unofficial mirror of emacs-devel@gnu.org 
 help / color / mirror / code / Atom feed
* `assert' macro
@ 2006-01-27 21:35 Johan Bockgård
  2006-01-29 19:17 ` John Paul Wallington
  0 siblings, 1 reply; 3+ messages in thread
From: Johan Bockgård @ 2006-01-27 21:35 UTC (permalink / raw)



The `assert' macro from cl-macs is used in several files in Emacs. It
expands into the following code.

    (macroexpand '(assert x))
     =>
    (progn (or x (signal 'cl-assertion-failed (list 'x))) nil)

Unless cl is loaded `cl-assertion-failed' is not a known error symbol
though, so if the assertion fails it will produce a "peculiar error"
that is impossible to catch. It should signal a known condition.

-- 
Johan Bockgård

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: `assert' macro
  2006-01-27 21:35 `assert' macro Johan Bockgård
@ 2006-01-29 19:17 ` John Paul Wallington
  2006-01-30 18:45   ` Richard M. Stallman
  0 siblings, 1 reply; 3+ messages in thread
From: John Paul Wallington @ 2006-01-29 19:17 UTC (permalink / raw)


bojohan+news@dd.chalmers.se (Johan Bockgård) writes:

> The `assert' macro from cl-macs is used in several files in Emacs. It
> expands into the following code.
>
>     (macroexpand '(assert x))
>      =>
>     (progn (or x (signal 'cl-assertion-failed (list 'x))) nil)
>
> Unless cl is loaded `cl-assertion-failed' is not a known error symbol
> though, so if the assertion fails it will produce a "peculiar error"
> that is impossible to catch. It should signal a known condition.

A minimal impact fix is to move the definition of that condition from
cl.el to subr.el.  Any objections ?


Index: lisp/subr.el
===================================================================
RCS file: /sources/emacs/emacs/lisp/subr.el,v
retrieving revision 1.494
diff -u -r1.494 subr.el
--- lisp/subr.el	26 Jan 2006 17:54:27 -0000	1.494
+++ lisp/subr.el	29 Jan 2006 19:09:26 -0000
@@ -1831,6 +1831,12 @@
 This variable is meaningful on MS-DOG and Windows NT.
 On those systems, it is automatically local in every buffer.
 On other systems, this variable is normally always nil.")
+
+;; The `assert' macro from the cl package signals
+;; `cl-assertion-failed' at runtime so always define it.
+(put 'cl-assertion-failed 'error-conditions '(error))
+(put 'cl-assertion-failed 'error-message "Assertion failed")
+
 \f
 ;;;; Misc. useful functions.

Index: lisp/emacs-lisp/cl.el
===================================================================
RCS file: /sources/emacs/emacs/lisp/emacs-lisp/cl.el,v
retrieving revision 1.46
diff -u -r1.46 cl.el
--- lisp/emacs-lisp/cl.el	6 Aug 2005 17:08:59 -0000	1.46
+++ lisp/emacs-lisp/cl.el	29 Jan 2006 19:09:27 -0000
@@ -580,9 +580,6 @@
 
 ;;; Miscellaneous.
 
-(put 'cl-assertion-failed 'error-conditions '(error))
-(put 'cl-assertion-failed 'error-message "Assertion failed")
-
 (defvar cl-fake-autoloads nil
   "Non-nil means don't make CL functions autoload.")
 

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: `assert' macro
  2006-01-29 19:17 ` John Paul Wallington
@ 2006-01-30 18:45   ` Richard M. Stallman
  0 siblings, 0 replies; 3+ messages in thread
From: Richard M. Stallman @ 2006-01-30 18:45 UTC (permalink / raw)
  Cc: emacs-devel

    A minimal impact fix is to move the definition of that condition from
    cl.el to subr.el.  Any objections ?

It seems good to me.

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2006-01-30 18:45 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2006-01-27 21:35 `assert' macro Johan Bockgård
2006-01-29 19:17 ` John Paul Wallington
2006-01-30 18:45   ` Richard M. Stallman

Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/emacs.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).