Re: Language Processor Generators

joshua@veritas.com (Joshua Levy)
Thu, 20 Feb 1992 20:40:38 GMT

          From comp.compilers

Related articles
Language Processor Generators ckminer@lance.colostate.edu (1992-02-18)
Re: Language Processor Generators pardo@cs.washington.edu (1992-02-19)
Re: Language Processor Generators joshua@veritas.com (1992-02-20)
| List of all articles for this month |
Newsgroups: comp.compilers
From: joshua@veritas.com (Joshua Levy)
Keywords: tools
Organization: VERITAS Software
References: 92-02-086
Date: Thu, 20 Feb 1992 20:40:38 GMT

ckminer@lance.colostate.edu (Chris Miner) writes:
>Lately there has been talk of little languages, and their compilers. I have
>also heard reference to the usefulness of Lisp, Lex and Yacc in compiler
>construction. ... With this mind, I though it would be good to have a tool
>which constructs a language processor, LP, from a specification of the
>language.


There is another way of attacking the whole problem. Language processors
do not need to be compilers at all, they could be interpreters. Such
interpreters can often be consructed using already existing, designed to
be extended, interpreters. I often use TCL (see comp.lang.tcl) for this
purpose, but elk also exists.


I have found that I can create a TCL based little language interpreter in
much less time than it would take me to build a compiler for the language,
IF the language fits well into what TCL already provides. This is because
TCL provides variables and control structures "for free" and is designed
to be extended.


Note that since all little languages built this way retain many properties
of the base interpreter, they made not be suitable for your research.


Joshua Levy (joshua@veritas.com)
--


Post a followup to this message

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