[COMMIT] Differentiate between symbol variables and functions, find-func

[Vladimir G. Ivanovic]

on 12/27/2008 01:41 AM Stephen J. Turnbull said the following:
> Vladimir G. Ivanovic writes:
> 
>  > 2. The construct "#'" is not Emacs Lisp, but a Common Lisp
>  >    extension.
> 
> It's also in XEmacs Lisp, which has adopted a number of Common Lisp
> primitives.  See src/lread.c, line 2560 (21.5 from Mercurial).

OK, then the docs are out of date. I'll submit a patch.

> 
>  >    Is the Common Lisp package always, always, always available?
> 
> Not in temacs.  Otherwise, yes.  See lisp/dumped-lisp.el, line 38.

So no "#'function" construct may be used in temacs. Is that OK?

> 
>  > 3. Since the function 'symbol-file' takes one mandatory argument and
>  >    one optional one, so calling 'symbol-file' with two arguments is
>  >    *always* allowed,
> 
> Not in 21.4.  This is a patch against a package, so it must work with
> 21.4.

OK.

on 12/27/2008 03:22 AM Aidan Kehoe said the following:
> Hi Vlad --

(I prefer "Vladimir".)

>  > 2. The construct "#'" is not Emacs Lisp, but a Common Lisp extension.
>
> GNU's ChangeLog says they incorporated support for it in C in 1995. The
> Lucid Emacs 19.0.23 beta, with a NEWS file dating from 19920528,
> incorporates it in C. It doesn't make sense to call it "not Emacs
Lisp" in
> 2008.

Then the docs are *way* out of date.

>
>  >    Is the Common Lisp package always, always, always available?
>  >    (Appendix D "Porting Common Lisp" in the Common Lisp Extension
info
>  >    file lists an alternative that's pure Emacs Lisp: use a regular
>  >    quote, e.g. "'symbol-file".)
>
> I'm aware of that; that misses the point of the #' construct,
though, since
> I used it to indicate that I'm interested in the symbol's associated
> function, not its associated variable.

OK. I only suggested the quote equivalence because I was unsure if the
share-quote construct was appropriate.

>
>  > 3. Since the function 'symbol-file' takes one mandatory argument and
>  >    one optional one, so calling 'symbol-file' with two arguments is
>  >    *always* allowed, so I don't understand the intent of the comment.
>
> Stephen covered that question well--thanks, Stephen!

Agreed, although I'm still bothered by the 21.4 vs 21.5 issue not
being mentioned.

--- Vladimir

-- 
Vladimir G. Ivanovic


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 259 bytes
Desc: OpenPGP digital signature
Url : http://calypso.tux.org/pipermail/xemacs-patches/attachments/20081227/a51bcaea/attachment-0001.bin