Writing 700 LoC is one thing. Spending two days writing 200 LoC of test 
framework followed by 400 LoC of actual test cases is another. I was 
beginning to wonder if I'd ever hit upon a good set of test 
abstractions. But now I seem to have found one, and I've written a test 
for every single function I've implemented.

But the really satisfying thing is... I actually found some bugs in the 
process. (And fixed them. Obviously.)

I also stumbled upon some poorly-documented edge-cases of the standard 
library functions that I'm comparing against, FWIW.

Post a reply to this message