(Assuming you meant to reply-all.) Does Ryan's syntax-local-eval suggestion help you?
On Tue, Sep 4, 2012 at 4:57 PM, André Matheus <amath...@mac.com> wrote: > > > Em 04 set 2012 às 13:25, Stephen Chang <stch...@ccs.neu.edu> escreveu: > >>> I want to create a macro to get fields of an object specified by a list >>> of >>> symbols. Like >>> >>> (get-fields '(x y) obj) -> (list (get-field x obj) (get-field y obj)) >> >> I'm not sure you want a macro for this? Will map suffice? >> >> (define (get-fields flds obj) >> (map (λ (fld) (get-field fld obj)) flds)) > > Oops, I misunderstood how to use get-field (and my example happened to > work because I had a field named "fld" in my test example). > > Yes, you need a macro. One way to do it is to match the "quote" directly: > > (define-syntax (test stx) > (syntax-case stx () > [(_ (quote (fld ...)) obj) > #'(list (get-field fld obj) ...)])) > > > Thanks for everyone who told about matching the quote, it worked like you > said but > then I discovered it's not what I need, I misunderstood my problem. What I > want to > do is have a macro that in fact takes a list of symbols and get all theses > symbol from > the object, like the list of symbols that field-names returns: > > (field-names obj) > '(y x) > > I want a macro that expands like this: > > (get-fields (field-names x) obj) --> (list (get-field x obj) (get-field y > obj)) > > I don't know, maybe I'll need a function first to compute (field-names obj) > and later > call the macro? > > Frankly this compile-time/run-time thing is confusing me a little. > > Thanks and sorry for the confusion. ____________________ Racket Users list: http://lists.racket-lang.org/users
