Manuel Serrano Manuel.Serrano
Mon Apr 11 06:14:20 EDT 2005

I'm currently working on an alternative proposal for SYNTAX-CASE. Presently,
I have no idea if this effort will yield to something useful or not. 
I'm coming to you because I have a though that I would like to share with you.

In Bigloo, I'm using a lot of macros (expanders more precisely) for
optimizing the compiled code. These expanders look like:

(define-expander + 
   (lambda (x e)   ;; x is the form to be macro-expansed, e is an expander
      (match-case x
         ((?- (and (? integer?) ?v1) (and (? integer?) ?v2))
          (+ v1 v2))
         ((?- (and (? fixnum?) ?v1) (and (? fixnum?) ?v2))
          `(+fixnum ,v1 ,v2))
         ((?- (and (? flonum?) ?v1) (and (? flonum?) ?v2))
          `(+flonum ,v1 ,v2))
         ((?- ?v1 . ?vs)
          `(+ ,(e v2 e) ,(e `(+ , at vs) e))))))
(I'm not pretending that this example is 100% correct, I'm just illustrating).
As far as I understand, such transformations cannot be expressed 
using SYNTAX-CASE. Am I wrong?


More information about the R6RS mailing list