At Fri, 2 Dec 2016 10:26:40 -0800, Matthew Butterick wrote:
> On Dec 1, 2016, 5:33 PM -0800, Matthew Flatt <mfl...@cs.utah.edu>, wrote:
> 
> > Assuming that "without lexical context" means "no scopes"[*], I don't
> > think there's a simpler way right now than walking through the syntax
> > object and making sure that each scope set is empty. And maybe there's
> > no easier way to check for an empty scope set than using
> > `bound-identifier=?` on an identifier with a scope set to test and an
> > identifier that definitely has an empty scope set.
> 
> Specifically, what I was trying to do was write a contract for the return 
> value of `read-syntax` that's narrower than just `syntax?` (which is 
> insufficient, since `read-syntax` has to return a stx object without 
> bindings).
> 
> One can make such a stx object with `(datum->syntax #f ···)` or 
> `(strip-context ···)`. So the mystery predicate here would return #t for stx 
> objects made either of those ways.

Just to make sure we're on the same page, this is the function that I
had in mind:

(define (empty-scopes-everywhere? e)
  (cond
   [(syntax? e)
    (and (bound-identifier=? (datum->syntax #f 'x)
                             (datum->syntax e 'x))
         (empty-scopes-everywhere? (syntax-e e)))]
   [(pair? e) (and (empty-scopes-everywhere? (car e))
                   (empty-scopes-everywhere? (cdr e)))]
   [(vector? e) (for/and ([elem (in-vector e)])
                  (empty-scopes-everywhere? elem))]
   [(box? e) (empty-scopes-everywhere? (unbox e))]
   [(prefab-struct-key e)
    (empty-scopes-everywhere? (struct->vector e))]
   [else #t]))


> PS is the name of `strip-context` obsolete? Should it be `strip-scope-sets`?

Maybe so. In retrospect, I like how "lexical context" is less connected
to an implementation, but the the different layers of terminology don't
line up precisely.

-- 
You received this message because you are subscribed to the Google Groups 
"Racket Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-dev+unsubscr...@googlegroups.com.
To post to this group, send email to racket-dev@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/racket-dev/584377c7.c952620a.95b47.061bSMTPIN_ADDED_MISSING%40gmr-mx.google.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to