Haskell-embedded OpenGL.

A simple, functional approach to OpenGL programming in Haskell.

All definitions that comprise HaGL are provided by the top-level Graphics.HaGL module.

HaGL: Haskell-embedded OpenGL

HaGL (Haskell-embedded OpenGL, pronounced "haggle") aims to provide a simple, elegant, and modular way to create lightweight OpenGL visualizations in Haskell.

By unifying vertex, fragment, and uniform processing into an expressively typed GLSL-like language consisting of pure, composable primitives, HaGL makes it easy to experiment with and prototype visual graphical demos.

To see it in action, please skip straight to the "Getting Started" guide.

Installation

Please make sure that you have installed the correct headers and static libraries for OpenGL development on your system, as well as freeglut for running the GLUT backend. (For example, in Debian/Ubuntu, ensure that the packages libgl-dev, lib-glu1-mesa-dev, and freeglut3 are installed.)

Then with the latest version of Haskell Cabal installed:

cabal update
cabal install HaGL

To instead build locally from the sources, issue:

cabal build HaGL

from the top-level directory.

The supporting library and examples can be built using:

cabal build HaGL-lib
cabal build HaGL-examples

(It is recommended that these packages be built not installed, as they are not yet stable and complete enough to be used without modifications.)

To check that all the packages have been installed correctly, run a single example as follows:

cabal run hagl-example interactive_cube

Usage

To learn how to use HaGL please refer to the "Getting Started" guide. The complete documentation can be found on Hackage.

Running the Test Suite

The test suite can be run using:

cabal test

Note that since some tests require launching a graphical window, the test suite cannot be run on a non-GUI system.

Intended Use and Alternatives

HaGL is best suited for the production of simple non-reactive visual animations; in this regard, its use-cases somewhat resemble those of the VisPy and Glumpy Python libraries.

At the moment, the only way to interpret HaGL code is as a GLUT application but other backends will be added in the future, as well as a backend-agnostic interface.

HaGL prioritizes approachability and simplicity over completeness and is likely unsuitable for use in a game engine. For more advanced graphics programming consider using, for example, LambdaCube 3D, FIR, or the Haskell OpenGL bindings.

Scope of OpenGL Support

HaGL models a simple OpenGL pipeline that supports basic vertex and fragment processing and exposes a generic API similar to that of GLSL that can be used for both shader programming and host numerical computation (of uniforms). In addition, it provides various operations of its own to model certain imperative constructs in a functional manner.

An "interpreter" for running HaGL code is provided in the form of GLUT backend built using the OpenGL and GLUT bindings for Haskell.

Features that are currently missing but could possibly be added in the future:

• Texture mapping using sampler objects
• Instanced rendering
• Tessellation shading
• Geometry shading
• Compute shading
• Non-GLUT backends and a backend-agnostic interface
• Integration with an FRP framework such as Yampa

Contributing

Contributions in the form of pull requests, suggestions, or feedback are welcome and appreciated.

The best way to keep the project alive is to demonstrate its usefulness through a wide range of interesting examples and tutorials. Another potential area to explore is the creation of high-level libraries for specific types of visual applications, as well as improving the supporting library that ships with HaGL.