[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: false warnings when compiling with lexical-binding and cl-lib.
From: |
Thierry Volpiatto |
Subject: |
Re: false warnings when compiling with lexical-binding and cl-lib. |
Date: |
Sat, 23 Nov 2013 17:24:10 +0100 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/24.3.50 (gnu/linux) |
Thierry Volpiatto <address@hidden> writes:
> Stefan Monnier <address@hidden> writes:
>
>>> This patch avoid adding a new binding by the 'into' clause when the
>>> binding have been already added by a 'with' clause.
>>> I wonder if this is correct, real common-lisp return an error in such
>>> case, WDYT?
>>
>> Oh, wait, then the warning is OK. It indicates that the `with lst' is
>> extraneous and unused. IOW, there's no bug to fix. We could change
>> cl-loop to signal an error, but it doesn't seem to be worth the trouble.
>
> Yes, maybe the error message can be a little more explicit, common-lisp send a
> message like this:
>
> ,----
> | Variable LST in INTO clause is a duplicate
> | current LOOP context: COLLECT I INTO LST FINALLY.
> | [Condition of type SB-INT:COMPILED-PROGRAM-ERROR]
> `----
>
> Also the warning is only at compile time, should we return an error on
> evaluation ?
> This can be detected and handled from the same function
> i.e (cl--loop-handle-accum)
--8<---------------cut here---------------start------------->8---
diff --git a/lisp/emacs-lisp/cl-macs.el b/lisp/emacs-lisp/cl-macs.el
index 2209297..3faf23b 100644
--- a/lisp/emacs-lisp/cl-macs.el
+++ b/lisp/emacs-lisp/cl-macs.el
@@ -1564,9 +1564,13 @@ If BODY is `setq', then use SPECS for assignments rather
than for bindings."
(defun cl--loop-handle-accum (def &optional func) ; uses loop-*
(if (eq (car cl--loop-args) 'into)
- (let ((var (cl--pop2 cl--loop-args)))
+ (let* ((var (cl--pop2 cl--loop-args))
+ (binding (list (list var def))))
(or (memq var cl--loop-accum-vars)
- (progn (push (list (list var def)) cl--loop-bindings)
+ (cl-assert (not (member binding cl--loop-bindings))
+ nil "Variable %s in INTO clause is a duplicate"
+ (upcase (symbol-name (caar binding))))
+ (progn (push binding cl--loop-bindings)
(push var cl--loop-accum-vars)))
var)
(or cl--loop-accum-var
--8<---------------cut here---------------end--------------->8---
--
Thierry
Get my Gnupg key:
gpg --keyserver pgp.mit.edu --recv-keys 59F29997