Re: C++ qualified method names

[Eli Zaretskii <eliz@gnu.org>]

> From: Yuri Khan <yuri.v.khan@gmail.com>
> Date: Fri, 26 May 2017 17:15:40 +0700
> Cc: Alan Mackenzie <acm@muc.de>, Emacs developers <emacs-devel@gnu.org>
> 
> >   void
> >   regcache::raw_supply_integer (int foo)
> >   {
> >     whatever (foo);
> >   }
> >
> > invoking "M-: (c-defun-name) RET" produces "raw_supply_integer".  Why
> > does it exclude the class qualification?  Is there a way to have that
> > included?  This is important e.g. when producing a ChangeLog entry for
> > such methods.
> 
> This is a good question that opens a can of worms.
> 
> What if the clasess is in a namespace? Do you want to include the
> namespace, too? …the whole chain of namespaces up to the root? What if
> some of them are inline namespaces?
> 
> What if the method is overloaded on argument types? …on method
> cv-qualifiers and ref-qualifiers?
> 
> What if the thing on which (c-defun-name) is a function template? …a
> class template? …an explicit specialization? …a partial
> specialization?

Since it is easier to delete than to type, I guess my answer would be
"all of the above".