In data giovedì 12 settembre 2024 10:31:56 CEST, Eli Zaretskii ha scritto:
> > From: Yuan Fu <casouri@gmail.com>
> > Date: Thu, 12 Sep 2024 00:51:20 -0700
> > Cc: Vincenzo Pupillo <v.pupillo@gmail.com>,
> > 
> >  Bug Report Emacs <bug-gnu-emacs@gnu.org>,
> >  72814@debbugs.gnu.org
> >  
> > > On Sep 11, 2024, at 5:05 AM, Eli Zaretskii <eliz@gnu.org> wrote:
> > >> From: Vincenzo Pupillo <v.pupillo@gmail.com>
> > >> Cc: Bug Report Emacs <bug-gnu-emacs@gnu.org>, 72814@debbugs.gnu.org
> > >> Date: Wed, 11 Sep 2024 11:40:17 +0200
> > >> 
> > >>> Thanks for taking this up! I won’t signal a warning if doxygen grammar
> > >>> isn’t found. Imagine a user without doxygen grammar, and din’t change
> > >>> c-ts-mode-enable-doxygen: they’ll get a warning whenever they open a
> > >>> C file. We should either set c-ts-mode-enable-doxygen to nil by
> > >>> default, or not warn when doxygen grammar doesn’t exist. Otherwise,
> > >>> the patch looks good to me.> >> 
> > >> Okay, fine. But Eli said something different...
> > > 
> > > Not necessarily.  All I said was "issue a user-friendly diagnostic".
> > > We could output the message into *Messages*, and do it only once per
> > > Emacs session.
> > 
> > Another idea: use tertiary value for c-ts-mode-enable-doxygen, t for
> > enable (and issue warning if doxygen isn’t present), nil for enable if
> > doxygen exists but don’t issue any warning, ‘disabled for disable. (I
> > don’t think there’s a convention for the three values of tertiary toggle,
> > is that right Eli?)
> That could also be OK, but I prefer something that does TRT by
> default.

Ciao,
in this new version of the patch, as default, the *-ts-mode-enable-doxygen are 
set to nil, and issue a diagnostic (one per session) when Not-nil and if the  
grammar isn't present. It is OK?

Vincenzo