Category archives: Snaking
Discussions of programming, particularly of programming Python
Just kidding (TurboGears template language vs the old-code mess...)
Written by
on
in
Snaking.
Okay, I've officially gone way over my time for the day. I wanted to see if I could use a templating language to clean up some of the old-code cruft without having the project's owner have a heart attack. A good bit of time was spent trying to get kid installed. It uses a Python ...
Ah, the piquant aroma of well-aged web code (Or, at least aged...)
Written by
on
in
Snaking.
Five and a half hours today restructuring a couple of web pages. These are basically huge (1000+ line) scripts with a handful of individual "functions", save that they aren't coded as actual functions, they're just conditional branches scattered throughout the script.
All the code's generated with print statements and embedded HTML strings (with all tags ...!-->!-->
The joys of long-deferred todos (You defer them because they are ugly and nasty...)
Written by
on
in
Snaking.
Today I finally tackled one of the longstanding todos in the VoIP project, namely the point of interface between ISP account activation and VoIP account activation. Since the VoIP system is written to be largely independent of the ISP billing system there's a certain "energy barrier" whenever there's something needing to be done in the ...
We need test code (Me writing a test for every extension will be a bit too slow...)
Written by
on
in
Snaking.
I continued working on OpenGL-ctypes for a few hours yesterday (until I fell asleep), just sitting down and writing test-cases for those extensions I have and using them to see if the extension modules are working. There's generally some work to do for every module that has a function using a pointer, out parameter, glGet* ...
Extensions move into CVS (OpenGL-ctypes grows by hundreds of modules in a few minutes...)
Written by
on
in
Snaking.
I've just checked the auto-generated extensions for OpenGL-ctypes into the Sourceforge CVS repository. The autogeneration is still quite crude, so expect any extension with glGet-like, pointer or image-based functionality to fail. I've also reworked the glget module to use the wrapper system, which makes it a much simpler module (and makes the patterns easier to ...
Extensions start to take shape (A bit of OpenGL-ctypes hacking...)
Written by
on
in
Snaking.
I continued work on the extension wrapping code for OpenGL-ctypes this evening. The arb multitexture and arb window pos extensions both work correctly. I've extended the auto-generator such that we can add custom code directly after the auto-generated stuff to customise operation.
Anyway, the generator module is checked into CVS for those who want to ...!-->!-->
Gentle hush of tickets falling (Moving toward launch as winter settles...)
Written by
on
in
Snaking,
Vindaloo.
Spent the entire day on the VoIP project, mostly on the Emergency Services (9-1-1) support. A bit of time toward the end on the proposal for the new business.
Dad's home from the hospital. Grams is still in surgery. I'm planning on dropping in to see her tomorrow afternoon.
Bit-rot sucks (One must maintain...)
Written by
on
in
Snaking.
Have spent most of the morning working against bit rot in various internal projects. Very tedious. Nothing as bad as with PyOpenGL (which I really need to spend some more time on (I just don't want to)), but enough to take up almost 3 hours of the day.
They still haven't come in to fix ...!-->!-->
Distractions, distractions (8+ hours into the working day without touching item 1 on the todo list...)
Written by
on
in
Snaking,
Tuxedo.
Today I was to rise, then work on the business plan. That was my entire todo list for the day. Instead I spent the bulk of the day on the VOIP project. There was a brief interlude where I provided free Unix lessons to one of our business partners.
Bit of time spent on learning ...!-->!-->
FreeGLUT support mostly working (Minor problems with the string-rendering...)
Written by
on
in
Snaking.
Okay, went with implementing the FreeGLUT wrappers. They are basically working now, though I can't test the mouse-wheel (I don't have a mouse-with-wheel configured on Linux).
Biggest annoyance is that only the first line of text is getting rendered with the string-rendering code. I tried using u_byte arrays (which is how arrays are normally handled ...!-->!-->