This is the mail archive of the
guile@sourceware.cygnus.com
mailing list for the Guile project.
Re: Inconsistant Results
- To: Mikael Djurfeldt <mdj at mdj dot nada dot kth dot se>
- Subject: Re: Inconsistant Results
- From: Dirk Herrmann <dirk at ida dot ing dot tu-bs dot de>
- Date: Thu, 22 Jun 2000 11:08:21 +0200 (MEST)
- cc: Jim Blandy <jimb at savonarola dot red-bean dot com>, Guile Mailing List <guile at sourceware dot cygnus dot com>
Probably another question that has been discussed several times,
but: Wouldn't it be better anyway to have a separate macro expansion
phase than to perform this on the fly during evaluation? This would
become necessary anyway if a compilation phase (to bytecode or treecode of
whatever) was to be added.
The current problem, as I see it, comes (partly) from the fact, that macro
expansion is not performed as a separate phase. If it was, the occurence
of a macro during evaluation would be an error, because for the evaluator
no macros would exist anymore.
Are threre special reasons _not_ to separate the macro expansion, or is it
just because we are waiting for someone (who is known to be a lazy
bastard) to do it?
Best regards
Dirk