Brian Harvey scripsit:
> > Get (or configure) a small Scheme that's -%bignums -%inexact.
>
> I submit that this whole configurable options thing doesn't look
> very Schemely.
You've way over-interpreted me. The + and - and % are metalanguage for
talking about feature groups. In practice, you'd do:
$ ./configure --nobignums --noratios
or maybe
$ myscheme -B -R
or maybe
$ ittybittyscheme
which you know has neither bignums or ratios.
> If you want a uniform interface for this sort of thing, it should be
> (feature-present? 'bignums)
> or
> (set-behavior! 'auto-promote-bignum #f)
Actually, I'm proposing SRFI 0:
(cond-expand
((or bignums ratios)
(error "Your Scheme does the Wrong Thing"))
(else
"Carry on, Sergeant!"))
> I gather that the point of the % business is that you want to make
> these decisions at compile time.
No, it's just metalanguage for use in English, to distinguish bignums,
the numbers, from %bignums, the feature of supporting bignums.
> But I submit that in the WG1-Edu model, at
> least, there's no such thing as "compile time."
There is still macroexpand time, which is a degenerate form of compile
time.
> Just the beautiful
> yin-yang mutual recursion of eval and apply. :-)
And there is no reason why WG1-Edu systems can't compile their inputs.
--
Eric Raymond is the Margaret Mead John Cowan
of the Open Source movement. [email protected]
--Bruce Perens, http://www.ccil.org/~cowan
some years ago
_______________________________________________
r6rs-discuss mailing list
[email protected]
http://lists.r6rs.org/cgi-bin/mailman/listinfo/r6rs-discuss
