[Openmcl-devel] Restart strangeness

[Gary Byers]

On Wed, 23 Jun 2004, Marco Baringer wrote:

> Is there any way SLIME can determine which restarts those "catch
> all" restarts are? or could we just add a note in the description of
> the restarts? quite a few people have been wondering why SLIME
> "dies" when those restarts are selected (when in fact they are doing
> exactly what they're supposed to do).

I don't think that there is, currently.  (I'd sort of like to either
supress them or highlight them in the TTY REPL as well.)

They're established by a RESTART-CASE inside a HANDLER-CASE in the
function CCL::RUN-PROCESS-INITIAL-FORM.  The restarts could at least
be given :REPORT clauses that make them a little less tempting, e.g.:

         ...
         (abort-break () :report "Reset this process")
         (abort () :report "Kill this process" (setq exited t)))
         ...

This would provide a way for people who really want to reset/kill
a process to conveniently do so, but would presumably make it harder
to choose the wrong ABORT restart when you just want to pop out of
the current break loop.