vertices = {{1.2f, 2.4f, 4.8f},
+                     {2.4f, 1.2f, 4.8f},
+                     {1.2f, 4.8f, 2.4f}},
+        .indices  = {0, 1, 2},
 
-    pw::geometry::indices_t idx = {
-        0, 1, 2,
-        0, 2, 3
-    };
+        .topology = pw::primitives::topology_type::triangle_list};
 
+    auto mesh = pw::mesh{
+        .geometry   = geom,
+        .attributes = {{.type = pw::attribute::normals,
+                        .data = std::vector{pw::vector{1.2f, 2.4f, 4.8f}}}}};
 
-    geo.set_vertices(vs);
-    geo.set_indices(idx);
-
-    geo.compute_normals();
-
-
-
-//    amesh.set_vertices(vs);
-
-//    for (auto v : amesh.vertices()) {
-//        std::cout << pw::serialize::matrix(v.transposed()) << std::endl;
-//    }
-
-//    auto scale = pw::matrix_transform<float>::scale_matrix({2,2,2});
-
-//    amesh.transform(scale);
-
-//    std::cout << "after scale" << std::endl;
-
-//    for (auto v : amesh.vertices()) {
-//        std::cout << pw::serialize::matrix(v.transposed()) << std::endl;
-//    }
-
-//    pw::axisangle aa;
-
-//    aa.axis = pw::vector3({ 0, 0, 1 });
-//    aa.angle = pw::deg_to_rad(90.0f);
-
-//    auto rot_mat = aa.to_matrix();
-
-//    amesh.transform(rot_mat);
-
-//    std::cout << "after rotate" << std::endl;
-
-//    for (auto v : amesh.vertices()) {
-//        std::cout << pw::serialize::matrix(v.transposed()) << std::endl;
-//    }
-
+    for (auto i : geom.indices) {
+        std::print("vertex[{}]({})\n", i, geom.vertices[i]);
+    }
 
+    auto aabb = pw::aabb<float, 3>::make_from_indexed_vertices(geom.vertices,
+                                                               geom.indices);
 
+    std::print("aabb.max = ({}) aabb.min=({})\n", aabb.max, aabb.min);
 }

src/core/tests/pwcore_test_quaternion.cpp

@@ -1,30 +1,46 @@
+#include <pw/core/debug.hpp>
 #include <pw/core/quaternion.hpp>
 #include <pw/core/serialize.hpp>
 
-#include <iostream>
+auto main() -> int {
 
-int main(int argc,char **argv) {
+    pw::debug::d() << "pixwerx.quaternion.test\n";
 
-	pw::quaternion_<float> qf;
+    auto q0 = pw::quaternion<float>{};
+    auto q1 = pw::quaternion{1.f, 2.f, 3.f, 4.f};
 
-	std::cout << "qf = " << pw::serialize::matrix(qf) << std::endl;
-    std::cout << "qf.matrix() = " << pw::serialize::matrix(qf.to_matrix()) << std::endl;
+    pw::debug::d() << "q0 = quaternion{} -> \n" << pw::serialize::to_string(q0);
+    pw::debug::d() << "q1 = quaternion{1,2,3,4} -> \n"
+                   << pw::serialize::to_string(q1);
 
-    std::cout << "qf.squared_norm() = " << qf.squared_norm() << std::endl;
-//    std::cout << "qf.dot(qf) = " << qf.dot(qf) << std::endl;
-	std::cout << "qf.conjugate() = " << pw::serialize::matrix(qf.conjugate()) << std::endl;
+    q1 = q1.normalized();
+    pw::debug::d() << "q1 = quaternion{1,2,3,4}.normalized() -> \n"
+                   << pw::serialize::to_string(q1);
 
-	pw::quaternionf qc = qf.conjugate();
-	std::cout << "qf.conjugate() (qc) = " << pw::serialize::matrix(qc.to_matrix()) << std::endl;
+    auto q0_x_q1 = q0 * q1;
+    pw::debug::d() << "q0 * q1 -> \n" << pw::serialize::to_string(q0_x_q1);
 
-	pw::quaternionf qi = qf.inverse();
-	std::cout << "qf.inverse() (qi) = " << pw::serialize::matrix(qi.to_matrix()) << std::endl;
+    auto q1_conj = q1.conjugate();
+    pw::debug::d() << "q1.conjugate() -> \n"
+                   << pw::serialize::to_string(q1_conj);
 
-	pw::quaternionf qmid = pw::quaternionf::normalized_lerp(qi,qf,0.5f);
-//	std::cout << "qmid.dot() (half between qf and qi) = " << pw::rad_to_deg(quaternionf::angle_between()) << std::endl;
+    auto q1_conj_nml = q1_conj.normalized();
+    pw::debug::d() << "q1.conjugate().normalized() -> \n"
+                   << pw::serialize::to_string(q1_conj_nml);
 
+    auto q1_conj_inv = q1_conj.inverse();
+    pw::debug::d() << "q1.conjugate().inverse() -> \n"
+                   << pw::serialize::to_string(q1_conj_inv);
 
+    pw::debug::d() << "quaternion::dot(q0,q1) -> " << q0.dot(q1);
 
+    auto lerp_q0_q1_half = decltype(q0)::lerp(q0, q1, 0.3f);
+    pw::debug::d() << "quaternion::lerp(q0,q1,0.3) -> \n"
+                   << pw::serialize::to_string(lerp_q0_q1_half);
+
+    auto slerp_q0_q1_half = decltype(q0)::slerp(q0, q1, 0.3f);
+    pw::debug::d() << "quaternion::slerp(q0,q1,0.3) -> \n"
+                   << pw::serialize::to_string(slerp_q0_q1_half);
 
     return 0;
 }

src/core/tests/pwcore_test_serialize.cpp

@@ -0,0 +1,12 @@
+#include <pw/core/debug.hpp>
+#include <pw/core/serialize.hpp>
+#include <pw/core/vector.hpp>
+
+#include <iostream>
+
+auto main() -> int {
+
+    auto vec = pw::vector{.333f, .444f, .555f};
+
+    std::cout << pw::serialize::to_string(vec) << '\n';
+}

src/core/tests/pwcore_test_transform_tools.cpp

@@ -1,18 +1,38 @@
 
-#include <pw/core/vector.hpp>
-#include <pw/core/serialize.hpp>
+#include <pw/core/debug.hpp>
 #include <pw/core/matrix_transform.hpp>
+#include <pw/core/serialize.hpp>
+#include <pw/core/vector.hpp>
 
-#include <iostream>
+auto main() -> int {
 
-int main(int argc,char **argv) {
+    pw::debug::d() << "pixwerx.matrix_transform.test\n";
 
-    auto perspective_mat = pw::matrix_transform<float>::perspective_projection(45.f,1.3f,10,100);
-    auto ortho_mat = pw::matrix_transform<float>::orthographic_frustum(-1,1,1,-1,10,100);
-    auto lookat_mat = pw::matrix_transform<float>::look_at(pw::vector3(0,0,5),pw::vector3(0,0,0),pw::vector3(0,1,0));
+    auto scale_123 =
+        pw::matrix_transform::scale_matrix<float>({1.f, 2.f, 3.f, 1.f});
+    pw::debug::d() << "matrix_transform::scale(1,2,3) -> \n"
+                   << pw::serialize::to_string(scale_123);
 
+    auto fm_default =
+        pw::matrix_transform::frustum_matrix(pw::frustum<float>{});
+    pw::debug::d() << "matrix_transform::frustum::default -> \n"
+                   << pw::serialize::to_string(fm_default);
 
-    std::cout << pw::serialize::matrix(perspective_mat) << std::endl;
-    std::cout << pw::serialize::matrix(ortho_mat) << std::endl;
-    std::cout << pw::serialize::matrix(lookat_mat) << std::endl;
+    auto fm_45deg_1dot3 = pw::matrix_transform::frustum_matrix(
+        pw::frustum<float>::make_perspective_symmetric(45.f, 1.3f, 0.1f,
+                                                       1000.f));
+    pw::debug::d() << "matrix_transform::frustum::45deg -> \n"
+                   << pw::serialize::to_string(fm_45deg_1dot3);
+
+    auto fm_ortho = pw::matrix_transform::frustum_matrix(
+        pw::frustum<float>::make_orthographic_symmetric(0.5f, 1.3f, 0.1f,
+                                                        1000.f));
+    pw::debug::d() << "matrix_transform::frustum::ortho -> \n"
+                   << pw::serialize::to_string(fm_ortho);
+
+    auto lookat_mat = pw::matrix_transform::look_at_matrix(
+        pw::vector{0.f, 0, 5}, pw::vector{0.f, 0, 0}, pw::vector{0.f, 1, 0});
+
+    pw::debug::d() << "matrix_transform::look_at -> \n"
+                   << pw::serialize::to_string(lookat_mat);
 }

src/core/tests/pwcore_test_vector.cpp

@@ -1,51 +1,88 @@
 
-#include <pw/core/vector.hpp>
+#include <pw/core/formatter.hpp>
 #include <pw/core/serialize.hpp>
+#include <pw/core/vector.hpp>
 
-#include <iostream>
+#include <print>
+#include <tuple>
+#include <vector>
 
-int main(int ,char **) {
+auto main() -> int {
 
-    pw::vector2_<float> v2_A = { 3.2, 1.2 };
-    pw::vector2_<float> v2_B = { 3.2, 1.2 };
+    auto v2_A = pw::vector{3.2, 1.2};
+    auto v2_B = pw::vector{6.4, 2.4};
 
-    auto AB_lerp = pw::vector2f::lerp(v2_A,v2_B,0.5);
+    auto AB_lerp = decltype(v2_A)::lerp(v2_A, v2_B, 0.5);
 
-    pw::vector4_<float> v4;
-    pw::vector3f v = pw::vector3f::backward();
+    std::print("lerp A:({}) B:({}) t:({}) -> {}\n",
+               pw::serialize::to_string(v2_A), pw::serialize::to_string(v2_B),
+               0.5, pw::serialize::to_string(AB_lerp));
 
-    v4.fill(1.5);
+    auto v4_14 = pw::vector4<float>::all(1.4);
+    auto v4_sq = pw::vector4<float>::sequence(1.4);
+    auto v3_fw = pw::vector3f::forward();
 
-    std::cout << "v4 = " << pw::serialize::matrix(v4) << std::endl;
+    std::print("all(1.4) -> {}\n", pw::serialize::to_string(v4_14));
+    std::print("sequence(1.4) -> {}\n", pw::serialize::to_string(v4_sq));
+    std::print("forward() -> {}\n", pw::serialize::to_string(v3_fw));
 
-//    std::cout << "rows()       : " << v4.rows() << std::endl;
-//    std::cout << "cols()       : " << v4.cols() << std::endl;
-    std::cout << "ptr()       : " << v4.ptr() << std::endl;
-    std::cout << "ptr()[0]    : " << v4.ptr()[0] << std::endl;
-    std::cout << "(0,0)      : " << v4(0,0) << std::endl;
+    auto v3_sw_1 = v4_sq.swizzle(0, 1); // xy
+    auto v3_sw_2 = v4_sq.swizzle(1, 0); // yx
+    auto v3_sw_3 = v4_sq.swizzle(0, 2); // xz
 
-    auto v3 = v4.xyz();
+    std::print("swizzle(0,1) aka xy -> {}\n",
+               pw::serialize::to_string(v3_sw_1));
+    std::print("swizzle(1,0) aka yx -> {}\n",
+               pw::serialize::to_string(v3_sw_2));
+    std::print("swizzle(0,2) aka xz -> {}\n",
+               pw::serialize::to_string(v3_sw_3));
 
-    auto v3_p = v4.project();
+    auto v3_up_1 = v3_fw.unproject(1);
+    auto v3_pj_1 = v3_up_1.project();
 
-    auto v3_h = v.homogenous();
+    std::print("unproject(1) -> {}\n", pw::serialize::to_string(v3_up_1));
+    std::print("project() -> {}\n", pw::serialize::to_string(v3_pj_1));
 
-//	auto v3_lerp = vector4f::
+    auto v3_nlz = v3_up_1.normalized();
+    std::print("normalized() -> {}\n", pw::serialize::to_string(v3_nlz));
 
-    std::cout << "v3 = " << pw::serialize::matrix(v3) << std::endl;
-
-    std::cout << "v3.normalized() = " << pw::serialize::matrix(v3.normalized()) << std::endl;
-
-    auto e1 = pw::vector3 { 2.0, 0.0, 0.0 };
-    auto e2 = pw::vector3 { 0.0, 0.0, 2.0 };
+    auto e1 = pw::vector{2.0, 0.0, 0.0};
+    auto e2 = pw::vector{0.0, 0.0, 2.0};
 
     auto e2_e1 = e1 - e2;
 
-    auto n_e1_e2 = pw::vector3::cross(e1,e2);
+    std::print("{} - {} -> {}\n", pw::serialize::to_string(e1),
+               pw::serialize::to_string(e2), pw::serialize::to_string(e2_e1));
 
-    std::cout << "e1xe2 " << pw::serialize::matrix(n_e1_e2) << std::endl;
+    auto e1_cross_e2 = e1.cross(e2);
+    std::print("{} x {} -> {}\n", pw::serialize::to_string(e1),
+               pw::serialize::to_string(e2),
+               pw::serialize::to_string(e1_cross_e2));
 
-    std::cout << "e1xe2 " << pw::serialize::matrix(e2_e1) << std::endl;
+    auto e1_dot_e2 = e1.dot(e2);
+
+    std::print("{} * {} -> {}\n", pw::serialize::to_string(e1),
+               pw::serialize::to_string(e2), e1_dot_e2);
+
+    auto vec3f = pw::vector{1.f, 2.f, 3.f};
+
+    std::print("y: {}\n", vec3f.get<1>());
+    std::print("tuple size {}\n", std::tuple_size_v<decltype(vec3f)>);
+
+    auto& [x, y, z] = vec3f;
+    std::print("x:{} y:{} z:{}\n", x, y, z);
+
+    x++;
+    y--;
+    z = 0;
+    std::print("x:{} y:{} z:{}\n", x, y, z);
+
+    auto min_ab{vec3f.min(pw::vector{1.3f, 0.9f, 2.3f})};
+
+    std::print("min((1.3f,0.9f,2.3f),({} {} {})) -> ({} {} {})\n", vec3f.x(),
+               vec3f.y(), vec3f.z(), min_ab.x(), min_ab.y(), min_ab.z());
+
+    std::print("{}", vec3f);
 
     return 0;
 }

src/deps/glad/CMakeLists.txt

@@ -1,11 +1,11 @@
-cmake_minimum_required(VERSION 3.8)
+cmake_minimum_required(VERSION 3.20)
 
 project(glad)
 
 add_library(glad STATIC src/glad.c)
 
 target_include_directories(
-	glad
-	PUBLIC
-	include
-	)
+ glad
+ PUBLIC
+ include
+ )

src/geometry/include/pw/geometry/primitives.hpp

@@ -1,21 +1,22 @@
 #ifndef PW_GEOMETRY_PRIMITIVES_HPP
 #define PW_GEOMETRY_PRIMITIVES_HPP
 
-#include <pw/core/geometry.hpp>
+#include <pw/core/primitives.hpp>
 
 namespace pw {
 
-	struct primitives {
+struct primitives {
 
-        static geometry box(real_t size_x, real_t size_y, real_t size_z);
+    static geometry box(real_t size_x, real_t size_y, real_t size_z);
 
-        static geometry sphere(real_t radius,int divisions_latitude,int divisions_longitude);
+    static geometry sphere(real_t radius, int divisions_latitude,
+                           int divisions_longitude);
 
-        static geometry cone();
-
-        static geometry pyramid();
-	};
+    static geometry cone();
 
+    static geometry pyramid();
 };
 
+}; // namespace pw
+
 #endif

src/io/include/pw/io/image_io.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -28,27 +28,22 @@
 
 namespace pw {
 
-class image_io {
-public:
-
+struct image_io final {
     static image_io& get();
 
-    image read(std::string const& uri,uint32_t flags = 0);
+    image read(std::string const& uri, uint32_t flags = 0);
 
-    bool write(const std::string &uri, const image &img, uint32_t flags = 0);
+    bool write(const std::string& uri, const image& img, uint32_t flags = 0);
 
     ~image_io();
 
-
-protected:
-
+  protected:
     struct impl;
     std::unique_ptr<impl> _impl;
 
     image_io();
-
 };
 
-}
+} // namespace pw
 
 #endif

src/io/src/image_io.cpp

@@ -6,77 +6,66 @@
 #include "stb_image.h"
 
 #define STB_IMAGE_WRITE_IMPLEMENTATION
- #include "stb_image_write.h"
+#include "stb_image_write.h"
 
 #include <future>
 
 namespace pw {
 
-struct image_io::impl
-{
+struct image_io::impl {
 
-    image read_impl(const std::string& uri,uint32_t flags)
-    {
-        int x{ 0 }, y{ 0 }, n{ 0 };
+    image read_impl(const std::string& uri, uint32_t flags) {
+        int x{0}, y{0}, n{0};
         const auto data = stbi_load(uri.c_str(), &x, &y, &n, 4);
 
         if (data) {
 
-			image r;
+            image r;
 
-            r.create(size(x,y),image::pixel_layout::RGBA8,reinterpret_cast<image::data_t*>(data));
+            r.create(size{x, y}.cast<image::size_type::value_type>(),
+                     image::pixel_layout::RGBA8,
+                     reinterpret_cast<image::data_t*>(data));
 
-			stbi_image_free(data);
+            stbi_image_free(data);
 
             return r;
-		} else {
+        } else {
 
-			debug::e() << stbi_failure_reason();
-		}
+            debug::e() << stbi_failure_reason();
+        }
 
         return image();
     }
 
-    bool write(const std::string& uri, const image& img, uint32_t flags)
-    {
-        auto res = stbi_write_png(uri.c_str(),img.size().width,img.size().height,
-                       image::components(img.layout()),
-                       img.data(),
-                       img.size().width * image::components(img.layout()));
+    bool write(const std::string& uri, const image& img, uint32_t flags) {
+        auto res =
+            stbi_write_png(uri.c_str(), img.size().width, img.size().height,
+                           image::components(img.layout()), img.data(),
+                           img.size().width * image::components(img.layout()));
 
         return 0 == res;
     }
-
 };
 
-
-image_io &image_io::get()
-{
+image_io& image_io::get() {
     static image_io instance;
     return instance;
 }
 
-image image_io::read(const std::string &uri, uint32_t flags)
-{
-    return _impl->read_impl(uri,flags);
+image image_io::read(const std::string& uri, uint32_t flags) {
+    return _impl->read_impl(uri, flags);
 }
 
-bool image_io::write(const std::string &uri, const image& img, uint32_t flags)
-{
-    return _impl->write(uri,img,flags);
+bool image_io::write(const std::string& uri, const image& img, uint32_t flags) {
+    return _impl->write(uri, img, flags);
 }
 
 //
 //
 //
 
-image_io::image_io()
-{
-    _impl = make_unique<impl>();
-}
+image_io::image_io() { _impl = make_unique<impl>(); }
 
-image_io::~image_io()
-{
-}
+image_io::~image_io() {}
 
-}
+} // namespace pw

src/scene/include/pw/scene/components/camera.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 1999-2020 Hartmut Seichter
+ * Copyright (C) 1999-2024 Hartmut Seichter
  *
  * Redistribution and use in source and binary forms, with or without
  * modification, are permitted provided that the following conditions
@@ -32,17 +32,15 @@
 namespace pw {
 
 struct camera {
-    matrix4x4 view = matrix4x4::identity();
-    matrix4x4 projection = matrix_transform<float>::look_at(vector3{0,0,0},
-                                                            vector3{0,0,1},
-                                                            vector3{0,1,0});
+    matrix4x4f projection = matrix4x4f::identity();
+    matrix4x4f view       = matrix_transform<float>::look_at_matrix(
+        vector3f{}, vector3f::x_axis(), vector3f::y_axis());
 
-    rectangle viewport = pw::rectangle({0, 0, 1, 1});
+    matrix4x4f viewport = pw::rectangle{{0.f, 0.f} {1.f, 1.f}};
 
     uint32_t mask = 0xFFFFFF;
 };
 
-}
-
+} // namespace pw
 
 #endif

src/scene/include/pw/scene/components/projection.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 1999-2020 Hartmut Seichter
+ * Copyright (C) 1999-2024 Hartmut Seichter
  *
  * Redistribution and use in source and binary forms, with or without
  * modification, are permitted provided that the following conditions
@@ -73,4 +73,3 @@ struct projection {
 }
 
 #endif
-

src/scene/include/pw/scene/components/relationship.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 1999-2020 Hartmut Seichter
+ * Copyright (C) 1999-2024 Hartmut Seichter
  *
  * Redistribution and use in source and binary forms, with or without
  * modification, are permitted provided that the following conditions
@@ -25,7 +25,6 @@
 #ifndef PW_SCENE_COMPONENTS_RELATIONSHIP_HPP
 #define PW_SCENE_COMPONENTS_RELATIONSHIP_HPP
 
-
 #include <pw/scene/scene.hpp>
 
 namespace pw {
@@ -34,22 +33,16 @@ namespace pw {
  * @brief entity relations are hidden and managed by the entity itself
  */
 struct parent {
-    parent() = default;
-    parent(const parent&) = default;
 
-    entt::entity entity { entt::null };
+    entt::entity entity{entt::null};
 
     operator bool() const { return entity != entt::null; }
 };
 
 struct children {
-    children() = default;
-    children(const children&) = default;
-
     std::vector<entt::entity> entities;
 };
 
-
-}
+} // namespace pw
 
 #endif

src/scene/include/pw/scene/components/transform.hpp

@@ -7,65 +7,64 @@
 
 namespace pw {
 
-struct transform {
+struct transform final {
 
-    transform() = default;
-    transform(const transform&) = default;
-
-	inline const matrix4x4& local() const { return _local; }
-    void set_local(const matrix4x4 &local)
-    {
-        _local = local;
-    }
-
-	inline const matrix4x4& global() const { return _global; }
-
-    void set_global(const matrix4x4 &global)
-    {
-        matrix4x4 diff = _global.inverse() * global;
-        _local = _local * diff;
+    void set_global(const matrix4x4f& global) {
+        matrix4x4f diff = _global.inverse() * global;
+        _local          = _local * diff;
 
         _global = global;
     }
 
-	inline transform& translate(const real_t &x, const real_t &y, const real_t &z) {
-        _local(0,3) += x;_local(1,3) += y;_local(2,3) += z;
-		return *this;
-	}
+    inline transform& translate(const float& x, const float& y,
+                                const float& z) {
+        _local[0][3] += x;
+        _local[1][3] += y;
+        _local[2][3] += z;
+        return *this;
+    }
 
-	inline transform& set_translation(const real_t &x, const real_t &y, const real_t &z) {
-		_local(0,3) = x;_local(1,3) = y;_local(2,3) = z;
-		return *this;
-	}
+    inline transform& set_translation(const float& x, const float& y,
+                                      const float& z) {
+        _local[0][3] = x;
+        _local[1][3] = y;
+        _local[2][3] = z;
+        return *this;
+    }
 
-	inline transform& rotate(const quaternion& q) {
-        _local = _local * q.to_matrix();
-		return *this;
-	}
+    inline transform& rotate(const quaternionf& q) {
+        _local = _local * q.to_matrix().unproject(1);
+        return *this;
+    }
 
-	inline transform& set_rotation(const quaternion& q) {
-		_local = q.to_matrix();
-		return *this;
-	}
+    inline transform& set_rotation(const quaternionf& q) {
+        _local = q.to_matrix().unproject(1);
+        return *this;
+    }
 
-	inline transform& scale(const real_t &sx, const real_t &sy, const real_t &sz) {
-		_local(0,0) *= sx; _local(1,1) *= sy; _local(2,2) *= sz;
-		return *this;
-	}
+    inline transform& scale(const float& sx, const float& sy, const float& sz) {
+        _local[0][0] *= sx;
+        _local[1][1] *= sy;
+        _local[2][2] *= sz;
+        return *this;
+    }
 
-	inline transform& set_scale(const real_t &sx, const real_t &sy, const real_t &sz) {
-		_local(0,0) = sx; _local(1,1) = sy; _local(2,2) = sz;
-		return *this;
-	}
+    inline transform& set_scale(const float& sx, const float& sy,
+                                const float& sz) {
+        _local[0][0] = sx;
+        _local[1][1] = sy;
+        _local[2][2] = sz;
+        return *this;
+    }
 
-	inline transform& scale(const real_t& uniform_scale) {
-		return scale(uniform_scale,uniform_scale,uniform_scale);
-	}
+    inline transform& scale(const float& uniform_scale) {
+        return scale(uniform_scale, uniform_scale, uniform_scale);
+    }
 
-    matrix4x4 _local = matrix4x4::identity();
-    matrix4x4 _global = matrix4x4::identity();
+    matrix4x4f _local  = matrix4x4f::identity();
+    matrix4x4f _global = matrix4x4f::identity();
 };
 
-}
+} // namespace pw
 
 #endif

src/scene/include/pw/scene/entity.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 1999-2020 Hartmut Seichter
+ * Copyright (C) 1999-2024 Hartmut Seichter
  *
  * Redistribution and use in source and binary forms, with or without
  * modification, are permitted provided that the following conditions
@@ -150,6 +150,6 @@ private:
 
 #include <pw/scene/components/relationship.hpp>
 #include <pw/scene/components/transform.hpp>
-#include <pw/scene/components/camera.hpp>
+// #include <pw/scene/components/camera.hpp>
 
 #endif

src/scene/include/pw/scene/scene.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 1999-2017 Hartmut Seichter
+ * Copyright (C) 1999-2024 Hartmut Seichter
  *
  * Redistribution and use in source and binary forms, with or without
  * modification, are permitted provided that the following conditions
@@ -33,9 +33,7 @@ namespace pw {
 
 class entity;
 
-class scene {
-public:
-
+struct scene final {
 
     using registry = entt::registry;
 

src/scene/src/entity.cpp

@@ -7,7 +7,6 @@ namespace pw {
 
 namespace detail {
 
-
 // entity identiy is hidden
 
 struct identity {
@@ -15,25 +14,23 @@ struct identity {
     identity(const std::string& name) : name(name) {}
     identity(const identity&) = default;
 
-    bool enabled = true;
-    std::string name = "";
-    std::vector<u_int32_t> tags = {};
+    bool enabled                 = true;
+    std::string name             = "";
+    std::vector<u_int32_t> tags  = {};
     std::vector<uint32_t> layers = {};
 };
 
-}
+} // namespace detail
 
-entity::entity(scene &s,const std::string& name)
-    : _registry(s._registry)
-    , _entity(s._registry ? _registry->create() : entt::null)
-{
+entity::entity(scene& s, const std::string& name)
+    : _registry(s._registry),
+      _entity(s._registry ? _registry->create() : entt::null) {
     if (s._registry) {
         this->add_component<detail::identity>(name);
     }
 }
 
-bool entity::add_child(entity &child)
-{
+bool entity::add_child(entity& child) {
     // not mixing scenes (yet)
     if (child._registry != this->_registry) {
         debug::e() << "Cannot mix entities from different scenes";
@@ -47,14 +44,13 @@ bool entity::add_child(entity &child)
     c.entities.push_back(child._entity);
 
     // declare parent
-    auto& p_r = child.get_or_create_component<parent>();
+    auto& p_r  = child.get_or_create_component<parent>();
     p_r.entity = _entity;
 
     return true;
 }
 
-bool entity::remove_child(entity& child)
-{
+bool entity::remove_child(entity& child) {
     // not mixing scenes (yet)
     if (child._registry != this->_registry) {
         debug::e() << "Cannot mix entities from different scenes";
@@ -62,19 +58,17 @@ bool entity::remove_child(entity& child)
     }
 
     // both need to have a relationship component
-    if (child.has_component<parent>() &&
-            this->has_component<children>())
-    {
+    if (child.has_component<parent>() && this->has_component<children>()) {
         // we need to check if the child is related to the parent
         auto r_p = child.get_component<parent>();
-        if (r_p.entity == _entity)
-        {
+        if (r_p.entity == _entity) {
             // now go ahead delete parent
             r_p.entity = entt::null;
 
             // now remove all children
             auto& c = this->get_component<children>();
-            c.entities.erase(std::remove(c.entities.begin(),c.entities.end(),child._entity));
+            c.entities.erase(std::remove(c.entities.begin(), c.entities.end(),
+                                         child._entity));
 
             // flag success
             return true;
@@ -83,38 +77,24 @@ bool entity::remove_child(entity& child)
     return false;
 }
 
-
-size_t entity::child_count() const
-{
-    return this->has_component<children>() ?
-                get_component<children>().entities.size() :
-                0;
+size_t entity::child_count() const {
+    return this->has_component<children>()
+               ? get_component<children>().entities.size()
+               : 0;
 }
 
-
-void entity::set_name(const std::string& n)
-{
-    auto &ident = get_or_create_component<detail::identity>();
-    ident.name = n;
+void entity::set_name(const std::string& n) {
+    auto& ident = get_or_create_component<detail::identity>();
+    ident.name  = n;
 }
 
-std::string entity::name()
-{
-    auto &ident = get_or_create_component<detail::identity>();
+std::string entity::name() {
+    auto& ident = get_or_create_component<detail::identity>();
     return ident.name;
 }
 
+entity::~entity() { this->destroy(); }
 
+void entity::destroy() { _registry->destroy(_entity); }
 
-entity::~entity()
-{
-    this->destroy();
-}
-
-void entity::destroy()
-{
-    _registry->destroy(_entity);
-}
-
-}
-
+} // namespace pw

src/scene/src/scene.cpp

@@ -4,23 +4,11 @@
 #include "pw/core/debug.hpp"
 #include "pw/core/serialize.hpp"
 
-#include "pw/core/geometry.hpp"
+#include "pw/core/mesh.hpp"
 #include "pw/core/matrix_transform.hpp"
 
 namespace pw {
 
-
-struct mesh {
-    geometry geom;
-};
-
-struct renderer {
-    std::unique_ptr<int> state;
-//    std::vector<matrials> mat;
-};
-
-
-
 scene::scene()
     : _registry{ std::make_shared<entt::registry>()}
 {
@@ -28,8 +16,6 @@ scene::scene()
 
 // size_t scene::count_all_enties() const
 // {
-//     return _registry->storage().
-
 // size_t scene::count_alive_enties() const
 // {
 //     return _registry->storage().alive();
@@ -41,7 +27,7 @@ void scene::update_transforms()
 
     //auto [t,r] = view.get<transform,parent>(entity);
 
-#if 1
+#if 0
 
     // search for nodes with parent and transform (only find leaf nodes)
     auto view = _registry->view<transform,parent>(entt::exclude<children>);
@@ -63,7 +49,7 @@ void scene::update_transforms()
         debug::d() << "\tpath length " << path.size();
         std::reverse(path.begin(),path.end());
 
-        auto m = pw::matrix4x4::identity();
+        auto m = pw::matrix4x4f::identity();
 
         for (auto& transform : path) {
 

src/system/include/pw/system/input.hpp

@@ -1,54 +1,46 @@
 #ifndef PW_SYSTEM_INPUT_HPP
 #define PW_SYSTEM_INPUT_HPP
 
+#include <cstdint>
 #include <pw/core/globals.hpp>
 #include <pw/core/point.hpp>
 
 namespace pw {
 
-class input {
-public:
+struct input final {
 
     static input& get();
 
-    point mouse_position() const { return _mouse_position; }
+    point<int32_t> mouse_position() const { return _mouse_position; }
 
     bool mouse_pressed() const { return _mouse_pressed; }
     int mouse_button() const { return _mouse_button; }
 
-
-    bool has_input() const { return !_input_string.empty();}
+    bool has_input() const { return !_input_string.empty(); }
     std::string input_string() const { return _input_string; }
 
-	~input() = default;
+    ~input() = default;
 
-	enum mouse_button_state {
-		pressed,
-		released
-	};
+    enum mouse_button_state { pressed, released };
 
     void reset();
 
-protected:
-
+  protected:
     friend class window;
 
     input() = default;
 
-private:
-
-    point _mouse_position;
-	int _mouse_button;
+  private:
+    point<int32_t> _mouse_position;
+    int _mouse_button;
     bool _mouse_pressed;
 
     int _key_code;
     bool _key_pressed;
 
-	std::string _input_string;
-
-
+    std::string _input_string;
 };
 
-}
+} // namespace pw
 
 #endif

src/system/include/pw/system/path.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -27,38 +27,33 @@
 
 namespace pw {
 
-
-class path {
-public:
-
+struct path final {
     using path_list = std::vector<std::string>;
 
-	static path& get();
-	~path();
+    static path& get();
+    ~path();
 
-	std::string separator() const;
+    std::string separator() const;
 
-	std::string executable_path() const;
+    std::string executable_path() const;
 
-	path_list resource_paths() const { return _resource_paths; }
+    path_list resource_paths() const { return _resource_paths; }
     void set_resource_paths(path_list pl);
 
-    std::string find_file(const std::string &filename) const;
+    std::string find_file(const std::string& filename) const;
 
-	std::string get_filename(const std::string &filepath, bool with_extension = true) const;
+    std::string get_filename(const std::string& filepath,
+                             bool with_extension = true) const;
 
-protected:
-
-	path_list _resource_paths;
-
-	path();
-
-	struct impl;
-	std::unique_ptr<impl> _impl;
+  private:
+    path_list _resource_paths;
 
+    path();
 
+    struct impl;
+    std::unique_ptr<impl> _impl;
 };
 
-}
+} // namespace pw
 
 #endif

src/system/include/pw/system/window.hpp

@@ -1,33 +1,33 @@
 #ifndef PW_SYSTEM_WINDOW_HPP
 #define PW_SYSTEM_WINDOW_HPP
 
+#include <cstdint>
 #include <pw/core/globals.hpp>
-#include <pw/core/size.hpp>
 #include <pw/core/point.hpp>
+#include <pw/core/size.hpp>
 
 #include <functional>
 
 namespace pw {
 
-class window {
-public:
+struct window final {
 
-    using on_update_t = std::function<void(window&)> ;
-    using on_resize_t = std::function<void(window&)> ;
+    using on_update_t = std::function<void(window&)>;
+    using on_resize_t = std::function<void(window&)>;
 
     window();
     ~window();
 
     bool update();
 
-	void set_title(const std::string& title);
+    void set_title(const std::string& title);
 
-	void set_size(const size& s);
-    pw::size size() const;
-    pw::size client_size() const;
+    void set_size(const size<int32_t>& s);
+    pw::size<int32_t> size() const;
+    pw::size<int32_t> client_size() const;
 
-	void set_position(const point& p);
-	point position() const;
+    void set_position(const point<int32_t>& p);
+    point<int32_t> position() const;
 
     typedef void drop_callback;
 
@@ -39,17 +39,15 @@ public:
 
     bool visible() const;
     void set_visible(bool is_visible);
-protected:
 
-	struct impl;
-	std::unique_ptr<impl> _impl;
+  protected:
+    struct impl;
+    std::unique_ptr<impl> _impl;
 
     on_update_t _on_update;
     on_resize_t _on_resize;
-
-
 };
 
-}
+} // namespace pw
 
 #endif

src/system/src/window.cpp

@@ -1,99 +1,100 @@
 #include "pw/system/window.hpp"
 
+// clang-format off
 #include "glad/glad.h"
 #include "GLFW/glfw3.h"
+// clang-format on
 
-#include "pw/visual/context.hpp"
-#include "pw/system/input.hpp"
 #include "pw/system/display.hpp"
+#include "pw/system/input.hpp"
+#include "pw/visual/context.hpp"
 
 #include "pw/core/debug.hpp"
 
 #include <cmath>
-#include <locale>
 #include <codecvt>
-
-#include <iostream>
+#include <cstdint>
+#include <locale>
 
 namespace pw {
 
-//struct window_context : context
+// struct window_context : context
 //{
-//    virtual bool make_current() override;
-//    virtual void resize() override;
-//    //	virtual context::size size() override;
-//    virtual void flush() override;
-//};
+//     virtual bool make_current() override;
+//     virtual void resize() override;
+//     //	virtual context::size size() override;
+//     virtual void flush() override;
+// };
 
-//bool window_context::make_current()
+// bool window_context::make_current()
 //{
-//}
+// }
 
-//void window_context::resize()
+// void window_context::resize()
 //{
-//}
+// }
 
-//void window_context::flush()
+// void window_context::flush()
 //{
-//}
+// }
 
 struct window::impl {
 
     window& _parent;
 
-    GLFWwindow *_window = nullptr;
+    GLFWwindow* _window = nullptr;
 
-    ::pw::size _old_size;
-    pointi _old_pos;
+    ::pw::size<int32_t> _old_size;
+    point<int32_t> _old_pos;
 
     //	window_context _context;
 
-    static void error_callback(int error, const char* description)
-    {
-        debug::e() <<  "GLFW error: " <<  description;
+    static void error_callback(int error, const char* description) {
+        debug::e() << "GLFW error: " << description;
     }
 
-    static void drop_callback(GLFWwindow* window, int count, const char** paths)
-    {
+    static void drop_callback(GLFWwindow* window, int count,
+                              const char** paths) {
         //        std::cout << __FUNCTION__ << std::endl;
         //		for (int i = 0; i < count; i++)
         //			std::cout << "\t" << paths[i] << std::endl;
     }
 
-    static void scroll_callback(GLFWwindow* window, double xoffset, double yoffset)
-    {
-        std::cout << __FUNCTION__ << std::endl;
+    static void scroll_callback(GLFWwindow* window, double xoffset,
+                                double yoffset) {
+        // std::cout << __FUNCTION__ << std::endl;
     }
 
-    static void mouse_button_callback(GLFWwindow* window, int button, int action, int mods)
-    {
+    static void mouse_button_callback(GLFWwindow* window, int button,
+                                      int action, int mods) {
         input::get()._mouse_button = button;
 
-        std::cout << __FUNCTION__ << " " << button << " " << action << " " << mods << std::endl;
+        // std::cout << __FUNCTION__ << " " << button << " " << action << " "
+        //           << mods << std::endl;
         //        input::get()._mouse_position
     }
 
-    static void cursor_pos_callback(GLFWwindow* window, double xpos, double ypos)
-    {
-        input::get()._mouse_position = pointd(xpos,ypos).cast<float>();
+    static void cursor_pos_callback(GLFWwindow* window, double xpos,
+                                    double ypos) {
+        input::get()._mouse_position = vector{xpos, ypos}.cast<int32_t>();
     }
 
-
-    static void key_callback(GLFWwindow *window,int key, int scancode, int action, int mods)
-    {
-        input::get()._key_code = scancode;
+    static void key_callback(GLFWwindow* window, int key, int scancode,
+                             int action, int mods) {
+        input::get()._key_code    = scancode;
         input::get()._key_pressed = action;
         // action 0,1,2
         //        std::cout << __FUNCTION__ << action <<  std::endl;
     }
 
-    //    static void character_callback(GLFWwindow* window, unsigned int codepoint)
+    //    static void character_callback(GLFWwindow* window, unsigned int
+    //    codepoint)
     //    {
     //        std::cout << __FUNCTION__ << std::endl;
     //    }
 
-    static void charmods_callback(GLFWwindow* window, unsigned int codepoint, int mods)
-    {
+    static void charmods_callback(GLFWwindow* window, unsigned int codepoint,
+                                  int mods) {
         // build the string from a Unicode code point
         std::wstring_convert<std::codecvt_utf8<char32_t>, char32_t> converter;
         std::string u8str = converter.to_bytes(codepoint);
@@ -101,28 +102,30 @@ struct window::impl {
         input::get()._input_string = u8str;
     }
 
-    static void framebuffer_size_callback(GLFWwindow* window, int width, int height)
-    {
-        window::impl* impl = static_cast<window::impl*>(glfwGetWindowUserPointer(window));
+    static void framebuffer_size_callback(GLFWwindow* window, int width,
+                                          int height) {
+        window::impl* impl =
+            static_cast<window::impl*>(glfwGetWindowUserPointer(window));
         impl->_parent._on_resize(impl->_parent);
 
-//        debug::d() << "window::frame_buffer_size_callback " << width << "x" << height;
+        //        debug::d() << "window::frame_buffer_size_callback " << width
+        //        << "x" << height;
 
         //        impl->on_resize(width,height);
 
-        //		std::cout << "framebuffer " << width << "x" << height << std::endl;
-//            glViewport(0, 0, width, height);
+        //		std::cout << "framebuffer " << width << "x" << height <<
+        // std::endl;
+        //            glViewport(0, 0, width, height);
     }
 
-    void update_display_list()
-    {
+    void update_display_list() {
         display::_displays.clear();
 
         // fetch all monitors
-        int monitor_count = 0;
+        int monitor_count      = 0;
         GLFWmonitor** monitors = glfwGetMonitors(&monitor_count);
 
-        for (int i = 0; i < monitor_count;i++) {
+        for (int i = 0; i < monitor_count; i++) {
             display d;
             d._name = std::string(glfwGetMonitorName(monitors[i]));
 
@@ -137,24 +140,21 @@ struct window::impl {
         //		glfwGetMonitorPos(
         //		glfwGetMonitorPhysicalSize(
         //		glfwGetMonitorName();
-
     }
 
-    impl(window& w)
-        : _parent(w)
-    {
+    impl(window& w) : _parent(w) {
         // initialize
-        if (!glfwInit())
-        {
+        if (!glfwInit()) {
             debug::e() << "Initalization error";
         }
 
-        int glfw_major, glfx_minor,glfw_rev;
-        glfwGetVersion(&glfw_major,&glfx_minor,&glfw_rev);
+        int glfw_major, glfx_minor, glfw_rev;
+        glfwGetVersion(&glfw_major, &glfx_minor, &glfw_rev);
 
-
-//        debug::d() << "GLFW " << glfw_major << "." << glfx_minor << "." << glfw_rev;
-//        debug::d() << "GLFW header " << GLFW_VERSION_MAJOR << "." << GLFW_VERSION_MINOR << "." << GLFW_VERSION_REVISION;
+        //        debug::d() << "GLFW " << glfw_major << "." << glfx_minor <<
+        //        "." << glfw_rev; debug::d() << "GLFW header " <<
+        //        GLFW_VERSION_MAJOR << "." << GLFW_VERSION_MINOR << "." <<
+        //        GLFW_VERSION_REVISION;
 
         update_display_list();
 
@@ -175,8 +175,7 @@ struct window::impl {
         glfwMakeContextCurrent(_window);
 
         // load opengl
-        if (!gladLoadGLLoader((GLADloadproc)glfwGetProcAddress))
-        {
+        if (!gladLoadGLLoader((GLADloadproc)glfwGetProcAddress)) {
             debug::e() << "glad couldn't get OpenGL API";
         }
 
@@ -184,14 +183,16 @@ struct window::impl {
         int major, minor, rev;
         major = glfwGetWindowAttrib(_window, GLFW_CONTEXT_VERSION_MAJOR);
         minor = glfwGetWindowAttrib(_window, GLFW_CONTEXT_VERSION_MINOR);
-        rev = glfwGetWindowAttrib(_window, GLFW_CONTEXT_REVISION);
+        rev   = glfwGetWindowAttrib(_window, GLFW_CONTEXT_REVISION);
 
         // maybe something to pass to the outside
-//        debug::d() << "OpenGL " << major << "." << minor << "." << rev;
+        //        debug::d() << "OpenGL " << major << "." << minor << "." <<
+        //        rev;
 
-        glfwSetWindowUserPointer(_window,this);
+        glfwSetWindowUserPointer(_window, this);
 
-        glfwSetFramebufferSizeCallback(_window, window::impl::framebuffer_size_callback);
+        glfwSetFramebufferSizeCallback(_window,
+                                       window::impl::framebuffer_size_callback);
         glfwSetKeyCallback(_window, window::impl::key_callback);
         //        glfwSetCharCallback(_window, character_callback);
         glfwSetCharModsCallback(_window, charmods_callback);
@@ -203,22 +204,17 @@ struct window::impl {
         glfwSetMouseButtonCallback(_window, mouse_button_callback);
         glfwSetScrollCallback(_window, scroll_callback);
 
-
         glfwSetErrorCallback(error_callback);
 
-        //glfwSetWindowCloseCallback(_window,close_callback);
+        // glfwSetWindowCloseCallback(_window,close_callback);
     }
 
-    ~impl()
-    {
-        glfwDestroyWindow(_window);
-    }
+    ~impl() { glfwDestroyWindow(_window); }
 
-    bool update()
-    {
-        if (_window && !glfwWindowShouldClose(_window))
-        {
-            // TODO lock an unlock the current input system to allow for late events coming in
+    bool update() {
+        if (_window && !glfwWindowShouldClose(_window)) {
+            // TODO lock an unlock the current input system to allow for late
+            // events coming in
             input::get().reset();
 
             // get new events
@@ -234,70 +230,60 @@ struct window::impl {
         return false;
     }
 
-    void set_title(const std::string& title)
-    {
-        glfwSetWindowTitle(_window,title.c_str());
+    void set_title(const std::string& title) {
+        glfwSetWindowTitle(_window, title.c_str());
     }
 
-    void set_size(const ::pw::size& s)
-    {
-        glfwSetWindowSize(_window,s.width,s.height);
+    void set_size(const ::pw::size<int32_t>& s) {
+        glfwSetWindowSize(_window, s.width, s.height);
     }
 
-    ::pw::size size() const
-    {
-        int w,h;
-        glfwGetWindowSize(_window,&w,&h);
-        return ::pw::size(w,h);
+    ::pw::size<int32_t> size() const {
+        int w, h;
+        glfwGetWindowSize(_window, &w, &h);
+        return ::pw::size<int32_t>{w, h};
     }
 
-
-    ::pw::size client_size() const
-    {
-        int w,h;
-        glfwGetFramebufferSize(_window,&w,&h);
-        return ::pw::size(w,h);
+    ::pw::size<int32_t> client_size() const {
+        int w, h;
+        glfwGetFramebufferSize(_window, &w, &h);
+        return ::pw::size<int32_t>{w, h};
     }
 
-    ::pw::point position() const
-    {
-        int x,y;
-        glfwGetWindowPos(_window,&x,&y);
-        return ::pw::point(x,y);
+    ::pw::point<int32_t> position() const {
+        int x, y;
+        glfwGetWindowPos(_window, &x, &y);
+        return {x, y};
     }
 
-    void set_position(const pointi& p)
-    {
-        glfwSetWindowPos(_window,p.x,p.y);
+    void set_position(const point<int32_t>& p) {
+        glfwSetWindowPos(_window, p.x(), p.y());
     }
 
-    void set_fullscreen(bool use_fullscreen)
-    {
+    void set_fullscreen(bool use_fullscreen) {
         if (fullscreen() == use_fullscreen)
             return;
 
-        if (use_fullscreen)
-        {
-            glfwGetWindowPos( _window, &_old_pos.x, &_old_pos.y );
-            glfwGetWindowSize( _window, &_old_size.width, &_old_size.height );
+        if (use_fullscreen) {
+            glfwGetWindowPos(_window, &_old_pos.x(), &_old_pos.y());
+            glfwGetWindowSize(_window, &_old_size.width, &_old_size.height);
 
-            GLFWmonitor * monitor = glfwGetPrimaryMonitor();
-            const GLFWvidmode * mode = glfwGetVideoMode(monitor);
+            GLFWmonitor* monitor    = glfwGetPrimaryMonitor();
+            const GLFWvidmode* mode = glfwGetVideoMode(monitor);
 
-            glfwSetWindowMonitor( _window, monitor, 0, 0, mode->width, mode->height, 0 );
+            glfwSetWindowMonitor(_window, monitor, 0, 0, mode->width,
+                                 mode->height, 0);
 
-        } else
-        {
+        } else {
 
-            glfwSetWindowMonitor( _window, nullptr, _old_pos.x,_old_pos.y, _old_size.width,_old_size.height,0);
+            glfwSetWindowMonitor(_window, nullptr, _old_pos.x(), _old_pos.y(),
+                                 _old_size.width, _old_size.height, 0);
         }
 
         //		glfwSetWindow
     }
 
-    bool fullscreen() const {
-        return glfwGetWindowMonitor(_window) != nullptr;
-    }
+    bool fullscreen() const { return glfwGetWindowMonitor(_window) != nullptr; }
 
     void set_visible(bool show) {
         (show) ? glfwShowWindow(_window) : glfwHideWindow(_window);
@@ -308,77 +294,39 @@ struct window::impl {
     }
 };
 
-
 //
 //
 //
 window::window()
-    : _impl(std::make_unique<window::impl>(*this))
-    , _on_update([](window&){})
-    , _on_resize([](window&){})
-{
-}
+    : _impl(std::make_unique<window::impl>(*this)), _on_update([](window&) {}),
+      _on_resize([](window&) {}) {}
 
-window::~window()
-{
-}
+window::~window() {}
 
-bool window::update()
-{
-    return _impl->update();
-}
+bool window::update() { return _impl->update(); }
 
-void window::set_title(const std::string& title)
-{
-    _impl->set_title(title);
-}
+void window::set_title(const std::string& title) { _impl->set_title(title); }
 
-void window::set_size(const ::pw::size& s)
-{
-    _impl->set_size(s.cast<int>());
-}
+void window::set_size(const ::pw::size<int32_t>& s) { _impl->set_size(s); }
 
-size window::size() const
-{
-    return _impl->size();
-}
+size<int32_t> window::size() const { return _impl->size(); }
 
-size window::client_size() const
-{
-    return _impl->client_size();
-}
+size<int32_t> window::client_size() const { return _impl->client_size(); }
 
-void window::set_position(const point &p)
-{
+void window::set_position(const point<int32_t>& p) {
     _impl->set_position(p.cast<int>());
 }
 
-point window::position() const
-{
-    return _impl->position();
-}
+point<int32_t> window::position() const { return _impl->position(); }
 
-bool window::fullscreen() const
-{
-    return _impl->fullscreen();
-}
+bool window::fullscreen() const { return _impl->fullscreen(); }
 
-void window::set_fullscreen(bool use_fullscreen)
-{
+void window::set_fullscreen(bool use_fullscreen) {
     _impl->set_fullscreen(use_fullscreen);
 }
 
-bool window::visible() const
-{
-    return _impl->visible();
-}
+bool window::visible() const { return _impl->visible(); }
 
-void window::set_visible(bool is_visible)
-{
-    _impl->set_visible(is_visible);
-}
+void window::set_visible(bool is_visible) { _impl->set_visible(is_visible); }
 
-
-
-
-}
+} // namespace pw

src/visual/CMakeLists.txt

@@ -38,4 +38,4 @@ target_include_directories(
     include
     )
 
-target_link_libraries(pwvisual pwscene glad)
+target_link_libraries(pwvisual pwcore pwscene glad)

src/visual/include/pw/visual/context.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999-2021 Hartmut Seichter
+ * Copyright (c) 1999-2024 Hartmut Seichter
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
@@ -8,8 +8,8 @@
  * 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 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,
@@ -24,17 +24,15 @@
 #ifndef PW_VISUAL_CONTEXT_HPP
 #define PW_VISUAL_CONTEXT_HPP
 
+#include <pw/core/color.hpp>
 #include <pw/core/point.hpp>
+#include <pw/core/rectangle.hpp>
 #include <pw/core/size.hpp>
 #include <pw/core/vector.hpp>
-#include <pw/core/rectangle.hpp>
-#include <pw/core/color.hpp>
-
 
 namespace pw {
 
-class context {
-public:
+struct context final {
 
     context();
     ~context();
@@ -42,24 +40,24 @@ public:
     void set_blend();
     void set_depth();
 
-    void set_viewport(const rectangle& v);
-    rectangle viewport() const;
+    void set_viewport(const rectangle<int32_t>& v);
+    rectangle<int32_t> viewport() const;
 
-    size viewport_size() const;
-    point viewport_origin() const;
+    size<int32_t> viewport_size() const;
+    point<int32_t> viewport_origin() const;
 
-    void set_clearcolor(const color &c);
+    void set_clearcolor(const color& c);
     color clearcolor() const;
 
     void clear();
 
     uint32_t get_error() const;
 
-protected:
+  protected:
     struct impl;
     std::unique_ptr<impl> _impl;
 };
 
-}
+} // namespace pw
 
 #endif

src/visual/include/pw/visual/framebuffer.hpp

@@ -6,22 +6,20 @@
 
 namespace pw {
 
-class framebuffer {
-public:
+struct framebuffer final {
     framebuffer();
     ~framebuffer();
 
-    bool create(const size& s);
+    bool create(const size<int32_t>& s);
 
     void bind();
     void blit();
     void unbind();
 
-protected:
-
+  protected:
     struct impl;
     std::unique_ptr<impl> _impl;
 };
 
-}
+} // namespace pw
 #endif

src/visual/include/pw/visual/pipeline.hpp

@@ -1,52 +1,42 @@
 #ifndef PW_VISUAL_PIPELINE_HPP
 #define PW_VISUAL_PIPELINE_HPP
 
-#include <pw/core/size.hpp>
-#include <pw/core/matrix.hpp>
-#include <pw/core/geometry.hpp>
 #include <pw/core/color.hpp>
+#include <pw/core/matrix.hpp>
+#include <pw/core/primitives.hpp>
+#include <pw/core/size.hpp>
 
-#include <pw/visual/shader.hpp>
 #include <pw/visual/renderer.hpp>
-
+#include <pw/visual/shader.hpp>
 
 #include <map>
 
 namespace pw {
 
-
-
-
 class pipeline {
 
-public:
-	pipeline();
+  public:
+    pipeline();
     ~pipeline();
 
     void draw();
 
-	bool create(size s);
-
-protected:
-	struct impl;
-	std::unique_ptr<impl> _impl;
+    bool create(size s);
 
+  protected:
+    struct impl;
+    std::unique_ptr<impl> _impl;
 };
 
-
-
 struct render_pass {
 
-    void submit(const geometry& g,
-                const matrix4x4& model_mat,
-                const matrix4x4& view_mat,
-                const matrix4x4& projection_mat);
+    void submit(const geometry& g, const matrix4x4& model_mat,
+                const matrix4x4& view_mat, const matrix4x4& projection_mat);
 
     shader _shader;
     renderer _renderer;
 };
 
-
-}
+} // namespace pw
 
 #endif

src/visual/include/pw/visual/renderer.hpp

@@ -1,39 +1,33 @@
 #ifndef PW_VISUAL_MESH_RENDERER_HPP
 #define PW_VISUAL_MESH_RENDERER_HPP
 
-#include <memory>
 #include <map>
+#include <memory>
 
 namespace pw {
 
-class geometry;
+struct mesh;
 
 /**
- * @brief builds a renderer for geometry
+ * @brief builds a renderer for meshes
  */
-class renderer {
-public:
-
+struct renderer final {
     renderer();
-    renderer(const geometry& m);
+    renderer(const mesh& m);
 
     ~renderer();
 
-    bool update(const geometry &m);
-	void release();
+    bool update(const mesh& m);
+    void release();
 
-	void draw();
-	bool ready() const;
-
-    uint64_t change_count() const;
-    void set_change_count(uint64_t n);
-protected:
-
-	struct impl;
-	std::unique_ptr<impl> _impl;
+    void draw();
+    bool ready() const;
 
+  protected:
+    struct impl;
+    std::unique_ptr<impl> _impl;
 };
 
-}
+} // namespace pw
 
 #endif

src/visual/include/pw/visual/shader.hpp

@@ -1,11 +1,10 @@
 #ifndef PW_VISUAL_SHADER_HPP
 #define PW_VISUAL_SHADER_HPP
 
+#include <pw/core/debug.hpp>
 #include <pw/core/globals.hpp>
 #include <pw/core/matrix.hpp>
 #include <pw/core/vector.hpp>
-#include <pw/core/debug.hpp>
-
 #include <pw/visual/texture.hpp>
 
 #include <map>
@@ -13,77 +12,72 @@
 
 namespace pw {
 
-class shader {
-public:
-
-	shader();
+struct shader final {
+    shader();
     ~shader();
 
-	enum class code_type {
-		vertex,
-		fragment,
-		geometry,
-		compute
-	};
+    enum class code_type { vertex, fragment, geometry, compute };
 
     void set_source(code_type t, const std::string& c) { _source[t] = c; }
-	std::string source(code_type t) { return _source[t]; }
+    std::string source(code_type t) { return _source[t]; }
 
-	bool ready() const;
+    bool ready() const;
 
-    shader& set_uniform_at_location(int location,float v);
+    using uniform_t = std::variant<bool, int, float, double, vector2f, vector3f,
+                                   vector4f, matrix4x4f>;
+    using uniform_entry_t = std::tuple<std::string, uniform_t, int>;
+
+
+    shader& set_uniform_at_location(int location, float v);
     shader& set_uniform_at_location(int location, uint32_t v);
-    shader &set_uniform_at_location(int location, int32_t v);
-    shader& set_uniform_at_location(int location,matrix4x4f const &v);
-    shader& set_uniform_at_location(int location,vector4f const &v);
-    shader& set_uniform_at_location(int location,texture const &t);
+    shader& set_uniform_at_location(int location, int32_t v);
+    shader& set_uniform_at_location(int location, matrix4x4f const& v);
+    shader& set_uniform_at_location(int location, vector4f const& v);
+    shader& set_uniform_at_location(int location, texture const& t);
 
     /**
      * @brief retrieves the position of a uniform
      * @param name of the uniform
      * @return position of the uniform or negative if it doesn't exist
      */
-	int uniform_location(std::string const & name) const;
+    int uniform_location(std::string const& name) const;
 
     /**
      * @brief check if a uniform with the given name exists
      * @param name of the uniform
      * @return true if found
      */
-    bool has_uniform(std::string const &name) const { return uniform_location(name) >= 0; }
+    bool has_uniform(std::string const& name) const {
+        return uniform_location(name) >= 0;
+    }
 
     /**
      * sets data of the
      */
-	template<typename T>
-    shader & set_uniform(std::string const & name, T &&value)
-	{    
-        return set_uniform_at_location( uniform_location(name), std::forward<T>(value) );
-	}
+    template <typename T>
+    shader& set_uniform(std::string const& name, T&& value) {
+        return set_uniform_at_location(uniform_location(name),
+                                       std::forward<T>(value));
+    }
 
-	bool build();
+    bool build();
 
     void use();
 
-	using uniform_t = std::variant<bool,int,float,double,vector2f,vector3f,vector4f,matrix4x4f>;
-	using uniform_entry_t = std::tuple<std::string,uniform_t,int>;
 
-	using uniform_cache_t = std::vector<uniform_entry_t>;
+    using uniform_cache_t = std::vector<uniform_entry_t>;
 
-	void set_uniforms(uniform_cache_t c);
+    void set_uniforms(uniform_cache_t c);
 
     uint32_t native_handle() const;
 
-protected:
-
-    std::unordered_map<code_type,std::string> _source;
-
-	struct impl;
-	std::unique_ptr<impl> _impl;
+  protected:
+    std::unordered_map<code_type, std::string> _source;
 
+    struct impl;
+    std::unique_ptr<impl> _impl;
 };
 
-
-}
+} // namespace pw
 
 #endif

src/visual/include/pw/visual/texture.hpp

@@ -1,70 +1,55 @@
 #ifndef PW_VISUAL_TEXTURE_HPP
 #define PW_VISUAL_TEXTURE_HPP
 
+#include <pw/core/debug.hpp>
 #include <pw/core/globals.hpp>
 #include <pw/core/image.hpp>
-#include <pw/core/debug.hpp>
 
 namespace pw {
 
 class texture {
-public:
-	enum class data_type {
-		color,
-		normal
-	};
+  public:
+    enum class data_type { color, normal };
 
-	enum class data_layout {
-		shape_1d,
-		shape_2d,
-		shape_3d
-	};
+    enum class data_layout { shape_1d, shape_2d, shape_3d };
 
-	enum texture_dimension {
-        dimension_r,
-		dimension_s,
-        dimension_t
-	};
+    enum texture_dimension { dimension_r, dimension_s, dimension_t };
 
-	enum wrap_mode {
-		wrap_repeat,
-		wrap_clamp,
-		wrap_clamp_to_edge,
-		wrap_clamp_to_repeat
-	};
+    enum wrap_mode {
+        wrap_repeat,
+        wrap_clamp,
+        wrap_clamp_to_edge,
+        wrap_clamp_to_repeat
+    };
 
     texture();
-    texture(const image& i,data_layout s);
+    texture(const image& i, data_layout s);
     ~texture();
 
     void set_shape(data_layout s);
-	data_layout shape() const { return _shape; }
+    data_layout shape() const { return _shape; }
 
-	void set_wrap(wrap_mode w);
-	wrap_mode wrap() const { return _wrap; }
+    void set_wrap(wrap_mode w);
+    wrap_mode wrap() const { return _wrap; }
 
     uint32_t native_handle() const;
 
-    bool create(const image &i);
-    void update(const image &i);
+    bool create(const image& i);
+    void update(const image& i);
 
     void bind();
     void unbind();
 
     bool valid() const;
 
-protected:
-
+  protected:
     data_layout _shape = data_layout::shape_2d;
-    wrap_mode _wrap = wrap_mode::wrap_clamp_to_edge;
-
-	struct impl;
-	unique_ptr<impl> _impl;
-
+    wrap_mode _wrap    = wrap_mode::wrap_clamp_to_edge;
 
+    struct impl;
+    unique_ptr<impl> _impl;
 };
 
-}
-
+} // namespace pw
 
 #endif

src/visual/src/context.cpp

@@ -2,91 +2,60 @@
 #include "pw/core/vector.hpp"
 
 #include "glad/glad.h"
-
+#include <cstdint>
 
 namespace pw {
 
 struct context::impl {
 
-    vector4f _clear_color { 0, 1, 0, 1};
+    vector4f _clear_color{0, 1, 0, 1};
 
-    void set_viewport(const rectangle& v)
-    {
-        glViewport(v.position.x,v.position.y,v.size.width,v.size.height);
+    void set_viewport(const rectangle<int32_t>& v) {
+        glViewport(v.position.x(), v.position.y(), v.size.width, v.size.height);
     }
 
-    const rectangle viewport()
-    {
-        float _viewport[4] = {0,0,1,1};
-        glGetFloatv(GL_VIEWPORT,&_viewport[0]);
-        return rectangle(_viewport);
+    const rectangle<int32_t> viewport() {
+        auto vp = rectangle<float>{0, 0, 1, 1};
+        glGetFloatv(GL_VIEWPORT, vp.position.data());
+        return vp.cast<int32_t>();
     }
 
-    void clear()
-    {
+    void clear() {
         glEnable(GL_CULL_FACE);
         glCullFace(GL_BACK);
         glFrontFace(GL_CCW);
 
-        glClearColor(_clear_color[0],_clear_color[1],_clear_color[2],_clear_color[3]);
+        glClearColor(_clear_color[0], _clear_color[1], _clear_color[2],
+                     _clear_color[3]);
         glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
-
-    }
-
-
-
-    uint32_t get_error() const
-    {
-        return glGetError();
     }
 
+    uint32_t get_error() const { return glGetError(); }
 };
 
-context::context()
-    : _impl(std::make_unique<impl>())
-{
-}
+context::context() : _impl(std::make_unique<impl>()) {}
 
-context::~context()
-{
-}
+context::~context() {}
 
-void context::set_blend()
-{
+void context::set_blend() {}
 
-}
-
-void context::set_viewport(const rectangle& v)
-{
+void context::set_viewport(const rectangle<int32_t>& v) {
     _impl->set_viewport(v);
 }
 
-rectangle context::viewport() const
-{
-    return _impl->viewport();
+rectangle<int32_t> context::viewport() const { return _impl->viewport(); }
+
+void context::clear() { _impl->clear(); }
+
+u_int32_t context::get_error() const { return _impl->get_error(); }
+
+color context::clearcolor() const {
+    // return _impl->_clear_color;
+    return color{};
 }
 
-void context::clear()
-{
-    _impl->clear();
+void context::set_clearcolor(const color& c) {
+    // _impl->_clear_color = c;
 }
 
-u_int32_t context::get_error() const
-{
-    return _impl->get_error();
-}
-
-color context::clearcolor() const
-{
-    return _impl->_clear_color;
-}
-
-void context::set_clearcolor(const color& c)
-{
-    _impl->_clear_color = c;
-}
-
-
-}
-
-
+} // namespace pw

src/visual/src/framebuffer.cpp

@@ -5,6 +5,7 @@
 
 #include "glad/glad.h"
 
+#include <cstdint>
 #include <limits>
 
 namespace pw {
@@ -13,7 +14,7 @@ namespace pw {
 
 struct framebuffer::impl {
 
-    size _size;
+    size<int32_t> _size;
 
     GLuint _fbo_draw { std::numeric_limits<GLuint>::max() };
     GLuint _fbo_msaa { std::numeric_limits<GLuint>::max() };
@@ -21,7 +22,7 @@ struct framebuffer::impl {
     GLuint _rbo_color { std::numeric_limits<GLuint>::max() };
     GLuint _rbo_depth { std::numeric_limits<GLuint>::max() };
 
-    bool create(const size &s)
+    bool create(const size<int32_t> &s)
     {
         // TODO  mak color resolution and
         _size = s;
@@ -107,7 +108,7 @@ framebuffer::~framebuffer()
 {
 }
 
-bool framebuffer::create(const size &s)
+bool framebuffer::create(const size<int32_t> &s)
 {
     return _impl->create(s);
 }

src/visual/src/pipeline.cpp

@@ -1,30 +1,26 @@
-#include "pw/core/size.hpp"
-#include "pw/core/matrix.hpp"
-#include "pw/core/geometry.hpp"
-#include "pw/core/time.hpp"
 #include "pw/core/axisangle.hpp"
-#include "pw/core/serialize.hpp"
+#include "pw/core/geometry.hpp"
+#include "pw/core/matrix.hpp"
 #include "pw/core/matrix_transform.hpp"
+#include "pw/core/serialize.hpp"
+#include "pw/core/size.hpp"
+#include "pw/core/time.hpp"
 
 #include "pw/core/debug.hpp"
-#include "pw/visual/pipeline.hpp"
-#include "pw/visual/shader.hpp"
-#include "pw/visual/renderer.hpp"
 #include "pw/visual/framebuffer.hpp"
+#include "pw/visual/pipeline.hpp"
+#include "pw/visual/renderer.hpp"
+#include "pw/visual/shader.hpp"
 #include "pw/visual/texture.hpp"
 
 #include "glad/glad.h"
 
 namespace pw {
 
-
-void render_pass::submit(const geometry &g,
-                         const matrix4x4 &model_mat,
-                         const matrix4x4 &view_mat,
-                         const matrix4x4 &projection_mat)
-{
-    if (!_shader.ready())
-    {     //
+void render_pass::submit(const geometry& g, const matrix4x4& model_mat,
+                         const matrix4x4& view_mat,
+                         const matrix4x4& projection_mat) {
+    if (!_shader.ready()) { //
         _shader.build();
     }
 
@@ -35,22 +31,18 @@ void render_pass::submit(const geometry &g,
 
     // new version with ...
     shader::uniform_cache_t us;
-//    us.push_back(std::make_tuple("input_color",mat._color,-1));
-    us.push_back(std::make_tuple("model",model_mat,-1));
-    us.push_back(std::make_tuple("view",view_mat,-1));
-    us.push_back(std::make_tuple("projection",projection_mat,-1));
+    //    us.push_back(std::make_tuple("input_color",mat._color,-1));
+    us.push_back(std::make_tuple("model", model_mat, -1));
+    us.push_back(std::make_tuple("view", view_mat, -1));
+    us.push_back(std::make_tuple("projection", projection_mat, -1));
 
     _shader.set_uniforms(us);
 
     _shader.use();
     _renderer.draw();
-
 }
 
-
-struct triangle_renderer
-{
-
+struct triangle_renderer {
 
     texture tex;
     shader shader_p;
@@ -58,15 +50,12 @@ struct triangle_renderer
     renderer r;
     time::tick_t tick;
 
-    triangle_renderer()
-    {
-    }
+    triangle_renderer() {}
 
-    void setup()
-    {
+    void setup() {
 
         const float z_val = -5.f;
-        const float s = 1.0f;
+        const float s     = 1.0f;
 
         //
         //		0 -- 1
@@ -75,23 +64,23 @@ struct triangle_renderer
 
         // geometry
         vector3_array vertices = {
-            {-s,  s,  z_val} // 0
-            ,{ s,  s,  z_val} // 1
-            ,{-s, -s,  z_val} // 2
-            ,{ s, -s,  z_val} // 3
+            {-s, s,  z_val}  // 0
+            ,
+            {s,  s,  z_val}  // 1
+            ,
+            {-s, -s, z_val}  // 2
+            ,
+            {s,  -s, z_val}  // 3
         };
 
         // topology / indices
-        geometry::indices_t indices = { 2, 1, 0,
-                                           2, 3, 1};
-
+        geometry::indices_t indices = {2, 1, 0, 2, 3, 1};
 
         amesh.set_indices(indices);
         amesh.set_vertices(vertices);
 
         amesh.compute_normals();
 
-
         const char* vertex_shader_2 = R"(
 									  #version 400
 									  uniform mat4 model;
@@ -106,7 +95,7 @@ struct triangle_renderer
 									  }
 									  )";
 
-        const char *fragment_shader_2 = R"(
+        const char* fragment_shader_2 = R"(
 										#version 400
 										uniform vec4 input_color = vec4(1.0, 0.0, 0.0, 1.0);
 										out vec4 frag_colour;
@@ -114,38 +103,37 @@ struct triangle_renderer
 										frag_colour = input_color;
 										})";
 
-        shader_p.set_source(shader::code_type::vertex,vertex_shader_2);
-        shader_p.set_source(shader::code_type::fragment,fragment_shader_2);
+        shader_p.set_source(shader::code_type::vertex, vertex_shader_2);
+        shader_p.set_source(shader::code_type::fragment, fragment_shader_2);
 
         if (!shader_p.build())
             exit(-1);
     }
 
-    void draw()
-    {
+    void draw() {
         if (!r.ready())
             r.create(amesh);
 
-
         // input needed here -
         // model view projection
 
         shader_p.use();
 
-        auto v_col = ping_pong(static_cast<float>(time::now()),1.0f);
+        auto v_col = ping_pong(static_cast<float>(time::now()), 1.0f);
 
-        vector4f col = {0.5f,1-v_col,v_col,1.0f};
+        vector4f col = {0.5f, 1 - v_col, v_col, 1.0f};
 
-        matrix4x4f model_mat; model_mat.set_identity();
+        matrix4x4f model_mat;
+        model_mat.set_identity();
 
-        auto v_angle = ping_pong(static_cast<float>(time::now()),2 * ::pw::pi<float>());
+        auto v_angle =
+            ping_pong(static_cast<float>(time::now()), 2 * ::pw::pi<float>());
 
-        axisangle rot(vector3::forward(),v_angle);
+        axisangle rot(vector3::forward(), v_angle);
         model_mat = rot.to_matrix();
 
-        matrix4x4f view_mat = matrix_transform<float>::look_at(vector3({0,0,0}),
-                                                               vector3::forward(),
-                                                               vector3::up());
+        matrix4x4f view_mat = matrix_transform<float>::look_at(
+            vector3({0, 0, 0}), vector3::forward(), vector3::up());
 
         // materials should carry this
 #if 1
@@ -154,35 +142,31 @@ struct triangle_renderer
         glFrontFace(GL_CCW);
 #endif
 
-
         // now bind textures
 
 #if 1
-        auto proj_mat = matrix_transform<float>::orthographic_projection(1.3,1.0,
-                                                                         0.2f,100.f);
-
+        auto proj_mat = matrix_transform<float>::orthographic_projection(
+            1.3, 1.0, 0.2f, 100.f);
 
 #else
-        auto proj_mat = matrix_transform<float>::perspective_projection(deg_to_rad(60.f),
-                                                                        1.3f,
-                                                                        0.2f,1000.f);
+        auto proj_mat = matrix_transform<float>::perspective_projection(
+            deg_to_rad(60.f), 1.3f, 0.2f, 1000.f);
 #endif
 
-
 #if 1
         // highly inefficient - should be cached -
-        shader_p.set_uniform("input_color",col);
-        shader_p.set_uniform("model",model_mat);
-        shader_p.set_uniform("view",view_mat);
-        shader_p.set_uniform("projection",proj_mat);
+        shader_p.set_uniform("input_color", col);
+        shader_p.set_uniform("model", model_mat);
+        shader_p.set_uniform("view", view_mat);
+        shader_p.set_uniform("projection", proj_mat);
 
 #else
         // new version with ...
         shader::uniform_cache_t us;
-        us.push_back(std::make_tuple("input_color",col,-1));
-        us.push_back(std::make_tuple("model",model_mat,-1));
-        us.push_back(std::make_tuple("view",view_mat,-1));
-        us.push_back(std::make_tuple("projection",proj_mat,-1));
+        us.push_back(std::make_tuple("input_color", col, -1));
+        us.push_back(std::make_tuple("model", model_mat, -1));
+        us.push_back(std::make_tuple("view", view_mat, -1));
+        us.push_back(std::make_tuple("projection", proj_mat, -1));
 
         shader_p.set_uniforms(us);
 
@@ -191,7 +175,7 @@ struct triangle_renderer
         r.draw();
 
         auto error = glGetError();
-        if (error != GL_NO_ERROR){
+        if (error != GL_NO_ERROR) {
             debug::e() << "GL error " << error;
         }
 
@@ -199,7 +183,6 @@ struct triangle_renderer
     }
 };
 
-
 struct pipeline::impl {
 
 #if 0
@@ -214,35 +197,30 @@ struct pipeline::impl {
 
     framebuffer fb;
 
-
-    //testing
+    // testing
     triangle_renderer tr;
 
     bool create(size s);
 
     void draw();
 
-
-
-    impl() = default;
+    impl()  = default;
     ~impl() = default;
-
 };
 
-GLuint generate_multisample_texture(const size& s,int samples)
-{
+GLuint generate_multisample_texture(const size& s, int samples) {
     GLuint texture;
     glGenTextures(1, &texture);
 
     glBindTexture(GL_TEXTURE_2D_MULTISAMPLE, texture);
-    glTexImage2DMultisample(GL_TEXTURE_2D_MULTISAMPLE, samples, GL_RGB8, s.width, s.height, GL_TRUE);
+    glTexImage2DMultisample(GL_TEXTURE_2D_MULTISAMPLE, samples, GL_RGB8,
+                            s.width, s.height, GL_TRUE);
     glBindTexture(GL_TEXTURE_2D_MULTISAMPLE, 0);
 
     return texture;
 }
 
-bool pipeline::impl::create(::pw::size s)
-{
+bool pipeline::impl::create(::pw::size s) {
 
 #if 0
     _size = s;
@@ -293,23 +271,14 @@ bool pipeline::impl::create(::pw::size s)
     glBindFramebuffer(GL_FRAMEBUFFER, 0);
 #endif
 
-
-
     tr.setup();
 
-
     return true;
 }
 
+void pipeline::impl::draw() {
 
-
-
-void pipeline::impl::draw()
-{
-
-
-
-    glClearColor(1.0,0,0,1);
+    glClearColor(1.0, 0, 0, 1);
     glClear(GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT);
 
     //        glViewport(0,0,800,600);
@@ -320,12 +289,8 @@ void pipeline::impl::draw()
 
     tr.draw();
 
-
-
     // reset
 
-
-
     // actuall blitting
 
 #if 0
@@ -346,7 +311,6 @@ void pipeline::impl::draw()
 
 #else
 
-
 #if 0
 
     glBindFramebuffer(GL_READ_FRAMEBUFFER, _fbo_msaa); // src FBO (multi-sample)
@@ -372,87 +336,76 @@ void pipeline::impl::draw()
     //    fb.unbind();
 
     GLenum err;
-    while((err = glGetError()) != GL_NO_ERROR)
-    {
+    while ((err = glGetError()) != GL_NO_ERROR) {
         std::string error;
 
-        switch(err) {
-        case GL_INVALID_OPERATION:      error="INVALID_OPERATION";      break;
-        case GL_INVALID_ENUM:           error="INVALID_ENUM";           break;
-        case GL_INVALID_VALUE:          error="INVALID_VALUE";          break;
-        case GL_OUT_OF_MEMORY:          error="OUT_OF_MEMORY";          break;
-        case GL_INVALID_FRAMEBUFFER_OPERATION:  error="INVALID_FRAMEBUFFER_OPERATION";  break;
+        switch (err) {
+        case GL_INVALID_OPERATION:
+            error = "INVALID_OPERATION";
+            break;
+        case GL_INVALID_ENUM:
+            error = "INVALID_ENUM";
+            break;
+        case GL_INVALID_VALUE:
+            error = "INVALID_VALUE";
+            break;
+        case GL_OUT_OF_MEMORY:
+            error = "OUT_OF_MEMORY";
+            break;
+        case GL_INVALID_FRAMEBUFFER_OPERATION:
+            error = "INVALID_FRAMEBUFFER_OPERATION";
+            break;
         }
 
         debug::e() << "OpenGL error:" << err << " " << error;
     }
-
 }
 
-
 //
 //
 //
 
-pipeline::pipeline()
-    : _impl(std::make_unique<pipeline::impl>())
-{
-}
+pipeline::pipeline() : _impl(std::make_unique<pipeline::impl>()) {}
 
-pipeline::~pipeline()
-{
+pipeline::~pipeline() {
     //
 }
 
-void pipeline::draw()
-{
-    _impl->draw();
-}
+void pipeline::draw() { _impl->draw(); }
 
-bool pipeline::create(size s)
-{
-    return _impl->create(s.cast<int>());
-}
+bool pipeline::create(size s) { return _impl->create(s.cast<int>()); }
 
-}
+} // namespace pw
 
+// class pipeline;
 
-
-
-//class pipeline;
-
-//class pass
+// class pass
 //{
-//public:
+// public:
 //	virtual void apply(pipeline& p);
-//};
+// };
 
-//class pipeline
+// class pipeline
 //{
 
-//public:
+// public:
 //	void apply()
 //	{
 //		for (auto p : _passes) p.apply(*this);
 //	}
 
-//protected:
+// protected:
 //	std::vector<pass> _passes;
-//};
+// };
 
-
-
-
-
-
-
-//pipeline >
+// pipeline >
 //		n*pass
 //		shadow_pass
-//      lighting_pass
+//       lighting_pass
 //		mesh_pass
 //		compositor_pass
 
-//compositor
-// render to fbo > add a
-// glBlitFramebuffer .. https://stackoverflow.com/questions/29254574/using-glblitframebuffer-to-display-a-texture
+// compositor
+//  render to fbo > add a
+//  glBlitFramebuffer ..
+//  https://stackoverflow.com/questions/29254574/using-glblitframebuffer-to-display-a-texture

src/visual/src/renderer.cpp

@@ -1,128 +1,115 @@
 #include "pw/visual/renderer.hpp"
 
-#include "pw/core/geometry.hpp"
-#include "pw/core/size.hpp"
 #include "pw/core/debug.hpp"
 #include "pw/core/matrix.hpp"
+#include "pw/core/mesh.hpp"
+#include "pw/core/primitives.hpp"
+#include "pw/core/size.hpp"
 
 #include "glad/glad.h"
 
 #include <algorithm>
-
+#include <atomic>
 
 namespace pw {
 
 struct renderer::impl {
 
-    uint64_t _change_count { 0 };
+    std::atomic<uint64_t> change_count{};
 
-    uint32_t _vao { 0 };
-    uint32_t _ebo { 0 };
-    std::vector<uint32_t> _vbos { };
+    uint32_t _vao{0};
+    uint32_t _ebo{0};
+    std::vector<uint32_t> _vbos{};
 
-
-    GLint _mesh_elements = { 0 };
+    GLint _mesh_elements = {0};
 
     impl() = default;
 
-    ~impl()
-    {
-        release();
-    }
+    ~impl() { release(); }
 
-    bool ready() const
-    {
+    bool ready() const {
         return glIsVertexArray != nullptr && GL_TRUE == glIsVertexArray(_vao);
     }
 
-    bool update(const geometry& m)
-    {
-        if (_change_count == m.change_count())
+    bool update(const mesh& m) {
+        if (this->change_count == m.change_count)
             return false;
 
         // reset if the renderer already in use
         if (ready())
             release();
 
-        _mesh_elements = m.indices().size();
+        _mesh_elements = m.geometry.indices.size();
 
         //
-        glGenVertexArrays(1,&_vao);
+        glGenVertexArrays(1, &_vao);
         glBindVertexArray(_vao);
 
         glGenBuffers(1, &_ebo);
 
-
         // TODO binding separate VBOs to the
         // vertexarray should be avoided
 
         // indices
         glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, _ebo);
         glBufferData(GL_ELEMENT_ARRAY_BUFFER,
-                     m.indices().size() * sizeof (uint32_t),
-                     m.indices().data(),
-                     GL_STATIC_DRAW);
+                     m.geometry.indices.size() * sizeof(uint32_t),
+                     m.geometry.indices.data(), GL_STATIC_DRAW);
 
-
-        _vbos.resize(_vbos.size()+1);
+        _vbos.resize(_vbos.size() + 1);
         glGenBuffers(1, &_vbos.back());
 
-
         // vertices
         glBindBuffer(GL_ARRAY_BUFFER, _vbos.back());
-        glVertexAttribPointer(_vbos.size()-1, vector3::coefficients, GL_FLOAT, GL_FALSE, 0, nullptr);
+        glVertexAttribPointer(_vbos.size() - 1, vector3f::coefficients, GL_FLOAT,
+                              GL_FALSE, 0, nullptr);
         glBufferData(GL_ARRAY_BUFFER,
-                     m.vertices().size() * sizeof(vector3),
-                     m.vertices().data(),
-                     GL_STATIC_DRAW);
-        glEnableVertexAttribArray(_vbos.size()-1);
+                     m.geometry.vertices.size() * sizeof(vector3f),
+                     m.geometry.vertices.data(), GL_STATIC_DRAW);
+        glEnableVertexAttribArray(_vbos.size() - 1);
 
-        debug::d() << "vertices at " << _vbos.size()-1;
+        debug::d() << "vertices at " << _vbos.size() - 1;
 
-        if (!m.normals().empty())
-        {
-            _vbos.resize(_vbos.size()+1);
+#if 0
+
+        if (!m.normals().empty()) {
+            _vbos.resize(_vbos.size() + 1);
             glGenBuffers(1, &_vbos.back());
 
             // normals
-            glBindBuffer(GL_ARRAY_BUFFER,_vbos.back());
-            glVertexAttribPointer(_vbos.size()-1, vector3::coefficients, GL_FLOAT, GL_FALSE, 0, nullptr);
-            glBufferData(GL_ARRAY_BUFFER,
-                         m.normals().size() * sizeof(vector3),
-                         m.normals().data(),
-                         GL_STATIC_DRAW);
-            glEnableVertexAttribArray(_vbos.size()-1);
-
-            debug::d() << "normals at " << _vbos.size()-1;
+            glBindBuffer(GL_ARRAY_BUFFER, _vbos.back());
+            glVertexAttribPointer(_vbos.size() - 1, vector3f::coefficients,
+                                  GL_FLOAT, GL_FALSE, 0, nullptr);
+            glBufferData(GL_ARRAY_BUFFER, m.normals().size() * sizeof(vector3f),
+                         m.normals().data(), GL_STATIC_DRAW);
+            glEnableVertexAttribArray(_vbos.size() - 1);
 
+            debug::d() << "normals at " << _vbos.size() - 1;
         }
 
-        if (!m.texture_coordinates().empty())
-        {
-            for (const auto& tc : m.texture_coordinates())
-            {
-                _vbos.resize(_vbos.size()+1);
+        if (!m.texture_coordinates().empty()) {
+            for (const auto& tc : m.texture_coordinates()) {
+                _vbos.resize(_vbos.size() + 1);
                 glGenBuffers(1, &_vbos.back());
 
                 // texture coordinates
-                glBindBuffer(GL_ARRAY_BUFFER,_vbos.back());
-                glVertexAttribPointer(_vbos.size()-1, vector2::coefficients, GL_FLOAT, GL_FALSE, 0, nullptr);
-                glBufferData(GL_ARRAY_BUFFER,
-                             tc.size() * sizeof(vector2),
-                             tc.data(),
-                             GL_STATIC_DRAW);
-                glEnableVertexAttribArray(_vbos.size()-1);
+                glBindBuffer(GL_ARRAY_BUFFER, _vbos.back());
+                glVertexAttribPointer(_vbos.size() - 1, vector2::coefficients,
+                                      GL_FLOAT, GL_FALSE, 0, nullptr);
+                glBufferData(GL_ARRAY_BUFFER, tc.size() * sizeof(vector2),
+                             tc.data(), GL_STATIC_DRAW);
+                glEnableVertexAttribArray(_vbos.size() - 1);
 
-                debug::d() << "texture coordinates at " << _vbos.size()-1;
+                debug::d() << "texture coordinates at " << _vbos.size() - 1;
             }
         }
 
+#endif
+
         // stop binding
         glBindVertexArray(0);
 
-
-       this->_change_count = m.change_count();
-
+        this->change_count.store(m.change_count.load());
 
 #if 1
         // get errors
@@ -136,19 +123,16 @@ struct renderer::impl {
         return ready();
     }
 
+    void release() {
 
-    void release()
-    {
+        for (auto& vbo : _vbos)
+            glDeleteBuffers(1, &vbo);
+        glDeleteBuffers(1, &_ebo);
 
-        for (auto vbo : _vbos)
-            glDeleteBuffers(1,&vbo);
-        glDeleteBuffers(1,&_ebo);
-
-        glDeleteVertexArrays(1,&_vao);
+        glDeleteVertexArrays(1, &_vao);
     }
 
-    void draw()
-    {
+    void draw() {
         glBindVertexArray(_vao);
         glDrawElements(GL_TRIANGLES, _mesh_elements, GL_UNSIGNED_INT, nullptr);
         glBindVertexArray(0);
@@ -163,57 +147,28 @@ struct renderer::impl {
     }
 
     //	GLint get_mode(vertex_array::)
-
 };
 
 //
 // Outer wrapper
 //
 
-renderer::renderer()
-    : _impl(std::make_unique<renderer::impl>())
-{
-}
+renderer::renderer() : _impl(std::make_unique<renderer::impl>()) {}
 
-renderer::renderer(const geometry &m)
-{
+renderer::renderer(const mesh& m) {
     renderer();
     // directly update
     _impl->update(m);
 }
 
-renderer::~renderer()
-{
-}
+renderer::~renderer() = default;
 
-bool renderer::ready() const
-{
-    return _impl->ready();
-}
+bool renderer::ready() const { return _impl->ready(); }
 
-bool renderer::update(const geometry &m)
-{
-    return _impl->update(m);
-}
+bool renderer::update(const mesh& m) { return _impl->update(m); }
 
-void renderer::release()
-{
-    _impl->release();
-}
+void renderer::release() { _impl->release(); }
 
-void renderer::draw()
-{
-    _impl->draw();
-}
+void renderer::draw() { _impl->draw(); }
 
-uint64_t renderer::change_count() const
-{
-    return _impl->_change_count;
-}
-
-void renderer::set_change_count(uint64_t n)
-{
-    _impl->_change_count = n;
-}
-
-}
+} // namespace pw

src/visual/src/shader.cpp

@@ -6,297 +6,240 @@
 
 namespace pw {
 
-struct shader::impl
-{
-	shader& _shader;
+struct shader::impl {
+    shader& _shader;
 
-	GLuint _shader_program;
-	std::vector<GLuint> _shader_stages;
+    GLuint _shader_program;
+    std::vector<GLuint> _shader_stages;
 
-	impl(shader& s)
-		: _shader(s)
-	{
-	}
+    impl(shader& s) : _shader(s) {}
 
-	~impl()
-	{
-        destroy();
-	}
+    ~impl() { destroy(); }
 
-	bool is_valid()
-	{
+    bool is_valid() {
         // we potentially haul in is_valid while no context is given
         return glIsProgram != nullptr && glIsProgram(_shader_program);
-	}
+    }
 
-	bool build()
-	{
-//        if (!is_valid()) return false;
+    bool build() {
+        //        if (!is_valid()) return false;
 
-        for (const auto & [type,code] : _shader._source)
-		{
-			GLuint shader_type = 0;
+        for (const auto& [type, code] : _shader._source) {
+            GLuint shader_type = 0;
             switch (type) {
-			case shader::code_type::vertex:
-				shader_type = GL_VERTEX_SHADER;
-				break;
-			case shader::code_type::compute:
-				shader_type = GL_COMPUTE_SHADER;
-				break;
-			case shader::code_type::geometry:
-				shader_type = GL_GEOMETRY_SHADER;
-				break;
-			case shader::code_type::fragment:
-				shader_type = GL_FRAGMENT_SHADER;
-				break;
+            case shader::code_type::vertex:
+                shader_type = GL_VERTEX_SHADER;
+                break;
+            case shader::code_type::compute:
+                shader_type = GL_COMPUTE_SHADER;
+                break;
+            case shader::code_type::geometry:
+                shader_type = GL_GEOMETRY_SHADER;
+                break;
+            case shader::code_type::fragment:
+                shader_type = GL_FRAGMENT_SHADER;
+                break;
             default:
                 debug::w() << " unknown shader type";
-			}
+            }
 
-			GLuint shaderId = glCreateShader(shader_type);
+            GLuint shaderId = glCreateShader(shader_type);
 
-            char* src = const_cast<char*>(code.c_str());
+            char* src  = const_cast<char*>(code.c_str());
             GLint size = static_cast<GLint>(code.length());
 
-			glShaderSource(shaderId , 1, &src, &size);
+            glShaderSource(shaderId, 1, &src, &size);
 
-			glCompileShader(shaderId);
+            glCompileShader(shaderId);
 
-			GLint is_compiled = GL_FALSE;
-			glGetShaderiv(shaderId, GL_COMPILE_STATUS, &is_compiled);
-			if(is_compiled == GL_FALSE)
-			{
+            GLint is_compiled = GL_FALSE;
+            glGetShaderiv(shaderId, GL_COMPILE_STATUS, &is_compiled);
+            if (is_compiled == GL_FALSE) {
 
-				GLint log_length;
+                GLint log_length;
 
-				glGetShaderiv(shaderId, GL_INFO_LOG_LENGTH, &log_length);
+                glGetShaderiv(shaderId, GL_INFO_LOG_LENGTH, &log_length);
 
-				std::vector<char> log_buffer(static_cast<size_t>(log_length));
+                std::vector<char> log_buffer(static_cast<size_t>(log_length));
 
-				glGetShaderInfoLog(shaderId, log_length, &log_length, log_buffer.data());
+                glGetShaderInfoLog(shaderId, log_length, &log_length,
+                                   log_buffer.data());
 
-				// TODO - handle errors!
-				debug::e() << log_buffer.data();
+                // TODO - handle errors!
+                debug::e() << log_buffer.data();
 
-				return false;
-			}
+                return false;
+            }
 
-			_shader_stages.push_back(shaderId);
+            _shader_stages.push_back(shaderId);
+        }
 
-		}
-
-		_shader_program = glCreateProgram();
-
-
-		for (auto s : _shader_stages)
-			glAttachShader(_shader_program,s);
+        _shader_program = glCreateProgram();
 
+        for (auto s : _shader_stages)
+            glAttachShader(_shader_program, s);
 
         // TODO attribute binding ...
 
-		/* Bind attribute index 0 (coordinates) to in_Position and attribute index 1 (color) to in_Color */
-		/* Attribute locations must be setup before calling glLinkProgram. */
-		//		 glBindAttribLocation(shaderprogram, 0, "in_Position");
-		//		 glBindAttribLocation(shaderprogram, 1, "in_Color");
+        /* Bind attribute index 0 (coordinates) to in_Position and attribute
+         * index 1 (color) to in_Color */
+        /* Attribute locations must be setup before calling glLinkProgram. */
+        //		 glBindAttribLocation(shaderprogram, 0, "in_Position");
+        //		 glBindAttribLocation(shaderprogram, 1, "in_Color");
 
-		glLinkProgram(_shader_program);
+        glLinkProgram(_shader_program);
 
-		GLint is_linked = 0;
-		glGetProgramiv(_shader_program, GL_LINK_STATUS, &is_linked);
-		if(is_linked == GL_FALSE)
-		{
+        GLint is_linked = 0;
+        glGetProgramiv(_shader_program, GL_LINK_STATUS, &is_linked);
+        if (is_linked == GL_FALSE) {
 
-			GLint log_length;
+            GLint log_length;
 
-			/* Noticed that glGetProgramiv is used to get the length for a shader program, not glGetShaderiv. */
-			glGetProgramiv(_shader_program, GL_INFO_LOG_LENGTH, &log_length);
+            /* Noticed that glGetProgramiv is used to get the length for a
+             * shader program, not glGetShaderiv. */
+            glGetProgramiv(_shader_program, GL_INFO_LOG_LENGTH, &log_length);
 
-			/* The maxLength includes the NULL character */
-			std::vector<char> info_log(static_cast<size_t>(log_length));
+            /* The maxLength includes the NULL character */
+            std::vector<char> info_log(static_cast<size_t>(log_length));
 
-			/* Notice that glGetProgramInfoLog, not glGetShaderInfoLog. */
-			glGetProgramInfoLog(_shader_program, log_length, &log_length, info_log.data());
+            /* Notice that glGetProgramInfoLog, not glGetShaderInfoLog. */
+            glGetProgramInfoLog(_shader_program, log_length, &log_length,
+                                info_log.data());
 
-			debug::e() << info_log.data();
+            debug::e() << info_log.data();
 
-			/* Handle the error in an appropriate way such as displaying a message or writing to a log file. */
-			/* In this simple program, we'll just leave */
+            /* Handle the error in an appropriate way such as displaying a
+             * message or writing to a log file. */
+            /* In this simple program, we'll just leave */
 
-			return false;
-		}
+            return false;
+        }
 
-		return true;
-	}
+        return true;
+    }
 
+    void use() { glUseProgram(_shader_program); }
 
-	void use()
-	{
-		glUseProgram(_shader_program);
-	}
-
-    void destroy()
-    {
+    void destroy() {
         // potentially the GL driver hasn't been loaded
         if (is_valid()) {
 
             // deleting and detaching should happen much earlier
 
-			for (auto s : _shader_stages)
-			{
-				glDeleteShader(s);
+            for (auto s : _shader_stages) {
+                glDeleteShader(s);
             }
 
             // only program needs to be deleted
 
             glDeleteProgram(_shader_program);
         }
-	}
-
-	int uniform_location(std::string const& name) const
-	{
-		return glGetUniformLocation(_shader_program,name.c_str());
-	}
-
-	void bind(int location,const matrix3x3f& m)
-    {
-		glUniformMatrix3fv(location,1,GL_FALSE,m.ptr());
-	}
-
-	void bind(int location,const matrix4x4f& m)
-	{
-		glUniformMatrix4fv(location,1,GL_FALSE,m.ptr());
-	}
-
-	void bind(int location,const vector4f& v)
-	{
-		glUniform4fv(location,1,v.ptr());
-	}
-
-	void bind(int location,const float& v)
-	{
-		glUniform1f(location,v);
-	}
-
-    void bind(int location,const uint32_t& i)
-    {
-        glUniform1ui(location,i);
     }
 
-    void bind(int location,const int32_t& i)
-    {
-        glUniform1i(location,i);
+    int uniform_location(std::string const& name) const {
+        return glGetUniformLocation(_shader_program, name.c_str());
     }
 
-    void bind(int location,const texture& v)
-    {
-        this->bind(location,(int)v.native_handle());
+    void bind(int location, const matrix3x3f& m) {
+        glUniformMatrix3fv(location, 1, GL_FALSE, m.data());
+    }
 
+    void bind(int location, const matrix4x4f& m) {
+        glUniformMatrix4fv(location, 1, GL_FALSE, m.data());
+    }
+
+    void bind(int location, const vector4f& v) {
+        glUniform4fv(location, 1, v.data());
+    }
+
+    void bind(int location, const float& v) { glUniform1f(location, v); }
+
+    void bind(int location, const uint32_t& i) { glUniform1ui(location, i); }
+
+    void bind(int location, const int32_t& i) { glUniform1i(location, i); }
+
+    void bind(int location, const texture& v) {
+        this->bind(location, (int)v.native_handle());
     }
 };
 
+shader::shader() { _impl = make_unique<impl>(*this); }
 
-shader::shader()
-{
-    _impl = make_unique<impl>(*this);
-}
+shader::~shader() {}
 
-shader::~shader()
-{
+bool shader::ready() const { return _impl->is_valid(); }
 
-}
-
-bool shader::ready() const
-{
-	return _impl->is_valid();
-}
-
-shader &shader::set_uniform_at_location(int location, float v)
-{
-    _impl->bind(location,v);
+shader& shader::set_uniform_at_location(int location, float v) {
+    _impl->bind(location, v);
     return *this;
 }
 
-shader &shader::set_uniform_at_location(int location, uint32_t v)
-{
-    _impl->bind(location,v); return *this;
+shader& shader::set_uniform_at_location(int location, uint32_t v) {
+    _impl->bind(location, v);
+    return *this;
 }
 
-shader &shader::set_uniform_at_location(int location, int32_t v)
-{
-    _impl->bind(location,v); return *this;
+shader& shader::set_uniform_at_location(int location, int32_t v) {
+    _impl->bind(location, v);
+    return *this;
 }
 
-
-shader &shader::set_uniform_at_location(int location, vector4f const &v)
-{
-	_impl->bind(location,v); return *this;
+shader& shader::set_uniform_at_location(int location, vector4f const& v) {
+    _impl->bind(location, v);
+    return *this;
 }
 
-shader &shader::set_uniform_at_location(int location, matrix4x4f const &v)
-{
-    _impl->bind(location,v); return *this;
+shader& shader::set_uniform_at_location(int location, matrix4x4f const& v) {
+    _impl->bind(location, v);
+    return *this;
 }
 
-
-shader &shader::set_uniform_at_location(int location, texture const &v)
-{
-    _impl->bind(location,v); return *this;
+shader& shader::set_uniform_at_location(int location, texture const& v) {
+    _impl->bind(location, v);
+    return *this;
 }
 
-bool shader::build()
-{
-	return _impl->build();
-}
+bool shader::build() { return _impl->build(); }
 
-void shader::use()
-{
-	_impl->use();
-}
+void shader::use() { _impl->use(); }
 
-void shader::set_uniforms(uniform_cache_t c)
-{
+void shader::set_uniforms(uniform_cache_t c) {
     // TODO rewrite in proper C++17
 
-	for (auto& u : c) {
-		// get name
-		std::string name = std::get<0>(u);
-		// get location
-		GLint loc = std::get<2>(u);
+    for (auto& u : c) {
+        // get name
+        std::string name = std::get<0>(u);
+        // get location
+        GLint loc = std::get<2>(u);
 
-		// if lower 0 check for location
-		if (loc < 0) {
-			loc = _impl->uniform_location(name);
-			std::get<2>(u) = loc; // cache location
-		}
+        // if lower 0 check for location
+        if (loc < 0) {
+            loc            = _impl->uniform_location(name);
+            std::get<2>(u) = loc; // cache location
+        }
 
-		auto var = std::get<1>(u);
-
-		std::visit([this,loc](auto&& arg) {
+        auto var = std::get<1>(u);
 
+        std::visit([this, loc](auto&& arg) {
             using T = std::decay_t<decltype(arg)>;
 
             // TODO query the std::variant of uniform_t
 
-			if constexpr ((std::is_same_v<T, vector4f>) ||
-				(std::is_same_v<T, matrix4x4f>) ||
-				(std::is_same_v<T, float>) ) {
-                    set_uniform_at_location( loc, std::forward<T>(arg));
-			} else {
+            if constexpr ((std::is_same_v<T, vector4f>) ||
+                          (std::is_same_v<T, matrix4x4f>) ||
+                          (std::is_same_v<T, float>)) {
+                set_uniform_at_location(loc, std::forward<T>(arg));
+            } else {
                 debug::e() << "unknown uniform type";
-			}
-		}, var);
-	}
+            }
+        }, var);
+    }
 }
 
-uint32_t shader::native_handle() const
-{
-	return _impl->_shader_program;
+uint32_t shader::native_handle() const { return _impl->_shader_program; }
+
+int shader::uniform_location(const std::string& name) const {
+    return _impl->uniform_location(name);
 }
 
-int shader::uniform_location(const std::string &name) const
-{
-	return _impl->uniform_location(name);
-}
-
-
-}
+} // namespace pw

tests/CMakeLists.txt

@@ -0,0 +1,18 @@
+include(FetchContent)
+
+FetchContent_Declare(snitch
+                     GIT_REPOSITORY https://github.com/snitch-org/snitch.git
+                     GIT_TAG        v1.3.2) # update version number as needed
+
+FetchContent_MakeAvailable(snitch)
+
+set(pixwerx_test_files
+    pw_test_core.cpp
+  )
+
+add_executable(pixwerx_tests ${pixwerx_test_files})
+
+target_link_libraries(pixwerx_tests PRIVATE
+  snitch::snitch
+  pwcore
+  )

tests/pw_test_core.cpp

@@ -0,0 +1,112 @@
+#include <snitch/snitch.hpp>
+
+#include <pw/core/matrix.hpp>
+#include <pw/core/vector.hpp>
+
+TEST_CASE("core", "[matrix]") {
+
+    auto mat = pw::matrix<float,2,2>{};
+
+    REQUIRE(mat[0][0] == 0);
+    REQUIRE(mat[0][1] == 0);
+    REQUIRE(mat[1][0] == 0);
+    REQUIRE(mat[1][1] == 0);
+
+    auto mat_1 = pw::matrix<float,2, 2>{
+        pw::vector{5.f, 4},
+        pw::vector{3.f, 2}
+    };
+
+    REQUIRE(mat_1[0][0] == 5);
+    REQUIRE(mat_1[0][1] == 4);
+    REQUIRE(mat_1[1][0] == 3);
+    REQUIRE(mat_1[1][1] == 2);
+
+    const auto val {42.42f};
+
+    auto mat_2 = pw::matrix<float,2,2>::all(val);
+
+    REQUIRE(mat_2[0][0] == val);
+    REQUIRE(mat_2[0][1] == val);
+    REQUIRE(mat_2[1][0] == val);
+    REQUIRE(mat_2[1][1] == val);
+
+    mat_2 *= 2;
+    REQUIRE(mat_2[0][0] == (val * 2));
+    REQUIRE(mat_2[0][1] == (val * 2));
+    REQUIRE(mat_2[1][0] == (val * 2));
+    REQUIRE(mat_2[1][1] == (val * 2));
+
+
+    mat_2 = pw::matrix<float, 2, 2>::identity();
+
+    REQUIRE(mat_2[0][0] == 1.0f);
+    REQUIRE(mat_2[0][1] == 0.0f);
+    REQUIRE(mat_2[1][0] == 0.0f);
+    REQUIRE(mat_2[1][1] == 1.0f);
+
+
+    auto mat_2_inv = mat_2.inverse();
+
+    REQUIRE(mat_2_inv[0][0] == 1.0f);
+    REQUIRE(mat_2_inv[0][1] == 0.0f);
+    REQUIRE(mat_2_inv[1][0] == 0.0f);
+    REQUIRE(mat_2_inv[1][1] == 1.0f);
+
+    // auto m44_2 = pw::matrix2x2::make(0, 1, 2, 3);
+
+    // REQUIRE(m44_2(0, 0) == 0.0f);
+    // REQUIRE(m44_2(1, 0) == 1.0f);
+    // REQUIRE(m44_2(0, 1) == 2.0f);
+    // REQUIRE(m44_2(1, 1) == 3.0f);
+
+
+    // pw::matrix2x2 m44_3 = {{},{1, 2, 3, 4}};
+
+    // REQUIRE(m44_3(0, 0) == 1.0f);
+    // REQUIRE(m44_3(1, 0) == 2.0f);
+    // REQUIRE(m44_3(0, 1) == 3.0f);
+    // REQUIRE(m44_3(1, 1) == 4.0f);
+
+}
+
+
+TEST_CASE("core", "[vector.matrix]") {
+
+    auto vec4_1 = pw::vector4<float>{};
+
+    REQUIRE(vec4_1[0] == 0.0f);
+    REQUIRE(vec4_1[1] == 0.0f);
+    REQUIRE(vec4_1[2] == 0.0f);
+    REQUIRE(vec4_1[3] == 0.0f);
+
+
+    // auto mat4_1 = pw::matrix_<4,1,float>{};
+
+
+    // mat4_1 = vec4_1; // down is easy
+
+    // vec4_1 = mat4_1;
+
+}
+
+TEST_CASE("core", "[vector]") {
+
+    // auto vec4_1 = pw::vector4_<float>{};
+    // auto vec4_2 = pw::vector4::make(1, 2, 3, 4);
+
+    // pw::vector4 vec4_3 = vec4_1 + vec4_2;
+
+    // REQUIRE(vec4_3[0] == 1.0f);
+    // REQUIRE(vec4_3[1] == 2.0f);
+    // REQUIRE(vec4_3[2] == 3.0f);
+    // REQUIRE(vec4_3[3] == 4.0f);
+
+    // auto vec4_4 = vec4_3 * 2;
+
+    // REQUIRE(vec4_4[0] == 1.0f * 2);
+    // REQUIRE(vec4_4[1] == 2.0f * 2);
+    // REQUIRE(vec4_4[2] == 3.0f * 2);
+    // REQUIRE(vec4_4[3] == 4.0f * 2);
+
+}