On Tue, 04 Nov 2003 16:34:27 +0100
Hrvoje Niksic <hniksic(a)xemacs.org> wrote:
junq <junq(a)ihubbell.com> writes:
>> What makes you think that XAE is a problem? Does the leak fail to
>> occur if and only if you comment out the loading of XAE?
>
> xemacs -q (which bypasses loading of init.el and init.el is where
> XAE is loaded) showed no memory growth,
Yes, but does your `init.el' do other things beside loading XAE? If
that is not the case, I'm curious why you think that XAE (out of all
things your init.el might load) is the one responsible.
It doesn't do much, but what do I know about it...?
init.el:
;; always end a file with a newline
(setq require-final-newline t)
;; don't let `next-line' add new lines in buffer
(setq next-line-add-newlines nil)
;; enable wheelmouse support by default
(when window-system
(mwheel-install))
(add-to-list 'load-path (expand-file-name
"/usr/share/xemacs/site-lisp/xae-1.0beta8/psgml"))
(add-to-list 'load-path (expand-file-name
"/usr/share/xemacs/site-lisp/xae-1.0beta8/lisp"))
(require 'xae)
The next logical step would be: `xemacs -q' + load XAE manually + see
if the leak occurs, with and without Perl.
> So where does this leave us?
If you're sure XAE makes a difference, I'll take a look at what it
does. But first I'd like to be sure that XAE is to blame.