Follow-up Comment #15, bug #66419 (group groff): I haven't heard anything more from Rob about the shape of my proposed fix (comment #13).
> I think I'd prefer instead to have a new entry point in libgroff called
> `font::open_resource()` to make it clearer what's going on at the call site,
> and omit the slash check from that. Both it and `font::open_file()` could be
> backed by a static function using an additional `bool` argument to enable the
> slash-rejecting logic. (I'm thinking about setting `errno`, too, since there
> are 2 other failure modes arising from invariant violations: the pointers
> `nm` or `device` could be null.)
So I reckon that is what I will try to do.
This approach would obviate the one I spitballed in comment #9, for _grops_ to
support a new `-U` flag, for the user to invoke _groff_ with it, and for
_groff_ to add logic to pass that option on to the output driver.
_______________________________________________________
Reply to this item at:
<https://savannah.gnu.org/bugs/?66419>
_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/
signature.asc
Description: PGP signature
