Archives December 2009

All three local PyCon presenters are booked in for the PyGTA on February 16th.  Should be pretty darn cool.

I've been working on something for my PyCon talk for *far* too long now.  I started in Nevow + Athena, which is a technology I'm moderately familiar with, having written Cinemon using it (though that was years ago now).  I got a nice single-person implementation done with Athena and then started looking into how to ...

I've been trying to push out the next beta of PyOpenGL 3.0.1 for a while now.  However, as noted before, I've now got a semi-automatic test suite... and it's failing on Win32.  So far all of the problems have come down to test-suite issues, but they've been exposing some annoying behaviors of the ATI Win32 ...

We started off the PyGTA meeting with brainstorming for fun projects that we could do in a "pure functional" manner. Lots of complex and involved things that we'd whip off in a few minutes. But first, just to be sure we were all on the same page, let's whip off a quicky...

• given a file ...

Just finished registering for PyCon 2010.  Early-bird corporate rate is only $450USD, very good value for the money.  Direct flights from Toronto run ~$920CDN (taking ~2 hrs), 1-stop flights are ~$600CDN (taking 6+ hrs).  Hotel is around $80/night or around $560 if you intend to arrive the night before the conference-proper and stay through the ...

So I'm once again trying to fix the PyOpenGL/OpenGLContext pydoc stuff.  They've been using a hacked up version of pydoc for years (with a brief excursion into epydoc with extensions) and somewhere along the way the changes to the structure of the libraries means that the output is now basically useless.  The key here is ...

I just finished writing the code for the 12th shader tutorial, which is basically taking the 11th tutorial, which is still a "single-file application" tutorial, and converting it to use the OpenGLContext Shader scenegraph nodes and generally look (slightly) more like a real-world rendering engine.

I like single-file tutorials, but really there's a point at ...

Beginning work on cleaning up the introductory shader tutorials.  Basically separating out the various component calculations into reusable functions and making it possible to use generic scenegraph objects for the geometry to be displayed.  Working on that because it really seems that beyond a certain point there's diminishing returns in working on the shadow tutorials ...

Grabbed some time this evening to clean up the shadow tutorial code.  It's now refactored into reasonably small methods, supports N lights, uses the cache to store textures/fbos, and generally is a lot less messy.  I'm currently debating what the next step in the tutorial should be.  My current thinking is to use shaders to ...