On Wed, Mar 28, 2018 at 10:32 AM, John Cowan <xxxxxx@ccil.org> wrote:


On Wed, Mar 28, 2018 at 3:57 PM, Shiro Kawai <xxxxxx@gmail.com> wrote:

Oops, you're right about unfulfilled cond-expand - R7RS states if no conditional is satisfied and there's no else clause, the behavior is unspecified.

If I remember correctly, that's because there's an existing divergence:  some implementations do nothing, others raise an error.  So you should explicitly say (else #t) or (else (syntax-error "unsupported implementation")), whichever makes sense in a particular context.
 

I see. That's unfortunate. srfi-0 requires unfulfilled feature conditional to be an error (although 'error' itself has broad meaning in R7RS).
We could move toward clarifying it in future srfis/rnrs.