2:12 a.m.
I'm bringing in a new version of semantic (to go with JDE 2.2.5) and
this snippet of code is giving me hell...
,----
| (defun semantic-clear-toplevel-cache ()
| "Clear the toplevel bovin cache for the current buffer."
| (interactive)
| (setq semantic-toplevel-bovine-cache nil)
| ;; Nuke all semantic overlays. This is faster than deleting based
| ;; on our data structure.
| (let ((l (overlay-lists)))
| (mapcar 'semantic-delete-overlay-maybe (car l))
| (mapcar 'semantic-delete-overlay-maybe (cdr l))
| )
| ;; Remove this hook which tracks if a buffer is up to date or not.
| (remove-hook 'after-change-functions 'semantic-change-function t)
| )
| (add-hook 'change-major-mode-hook 'semantic-clear-toplevel-cache)
`----
eval-defun gives no error, but it gives...
'Symbols function definition is void: overlay-lists'
whenever 'change-major-mode-hook' is called.
If what I've given here isn't enough to debug this problem I'll put
the semantic-1.3 tarball on
ftp.xemacs.org/pub/xemacs/beta/experimental/.
--
|---<Regards, Steve Youngs>-----------[GnuPG KeyID: EFD82ED2]---|
| XEmacs - It's not just an editor... |
| It's a way of life. |
|----------------------------------<mailto:youngs＠xemacs.org>---|
3:53 a.m.
>>>> "Steve" == Steve Youngs
<youngs(a)xemacs.org> writes:
Steve> eval-defun gives no error, but it gives...
Steve> 'Symbols function definition is void:
Steve> overlay-lists'
Overlays are defined in fsf-compat/overlay.el. You probably need a
(require 'overlay) somewhere; overlays are the FSF native way of
expressing what we call extents. They are severely deprecated, to the
extent that overlay.el provides _no_ autoloads (presumably as a matter
of policy). Cf the comments in that library.
HTH
--
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091
_________________ _________________ _________________ _________________
What are those straight lines for? "XEmacs rules."
6:58 a.m.
* Stephen J Turnbull <turnbull(a)sk.tsukuba.ac.jp> writes:
>>>>> "Steve" == Steve Youngs
<youngs(a)xemacs.org> writes:
Steve> eval-defun gives no error, but it
gives...
Steve> 'Symbols function definition is void:
Steve> overlay-lists'
Overlays are defined in fsf-compat/overlay.el. You probably need a
(require 'overlay) somewhere; overlays are the FSF native way of
expressing what we call extents. They are severely deprecated, to the
extent that overlay.el provides _no_ autoloads (presumably as a matter
of policy). Cf the comments in that library.
Thanks, that fixed it.
--
|---<Regards, Steve Youngs>-----------[GnuPG KeyID: EFD82ED2]---|
| XEmacs - It's not just an editor... |
| It's a way of life. |
|----------------------------------<mailto:youngs＠xemacs.org>---|
10:22 a.m.
Steve Youngs <youngs(a)xemacs.org> writes:
I'm bringing in a new version of semantic (to go with JDE 2.2.5)
and
this snippet of code is giving me hell...
Two points:
* JDE 2.2.5 doesn't run properly with semantic 1.3, the next version
of JDE 2.2.6 will (according to the JDE mailing list).
* semantic 1.3 actually contains some code to use overlays OR
extends. It checks with this code:
,----
| (if (fboundp 'make-overlay)
| (progn
| (defalias 'semantic-make-overlay 'make-overlay)
| [more aliases]
| )
| (defalias 'semantic-make-overlay 'make-extent)
| [more aliases]
| )
`----
There is the same problem with the newest version of speedbar, too.
And for some reason overlay.el get's loaded in my configuration, so
speedbar (or semantic for that matter) initialize the overlay aliases
instead of the extend aliases. And those don't work well enough in
XEmacs. I suggest to check simply for (featurep 'xemacs) wich is
kinda dirty but should work well.
I CC'ed Eric Ludlam, who is the author of semantic and speedbar to
avoid differences between the author version and XEmacs version,
especially if the author tries to be compatible with XEmacs.
Gerd
12:59 p.m.
New subject: Re[1]: semantic lisp code help please
>> Gerd Boerrigter <gerd.boerrigter(a)i2c-systems.com> seems
to think that:
Steve Youngs <youngs(a)xemacs.org> writes:
> I'm bringing in a new version of semantic (to go with JDE 2.2.5) and
> this snippet of code is giving me hell...
Two points:
* JDE 2.2.5 doesn't run properly with semantic 1.3, the next version
of JDE 2.2.6 will (according to the JDE mailing list).
* semantic 1.3 actually contains some code to use overlays OR
extends. It checks with this code:
[ ... ]
I CC'ed Eric Ludlam, who is the author of semantic and speedbar
to
avoid differences between the author version and XEmacs version,
especially if the author tries to be compatible with XEmacs.
[ ... ]
Hi,
I'll be making the changes recommended by Gerd for future versions
of both tools so that XEmacs teams won't need to make any changes.
(At least, that is my hope.) I don't understand why the overlay
compatibility tool can't handle semantic's demands. I didn't do
anything beyond aliasing the extent code, and things work fine (as far
as I can tell.)
On a side note, I had signed up for XEmacs CVS access a while ago,
but I don't think my account was ever set up. (I had sent in a ssh
key)
Eric
--
|\/\/\___/\/\/| Eric Ludlam zappo(a)gnu.org; zappo(a)ultranet.com
\____ o o ____/ Homepage: http://www.ultranet.com/~zappo
) ( Trebuchets: http://www.ultranet.com/~zappo/trebuchet.shtml
( * * ) GNU: http://www.gnu.org
\___/ Trevor Shea Ludlam: Oct 16th, 1999, 5lbs 11oz 18in
5:31 a.m.
New subject: Re[1]: semantic lisp code help please
"Eric M. Ludlam" <zappo(a)ultranet.com> writes:
I'll be making the changes recommended by Gerd for future
versions
of both tools so that XEmacs teams won't need to make any changes.
(At least, that is my hope.) I don't understand why the overlay
compatibility tool can't handle semantic's demands. I didn't do
anything beyond aliasing the extent code, and things work fine (as far
as I can tell.)
Well, using fsf-compat library (overlay.el is part of it) is
deprecated in XEmacs lisp.
Are you interested in adding your checkdoc.el to XEmacs
package? I'm asking because I made some change for it to
work with XEmacs in my workspace. (Mininum change is to use
characterp instead of integerp in line 647.)
--
Yoshiki Hayashi
12:57 p.m.
New subject: Re[1]: semantic lisp code help please
>> Yoshiki Hayashi <yoshiki(a)xemacs.org> seems to think
that:
"Eric M. Ludlam" <zappo(a)ultranet.com> writes:
> I'll be making the changes recommended by Gerd for future versions
> of both tools so that XEmacs teams won't need to make any changes.
> (At least, that is my hope.) I don't understand why the overlay
> compatibility tool can't handle semantic's demands. I didn't do
> anything beyond aliasing the extent code, and things work fine (as far
> as I can tell.)
Well, using fsf-compat library (overlay.el is part of it) is
deprecated in XEmacs lisp.
Hi,
I did not try to use overlay.el but checked for the existence of the
function `make-overlay' which the compat library apparently provides.
I don't explicitly load any XEmacs compatibility items in my .emacs
when I test against it.
Are you interested in adding your checkdoc.el to XEmacs
package? I'm asking because I made some change for it to
work with XEmacs in my workspace. (Mininum change is to use
characterp instead of integerp in line 647.)
That is fine with me. Feel free to send me the patch. I have several
other small changes that I haven't published yet.
If anyone is ever interested in the latest goings on of any of my
software, feel free to go here to check the latest RCS logs:
http://choochoo.ultranet.com/~zappo/fsf.shtml
Enjoy
Eric
--
|\/\/\___/\/\/| Eric Ludlam zappo(a)gnu.org; zappo(a)ultranet.com
\____ o o ____/ Homepage: http://www.ultranet.com/~zappo
) ( Trebuchets: http://www.ultranet.com/~zappo/trebuchet.shtml
( * * ) GNU: http://www.gnu.org
\___/ Trevor Shea Ludlam: Oct 16th, 1999, 5lbs 11oz 18in
11:59 a.m.
New subject: Re[1]: semantic lisp code help please
"Eric M. Ludlam" <zappo(a)ultranet.com> writes:
>Are you interested in adding your checkdoc.el to XEmacs
>package? I'm asking because I made some change for it to
>work with XEmacs in my workspace. (Mininum change is to use
>characterp instead of integerp in line 647.)
That is fine with me. Feel free to send me the patch. I have several
other small changes that I haven't published yet.
Here's the patch. Changes are:
1. New variable checkdoc-xemacs-p.
2. Avoid integerp check. Return value of read-char is
always a character on XEmacs.
If this patch is OK to you, I'd like to add this to
CVS XEmacs package tree. Steve, where would you suggest to
add? I think xemacs-devel package is a good candidate.
Once it's added to the CVS tree, I'll ask Martin to change
CVS write access and after that it will be maintained by
you.
If anyone is ever interested in the latest goings on of any of my
software, feel free to go here to check the latest RCS logs:
http://choochoo.ultranet.com/~zappo/fsf.shtml
Wow. I'd like to see every software incorporated into XEmacs
package.
--- checkdoc.el.orig Wed Oct 18 19:46:00 2000
+++ checkdoc.el Wed Oct 18 01:05:41 2000
＠＠ -189,6 +189,8 ＠＠
(defmacro defcustom (var value doc &rest args)
(` (defvar (, var) (, value) (, doc))))))
+(defvar checkdoc-xemacs-p (string-match "XEmacs" emacs-version))
+
(defcustom checkdoc-autofix-flag 'semiautomatic
"*Non-nil means attempt auto-fixing of doc strings.
If this value is the symbol `query', then the user is queried before
＠＠ -445,7 +447,7 ＠＠
;;; Compatibility
;;
-(if (string-match "X[Ee]macs" emacs-version)
+(if checkdoc-xemacs-p
(progn
(defalias 'checkdoc-make-overlay 'make-extent)
(defalias 'checkdoc-overlay-put 'set-extent-property)
＠＠ -650,8 +652,10 ＠＠
(goto-char (checkdoc-error-start (car (car err-list))))
(if (not (pos-visible-in-window-p))
(recenter (- (window-height) 2)))
- (setq c (checkdoc-read-event)))1
- (if (not (integerp c)) (setq c ??))
+ (setq c (checkdoc-read-event)))
+ (if (and (not checkdoc-xemacs-p)
+ (not (integerp c)))
+ (setq c ??))
(cond
;; Exit condition
((checkdoc-char= c ?\C-g) (signal 'quit nil))
＠＠ -1193,7 +1197,7 ＠＠
;; Override some bindings
(define-key map "\C-\M-x" 'checkdoc-eval-defun)
;(define-key map "\C-x`" 'checkdoc-continue)
- (if (not (string-match "XEmacs" emacs-version))
+ (if (not checkdoc-xemacs-p)
(define-key map [menu-bar emacs-lisp eval-buffer]
'checkdoc-eval-current-buffer))
;; Add some new bindings under C-c ?
＠＠ -2660,7 +2664,7 ＠＠
(if checkdoc-output-mode-map
nil
(setq checkdoc-output-mode-map (make-sparse-keymap))
- (if (not (string-match "XEmacs" emacs-version))
+ (if (not checkdoc-xemacs-p)
(define-key checkdoc-output-mode-map [mouse-2]
'checkdoc-find-error-mouse))
(define-key checkdoc-output-mode-map "\C-c\C-c" 'checkdoc-find-error)
--
Yoshiki Hayashi
4:42 p.m.
New subject: Re[1]: semantic lisp code help please
* Yoshiki Hayashi <yoshiki(a)xemacs.org> writes:
"Eric M. Ludlam" <zappo(a)ultranet.com> writes:
> >Are you interested in adding your checkdoc.el to XEmacs
> >package? I'm asking because I made some change for it to
> >work with XEmacs in my workspace. (Mininum change is to use
> >characterp instead of integerp in line 647.)
>
> That is fine with me. Feel free to send me the patch. I have several
> other small changes that I haven't published yet.
Here's the patch. Changes are:
1. New variable checkdoc-xemacs-p.
2. Avoid integerp check. Return value of read-char is
always a character on XEmacs.
If this patch is OK to you, I'd like to add this to
CVS XEmacs package tree. Steve, where would you suggest to
add? I think xemacs-devel package is a good candidate.
Yes, I agree, xemacs-devel would be the most appropriate place for
checkdoc.
--
|---<Regards, Steve Youngs>-----------[GnuPG KeyID: 59E4C4B2]---|
| XEmacs - It's not just an editor... |
| It's a way of life. |
|----------------------------------<mailto:youngs＠xemacs.org>---|
8802
days inactive
8809
days old
8 comments
5 participants
participants (5)
-
Eric M. Ludlam -
Gerd Boerrigter -
Stephen J. Turnbull -
Steve Youngs -
Yoshiki Hayashi