This is the mail archive of the
guile@cygnus.com
mailing list for the Guile project.
Re: Translators and (re)interpreters
- To: Lynn Winebarger <owinebar@free-expression.org>
- Subject: Re: Translators and (re)interpreters
- From: Allover Stripes <allover@hedgehog.dyn.ez-ip.net>
- Date: 05 Jul 1999 00:58:29 -0400
- Cc: Lalo Martins <lalo@webcom.com>, guile@cygnus.com
- References: <Pine.LNX.4.10.9906301421060.19891-100000@se232.math.indiana.edu>
Lynn Winebarger <owinebar@free-expression.org> writes:
> On Wed, 30 Jun 1999, Lalo Martins wrote:
> > BTW, what happened to these? I would dare to suggest that we
> > create a translator/interpreter "task force", perhaps with its
> > own list. THis is a very ugly example of vapourware... we
> > advertise it as being one of the coolest features of Guile, but
> > it doesn't exist!!!
We do need this; what we in fact need is people who would rather
program in some other dialect than scheme, and get *them* to maintain
the translators from other dialects. Me I like Scheme; but for
interactive use I'd rather have something that makes the usual actions
shorter (I hate matching parens on the repl).
> A nice place to start would be having some standard parser generator.
> Slib has one, but it's a "Pratt's style" parser generator, and I'm not
> familiar with those.
> Any suggestions? (alternatively, documentation on a "Pratt's style"
> parser generator).
Guile already has a parser generator; they are the lang module used to
implement ctax. Both it and ctax are constantly bitrotting. Also, it
was vary slow, largely as a result of trying to be to dynamic --- the
lexer and parser were regenerated every time guile was restarted.
They were evaluated as lazily as possible, but that didn't help much.
So by all means grab another parser generator.
Andrew