targon

> They also tend to be the least buggy parts, I would handwave that maybe 1 in a thousand bugs were in the lexer/parser.

Ehh, I don't know... https://johanengelen.github.io/ldc/2018/01/14/Fuzzing-with-L...

From the article:

> My expectation was that there would be a bug deep down with some rare cornercase of a cornercase, and that the fuzzer would have to run for hours and hours covering more and more of the lexer to, perhaps, finally find a bug. So I was preparing for a long fuzzing session, wanted to go to bed and kick this one off but… The fuzzer found a first failure within a second.

> Running the fuzzer several times, sometimes it still hadn’t found a bug after 15 seconds, but usually it found the bug within a second; the process is (semi-)random so that’s expected.