3:32 a.m.
Hello,
I'm trying to use refill-mode in xemacs21 (xemacs-21.4.14-mule, from
the debian package). I've tried a bunch of different versions from
the mode's CVS (you can get a representative version at
http://savannah.gnu.org/cgi-bin/viewcvs/*checkout*/emacs/emacs/lisp/textm...).
All of them work fine with FSF Emacs (which I don't use, ubt do
have lying around); none of them will load properly with Xemacs. The
*backtrace* buffer resulting from running M-x eval-current-buffer from
inside the refill.el buffer is:
---------------------------------------------------------------------------
Signaling: (wrong-number-of-arguments #<compiled-function (mode doc &optional
init-value lighter keymap) "...(85)" [keymap-doc mode-doc keymap-name mode-name
keymap mode symbol-name format "Non-nil if %s mode is enabled." "-map"
"Keymap for %s mode." progn defvar make-variable-buffer-local quote intern cond
and keymapp listp easy-mmode-define-keymap t error "Invalid keymap %S"
macroexpand easy-mmode-define-toggle add-minor-mode init-value doc lighter] 12
("/usr/share/xemacs21/xemacs-packages/lisp/xemacs-base/easy-mmode.elc" .
2037)> 8)
#<compiled-function (mode doc &optional init-value lighter keymap)
"...(85)" [keymap-doc mode-doc keymap-name mode-name keymap mode symbol-name
format "Non-nil if %s mode is enabled." "-map" "Keymap for %s
mode." progn defvar make-variable-buffer-local quote intern cond and keymapp listp
easy-mmode-define-keymap t error "Invalid keymap %S" macroexpand
easy-mmode-define-toggle add-minor-mode init-value doc lighter] 12
("/usr/share/xemacs21/xemacs-packages/lisp/xemacs-base/easy-mmode.elc" .
2037)>(refill-mode "Toggle Refill minor mode.\nWith prefix arg, turn Refill mode
on iff arg is positive.\n\nWhen Refill mode is on, the current paragraph will be formatted
when\nchanges are made within it. Self-inserting characters only cause\nrefilling if they
would cause auto-filling." nil " Refill" (quote (("" .
backward-delete-char-untabify))) (when refill-ignorable-overlay (delete-overlay
refill-ignorable-overlay) (kill-local-variable (quote refill-ignorable-overlay))) (when
(local-variable-p (quote refill-late-fill-paragraph-function)) (setq
fill-paragraph-function refill-late-fill-paragraph-function) (kill-local-variable (quote
refill-late-fill-paragraph-function))) (if refill-mode (progn (add-hook ... ... nil t)
(add-hook ... ... nil t) (add-hook ... ... nil t) (set ... fill-paragraph-function) (set
... ...) (set ... ...) (setq refill-ignorable-overlay ...) (overlay-put
refill-ignorable-overlay ... ...) (overlay-put refill-ignorable-overlay ... ...)
(auto-fill-mode 0)) (remove-hook (quote after-change-functions) (quote
refill-after-change-function) t) (remove-hook (quote post-command-hook) (quote
refill-post-command-function) t) (kill-local-variable (quote
backward-delete-char-untabify-method))))
(define-minor-mode refill-mode "Toggle Refill minor mode.\nWith prefix arg, turn
Refill mode on iff arg is positive.\n\nWhen Refill mode is on, the current paragraph will
be formatted when\nchanges are made within it. Self-inserting characters only
cause\nrefilling if they would cause auto-filling." nil " Refill" (quote
(...)) (when refill-ignorable-overlay (delete-overlay refill-ignorable-overlay)
(kill-local-variable ...)) (when (local-variable-p ...) (setq fill-paragraph-function
refill-late-fill-paragraph-function) (kill-local-variable ...)) (if refill-mode (progn ...
... ... ... ... ... ... ... ... ...) (remove-hook ... ... t) (remove-hook ... ... t)
(kill-local-variable ...)))
eval-buffer(#<buffer "refill.el"> nil)
#<compiled-function (&optional printflag) "...(5)" [printflag
eval-buffer] 3 826100 nil>()
call-interactively(eval-current-buffer)
command-execute(eval-current-buffer t)
execute-extended-command(nil)
call-interactively(execute-extended-command)
-----------------------------------------------------
Alas, I'm too incompetent to understand this. Can you guys make sense
of it? Does this reflect a bug in Xemacs, or a design decision of some
kind that refill-mode doesn't take into account? I see from the
archives that there have been issues around refill.el in the past, but
the net suggests that the problems ought to have been fixed. Am I
missing something?
thanks,
matt
3:46 a.m.
Matt Price <matt.price(a)utoronto.ca> wrote:
I'm trying to use refill-mode in xemacs21 (xemacs-21.4.14-mule,
from
the debian package). I've tried a bunch of different versions from
the mode's CVS (you can get a representative version at
http://savannah.gnu.org/cgi-bin/viewcvs/*checkout*/emacs/emacs/lisp/textm...).
All of them work fine with FSF Emacs (which I don't use, ubt do
have lying around); none of them will load properly with Xemacs. The
*backtrace* buffer resulting from running M-x eval-current-buffer from
inside the refill.el buffer is:
---------------------------------------------------------------------------
Signaling: (wrong-number-of-arguments
[snip the rest]
The immediate problem is that our version of define-minor-mode is
missing one argument that is supported by Emacs. I recently submitted a
patch to fix that. It is in CVS now. It will be in the next release of
the xemacs-base package, which I hear is coming out Real Soon Now.
Watch for the release announcement, upgrade your xemacs-base package,
and then try again and let us know how it goes. There may well be other
problems lurking.
Thanks for the report.
--
Jerry James
Email: james(a)eecs.ku.edu -or- jamesj(a)acm.org
WWW: http://people.eecs.ku.edu/~james/
3:34 p.m.
On Wed, Jan 28, 2004 at 09:46:11PM -0600, Jerry James wrote:
Matt Price <matt.price(a)utoronto.ca> wrote:
> I'm trying to use refill-mode in xemacs21 (xemacs-21.4.14-mule, from
> the debian package). I've tried a bunch of different versions from
> All of them work fine with FSF Emacs (which I don't use, ubt
do
> have lying around); none of them will load properly with Xemacs. The
[snip the rest]
The immediate problem is that our version of define-minor-mode is
missing one argument that is supported by Emacs. I recently submitted a
patch to fix that. It is in CVS now. It will be in the next release of
the xemacs-base package, which I hear is coming out Real Soon Now.
Watch for the release announcement, upgrade your xemacs-base package,
and then try again and let us know how it goes. There may well be other
problems lurking.
Thanks to both of y'all for the lightning-fast response! I will keep
an eye on the xemacs release announcement. Should I be looking for
21.4.15?
As a lowly user, I have been pretty reliant on the debian xemacs
packages -- so I'm not entirely certain what I need to do if I want to
inxtall xemacs-base on my own. In my distribution I find this
directory:
/usr/share/xemacs21/xemacs-packages/lisp/xemacs-base
would installing a new version of xemacs-base be as simple as
replacing that directory with a new one? Or is there something more
complicated involved?
Thanks much,
matt
4:18 p.m.
>>>> "Matt" == Matt Price
<matt.price(a)utoronto.ca> writes:
Matt> would installing a new version of xemacs-base be as simple
Matt> as replacing that directory with a new one? Or is there
Matt> something more complicated involved?
It should be as simple as
sudo xemacs -vanilla
and selecting a mirror from Tools->Packages->Set Download Site,
then update the list from Tools->Packages->Update Package Index,
and finally run M-x list-packages and select the packages you want
updated.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Ask not how you can "do" free software business;
ask what your business can "do for" free software.
1 a.m.
On Fri, Jan 30, 2004 at 01:18:33AM +0900, Stephen J. Turnbull wrote:
>>>>> "Matt" == Matt Price
<matt.price(a)utoronto.ca> writes:
Matt> would installing a new version of xemacs-base be as simple
Matt> as replacing that directory with a new one? Or is there
Matt> something more complicated involved?
It should be as simple as
sudo xemacs -vanilla
and selecting a mirror from Tools->Packages->Set Download Site,
then update the list from Tools->Packages->Update Package Index,
and finally run M-x list-packages and select the packages you want
updated.
cool, I'll check that out! for now I've grabbed easy-mmode.el from
cvs, put a copy in my .xemacs file, and added a "load" statement in
init.el -- seems to work great (I know it's an ugly ugly hack, but
this way I don't have to interfere with my distribution, on which I am
codependent)!
thanks again,
matt
6:40 p.m.
New subject: gaah! refill strikes again!
Hi,
I'm running into more problems with refill-mode than I thought. I
don't know whether I just missed this the first time; but in any case
when refill_mode loadS (NOT when it's evaluated, though) I get the
following error in the minibuffer:
Wrong number of arguments: local-variable-p, 1
... and then no actual refillingtakes place. I need refill.el to make
longlines.el useful (and I need longlines to be able to use emacs to
write stuff that will later go into a word processor). Anyway, I'm
not sure what the problem is. Just in case I'll repost the url for
refill.el:
http://savannah.gnu.org/cgi-bin/viewcvs/emacs/emacs/lisp/textmodes/refill.el
Anyway, if you can see what's happening with 'local-variable-p' I'd be
realll grateful.
thanks,
matt
6:40 a.m.
New subject: gaah! refill strikes again!
>>>> "Matt" == Matt Price
<matt.price(a)utoronto.ca> writes:
Matt> Wrong number of arguments: local-variable-p, 1
Probably GNU Emacs's local-variable-p allows the BUFFER argument to be
defaulted. I don't currently have GNU Emacs running (space-constraint
after a disk crash) so you'll have to compare their default semantics
for BUFFER.
For every (local-variable-p foo) I would try substituting
(local-variable-p foo nil).
NB: I think allowing a default is a bad idea, since the usual default
for a buffer argument is (current-buffer), but our local-variable-p
defaults to "check whether this variable has a local value anywhere",
not "check whether the value in the current buffer was set locally".
I think it would be a good idea to report it as a bug to the
maintainer of refill.el, because there is a reasonably good chance
that XEmacs will decide not to allow a default for that argument.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Ask not how you can "do" free software business;
ask what your business can "do for" free software.
4:25 a.m.
>>>> "Matt" == Matt Price
<matt.price(a)utoronto.ca> writes:
Matt> M-x eval-current-buffer from inside the refill.el buffer is:
The last sane thing (ie, going up from the bottom) is
define-minor-mode, which is in the process of being synched to GNU.
Maybe all you have to do is wait, or get the more recent version of
define-minor-mode (I think Jerry James <james(a)xemacs.org> is point man
on this one).
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Ask not how you can "do" free software business;
ask what your business can "do for" free software.
7601
days inactive
7603
days old
7 comments
3 participants
participants (3)
-
Jerry James -
Matt Price -
Stephen J. Turnbull