bug#9482: Documentation bug: Emacs Lisp Manual should say quoting does not cons explicitly in quoting section. related to bug#9469

bug#9482: Documentation bug: Emacs Lisp Manual should say quoting does not cons explicitly in quoting section. related to bug#9469

[Le Wang]

Relevant page here:
http://www.gnu.org/s/emacs/manual/html_node/elisp/Quoting.html

I was confused and always assumed that

'(a . b)

is simply short hand for

(cons 'a 'b)

Reading stackexchange, it seems at least some other people are under
this impression as well.  This works a majority of the time, but when
it fails it's really surprising.

In fact, the manual already has an example of this pit-fall in the
nconc section: http://www.gnu.org/software/emacs/elisp/html_node/Rearrangement.html

I think the quoting section should give an explicit warning with a
reference to the nconc example.



-- 
Le

bug#9482: specific manual text suggestion

[Le Wang]

Hi Stefan,

As I mentioned in this bug.  The manual already has an example of this
trap in the nconc section.  Maybe just a pointer to it.

So my suggestion for specific text would be:

"WARNING: '(a . b) is not short-hand for (cons 'a 'b).  The former
does not cons.  Quoting should be reserved for constants that will
*NEVER* change.  See <link>nconc</link> for example of unexpected
results when a quoted object is modified."

My terminology may be not quite right, but if I read something like
this, the difference would be much clearer.

-- 
Le

bug#9482: specific manual text suggestion

[Stefan Monnier]

> My terminology may be not quite right, but if I read something like
> this, the difference would be much clearer.

Thanks for your sample text.  I tweaked it a bit to make it hopefully
even more explicit, see patch below.


        Stefan


--- src/eval.c	2011-09-09 01:06:52 +0000
+++ src/eval.c	2011-09-16 14:09:39 +0000
@@ -475,6 +475,14 @@
 
 DEFUN ("quote", Fquote, Squote, 1, UNEVALLED, 0,
        doc: /* Return the argument, without evaluating it.  `(quote x)' yields `x'.
+Warning: `quote' does not construct its return value, but just returns
+the value that was pre-constructed by the Lisp reader (see info node
+`(elisp)Printed Representation').
+This means that '(a . b) is not identical to (cons 'a 'b): the former
+does not cons.  Quoting should be reserved for constants that will
+never be modified by side-effects, unless you like self-modifying
+code.  See the common pitfall in info node `(elisp)Rearrangement' for
+an example of unexpected results when a quoted object is modified.
 usage: (quote ARG)  */)
   (Lisp_Object args)
 {