On Wed 18 May 2011 16:39, Alaric Snell-Pym <alaric@snell-pym.org.uk> writes:

> In which case, the re-raise if no clause matches would rely on having
> preserved the original dynamic state by keeping a copy of the
> continuation of RAISE around. Or, rather, a continuation captured just
> within RAISE before the handler is invoked that, if it actually
> continues, causes the re-raise - having re-wound the dynamic state...

That is crazy :)  We're not talking just about parameters; there are
dynamic-wind guards, etc to think about, and rewinding those does not
sound like something that you want to do as part of your error-handling
mechanism.

Could this not be a bug in the SRFI-34 spec?

  http://srfi.schemers.org/srfi-34/mail-archive/msg00013.html

Andy
--
http://wingolog.org/

_______________________________________________
Scheme-reports mailing list
Scheme-reports@scheme-reports.org
http://lists.scheme-reports.org/cgi-bin/mailman/listinfo/scheme-reports