Fast, Super Slim, Embeddable, Headers only C++11
vector graphics library, that can run on any 32/64 bits computer without FPU or GPU. No standard library required.
check out our website at micro-gl.github.io/docs/microgl
A circle with Texture fill and Gradient stroke
using number = float;
using Canvas24= canvas<bitmap<coder::RGB888_PACKED_32>, CANVAS_OPT_32_BIT>;
using Texture24= sampling::texture<bitmap<coder::RGB888_ARRAY>, sampling::texture_filter::NearestNeighboor>;
fast_radial_gradient<number> gradient{0.5, 0.5, 0.75};
linear_gradient_2_colors<120> gradient2Colors{{255,0,255},
{255,0,0}};
flat_color<> flatColor{{133,133,133, 255}};
auto img_2 = Resources::loadImageFromCompressedPath("images/uv_256.png");
Canvas24 canvas(W, H);;
Texture24 tex_uv{new bitmap<coder::RGB888_ARRAY>(img_2.data, img_2.width, img_2.height)};
gradient.addStop(0.0f, {255,0,0});
gradient.addStop(0.45f, {255,0,0});
gradient.addStop(0.50f, {0,255,0});
gradient.addStop(1.f, {255,0,255});
canvas.clear({255,255,255,255});
canvas.drawCircle<blendmode::Normal, porterduff::FastSourceOverOnOpaque, true, number>(
tex_uv, // fill
gradient2Colors, // stroke
200+0, 200+0,
150+t, 10, 255
);
A Rectangle, that samples a circle
using Canvas24= canvas<bitmap<coder::RGB888_PACKED_32>>;
using number = float;
// using number = Q<12>;
circle_sampler<number> sampler{};
sampler.updatePoints({0.50,0.5}, 0.5, 0.10);
sampler.color_fill= {0, 0, 0, 255};
sampler.color_background= {255, 255, 255, 0};
sampler.color_stroke= {255, 255, 255, 255};
Canvas24 canvas(512, 512);
canvas.clear({255,0,255,255});
canvas.drawRect<blendmode::Normal, porterduff::FastSourceOverOnOpaque, false, number>(
sampler,
0, 0, 300, 300);
A path tessellation with red color
template <typename number>
path_t<number> path_arc_animation() {
path_t<number> path{};
int div=32; //4
path.arc({200,200}, 100,
math::deg_to_rad(0.0f),
math::deg_to_rad(360.0f),
false, div).closePath();
path.arc({250,200}, 50,
math::deg_to_rad(0.0f),
math::deg_to_rad(360.0f),
true, div).closePath();
path.moveTo({150,150});
path.arc({150+0,150}, 50+t-0,
math::deg_to_rad(0.0f),
math::deg_to_rad(360.0f),
false, div);//.closePath();
return path;
}
// Choose your own number format
// using number = float;
// using number = double;
// using number = Q<15, long long>;
using number = Q<8, microgl::ints::int32_t, microgl::ints::int64_t, 1>;
// using number = Q<2, microgl::ints::int64_t>;
// using number = Q<4, microgl::ints::int32_t>;
// using number = Q<12>;
using Canvas24= canvas<bitmap<RGB888_PACKED_32>>;
sampling::flat_color<> color_red {{ 255, 0, 255, 255}};
Canvas24 canvas(255, 255);
canvas.clear({255, 255, 255, 255});
canvas.drawPathFill<blendmode::Normal, porterduff::FastSourceOverOnOpaque, false, true>(
color_red,
matrix_3x3<number>::identity(),
path,
microtess::fill_rule::even_odd,
microtess::tess_quality::prettier_with_extra_vertices,
255
);
Language
C++11
, does not use anySTL
library and standard library.- headers only library
- most functions are templates, which means you compile only what is used
- compile-time polymorphism (no runtime virtual methods)
- support for 32 bits and 64 bits
- self contained and modular.
Graphics
- very modular design. if something does not exist, you can implement your own
- over 10 blending modes impemented, you can wrote your own and supply as well
porter-duff
alpha compositing- support for 2d sampler
- can support any pixel coding you need, we implemeted a lot out of the box
- can render 3d geometries
- we support 3d shaders
- no need for GPU, it is all CPU
- no need for FPU
- support for any number system including
Q
numbers (fixed point), no need to use float points
Geometry
- fast implementation for common shapes (rect, rounded rectangles, circles, triangles)
- batch triangles like GPU would
- support polygons of any type convex, simple, and complex polygons
- path fills (with even-odd or zero-one fill rule) in any precision
- stroke fills
- quadrilaterals
- bezier patches
- we providea dynamic array, but you can plug your own.
microgl
is a headers only library, which gives the following install possibilities:
- Using
cmake
to invoke theinstall
target, that will copy everything in your system via
$ mkdir cmake-build-release
$ cd cmake-build-release
$ cmake -DCMAKE_BUILD_TYPE=Release ..
$ cmake --install .
- Copying the
include/microgl
to anywhere you want. - Install via vcpkg.
Following options are available:
- copy the project to a sub folder of your project. Inside your
CMakeLists.txt
add
add_subdirectory(/path/to/microgl)
target_link_libraries(your_app microgl)
- If you installed
micro{gl}
with option 1 (see above) at your system, you can instead
find_package(microgl CONFIG REQUIRED)
target_link_libraries(your_app microgl::microgl)
- If you have not installed, you can add in your app's
CMakeLists.txt
target_include_directories(app path/to/microgl/folder/include/)
- If you manually copied the
include/microgl
to the default system include path,
you can usecmake/Findmicrogl.cmake
to automatically create the cmake targets
list(APPEND CMAKE_MODULE_PATH ${PROJECT_SOURCE_DIR}/path/to/Findmicrogl/folder)
find_package(microgl REQUIRED)
target_link_libraries(your_app microgl::microgl)
- Just copy the
include/microgl
into a sub folder of your project and include the header
files you need with relative path in your source files. - If you have vcpkg installed, install the
micro{gl}
using thevcpkg install microgl
command, and then you can use option 2 (see above) to find it. See issue micro-gl#33 for more details.
First make sure you have
There are two ways:
- Use your favourite IDE to load the root
CMakeLists.txt
file, and then it
will pick up all of the targets, including the examples - Using the command line:
$ mkdir cmake-build-release
$ cd cmake-build-release
$ cmake -DCMAKE_BUILD_TYPE=Release ..
$ cmake --build . --target <example_name>
$ ../examples/bin/example_name
Author: Tomer Shalev, [email protected], all rights reserved (2021)
🙏🏻 Thanks to JetBrains cLion Support Team for granting this project an OSS license of cLion