GNU bug report logs - #5805
23.1; abbrev-insert does not protext itself with save-excursion

Previous Next

Package: emacs;

Reported by: "Maguire, Andrew (GE Infra, Energy)" <andrew.maguire <at> ge.com>

Date: Tue, 30 Mar 2010 16:37:02 UTC

Severity: normal

Tags: patch

Done: Stefan Monnier <monnier <at> iro.umontreal.ca>

Bug is archived. No further changes may be made.

Full log

View this message in rfc822 format

From: Stefan Monnier <monnier <at> iro.umontreal.ca>
To: Bob Nnamtrop <bobnnamtrop <at> gmail.com>
Cc: 5805 <at> debbugs.gnu.org
Subject: bug#5805: 23.3 abbrev-insert needs a limited save-excursion
Date: Thu, 07 Jul 2011 21:03:03 -0400

> Here are two patches that do what I explained above. I am new to elisp
> so be kind :-) Note that these apply against the trunk (sorry to be
> switching back of forth).

I still haven't thought about the save-excursion issue, but for the
other, I think the fix should be complete in viper without changing
expand-abbrev.
The way it work in "normal Emacs" is the self-insert-command only calls
expand-abbrev if (as its docstring explains) the char inserted doesn't
have word syntax and the previous one does.  The corresponding code
(it's in C):

  if (!NILP (BVAR (current_buffer, abbrev_mode))
      && synt != Sword
      && NILP (BVAR (current_buffer, read_only))
      && PT > BEGV
      && (SYNTAX (!NILP (BVAR (current_buffer, enable_multibyte_characters))
		  ? XFASTINT (Fprevious_char ())
		  : UNIBYTE_TO_CHAR (XFASTINT (Fprevious_char ())))
	  == Sword))
    {


-- Stefan
This bug report was last modified 13 years and 323 days ago.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.