[slim-vim] first try, some issues

[Brad Beveridge]

On 24/07/06, Larry Clapp <larry at theclapp.org> wrote:
> On Sun, Jul 23, 2006 at 08:56:12PM -0700, Brad Beveridge wrote:
> > > I would like to see a couple minor usability enhancements.
<SNIP>
> I saw that you fixed this yourself, but I'd like to discuss a
> different approach.
>
> You wrapped each mapping in a call to VIM-SLIME::TEST-CONNECTED.
You're right, that was a poor way to do it - lazy me.  Testing for
slime-connected-p is the correct way (and, huh, even an existing
function!)
Defining a def-toplevel-command macro and renaming the effected
functions is probably the way to go.  The tests probably ought to live
largely in the Slime code, so that other code bases that use SlimVim
won't need to add their own stuff.  We could possibly check for
connectedness as low as SLIME-DISPATCH-EVENT.

> (I can certainly make these changes myself and send you a patch, but
> slim-vim has been your baby so far, and I thought that would be rude.)
Please do, and the other patches.  Slim-vim was never meant to be mine
alone, so patch away - I play well with others :)

> Patch in hand for part of this; it doesn't deal with the remote Lisp
> closing the socket yet.  See :help VimLeave.
I'm not sure how Slime deals with this - not very gracefully I
suspect.  In fact, to close a connection it appears to just kill the
process!

With these set of patches in mind, I think we are shaping up pretty
nicely - enough for an alpha release early next month perhaps.

Cheers
Brad