April 2001 - XEmacs-Beta - XEmacs Mailing List Archives

Alex Schroeder

I recently asked on this list wether you'd be interested in adding goto-addr to XEmacs but got no reply. I went ahead and so now I'm putting my mouth where the code is. :) goto-addr.el is currently part of Emacs and uses overlays. I've made the necessary changes in the source. If you are interested, I'll submit the changes back to the Emacs maintainer. I got the latest goto-addr.el from him to do exactly this, so I don't expect any problems on the Emacs side. In order to test, I've used a buffer with a mail address and a URL, like this: alex(a)gnu.org and http://www.gnus.org/ and I evaluated the following code: (load-file "~/elisp/goto-addr/goto-addr.el") (goto-address) (setq mail-user-agent 'sendmail-user-agent) This allowed me to use mouse-2 or C-c RET to open Netscape (this depends on your browse-url setup) or create a mail buffer (and this depends on your compose-mail setup). Beyond requiring 'overlay, there is other compatibility cruft as well. Here's the entire list of XEmacs related code. I've tried to find equivalents for line-beginning-position and line-end-position in the XEmacs Lispref manual but couldn't find anything. A defalias would be nicer, of course. And if you know of a simple workaround to the mouse-2 / button2 problem, please let me know. ;; XEmacs needs the following definitions. (unless (fboundp 'overlays-in) (require 'overlay)) (unless (fboundp 'line-beginning-position) (defun line-beginning-position () "Compatibility function." (save-excursion (beginning-of-line) (point)))) (unless (fboundp 'line-end-position) (defun line-end-position () "Compatibility function." (save-excursion (end-of-line) (point)))) (unless (fboundp 'match-string-no-properties) (defalias 'match-string-no-properties 'match-string)) (defvar goto-address-highlight-keymap (let ((m (make-sparse-keymap))) (if (featurep 'xemacs) (define-key m (kbd "<button2>") 'goto-address-at-mouse) (define-key m (kbd "<mouse-2>") 'goto-address-at-mouse)) (define-key m (kbd "C-c RET") 'goto-address-at-point) m) "keymap to hold goto-addr's mouse key defs under highlighted URLs.") Anyway, I'd appreciate some feedback. I am not on the xemacs-beta mailing list, so please include me in the CC. :) Alex.

2 participants
2 comments

X event loop 23 years, 7 months

Gunnar Evermann

I need some advice from people who understand X. I am trying to debug a problem with the C-g detection in xemacs. The main part of this is in x_check_for_quit_char() which does the following: ------------------------------------------------------------ XtAppPending (Xt_app_con); XEventsQueued (display, QueuedAfterReading); queued = XCheckIfEvent (display, &event, quit_char_predicate, (char *) &critical_quit); ------------------------------------------------------------ To be able to interrupt system calls (like a hung connect()) with C-g the above is called _once_ from a SIGALARM signal handler. The problem is that XCheckIfEvent() doesn't seem to find all events. For example if I run all this from a debugger then what I see is that XCheckIfEvent() checks two events, but if I then immediately call XCheckIfEvent() again there is suddenly another event in the queue. If I call it once more then there are four (in my example finally including the C-g). I assume the idea of the XtAppPending() and XEventsQueued() calls was to avoid exactly this situation by making sure that all events are available in the queue. Obviously this doesn't work. Can anybody shed some light on this? Is there any way to make sure we really check all events? Gunnar P.S.: In case this is relevant: before all this we called fcntl (filedesc, F_SETFL, fcntl (filedesc, F_GETFL, 0) & ~FASYNC);

3 participants
3 comments

Certain keys crash XEmacs with MacOS X 10.0.1 23 years, 7 months

Eric Knauel

This bug report will be sent to the XEmacs Development Team, not to your local site managers!! Please write in English, because the XEmacs maintainers do not have translators to read other languages for them. In XEmacs 21.2 (beta46) "Urania" [Lucid] (powerpc-apple-darwin1.3) of Tue Apr 3 2001 on gryffindor configured using `configure --pdump' Please describe exactly what actions triggered the bug and the precise symptoms of the bug: Pressing the apostrophe key leads to a crash. That seems to independent from the buffer mode, it happens in gnus and in all other buffers. I can reproduce this behaviour at any time I want (which is not very often ;-). Lisp backtrace follows: #<compiled-function (key value) "...(33)" [keymap value keymapp xlib-input-method-bug-workaround sequencep 1 lookup-key define-key] 4>(#<INTERNAL OBJECT (XEmacs bug?) (symbol-value-forward type 10633880) 0xf10dc0> [yen]) map-keymap(#<compiled-function (key value) "...(33)" [keymap value keymapp xlib-input-method-bug-workaround sequencep 1 lookup-key define-key] 4> #<keymap (- #<keymap compose-map size 63 0x25ea>) size 17 0x2618>) # bind (keymap) xlib-input-method-bug-workaround(#<keymap (- #<keymap compose-map size 63 0x25ea>) size 17 0x2618>) # bind (value key) #<compiled-function (key value) "...(33)" [keymap value keymapp xlib-input-method-bug-workaround sequencep 1 lookup-key define-key] 4>(- #<keymap (- #<keymap compose-map size 63 0x25ea>) size 17 0x2618>) map-keymap(#<compiled-function (key value) "...(33)" [keymap value keymapp xlib-input-method-bug-workaround sequencep 1 lookup-key define-key] 4> #<keymap compose-map size 63 0x25ea>) # bind (keymap) xlib-input-method-bug-workaround(#<keymap compose-map size 63 0x25ea>) byte-code("..." [compose-map xlib-input-method-bug-workaround unintern] 2) # bind (current-load-list) # (unwind-protect ...) # bind (load-file-name) # (unwind-protect ...) # (unwind-protect ...) # (unwind-protect ...) # (unwind-protect ...) # (unwind-protect ...) load-internal("x-compose" nil nil nil nil nil) # bind (nosuffix nomessage noerror file) load("x-compose" nil nil nil) # (unwind-protect ...) # (condition-case ... . error) # (catch top-level ...) Abort (core dumped) I loaded the core file with gdb (like XEmacs said) but "where" simply says "No stack.". Recent keystrokes: button1 button1up misc-user Recent messages (most recent first): Loading emacsbug...done Loading emacsbug... Loading gnuserv...done Loading gnuserv... Loading font-lock...done Loading font-lock... Loading advice...done Loading advice... Loading smtpmail...done Loading smtpmail...

4 participants
5 comments

21.4.0 - someone please run autoconf 23 years, 7 months

Paul Stodghill

I just checked out "release-21-4" and it looks like "configure" and "configure.in" are out of sync. Could someone run autoconf and check in an updated "configure"? Thanks.

2 participants
1 comments

FYI: possibly better Postscript mode 23 years, 7 months

Hannu Koivisto

Greetings, I just searched for a Postscript mode for Emacs use and I found two: the one in the XEmacs packages CVS repository (postscript.el) and one at <URL:http://odur.let.rug.nl/~kleiweg/postscript/> (ps-mode.el). I don't really write Postscript and I didn't try the XEmacs one (I just took a look at the code) so don't rely on this or anything but to me the latter looks cooler; it can be customized (I don't use customize so I don't care but newbies might...), it has a fancy menu, possibly fancier fontification (again, I didn't try postscript.el's fontification, I just looked at the patterns), templates and similar helper stuff. It will also be part of Emacs 21. I just thought to let you know just in case someone caring about XEmacs and Postscript would like to examine whether postscript.el in XEmacs should be replaced with ps-mode.el. -- Hannu Please don't send copies of list mail

2 participants
1 comments

new flyspell available 23 years, 7 months

Gunnar Evermann

There is a new version (1.5d) of flyspell available from http://kaolin.unice.fr/~serrano/emacs/emacs.html We currently have 1.4d. Maybe someone who actually used this could do the merge and maybe even convince the author to commit directly to our CVS :-) Gunnar

3 participants
2 comments

Re: Building XEmacs 21.x for Mac OS X? 23 years, 7 months

Doug Alcorn

Martin Buchholz <martin(a)xemacs.org> writes: > You provide me access to your machine, and I'll do the port. I've > been looking for a MacOS X machine for a while. What ever happened with this? I saw a bug-report from someone running on Mac OS X. Was that using X11? Was this offer to "Carbonize" XEmacs. I would gladly give you access to my Mac OS X box if you would do a carbonized version of XEmacs. Is secure shell access enough? -- (__) Doug Alcorn (mailto:doug＠lathi.net http://www.lathi.net) oo / PGP 02B3 1E26 BCF2 9AAF 93F1 61D7 450C B264 3E63 D543 |_/ If you're a capitalist and you have the best goods and they're free, you don't have to proselytize, you just have to wait.

1 participants
0 comments

Re: 21.4.0 build failure on Solaris 23 years, 7 months

Raymond Toy

>>>>> "Raymond" == Raymond Toy <toy(a)rtp.ericsson.se> writes: >>>>> "Stephen" == Stephen J Turnbull <turnbull(a)sk.tsukuba.ac.jp> writes: >>>>> "Ray" == Raymond Toy <toy(a)rtp.ericsson.se> writes: Ray> Loading /apps/public/XEmacs/lib/mule-packages/lisp/egg-its/dumped-lisp.el... Stephen> Er, WTF? Packages should not be dumped IMHO, and _especially_ the Stephen> Japanese-only input methods should be dumped only at sites that use Stephen> them by explicit choice of the admin. I assume you are using Stephen> everything straight off the XEmacs FTP site? Raymond> Everything straight from the XEmacs FTP site. And, since I can't read Raymond> Japanese, I did nothing to explicitly dump Japanese input methods. Stephen> I'm inclined to consider this a package bug. Raymond> Any hints on how to get past this? Even though I'm illiterate, it is Raymond> rather cute to see messages and stuff in other languages. FWIW, I just reinstalled egg-its-1.13, leim-1.11 mule-base-1.30 and skk-1.11 since that's what I had handy. 21.4.0 builds fine now. Haven't tested it yet, but I confident it works well enough. :-) Ray

1 participants
0 comments

Re: [Isaac Hollander <ysh@mindspring.com>] Sun package is deadwood? 23 years, 7 months

Hollander, Isaac (IT)

I prefer ddd myself, but I've got users who insist on workshop. Your function, with minor changes for my environment [1] , works perfectly. I added it to default.el in the site-packages directory. Thanks for the assist. I continue to believe that code in the sumo package tarball such as the Sun directory should work on 21.1.x and 21.4.x unmodified. I.e. M-x workshop-start should Just Work if you build xemacs with workshop support. Right now, lisp hacking such as below is still required... Thanks to Gunnar Everson and Charles Hines for their responses as well... Isaac [1] Here's the code. I commented out (interactive) to avoid error messages in the modline if ttsession is already running. (defun ms-start-workshop() "Initializes Sun Workshop. Use M-x workshop-start to invoke the debugger" ; (interactive) (when (featurep 'tooltalk) (load "tooltalk-macros") (load "tooltalk-util") (load "tooltalk-init")) (when (featurep 'sparcworks) (load "sunpro-init") (load "ring") (load "comint") (load "annotations") (sunpro-startup)) (setq bar-cursor nil)) (when (and (>= emacs-major-version 21) (boundp 'ms-i-really-want-workshop) ms-i-really-want-workshop) (ms-start-workshop)) >>>>> "Jake" == Jake Colman <colman(a)ppllc.com> writes: Jake> I usually simply lurk this group but I'll jump in on this. Jake> I use XEmacs 21.1.x regularly and with only some minor setup on Jake> my part can integrate workshop quite nicely. Although I Jake> personally do not use Workshop, preferring gud and dbx, I've Jake> made it work for several of my colleagues Jake> Provided that XEmacs 21.1.x is in the PATH ahead of any of the Jake> Sun Workshop directories, the following function should work Jake> like a charm: Jake> ;; Jake> ;; Initialization required for interfacing with Sun's Workshop. Jake> ;; Do M-x workshop-start to initiate a debugging session. Jake> ;; Jake> (defun jnc-workshop() Jake> "Initializes Sun Workshop. Use M-x workshop-start to invoke the debugger" Jake> (interactive) Jake> (when (featurep 'tooltalk) Jake> (load "tooltalk-macros") Jake> (load "tooltalk-util") Jake> (load "tooltalk-init")) Jake> (when (featurep 'sparcworks) Jake> (load "sunpro-init") Jake> (load "ring") Jake> (load "comint") Jake> (load "annotations") Jake> (sunpro-startup)) Jake> (setq bar-cursor nil))

1 participants
0 comments

Recursive entries in package-index.LATEST 23 years, 7 months

James N. Potts

While parsing the package-index file to build a nice tree of who requires who, I found that some of the entries contain themselves. For example: (package-get-update-base-entry (quote (gnus --^^^^ (standards-version 1.1 version "1.55" author-version "5.8.8" date "2001-02-14" build-date "2001-02-18" maintainer "XEmacs Development Team <xemacs-beta(a)xemacs.org>" distribution xemacs priority medium category "comm" dump nil description "The Gnus Newsreader and Mailreader." filename "gnus-1.55-pkg.tar.gz" md5sum "926b8cb1b770bf367d260f7b1f33b017" size 2353348 provides (gnus message) requires (gnus w3 mh-e mailcrypt rmail eterm mail-lib --------------^^^^ xemacs-base fsf-compat) type regular )) )) Is there a reason for this, or is it an accident? -Jim

2 participants
1 comments

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

XEmacs-Beta April 2001