On Mon, Jan 2, 2012 at 11:31 PM, Alex Shinn <[email protected]> wrote:

> On Tue, Jan 3, 2012 at 1:01 PM, John Cowan <[email protected]> wrote:
> > Stefan Edwards scripsit:
> >
> >> What about loading alternative implementation of a feature, such as
> >> falling back onto the reference implementation of a SRFI if the Scheme
> >> you're using doesn't provide one? I actually think this would ease
> >> writing libraries likes SLIB or Nausicaa that support multiple Scheme
> >> implementations and have to Spackle the gaps in features between them.
> >
> > If possible that should be done using cond-expand in the module
> > language.
>
> Exactly.  That's the primary (and currently only) use of features.
> We're talking about potential secondary uses.
>

Perhaps I'm wrong, but I thought SRFI-0/7 only worked on features specified
by SRFIs, and not
things such as POSIX interaction, non-SRFI-based threading, implementation
specific module
styles, &c. Even just having a bog-standard methods for accessing the
implementation name
would help, I think.


> --
> Alex
>



-- 
====
Q. How many Prolog programmers does it take to change a lightbulb?
A. No.
_______________________________________________
Scheme-reports mailing list
[email protected]
http://lists.scheme-reports.org/cgi-bin/mailman/listinfo/scheme-reports

Reply via email to