tags 24618 + patch
quit

Philipp Stephani <p.stephani2@gmail.com> writes:

> Consider the following code:
>
> (condition-case err
>     (signal 'does-not-exist '(1 2))
>   (error (print err)))
>
> The signal is not caught by condition-case because it has no error
> conditions.  This makes it impossible to reliably catch all signals
> (without abusing the debugger).

Yeah, I noticed this when I started looking at fixing ert to stop
abusing the debugger like this (Bug#30745 and Bug#11218).

> I propose that an error condition of 't' in `condition-case' should be
> interpreted as 'all conditions'.

Seems easy enough: