[slim-vim] thanks, feedback

[Larry Clapp]

On Fri, Aug 11, 2006 at 01:50:48PM +0100, Serafeim Zanikolas wrote:
> Friday 11 August 2006 13:25, Larry Clapp:
> > <leader>x is the Slim-vim command and talks to the remote Lisp,
> > and doesn't update the --lisp-repl-output-- buffer. ??<leader>ee
> > is the ecl-repl command and talks to the embedded Lisp, and
> > updates the --lisp-repl-output-- buffer.
> 
> This may sound naive but I can't help asking: why is there a need
> for both an embedded lisp and a remote lisp? Is it possible to use
> only the embedded lisp? (I see that the bindings currently get setup
> only after connecting to swank)

It's not naive.  You *can* use only the embedded Lisp to develop Lisp
projects, and I've done a lot of small-scale fiddling with Lisp using
just the embedded ECL.

But the embedded ECL has few debugging facilities.  Vim is
single-threaded and not terribly re-entrant, so anytime you call an
ECL function you have to wait for it to return before you can interact
with the user.  (Or, alternatively, we just don't know how to do that
yet.  Vim7 has a function that I don't think vim6 had, called confirm,
that might come in handy.  But even then, you still can't do like
Emacs does with Elisp or with Slime where you can enter a "recursive
edit" and have all the facilities of Vim + ECL at your fingertips.)

So yes, you can do a lot of Lisp development with the ECL embedded in
Vim, as long as you don't mind a not-very-Lispy approach to
testing/debugging code, and you don't mind the risk that your code
will crash the editor.

If you want a more Lispy approach to developing a Lisp project, well,
use Slim-Vim.  :)

-- Larry