First of all, #lang racket is not Scheme, nor does it claim to be. However, #lang r5rs does attempt to be Scheme, and it exhibits the same behavior as #lang racket in this case, so your question is still valid. I would argue that, in this case, Racket gets it right, and Bigloo and Kawa are noncomformant.
In both R5RS and R7RS, the way syntax-rules matches “literals” is quite explicit[1]: “[…] an input expression E matches a pattern P if and only if […] P is a literal identifier and E is an identifier with the same binding […]” The wording “with the same binding” is important. This means literals must not be matched by datum (that is, by their name) but by their lexical binding. When you redefine `else`, you have created a new binding that is distinct from the binding (or lack thereof) provided by the Scheme standard library. The `cond` form is listed as a “derived expression”, and an implementation of `cond` using `syntax-rules` is provided by the report[2]. However, it does not appear that Bigloo or Kawa properly implement the same binding restriction mandated by the standard, so they are nonconformant. I also tested two other Schemes, somewhat randomly chosen. Chibi Scheme exhibits the same incorrect behavior as Bigloo and Kawa, but Larceny has the same behavior as Racket. It would be interesting to see what other Schemes do, but the practical takeaway seems to be that this part of the standard is inconsistently implemented, and you cannot depend on it if you want to (for whatever reason) write portable Scheme. [1]: http://docs.racket-lang.org/r5rs/r5rs-std/r5rs-Z-H-7.html#%_sec_4.3.2 [2]: http://docs.racket-lang.org/r5rs/r5rs-std/r5rs-Z-H-10.html#%_sec_7.3 > On Oct 27, 2016, at 15:46, Damien Mattei <damien.mat...@gmail.com> wrote: > > Hi, > > why this does not work with DrRacket: > > #lang racket > (define-syntax then > (syntax-rules () > ((_ ev) ev) > ((_ ev ...) (begin ev ...)))) > > (define-syntax else > (syntax-rules () > ((_ ev) ev) > ((_ ev ...) (begin ev ...)))) > > (if #f > (then > 1 > 2 > 3) > (else > 3 > 4 > 5)) > > (cond (#f 'never) > (else 'always)) > > . else: bad syntax in: else > > it works with others schemes as in bigloo or kawa: > > ------------------------------------------------------------------------------ > Bigloo (4.1a) > ,--^, > `a practical Scheme compiler' _ ___/ /|/ > > Sam 26 sep 2015 04:59:46 CEST ,;'( )__, ) ' > > Inria -- Sophia Antipolis ;; // L__. > > email: big...@lists-sop.inria.fr ' \ / ' > > url: http://www-sop.inria.fr/indes/fp/Bigloo ^ ^ > > ------------------------------------------------------------------------------ > > > 1:=> (define-syntax then > (syntax-rules () > ((_ ev) ev) > ((_ ev ...) (begin ev ...)))) > #unspecified > 1:=> (define-syntax else > (syntax-rules () > ((_ ev) ev) > ((_ ev ...) (begin ev ...)))) > #unspecified > 1:=> (if #f > (then > 1 > 2 > 3) > (else > 3 > 4 > 5)) > 5 > 1:=> (cond (#f 'never) > (else 'always)) > always > > Damien -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
