April 2001 - XEmacs-Beta - XEmacs Mailing List Archives

Guile support in XEmacs. 23 years, 7 months

monkeyiq

Hi, I have seen the links to gemacs and other works for using scheme as a macro language in GNU Emacs. I am wondering if there is any work on doing this in XEmacs? The problem that I am having is that I quite enjoy using XEmacs but would like to write a few packages and integrate by XEmacs and witme usage (witme == filemgr + scheme + evas) and would obviously like to use only scheme in this utopian desktop. I am wondering if there is any works like gemacs to use scheme along side Emacs lisp as another interpreter in XEmacs? -- --------------------------------------------------- It's the question, http://witme.sourceforge.net If you think education is expensive, try ignorance. -- Derek Bok, president of Harvard

9 participants
11 comments

Re: Agony, persistent solaris8 hang of doom: Can I help debug this? 23 years, 7 months

John M. Adams

Andy Piper <andyp(a)bea.com> writes: > Try and get a C stacktrace for the hang (gdb) where #0 0xfec18224 in _poll () from /usr/lib/libc.so.1 #1 0xfebcb808 in select () from /usr/lib/libc.so.1 #2 0xfef1bdc0 in _XWaitForReadable () from /opt/X11R5/lib/libX11.so.4 #3 0xfef1bbd8 in _XRead () from /opt/X11R5/lib/libX11.so.4 #4 0xfef265fc in _XEventsQueued () from /opt/X11R5/lib/libX11.so.4 #5 0xfef35ae4 in XEventsQueued () from /opt/X11R5/lib/libX11.so.4 #6 0x1daed0 in emacs_Xt_quit_p () at event-Xt.c:2829 #7 0xe1130 in event_stream_quit_p () at event-stream.c:596 #8 0x1b13f4 in check_what_happened () at signal.c:497 #9 0x97680 in Ffuncall (nargs=1, args=0xffbee748) at eval.c:3474 #10 0x9bc58 in call1 (fn=5050196, arg0=11554280) at eval.c:4183 #11 0xdc954 in execute_internal_event (event=11554280) at event-stream.c:2939 #12 0xdf544 in Fdispatch_event (event=11554280) at event-stream.c:4170 #13 0x7009c in Fcommand_loop_1 () at cmdloop.c:583 #14 0x70334 in command_loop_1 (dummy=5393664) at cmdloop.c:494 #15 0x934f8 in condition_case_1 (handlers=3018748, bfun=0x702e8 <command_loop_1>, barg=3254276, hfun=0x70394 <cmd_error>, harg=3254276) at eval.c:1651 #16 0x704a4 in command_loop_2 (dummy=3254276) at cmdloop.c:256 #17 0x9c398 in internal_catch (tag=3334340, func=0x70458 <command_loop_2>, arg=3254276, threw=0x0) at eval.c:1317 #18 0x6f7b4 in initial_command_loop (load_me=3254276) at cmdloop.c:305 #19 0x8edc0 in sort_args (argc=3064832, argv=0xffbeedac) at emacs.c:2253 If you look at truss while moving the mouse or typing you see this sort of thing. There is no display activity. poll(0xFFBEE1A8, 1, -1) (sleeping...) Received signal #22, SIGPOLL, in poll() [caught] siginfo: SIGPOLL POLL_IN fd=0 band=0 poll(0xFFBEE1A8, 1, -1) Err#4 EINTR setcontext(0xFFBEDE90) Received signal #22, SIGPOLL [caught] siginfo: SIGPOLL POLL_IN fd=0 band=0 setcontext(0xFFBEDE90) Received signal #22, SIGPOLL, in poll() [caught] siginfo: SIGPOLL POLL_IN fd=0 band=0 poll(0xFFBEE1A8, 1, -1) Err#4 EINTR setcontext(0xFFBEDE90) Received signal #22, SIGPOLL, in poll() [caught] siginfo: SIGPOLL POLL_IN fd=0 band=0 Anything else? > > andy > > At 02:42 PM 4/26/01 -0400, John M. Adams wrote: > >Help, I'm dyin' out here. > > > >I used to use a ga release 21 version of xemacs until I upgraded to > >solaris8. At that time, I started getting random hangs. Since then > >I've used many different betas and have the same experience. I > >reported a bug a while back. > > > >It's easy to reproduce this over the course of a day or two. If > >someone could give me a few hints, I'd be happy to poke around in the > >broken running image. > > > >I'm using xemacs-21.2-b45 which I know is behind. If we think this is > >fixed, fine. Otherwise, I'll be happy to test the latest beta. > > > >Thanks alot. > > > >This problem REALLY hurts. I'd like to help as much as possible. > > > >-- > >John M. Adams -- John M. Adams

5 participants
4 comments

Agony, persistent solaris8 hang of doom: Can I help debug this? 23 years, 7 months

John M. Adams

Help, I'm dyin' out here. I used to use a ga release 21 version of xemacs until I upgraded to solaris8. At that time, I started getting random hangs. Since then I've used many different betas and have the same experience. I reported a bug a while back. It's easy to reproduce this over the course of a day or two. If someone could give me a few hints, I'd be happy to poke around in the broken running image. I'm using xemacs-21.2-b45 which I know is behind. If we think this is fixed, fine. Otherwise, I'll be happy to test the latest beta. Thanks alot. This problem REALLY hurts. I'd like to help as much as possible. -- John M. Adams

3 participants
2 comments

new version of feedmail (v10) 23 years, 7 months

WJCarpenter

A new version, patchlevel 10, of feedmail.el is available here: http://www.carpenter.org/feedmail This is a "safe" release, by which I mean that it doesn't have any exotic new stuff that hasn't been in use for a while. (There was a very short cycle between patchlevel 9 and patchlevel 10 because of some important internationalization things brought to my attention.) -- bill(a)carpenter.ORG (WJCarpenter) PGP 0x91865119 38 95 1B 69 C9 C6 3D 25 73 46 32 04 69 D6 ED F3 The short form of what it is: ;;; feedmail.el --- assist other email packages to massage outgoing messages ;;; A replacement for parts of GNUemacs' sendmail.el (specifically, ;;; it's what handles your outgoing mail after you hit C-c C-c in mail ;;; mode). See below for a list of additional features, including the ;;; ability to queue messages for later sending. If you are using ;;; fakemail as a subprocess, you can switch to feedmail and eliminate ;;; the use of fakemail. Here is the changelog: ;; Change log: ;; original, 31 March 1991 ;; patchlevel 1, 5 April 1991 ;; patchlevel 2, 24 May 1991 ;; 5-may-92 jwz Conditionalized calling expand-mail-aliases, since that ;; function doesn't exist in Lucid GNU Emacs or when using ;; mail-abbrevs.el. ;; patchlevel 3, 3 October 1996 ;; added queue stuff; still works in v18 ;; patchlevel 4, issued by someone else ;; patchlevel 5, issued by someone else ;; patchlevel 6, not issued as far as I know ;; patchlevel 7, 20 May 1997 ;; abandon futile support of GNUemacs v18 (sorry if that hurts you) ;; provide a DATE: header by default ;; provide a default for generating MESSAGE-ID: header contents ;; and use it by default (slightly changed API) ;; return value from feedmail-run-the-queue ;; new wrapper function feedmail-run-the-queue-no-prompts ;; user-mail-address as default for FROM: ;; properly deal with RESENT-{TO,CC,BCC} ;; BCC and RESENT-* now included in smart filling ;; limited support for a "drafts" directory ;; user-configurable default message action ;; allow timeout for confirmation prompt (where available) ;; move FCC handling to as late as possible to get max ;; header munging in the saved file ;; work around sendmail.el's prompts when working from queue ;; more reliably detect voluntary user bailouts ;; offer to save modified buffers visiting queue files ;; offer to delete old file copies of messages being queued ;; offer to delete queue files when sending immediately ;; queue filename convention preserves queue order ;; default queue and draft directory names that work on VMS ;; deduced address list now really a list, not a string (API change) ;; no more address buffer ;; when sending immediately, brief reminder of queue/draft counts ;; copy trace of smtpmail stuff to feedmail error buffer on no-go ;; more granularity on when to confirm sending ;; pause a bit for errors while running queue ;; try to clean up some pesky auto-save files from the ;; queue/draft directories ;; feedmail-force-expand-mail-aliases in case you can't figure ;; any other way ;; cleanup some of my sloppiness about case-fold-search (a strange ;; variable) ;; best effort following coding conventions from GNUemacs ;; elisp manual appendix ;; "customize" (see custom.el) ;; when user selects "immediate send", clear action prompt since ;; hooks may take a while to operate, and user may think the ;; response didn't take ;; fixes to the argument conventions for the ;; feedmail-queue-runner-* functions; allows ;; feedmail-run-the-queue[-no-prompts] to properly be called ;; non-interactively ;; eliminate reliance on directory-sep-char and feedmail-sep-thing ;; tweak smart filling (reminded of comma problem by levitte(a)lp.se) ;; option to control writing in text vs binary mode ;; patchlevel 8, 15 June 1998 ;; reliable re-editing of text-mode (vs binary) queued messages ;; user option to keep BCC: in FCC: copy (keep by default) ;; user option to delete body from FCC: copy (keep by default) ;; feedmail-deduce-bcc-where for envelope (API change for ;; feedmail-deduce-address list) ;; feedmail-queue-alternative-mail-header-separator ;; at message action prompt, "I"/"S" bypass message confirmation prompt ;; feedmail-mail-send-hook-splitter, feedmail-mail-send-hook, ;; feedmail-mail-send-hook-queued ;; user can supply stuff for message action prompt ;; variable feedmail-queue-runner-confirm-global, function feedmail-run-the-queue-global-prompt ;; bugfix: absolute argument to directory-files (tracked down for me ;; by gray(a)austin.apc.slb.com (Douglas Gray Stephens)); relative ;; pathnames can tickle stuff in ange-ftp remote directories ;; (perhaps because feedmail is careless about its working ;; directory) ;; feedmail-deduce-envelope-from ;; always supply envelope "from" (user-mail-address) to sendmail ;; feedmail-message-id-suffix ;; feedmail-queue-reminder, feedmail-queue-reminder-alist (after suggestions ;; and/or code fragments from tonyl(a)Eng.Sun.COM (Tony Lam) and ;; burge(a)newvision.com (Shane Burgess); bumped up the default value of ;; feedmail-queue-chatty-sit-for since info is more complex sometimes ;; feedmail-enable-spray (individual transmissions, crude mailmerge) ;; blank SUBJECT: no longer a special case; see feedmail-nuke-empty-headers ;; fiddle-plexes data structure used lots of places; see feedmail-fiddle-plex-blurb ;; feedmail-fiddle-plex-user-list ;; feedmail-is-a-resend ;; honor mail-from-style in constructing default for feedmail-from-line ;; re-implement feedmail-from-line and feedmail-sender-line with ;; fiddle-plexes; slightly modified semantics for feedmail-sender-line ;; feedmail-queue-default-file-slug; tidy up some other slug details ;; feedmail-queue-auto-file-nuke ;; feedmail-queue-express-to-queue and feedmail-queue-express-to-draft ;; strong versions of "q"ueue and "d"raft answers (always make a new file) ;; patchlevel 9, 23 March 2001 ;; feedmail-queue-buffer-file-name to work around undesirable mail-send prompt ;; at message action prompt, can scroll message buffer with "<" and ">"; ;; C-v no longer scrolls help buffer ;; conditionalize (discard-input) in message action prompt to avoid killing ;; define-kbd-macro ;; fixed error if feedmail-x-mailer-line was nil ;; feedmail-binmail-template only uses /bin/rmail if it exists ;; relocate feedmail-queue-alternative-mail-header-separator stuff ;; added feedmail-vm-mail-mode, which make a good auto-mode-alist entry ;; for FQM files if you're a VM user ;; change buffer-substring calls to buffer-substring-no-properties for ;; speed-up (suggested by Howard Melman <howard(a)silverstream.com>) ;; feedmail-sendmail-f-doesnt-sell-me-out to contol "-f" in call to sendmail ;; in feedmail-buffer-to-sendmail ;; better trapping of odd conditions during the running of the queue; ;; thanks to Yigal Hochberg for helping me test much of this by remote ;; control ;; feedmail-debug and feedmail-debug-sit-for ;; feedmail-display-full-frame ;; feedmail-queue-express-hook ;; added example function feedmail-spray-via-bbdb ;; use expand-file-name for setting default directory names ;; define feedmail-binmail-linuxish-template as a suggestion for ;; the value of feedmail-binmail-template on Linux and maybe other ;; systems with non-classic /bin/[r]mail behavior ;; guard against nil user-mail-address in generating MESSAGE-ID: ;; feedmail-queue-slug-suspect-regexp is now a variable to ;; accomodate non-ASCII environments (thanks to ;; Makoto.Nakagawa(a)jp.compaq.com for this suggestion) ;; feedmail-buffer-to-smtp, to parallel feedmail-buffer-to-smtpmail ;; patchlevel 10, 22 April 2001 ;; DATE: and MESSAGE-ID stuff now forces system-time-locale to "C" ;; (brought to my attention by Makoto.Nakagawa(a)jp.compaq.com)

1 participants
0 comments

initializations in .emacs file 23 years, 7 months

Jeff Mincy

In my .emacs file I have initializations to smash various modes in various ways. For example: (setq sql-mode-font-lock-keywords sql-mode-oracle-font-lock-keywords) (define-key tcl-mode-map [(meta backspace)] 'backward-kill-word) (define-key makefile-mode-map "$" 'self-insert-command) (define-key makefile-mode-map "\^x\$" 'makefile-insert-macro-ref) However doing these things require that I load all of the packages first. (require 'sql) (require 'tcl) (require 'makefile) Is there some standard way to do initializations like these (that require that the file be loaded first) without causing the file to be loaded? Basically what I want is a load-hook, but most packages do not have load hooks (comint has one). I hacked something up to do what I wanted, but there's got to be a better way. -jeff Example: ;; stuff to do after sql gets loaded (after-provided 'sql '(lambda () (setq sql-mode-font-lock-keywords sql-mode-oracle-font-lock-keywords))) ;; stuff to do after tcl gets loaded (after-provided 'tcl '(lambda () (define-key tcl-mode-map [(meta backspace)] 'backward-kill-word))) ;; stuff to do after makefile gets loaded (after-provided 'makefile '(lambda () (define-key makefile-mode-map "$" 'self-insert-command) (define-key makefile-mode-map "\^x\$" 'makefile-insert-macro-ref))) ;; Keep track of nesting level of require/load. Don't do after-provide stuff until we ;; know that we are done loading the files. Some files do the provide at the top of the file (defvar load-level 0) ;; Track packages that have been provided. (defvar provided ()) (put 'after-provided 'lisp-indent-function 1) (defun after-provided (symbol function) (if (memq symbol features) ;; Package has already been loaded, just do it. (progn ;;(message "already loaded %s" symbol) (funcall function)) ;; Keep function around until package gets provided. (setf (get symbol 'after-provide) (cons function (get symbol 'after-provide))))) (defun provided (symbol) (while (not (null (get symbol 'after-provide))) ;;(message "after %s" symbol) (funcall (pop (get symbol 'after-provide))))) (defadvice load (around first activate protect) (incf load-level) ad-do-it (decf load-level) ;; If load has caused some packages to have been provided then do after stuff, when done loading. (when (zerop load-level) (while (not (null provided)) (provided (pop provided))))) (defadvice provide (around first activate protect) (incf load-level) ad-do-it (decf load-level) (push (ad-get-arg 0) provided) ;; Some packages to have been provided do after stuff, when done loading. ;; Doing a (provide 'foo) at top level will cause the after stuff to run. (when (zerop load-level) (while (not (null provided)) (provided (pop provided))))) (defadvice require (around first activate protect) (incf load-level) ad-do-it (decf load-level) ;; We're done doing require, look for packages that have been provided. (when (zerop load-level) (while (not (null provided)) (provided (pop provided))) ;; If package already loaded just do the cleanup, this also catches the case where ;; the package loaded didn't do a provide - underlying require already gave error. (provided (ad-get-arg 0))))

5 participants
5 comments

subscribe 23 years, 7 months

Jason.D.Karney＠clev.frb.org

subscribe

1 participants
0 comments

little documentation patch (for 21.2.47, 21.4.1, 21.5.0) 23 years, 7 months

John H Palmieri

I've attached a little patch for the xemacs info files; could someone apply it? (Or apply the corresponding patch to the texinfo source file...) The patch changes 'default-frame-alist' to 'default-frame-plist' where it appears, along with a sentence saying that the former variable is now considered obsolete. -- J. H. Palmieri Dept of Mathematics, Box 354350 mailto:palmieri＠math.washington.edu University of Washington http://www.math.washington.edu/~palmieri/ Seattle, WA 98195-4350

4 participants
4 comments

Cygwin 1.3.1 causes link with both linuxplay.o and ntplay.o 23 years, 7 months

Bush, David

The just released Cygwin version 1.3.1 now has a soundcard.h in /usr/include/sys. This causes configure to deduce that this is a linux system and add linuxplay.o to the objects to be linked. Since ntplay.o is automatically included, this causes a link conflict. I'm looking configure.in now. david David Bush Software Engineer Dialogic, an Intel Company Phone: (703) 633-3384 Fax: (703) 633-3319 mailto:David.Bush＠Dialogic.com http://www.dialogic.com

1 participants
0 comments

XEmacs crashes at char-equal!!!!!!!!!!! 23 years, 7 months

Konstantin Malakhanov

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 (beta40) "Persephone" [Lucid] (i586-pc-win32) of Tue Jan 16 2001 on NEVERYON configured using `configure UNKNOWN' Please describe exactly what actions triggered the bug and the precise symptoms of the bug: XEmacs crashes at the function char-equal. I just need to eval in *scratch* buffer: (char-equal ? ? ) and voila - I get Dr. Watson. Funny thing is, char-equal is used in the function of mine, which I used quite often. Until today, it has worked!!! Recent keystrokes: button1 button1up misc-user misc-user misc-user misc-user RET y button2 button2up down RET down RET SPC up up C-u B DEL y e s RET q RET RET SPC down RET SPC up up C-u B DEL y e s RET q C-h f c h a r - e q TAB RET q misc-user Recent messages (most recent first): Loading mail-abbrevs... Loading emacsbug...done Loading emacsbug... Loading view-less...done Loading view-less... nnimap: Setting marks in xemacs...done nnimap: Setting marks in xemacs... nnimap: Fetching (part of) article 1794... Prefetching article 1794 in group nnimap+mail:xemacs nnimap: Fetching (part of) article 1795...done -- Konstantin Malakhanov ISOWARE GmbH, Feithstr. 142, 58097 Hagen, Germany email: km(a)isoware.de Tel: 02331/8019-71 FAX: 02331/8019-99

2 participants
1 comments

regular expression question 23 years, 7 months

Brady Montz

Can anyone explain to me why: (string-match "\$ *a+\$+" "a a") returns 0 and (string-match "\$?: *a+\$+" "a a") returns nil while both (string-match "\$ *a\$+" "a a") and (string-match "\$?: *a\$+" "a a") return 0? I've tried it on xemacs-21.1.13 and xemacs-21.4.1 I'm confused. It seems to be that the result should be 0 in all cases. -- Brady Montz bradym(a)balestra.org

5 participants
15 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