Per Bothner scripsit: > Kawa isn't quite consistent here. I think it is plausible > to make (set! bar (set! foo 32)) be a compile-time error (or at > least warning), since it is very likely to a bug, and unlikely > to be actually useful.
Indeed, it's not the best example: a better one is a procedure that returns an undefined value, rather than a primitive syntax form. The idea is that it should be possible to call an unknown procedure without having to defend oneself against receiving multiple values quite as carefully as would be required if all mutators truly returned no values (as opposed to a no-value value). There are still things like exact-integer-sqrt, of course, that can trip you up. > if we didn't have to worry about raise-continuable. If I were you, I'd just say that raise-continuable is not supported. It's part of the close JVM integration that Kawa is intended to achieve. -- A poetical purist named Cowan [that's me: [email protected]] Once put the rest of us dowan. [on xml-dev] "Your verse would be sweeter http://www.ccil.org/~cowan If it only had metre And rhymes that didn't force me to frowan." [overpacked line!] --Michael Kay _______________________________________________ Scheme-reports mailing list [email protected] http://lists.scheme-reports.org/cgi-bin/mailman/listinfo/scheme-reports
