Daphne,

On 2024-10-24 21:04 +0200, Daphne Preston-Kendal wrote:
> Why is the use of restarters limited to assertion violations?

Only the restarters installed by the ‘restartable’ forms are currently
restricted to handling assertion violations, because we conceived them
as allowing the user to retry application with new arguments. There
is, however, no reason that I can see for limiting them to &assertion
conditions. ‘serious-condition?’ seems like a good default criterion
to me, whether or not we end up allowing restarters to select which
condition types to restart.

‘restarter-guard’ says nothing about the type of the triggering
exception’s condition.

The broader question (noted in the Issues) is whether it’s useful for
a restarter to handle only certain conditions. I tend to think it is,
but it may be tricky to work this feature into the SRFI 255 syntax.

Regards,

Wolf

--
Wolfgang Corcoran-Mathe  <xxxxxx@sigwinch.xyz>