1:46 p.m.
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.
Please describe as succinctly as possible:
- What happened.
- What you thought should have happened.
- Precisely what you were doing at the time.
Please also include any C or lisp back-traces that you may have.
================================================================
Dear Bug Team!
I'm a Russian user using KOI8-R codepage and non-MULE XEmacs. I have
troubles in i-search:
1. In dired. I enter i-search, and for ascii it works fine, but the
first Cyrillic letter aborts it and writes a message like:
``Cyrillic_te not defined.'' or ``т not defined.'' -- it depends,
whether it's X (in my case, localized with xkb) or tty (rxvt, linux
console, whatever).
On X[1] it's reproduceable with `xemacs -q` and (require 'ru-keys),
on tty[2] -- with `xemacs -q` and (set-input-mode (car
(current-input-mode)) (nth 1 (current-input-mode)) 0). (Without these
forms, `xemacs -q` has no ability to enter Cyrillic letters.)
Backtraces:
[1]: Signaling: (undefined-keystroke-sequence [#<keypress-event Cyrillic_te>])
[2]: Signaling: (undefined-keystroke-sequence [#<keypress-event т>])
2. In view-mode on tty (in X everything is fine). Same story with
reproducing. Same backtrace, as [2].
Please note, that switching the language does not abort i-search here:
it's bound to ru-keys-noop in X by ru-keys.
================================================================
System Info to help track down your bug:
---------------------------------------
uname -a: Linux unix.home 2.4.20 #11 Sat Nov 30 03:51:36 MSK 2002 i686 unknown
../configure '--with-gpm=no' '--with-tiff=no'
'--mail-locking=flock' '--with-site-lisp=yes' '--with-modules=no'
'--with-widgets=athena' '--with-dialogs=athena' '--pdump'
XEmacs 21.4.11 "Native Windows TTY Support" configured for `i686-pc-linux'.
Compilation / Installation:
Source code location: /var/tmp/xemacs-21.4
Installation prefix: /usr/local
Runtime library search path: /usr/lib:/lib:/usr/local/lib:/usr/X11R6/lib
Operating system description file: `s/linux.h'
Machine description file: `m/intel386.h'
Compiler: gcc -O2 -pipe -march=athlon-tbird
Relocating allocator for buffers: no
GNU version of malloc: yes
- Using Doug Lea's new malloc from the GNU C Library.
Window System:
Compiling in support for the X window system:
- X Windows headers location: /usr/X11/include
- X Windows libraries location: /usr/X11R6/lib
- Handling WM_COMMAND properly.
Compiling in support for the Athena widget set:
- Athena headers location: X11/Xaw
- Athena library to link: Xaw
Using Lucid menubars.
Using Lucid scrollbars.
Using Athena dialog boxes.
Using Athena native widgets.
TTY:
Compiling in support for ncurses.
Images:
Compiling in support for GIF images (builtin).
Compiling in support for XPM images.
Compiling in support for PNG images.
Compiling in support for JPEG images.
Compiling in support for X-Face message headers.
Sound:
Compiling in support for sound (native).
Databases:
Compiling in support for Berkeley database.
Internationalization:
Mail:
Compiling in support for "flock" mail spool file locking method.
Other Features:
Inhibiting IPv6 canonicalization at startup.
Using the new portable dumper.
Load-Path Lisp Shadows:
----------------------
(/usr/local/lib/xemacs/site-packages/lisp/gnus/pop3
/usr/local/lib/xemacs/xemacs-packages/lisp/mail-lib/pop3
/usr/local/lib/xemacs/site-packages/lisp/gnus/rfc2104
/usr/local/lib/xemacs/xemacs-packages/lisp/mail-lib/rfc2104
/usr/local/lib/xemacs/site-packages/lisp/gnus/starttls
/usr/local/lib/xemacs/xemacs-packages/lisp/mail-lib/starttls)
Installed XEmacs Packages:
-------------------------
((xemacs-devel:version 1.4 :type single-file)
(xemacs-base :version 1.58 :type regular)
(view-process :version 1.11 :type regular)
(vc :version 1.3 :type regular)
(tramp :version 1.02 :type regular)
(textools :version 1.12 :type regular)
(text-modes :version 1.37 :type single-file)
(texinfo :version 1.2 :type regular)
(supercite :version 1.19 :type regular)
(strokes :version 1.08 :type regular)
(speedbar :version 1.22 :type regular)
(sounds-au :version 1.1 :type regular)
(sh-script :version 1.13 :type regular)
(sgml :version 1.08 :type regular)
(scheme :version 1.11 :type regular)
(reftex :version 1.24 :type regular)
(prog-modes
:version
1
.49
:type
single-file)
(pcomplete :version 1.01 :type regular)
(pcl-cvs :version 1.58 :type regular)
(os-utils :version 1.26 :type single-file)
(net-utils :version 1.23 :type single-file)
(misc-games :version 1.15 :type single-file)
(mine :version 1.14 :type regular)
(mail-lib :version 1.43 :type regular)
(igrep :version 1.08 :type regular)
(hm--html-menus :version 1.17 :type regular)
(games :version 1.13 :type regular)
(fsf-compat :version 1.11 :type single-file)
(frame-icon :version 1.09 :type regular)
(footnote :version 1.13 :type regular)
(eterm :version 1.13 :type regular)
(eshell :version 1.02 :type regular)
(emerge :version 1.09 :type regular)
(elib :version 1.1 :type single-file)
(efs :version 1.27 :type regular)
(edit-utils :version 1.76 :type single-file)
(ediff :version 1.36 :type regular)
(edebug :version 1.14 :type regular)
(dired :version 1.11 :type regular)
(debug :version 1.14 :type regular)
(cookie :version 1.13 :type regular)
(cc-mode :version 1.27 :type regular)
(calendar :version 1.18 :type regular)
(calc :version 1.18 :type regular)
(c-support :version 1.16 :type single-file)
(bbdb :version 1.17 :type regular)
(auctex :version 1.29 :type regular)
(apel :version 1.25 :type regular))
Features:
--------
(gnus-ems gnus-xmas efs-cu xemacsbug shadow gnuserv mouse-wheel
lazy-lock permanent-buffers tramp advice advice-preload shell
format-spec timer comment-region-or-line info cperl-mode cus-edit
wid-edit hyper-apropos dired-xemacs dired dired-faces header-action
message-utils bbdb timezone supercite mail-extr regi message messagexmas
mml mml-sec mml-smime smime comint ring dig mml1991 mml2015 mm-decode
executable mm-bodies mm-encode mailcap mail-parse rfc2045 rfc2231
rfc2047 gnus-util netrc qp ietf-drums time-date parse-time mail-abbrevs
sendmail rfc822 nnheader nnheaderxm mm-util mail-prsvr mail-utils
mailheader canlock gnus-load cus-load paren blink-paren font-lock
cus-face filladapt crypt crypt++ ru-keys xemacs-devel-autoloads
xemacs-base-autoloads view-process-autoloads vc-autoloads
tramp-autoloads textools-autoloads text-modes-autoloads
texinfo-autoloads supercite-autoloads strokes-autoloads
speedbar-autoloads sounds-au-autoloads sh-script-autoloads
sgml-autoloads scheme-autoloads reftex-autoloads prog-modes-autoloads
pcomplete-autoloads pcl-cvs-autoloads os-utils-autoloads
net-utils-autoloads misc-games-autoloads mine-autoloads
mail-lib-autoloads igrep-autoloads hm--html-menus-autoloads
games-autoloads fsf-compat-autoloads frame-icon-autoloads
footnote-autoloads eterm-autoloads eshell-autoloads emerge-autoloads
elib-autoloads efs-autoloads edit-utils-autoloads ediff-autoloads
edebug-autoloads dired-autoloads debug-autoloads cookie-autoloads
cc-mode-autoloads calendar-autoloads calc-autoloads c-support-autoloads
bbdb-autoloads auctex-autoloads apel-autoloads lisp-autoloads loadhist
auto-show fontl-hooks x-iso8859-1 gutter-items menubar-items x-menubar
mode-motion mouse itimer auto-save lisp-mode easymenu iso8859-1 page
buff-menu lib-complete help-nomule cus-file derived frame text-props
obsolete cus-start custom widget cl-extra mini-cl cl cl-19 packages
backquote very-early-lisp lucid-scrollbars cut-buffer lucid-menubars
athena-dialogs x c-balloon-help tty-frames tty toolbar native-sound
scrollbar unix-processes multicast network-streams subprocesses
menu-accelerator-support menubar berkeley-db md5 xemacs gutter png gif
jpeg xface xpm xbm lisp-float-type linux dialog devices window-system
base64)
Recent keystrokes:
-----------------
M-x r e p o TAB t TAB BS r t - x e m TAB RET
Recent messages (most recent first):
-----------------------------------
Fontifying *mail to xemacs-beta＠xemacs.org*...
Loading gnus-xmas...done
Loading gnus-xmas...
Loading gnus-ems...
Loading efs-cu...done
Loading efs-cu...
Loading xemacsbug...done
Loading xemacsbug...
Making completion list...
Loading gnuserv...done
Fontifying *scratch*... done.
Loading gnuserv...
Loading lazy-lock...done
Fontifying *scratch*... (syntactically).
Fontifying *scratch*...
Loading lazy-lock...
Loading permanent-buffers...done
Loading permanent-buffers...
Loading /home/deskpot/.xemacs/faces...done
Loading /home/deskpot/.xemacs/faces...
6:08 a.m.
Alexey Mahotkin <alexm(a)xemacs.org> is working on a similar problem.
You might want to ask on the xemacs-users-ru(a)xemacs.org list, too.
VK> [1]: Signaling: (undefined-keystroke-sequence [#<keypress-event
Cyrillic_te>])
VK> [2]: Signaling: (undefined-keystroke-sequence [#<keypress-event т>])
I don't understand how ru-keys works. But I would think for X
(define-key global-map [Cyrillic_te] #'self-insert-command)
(put Cyrillic_te 'ascii-character ?т)
and for TTY
(define-key global-map [?т] #'self-insert-command)
should give correct results. AFAIK isearch does not override
global-map as a whole. Of course you have to repeat for all
characters, which is annoying; if you send me a file I'll see about
getting it put somewhere sensible.
--
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.
6:52 a.m.
Sorry, in the last message I've misspelt the xemacs-users-ru address, so,
Stephen and xemacs-beta, please discard that one in the favor of this
one -- I'm resending it with the correct address.
>>>> "SJT" == Stephen J Turnbull writes:
SJT> Alexey Mahotkin <alexm(a)xemacs.org> is working on a similar problem.
SJT> You might want to ask on the xemacs-users-ru(a)xemacs.org list, too.
I think, it's a good idea to CC the latter. So doing so.
VK> [1]: Signaling: (undefined-keystroke-sequence [#<keypress-event
VK> Cyrillic_te>])
VK> [2]: Signaling: (undefined-keystroke-sequence [#<keypress-event т>])
SJT> I don't understand how ru-keys works. But I would think for X
Ok, let's leave ru-keys. AFAIK, on tty it sets up only the case table.
It's simply `xemacs -q -nw` with:
(set-input-mode (car (current-input-mode))
(nth 1 (current-input-mode))
0)
The problem is present, while Cyrillic letters (and, say, umlauts --
I've heard, there's a similar problem for German users) work in most
places -- say, in i-search in fundamental-mode.
But not in i-search in dired or view-mode, or gnus-summary-mode
(according to the news:gnu.emacs.gnus newsgroup, the problem with the
latter also occurs in MULE-enabled XEmacs -- but not in GNU Emacs; Gnus
towers are also notified of this bug but in today's Gnus snapshot
there's nothing to work around this)...
SJT> (define-key global-map [Cyrillic_te] #'self-insert-command)
SJT> (put Cyrillic_te 'ascii-character ?т)
Maybe, a typo in the second string? It does not evaluate here.
SJT> and for TTY
SJT> (define-key global-map [?т] #'self-insert-command)
SJT> should give correct results.
I'm afraid, it doesn't. =(( Alike strings seem not to do _anything_
here.
SJT> AFAIK isearch does not override global-map as a whole.
Let me repeat: the problem appears only in i-search in certain modes;
others are OK.
---Vas
6:52 a.m.
>>>> "SJT" == Stephen J Turnbull writes:
SJT> Alexey Mahotkin <alexm(a)xemacs.org> is working on a similar problem.
SJT> You might want to ask on the xemacs-users-ru(a)xemacs.org list, too.
I think, it's a good idea to CC the latter. So doing so.
VK> [1]: Signaling: (undefined-keystroke-sequence [#<keypress-event
VK> Cyrillic_te>])
VK> [2]: Signaling: (undefined-keystroke-sequence [#<keypress-event т>])
SJT> I don't understand how ru-keys works. But I would think for X
Ok, let's leave ru-keys. AFAIK, on tty it sets up only the case table.
It's simply `xemacs -q -nw` with:
(set-input-mode (car (current-input-mode))
(nth 1 (current-input-mode))
0)
The problem is present, while Cyrillic letters (and, say, umlauts --
I've heard, there's a similar problem for German users) work in most
places -- say, in i-search in fundamental-mode.
But not in i-search in dired or view-mode, or gnus-summary-mode
(according to the news:gnu.emacs.gnus newsgroup, the problem with the
latter also occurs in MULE-enabled XEmacs -- but not in GNU Emacs; Gnus
towers are also notified of this bug but in today's Gnus snapshot
there's nothing to work around this)...
SJT> (define-key global-map [Cyrillic_te] #'self-insert-command)
SJT> (put Cyrillic_te 'ascii-character ?т)
Maybe, a typo in the second string? It does not evaluate here.
SJT> and for TTY
SJT> (define-key global-map [?т] #'self-insert-command)
SJT> should give correct results.
I'm afraid, it doesn't. =(( Alike strings seem not to do _anything_
here.
SJT> AFAIK isearch does not override global-map as a whole.
Let me repeat: the problem appears only in i-search in certain modes;
others are OK.
---Vas
7:56 a.m.
>>>> "VK" == Vasily Korytov
<deskpot(a)myrealbox.com> writes:
VK> But not in i-search in dired or view-mode, or
VK> gnus-summary-mode (according to the news:gnu.emacs.gnus
VK> newsgroup, the problem with the latter also occurs in
VK> MULE-enabled XEmacs -- but not in GNU Emacs;
Ah, OK, all of those are likely to override the whole global-map since
they use many letters for commands and don't want others to ever
self-insert. Hm ... dired uses suppress-keymap for this.
SJT> (define-key global-map [Cyrillic_te] #'self-insert-command)
SJT> (put Cyrillic_te 'ascii-character ?т)
VK> Maybe, a typo in the second string? It does not evaluate here.
You're right, the Cyrillic_te needs to be quoted
(put 'Cyrillic_te 'ascii-character ?т)
However, this is evidently irrelevant; if the keysym were unrecognized
you couldn't use isearch in _any_ mode. Don't do either of those
define-keys or the put 'ascii-character. It shouldn't hurt but it
can't help.
I'd be curious to know if quoting a Cyrillic character with C-q
helps. (Just an experiment and maybe a workaround for the occasional
case where you only need a couple of Cyrillic characters for the
search.)
Try M-x find-library RET isearch-mode RET, find the variable
isearch-mode-keymap, and in the loop where it binds most stuff to
'isearch-printing-character, change the while loop bound from 127 to
255 [this is non-Mule, right?], and M-x byte-compile-and-load-file RET
RET. Then try it.
If that gives satisfactory results, to get it in future invocations of
XEmacs you'll have to redump XEmacs (rm src/xemacs; make &
get-coffee-or-beverage-of-choice ;-) because isearch-mode is
preloaded.
The current code is very very wrong, but I'm not sure what is the
Right Thing to do yet. However, I think the above workaround is safe,
and if it works for you I will recommend it to Vin for inclusion in
21.4. (Protected by (if (featurep 'mule) 127 255) until we know it is
safe for Mule---if someone out there in Cyrillic or ISO 8859-land who
uses Mule would confirm it doesn't blow up your terminal I'd love to
hear about it! Note: it will not help for Cyrillic under Mule because
Cyrillic doesn't map to 128--255 under Mule.)
--
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.
5:01 p.m.
>>>> "SJT" == Stephen J Turnbull writes:
I've finally guessed. =))
SJT> AFAIK isearch does not override global-map as a whole.
Yes, and it's the problem. (suppress-keymap ...) in a nutshell undefines
all Cyrillic keys (as well as ASCII ones) -- but isearch, when called
From a mode with such a map, needs them -- and doesn't know, how to
restore them.
There's a block in isearch-mode.el:
--8<------------------------schnipp------------------------->8---
;; Bind all printing characters to `isearch-printing-char'.
;; This isn't normally necessary, but if a printing character were
;; bound to something other than self-insert-command in global-map,
;; then it would terminate the search and be executed without this.
(let ((i 32)
(str (make-string 1 0)))
(while (< i 127)
(aset str 0 i)
(define-key map str 'isearch-printing-char)
(setq i (1+ i))))
--8<------------------------schnapp------------------------->8---
Is seems, the only thing one needs is to add Cyrillic initialization
there. =)) I can write mapping for all letters keysums, as Stephen
suggests, like this:
(define-key isearch-mode-map [т] 'isearch-printing-char)
But would it be grecefil? IIRC, GNU Emacs doesn't have this problem. Can
anyone say, how this case is worked around there? If there's no a better
way for two days, I'll post a quick Russian work-around, as described
before.
---Vas
6:24 p.m.
>>>> "VK" == Vasily Korytov writes:
VK> If there's no a better way for two days, I'll post a quick Russian
VK> work-around, as described before.
I decided not to wait and wrote it now. It works here. =))
---Vas
--8<------------------------schnipp------------------------->8---
;; isearch in XEmacs (at least, up to 21.4.11) is broken: in modes, that
;; have (suppress-keymap ...) in their map, it can accept only ASCII
;; letters. This hack seems to fix it.
;; Public domain. Originally by Vasily Korytov.
;; I use Russian letters and keysums. If you don't, change this:
;;
(defconst self-insert-list
(list
'Cyrillic_IO 'Ё
'Cyrillic_io 'ё
'Cyrillic_SHORTI 'Й
'Cyrillic_shorti 'й
'Cyrillic_TSE 'Ц
'Cyrillic_tse 'ц
'Cyrillic_U 'У
'Cyrillic_u 'у
'Cyrillic_KA 'К
'Cyrillic_ka 'к
'Cyrillic_IE 'Е
'Cyrillic_ie 'е
'Cyrillic_EN 'Н
'Cyrillic_en 'н
'Cyrillic_GHE 'Г
'Cyrillic_ghe 'г
'Cyrillic_SHA 'Ш
'Cyrillic_sha 'ш
'Cyrillic_SHCHA 'Щ
'Cyrillic_shcha 'щ
'Cyrillic_ZE 'З
'Cyrillic_ze 'з
'Cyrillic_HA 'Х
'Cyrillic_ha 'х
'Cyrillic_HARDSIGN 'Ъ
'Cyrillic_hardsign 'ъ
'Cyrillic_EF 'Ф
'Cyrillic_ef 'ф
'Cyrillic_YERU 'Ы
'Cyrillic_yeru 'ы
'Cyrillic_VE 'В
'Cyrillic_ve 'в
'Cyrillic_A 'А
'Cyrillic_a 'а
'Cyrillic_PE 'П
'Cyrillic_pe 'п
'Cyrillic_ER 'Р
'Cyrillic_er 'р
'Cyrillic_O 'О
'Cyrillic_o 'о
'Cyrillic_EL 'Л
'Cyrillic_el 'л
'Cyrillic_DE 'Д
'Cyrillic_de 'д
'Cyrillic_ZHE 'Ж
'Cyrillic_zhe 'ж
'Cyrillic_E 'Э
'Cyrillic_e 'э
'Cyrillic_YA 'Я
'Cyrillic_ya 'я
'Cyrillic_CHE 'Ч
'Cyrillic_che 'ч
'Cyrillic_ES 'С
'Cyrillic_es 'с
'Cyrillic_EM 'М
'Cyrillic_em 'м
'Cyrillic_I 'И
'Cyrillic_i 'и
'Cyrillic_TE 'Т
'Cyrillic_te 'т
'Cyrillic_SOFTSIGN 'Ь
'Cyrillic_softsign 'ь
'Cyrillic_BE 'Б
'Cyrillic_be 'б
'Cyrillic_YU 'Ю
'Cyrillic_yu 'ю
)
"List of non-ASCII letters, that are in use.")
(let ((list self-insert-list))
(while list
(define-key isearch-mode-map (car list) 'isearch-printing-char)
(setq list (cdr list))))
(provide 'isearch-vk-fix-keymap)
--8<------------------------schnapp------------------------->8---
3:02 a.m.
>>>> "VK" == Vasily Korytov
<deskpot(a)myrealbox.com> writes:
VK> I decided not to wait and wrote it now. It works here. =))
I think it's preferable to change the `127' in the original loop to
`(if (featurep 'mule) 127 255)'. This works for everybody in a
non-Mule XEmacs, including non-ISO-conformant charsets like
Windows-125x and (IIRC) KOI8 which put graphic characters in the C1
set (== code points 0x80--0x9F).
The reason for the featurep is that I want to fix all of Mule (if it's
broken in a similar way), not just ISO 8859-1.
Many thanks for testing it.
--
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.
8105
days inactive
8108
days old
7 comments
2 participants
participants (2)
-
Stephen J. Turnbull -
Vasily Korytov