[slim-vim] the slime- prefix

[Brad Beveridge]

On 26/07/06, Larry Clapp <larry at theclapp.org> wrote:
> On Tue, Jul 25, 2006 at 08:48:10AM -0700, Brad Beveridge wrote:
> > On 25/07/06, Larry Clapp <larry at theclapp.org> wrote:
> > > A note on the "slime-" prefix naming convention (slime-connect,
> > > slime-disconnect, slime-this, slime-that): slime.el uses that
> > > naming convention because Elisp doesn't have packages.  Since we
> > > do, I would rather just use the package name: slime:this and
> > > slime:that.  It bugs me to have to say slime:slime-this and
> > > slime:slime-that, you know?
> > >
> > > It *has* occurred to me that keeping the exact same function names
> > > *may* make it easier to keep in sync with slime.el, so I
> > > acknowledge that reasonable people might disagree and that making
> > > such a change involves trade-offs.
> > >
> > > What does the group think?
> >
> > I could probably be convinced either way.  Initially I wanted to
> > mimic slime.el where possible, but I think that packaging is the way
> > to go.  However, even within the package I think I would like to use
> > fully qualified names.
> >
> > eg
> > (defpackage #:slime
> >  (:export #:connect))
> > (in-package #:slime)
> > (defun slime:connect () ....)
> >
> > That way you can grep slime.el and slime.lisp for functions with the
> > regular expression slime.foo, and match both.
>
> That works for me.  Good thinking.
>
> In general, I think a package shouldn't fully-qualify its own symbols,
> but in this case I think it serves a purpose.  Possibly this will fail
> the test of time, but for now it seems like a good compromise.

Actually, I'm going to change my mind again :)
It turns out that in order to do slime:foo, even internally to the
package, we would need to export foo.  That's fine for the external
functions, but not for every single function.  So I am now in favour
of dropping the SLIME- prefix entirely.

Cheers
Brad