ci: Try another approach to docs workflow

I hate CI.

Signed-off-by: Slendi <slendi@socopon.com>

.clang-format

@@ -0,0 +1,51 @@
+---
+Language: Cpp
+BasedOnStyle: LLVM
+
+IndentWidth: 4
+TabWidth: 4
+UseTab: ForIndentation
+
+ColumnLimit: 80
+AlignAfterOpenBracket: DontAlign
+AlignOperands: false
+BreakBeforeBinaryOperators: All
+ContinuationIndentWidth: 4
+BreakBeforeBraces: Custom
+BraceWrapping:
+  AfterCaseLabel: false
+  AfterClass: true
+  AfterControlStatement: Never
+  AfterEnum: true
+  AfterFunction: true
+  AfterNamespace: true
+  AfterStruct: true
+  AfterUnion: true
+  BeforeCatch: false
+  BeforeElse: false
+  SplitEmptyFunction: false
+  SplitEmptyRecord: false
+  SplitEmptyNamespace: false
+AllowShortIfStatementsOnASingleLine: false
+AllowShortLoopsOnASingleLine: false
+AllowShortFunctionsOnASingleLine: Inline
+AllowShortBlocksOnASingleLine: Empty
+SpaceInEmptyBlock: true
+BinPackArguments: false
+BinPackParameters: false
+Cpp11BracedListStyle: false
+SpaceBeforeCpp11BracedList: true
+IndentRequiresClause: false
+RequiresClausePosition: OwnLine
+
+PointerAlignment: Right
+ReferenceAlignment: Right
+IndentAccessModifiers: false
+AccessModifierOffset: -4
+
+SortIncludes: CaseSensitive
+SpaceAfterTemplateKeyword: false
+AlignEscapedNewlines: DontAlign
+AlignConsecutiveAssignments: false
+AlignConsecutiveDeclarations: false
+...

.github/workflows/build.yml

@@ -21,7 +21,7 @@ jobs:
       - name: Install Nix
         uses: DeterminateSystems/nix-installer-action@main
       - name: Building default package
-        run: nix build .#default
+        run: nix build .#default --print-build-logs
       - name: Upload artifact
         uses: actions/upload-artifact@v4
         with:

.github/workflows/docs.yml

@@ -0,0 +1,50 @@
+name: Deploy Doxygen docs
+
+on:
+  push:
+    branches: [master]
+
+permissions:
+  contents: read
+  pages: write
+  id-token: write
+
+concurrency:
+  group: "pages"
+  cancel-in-progress: true
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Install Doxygen
+        run: sudo apt-get update && sudo apt-get install -y doxygen graphviz
+
+      - name: Create output directory
+        run: mkdir -p build/docs
+
+      - name: Generate docs
+        run: doxygen docs/Doxyfile
+
+      - name: Disable Jekyll
+        run: touch build/docs/html/.nojekyll
+
+      - name: Upload Pages artifact
+        uses: actions/upload-pages-artifact@v3
+        with:
+          path: build/docs/html
+
+  deploy:
+    environment:
+      name: github-pages
+      url: ${{ steps.deployment.outputs.page_url }}
+    runs-on: ubuntu-latest
+    needs: build
+
+    steps:
+      - name: Deploy to GitHub Pages
+        id: deployment
+        uses: actions/deploy-pages@v4

CMakeLists.txt

@@ -1,5 +1,5 @@
 cmake_minimum_required(VERSION 3.15)
-project(SmathExamples LANGUAGES CXX VERSION 0.1.0)
+project(SmathExamples LANGUAGES CXX VERSION 1.0.0)
 
 set(CMAKE_CXX_STANDARD 23)
 set(CMAKE_CXX_STANDARD_REQUIRED ON)
@@ -7,6 +7,8 @@ set(CMAKE_CXX_STANDARD_REQUIRED ON)
 option(SMATH_BUILD_EXAMPLES "Build example programs" ON)
 option(SMATH_BUILD_TESTS "Build unit tests" ON)
 option(SMATH_BUILD_MODULES "Enable C++20 modules support" OFF)
+option(SMATH_ENABLE_LEGACY_HEADER "Install legacy include <smath.hpp> shim" OFF)
+option(SMATH_INSTALL_INTEROP_HEADERS "Install interop headers" OFF)
 
 if(SMATH_BUILD_MODULES)
 	message(STATUS "Building smath C++ module")
@@ -17,6 +19,7 @@ add_library(smath INTERFACE)
 target_include_directories(smath
 	INTERFACE
 		$<BUILD_INTERFACE:${CMAKE_CURRENT_SOURCE_DIR}/include>
+		$<$<BOOL:${SMATH_ENABLE_LEGACY_HEADER}>:$<BUILD_INTERFACE:${CMAKE_CURRENT_BINARY_DIR}/include>>
 		$<INSTALL_INTERFACE:include>
 )
 add_library(smath::smath ALIAS smath)
@@ -25,7 +28,21 @@ include(CMakePackageConfigHelpers)
 install(TARGETS smath
 	EXPORT smathTargets
 )
-install(DIRECTORY include/ DESTINATION include)
+install(FILES include/smath/smath.hpp DESTINATION include/smath)
+
+if(SMATH_INSTALL_INTEROP_HEADERS)
+	install(DIRECTORY include/smath/interop DESTINATION include/smath)
+endif()
+
+if(SMATH_ENABLE_LEGACY_HEADER)
+	file(MAKE_DIRECTORY "${CMAKE_CURRENT_BINARY_DIR}/include")
+	configure_file(
+		"${CMAKE_CURRENT_SOURCE_DIR}/cmake/smath_legacy_header.hpp.in"
+		"${CMAKE_CURRENT_BINARY_DIR}/include/smath.hpp"
+		@ONLY
+	)
+	install(FILES "${CMAKE_CURRENT_BINARY_DIR}/include/smath.hpp" DESTINATION include)
+endif()
 
 install(EXPORT smathTargets
 	NAMESPACE smath::
@@ -96,4 +113,3 @@ if(SMATH_BUILD_TESTS)
 	include(GoogleTest)
 	gtest_discover_tests(smath_tests)
 endif()
-

README.md

@@ -1,6 +1,16 @@
-# smath
+<p align="center">
-
+  <img alt="smath logo" width="100%" src="https://github.com/user-attachments/assets/d6f2ef4b-eca0-4004-9099-37423528bcba" />
-Single-file, header-only linear algebra math library for C++23.
+  <br><br>
+  <a href="https://github.com/slendidev/smath/actions/workflows/build.yml">
+		<img src="https://github.com/slendidev/smath/actions/workflows/build.yml/badge.svg" alt="Build">
+	</a>
+	<a href="LICENSE.txt">
+		<img src="https://img.shields.io/badge/license-Apache%202.0-blue.svg" alt="License: Apache-2.0">
+	</a>
+	<a href="https://en.cppreference.com/w/cpp/23">
+		<img src="https://img.shields.io/badge/C%2B%2B-23-00599C.svg" alt="C++23">
+	</a>
+</p>
 
 ## Features
 
@@ -12,8 +22,51 @@ Single-file, header-only linear algebra math library for C++23.
 - Angle helpers `rad/deg/turns` respecting a configurable base unit via the macro `SMATH_ANGLE_UNIT`.
 - Optional implicit conversions.
 - Packing utilities for normalized RGBA (`pack_unorm4x8`, `unpack_snorm4x8`, etc.).
+- C++20 modules support.
+- Built-in interop adapter and optional interop headers for various libraries.
+
+## Quick Start
+
+Create `main.cpp`:
+
+```cpp
+#include <print>
+#include <smath/smath.hpp>
+
+int main() {
+  using namespace smath;
+
+  Vec3 a{1.0f, 2.0f, 3.0f};
+  Vec3 b{3.0f, 2.0f, 1.0f};
+
+  std::println("a + b = {}", a + b);
+  std::println("dot(a, b) = {}", a.dot(b));
+  std::println("normalized(a) = {}", a.normalized());
+}
+```
+
+Build and run:
+
+```bash
+g++ -std=c++23 -Iinclude main.cpp -o quickstart
+./quickstart
+```
+
+## Interop Headers
+
+smath ships optional interop adapters under `include/smath/interop/`.
+
+When installing with CMake, interop headers are gated behind:
+
+- `-DSMATH_INSTALL_INTEROP_HEADERS=ON`
+
+All of them plug into the same generic API defined in `smath.hpp`:
+
+```cpp
+auto v = smath::from_external(external_value);
+auto ext = smath::to_external<ExternalType>(smath_value);
+```
 
 ## License
 
-This library is licensed under the Apache License 2.0. See the (LICENSE.txt)[LICENSE.txt] file for more details.
+This library is licensed under the Apache License 2.0. See the [LICENSE.txt](LICENSE.txt) file for more details.
-

cmake/smath_legacy_header.hpp.in

@@ -0,0 +1,3 @@
+#pragma once
+
+#include <smath/smath.hpp>

docs/Doxyfile

@@ -0,0 +1,27 @@
+PROJECT_NAME           = "smath"
+OUTPUT_DIRECTORY       = build/docs
+CREATE_SUBDIRS         = NO
+
+INPUT                  = include README.md
+USE_MDFILE_AS_MAINPAGE = README.md
+TOC_INCLUDE_HEADINGS   = 0
+RECURSIVE              = YES
+FILE_PATTERNS          = *.hpp *.h *.md
+EXCLUDE                = include/smath/interop
+
+EXTRACT_ALL            = YES
+EXTRACT_PRIVATE        = NO
+EXTRACT_STATIC         = NO
+
+GENERATE_HTML          = YES
+GENERATE_LATEX         = NO
+GENERATE_MAN           = YES
+GENERATE_TREEVIEW      = YES
+DISABLE_INDEX          = NO
+FULL_SIDEBAR           = NO
+HTML_COLORSTYLE        = LIGHT
+HTML_EXTRA_STYLESHEET  = docs/vendor/doxygen-awesome.css
+
+WARN_IF_UNDOCUMENTED   = YES
+QUIET                  = NO
+EXCLUDE_SYMBOLS        = std smath::detail smath::interop_adapter interop_adapter smath::from_external from_external smath::to_external to_external

examples/projection.cpp

@@ -10,205 +10,219 @@
 #include <Windows.h>
 #endif
 
-#include <smath.hpp>
+#include <smath/smath.hpp>
 using smath::Vec2;
 
-enum Color : uint8_t {
+enum Color : uint8_t
-  CLR_NONE = 0,  // default
+{
-  CLR_AXES = 90, // bright black (gray)
+	CLR_NONE = 0,  // default
-  CLR_A = 32,    // green
+	CLR_AXES = 90, // bright black (gray)
-  CLR_B = 33,    // yellow
+	CLR_A = 32,    // green
-  CLR_P = 35,    // magenta
+	CLR_B = 33,    // yellow
-  CLR_DOT = 36   // cyan
+	CLR_P = 35,    // magenta
+	CLR_DOT = 36   // cyan
 };
 
-struct Cell {
+struct Cell
-  char ch{' '};
+{
-  uint8_t color{CLR_NONE};
+	char ch { ' ' };
-  int prio{0};
+	uint8_t color { CLR_NONE };
+	int prio { 0 };
 };
 
-struct Canvas {
+struct Canvas
-  int w, h;
+{
-  std::vector<Cell> pix;
+	int w, h;
-  Canvas(int W, int H) : w(W), h(H), pix(W * H) {}
+	std::vector<Cell> pix;
+	Canvas(int W, int H) : w(W), h(H), pix(W * H) { }
 
-  void put(int x, int y, char c, int prio, uint8_t color) {
+	void put(int x, int y, char c, int prio, uint8_t color)
-    if (x < 0 || x >= w || y < 0 || y >= h)
+	{
-      return;
+		if (x < 0 || x >= w || y < 0 || y >= h)
-    Cell &cell = pix[y * w + x];
+			return;
-    if (prio >= cell.prio) {
+		Cell &cell = pix[y * w + x];
-      cell.ch = c;
+		if (prio >= cell.prio) {
-      cell.prio = prio;
+			cell.ch = c;
-      cell.color = color;
+			cell.prio = prio;
-    }
+			cell.color = color;
-  }
+		}
-  void hline(int y, char c, int prio, uint8_t color) {
+	}
-    for (int x = 0; x < w; ++x)
+	void hline(int y, char c, int prio, uint8_t color)
-      put(x, y, c, prio, color);
+	{
-  }
+		for (int x = 0; x < w; ++x)
-  void vline(int x, char c, int prio, uint8_t color) {
+			put(x, y, c, prio, color);
-    for (int y = 0; y < h; ++y)
+	}
-      put(x, y, c, prio, color);
+	void vline(int x, char c, int prio, uint8_t color)
-  }
+	{
+		for (int y = 0; y < h; ++y)
+			put(x, y, c, prio, color);
+	}
 
-  void line_dir(int x0, int y0, int x1, int y1, int prio, uint8_t color) {
+	void line_dir(int x0, int y0, int x1, int y1, int prio, uint8_t color)
-    int dx = std::abs(x1 - x0), sx = x0 < x1 ? 1 : -1;
+	{
-    int dy = -std::abs(y1 - y0), sy = y0 < y1 ? 1 : -1;
+		int dx = std::abs(x1 - x0), sx = x0 < x1 ? 1 : -1;
-    int err = dx + dy;
+		int dy = -std::abs(y1 - y0), sy = y0 < y1 ? 1 : -1;
+		int err = dx + dy;
 
-    int px = x0, py = y0;
+		int px = x0, py = y0;
-    put(px, py, '+', prio, color);
+		put(px, py, '+', prio, color);
 
-    while (true) {
+		while (true) {
-      if (px == x1 && py == y1)
+			if (px == x1 && py == y1)
-        break;
+				break;
-      int e2 = 2 * err;
+			int e2 = 2 * err;
-      int nx = px, ny = py;
+			int nx = px, ny = py;
-      if (e2 >= dy) {
+			if (e2 >= dy) {
-        err += dy;
+				err += dy;
-        nx += sx;
+				nx += sx;
-      }
+			}
-      if (e2 <= dx) {
+			if (e2 <= dx) {
-        err += dx;
+				err += dx;
-        ny += sy;
+				ny += sy;
-      }
+			}
 
-      int sdx = nx - px;
+			int sdx = nx - px;
-      int sdy = ny - py;
+			int sdy = ny - py;
-      int adx = std::abs(sdx);
+			int adx = std::abs(sdx);
-      int ady = std::abs(sdy);
+			int ady = std::abs(sdy);
-      char g;
+			char g;
-      if (adx >= 2 * ady)
+			if (adx >= 2 * ady)
-        g = '-';
+				g = '-';
-      else if (ady >= 2 * adx)
+			else if (ady >= 2 * adx)
-        g = '|';
+				g = '|';
-      else
+			else
-        g = ((sdx > 0 && sdy > 0) || (sdx < 0 && sdy < 0)) ? '\\' : '/';
+				g = ((sdx > 0 && sdy > 0) || (sdx < 0 && sdy < 0)) ? '\\' : '/';
 
-      put(nx, ny, g, prio, color);
+			put(nx, ny, g, prio, color);
-      px = nx;
+			px = nx;
-      py = ny;
+			py = ny;
-    }
+		}
-  }
+	}
 
-  void flush() const {
+	void flush() const
-    uint8_t cur = 255;
+	{
-    for (int y = 0; y < h; ++y) {
+		uint8_t cur = 255;
-      for (int x = 0; x < w; ++x) {
+		for (int y = 0; y < h; ++y) {
-        const Cell &c = pix[y * w + x];
+			for (int x = 0; x < w; ++x) {
-        if (c.color != cur) {
+				const Cell &c = pix[y * w + x];
-          if (c.color == CLR_NONE)
+				if (c.color != cur) {
-            std::print("\x1b[0m");
+					if (c.color == CLR_NONE)
-          else
+						std::print("\x1b[0m");
-            std::print("\x1b[{}m", c.color);
+					else
-          cur = c.color;
+						std::print("\x1b[{}m", c.color);
-        }
+					cur = c.color;
-        std::print("{}", c.ch);
+				}
-      }
+				std::print("{}", c.ch);
-      std::println("");
+			}
-    }
+			std::println("");
-    std::print("\x1b[0m");
+		}
-  }
+		std::print("\x1b[0m");
+	}
 };
 
-struct Mapper {
+struct Mapper
-  int W, H;
+{
-  double scale;
+	int W, H;
-  int cx, cy;
+	double scale;
-  Mapper(int w, int h, double s) : W(w), H(h), scale(s), cx(w / 2), cy(h / 2) {}
+	int cx, cy;
-  std::pair<int, int> map(Vec2 v) const {
+	Mapper(int w, int h, double s) : W(w), H(h), scale(s), cx(w / 2), cy(h / 2)
-    int x = cx + (int)std::llround(v.x() * scale);
+	{ }
-    int y = cy - (int)std::llround(v.y() * scale);
+	std::pair<int, int> map(Vec2 v) const
-    return {x, y};
+	{
-  }
+		int x = cx + (int)std::llround(v.x() * scale);
+		int y = cy - (int)std::llround(v.y() * scale);
+		return { x, y };
+	}
 };
 
-int main() {
+int main()
+{
 #if defined(_WIN32)
-  HANDLE h = GetStdHandle(STD_OUTPUT_HANDLE);
+	HANDLE h = GetStdHandle(STD_OUTPUT_HANDLE);
-  if (h != INVALID_HANDLE_VALUE) {
+	if (h != INVALID_HANDLE_VALUE) {
-    DWORD m;
+		DWORD m;
-    if (GetConsoleMode(h, &m))
+		if (GetConsoleMode(h, &m))
-      SetConsoleMode(h, m | ENABLE_VIRTUAL_TERMINAL_PROCESSING);
+			SetConsoleMode(h, m | ENABLE_VIRTUAL_TERMINAL_PROCESSING);
-  }
+	}
 #endif
 
-  std::print("Enter vector a (ax ay): ");
+	std::print("Enter vector a (ax ay): ");
-  double ax, ay;
+	double ax, ay;
-  if (!(std::cin >> ax >> ay))
+	if (!(std::cin >> ax >> ay))
-    return 0;
+		return 0;
-  std::print("Enter vector b (bx by): ");
+	std::print("Enter vector b (bx by): ");
-  double bx, by;
+	double bx, by;
-  if (!(std::cin >> bx >> by))
+	if (!(std::cin >> bx >> by))
-    return 0;
+		return 0;
 
-  Vec2 a{(float)ax, (float)ay};
+	Vec2 a { (float)ax, (float)ay };
-  Vec2 b{(float)bx, (float)by};
+	Vec2 b { (float)bx, (float)by };
-  Vec2 p = a.project_onto(b);
+	Vec2 p = a.project_onto(b);
 
-  std::println("\nResults:");
+	std::println("\nResults:");
-  std::println("a = {}", a);
+	std::println("a = {}", a);
-  std::println("b = {}", b);
+	std::println("b = {}", b);
-  std::println("proj_b(a) = p = {}", p);
+	std::println("proj_b(a) = p = {}", p);
-  std::println("|a|={}  |b|={}  |p|={}", a.magnitude(), b.magnitude(),
+	std::println(
-               p.magnitude());
+	    "|a|={}  |b|={}  |p|={}", a.magnitude(), b.magnitude(), p.magnitude());
-  double dot = a.dot(b);
+	double dot = a.dot(b);
-  double ang = (a.magnitude() > 0 && b.magnitude() > 0)
+	double ang = (a.magnitude() > 0 && b.magnitude() > 0)
-                   ? std::acos(std::clamp(dot / (a.magnitude() * b.magnitude()),
+	    ? std::acos(
-                                          -1.0, 1.0)) *
+	          std::clamp(dot / (a.magnitude() * b.magnitude()), -1.0, 1.0))
-                         180.0 / M_PI
+	        * 180.0 / M_PI
-                   : 0.0;
+	    : 0.0;
-  std::println("a·b={}  angle(a,b)={} deg\n", dot, ang);
+	std::println("a·b={}  angle(a,b)={} deg\n", dot, ang);
 
-  const int W = 73, H = 33;
+	const int W = 73, H = 33;
-  double maxr = std::max({(double)a.magnitude(), (double)b.magnitude(),
+	double maxr = std::max({ (double)a.magnitude(),
-                          (double)p.magnitude(), 1.0});
+	    (double)b.magnitude(),
-  double usable = 0.45 * std::min(W, H);
+	    (double)p.magnitude(),
-  double scale = usable / maxr;
+	    1.0 });
+	double usable = 0.45 * std::min(W, H);
+	double scale = usable / maxr;
 
-  Canvas cvs(W, H);
+	Canvas cvs(W, H);
-  Mapper mp(W, H, scale);
+	Mapper mp(W, H, scale);
 
-  int pr_axes = 1;
+	int pr_axes = 1;
-  auto [cx, cy] = mp.map({0, 0});
+	auto [cx, cy] = mp.map({ 0, 0 });
-  cvs.hline(H / 2, '-', pr_axes, CLR_AXES);
+	cvs.hline(H / 2, '-', pr_axes, CLR_AXES);
-  cvs.vline(W / 2, '|', pr_axes, CLR_AXES);
+	cvs.vline(W / 2, '|', pr_axes, CLR_AXES);
-  cvs.put(W / 2, H / 2, 'O', pr_axes + 1, CLR_AXES);
+	cvs.put(W / 2, H / 2, 'O', pr_axes + 1, CLR_AXES);
 
-  auto [ox, oy] = mp.map({0, 0});
+	auto [ox, oy] = mp.map({ 0, 0 });
-  auto [ax1, ay1] = mp.map(a);
+	auto [ax1, ay1] = mp.map(a);
-  auto [bx1, by1] = mp.map(b);
+	auto [bx1, by1] = mp.map(b);
-  auto [px1, py1] = mp.map(p);
+	auto [px1, py1] = mp.map(p);
 
-  int pr_a = 3, pr_b = 3, pr_p = 4;
+	int pr_a = 3, pr_b = 3, pr_p = 4;
 
-  cvs.line_dir(ox, oy, ax1, ay1, pr_a, CLR_A);
+	cvs.line_dir(ox, oy, ax1, ay1, pr_a, CLR_A);
-  cvs.line_dir(ox, oy, bx1, by1, pr_b, CLR_B);
+	cvs.line_dir(ox, oy, bx1, by1, pr_b, CLR_B);
-  cvs.line_dir(ox, oy, px1, py1, pr_p, CLR_P);
+	cvs.line_dir(ox, oy, px1, py1, pr_p, CLR_P);
 
-  if (!a.approx_equal(p) && b.magnitude() > 0) {
+	if (!a.approx_equal(p) && b.magnitude() > 0) {
-    int steps = std::max(std::abs(ax1 - px1), std::abs(ay1 - py1));
+		int steps = std::max(std::abs(ax1 - px1), std::abs(ay1 - py1));
-    for (int i = 0; i <= steps; ++i) {
+		for (int i = 0; i <= steps; ++i) {
-      double t = steps ? double(i) / steps : 0.0;
+			double t = steps ? double(i) / steps : 0.0;
-      int x = (int)std::llround(px1 + t * (ax1 - px1));
+			int x = (int)std::llround(px1 + t * (ax1 - px1));
-      int y = (int)std::llround(py1 + t * (ay1 - py1));
+			int y = (int)std::llround(py1 + t * (ay1 - py1));
-      if (i % 2 == 0)
+			if (i % 2 == 0)
-        cvs.put(x, y, '.', 2, CLR_DOT);
+				cvs.put(x, y, '.', 2, CLR_DOT);
-    }
+		}
-  }
+	}
 
-  auto place_label = [&](Vec2 v, const char *txt, uint8_t c, int pr) {
+	auto place_label = [&](Vec2 v, const char *txt, uint8_t c, int pr) {
-    auto tip = mp.map(v * 1.05f);
+		auto tip = mp.map(v * 1.05f);
-    for (int i = 0; txt[i]; ++i)
+		for (int i = 0; txt[i]; ++i)
-      cvs.put(tip.first + i, tip.second, txt[i], pr, c);
+			cvs.put(tip.first + i, tip.second, txt[i], pr, c);
-  };
+	};
-  place_label(a, "A", CLR_A, pr_a + 1);
+	place_label(a, "A", CLR_A, pr_a + 1);
-  place_label(b, "B", CLR_B, pr_b + 1);
+	place_label(b, "B", CLR_B, pr_b + 1);
-  place_label(p, "P", CLR_P, pr_p + 1);
+	place_label(p, "P", CLR_P, pr_p + 1);
 
-  std::println("Legend: axes(gray), a(green), b(yellow), p=proj(magenta), "
+	std::println("Legend: axes(gray), a(green), b(yellow), p=proj(magenta), "
-               "dotted=cross-perp\n");
+	             "dotted=cross-perp\n");
-  cvs.flush();
+	cvs.flush();
-  return 0;
+	return 0;
 }

examples/random_steps.cpp

@@ -22,22 +22,23 @@
 #include <print>
 #include <random>
 
-#include <smath.hpp>
+#include <smath/smath.hpp>
 
-auto main() -> int {
+auto main() -> int
-  using namespace smath;
+{
+	using namespace smath;
 
-  Vec2d point;
+	Vec2d point;
-  std::random_device rd;
+	std::random_device rd;
-  std::mt19937 rng{rd()};
+	std::mt19937 rng { rd() };
-  std::uniform_real_distribution<> dis(-5, 5);
+	std::uniform_real_distribution<> dis(-5, 5);
-  int i = 0;
+	int i = 0;
-  do {
+	do {
-    Vec2d add{dis(rng), dis(rng)};
+		Vec2d add { dis(rng), dis(rng) };
-    auto const n = point + add;
+		auto const n = point + add;
-    std::println("{}: {:.2f} + {:.2f} -> {:.2f}", i, point, add, n);
+		std::println("{}: {:.2f} + {:.2f} -> {:.2f}", i, point, add, n);
-    point = n;
+		point = n;
 
-    i++;
+		i++;
-  } while (i < 15);
+	} while (i < 15);
 }

examples/tour.cpp

@@ -17,53 +17,54 @@
 
 // #define SMATH_IMPLICIT_CONVERSIONS
 
-#include <smath.hpp>
+#include <smath/smath.hpp>
 
-int main() {
+int main()
-  using namespace smath;
+{
-  Vec3 v{1, 2, 3};
+	using namespace smath;
-  std::println("v: {}", v);
+	Vec3 v { 1, 2, 3 };
-  auto v2 = swizzle<"zyx">(v);
+	std::println("v: {}", v);
-  std::println("v2: {}", v2);
+	auto v2 = swizzle<"zyx">(v);
-  std::println("+: {}", v + v2);
+	std::println("v2: {}", v2);
-  std::println("-: {}", v - v2);
+	std::println("+: {}", v + v2);
-  std::println("*: {}", v * v2);
+	std::println("-: {}", v - v2);
-  std::println("/: {}", v / v2);
+	std::println("*: {}", v * v2);
-  std::println("dot: {}", v.dot(v2));
+	std::println("/: {}", v / v2);
-  std::println("rrggbb: {}", swizzle<"rrggbb">(v));
+	std::println("dot: {}", v.dot(v2));
-  std::println("Magnitude: {}", v.magnitude());
+	std::println("rrggbb: {}", swizzle<"rrggbb">(v));
-  std::println("Normalized: {}", v.normalized());
+	std::println("Magnitude: {}", v.magnitude());
-  std::println("(alias) Unit: {}", v.unit());
+	std::println("Normalized: {}", v.normalized());
-  std::println("(alias) Normalize: {}", v.normalize());
+	std::println("(alias) Unit: {}", v.unit());
-  std::println("(alias) Length: {}", v.length());
+	std::println("(alias) Normalize: {}", v.normalize());
-  std::println("std::get<1>(v): {}", std::get<1>(v));
+	std::println("(alias) Length: {}", v.length());
-  auto [x, y, z] = v;
+	std::println("std::get<1>(v): {}", std::get<1>(v));
-  std::println("Bindings: [{}, {}, {}]", x, y, z);
+	auto [x, y, z] = v;
-  float x1{}, y1{}, z1{};
+	std::println("Bindings: [{}, {}, {}]", x, y, z);
-  v.unpack(x1, y1, z1);
+	float x1 {}, y1 {}, z1 {};
-  std::println("Unpacked: {}, {}, {}", x1, y1, z1);
+	v.unpack(x1, y1, z1);
+	std::println("Unpacked: {}, {}, {}", x1, y1, z1);
 
-  // Let's mix and match!
+	// Let's mix and match!
-  Vec<6> v3(v, 7, swizzle<"zy">(v2));
+	Vec<6> v3(v, 7, swizzle<"zy">(v2));
-  std::println("{{v, 7, XZ(v2)}}: {}", v3);
+	std::println("{{v, 7, XZ(v2)}}: {}", v3);
 
-  // Scalar operations
+	// Scalar operations
-  std::println("v + 3: {}", v + 3);
+	std::println("v + 3: {}", v + 3);
-  std::println("v - 3: {}", v - 3);
+	std::println("v - 3: {}", v - 3);
-  std::println("v * 3: {}", v * 3);
+	std::println("v * 3: {}", v * 3);
-  std::println("v / 3: {}", v / 3);
+	std::println("v / 3: {}", v / 3);
 
-  std::println("3 + v: {}", 3 + v);
+	std::println("3 + v: {}", 3 + v);
-  std::println("3 - v: {}", 3 - v);
+	std::println("3 - v: {}", 3 - v);
-  std::println("3 * v: {}", 3 * v);
+	std::println("3 * v: {}", 3 * v);
-  std::println("3 / v: {}", 3 / v);
+	std::println("3 / v: {}", 3 / v);
 
-  // Casting
+	// Casting
-  auto v4 = static_cast<Vec3d>(v);
+	auto v4 = static_cast<Vec3d>(v);
 #ifdef SMATH_IMPLICIT_CONVERSIONS
-  Vec3d v5 = v;
+	Vec3d v5 = v;
 #else
-  Vec3d v5 = static_cast<Vec3d>(v);
+	Vec3d v5 = static_cast<Vec3d>(v);
 #endif // SMATH_IMPLICIT_CONVERSIONS
-  std::println("Are v4 and v5 same? {}", v4 == v5);
+	std::println("Are v4 and v5 same? {}", v4 == v5);
 }

flake.nix

@@ -27,6 +27,7 @@
             clang-tools
             lldb
             pkg-config
+            doxygen
           ];
         };
 
@@ -55,10 +56,14 @@
             })
           ];
 
+          extraCMakeFlags = [
+            "-DSMATH_INSTALL_INTEROP_HEADERS=ON"
+          ];
+
           installPhase = ''
             runHook preInstall
-            mkdir -p $out/include
+            mkdir -p $out/include/
-            cp ../include/smath.hpp $out/include/
+            cp -r ../include/smath/ $out/include/
             runHook postInstall
           '';
 

include/smath/interop/eigen.hpp

@@ -0,0 +1,131 @@
+/*
+ * smath - Single-file linear algebra math library for C++23.
+ *
+ * Copyright 2025 Slendi <slendi@socopon.com>
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *   http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#pragma once
+
+#include <Eigen/Core>
+#include <Eigen/Geometry>
+
+#include <smath/smath.hpp>
+
+namespace smath
+{
+
+template<class Scalar, int N, int Options, int MaxRows, int MaxCols>
+requires(std::is_arithmetic_v<Scalar> && (N > 0))
+struct interop_adapter<Eigen::Matrix<Scalar, N, 1, Options, MaxRows, MaxCols>>
+{
+	using smath_type = Vec<static_cast<std::size_t>(N), Scalar>;
+
+	static constexpr auto to_smath(
+	    Eigen::Matrix<Scalar, N, 1, Options, MaxRows, MaxCols> const &v)
+	    -> smath_type
+	{
+		smath_type out {};
+		for (std::size_t i = 0; i < static_cast<std::size_t>(N); ++i)
+			out[i] = v(static_cast<int>(i));
+		return out;
+	}
+
+	static constexpr auto from_smath(smath_type const &v)
+	    -> Eigen::Matrix<Scalar, N, 1, Options, MaxRows, MaxCols>
+	{
+		Eigen::Matrix<Scalar, N, 1, Options, MaxRows, MaxCols> out;
+		for (std::size_t i = 0; i < static_cast<std::size_t>(N); ++i)
+			out(static_cast<int>(i)) = v[i];
+		return out;
+	}
+};
+
+template<class Scalar, int N, int Options, int MaxRows, int MaxCols>
+requires(std::is_arithmetic_v<Scalar> && (N > 0))
+struct interop_adapter<Eigen::Matrix<Scalar, 1, N, Options, MaxRows, MaxCols>>
+{
+	using smath_type = Vec<static_cast<std::size_t>(N), Scalar>;
+
+	static constexpr auto to_smath(
+	    Eigen::Matrix<Scalar, 1, N, Options, MaxRows, MaxCols> const &v)
+	    -> smath_type
+	{
+		smath_type out {};
+		for (std::size_t i = 0; i < static_cast<std::size_t>(N); ++i)
+			out[i] = v(static_cast<int>(i));
+		return out;
+	}
+
+	static constexpr auto from_smath(smath_type const &v)
+	    -> Eigen::Matrix<Scalar, 1, N, Options, MaxRows, MaxCols>
+	{
+		Eigen::Matrix<Scalar, 1, N, Options, MaxRows, MaxCols> out;
+		for (std::size_t i = 0; i < static_cast<std::size_t>(N); ++i)
+			out(static_cast<int>(i)) = v[i];
+		return out;
+	}
+};
+
+template<class Scalar, int R, int C, int Options, int MaxRows, int MaxCols>
+requires(std::is_arithmetic_v<Scalar> && (R > 1) && (C > 1))
+struct interop_adapter<Eigen::Matrix<Scalar, R, C, Options, MaxRows, MaxCols>>
+{
+	using smath_type
+	    = Mat<static_cast<std::size_t>(R), static_cast<std::size_t>(C), Scalar>;
+
+	static constexpr auto to_smath(
+	    Eigen::Matrix<Scalar, R, C, Options, MaxRows, MaxCols> const &m)
+	    -> smath_type
+	{
+		smath_type out {};
+		for (std::size_t c = 0; c < static_cast<std::size_t>(C); ++c) {
+			for (std::size_t r = 0; r < static_cast<std::size_t>(R); ++r)
+				out[r, c] = m(static_cast<int>(r), static_cast<int>(c));
+		}
+		return out;
+	}
+
+	static constexpr auto from_smath(smath_type const &m)
+	    -> Eigen::Matrix<Scalar, R, C, Options, MaxRows, MaxCols>
+	{
+		Eigen::Matrix<Scalar, R, C, Options, MaxRows, MaxCols> out;
+		for (std::size_t c = 0; c < static_cast<std::size_t>(C); ++c) {
+			for (std::size_t r = 0; r < static_cast<std::size_t>(R); ++r)
+				out(static_cast<int>(r), static_cast<int>(c)) = m[r, c];
+		}
+		return out;
+	}
+};
+
+template<class Scalar, int Options>
+requires std::is_arithmetic_v<Scalar>
+struct interop_adapter<Eigen::Quaternion<Scalar, Options>>
+{
+	using smath_type = Quaternion<Scalar>;
+
+	static constexpr auto to_smath(Eigen::Quaternion<Scalar, Options> const &q)
+	    -> smath_type
+	{
+		return { q.x(), q.y(), q.z(), q.w() };
+	}
+
+	static constexpr auto from_smath(smath_type const &q)
+	    -> Eigen::Quaternion<Scalar, Options>
+	{
+		return { q.w(), q.x(), q.y(), q.z() };
+	}
+};
+
+} // namespace smath

include/smath/interop/glm.hpp

@@ -0,0 +1,103 @@
+/*
+ * smath - Single-file linear algebra math library for C++23.
+ *
+ * Copyright 2025 Slendi <slendi@socopon.com>
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *   http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#pragma once
+
+#include <glm/glm.hpp>
+#include <glm/gtc/quaternion.hpp>
+
+#include <smath/smath.hpp>
+
+namespace smath
+{
+
+template<glm::length_t L, class T, glm::qualifier Q>
+requires std::is_arithmetic_v<T>
+struct interop_adapter<glm::vec<L, T, Q>>
+{
+	using smath_type = Vec<static_cast<std::size_t>(L), T>;
+
+	static constexpr auto to_smath(glm::vec<L, T, Q> const &v) -> smath_type
+	{
+		smath_type out {};
+		for (std::size_t i = 0; i < static_cast<std::size_t>(L); ++i)
+			out[i] = v[static_cast<glm::length_t>(i)];
+		return out;
+	}
+
+	static constexpr auto from_smath(smath_type const &v) -> glm::vec<L, T, Q>
+	{
+		glm::vec<L, T, Q> out {};
+		for (std::size_t i = 0; i < static_cast<std::size_t>(L); ++i)
+			out[static_cast<glm::length_t>(i)] = v[i];
+		return out;
+	}
+};
+
+template<glm::length_t C, glm::length_t R, class T, glm::qualifier Q>
+requires std::is_arithmetic_v<T>
+struct interop_adapter<glm::mat<C, R, T, Q>>
+{
+	using smath_type
+	    = Mat<static_cast<std::size_t>(R), static_cast<std::size_t>(C), T>;
+
+	static constexpr auto to_smath(glm::mat<C, R, T, Q> const &m) -> smath_type
+	{
+		smath_type out {};
+		for (std::size_t c = 0; c < static_cast<std::size_t>(C); ++c) {
+			for (std::size_t r = 0; r < static_cast<std::size_t>(R); ++r) {
+				out[r, c] = m[static_cast<glm::length_t>(c)]
+				             [static_cast<glm::length_t>(r)];
+			}
+		}
+		return out;
+	}
+
+	static constexpr auto from_smath(smath_type const &m)
+	    -> glm::mat<C, R, T, Q>
+	{
+		glm::mat<C, R, T, Q> out {};
+		for (std::size_t c = 0; c < static_cast<std::size_t>(C); ++c) {
+			for (std::size_t r = 0; r < static_cast<std::size_t>(R); ++r) {
+				out[static_cast<glm::length_t>(c)]
+				   [static_cast<glm::length_t>(r)]
+				    = m[r, c];
+			}
+		}
+		return out;
+	}
+};
+
+template<class T, glm::qualifier Q>
+requires std::is_arithmetic_v<T>
+struct interop_adapter<glm::qua<T, Q>>
+{
+	using smath_type = Quaternion<T>;
+
+	static constexpr auto to_smath(glm::qua<T, Q> const &q) -> smath_type
+	{
+		return { q.x, q.y, q.z, q.w };
+	}
+
+	static constexpr auto from_smath(smath_type const &q) -> glm::qua<T, Q>
+	{
+		return { q.w(), q.x(), q.y(), q.z() };
+	}
+};
+
+} // namespace smath

include/smath/interop/hmm.hpp

@@ -0,0 +1,125 @@
+/*
+ * smath - Single-file linear algebra math library for C++23.
+ *
+ * Copyright 2025 Slendi <slendi@socopon.com>
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *   http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#pragma once
+
+#include <HandmadeMath.h>
+
+#include <smath/smath.hpp>
+
+namespace smath
+{
+
+template<> struct interop_adapter<HMM_Vec2>
+{
+	using smath_type = Vec<2, float>;
+
+	static constexpr auto to_smath(HMM_Vec2 const &v) -> smath_type
+	{
+		return { v.X, v.Y };
+	}
+
+	static constexpr auto from_smath(smath_type const &v) -> HMM_Vec2
+	{
+		return { v.x(), v.y() };
+	}
+};
+
+template<> struct interop_adapter<HMM_Vec3>
+{
+	using smath_type = Vec<3, float>;
+
+	static constexpr auto to_smath(HMM_Vec3 const &v) -> smath_type
+	{
+		return { v.X, v.Y, v.Z };
+	}
+
+	static constexpr auto from_smath(smath_type const &v) -> HMM_Vec3
+	{
+		return { v.x(), v.y(), v.z() };
+	}
+};
+
+template<> struct interop_adapter<HMM_Vec4>
+{
+	using smath_type = Vec<4, float>;
+
+	static constexpr auto to_smath(HMM_Vec4 const &v) -> smath_type
+	{
+		return { v.X, v.Y, v.Z, v.W };
+	}
+
+	static constexpr auto from_smath(smath_type const &v) -> HMM_Vec4
+	{
+		return { v.x(), v.y(), v.z(), v.w() };
+	}
+};
+
+template<> struct interop_adapter<HMM_Mat4>
+{
+	using smath_type = Mat<4, 4, float>;
+
+	static constexpr auto to_smath(HMM_Mat4 const &m) -> smath_type
+	{
+		smath_type out {};
+		out[0, 0] = m.Columns[0].X;
+		out[1, 0] = m.Columns[0].Y;
+		out[2, 0] = m.Columns[0].Z;
+		out[3, 0] = m.Columns[0].W;
+		out[0, 1] = m.Columns[1].X;
+		out[1, 1] = m.Columns[1].Y;
+		out[2, 1] = m.Columns[1].Z;
+		out[3, 1] = m.Columns[1].W;
+		out[0, 2] = m.Columns[2].X;
+		out[1, 2] = m.Columns[2].Y;
+		out[2, 2] = m.Columns[2].Z;
+		out[3, 2] = m.Columns[2].W;
+		out[0, 3] = m.Columns[3].X;
+		out[1, 3] = m.Columns[3].Y;
+		out[2, 3] = m.Columns[3].Z;
+		out[3, 3] = m.Columns[3].W;
+		return out;
+	}
+
+	static constexpr auto from_smath(smath_type const &m) -> HMM_Mat4
+	{
+		HMM_Mat4 out {};
+		out.Columns[0] = { m[0, 0], m[1, 0], m[2, 0], m[3, 0] };
+		out.Columns[1] = { m[0, 1], m[1, 1], m[2, 1], m[3, 1] };
+		out.Columns[2] = { m[0, 2], m[1, 2], m[2, 2], m[3, 2] };
+		out.Columns[3] = { m[0, 3], m[1, 3], m[2, 3], m[3, 3] };
+		return out;
+	}
+};
+
+template<> struct interop_adapter<HMM_Quat>
+{
+	using smath_type = Quaternion<float>;
+
+	static constexpr auto to_smath(HMM_Quat const &q) -> smath_type
+	{
+		return { q.X, q.Y, q.Z, q.W };
+	}
+
+	static constexpr auto from_smath(smath_type const &q) -> HMM_Quat
+	{
+		return { q.x(), q.y(), q.z(), q.w() };
+	}
+};
+
+} // namespace smath

include/smath/interop/raylib.hpp

@@ -0,0 +1,122 @@
+/*
+ * smath - Single-file linear algebra math library for C++23.
+ *
+ * Copyright 2025 Slendi <slendi@socopon.com>
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *   http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#pragma once
+
+#include <raylib.h>
+
+#include <smath/smath.hpp>
+
+namespace smath
+{
+
+template<> struct interop_adapter<Vector2>
+{
+	using smath_type = Vec<2, float>;
+
+	static constexpr auto to_smath(Vector2 const &v) -> smath_type
+	{
+		return { v.x, v.y };
+	}
+
+	static constexpr auto from_smath(smath_type const &v) -> Vector2
+	{
+		return { v.x(), v.y() };
+	}
+};
+
+template<> struct interop_adapter<Vector3>
+{
+	using smath_type = Vec<3, float>;
+
+	static constexpr auto to_smath(Vector3 const &v) -> smath_type
+	{
+		return { v.x, v.y, v.z };
+	}
+
+	static constexpr auto from_smath(smath_type const &v) -> Vector3
+	{
+		return { v.x(), v.y(), v.z() };
+	}
+};
+
+template<> struct interop_adapter<Vector4>
+{
+	using smath_type = Vec<4, float>;
+
+	static constexpr auto to_smath(Vector4 const &v) -> smath_type
+	{
+		return { v.x, v.y, v.z, v.w };
+	}
+
+	static constexpr auto from_smath(smath_type const &v) -> Vector4
+	{
+		return { v.x(), v.y(), v.z(), v.w() };
+	}
+};
+
+template<> struct interop_adapter<Matrix>
+{
+	using smath_type = Mat<4, 4, float>;
+
+	static constexpr auto to_smath(Matrix const &m) -> smath_type
+	{
+		smath_type out {};
+		out[0, 0] = m.m0;
+		out[1, 0] = m.m1;
+		out[2, 0] = m.m2;
+		out[3, 0] = m.m3;
+		out[0, 1] = m.m4;
+		out[1, 1] = m.m5;
+		out[2, 1] = m.m6;
+		out[3, 1] = m.m7;
+		out[0, 2] = m.m8;
+		out[1, 2] = m.m9;
+		out[2, 2] = m.m10;
+		out[3, 2] = m.m11;
+		out[0, 3] = m.m12;
+		out[1, 3] = m.m13;
+		out[2, 3] = m.m14;
+		out[3, 3] = m.m15;
+		return out;
+	}
+
+	static constexpr auto from_smath(smath_type const &m) -> Matrix
+	{
+		return {
+			m[0, 0],
+			m[0, 1],
+			m[0, 2],
+			m[0, 3],
+			m[1, 0],
+			m[1, 1],
+			m[1, 2],
+			m[1, 3],
+			m[2, 0],
+			m[2, 1],
+			m[2, 2],
+			m[2, 3],
+			m[3, 0],
+			m[3, 1],
+			m[3, 2],
+			m[3, 3],
+		};
+	}
+};
+
+} // namespace smath

include/smath/interop/sfml.hpp

@@ -0,0 +1,63 @@
+/*
+ * smath - Single-file linear algebra math library for C++23.
+ *
+ * Copyright 2025 Slendi <slendi@socopon.com>
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *   http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#pragma once
+
+#include <SFML/System/Vector2.hpp>
+#include <SFML/System/Vector3.hpp>
+
+#include <smath/smath.hpp>
+
+namespace smath
+{
+
+template<class T>
+requires std::is_arithmetic_v<T>
+struct interop_adapter<sf::Vector2<T>>
+{
+	using smath_type = Vec<2, T>;
+
+	static constexpr auto to_smath(sf::Vector2<T> const &v) -> smath_type
+	{
+		return { v.x, v.y };
+	}
+
+	static constexpr auto from_smath(smath_type const &v) -> sf::Vector2<T>
+	{
+		return { v.x(), v.y() };
+	}
+};
+
+template<class T>
+requires std::is_arithmetic_v<T>
+struct interop_adapter<sf::Vector3<T>>
+{
+	using smath_type = Vec<3, T>;
+
+	static constexpr auto to_smath(sf::Vector3<T> const &v) -> smath_type
+	{
+		return { v.x, v.y, v.z };
+	}
+
+	static constexpr auto from_smath(smath_type const &v) -> sf::Vector3<T>
+	{
+		return { v.x(), v.y(), v.z() };
+	}
+};
+
+} // namespace smath

include/smath/smath.hpp

@@ -17,6 +17,7 @@
  *
  * You can define the following macros to change functionality:
  * - SMATH_IMPLICIT_CONVERSIONS
+ * - SMATH_ANGLE_UNIT (=rad, =deg, =turns)
  */
 
 #pragma once
@@ -32,20 +33,23 @@
 #include <type_traits>
 
 #ifndef SMATH_ANGLE_UNIT
-#	define SMATH_ANGLE_UNIT rad
+#define SMATH_ANGLE_UNIT rad
 #endif // SMATH_ANGLE_UNIT
 
-namespace smath {
+namespace smath
+{
 
 template<std::size_t N, typename T>
-requires std::is_arithmetic_v<T> struct Vec;
+requires std::is_arithmetic_v<T>
+struct Vec;
 
-namespace detail {
+namespace detail
+{
 
 #define SMATH_STR(x) #x
 #define SMATH_XSTR(x) SMATH_STR(x)
 
-consteval bool streq(const char *a, const char *b)
+consteval auto streq(const char *a, const char *b) -> bool
 {
 	for (;; ++a, ++b) {
 		if (*a != *b)
@@ -55,13 +59,14 @@ consteval bool streq(const char *a, const char *b)
 	}
 }
 
-enum class AngularUnit {
+enum class AngularUnit
+{
 	Radians,
 	Degrees,
 	Turns,
 };
 
-consteval std::optional<AngularUnit> parse_unit(char const *s)
+consteval auto parse_unit(char const *s) -> std::optional<AngularUnit>
 {
 	if (streq(s, "rad"))
 		return AngularUnit::Radians;
@@ -76,7 +81,8 @@ constexpr auto SMATH_ANGLE_UNIT_ID = parse_unit(SMATH_XSTR(SMATH_ANGLE_UNIT));
 static_assert(SMATH_ANGLE_UNIT_ID != std::nullopt,
     "Invalid SMATH_ANGLE_UNIT. Should be rad, deg, or turns.");
 
-template<std::size_t N> struct FixedString {
+template<std::size_t N> struct FixedString
+{
 	char data[N] {};
 	static constexpr std::size_t size = N - 1;
 	constexpr FixedString(char const (&s)[N])
@@ -86,8 +92,10 @@ template<std::size_t N> struct FixedString {
 	}
 	constexpr char operator[](std::size_t i) const { return data[i]; }
 };
-template<class X> struct is_Vec : std::false_type { };
+template<class X> struct is_Vec : std::false_type
-template<std::size_t M, class U> struct is_Vec<Vec<M, U>> : std::true_type { };
+{ };
+template<std::size_t M, class U> struct is_Vec<Vec<M, U>> : std::true_type
+{ };
 template<class X>
 inline constexpr bool is_Vec_v = is_Vec<std::remove_cvref_t<X>>::value;
 template<class X>
@@ -95,7 +103,8 @@ inline constexpr bool is_scalar_v
     = std::is_arithmetic_v<std::remove_cvref_t<X>>;
 template<class X> struct Vec_size;
 template<std::size_t M, class U>
-struct Vec_size<Vec<M, U>> : std::integral_constant<std::size_t, M> { };
+struct Vec_size<Vec<M, U>> : std::integral_constant<std::size_t, M>
+{ };
 
 template<class T> constexpr auto pack_unorm8(T v) -> std::uint8_t
 {
@@ -144,9 +153,16 @@ template<class T> constexpr auto unpack_snorm8(std::int8_t b) -> T
 } // namespace detail
 
 template<std::size_t N, typename T = float>
-requires std::is_arithmetic_v<T> struct Vec : std::array<T, N> {
+requires std::is_arithmetic_v<T>
+/**
+ * @brief Fixed-size arithmetic vector.
+ * @tparam N Number of components.
+ * @tparam T Component type.
+ */
+struct Vec : std::array<T, N>
+{
 private:
-	template<class X> static consteval std::size_t extent()
+	template<class X> static consteval auto extent() -> std::size_t
 	{
 		if constexpr (detail::is_Vec_v<X>)
 			return detail::Vec_size<std::remove_cvref_t<X>>::value;
@@ -155,19 +171,24 @@ private:
 		else
 			return 0; // Should be unreachable
 	}
-	template<class... Args> static consteval std::size_t total_extent()
+	template<class... Args> static consteval auto total_extent() -> std::size_t
 	{
 		return (extent<Args>() + ... + 0);
 	}
 
 public:
 	// Constructors
+	/** @brief Constructs a zero-initialized vector. */
 	constexpr Vec() noexcept
 	{
 		for (auto &v : *this)
 			v = T(0);
 	}
 
+	/**
+	 * @brief Fills all components with the same scalar.
+	 * @param s Scalar value assigned to every component.
+	 */
 	explicit constexpr Vec(T const &s) noexcept
 	{
 		for (auto &v : *this)
@@ -178,6 +199,10 @@ public:
 	requires((detail::is_scalar_v<Args> || detail::is_Vec_v<Args>) && ...)
 	    && (total_extent<Args...>() == N)
 	    && (!(sizeof...(Args) == 1 && (detail::is_Vec_v<Args> && ...)))
+	/**
+	 * @brief Constructs from scalars and/or vectors whose flattened extent is
+	 * N.
+	 */
 	constexpr Vec(Args &&...args) noexcept
 	{
 		std::size_t i = 0;
@@ -188,12 +213,9 @@ public:
 	// NOTE: This can (probably) be improved with C++26 reflection in the
 	// future.
 #define VEC_ACC(component, req, idx) \
-	constexpr auto component() noexcept -> T & \
+	constexpr auto component() noexcept -> T &requires(N >= req) { \
-	requires(N >= req) \
-	{ \
 		return (*this)[idx]; \
-	} \
+	} constexpr auto component() const->T const & \
-	constexpr auto component() const -> T const & \
 	requires(N >= req) \
 	{ \
 		return (*this)[idx]; \
@@ -225,7 +247,11 @@ public:
 		((args = (*this)[Is]), ...);
 	}
 
-	template<class... Args> constexpr void unpack(Args &...args) noexcept
+	template<class... Args>
+	/**
+	 * @brief Unpacks components into output references.
+	 */
+	constexpr auto unpack(Args &...args) noexcept -> void
 	{
 		unpack_impl(std::index_sequence_for<Args...> {}, args...);
 	}
@@ -284,13 +310,13 @@ public:
 	VEC_OP(/)
 #undef VEC_OP
 #define VEC_OP_ASSIGN(sym) \
-	constexpr Vec &operator sym## = (Vec const &rhs) noexcept \
+	constexpr Vec &operator sym##=(Vec const &rhs) noexcept \
 	{ \
 		for (std::size_t i = 0; i < N; ++i) \
 			(*this)[i] sym## = rhs[i]; \
 		return *this; \
 	} \
-	constexpr Vec &operator sym## = (T const &s) noexcept \
+	constexpr Vec &operator sym##=(T const &s) noexcept \
 	{ \
 		for (std::size_t i = 0; i < N; ++i) \
 			(*this)[i] sym## = s; \
@@ -315,6 +341,7 @@ public:
 		return !(*this == v);
 	}
 
+	/** @brief Returns Euclidean magnitude. */
 	constexpr auto magnitude() const noexcept -> T
 	requires std::is_floating_point_v<T>
 	{
@@ -323,6 +350,7 @@ public:
 			total += v * v;
 		return std::sqrt(total);
 	}
+	/** @brief Alias for magnitude(). */
 	constexpr auto length() const noexcept -> T
 	requires std::is_floating_point_v<T>
 	{
@@ -331,6 +359,9 @@ public:
 
 	template<typename U = T>
 	requires std::is_floating_point_v<U>
+	/**
+	 * @brief Normalizes with zero fallback for very small magnitudes.
+	 */
 	constexpr auto normalized_safe(U eps = EPS_DEFAULT) const noexcept -> Vec
 	{
 		auto m = magnitude();
@@ -338,27 +369,32 @@ public:
 	}
 	template<typename U = T>
 	requires std::is_floating_point_v<U>
+	/** @brief Alias for normalized_safe(). */
 	constexpr auto normalize_safe(U eps = EPS_DEFAULT) const noexcept -> Vec
 	{
 		return normalized_safe(eps);
 	}
 
+	/** @brief Returns normalized vector; undefined for zero magnitude. */
 	[[nodiscard]] constexpr auto normalized() noexcept -> Vec<N, T>
 	requires std::is_floating_point_v<T>
 	{
 		return (*this) / this->magnitude();
 	}
+	/** @brief Alias for normalized(). */
 	[[nodiscard]] constexpr auto normalize() noexcept -> Vec<N, T>
 	requires std::is_floating_point_v<T>
 	{
 		return this->normalized();
 	}
+	/** @brief Alias for normalized(). */
 	[[nodiscard]] constexpr auto unit() noexcept -> Vec<N, T>
 	requires std::is_floating_point_v<T>
 	{
 		return this->normalized();
 	}
 
+	/** @brief Dot product with `other`. */
 	[[nodiscard]] constexpr auto dot(Vec<N, T> const &other) const noexcept -> T
 	{
 		T res = 0;
@@ -369,6 +405,13 @@ public:
 	}
 
 	static constexpr T EPS_DEFAULT = T(1e-6);
+
+	/**
+	 * @brief Approximate equality with per-component absolute epsilon.
+	 * @param rhs Vec to compare against.
+	 * @param eps Absolute tolerance per component.
+	 * @return `true` when all components are approximately equal.
+	 */
 	template<class U = T>
 	requires std::is_floating_point_v<U>
 	[[nodiscard]] constexpr auto approx_equal(
@@ -382,6 +425,7 @@ public:
 	}
 
 	template<class U = T>
+	/** @brief Magnitude computed in promoted type `U` (or `double`). */
 	constexpr auto magnitude_promoted() const noexcept
 	requires std::is_floating_point_v<T>
 	{
@@ -393,19 +437,25 @@ public:
 	}
 
 	template<typename U = T>
-	requires(N == 3) constexpr auto cross(Vec const &r) const noexcept -> Vec
+	requires(N == 3)
+	/** @brief 3D cross product. */
+	constexpr auto cross(Vec const &r) const noexcept -> Vec
 	{
-		return { (*this)[1] * r[2] - (*this)[2] * r[1],
+		return {
+			(*this)[1] * r[2] - (*this)[2] * r[1],
 			(*this)[2] * r[0] - (*this)[0] * r[2],
-			(*this)[0] * r[1] - (*this)[1] * r[0] };
+			(*this)[0] * r[1] - (*this)[1] * r[0],
+		};
 	}
 
+	/** @brief Distance between this vector and `r`. */
 	constexpr auto distance(Vec const &r) const noexcept -> T
 	requires std::is_floating_point_v<T>
 	{
 		return (*this - r).magnitude();
 	}
 
+	/** @brief Projection of this vector onto `n`. */
 	constexpr auto project_onto(Vec const &n) const noexcept -> Vec
 	requires std::is_floating_point_v<T>
 	{
@@ -425,8 +475,8 @@ public:
 
 	template<class U>
 	requires(std::is_arithmetic_v<U> && N >= 1)
-	constexpr explicit(!std::is_convertible_v<T, U>)
+	constexpr explicit(
-	operator Vec<N, U>() const noexcept
+	    !std::is_convertible_v<T, U>) operator Vec<N, U>() const noexcept
 	{
 		Vec<N, U> r {};
 		for (std::size_t i = 0; i < N; ++i)
@@ -444,26 +494,26 @@ public:
 	}
 
 private:
-	constexpr void fill_one(std::size_t &i, T const &v) noexcept
+	constexpr auto fill_one(std::size_t &i, T const &v) noexcept -> void
 	{
 		(*this)[i++] = v;
 	}
 #ifdef SMATH_IMPLICIT_CONVERSIONS
 	template<class U>
-	requires std::is_arithmetic_v<U> && (!std::is_same_v<U, T>)constexpr void
+	requires std::is_arithmetic_v<U> && (!std::is_same_v<U, T>)
-	fill_one(std::size_t &i, const U &v) noexcept
+	constexpr auto fill_one(std::size_t &i, const U &v) noexcept -> void
 	{
 		(*this)[i++] = static_cast<T>(v);
 	}
 	template<std::size_t M, class U>
-	constexpr void fill_one(std::size_t &i, Vec<M, U> const &v) noexcept
+	constexpr auto fill_one(std::size_t &i, Vec<M, U> const &v) noexcept -> void
 	{
 		for (std::size_t k = 0; k < M; ++k)
 			(*this)[i++] = static_cast<T>(v[k]);
 	}
 #endif // SMATH_IMPLICIT_CONVERSIONS
 	template<std::size_t M>
-	constexpr void fill_one(std::size_t &i, const Vec<M, T> &v) noexcept
+	constexpr auto fill_one(std::size_t &i, const Vec<M, T> &v) noexcept -> void
 	{
 		for (std::size_t k = 0; k < M; ++k)
 			(*this)[i++] = static_cast<T>(v[k]);
@@ -495,9 +545,15 @@ constexpr T const &&get(Vec<N, T> const &&v) noexcept
 
 template<std::size_t N, typename T = float>
 requires std::is_arithmetic_v<T>
+/**
+ * @brief Return type used by swizzle operations.
+ *
+ * Produces `T` when `N == 1`, otherwise `Vec<N, T>`.
+ */
 using VecOrScalar = std::conditional_t<N == 1, T, Vec<N, T>>;
 
-namespace detail {
+namespace detail
+{
 
 consteval auto char_to_idx(char c) -> std::size_t
 {
@@ -567,22 +623,39 @@ concept ValidSwizzle
 
 template<detail::FixedString S, std::size_t N, typename T>
 requires detail::ValidSwizzle<S, N>
+/**
+ * @brief Compile-time swizzle selection.
+ * @tparam S Swizzle pattern like `"xy"`, `"rgba"`, or `"stp"`.
+ * @param v Source vector.
+ * @return Swizzled scalar/vector according to `S`.
+ */
 constexpr auto swizzle(Vec<N, T> const &v) -> VecOrScalar<S.size, T>
 {
 	return detail::swizzle_impl<S>(v, std::make_index_sequence<S.size> {});
 }
 
+/** @brief 2D float vector alias. */
 using Vec2 = Vec<2>;
+/** @brief 3D float vector alias. */
 using Vec3 = Vec<3>;
+/** @brief 4D float vector alias. */
 using Vec4 = Vec<4>;
 
+/** @brief 2D double vector alias. */
 using Vec2d = Vec<2, double>;
+/** @brief 3D double vector alias. */
 using Vec3d = Vec<3, double>;
+/** @brief 4D double vector alias. */
 using Vec4d = Vec<4, double>;
 
 template<class T>
 using angle_ret_t = std::conditional_t<std::is_same_v<T, float>, float, double>;
 
+/**
+ * @brief Convert degrees to the configured angle unit.
+ * @param value Angle in degrees.
+ * @return Angle in `SMATH_ANGLE_UNIT`.
+ */
 template<class T> constexpr auto deg(T value)
 {
 	using R = angle_ret_t<T>;
@@ -597,6 +670,11 @@ template<class T> constexpr auto deg(T value)
 	}
 }
 
+/**
+ * @brief Convert radians to the configured angle unit.
+ * @param value Angle in radians.
+ * @return Angle in `SMATH_ANGLE_UNIT`.
+ */
 template<class T> constexpr auto rad(T value)
 {
 	using R = angle_ret_t<T>;
@@ -612,6 +690,11 @@ template<class T> constexpr auto rad(T value)
 	}
 }
 
+/**
+ * @brief Convert turns to the configured angle unit.
+ * @param value Angle in turns.
+ * @return Angle in `SMATH_ANGLE_UNIT`.
+ */
 template<class T> constexpr auto turns(T value)
 {
 	using R = angle_ret_t<T>;
@@ -627,14 +710,22 @@ template<class T> constexpr auto turns(T value)
 	}
 }
 template<std::size_t R, std::size_t C, typename T>
-requires std::is_arithmetic_v<T> struct Mat;
+requires std::is_arithmetic_v<T>
+struct Mat;
 
-template<class T> struct Quaternion : Vec<4, T> {
+/**
+ * @brief Quaternion represented as `(x, y, z, w)`.
+ * @tparam T Component type.
+ */
+template<class T> struct Quaternion : Vec<4, T>
+{
 	using Base = Vec<4, T>;
 	using Base::Base;
 	using Base::operator=;
 
+	/** @brief Returns this quaternion as its vector base type. */
 	constexpr Base &vec() noexcept { return *this; }
+	/** @brief Returns this quaternion as its vector base type. */
 	constexpr Base const &vec() const noexcept { return *this; }
 
 	constexpr T &x() noexcept { return Base::x(); }
@@ -646,6 +737,11 @@ template<class T> struct Quaternion : Vec<4, T> {
 	constexpr T &w() noexcept { return Base::w(); }
 	constexpr T const &w() const noexcept { return Base::w(); }
 
+	/**
+	 * @brief Hamilton product.
+	 * @param rhs Right-hand quaternion.
+	 * @return Product quaternion.
+	 */
 	constexpr auto operator*(Quaternion const &rhs) const noexcept -> Quaternion
 	{
 		Quaternion r;
@@ -663,6 +759,10 @@ template<class T> struct Quaternion : Vec<4, T> {
 		return r;
 	}
 
+	/**
+	 * @brief Converts this quaternion to a 4x4 rotation matrix.
+	 * @return Homogeneous rotation matrix.
+	 */
 	[[nodiscard]] constexpr auto as_matrix() const noexcept -> Mat<4, 4, T>
 	{
 		auto const xx = x() * x();
@@ -683,6 +783,12 @@ template<class T> struct Quaternion : Vec<4, T> {
 		};
 	}
 
+	/**
+	 * @brief Builds a quaternion from axis-angle input.
+	 * @param axis Rotation axis.
+	 * @param angle Rotation angle in configured angle units.
+	 * @return Quaternion representing the rotation.
+	 */
 	[[nodiscard]] static constexpr auto from_axis_angle(
 	    Vec<3, T> const &axis, T const angle) noexcept -> Quaternion
 	{
@@ -702,6 +808,10 @@ template<class T> struct Quaternion : Vec<4, T> {
 
 template<class T>
 requires std::is_floating_point_v<T>
+/**
+ * @brief Packs `[0, 1]` RGBA values into 4x8-bit UNORM.
+ * @return Packed 32-bit value in RGBA byte order.
+ */
 [[nodiscard]] constexpr auto pack_unorm4x8(Vec<4, T> const &v) -> std::uint32_t
 {
 	std::uint32_t r = detail::pack_unorm8(v[0]);
@@ -714,6 +824,10 @@ requires std::is_floating_point_v<T>
 
 template<class T>
 requires std::is_floating_point_v<T>
+/**
+ * @brief Packs `[-1, 1]` RGBA values into 4x8-bit SNORM.
+ * @return Packed 32-bit value in RGBA byte order.
+ */
 [[nodiscard]] constexpr auto pack_snorm4x8(Vec<4, T> const &v) -> std::uint32_t
 {
 	std::uint32_t r = static_cast<std::uint8_t>(detail::pack_snorm8(v[0]));
@@ -726,6 +840,10 @@ requires std::is_floating_point_v<T>
 
 template<class T = float>
 requires std::is_floating_point_v<T>
+/**
+ * @brief Unpacks a 4x8-bit UNORM RGBA value.
+ * @return RGBA vector with components in `[0, 1]`.
+ */
 [[nodiscard]] constexpr auto unpack_unorm4x8(std::uint32_t packed) -> Vec<4, T>
 {
 	std::uint8_t r = static_cast<std::uint8_t>(packed & 0xFFu);
@@ -743,6 +861,10 @@ requires std::is_floating_point_v<T>
 
 template<class T = float>
 requires std::is_floating_point_v<T>
+/**
+ * @brief Unpacks a 4x8-bit SNORM RGBA value.
+ * @return RGBA vector with components in `[-1, 1]`.
+ */
 [[nodiscard]] constexpr auto unpack_snorm4x8(std::uint32_t packed) -> Vec<4, T>
 {
 	std::int8_t r = static_cast<std::int8_t>(packed & 0xFFu);
@@ -759,7 +881,15 @@ requires std::is_floating_point_v<T>
 }
 
 template<std::size_t R, std::size_t C, typename T = float>
-requires std::is_arithmetic_v<T> struct Mat : std::array<Vec<R, T>, C> {
+requires std::is_arithmetic_v<T>
+/**
+ * @brief Column-major matrix with `R` rows and `C` columns.
+ * @tparam R Row count.
+ * @tparam C Column count.
+ * @tparam T Scalar type.
+ */
+struct Mat : std::array<Vec<R, T>, C>
+{
 	using Base = std::array<Vec<R, T>, C>;
 	using Base::operator[];
 
@@ -774,12 +904,17 @@ requires std::is_arithmetic_v<T> struct Mat : std::array<Vec<R, T>, C> {
 		return col(column)[row];
 	}
 
+	/** @brief Constructs a zero matrix. */
 	constexpr Mat() noexcept
 	{
 		for (auto &col : *this)
 			col = Vec<R, T> {};
 	}
 
+	/**
+	 * @brief Constructs a diagonal matrix.
+	 * @param diag Value used for every diagonal component.
+	 */
 	constexpr explicit Mat(T const &diag) noexcept
 	requires(R == C)
 	{
@@ -792,10 +927,12 @@ requires std::is_arithmetic_v<T> struct Mat : std::array<Vec<R, T>, C> {
 	template<typename... Cols>
 	requires(sizeof...(Cols) == C
 	    && (std::same_as<std::remove_cvref_t<Cols>, Vec<R, T>> && ...))
-	constexpr Mat(Cols const &...cols) noexcept
+	/**
-	    : Base { cols... }
+	 * @brief Constructs from explicit column vectors.
-	{
+	 * @param cols Columns in column-major order.
-	}
+	 */
+	constexpr Mat(Cols const &...cols) noexcept : Base { cols... }
+	{ }
 
 	constexpr auto col(std::size_t j) noexcept -> Vec<R, T> &
 	{
@@ -890,6 +1027,13 @@ requires std::is_arithmetic_v<T> struct Mat : std::array<Vec<R, T>, C> {
 	}
 
 	static constexpr T EPS_DEFAULT = T(1e-6);
+
+	/**
+	 * @brief Approximate matrix equality check.
+	 * @param rhs Matrix to compare against.
+	 * @param eps Absolute tolerance per component.
+	 * @return `true` when all columns are approximately equal.
+	 */
 	template<class U = T>
 	requires std::is_floating_point_v<U>
 	[[nodiscard]] constexpr auto approx_equal(
@@ -901,6 +1045,7 @@ requires std::is_arithmetic_v<T> struct Mat : std::array<Vec<R, T>, C> {
 		return true;
 	}
 
+	/** @brief Returns the transposed matrix. */
 	[[nodiscard]] constexpr auto transposed() const noexcept -> Mat<C, R, T>
 	{
 		Mat<C, R, T> r {};
@@ -910,6 +1055,7 @@ requires std::is_arithmetic_v<T> struct Mat : std::array<Vec<R, T>, C> {
 		return r;
 	}
 
+	/** @brief Returns an identity matrix. */
 	[[nodiscard]] static constexpr auto identity() noexcept -> Mat<R, C, T>
 	requires(R == C)
 	{
@@ -920,15 +1066,25 @@ requires std::is_arithmetic_v<T> struct Mat : std::array<Vec<R, T>, C> {
 	}
 };
 
+/** @brief 2x2 float matrix alias. */
 using Mat2 = Mat<2, 2>;
+/** @brief 3x3 float matrix alias. */
 using Mat3 = Mat<3, 3>;
+/** @brief 4x4 float matrix alias. */
 using Mat4 = Mat<4, 4>;
 
+/** @brief 2x2 double matrix alias. */
 using Mat2d = Mat<2, 2, double>;
+/** @brief 3x3 double matrix alias. */
 using Mat3d = Mat<3, 3, double>;
+/** @brief 4x4 double matrix alias. */
 using Mat4d = Mat<4, 4, double>;
 
 template<std::size_t R, std::size_t C, typename T>
+/**
+ * @brief Matrix-vector multiplication.
+ * @return Product vector.
+ */
 [[nodiscard]] constexpr Vec<R, T> operator*(
     Mat<R, C, T> const &m, Vec<C, T> const &v) noexcept
 {
@@ -938,7 +1094,10 @@ template<std::size_t R, std::size_t C, typename T>
 	return out;
 }
 
-// Matrix * Matrix
+/**
+ * @brief Matrix-matrix multiplication.
+ * @return Product matrix.
+ */
 template<std::size_t R, std::size_t C, std::size_t K, typename T>
 [[nodiscard]] constexpr Mat<R, K, T> operator*(
     Mat<R, C, T> const &a, Mat<C, K, T> const &b) noexcept
@@ -955,7 +1114,10 @@ template<std::size_t R, std::size_t C, std::size_t K, typename T>
 	return out;
 }
 
-// Mat3 transformations
+/** @name 2D transforms (Mat3)
+ *  @{ */
+
+/** @brief Applies translation to an existing 3x3 transform. */
 template<typename T>
 [[nodiscard]] inline auto translate(Mat<3, 3, T> const &m, Vec<2, T> const &v)
     -> Mat<3, 3, T>
@@ -965,6 +1127,7 @@ template<typename T>
 	return res;
 }
 
+/** @brief Creates a 3x3 translation matrix. */
 template<typename T>
 [[nodiscard]] inline auto translate(Vec<2, T> const &v) -> Mat<3, 3, T>
 {
@@ -974,6 +1137,7 @@ template<typename T>
 	return res;
 }
 
+/** @brief Applies rotation to an existing 3x3 transform. */
 template<typename T>
 [[nodiscard]] inline auto rotate(Mat<3, 3, T> const &m, T const angle)
     -> Mat<3, 3, T>
@@ -990,6 +1154,7 @@ template<typename T>
 	return res;
 }
 
+/** @brief Applies scaling to an existing 3x3 transform. */
 template<typename T>
 [[nodiscard]] inline auto scale(Mat<3, 3, T> const &m, Vec<2, T> const &v)
     -> Mat<3, 3, T>
@@ -1003,6 +1168,7 @@ template<typename T>
 	return res;
 }
 
+/** @brief Applies X shear to an existing 3x3 transform. */
 template<typename T>
 [[nodiscard]] inline auto shear_x(Mat<3, 3, T> const &m, T const v)
     -> Mat<3, 3, T>
@@ -1012,6 +1178,7 @@ template<typename T>
 	return m * res;
 }
 
+/** @brief Applies Y shear to an existing 3x3 transform. */
 template<typename T>
 [[nodiscard]] inline auto shear_y(Mat<3, 3, T> const &m, T const v)
     -> Mat<3, 3, T>
@@ -1021,7 +1188,12 @@ template<typename T>
 	return m * res;
 }
 
-// Mat4 transformations
+/** @} */
+
+/** @name 3D transforms (Mat4)
+ *  @{ */
+
+/** @brief Applies translation to an existing 4x4 transform. */
 template<typename T>
 [[nodiscard]] inline auto translate(Mat<4, 4, T> const &m, Vec<3, T> const &v)
     -> Mat<4, 4, T>
@@ -1031,6 +1203,7 @@ template<typename T>
 	return res;
 }
 
+/** @brief Creates a 4x4 translation matrix. */
 template<typename T>
 [[nodiscard]] inline auto translate(Vec<3, T> const &v) -> Mat<4, 4, T>
 {
@@ -1041,6 +1214,7 @@ template<typename T>
 	return res;
 }
 
+/** @brief Applies rotation to an existing 4x4 transform. */
 template<typename T>
 [[nodiscard]] inline auto rotate(Mat<4, 4, T> const &m, T const angle)
     -> Mat<4, 4, T>
@@ -1058,6 +1232,7 @@ template<typename T>
 	return res;
 }
 
+/** @brief Applies scaling to an existing 4x4 transform. */
 template<typename T>
 [[nodiscard]] inline auto scale(Mat<4, 4, T> const &m, Vec<3, T> const &v)
     -> Mat<4, 4, T>
@@ -1072,6 +1247,7 @@ template<typename T>
 	return res;
 }
 
+/** @brief Applies X shear to an existing 4x4 transform. */
 template<typename T>
 [[nodiscard]] inline auto shear_x(Mat<4, 4, T> const &m, T const v)
     -> Mat<4, 4, T>
@@ -1081,6 +1257,7 @@ template<typename T>
 	return m * res;
 }
 
+/** @brief Applies Y shear to an existing 4x4 transform. */
 template<typename T>
 [[nodiscard]] inline auto shear_y(Mat<4, 4, T> const &m, T const v)
     -> Mat<4, 4, T>
@@ -1090,6 +1267,7 @@ template<typename T>
 	return m * res;
 }
 
+/** @brief Applies Z shear to an existing 4x4 transform. */
 template<typename T>
 [[nodiscard]] inline auto shear_z(Mat<4, 4, T> const &m, T const v)
     -> Mat<4, 4, T>
@@ -1099,9 +1277,24 @@ template<typename T>
 	return m * res;
 }
 
+/**
+ * @brief Builds an orthographic projection matrix.
+ * @param left Left clipping plane.
+ * @param right Right clipping plane.
+ * @param bottom Bottom clipping plane.
+ * @param top Top clipping plane.
+ * @param near Near clipping plane.
+ * @param far Far clipping plane.
+ * @param flip_z_axis Whether to flip the output Z axis convention.
+ * @return Orthographic projection matrix.
+ */
 template<typename T>
-[[nodiscard]] inline auto matrix_ortho3d(T const left, T const right,
+[[nodiscard]] inline auto matrix_ortho3d(T const left,
-    T const bottom, T const top, T const near, T const far,
+    T const right,
+    T const bottom,
+    T const top,
+    T const near,
+    T const far,
     bool const flip_z_axis = true) -> Mat<4, 4, T>
 {
 	Mat<4, 4, T> res {};
@@ -1121,6 +1314,9 @@ template<typename T>
 	return res;
 }
 
+/**
+ * @brief Builds a finite perspective projection matrix.
+ */
 template<typename T>
 inline auto matrix_perspective(
     T fovy, T aspect, T znear, T zfar, bool flip_z_axis = false) -> Mat<4, 4, T>
@@ -1147,9 +1343,13 @@ inline auto matrix_perspective(
 	return m;
 }
 
+/**
+ * @brief Builds a right-handed look-at view matrix.
+ */
 template<typename T>
-[[nodiscard]] inline auto matrix_look_at(Vec<3, T> const eye,
+[[nodiscard]] inline auto matrix_look_at(
-    Vec<3, T> const center, Vec<3, T> const up) -> Mat<4, 4, T>
+    Vec<3, T> const eye, Vec<3, T> const center, Vec<3, T> const up)
+    -> Mat<4, 4, T>
 {
 	auto f = (center - eye).normalized_safe();
 	auto s = f.cross(up).normalized_safe();
@@ -1163,9 +1363,13 @@ template<typename T>
 	};
 }
 
+/**
+ * @brief Builds a perspective matrix with an infinite far plane.
+ */
 template<typename T>
-[[nodiscard]] inline auto matrix_infinite_perspective(T const fovy,
+[[nodiscard]] inline auto matrix_infinite_perspective(
-    T const aspect, T const znear, bool flip_z_axis = false) -> Mat<4, 4, T>
+    T const fovy, T const aspect, T const znear, bool flip_z_axis = false)
+    -> Mat<4, 4, T>
 {
 	Mat<4, 4, T> m {};
 
@@ -1186,11 +1390,38 @@ template<typename T>
 	return m;
 }
 
+/** @} */
+
+/**
+ * @brief Extension point for external math-type interoperability.
+ */
+template<class Ext> struct interop_adapter;
+
+/**
+ * @brief Converts an external value into its mapped smath type.
+ */
+template<class Ext>
+constexpr auto from_external(Ext const &ext) ->
+    typename interop_adapter<Ext>::smath_type
+{
+	return interop_adapter<Ext>::to_smath(ext);
+}
+
+/**
+ * @brief Converts a smath value into an external type.
+ */
+template<class Ext, class SMathT>
+constexpr auto to_external(SMathT const &value) -> Ext
+{
+	return interop_adapter<Ext>::from_smath(value);
+}
+
 } // namespace smath
 
 template<std::size_t N, typename T>
 requires std::formattable<T, char>
-struct std::formatter<smath::Vec<N, T>> : std::formatter<T> {
+struct std::formatter<smath::Vec<N, T>> : std::formatter<T>
+{
 	constexpr auto parse(std::format_parse_context &ctx)
 	{
 		return std::formatter<T>::parse(ctx);
@@ -1213,12 +1444,14 @@ struct std::formatter<smath::Vec<N, T>> : std::formatter<T> {
 	}
 };
 
-namespace std {
+namespace std
+{
 template<size_t N, class T>
-struct tuple_size<smath::Vec<N, T>> : std::integral_constant<size_t, N> { };
+struct tuple_size<smath::Vec<N, T>> : std::integral_constant<size_t, N>
+{ };
 
-template<size_t I, size_t N, class T>
+template<size_t I, size_t N, class T> struct tuple_element<I, smath::Vec<N, T>>
-struct tuple_element<I, smath::Vec<N, T>> {
+{
 	static_assert(I < N);
 	using type = T;
 };

src/modules/smath.cppm

@@ -1,6 +1,6 @@
 module;
 
-#include "smath.hpp"
+#include "smath/smath.hpp"
 
 export module smath;
 

tests/angles.cpp

@@ -1,15 +1,18 @@
 #include <gtest/gtest.h>
 
-#include <smath.hpp>
+#include <smath/smath.hpp>
 
-TEST(AngleReturnRadians, DegInput) {
+TEST(AngleReturnRadians, DegInput)
-  EXPECT_NEAR(smath::deg(180.0), std::numbers::pi, 1e-12);
+{
+	EXPECT_NEAR(smath::deg(180.0), std::numbers::pi, 1e-12);
 }
 
-TEST(AngleReturnRadians, RadInput) {
+TEST(AngleReturnRadians, RadInput)
-  EXPECT_DOUBLE_EQ(smath::rad(std::numbers::pi), std::numbers::pi);
+{
+	EXPECT_DOUBLE_EQ(smath::rad(std::numbers::pi), std::numbers::pi);
 }
 
-TEST(AngleReturnRadians, TurnsInput) {
+TEST(AngleReturnRadians, TurnsInput)
-  EXPECT_NEAR(smath::turns(0.5), std::numbers::pi, 1e-12);
+{
+	EXPECT_NEAR(smath::turns(0.5), std::numbers::pi, 1e-12);
 }

tests/interop.cpp

@@ -0,0 +1,110 @@
+#include <type_traits>
+
+#include <gtest/gtest.h>
+
+#include <smath/smath.hpp>
+
+struct ExternalVec3f
+{
+	float x;
+	float y;
+	float z;
+};
+
+struct ExternalMat2f
+{
+	float m00;
+	float m01;
+	float m10;
+	float m11;
+};
+
+namespace smath
+{
+
+template<> struct interop_adapter<ExternalVec3f>
+{
+	using smath_type = Vec<3, float>;
+
+	static constexpr auto to_smath(ExternalVec3f const &v) -> smath_type
+	{
+		return { v.x, v.y, v.z };
+	}
+
+	static constexpr auto from_smath(smath_type const &v) -> ExternalVec3f
+	{
+		return { v[0], v[1], v[2] };
+	}
+};
+
+template<> struct interop_adapter<ExternalMat2f>
+{
+	using smath_type = Mat<2, 2, float>;
+
+	static constexpr auto to_smath(ExternalMat2f const &m) -> smath_type
+	{
+		smath_type out {};
+		out(0, 0) = m.m00;
+		out(0, 1) = m.m01;
+		out(1, 0) = m.m10;
+		out(1, 1) = m.m11;
+		return out;
+	}
+
+	static constexpr auto from_smath(smath_type const &m) -> ExternalMat2f
+	{
+		return { m(0, 0), m(0, 1), m(1, 0), m(1, 1) };
+	}
+};
+
+} // namespace smath
+
+TEST(Interop, FromExternalVec)
+{
+	ExternalVec3f ext { 1.0f, 2.0f, 3.0f };
+	auto v = smath::from_external(ext);
+
+	static_assert(std::is_same_v<decltype(v), smath::Vec<3, float>>);
+	EXPECT_EQ(v[0], 1.0f);
+	EXPECT_EQ(v[1], 2.0f);
+	EXPECT_EQ(v[2], 3.0f);
+}
+
+TEST(Interop, ToExternalVec)
+{
+	smath::Vec3 v { 4.0f, 5.0f, 6.0f };
+	auto ext = smath::to_external<ExternalVec3f>(v);
+
+	EXPECT_EQ(ext.x, 4.0f);
+	EXPECT_EQ(ext.y, 5.0f);
+	EXPECT_EQ(ext.z, 6.0f);
+}
+
+TEST(Interop, RoundtripVec)
+{
+	ExternalVec3f ext { 7.0f, 8.0f, 9.0f };
+	auto v = smath::from_external(ext);
+	auto ext2 = smath::to_external<ExternalVec3f>(v);
+
+	EXPECT_EQ(ext2.x, 7.0f);
+	EXPECT_EQ(ext2.y, 8.0f);
+	EXPECT_EQ(ext2.z, 9.0f);
+}
+
+TEST(Interop, MatrixConversion)
+{
+	ExternalMat2f ext { 1.0f, 2.0f, 3.0f, 4.0f };
+	auto m = smath::from_external(ext);
+
+	static_assert(std::is_same_v<decltype(m), smath::Mat<2, 2, float>>);
+	EXPECT_EQ(m(0, 0), 1.0f);
+	EXPECT_EQ(m(0, 1), 2.0f);
+	EXPECT_EQ(m(1, 0), 3.0f);
+	EXPECT_EQ(m(1, 1), 4.0f);
+
+	auto ext2 = smath::to_external<ExternalMat2f>(m);
+	EXPECT_EQ(ext2.m00, 1.0f);
+	EXPECT_EQ(ext2.m01, 2.0f);
+	EXPECT_EQ(ext2.m10, 3.0f);
+	EXPECT_EQ(ext2.m11, 4.0f);
+}

tests/vec.cpp

@@ -4,176 +4,191 @@
 
 #include <gtest/gtest.h>
 
-#include <smath.hpp>
+#include <smath/smath.hpp>
 
 using smath::Vec;
 using smath::Vec2;
 using smath::Vec3;
 using smath::Vec4;
 
-template <class T>
+template<class T>
-static void ExpectVecNear(const Vec<3, T> &a, const Vec<3, T> &b,
+static void ExpectVecNear(
-                          T eps = T(1e-6)) {
+    const Vec<3, T> &a, const Vec<3, T> &b, T eps = T(1e-6))
-  for (int i = 0; i < 3; ++i)
+{
-    EXPECT_NEAR(double(a[i]), double(b[i]), double(eps));
+	for (int i = 0; i < 3; ++i)
+		EXPECT_NEAR(double(a[i]), double(b[i]), double(eps));
 }
 
 // Constructors and accessors
-TEST(Vec, DefaultZero) {
+TEST(Vec, DefaultZero)
-  Vec3 v;
+{
-  EXPECT_EQ(v[0], 0.0f);
+	Vec3 v;
-  EXPECT_EQ(v[1], 0.0f);
+	EXPECT_EQ(v[0], 0.0f);
-  EXPECT_EQ(v[2], 0.0f);
+	EXPECT_EQ(v[1], 0.0f);
+	EXPECT_EQ(v[2], 0.0f);
 }
 
-TEST(Vec, ScalarFillCtor) {
+TEST(Vec, ScalarFillCtor)
-  Vec4 v{2.0f};
+{
-  EXPECT_EQ(v.x(), 2.0f);
+	Vec4 v { 2.0f };
-  EXPECT_EQ(v.y(), 2.0f);
+	EXPECT_EQ(v.x(), 2.0f);
-  EXPECT_EQ(v.z(), 2.0f);
+	EXPECT_EQ(v.y(), 2.0f);
-  EXPECT_EQ(v.w(), 2.0f);
+	EXPECT_EQ(v.z(), 2.0f);
+	EXPECT_EQ(v.w(), 2.0f);
 }
 
-TEST(Vec, VariadicCtorScalarsAndSubvectors) {
+TEST(Vec, VariadicCtorScalarsAndSubvectors)
-  Vec2 a{1.0f, 2.0f};
+{
-  Vec2 b{3.0f, 4.0f};
+	Vec2 a { 1.0f, 2.0f };
-  Vec4 v{a, b};
+	Vec2 b { 3.0f, 4.0f };
-  EXPECT_EQ(v.r(), 1.0f);
+	Vec4 v { a, b };
-  EXPECT_EQ(v.g(), 2.0f);
+	EXPECT_EQ(v.r(), 1.0f);
-  EXPECT_EQ(v.b(), 3.0f);
+	EXPECT_EQ(v.g(), 2.0f);
-  EXPECT_EQ(v.a(), 4.0f);
+	EXPECT_EQ(v.b(), 3.0f);
+	EXPECT_EQ(v.a(), 4.0f);
 }
 
-TEST(Vec, NamedAccessorsAliases) {
+TEST(Vec, NamedAccessorsAliases)
-  Vec3 v{1.0f, 2.0f, 3.0f};
+{
-  EXPECT_EQ(v.x(), v.r());
+	Vec3 v { 1.0f, 2.0f, 3.0f };
-  EXPECT_EQ(v.y(), v.g());
+	EXPECT_EQ(v.x(), v.r());
-  EXPECT_EQ(v.z(), v.b());
+	EXPECT_EQ(v.y(), v.g());
+	EXPECT_EQ(v.z(), v.b());
 }
 
 // Arithmetic
-TEST(Vec, ElementwiseAndScalarOps) {
+TEST(Vec, ElementwiseAndScalarOps)
-  Vec3 a{1.0f, 2.0f, 3.0f};
+{
-  Vec3 b{4.0f, 5.0f, 6.0f};
+	Vec3 a { 1.0f, 2.0f, 3.0f };
+	Vec3 b { 4.0f, 5.0f, 6.0f };
 
-  auto s1 = a + b;
+	auto s1 = a + b;
-  EXPECT_EQ(s1[0], 5.0f);
+	EXPECT_EQ(s1[0], 5.0f);
-  EXPECT_EQ(s1[1], 7.0f);
+	EXPECT_EQ(s1[1], 7.0f);
-  EXPECT_EQ(s1[2], 9.0f);
+	EXPECT_EQ(s1[2], 9.0f);
 
-  auto s2 = a * 2.0f;
+	auto s2 = a * 2.0f;
-  EXPECT_EQ(s2[0], 2.0f);
+	EXPECT_EQ(s2[0], 2.0f);
-  EXPECT_EQ(s2[1], 4.0f);
+	EXPECT_EQ(s2[1], 4.0f);
-  EXPECT_EQ(s2[2], 6.0f);
+	EXPECT_EQ(s2[2], 6.0f);
 
-  auto s3 = 2.0f * a; // RHS overloads
+	auto s3 = 2.0f * a; // RHS overloads
-  EXPECT_EQ(s3[0], 2.0f);
+	EXPECT_EQ(s3[0], 2.0f);
-  EXPECT_EQ(s3[1], 4.0f);
+	EXPECT_EQ(s3[1], 4.0f);
-  EXPECT_EQ(s3[2], 6.0f);
+	EXPECT_EQ(s3[2], 6.0f);
 
-  Vec3 c{1.0f, 2.0f, 3.0f};
+	Vec3 c { 1.0f, 2.0f, 3.0f };
-  c += Vec3{1.0f, 1.0f, 1.0f};
+	c += Vec3 { 1.0f, 1.0f, 1.0f };
-  EXPECT_EQ(c[0], 2.0f);
+	EXPECT_EQ(c[0], 2.0f);
-  EXPECT_EQ(c[1], 3.0f);
+	EXPECT_EQ(c[1], 3.0f);
-  EXPECT_EQ(c[2], 4.0f);
+	EXPECT_EQ(c[2], 4.0f);
 
-  c *= 2.0f;
+	c *= 2.0f;
-  EXPECT_EQ(c[0], 4.0f);
+	EXPECT_EQ(c[0], 4.0f);
-  EXPECT_EQ(c[1], 6.0f);
+	EXPECT_EQ(c[1], 6.0f);
-  EXPECT_EQ(c[2], 8.0f);
+	EXPECT_EQ(c[2], 8.0f);
 }
 
 // Length, dot, cross, normalize
-TEST(Vec, MagnitudeAndDot) {
+TEST(Vec, MagnitudeAndDot)
-  Vec3 v{3.0f, 4.0f, 12.0f};
+{
-  EXPECT_FLOAT_EQ(v.magnitude(), 13.0f);
+	Vec3 v { 3.0f, 4.0f, 12.0f };
-  EXPECT_FLOAT_EQ(v.length(), 13.0f);
+	EXPECT_FLOAT_EQ(v.magnitude(), 13.0f);
+	EXPECT_FLOAT_EQ(v.length(), 13.0f);
 
-  Vec3 u{1.0f, 0.0f, 2.0f};
+	Vec3 u { 1.0f, 0.0f, 2.0f };
-  EXPECT_FLOAT_EQ(v.dot(u), 27.0f);
+	EXPECT_FLOAT_EQ(v.dot(u), 27.0f);
 }
 
-TEST(Vec, Cross3D) {
+TEST(Vec, Cross3D)
-  Vec3 x{1.0f, 0.0f, 0.0f};
+{
-  Vec3 y{0.0f, 1.0f, 0.0f};
+	Vec3 x { 1.0f, 0.0f, 0.0f };
-  auto z = x.cross(y);
+	Vec3 y { 0.0f, 1.0f, 0.0f };
-  EXPECT_EQ(z[0], 0.0f);
+	auto z = x.cross(y);
-  EXPECT_EQ(z[1], 0.0f);
+	EXPECT_EQ(z[0], 0.0f);
-  EXPECT_EQ(z[2], 1.0f);
+	EXPECT_EQ(z[1], 0.0f);
+	EXPECT_EQ(z[2], 1.0f);
 }
 
-TEST(Vec, NormalizeAndSafeNormalize) {
+TEST(Vec, NormalizeAndSafeNormalize)
-  Vec3 v{10.0f, 0.0f, 0.0f};
+{
-  auto n = v.normalized();
+	Vec3 v { 10.0f, 0.0f, 0.0f };
-  auto ns = v.normalized_safe();
+	auto n = v.normalized();
-  ExpectVecNear(n, Vec3{1.0f, 0.0f, 0.0f});
+	auto ns = v.normalized_safe();
+	ExpectVecNear(n, Vec3 { 1.0f, 0.0f, 0.0f });
 
-  Vec3 zero{};
+	Vec3 zero {};
-  auto zs = zero.normalized_safe();
+	auto zs = zero.normalized_safe();
-  EXPECT_EQ(zs[0], 0.0f);
+	EXPECT_EQ(zs[0], 0.0f);
-  EXPECT_EQ(zs[1], 0.0f);
+	EXPECT_EQ(zs[1], 0.0f);
-  EXPECT_EQ(zs[2], 0.0f);
+	EXPECT_EQ(zs[2], 0.0f);
 }
 
-TEST(Vec, DistanceAndProjection) {
+TEST(Vec, DistanceAndProjection)
-  Vec3 a{1.0f, 2.0f, 3.0f};
+{
-  Vec3 b{4.0f, 6.0f, 3.0f};
+	Vec3 a { 1.0f, 2.0f, 3.0f };
-  EXPECT_FLOAT_EQ(a.distance(b), 5.0f);
+	Vec3 b { 4.0f, 6.0f, 3.0f };
+	EXPECT_FLOAT_EQ(a.distance(b), 5.0f);
 
-  Vec3 n{2.0f, 0.0f, 0.0f};   // onto x-axis scaled
+	Vec3 n { 2.0f, 0.0f, 0.0f }; // onto x-axis scaled
-  auto p = a.project_onto(n); // (a·n)/(n·n) * n = (2)/4 * n = 0.5 * n
+	auto p = a.project_onto(n);  // (a·n)/(n·n) * n = (2)/4 * n = 0.5 * n
-  ExpectVecNear(p, Vec3{1.0f, 0.0f, 0.0f});
+	ExpectVecNear(p, Vec3 { 1.0f, 0.0f, 0.0f });
 }
 
 // Approx equal
-TEST(Vec, ApproxEqual) {
+TEST(Vec, ApproxEqual)
-  Vec3 a{1.0f, 2.0f, 3.0f};
+{
-  Vec3 b{1.0f + 1e-7f, 2.0f - 1e-7f, 3.0f};
+	Vec3 a { 1.0f, 2.0f, 3.0f };
-  EXPECT_TRUE(a.approx_equal(b, 1e-6f));
+	Vec3 b { 1.0f + 1e-7f, 2.0f - 1e-7f, 3.0f };
-  EXPECT_FALSE(a.approx_equal(b, 1e-9f));
+	EXPECT_TRUE(a.approx_equal(b, 1e-6f));
+	EXPECT_FALSE(a.approx_equal(b, 1e-9f));
 }
 
 // std::get & tuple interop
-TEST(Vec, StdGetAndTuple) {
+TEST(Vec, StdGetAndTuple)
-  Vec3 v{7.0f, 8.0f, 9.0f};
+{
-  static_assert(std::tuple_size_v<Vec3> == 3);
+	Vec3 v { 7.0f, 8.0f, 9.0f };
-  static_assert(std::is_same_v<std::tuple_element_t<1, Vec3>, float>);
+	static_assert(std::tuple_size_v<Vec3> == 3);
-  EXPECT_EQ(std::get<0>(v), 7.0f);
+	static_assert(std::is_same_v<std::tuple_element_t<1, Vec3>, float>);
-  EXPECT_EQ(std::get<1>(v), 8.0f);
+	EXPECT_EQ(std::get<0>(v), 7.0f);
-  EXPECT_EQ(std::get<2>(v), 9.0f);
+	EXPECT_EQ(std::get<1>(v), 8.0f);
+	EXPECT_EQ(std::get<2>(v), 9.0f);
 }
 
 // Swizzle
-TEST(Vec, SwizzleBasic) {
+TEST(Vec, SwizzleBasic)
-  const Vec3 v{1.0f, 2.0f, 3.0f};
+{
+	const Vec3 v { 1.0f, 2.0f, 3.0f };
 
-  auto yz = smath::swizzle<"yz">(v);
+	auto yz = smath::swizzle<"yz">(v);
-  EXPECT_EQ(yz[0], 2.0f);
+	EXPECT_EQ(yz[0], 2.0f);
-  EXPECT_EQ(yz[1], 3.0f);
+	EXPECT_EQ(yz[1], 3.0f);
 
-  auto rxx = smath::swizzle<"xxy">(v);
+	auto rxx = smath::swizzle<"xxy">(v);
-  EXPECT_EQ(rxx[0], 1.0f);
+	EXPECT_EQ(rxx[0], 1.0f);
-  EXPECT_EQ(rxx[1], 1.0f);
+	EXPECT_EQ(rxx[1], 1.0f);
-  EXPECT_EQ(rxx[2], 2.0f);
+	EXPECT_EQ(rxx[2], 2.0f);
 }
 
 // std::formatter
-TEST(Vec, Formatter) {
+TEST(Vec, Formatter)
-  smath::Vec<3, int> vi{1, 2, 3};
+{
-  std::string s = std::format("{}", vi);
+	smath::Vec<3, int> vi { 1, 2, 3 };
-  EXPECT_EQ(s, "{1, 2, 3}");
+	std::string s = std::format("{}", vi);
+	EXPECT_EQ(s, "{1, 2, 3}");
 }
 
 // Conversions
-TEST(Vec, ExplicitConversionBetweenScalarTypes) {
+TEST(Vec, ExplicitConversionBetweenScalarTypes)
-  smath::Vec<3, int> vi{1, 2, 3};
+{
-  smath::Vec<3, float> vf{vi};
+	smath::Vec<3, int> vi { 1, 2, 3 };
-  EXPECT_EQ(vf[0], 1.0f);
+	smath::Vec<3, float> vf { vi };
-  EXPECT_EQ(vf[1], 2.0f);
+	EXPECT_EQ(vf[0], 1.0f);
-  EXPECT_EQ(vf[2], 3.0f);
+	EXPECT_EQ(vf[1], 2.0f);
+	EXPECT_EQ(vf[2], 3.0f);
 
-  auto vi2 = static_cast<smath::Vec<3, int>>(vf);
+	auto vi2 = static_cast<smath::Vec<3, int>>(vf);
-  EXPECT_EQ(vi2[0], 1);
+	EXPECT_EQ(vi2[0], 1);
-  EXPECT_EQ(vi2[1], 2);
+	EXPECT_EQ(vi2[1], 2);
-  EXPECT_EQ(vi2[2], 3);
+	EXPECT_EQ(vi2[2], 3);
 }