>>>> "Stephen" == Stephen J Turnbull
<stephen(a)xemacs.org> writes:
>>>> "Paul" == Paul Thompson
<paul(a)wubios.wustl.edu> writes:
Paul> I have reported this bug 5-6
times.
Stephen> AFAIK this doesn't happen to anyone else. It's extremely
Stephen> hard to fix a bug that cannot be replicated. There's
Stephen> only one person I know of who could fix this easily: you.
Paul> XEmacs 21.4.6 \"Common Lisp (Windows [1])\" configured for
Paul> `i586-pc-win32'.
Paul> Load-Path Lisp Shadows:
Paul> ----------------------
Paul> (e:\programs\xemacs\XEmacs-21.4.3\lisp\x-win-xfree86
Paul> e:\programs\xemacs\XEmacs-21.4.6\lisp\x-win-xfree86
Stephen> yada yada yada ...
Stephen> It looks like your XEmacs 21.4.6 is using stuff
Stephen> byte-compiled for 21.4.3. This should work, but it's not
Stephen> a good thing to be happening. I don't understand why
Stephen> this is happening. The two versions should be able to
Stephen> coexist without tromping on each others' toes. Do you
Stephen> have any XEmacs-related environment variables set, in
Stephen> particular EMACSLOADPATH or EMACSPACKAGEPATH? Or maybe
Stephen> it's something in the Windows registry (which I'm
Stephen> learning to loathe).
Paul> e:\programs\xemacs\XEmacs-21.4.3\lisp\lisp
Paul> e:\programs\xemacs\site-version\lisp\lisp
Stephen> And what's this "site-version" stuff? I've never heard
Stephen> of it before, but you seem to have quite a lot of it.
Stephen> Again, a default install should _not_ find that
Stephen> site-version stuff AFAIK (but I don't work with Windows
Stephen> much).
I'm working almost exclusively with native Windows XEmacs but only
know about the standard mule-packages, site-packages, and
xemacs-packages hierarchies.
This site-version stuff must be someones invention outside of the
XEmacs development team.
Paul> e:\programs\xemacs\site-version\lisp\eval-reg
Paul> e:\programs\xemacs\xemacs-packages\lisp\edebug\eval-reg
Stephen> And it's shadowing the optional packages, too.
Paul> Invalid byte code: "variable reference to constant symbol :prefix"
Some Gnu Emacs elcs were responsible for this "Invalid byte code", right?
Adrian
Paul> Loading advice...
Stephen> And it looks like something in your init file is doing
Stephen> something evil. Loading advice the very first thing is
Stephen> not reassuring. I sorta suspect that has something to do
Stephen> with it. (Unless your system is just hosed; you're the
Stephen> guy who reported the gdi32.dll problem, too, aren't you?)
Stephen> Can you replicate with "xemacs -vanilla"? With a clean
Stephen> environment?
Stephen> --
Stephen> Institute of Policy and Planning Sciences
http://turnbull.sk.tsukuba.ac.jp
Stephen> University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573
JAPAN
Stephen> Don't ask how you can "do" free software
business;
Stephen> ask what your business can "do for" free
software.
--
Adrian Aichner
mailto:adrianï¼ xemacs.org
http://www.xemacs.org/