I'm getting started with SLIME from CVS. It's successfully installed.
If I start SLIME, create a new buffer Lisp buffer, and define a function, completion doesn't work for that function's name until the defun is evaluated.
Is this the expected behavior?
If so, is there any way to "enable" completion for not-yet-evaluated symbols?
(More details appear following the salutation.)
Thanks,
Walter
_______________________________________________________
Details:
I've bound TAB in the following way:
(define-key slime-mode-map (kbd "<tab>") 'slime-indent-and-complete-symbol)))
If I define a function, say:
(defun square (x) (* x x))
Then type
(squ <TAB>
...the following appears in the minibuffer:
[Can't find completion for "squ"]
eof
On 5/19/10, Walter GR waltergr@aol.com wrote:
I'm getting started with SLIME from CVS. It's successfully installed.
If I start SLIME, create a new buffer Lisp buffer, and define a function, completion doesn't work for that function's name until the defun is evaluated.
Is this the expected behavior?
If so, is there any way to "enable" completion for not-yet-evaluated symbols?
Yeah, Lisp has no way of knowing about something it hasn't seen yet. However, you source code sits inside Emacs, so the Emacs completion (which searchs all buffers) will work. Try M-/ (dabbrev-expand).
Yong.
Walter GR waltergr@aol.com writes:
I'm getting started with SLIME from CVS. It's successfully installed.
If I start SLIME, create a new buffer Lisp buffer, and define a function, completion doesn't work for that function's name until the defun is evaluated.
Is this the expected behavior?
Yes.
If so, is there any way to "enable" completion for not-yet-evaluated symbols?
Currently, no. And I doubt there is any need. Slime uses lisp implementation for introspection, because it's not so easy to do static analysis on Lisp. And defining a function is as easy as pressing "C-c C-c", or "C-c C-k" so that the whole file would be compiled.