lisp-completion question

lisp-completion question

[Thierry Volpiatto]

Hi,

It seem `lisp-completion-at-point' is no more returning :predicate,
(accessible in emacs-24.3 with (plist-get plist :predicate))
is it wanted?
Note: `lisp-complete-symbol' is still using this with no effect of course. 

In `lisp--local-variables' e.g &optional is returned, is it wanted?
otherwise, the patch below fix it:

diff --git a/lisp/emacs-lisp/lisp.el b/lisp/emacs-lisp/lisp.el
index b37a811..af8433a 100644
--- a/lisp/emacs-lisp/lisp.el
+++ b/lisp/emacs-lisp/lisp.el
@@ -725,7 +725,7 @@ considered."
         (delq nil
               (mapcar (lambda (var)
                         (and (symbolp var)
-                             (not (string-match (symbol-name var) "\\`[&_]"))
+                             (not (string-match "\\`[&_]" (symbol-name var)))
                              ;; Eliminate uninterned vars.
                              (intern-soft var)
                              var))


Thanks.

-- 
Thierry
Get my Gnupg key:
gpg --keyserver pgp.mit.edu --recv-keys 59F29997 

Re: lisp-completion question

[Stefan Monnier]

> It seem `lisp-completion-at-point' is no more returning :predicate,
> (accessible in emacs-24.3 with (plist-get plist :predicate))
> is it wanted?

It sometimes does, sometimes doesn't.  Do you see a difference in the
resulting behavior?

> In `lisp--local-variables' e.g &optional is returned, is it wanted?

No.

> otherwise, the patch below fix it:
> diff --git a/lisp/emacs-lisp/lisp.el b/lisp/emacs-lisp/lisp.el
> index b37a811..af8433a 100644
> --- a/lisp/emacs-lisp/lisp.el
> +++ b/lisp/emacs-lisp/lisp.el
> @@ -725,7 +725,7 @@ considered."
>          (delq nil
>                (mapcar (lambda (var)
>                          (and (symbolp var)
> -                             (not (string-match (symbol-name var) "\\`[&_]"))
> +                             (not (string-match "\\`[&_]" (symbol-name var)))
>                               ;; Eliminate uninterned vars.
>                               (intern-soft var)
>                               var))

Very much so, indeed, thanks for catching this typo.  Feel free to
install that patch,


        Stefan

Re: lisp-completion question

[Thierry Volpiatto]

Hello Stefan,

Stefan Monnier <monnier@IRO.UMontreal.CA> writes:

>> It seem `lisp-completion-at-point' is no more returning :predicate,
>> (accessible in emacs-24.3 with (plist-get plist :predicate))
>> is it wanted?
>
> It sometimes does, sometimes doesn't.  Do you see a difference in the
> resulting behavior?

Yes, for me here it is never returned and as result, (predicate is nil)
I get all the unbounded symbol in the resulting lists, so I dont use
anymore this function, I must use my own completing function with a
predicate determined by a function based on emacs-24.3 code.

> Very much so, indeed, thanks for catching this typo.  Feel free to
> install that patch,

Please, do it, I don't use anymore bzr.

Thanks.

-- 
Thierry
Get my Gnupg key:
gpg --keyserver pgp.mit.edu --recv-keys 59F29997 

Re: lisp-completion question

[Stefan Monnier]

> I get all the unbounded symbol in the resulting lists,

Sounds like a bug, indeed.  Can you give me a concrete example?

>> Very much so, indeed, thanks for catching this typo.  Feel free to
>> install that patch,
> Please, do it, I don't use anymore bzr.

Presumably you can now use Git to access the Bzr repository.


        Stefan

Re: lisp-completion question

[Thierry Volpiatto]

Stefan Monnier <monnier@IRO.UMontreal.CA> writes:

>> I get all the unbounded symbol in the resulting lists,
>
> Sounds like a bug, indeed.  Can you give me a concrete example?

;; emacs-24.3

helm!
M-: (lisp-completion-at-point)
=>[...]
    :predicate
    #[(sym)
      "\301\b!\206

Re: lisp-completion question

[Thierry Volpiatto]

Sorry, the previous mail was truncated.

;; emacs-24.3

helm!
M-: (lisp-completion-at-point)
=>[...]
    :predicate
    #[(sym)
      "\301\b!\206.

Re: lisp-completion question

[Stefan Monnier]

> helm!
> M-: (lisp-completion-at-point)
> =>[...]
>     :predicate
>     #[(sym)
>       "\301\b!\206\302\b!\206\303\b!\207"
>       [sym boundp fboundp symbol-plist]
>       2]

No, I really mean a concrete example of difference in the resulting set
of candidates.  Whether a predicate is used or not is an internal detail.


        Stefsan