bug#73524: 29.4; bug-reference-mode matches regexp case-insensitively

[Tassilo Horn <tsdh@gnu.org>]

Eli Zaretskii <eliz@gnu.org> writes:

>> >> If we want to always match case-insensitively in bug-reference,
>> >> then we could bind case-fold-search to t in bug-reference-fontify,
>> >> no?
>> 
>> > Sure, if we are convinced that there's really no scenario where bug
>> > references are case-sensitive.  I would think this is very uncommon
>> > but not impossible.
>> 
>> What about the other examples in the default, i.e. "RFE" and "PR"?
>> I'm not completely convinced that matching their lower-case form
>> would be a good thing.
>
> I see in binutils-gdb repository most "PR" are in upper-case, but
> all.
> [...]
> Not sure about RFE (where is that used?)

Request for enhancement?

Anyway, I also think that bug#123 is probably better matched
case-insensitively while RFC 2616 or CVE-2024-5742 are commonly
all-caps.  But in the end it's a user's choice if she prefers to have a
false positive or a false negative when something is or just looks like
a bug reference.  I'd rather have a false positive but that's just me.

So I'd vote for a defcustom bug-reference-case-fold-search defaulting to
t that users can set according to their preference or conventions.

And I wouldn't read too much into the default value of
bug-reference-bug-regexp.  It's just a value that works for us (Emacs or
other GNU projects using debbugs).

Bye,
  Tassilo