Re:My First Fifteen Compilers
by codr7 on Friday July 12, 2019 @10:51AM
(#20421146)
I assume you're aiming to make your own compiler/interpreter.
Consider postponing the deep-dive into theory until you've tried solving the problem using what you already know. Once you know your way around a problem, digesting theory becomes much easier.
On one level, there's nothing magical about compilers and interpreters. It's the same old regular code [0] solving the same kinds of problems.
What makes it tricky is working on a meta-level in parallel, visualizing what's happening in two dimensions. Like writing Lisp macros if that makes any sense.
I think it's a shame that so many get stuck on parsing and type theory. The top priority should be to get a feedback loop up and running.
Consider postponing the deep-dive into theory until you've tried solving the problem using what you already know. Once you know your way around a problem, digesting theory becomes much easier.
On one level, there's nothing magical about compilers and interpreters. It's the same old regular code [0] solving the same kinds of problems.
What makes it tricky is working on a meta-level in parallel, visualizing what's happening in two dimensions. Like writing Lisp macros if that makes any sense.
I think it's a shame that so many get stuck on parsing and type theory. The top priority should be to get a feedback loop up and running.
[0] https://github.com/codr7/cidk/blob/master/src/cidk/read.cpp