[TI ASM] Parser and Compiler

coelurus · Post by **coelurus** » Fri 19 May, 2006 9:25 am

Full parsers/grammarthingies/syntaxmabobs/whatnot can be rather tedious for high level languages, take CoBB's piece of advice and use free tools (flex+bison are pretty common). Even gcc used flex+bison until only recently (4.0.1?).

Consider also his advice regarding planning. The hard part with compilers isn't the frontend, but the backend. The frontend feeds the backend, the meaning of that should be pretty obvious

ssartell · Post by **ssartell** » Fri 19 May, 2006 2:59 pm

Look, I'm doing this for my own enjoyment, so if I want to try my hand at whatever part so be it. Like I said before (and I mean it even more so now after the last few comments), if someone would like to try it themselves then by all means go ahead. I sounds like there are several people who are more qualified/competent/organized than me who could no doubt make a better language than me in probably less time than me. You guys seem to have all the experience in the world regarding high level languages. And I'm sure the community would eat it up if one of you guys announced you were working on one.

CoBB · Post by **CoBB** » Fri 19 May, 2006 6:50 pm

Anything for enjoyment!

Since I had to go through a semester on formal languages, I don't really associate parsing with any kind of fun, but that's my personal problem.

coelurus · Post by **coelurus** » Fri 19 May, 2006 9:52 pm

Parsing is the worst thing I know of, since it involves heaploads of human intervention

Anyway, we understand this is a project you work on for fun (everything related to TI calcs is just for fun, right? Right?). My intention was only to clarify some input about overcoming the time-consuming bits and common pitfalls (geez, I gotta stop using those mongo-statements

).

Btw, I know hardly anything about high level languages, I've only written a script compiler "for fun".