Re: What is the future of Compiler technology?

On Sun, 16 Jul 2006, Oliver Zeigermann wrote:


"[..]


[..] However, there are quite a number of mini-languages and/or
mini-parsers needed. You would not want to hire a compiler/parser
designer for such a task, but you still want some maintainable code from
your "ordinary" programmer. This requires compiler generators to be
easier to use and understand."


LL(k) compiler generators are easy to use and understand. Maintaining
a hacked together parser by someone who does not understand parsing might
not be easy to use or understand.


" Who *really* understands yacc?


[..]"


People who have studied LALR(k) theory.


John R. Levine wrote:


"[Anyone who's taken a decent undergraduate compiler course should
understand
yacc. [..]"


Is an undergraduate compiler course dealing with LALR(k) and LL(k)
theoretically but whose practical parsing coursework assignments are
exclusively LL(k) not decent?


I was surprised a few years ago at how ignorant senior professionals in
the Association of C & C++ Users were of long established parsing theory.
[If the coverage of LALR parsing isn't good enough to understand how a
yacc parser works, I'd say no, it's not decent. LR parsers just
aren't that complicated, a regex-like state machine and a state
stack. -John]

Return to the comp.compilers page.
Search the comp.compilers archives again.