At Mon, 8 Jul 2013 17:01:24 -0400, Sam Tobin-Hochstadt wrote: > On Mon, Jul 8, 2013 at 2:39 PM, <mfl...@racket-lang.org> wrote: > > > --- > OLD/pkgs/typed-racket-pkgs/typed-racket-lib/typed-racket/typecheck/tc-toplevel.r > kt > > +++ > NEW/pkgs/typed-racket-pkgs/typed-racket-lib/typed-racket/typecheck/tc-toplevel.r > kt > > @@ -392,7 +392,8 @@ > > (match a > > [(list from to) > > #`(add-alias (quote-syntax #,from) (quote-syntax > #,to))])))) > > - (begin-for-syntax (add-mod! (quote-module-path)))) > > + (begin-for-syntax (add-mod! > > (variable-reference->module-path-index > > + (#%variable-reference))))) > > #`(begin > > #,(if (null? (syntax-e #'(new-provs ...))) > > #'(begin) > > One thing I liked about the old code here is that > `(quote-module-path)` has easier-to-explain semantics. Should it be > changed, or should there be some other form in `syntax/location` that > does what you've done here?
I don't think that `(quote-module-path)' should change. A new form in `syntax/location' or some other library would make sense. I didn't add `(quote-module-path-index)', because a module-path index cannot be marshaled in the same way as the other `(quote-....)' results, and so I wasn't sure that it belonged there or what it should be called. _________________________ Racket Developers list: http://lists.racket-lang.org/dev
