[Openmcl-devel] Bug in HANDLER-CASE?

[Gary Byers]

On Fri, 25 Aug 2006, Kai Kaminski wrote:

> [I already sent this mail a few minutes ago, only to find out that I'm
> subscribed with a different email address. Sorry for the
> inconvenience.]
>
> Hi,
>
> I just tried to use the latest Slime with OpenMCL 1.0 (ppc32) and it
> choked on a HANDLER-CASE form in swank.lisp. The problem is that the
> :NO-ERROR clause doesn't come last. A minimal test case is
>
> (handler-case (read)
>  (:no-error (c) (print c))
>  (t (c) (print c)))
>
> As far as I understand the Hyperspec, this should be legal (see the
> section on HANDLER-CASE and Section 1.4.1.2.1).
>

Yes; it looks like it's going to great lengths to insist that :NO-ERROR
appear last. (It'd be reasonable to insist that it appear at most once.)

> I'll send a patch for Slime to slime-devel shortly.
>
> Since this is - as far as I remember - my first mail to this list, I'd
> like to thank everyone involved with OpenMCL's development for their
> great work.

Except, of course, for whoever wrote HANDLER-CASE (:)).

>
> Thanks,
> Kai
> _______________________________________________
> Openmcl-devel mailing list
> Openmcl-devel at clozure.com
> http://clozure.com/mailman/listinfo/openmcl-devel
>
>