The JyOpenGL project is a fork of PyOpenGL containing adjustments that allow it to run on Jython. Note that JyOpenGL is not a port, since the adjustments are so minimal that - as of this writing - they don't touch CPython compatibility. So JyOpenGL is workable on Jython and CPython at the same time. However it is not explicitly tested on CPython. Still JyOpenGL was created to be a separate project, because we will add additional Jython- and Java-intergration features, an AWT-based rendering frontend and Jython-style deployment structure, e.g. a build-script to create a Jython-compliant jar-file containing a precompiled JyOpenGL library.

To use JyOpenGL on Jython, a sufficiently complete ctypes implementation for Jython is required. As of this writing Jython's builtin JFFI-based ctypes implementation is not sufficient for this purpose.

Currently the only known workable way is to use JyNI (www.jyni.org) with native ctypes. While JyNI also supports other posix platforms, the described setup was only tested on Linux (LMDE2, 64 bit) yet. Detailed setup instructions for this configuration will be appended to this section in near future.

PyOpenGL is normally distributed via PyPI using standard pip:

$ pip install PyOpenGL PyOpenGL_accelerate

You can install this repository by branching/cloning and running setup.py:

$ cd pyopengl
$ python setup.py develop
$ cd accelerate
$ python setup.py develop

Note that to compile PyOpenGL_accelerate you will need to have a functioning Python extension-compiling environment.

If you are new to PyOpenGL, you likely want to start with the OpenGLContext tutorial page. Those tutorials require OpenGLContext, (which is a big wrapper including a whole scenegraph engine, VRML97 parser, lots of demos, etc) you can install that with:

$ pip2.7 install "OpenGLContext-full==3.1.1

Or you can clone it (including the tutorial sources) with:

$ bzr branch lp:openglcontext

The documentation pages are useful for looking up the parameters and semantics of PyOpenGL calls.

You can run the PyOpenGL test suite only if you have prebuilt Pygame and Numpy wheels, along with Python 2.7, 3.4 and 3.5. The wheels for the test suite to use should be stored in a directory called "wheelhouse" at the same level as the root checkout here.

To build the wheels on Ubuntu:

$ hg clone https://bitbucket.org/pygame/pygame
$ apt-get build-dep pygame python-numpy
$ pip2.7 wheel ./pygame numpy
$ pip3.4 wheel ./pygame numpy
$ pip3.5 wheel ./pygame numpy

if you do that in the same directory where you checked out pyopengl you will have all of your wheels in the directory the pyopengl tox suite is expecting.

You'll obviously need tox installed to run tox, which looks like this:

$ tox

The result being a lot of tests being run in a matrix of environments, with Python versions:

• 2.7
• 3.5
• 3.4

Where we test with and without the accelerate module and with and without numpy installed in the environment.