I stopped working on the code for many weeks because I had other things to on my mind, but also because the work I planned was not super exciting.
I want to trust my “basic” libraries, math, core and image. They contain relatively simple code that will be the building blocks of the rest of the project. So I want to make sure I eliminate most of the obvious bugs before hand. Other libraries have much higher-level code and won’t need the same amount of precision testing, I will likely make different sort of tests for them.
But for that, I must complete the unit tests for those libraries, and I have a lot work ahead of me:
- test_math: 95 out of 151 tests written, I just added helper functions to generate random vectors and matrices. I truly hate myself for wanting to write everything myself, writing tests for this is just awful
- test_core: only 13 tests so far, and the library has like 447 functions… That will take a while but it will be infinitely better than writing math tests
- test_image: I will probably skip a few functions there (like the noise stuff), but that is very important to make sure the different image formats are working
I can’t help but think all this is very tedious, and I have half a mind to just throw it all away… But at the same time, when I decide to make the project cross-platform, the tests will help securing the code.
It’s been nagging me for months now, so I will push through and finally move on to something better!