bug#10475: 24.0.92; `C-h v' displays `*' indicating user option

[Drew Adams <drew.adams@oracle.com>]

> > (defcustom foo nil
> >   "*A foobar function or nil."
> >   :type '(choice (const :tag "None" nil) function)
> >   :group 'convenience)
> >
> > Then `C-h v foo'.  The `*' indicating that this is a user variable
> > (e.g. usable by `set-variable') is printed as if it were part of the
> > doc.  It should be removed.
> 
> I don't think so.  Instead we should remove all the superfluous "*"s in
> the source code.  I'll submit a bug report.
> 
> > In addition, `user-variable-p' is the test used for functions such as
> > `set-variable', and it`user-variable-p' recognizes the `*' as defining a
> > user variable.  This is correct behavior.  The only problem is that
> > `describe-variable' should not treat this `*' as if it were part of the
> > doc text.
> 
> It no longer does that.  It's now an alias for custom-variable-p.

You're missing the point.  "It no longer does that..."
Existing code written (and usable) in earlier releases,
where `*' here DOES have that meaning and effect, should
not be handled by recent Emacs differently wrt `C-h v'.

`C-h v' should behave the same across releases in this regard.