This is the mail archive of the
guile@sources.redhat.com
mailing list for the Guile project.
Re: On #&optional, etc
- To: Marius Vollmer <mvo at zagadka dot ping dot de>
- Subject: Re: On #&optional, etc
- From: Maciej Stachowiak <mjs at eazel dot com>
- Date: 14 Aug 2000 16:27:11 -0700
- Cc: guile at sources dot redhat dot com
- References: <87lmxz8mpk.fsf@zagadka.ping.de>
Marius Vollmer <mvo@zagadka.ping.de> writes:
> Hi,
>
> look was Kent Pitman has to say about &optional in Common Lisp.
>
> > Newsgroups: comp.lang.lisp
> > From: Kent M Pitman <pitman@world.std.com>
> > Subject: Re: RFC: Lisp/Scheme with less parentheses through Python-like
> significant indentation?
> > Message-ID: <sfwsns7ltwv.fsf@world.std.com>
> > Date: Mon, 14 Aug 2000 15:54:56 GMT
> >
> > CL made a horrible error in having the &keywords at all, and Dylan copied the
> > idea to #keyword over my offered objections at the appropriate design time.
> >
> > [...]
> >
> > In Lisp, for example, (defun foo (x :optional y) ...)
> > would have been better than &optional, and would have cleared the namespace
> > of an irritating hole that it now has, where some symbols are inelegantly
> > dependent on their names.
>
> This reminded me that Guile has in (ice-9 optargs) the special object
> #&optional, etc, which are neither symbols nor keywords. I want to
> remove them and replace them with proper keywords like #:optional.
> [I'll think about a tranbsition mechanism.]
Transition should be feasible by just making the #& versions read as
the corresponding keywords, but deprecating this reader extension.
> Comments?
As the original author of the optargs packahe, I think it sounds good.
- Maciej