Re: [Emacs-diffs] master 2056db3: Rationalize use of c[ad]+r, expunging cl-c[ad]\{3, 4\}r.

[Alan Mackenzie]

Hello, Stefan.

On Mon, Apr 06, 2015 at 09:03:03AM -0400, Stefan Monnier wrote:
> >> >     * emacs-lisp/cl-lib.el (cl-caaar, etc): Rename to caaar, etc.
> >> This is wrong.  cl-lib's definitions should *all* start with "cl-".
> > Hmmm.  `cl-cdaar' is a nonsense.  `eudc-cdaar' even more so.  They
> > _must_ simply be `cdaar', surely?  There are enough instances of these
> > functions in the code to matter (~311).

> No, there's nothing wrong with "cl-cXXXr".

The cl- prefix often carries overtones of "and something extra": e.g.
cl-mapcar is more general than mapcar, and so is cl-member more general
than just member.  cl-caddr is however identical to caddr.  But seeing
cl-caddr in source raises the subliminal question "what's it got that
caddr doesn't?"

> > Perhaps the answer is to put them somewhere other than cl*.el.  Maybe
> > not subr.el, but perhaps some other similar place?

> If you want them without "cl-" then they need to go to subr.el.
> As I said, I'm not yet convinced it's a good idea, but I'm not
> necessarily dead set against it.

Let's just put them all in subr.el, then.  All the cXXr definitions would
then be together.  The defaliases can stay in cl-lib.el.  I can't think
of anything that might go wrong.

>         Stefan

-- 
Alan Mackenzie (Nuremberg, Germany).