Since it's impossible (correct me if I'm wrong) to portably hide all
those little auxiliary libraries in a single file, I prefer the eval
approach (actually I don't even want to have to have a separate
library just for free-identifier-bound?, but if I'm willing to import
(rnrs eval) I can simply define free-identifier-bound? where I need
it)
Is this version better (i.e. correct as far as you can tell)? I
haven't bothered with phasing annotations - let me know what they
should be, if you know.
(library (free-identifier-bound?)
(export free-identifier-bound?)
(import (rnrs) (rnrs eval))
(define free-identifier-bound?
(let ((a:syntax (eval '(a:syntax a:syntax) (environment '(prefix
(only (rnrs) syntax) a:))))
(b:syntax (eval '(b:syntax b:syntax) (environment '(prefix
(only (rnrs) syntax) b:)))))
(lambda (id)
(unless (identifier? id) (assertion-violation
'free-identifier-bound? "not an identifier" id))
(or
(free-identifier=? id a:syntax)
(free-identifier=? id b:syntax)
(let ((sym (syntax->datum id)))
(not (or
(free-identifier=? id (datum->syntax a:syntax sym))
(free-identifier=? id (datum->syntax b:syntax sym))))))))))
On Mon, Jun 22, 2009 at 7:35 PM, Derick
Eddington<[email protected]> wrote:
>
>> (library (some suitable name)
>> (export free-identifier-bound?)
>> (import (rnrs) (rnrs eval))
>> (define free-identifier-bound?
>> (let ((empty-ctxt (eval '(syntax id) (environment '(only (rnrs)
>> syntax)))))
>> (lambda (id)
>> (unless (identifier? id)
>> (assertion-violation 'free-identifier-bound? "not an identifier"
>> id))
>> (or (free-identifier=? id #'syntax)
>> (not (free-identifier=? id
>> (datum->syntax empty-ctxt
>> (syntax->datum id)))))))))
_______________________________________________
r6rs-discuss mailing list
[email protected]
http://lists.r6rs.org/cgi-bin/mailman/listinfo/r6rs-discuss
