1 ;;; mule-cmds.el --- Commands for mulitilingual environment
3 ;; Copyright (C) 1995 Electrotechnical Laboratory, JAPAN.
4 ;; Licensed to the Free Software Foundation.
5 ;; Copyright (C) 2000 Free Software Foundation, Inc.
7 ;; Keywords: mule, multilingual
9 ;; This file is part of GNU Emacs.
11 ;; GNU Emacs is free software; you can redistribute it and/or modify
12 ;; it under the terms of the GNU General Public License as published by
13 ;; the Free Software Foundation; either version 2, or (at your option)
16 ;; GNU Emacs is distributed in the hope that it will be useful,
17 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
18 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
19 ;; GNU General Public License for more details.
21 ;; You should have received a copy of the GNU General Public License
22 ;; along with GNU Emacs; see the file COPYING. If not, write to the
23 ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330,
24 ;; Boston, MA 02111-1307, USA.
28 (eval-when-compile (defvar dos-codepage))
30 ;;; MULE related key bindings and menus.
32 (defvar mule-keymap (make-sparse-keymap)
33 "Keymap for Mule (Multilingual environment) specific commands.")
35 ;; Keep "C-x C-m ..." for mule specific commands.
36 (define-key ctl-x-map "\C-m" mule-keymap)
38 (define-key mule-keymap "f" 'set-buffer-file-coding-system)
39 (define-key mule-keymap "t" 'set-terminal-coding-system)
40 (define-key mule-keymap "k" 'set-keyboard-coding-system)
41 (define-key mule-keymap "p" 'set-buffer-process-coding-system)
42 (define-key mule-keymap "x" 'set-selection-coding-system)
43 (define-key mule-keymap "X" 'set-next-selection-coding-system)
44 (define-key mule-keymap "\C-\\" 'set-input-method)
45 (define-key mule-keymap "c" 'universal-coding-system-argument)
46 (define-key mule-keymap "l" 'set-language-environment)
48 (define-key help-map "\C-L" 'describe-language-environment)
49 (define-key help-map "L" 'describe-language-environment)
50 (define-key help-map "\C-\\" 'describe-input-method)
51 (define-key help-map "I" 'describe-input-method)
52 (define-key help-map "C" 'describe-coding-system)
53 (define-key help-map "h" 'view-hello-file)
55 (defvar mule-menu-keymap
56 (make-sparse-keymap "Mule (Multilingual Environment)")
57 "Keymap for Mule (Multilingual environment) menu specific commands.")
59 (defvar describe-language-environment-map
60 (make-sparse-keymap "Describe Language Environment"))
62 (defvar setup-language-environment-map
63 (make-sparse-keymap "Set Language Environment"))
65 (defvar set-coding-system-map
66 (make-sparse-keymap "Set Coding System"))
68 (define-key-after mule-menu-keymap [set-language-environment]
69 (list 'menu-item "Set Language Environment" setup-language-environment-map
70 :help "Multilingual environment suitable for a specific language"))
71 (define-key-after mule-menu-keymap [mouse-set-font]
72 '(menu-item "Set Font/Fontset" mouse-set-font
73 :visible (fboundp 'generate-fontset-menu)
74 :help "Select a font from list of known fonts/fontsets"))
75 (define-key-after mule-menu-keymap [separator-mule]
78 (define-key-after mule-menu-keymap [toggle-input-method]
79 '(menu-item "Toggle Input Method" toggle-input-method)
81 (define-key-after mule-menu-keymap [set-input-method]
82 '(menu-item "Select Input Method..." set-input-method)
84 (define-key-after mule-menu-keymap [describe-input-method]
85 '(menu-item "Describe Input Method" describe-input-method))
86 (define-key-after mule-menu-keymap [separator-input-method]
89 (define-key-after mule-menu-keymap [set-various-coding-system]
90 (list 'menu-item "Set Coding Systems" set-coding-system-map
91 :enable 'enable-multibyte-characters))
92 (define-key-after mule-menu-keymap [view-hello-file]
93 '(menu-item "Show Multi-lingual Text" view-hello-file
94 :enable (file-readable-p
95 (expand-file-name "HELLO" data-directory))
96 :help "Display file which says HELLO in many languages")
98 (define-key-after mule-menu-keymap [separator-coding-system]
101 (define-key-after mule-menu-keymap [describe-language-environment]
102 (list 'menu-item "Describe Language Environment"
103 describe-language-environment-map
104 :help "Show multilingual settings for a specific language")
106 (define-key-after mule-menu-keymap [describe-input-method]
107 '(menu-item "Describe Input Method..." describe-input-method
108 :help "Keyboard layout for a specific input method")
110 (define-key-after mule-menu-keymap [describe-coding-system]
111 '(menu-item "Describe Coding System..." describe-coding-system)
113 (define-key-after mule-menu-keymap [list-character-sets]
114 '(menu-item "List Character Sets" list-character-sets
115 :help "Show table of available character sets"))
116 (define-key-after mule-menu-keymap [mule-diag]
117 '(menu-item "Show All of Mule Status" mule-diag
118 :help "Display multilingual environment settings")
121 (define-key-after set-coding-system-map [set-buffer-file-coding-system]
122 '(menu-item "For Saving this Buffer" set-buffer-file-coding-system
123 :help "How to encode this buffer on disk")
125 (define-key-after set-coding-system-map [universal-coding-system-argument]
126 '(menu-item "For Next Command" universal-coding-system-argument
127 :help "Coding system to be used by next command")
129 (define-key-after set-coding-system-map [set-terminal-coding-system]
130 '(menu-item "For Terminal" set-terminal-coding-system
131 :enable (null (memq window-system '(x w32 mac)))
132 :help "How to encode terminal output")
134 (define-key-after set-coding-system-map [set-keyboard-coding-system]
135 '(menu-item "For Keyboard" set-keyboard-coding-system
136 :help "How to decode keyboard input")
138 (define-key-after set-coding-system-map [set-buffer-process-coding-system]
139 '(menu-item "For I/O with Subprocess" set-buffer-process-coding-system
140 :visible (fboundp 'start-process)
141 :enable (get-buffer-process (current-buffer))
142 :help "How to en/decode I/O from/to subprocess connected to this buffer")
144 (define-key-after set-coding-system-map [set-selection-coding-system]
145 '(menu-item "For X Selections/Clipboard" set-selection-coding-system
146 :visible (display-selections-p)
147 :help "How to en/decode data to/from selection/clipboard")
149 (define-key-after set-coding-system-map [set-next-selection-coding-system]
150 '(menu-item "For Next X Selection" set-next-selection-coding-system
151 :visible (display-selections-p)
152 :help "How to en/decode next selection/clipboard operation")
154 (define-key setup-language-environment-map
155 [Default] '(menu-item "Default" setup-specified-language-environment))
157 (define-key describe-language-environment-map
158 [Default] '(menu-item "Default" describe-specified-language-support))
160 ;; This should be a single character key binding because users use it
161 ;; very frequently while editing multilingual text. Now we can use
162 ;; only two such keys: "\C-\\" and "\C-^", but the latter is not
163 ;; convenient because it requires shifting on most keyboards. An
164 ;; alternative is "\C-\]" which is now bound to `abort-recursive-edit'
165 ;; but it won't be used that frequently.
166 (define-key global-map "\C-\\" 'toggle-input-method)
168 ;;; This is no good because people often type Shift-SPC
169 ;;; meaning to type SPC. -- rms.
170 ;;; ;; Here's an alternative key binding for X users (Shift-SPACE).
171 ;;; (define-key global-map [?\S- ] 'toggle-input-method)
173 (defun coding-system-change-eol-conversion (coding-system eol-type)
174 "Return a coding system which differs from CODING-SYSTEM in eol conversion.
175 The returned coding system converts end-of-line by EOL-TYPE
176 but text as the same way as CODING-SYSTEM.
177 EOL-TYPE should be `unix', `dos', `mac', or nil.
178 If EOL-TYPE is nil, the returned coding system detects
179 how end-of-line is formatted automatically while decoding.
181 EOL-TYPE can be specified by an integer 0, 1, or 2.
182 They means `unix', `dos', and `mac' respectively."
183 (if (symbolp eol-type)
184 (setq eol-type (cond ((eq eol-type 'unix) 0)
185 ((eq eol-type 'dos) 1)
186 ((eq eol-type 'mac) 2)
188 (let ((orig-eol-type (coding-system-eol-type coding-system)))
189 (if (vectorp orig-eol-type)
192 (aref orig-eol-type eol-type))
193 (let ((base (coding-system-base coding-system)))
196 (if (= eol-type orig-eol-type)
198 (setq orig-eol-type (coding-system-eol-type base))
199 (if (vectorp orig-eol-type)
200 (aref orig-eol-type eol-type))))))))
202 (defun coding-system-change-text-conversion (coding-system coding)
203 "Return a coding system which differs from CODING-SYSTEM in text conversion.
204 The returned coding system converts text by CODING
205 but end-of-line as the same way as CODING-SYSTEM.
206 If CODING is nil, the returned coding system detects
207 how text is formatted automatically while decoding."
209 (coding-system-base coding-system)
210 (let ((eol-type (coding-system-eol-type coding-system)))
211 (coding-system-change-eol-conversion
213 (if (numberp eol-type) (aref [unix dos mac] eol-type))))))
215 (defun toggle-enable-multibyte-characters (&optional arg)
216 "Change whether this buffer uses multibyte characters.
217 With arg, use multibyte characters if the arg is positive.
219 Note that this command does not convert the byte contents of
220 the buffer; it only changes the way those bytes are interpreted.
221 In general, therefore, this command *changes* the sequence of
222 characters that the current buffer contains.
224 We suggest you avoid using use this command unless you know what you
225 are doing. If you use it by mistake, and the buffer is now displayed
226 wrong, use this command again to toggle back to the right mode."
229 (if (null arg) (null enable-multibyte-characters)
230 (> (prefix-numeric-value arg) 0))))
231 (set-buffer-multibyte new-flag))
232 (force-mode-line-update))
234 (defun view-hello-file ()
235 "Display the HELLO file which list up many languages and characters."
237 ;; We have to decode the file in any environment.
238 (let ((default-enable-multibyte-characters t)
239 (coding-system-for-read 'iso-2022-7bit))
240 (find-file-read-only (expand-file-name "HELLO" data-directory))))
242 (defun universal-coding-system-argument ()
243 "Execute an I/O command using the specified coding system."
245 (let* ((default (and buffer-file-coding-system
246 (not (eq (coding-system-type buffer-file-coding-system)
248 buffer-file-coding-system))
249 (coding-system (read-coding-system
251 (format "Coding system for following command (default, %s): " default)
252 "Coding system for following command: ")
254 (keyseq (read-key-sequence
255 (format "Command to execute with %s:" coding-system)))
256 (cmd (key-binding keyseq)))
257 (let ((coding-system-for-read coding-system)
258 (coding-system-for-write coding-system))
260 (call-interactively cmd))))
262 (defun set-default-coding-systems (coding-system)
263 "Set default value of various coding systems to CODING-SYSTEM.
264 This sets the following coding systems:
265 o coding system of a newly created buffer
266 o default coding system for subprocess I/O
267 This also sets the following values:
268 o default value used as file-name-coding-system for converting file names.
269 o default value for the command `set-terminal-coding-system' (not on MSDOS)
270 o default value for the command `set-keyboard-coding-system'."
271 (check-coding-system coding-system)
272 (setq-default buffer-file-coding-system coding-system)
273 (if default-enable-multibyte-characters
274 (setq default-file-name-coding-system coding-system))
275 ;; If coding-system is nil, honor that on MS-DOS as well, so
276 ;; that they could reset the terminal coding system.
277 (unless (and (eq window-system 'pc) coding-system)
278 (setq default-terminal-coding-system coding-system))
279 (setq default-keyboard-coding-system coding-system)
280 (setq default-process-coding-system (cons coding-system coding-system)))
282 (defalias 'update-iso-coding-systems 'update-coding-systems-internal)
283 (make-obsolete 'update-iso-coding-systems 'update-coding-systems-internal "20.3")
285 (defun prefer-coding-system (coding-system)
286 "Add CODING-SYSTEM at the front of the priority list for automatic detection.
287 This also sets the following coding systems:
288 o coding system of a newly created buffer
289 o default coding system for subprocess I/O
290 This also sets the following values:
291 o default value used as file-name-coding-system for converting file names.
292 o default value for the command `set-terminal-coding-system' (not on MSDOS)
293 o default value for the command `set-keyboard-coding-system'
295 If CODING-SYSTEM specifies a certain type of EOL conversion, the coding
296 systems set by this function will use that type of EOL conversion.
298 This command does not change the default value of terminal coding system
299 for MS-DOS terminal, because DOS terminals only support a single coding
300 system, and Emacs automatically sets the default to that coding system at
302 (interactive "zPrefer coding system: ")
303 (if (not (and coding-system (coding-system-p coding-system)))
304 (error "Invalid coding system `%s'" coding-system))
305 (let ((coding-category (coding-system-category coding-system))
306 (base (coding-system-base coding-system))
307 (eol-type (coding-system-eol-type coding-system)))
308 (if (not coding-category)
309 ;; CODING-SYSTEM is no-conversion or undecided.
310 (error "Can't prefer the coding system `%s'" coding-system))
311 (set coding-category (or base coding-system))
312 (update-coding-systems-internal)
313 (or (eq coding-category (car coding-category-list))
314 ;; We must change the order.
315 (set-coding-priority (list coding-category)))
316 (if (and base (interactive-p))
317 (message "Highest priority is set to %s (base of %s)"
319 ;; If they asked for specific EOL conversion, honor that.
320 (if (memq eol-type '(0 1 2))
322 (coding-system-change-eol-conversion base eol-type))
323 (setq coding-system base))
324 (set-default-coding-systems coding-system)))
326 (defvar sort-coding-systems-predicate nil
327 "If non-nil, a predicate function to sort coding systems.
329 It is called with two coding systems, and should return t if the first
330 one is \"less\" than the second.
332 The function `sort-coding-systems' use it.")
334 (defun sort-coding-systems (codings)
335 "Sort coding system list CODINGS by a priority of each coding system.
337 If a coding system is most preferred, it has the highest priority.
338 Otherwise, a coding system corresponds to some MIME charset has higher
339 priorities. Among them, a coding system included in `coding-system'
340 key of the current language environment has higher priorities. See
341 also the documentation of `language-info-alist'.
343 If the variable `sort-coding-systems-predicate' (which see) is
344 non-nil, it is used to sort CODINGS in the different way than above."
345 (if sort-coding-systems-predicate
346 (sort codings sort-coding-systems-predicate)
347 (let* ((most-preferred (symbol-value (car coding-category-list)))
348 (lang-preferred (get-language-info current-language-environment
352 (let ((base (coding-system-base x)))
353 (+ (if (eq base most-preferred) 64 0)
354 (let ((mime (coding-system-get base 'mime-charset)))
356 (if (string-match "^x-" (symbol-name mime))
359 (if (memq base lang-preferred) 8 0)
360 (if (string-match "-with-esc$" (symbol-name base))
362 (if (eq (coding-system-type base) 2)
363 ;; For ISO based coding systems, prefer
364 ;; one that doesn't use escape sequences.
365 (let ((flags (coding-system-flags base)))
366 (if (or (consp (aref flags 0))
367 (consp (aref flags 1))
368 (consp (aref flags 2))
369 (consp (aref flags 3)))
370 (if (or (aref flags 8) (aref flags 9))
375 (sort codings (function (lambda (x y)
376 (> (funcall func x) (funcall func y))))))))
378 (defun find-coding-systems-region (from to)
379 "Return a list of proper coding systems to encode a text between FROM and TO.
380 All coding systems in the list can safely encode any multibyte characters
383 If the text contains no multibyte characters, return a list of a single
384 element `undecided'."
385 (let ((codings (find-coding-systems-region-internal from to)))
387 ;; The text contains only ASCII characters. Any coding
390 ;; We need copy-sequence because sorting will alter the argument.
391 (sort-coding-systems (copy-sequence codings)))))
393 (defun find-coding-systems-string (string)
394 "Return a list of proper coding systems to encode STRING.
395 All coding systems in the list can safely encode any multibyte characters
398 If STRING contains no multibyte characters, return a list of a single
399 element `undecided'."
400 (find-coding-systems-region string nil))
402 (defun find-coding-systems-for-charsets (charsets)
403 "Return a list of proper coding systems to encode characters of CHARSETS.
404 CHARSETS is a list of character sets."
405 (cond ((or (null charsets)
406 (and (= (length charsets) 1)
407 (eq 'ascii (car charsets))))
409 ((or (memq 'eight-bit-control charsets)
410 (memq 'eight-bit-graphic charsets))
411 '(raw-text emacs-mule))
415 (while (and codings charsets)
416 (setq charset (car charsets) charsets (cdr charsets))
417 (unless (eq charset 'ascii)
418 (setq l (aref char-coding-system-table (make-char charset)))
423 (if (memq (car codings) l)
424 (setq ll (cons (car codings) ll)))
425 (setq codings (cdr codings)))
426 (setq codings ll)))))
428 (char-table-extra-slot char-coding-system-table 0))))))
430 (defun find-multibyte-characters (from to &optional maxcount excludes)
431 "Find multibyte characters in the region specified by FROM and TO.
432 If FROM is a string, find multibyte characters in the string.
433 The return value is an alist of the following format:
434 ((CHARSET COUNT CHAR ...) ...)
436 CHARSET is a character set,
437 COUNT is a number of characters,
438 CHARs are found characters of the character set.
439 Optional 3rd arg MAXCOUNT limits how many CHARs are put in the above list.
440 Optional 4th arg EXCLUDE is a list of character sets to be ignored.
442 For invalid characters, CHARs are actually strings."
447 (while (setq idx (string-match "[^\000-\177]" from idx))
448 (setq char (aref from idx)
449 charset (char-charset char))
450 (if (eq charset 'unknown)
451 (setq char (match-string 0)))
452 (if (or (memq charset '(unknown
453 eight-bit-control eight-bit-graphic))
454 (not (or (eq excludes t) (memq charset excludes))))
455 (let ((slot (assq charset chars)))
457 (if (not (memq char (nthcdr 2 slot)))
458 (let ((count (nth 1 slot)))
459 (setcar (cdr slot) (1+ count))
460 (if (or (not maxcount) (< count maxcount))
461 (nconc slot (list char)))))
462 (setq chars (cons (list charset 1 char) chars)))))
463 (setq idx (1+ idx))))
466 (while (re-search-forward "[^\000-\177]" to t)
467 (setq char (preceding-char)
468 charset (char-charset char))
469 (if (eq charset 'unknown)
470 (setq char (match-string 0)))
471 (if (or (memq charset '(unknown eight-bit-control eight-bit-graphic))
472 (not (or (eq excludes t) (memq charset excludes))))
473 (let ((slot (assq charset chars)))
475 (if (not (member char (nthcdr 2 slot)))
476 (let ((count (nth 1 slot)))
477 (setcar (cdr slot) (1+ count))
478 (if (or (not maxcount) (< count maxcount))
479 (nconc slot (list char)))))
480 (setq chars (cons (list charset 1 char) chars))))))))
483 (defvar last-coding-system-specified nil
484 "Most recent coding system explicitly specified by the user when asked.
485 This variable is set whenever Emacs asks the user which coding system
486 to use in order to write a file. If you set it to nil explicitly,
487 then call `write-region', then afterward this variable will be non-nil
488 only if the user was explicitly asked and specified a coding system.")
490 (defvar select-safe-coding-system-accept-default-p nil
491 "If non-nil, a function to control the behaviour of coding system selection.
492 The meaning is the same as the argument ACCEPT-DEFAULT-P of the
493 function `select-safe-coding-system' (which see). This variable
494 overrides that argument.")
496 (defun select-safe-coding-system (from to &optional default-coding-system
498 "Ask a user to select a safe coding system from candidates.
499 The candidates of coding systems which can safely encode a text
500 between FROM and TO are shown in a popup window. Among them, the most
501 proper one is suggested as the default.
503 The list of `buffer-file-coding-system' of the current buffer and the
504 most preferred coding system (if it corresponds to a MIME charset) is
505 treated as the default coding system list. Among them, the first one
506 that safely encodes the text is silently selected and returned without
507 any user interaction. See also the command `prefer-coding-system'.
509 Optional 3rd arg DEFAULT-CODING-SYSTEM specifies a coding system or a
510 list of coding systems to be prepended to the default coding system
513 Optional 4th arg ACCEPT-DEFAULT-P, if non-nil, is a function to
514 determine the acceptability of the silently selected coding system.
515 It is called with that coding system, and should return nil if it
516 should not be silently selected and thus user interaction is required.
518 The variable `select-safe-coding-system-accept-default-p', if
519 non-nil, overrides ACCEPT-DEFAULT-P.
521 Kludgy feature: if FROM is a string, the string is the target text,
523 (if (and default-coding-system
524 (not (listp default-coding-system)))
525 (setq default-coding-system (list default-coding-system)))
527 ;; Change elements of the list to (coding . base-coding).
528 (setq default-coding-system
529 (mapcar (function (lambda (x) (cons x (coding-system-base x))))
530 default-coding-system))
532 ;; If buffer-file-coding-system is not nil nor undecided, append it
534 (if buffer-file-coding-system
535 (let ((base (coding-system-base buffer-file-coding-system)))
536 (or (eq base 'undecided)
537 (assq buffer-file-coding-system default-coding-system)
538 (rassq base default-coding-system)
539 (setq default-coding-system
540 (append default-coding-system
541 (list (cons buffer-file-coding-system base)))))))
543 ;; If the most preferred coding system has the property mime-charset,
544 ;; append it to the defaults.
545 (let* ((preferred (symbol-value (car coding-category-list)))
546 (base (coding-system-base preferred)))
547 (and (coding-system-get preferred 'mime-charset)
548 (not (assq preferred default-coding-system))
549 (not (rassq base default-coding-system))
550 (setq default-coding-system
551 (append default-coding-system (list (cons preferred base))))))
553 (if select-safe-coding-system-accept-default-p
554 (setq accept-default-p select-safe-coding-system-accept-default-p))
556 (let ((codings (find-coding-systems-region from to))
558 (l default-coding-system))
559 (if (eq (car codings) 'undecided)
560 ;; Any coding system is ok.
561 (setq coding-system t)
563 (while (and l (not coding-system))
564 (if (memq (cdr (car l)) codings)
565 (setq coding-system (car (car l)))
567 (if (and coding-system accept-default-p)
568 (or (funcall accept-default-p coding-system)
569 (setq coding-system (list coding-system)))))
571 ;; If all the defaults failed, ask a user.
572 (when (or (not coding-system) (consp coding-system))
573 ;; At first, change each coding system to the corresponding
574 ;; mime-charset name if it is also a coding system. Such a name
575 ;; is more friendly to users.
579 (setq mime-charset (coding-system-get (car l) 'mime-charset))
580 (if (and mime-charset (coding-system-p mime-charset))
581 (setcar l mime-charset))
584 ;; Then ask users to select one form CODINGS.
586 (save-window-excursion
587 (with-output-to-temp-buffer "*Warning*"
589 (set-buffer standard-output)
590 (insert "The following default coding systems were tried:\n")
593 (mapcar (function (lambda (x) (princ " ") (princ (car x))))
594 default-coding-system)
596 (fill-region-as-paragraph pos (point)))
598 (if (consp coding-system)
599 (concat (format "%s safely encodes the target text,\n"
601 "but it is not recommended for encoding text in this context,\n"
602 "e.g., for sending an email message.\n")
603 "However, none of them safely encodes the target text.\n"))
604 (insert (if (consp coding-system)
605 "\nSelect the above, or "
607 "one of the following safe coding systems:\n")
610 (mapcar (function (lambda (x) (princ " ") (princ x)))
613 (fill-region-as-paragraph pos (point)))))
615 ;; Read a coding system.
616 (if (consp coding-system)
617 (setq codings (cons (car coding-system) codings)))
618 (let* ((safe-names (mapcar (lambda (x) (list (symbol-name x)))
620 (name (completing-read
621 (format "Select coding system (default %s): "
623 safe-names nil t nil nil
624 (car (car safe-names)))))
625 (setq last-coding-system-specified (intern name)
626 coding-system last-coding-system-specified)))
627 (kill-buffer "*Warning*")))
629 (if (vectorp (coding-system-eol-type coding-system))
630 (let ((eol (coding-system-eol-type buffer-file-coding-system)))
633 (coding-system-change-eol-conversion coding-system eol)))))
635 (if (eq coding-system t)
636 (setq coding-system buffer-file-coding-system))
639 (setq select-safe-coding-system-function 'select-safe-coding-system)
641 (defun select-message-coding-system ()
642 "Return a coding system to encode the outgoing message of the current buffer.
643 It at first tries the first coding system found in these variables
645 (1) local value of `buffer-file-coding-system'
646 (2) value of `sendmail-coding-system'
647 (3) value of `default-sendmail-coding-system'
648 (4) value of `default-buffer-file-coding-system'
649 If the found coding system can't encode the current buffer,
650 or none of them are bound to a coding system,
651 it asks the user to select a proper coding system."
652 (let ((coding (or (and (local-variable-p 'buffer-file-coding-system)
653 buffer-file-coding-system)
654 sendmail-coding-system
655 default-sendmail-coding-system
656 default-buffer-file-coding-system)))
657 (if (eq coding 'no-conversion)
658 ;; We should never use no-conversion for outgoing mails.
660 (if (fboundp select-safe-coding-system-function)
661 (funcall select-safe-coding-system-function
662 (point-min) (point-max) coding
663 (function (lambda (x) (coding-system-get x 'mime-charset))))
666 ;;; Language support stuff.
668 (defvar language-info-alist nil
669 "Alist of language environment definitions.
670 Each element looks like:
671 (LANGUAGE-NAME . ((KEY . INFO) ...))
672 where LANGUAGE-NAME is a string, the name of the language environment,
673 KEY is a symbol denoting the kind of information, and
674 INFO is the data associated with KEY.
675 Meaningful values for KEY include
677 documentation value is documentation of what this language environment
678 is meant for, and how to use it.
679 charset value is a list of the character sets used by this
680 language environment.
681 sample-text value is one line of text,
682 written using those character sets,
683 appropriate for this language environment.
684 setup-function value is a function to call to switch to this
685 language environment.
686 exit-function value is a function to call to leave this
687 language environment.
688 coding-system value is a list of coding systems that are good
689 for saving text written in this language environment.
690 This list serves as suggestions to the user;
691 in effect, as a kind of documentation.
692 coding-priority value is a list of coding systems for this language
693 environment, in order of decreasing priority.
694 This is used to set up the coding system priority
695 list when you switch to this language environment.
697 value is a translation table to be set in the
698 variable `nonascii-translation-table' in this
699 language environment, or a character set from
700 which `nonascii-insert-offset' is calculated.
701 input-method value is a default input method for this language
703 features value is a list of features requested in this
704 language environment.
706 The following keys take effect only when multibyte characters are
707 globally disabled, i.e. the value of `default-enable-multibyte-characters'
710 unibyte-syntax value is a library name to load to set
711 unibyte 8-bit character syntaxes for this
712 language environment.
714 unibyte-display value is a coding system to encode characters
715 for the terminal. Characters in the range
716 of 160 to 255 display not as octal escapes,
717 but as non-ASCII characters in this language
720 (defun get-language-info (lang-env key)
721 "Return information listed under KEY for language environment LANG-ENV.
722 KEY is a symbol denoting the kind of information.
723 For a list of useful values for KEY and their meanings,
724 see `language-info-alist'."
725 (if (symbolp lang-env)
726 (setq lang-env (symbol-name lang-env)))
727 (let ((lang-slot (assoc-ignore-case lang-env language-info-alist)))
729 (cdr (assq key (cdr lang-slot))))))
731 (defun set-language-info (lang-env key info)
732 "Modify part of the definition of language environment LANG-ENV.
733 Specifically, this stores the information INFO under KEY
734 in the definition of this language environment.
735 KEY is a symbol denoting the kind of information.
736 INFO is the value for that information.
738 For a list of useful values for KEY and their meanings,
739 see `language-info-alist'."
740 (if (symbolp lang-env)
741 (setq lang-env (symbol-name lang-env)))
742 (let (lang-slot key-slot)
743 (setq lang-slot (assoc lang-env language-info-alist))
744 (if (null lang-slot) ; If no slot for the language, add it.
745 (setq lang-slot (list lang-env)
746 language-info-alist (cons lang-slot language-info-alist)))
747 (setq key-slot (assq key lang-slot))
748 (if (null key-slot) ; If no slot for the key, add it.
750 (setq key-slot (list key))
751 (setcdr lang-slot (cons key-slot (cdr lang-slot)))))
752 (setcdr key-slot (purecopy info))))
754 (defun set-language-info-alist (lang-env alist &optional parents)
755 "Store ALIST as the definition of language environment LANG-ENV.
756 ALIST is an alist of KEY and INFO values. See the documentation of
757 `language-info-alist' for the meanings of KEY and INFO.
759 Optional arg PARENTS is a list of parent menu names; it specifies
760 where to put this language environment in the
761 Describe Language Environment and Set Language Environment menus.
762 For example, (\"European\") means to put this language environment
763 in the European submenu in each of those two menus."
764 (if (symbolp lang-env)
765 (setq lang-env (symbol-name lang-env)))
766 (let ((describe-map describe-language-environment-map)
767 (setup-map setup-language-environment-map))
770 map parent-symbol parent prompt)
772 (if (symbolp (setq parent-symbol (car l)))
773 (setq parent (symbol-name parent))
774 (setq parent parent-symbol parent-symbol (intern parent)))
775 (setq map (lookup-key describe-map (vector parent-symbol)))
776 ;; This prompt string is for define-prefix-command, so
777 ;; that the map it creates will be suitable for a menu.
778 (or map (setq prompt (format "%s Environment" parent)))
781 (setq map (intern (format "describe-%s-environment-map"
783 (define-prefix-command map nil prompt)
784 (define-key-after describe-map (vector parent-symbol)
785 (cons parent map) t)))
786 (setq describe-map (symbol-value map))
787 (setq map (lookup-key setup-map (vector parent-symbol)))
790 (setq map (intern (format "setup-%s-environment-map"
792 (define-prefix-command map nil prompt)
793 (define-key-after setup-map (vector parent-symbol)
794 (cons parent map) t)))
795 (setq setup-map (symbol-value map))
798 ;; Set up menu items for this language env.
799 (let ((doc (assq 'documentation alist)))
801 (define-key-after describe-map (vector (intern lang-env))
802 (cons lang-env 'describe-specified-language-support) t)))
803 (define-key-after setup-map (vector (intern lang-env))
804 (cons lang-env 'setup-specified-language-environment) t)
807 (set-language-info lang-env (car (car alist)) (cdr (car alist)))
808 (setq alist (cdr alist)))))
810 (defun read-language-name (key prompt &optional default)
811 "Read a language environment name which has information for KEY.
812 If KEY is nil, read any language environment.
813 Prompt with PROMPT. DEFAULT is the default choice of language environment.
814 This returns a language environment name as a string."
815 (let* ((completion-ignore-case t)
816 (name (completing-read prompt
819 (function (lambda (elm) (assq key elm))))
821 (if (and (> (length name) 0)
823 (get-language-info name key)))
826 ;;; Multilingual input methods.
828 (defconst leim-list-file-name "leim-list.el"
829 "Name of LEIM list file.
830 This file contains a list of libraries of Emacs input methods (LEIM)
831 in the format of Lisp expression for registering each input method.
832 Emacs loads this file at startup time.")
834 (defvar leim-list-header (format
835 ";;; %s -- list of LEIM (Library of Emacs Input Method)
837 ;; This file contains a list of LEIM (Library of Emacs Input Method)
838 ;; in the same directory as this file. Loading this file registers
839 ;; the whole input methods in Emacs.
841 ;; Each entry has the form:
842 ;; (register-input-method
843 ;; INPUT-METHOD LANGUAGE-NAME ACTIVATE-FUNC
846 ;; See the function `register-input-method' for the meanings of arguments.
848 ;; If this directory is included in load-path, Emacs automatically
849 ;; loads this file at startup time.
853 "Header to be inserted in LEIM list file.")
855 (defvar leim-list-entry-regexp "^(register-input-method"
856 "Regexp matching head of each entry in LEIM list file.
857 See also the variable `leim-list-header'")
859 (defvar update-leim-list-functions
860 '(quail-update-leim-list-file)
861 "List of functions to call to update LEIM list file.
862 Each function is called with one arg, LEIM directory name.")
864 (defun update-leim-list-file (&rest dirs)
865 "Update LEIM list file in directories DIRS."
866 (let ((functions update-leim-list-functions))
868 (apply (car functions) dirs)
869 (setq functions (cdr functions)))))
871 (defvar current-input-method nil
872 "The current input method for multilingual text.
873 If nil, that means no input method is activated now.")
874 (make-variable-buffer-local 'current-input-method)
875 (put 'current-input-method 'permanent-local t)
877 (defvar current-input-method-title nil
878 "Title string of the current input method shown in mode line.")
879 (make-variable-buffer-local 'current-input-method-title)
880 (put 'current-input-method-title 'permanent-local t)
882 (defcustom default-input-method nil
883 "*Default input method for multilingual text (a string).
884 This is the input method activated automatically by the command
885 `toggle-input-method' (\\[toggle-input-method])."
887 :type '(choice (const nil) string)
888 :set-after '(current-language-environment))
890 (put 'input-method-function 'permanent-local t)
892 (defvar input-method-history nil
893 "History list for some commands that read input methods.")
894 (make-variable-buffer-local 'input-method-history)
895 (put 'input-method-history 'permanent-local t)
897 (defvar inactivate-current-input-method-function nil
898 "Function to call for inactivating the current input method.
899 Every input method should set this to an appropriate value when activated.
900 This function is called with no argument.
902 This function should never change the value of `current-input-method'.
903 It is set to nil by the function `inactivate-input-method'.")
904 (make-variable-buffer-local 'inactivate-current-input-method-function)
905 (put 'inactivate-current-input-method-function 'permanent-local t)
907 (defvar describe-current-input-method-function nil
908 "Function to call for describing the current input method.
909 This function is called with no argument.")
910 (make-variable-buffer-local 'describe-current-input-method-function)
911 (put 'describe-current-input-method-function 'permanent-local t)
913 (defvar input-method-alist nil
914 "Alist of input method names vs how to use them.
915 Each element has the form:
916 (INPUT-METHOD LANGUAGE-ENV ACTIVATE-FUNC TITLE DESCRIPTION ARGS...)
917 See the function `register-input-method' for the meanings of the elements.")
919 (defun register-input-method (input-method lang-env &rest args)
920 "Register INPUT-METHOD as an input method for language environment ENV.
921 INPUT-METHOD and LANG-ENV are symbols or strings.
923 The remaining arguments are:
924 ACTIVATE-FUNC, TITLE, DESCRIPTION, and ARGS...
925 ACTIVATE-FUNC is a function to call to activate this method.
926 TITLE is a string to show in the mode line when this method is active.
927 DESCRIPTION is a string describing this method and what it is good for.
928 The ARGS, if any, are passed as arguments to ACTIVATE-FUNC.
929 All told, the arguments to ACTIVATE-FUNC are INPUT-METHOD and the ARGS.
931 This function is mainly used in the file \"leim-list.el\" which is
932 created at building time of emacs, registering all quail input methods
933 contained in the emacs distribution.
935 In case you want to register a new quail input method by yourself, be
936 careful to use the same input method title as given in the third
937 parameter of `quail-define-package' (if the values are different, the
938 string specified in this function takes precedence).
940 The commands `describe-input-method' and `list-input-methods' need
941 this duplicated values to show some information about input methods
942 without loading the affected quail packages."
943 (if (symbolp lang-env)
944 (setq lang-env (symbol-name lang-env)))
945 (if (symbolp input-method)
946 (setq input-method (symbol-name input-method)))
947 (let ((info (cons lang-env args))
948 (slot (assoc input-method input-method-alist)))
951 (setq slot (cons input-method info))
952 (setq input-method-alist (cons slot input-method-alist)))))
954 (defun read-input-method-name (prompt &optional default inhibit-null)
955 "Read a name of input method from a minibuffer prompting with PROMPT.
956 If DEFAULT is non-nil, use that as the default,
957 and substitute it into PROMPT at the first `%s'.
958 If INHIBIT-NULL is non-nil, null input signals an error.
960 The return value is a string."
962 (setq prompt (format prompt default)))
963 (let* ((completion-ignore-case t)
964 ;; This binding is necessary because input-method-history is
966 (input-method (completing-read prompt input-method-alist
967 nil t nil 'input-method-history
969 (if (and input-method (symbolp input-method))
970 (setq input-method (symbol-name input-method)))
971 (if (> (length input-method) 0)
974 (error "No valid input method is specified")))))
976 (defun activate-input-method (input-method)
977 "Switch to input method INPUT-METHOD for the current buffer.
978 If some other input method is already active, turn it off first.
979 If INPUT-METHOD is nil, deactivate any current input method."
980 (if (and input-method (symbolp input-method))
981 (setq input-method (symbol-name input-method)))
982 (if (and current-input-method
983 (not (string= current-input-method input-method)))
984 (inactivate-input-method))
985 (unless (or current-input-method (null input-method))
986 (let ((slot (assoc input-method input-method-alist)))
988 (error "Can't activate input method `%s'" input-method))
989 (let ((func (nth 2 slot)))
991 (apply (nth 2 slot) input-method (nthcdr 5 slot))
992 (if (and (consp func) (symbolp (car func)) (symbolp (cdr func)))
995 (apply (car func) input-method (nthcdr 5 slot)))
996 (error "Can't activate input method `%s'" input-method))))
997 (setq current-input-method input-method)
998 (setq current-input-method-title (nth 3 slot))
1000 (run-hooks 'input-method-activate-hook)
1001 (force-mode-line-update)))))
1003 (defun inactivate-input-method ()
1004 "Turn off the current input method."
1005 (when current-input-method
1006 (if input-method-history
1007 (unless (string= current-input-method (car input-method-history))
1008 (setq input-method-history
1009 (cons current-input-method
1010 (delete current-input-method input-method-history))))
1011 (setq input-method-history (list current-input-method)))
1013 (funcall inactivate-current-input-method-function)
1015 (run-hooks 'input-method-inactivate-hook)
1016 (setq current-input-method nil
1017 current-input-method-title nil)
1018 (force-mode-line-update)))))
1020 (defun set-input-method (input-method)
1021 "Select and activate input method INPUT-METHOD for the current buffer.
1022 This also sets the default input method to the one you specify."
1024 (let* ((default (or (car input-method-history) default-input-method)))
1025 (list (read-input-method-name
1026 (if default "Select input method (default %s): " "Select input method: ")
1028 (activate-input-method input-method)
1029 (setq default-input-method input-method))
1031 (defun toggle-input-method (&optional arg)
1032 "Turn on or off a multilingual text input method for the current buffer.
1034 With no prefix argument, if an input method is currently activated,
1035 turn it off. Otherwise, activate an input method -- the one most
1036 recently used, or the one specified in `default-input-method', or
1037 the one read from the minibuffer.
1039 With a prefix argument, read an input method from the minibuffer and
1042 The default is to use the most recent input method specified
1043 \(not including the currently active input method, if any)."
1045 (if (and current-input-method (not arg))
1046 (inactivate-input-method)
1047 (let ((default (or (car input-method-history) default-input-method)))
1048 (if (and arg default (equal current-input-method default)
1049 (> (length input-method-history) 1))
1050 (setq default (nth 1 input-method-history)))
1051 (activate-input-method
1052 (if (or arg (not default))
1054 (read-input-method-name
1055 (if default "Input method (default %s): " "Input method: " )
1058 (or default-input-method
1059 (setq default-input-method current-input-method)))))
1061 (defun describe-input-method (input-method)
1062 "Describe input method INPUT-METHOD."
1064 (list (read-input-method-name
1065 "Describe input method (default, current choice): ")))
1066 (if (and input-method (symbolp input-method))
1067 (setq input-method (symbol-name input-method)))
1068 (if (null input-method)
1069 (describe-current-input-method)
1070 (with-output-to-temp-buffer "*Help*"
1071 (let ((elt (assoc input-method input-method-alist)))
1072 (princ (format "Input method: %s (`%s' in mode line) for %s\n %s\n"
1073 input-method (nth 3 elt) (nth 1 elt) (nth 4 elt)))))))
1075 (defun describe-current-input-method ()
1076 "Describe the input method currently in use."
1077 (if current-input-method
1078 (if (and (symbolp describe-current-input-method-function)
1079 (fboundp describe-current-input-method-function))
1080 (funcall describe-current-input-method-function)
1081 (message "No way to describe the current input method `%s'"
1082 current-input-method)
1084 (error "No input method is activated now")))
1086 (defun read-multilingual-string (prompt &optional initial-input input-method)
1087 "Read a multilingual string from minibuffer, prompting with string PROMPT.
1088 The input method selected last time is activated in minibuffer.
1089 If optional second arg INITIAL-INPUT is non-nil, insert it in the minibuffer
1091 Optional 3rd argument INPUT-METHOD specifies the input method
1092 to be activated instead of the one selected last time. It is a symbol
1096 current-input-method
1097 default-input-method
1098 (read-input-method-name "Input method: " nil t)))
1099 (if (and input-method (symbolp input-method))
1100 (setq input-method (symbol-name input-method)))
1101 (let ((prev-input-method current-input-method))
1104 (activate-input-method input-method)
1105 (read-string prompt initial-input nil nil t))
1106 (activate-input-method prev-input-method))))
1108 ;; Variables to control behavior of input methods. All input methods
1109 ;; should react to these variables.
1111 (defcustom input-method-verbose-flag 'default
1112 "*A flag to control extra guidance given by input methods.
1113 The value should be nil, t, `complex-only', or `default'.
1115 The extra guidance is done by showing list of available keys in echo
1116 area. When you use the input method in the minibuffer, the guidance
1117 is shown at the bottom short window (split from the existing window).
1119 If the value is t, extra guidance is always given, if the value is
1120 nil, extra guidance is always suppressed.
1122 If the value is `complex-only', only complex input methods such as
1123 `chinese-py' and `japanese' give extra guidance.
1125 If the value is `default', complex input methods always give extra
1126 guidance, but simple input methods give it only when you are not in
1129 See also the variable `input-method-highlight-flag'."
1130 :type '(choice (const t) (const nil) (const complex-only) (const default))
1133 (defcustom input-method-highlight-flag t
1134 "*If this flag is non-nil, input methods highlight partially-entered text.
1135 For instance, while you are in the middle of a Quail input method sequence,
1136 the text inserted so far is temporarily underlined.
1137 The underlining goes away when you finish or abort the input method sequence.
1138 See also the variable `input-method-verbose-flag'."
1142 (defvar input-method-activate-hook nil
1143 "Normal hook run just after an input method is activated.
1145 The variable `current-input-method' keeps the input method name
1148 (defvar input-method-inactivate-hook nil
1149 "Normal hook run just after an input method is inactivated.
1151 The variable `current-input-method' still keeps the input method name
1154 (defvar input-method-after-insert-chunk-hook nil
1155 "Normal hook run just after an input method insert some chunk of text.")
1157 (defvar input-method-exit-on-first-char nil
1158 "This flag controls a timing when an input method returns.
1159 Usually, the input method does not return while there's a possibility
1160 that it may find a different translation if a user types another key.
1161 But, it this flag is non-nil, the input method returns as soon as
1162 the current key sequence gets long enough to have some valid translation.")
1164 (defvar input-method-use-echo-area nil
1165 "This flag controls how an input method shows an intermediate key sequence.
1166 Usually, the input method inserts the intermediate key sequence,
1167 or candidate translations corresponding to the sequence,
1168 at point in the current buffer.
1169 But, if this flag is non-nil, it displays them in echo area instead.")
1171 (defvar input-method-exit-on-invalid-key nil
1172 "This flag controls the behaviour of an input method on invalid key input.
1173 Usually, when a user types a key which doesn't start any character
1174 handled by the input method, the key is handled by turning off the
1175 input method temporarily. After that key, the input method is re-enabled.
1176 But, if this flag is non-nil, the input method is never back on.")
1179 (defvar set-language-environment-hook nil
1180 "Normal hook run after some language environment is set.
1182 When you set some hook function here, that effect usually should not
1183 be inherited to another language environment. So, you had better set
1184 another function in `exit-language-environment-hook' (which see) to
1185 cancel the effect.")
1187 (defvar exit-language-environment-hook nil
1188 "Normal hook run after exiting from some language environment.
1189 When this hook is run, the variable `current-language-environment'
1190 is still bound to the language environment being exited.
1192 This hook is mainly used for canceling the effect of
1193 `set-language-environment-hook' (which-see).")
1195 (put 'setup-specified-language-environment 'apropos-inhibit t)
1197 (defun setup-specified-language-environment ()
1198 "Switch to a specified language environment."
1200 (let (language-name)
1201 (if (and (symbolp last-command-event)
1202 (or (not (eq last-command-event 'Default))
1203 (setq last-command-event 'English))
1204 (setq language-name (symbol-name last-command-event)))
1205 (set-language-environment language-name)
1206 (error "Bogus calling sequence"))))
1208 (defcustom current-language-environment "English"
1209 "The last language environment specified with `set-language-environment'.
1210 This variable should be set only with \\[customize], which is equivalent
1211 to using the function `set-language-environment'."
1212 :link '(custom-manual "(emacs)Language Environments")
1213 :set (lambda (symbol value) (set-language-environment value))
1215 (or (car-safe (assoc-ignore-case
1216 (if (symbolp current-language-environment)
1217 (symbol-name current-language-environment)
1218 current-language-environment)
1219 language-info-alist))
1221 :type (cons 'choice (mapcar (lambda (lang)
1222 (list 'const (car lang)))
1223 language-info-alist))
1224 :initialize 'custom-initialize-default
1228 (defun reset-language-environment ()
1229 "Reset multilingual environment of Emacs to the default status.
1231 The default status is as follows:
1233 The default value of buffer-file-coding-system is nil.
1234 The default coding system for process I/O is nil.
1235 The default value for the command `set-terminal-coding-system' is nil.
1236 The default value for the command `set-keyboard-coding-system' is nil.
1238 The order of priorities of coding categories and the coding system
1239 bound to each category are as follows
1240 coding category coding system
1241 --------------------------------------------------
1242 coding-category-iso-8-2 iso-latin-1
1243 coding-category-iso-8-1 iso-latin-1
1244 coding-category-iso-7-tight iso-2022-jp
1245 coding-category-iso-7 iso-2022-7bit
1246 coding-category-iso-7-else iso-2022-7bit-lock
1247 coding-category-iso-8-else iso-2022-8bit-ss2
1248 coding-category-emacs-mule emacs-mule
1249 coding-category-raw-text raw-text
1250 coding-category-sjis japanese-shift-jis
1251 coding-category-big5 chinese-big5
1252 coding-category-ccl nil
1253 coding-category-binary no-conversion
1256 ;; This function formerly set default-enable-multibyte-characters to t,
1257 ;; but that is incorrect. It should not alter the unibyte/multibyte choice.
1259 (setq coding-category-iso-7-tight 'iso-2022-jp
1260 coding-category-iso-7 'iso-2022-7bit
1261 coding-category-iso-8-1 'iso-latin-1
1262 coding-category-iso-8-2 'iso-latin-1
1263 coding-category-iso-7-else 'iso-2022-7bit-lock
1264 coding-category-iso-8-else 'iso-2022-8bit-ss2
1265 coding-category-emacs-mule 'emacs-mule
1266 coding-category-raw-text 'raw-text
1267 coding-category-sjis 'japanese-shift-jis
1268 coding-category-big5 'chinese-big5
1269 coding-category-utf-8 nil
1270 coding-category-utf-16-be nil
1271 coding-category-utf-16-le nil
1272 coding-category-ccl nil
1273 coding-category-binary 'no-conversion)
1275 (set-coding-priority
1276 '(coding-category-iso-8-1
1277 coding-category-iso-8-2
1278 coding-category-iso-7-tight
1279 coding-category-iso-7
1280 coding-category-iso-7-else
1281 coding-category-iso-8-else
1282 coding-category-emacs-mule
1283 coding-category-raw-text
1284 coding-category-sjis
1285 coding-category-big5
1287 coding-category-binary
1288 coding-category-utf-16-be
1289 coding-category-utf-16-le
1290 coding-category-utf-8))
1292 (update-coding-systems-internal)
1294 (set-default-coding-systems nil)
1295 (setq default-sendmail-coding-system 'iso-latin-1)
1297 ;; Don't alter the terminal and keyboard coding systems here.
1298 ;; The terminal still supports the same coding system
1299 ;; that it supported a minute ago.
1300 ;;; (set-terminal-coding-system-internal nil)
1301 ;;; (set-keyboard-coding-system-internal nil)
1303 (setq nonascii-translation-table nil
1304 nonascii-insert-offset 0))
1306 (defun set-display-table-and-terminal-coding-system (language-name)
1307 "Set up the display table and terminal coding system for LANGUAGE-NAME."
1308 (let ((coding (get-language-info language-name 'unibyte-display)))
1310 (standard-display-european-internal)
1311 (standard-display-default (if (eq window-system 'pc) 128 160) 255)
1312 (aset standard-display-table 146 nil))
1313 (or (eq window-system 'pc)
1314 (set-terminal-coding-system coding))))
1316 (defun set-language-environment (language-name)
1317 "Set up multi-lingual environment for using LANGUAGE-NAME.
1318 This sets the coding system priority and the default input method
1319 and sometimes other things. LANGUAGE-NAME should be a string
1320 which is the name of a language environment. For example, \"Latin-1\"
1321 specifies the character set for the major languages of Western Europe."
1322 (interactive (list (read-language-name
1324 "Set language environment (default, English): ")))
1326 (if (symbolp language-name)
1327 (setq language-name (symbol-name language-name)))
1328 (setq language-name "English"))
1329 (or (assoc-ignore-case language-name language-info-alist)
1330 (error "Language environment not defined: %S" language-name))
1331 (if current-language-environment
1332 (let ((func (get-language-info current-language-environment
1334 (run-hooks 'exit-language-environment-hook)
1335 (if (fboundp func) (funcall func))))
1336 (let ((default-eol-type (coding-system-eol-type
1337 default-buffer-file-coding-system)))
1338 (reset-language-environment)
1340 (setq current-language-environment language-name)
1341 (set-language-environment-coding-systems language-name default-eol-type))
1342 (let ((input-method (get-language-info language-name 'input-method)))
1344 (setq default-input-method input-method)
1345 (if input-method-history
1346 (setq input-method-history
1348 (delete input-method input-method-history))))))
1349 (let ((nonascii (get-language-info language-name 'nonascii-translation))
1351 (if (eq window-system 'pc)
1353 (format "cp%d-nonascii-translation-table" dos-codepage)))))
1355 ((char-table-p nonascii)
1356 (setq nonascii-translation-table nonascii))
1357 ((and (eq window-system 'pc) (boundp dos-table))
1358 ;; DOS terminals' default is to use a special non-ASCII translation
1359 ;; table as appropriate for the installed codepage.
1360 (setq nonascii-translation-table (symbol-value dos-table)))
1361 ((charsetp nonascii)
1362 (setq nonascii-insert-offset (- (make-char nonascii) 128)))))
1364 ;; Unibyte setups if necessary.
1365 (unless default-enable-multibyte-characters
1366 ;; Syntax and case table.
1367 (let ((syntax (get-language-info language-name 'unibyte-syntax)))
1369 (let ((set-case-syntax-set-multibyte nil))
1370 (load syntax nil t))
1371 ;; No information for syntax and case. Reset to the defaults.
1372 (let ((syntax-table (standard-syntax-table))
1373 (case-table (standard-case-table))
1374 (ch (if (eq window-system 'pc) 128 160)))
1376 (modify-syntax-entry ch " " syntax-table)
1377 (aset case-table ch ch)
1379 (set-char-table-extra-slot case-table 0 nil)
1380 (set-char-table-extra-slot case-table 1 nil)
1381 (set-char-table-extra-slot case-table 2 nil))
1382 (set-standard-case-table (standard-case-table))
1383 (let ((list (buffer-list)))
1385 (with-current-buffer (car list)
1386 (set-case-table (standard-case-table)))
1387 (setq list (cdr list))))))
1388 (set-display-table-and-terminal-coding-system language-name))
1390 (let ((required-features (get-language-info language-name 'features)))
1391 (while required-features
1392 (require (car required-features))
1393 (setq required-features (cdr required-features))))
1394 (let ((func (get-language-info language-name 'setup-function)))
1397 (run-hooks 'set-language-environment-hook)
1398 (force-mode-line-update t))
1400 (defun standard-display-european-internal ()
1401 ;; Actually set up direct output of non-ASCII characters.
1402 (standard-display-8bit (if (eq window-system 'pc) 128 160) 255)
1403 ;; Unibyte Emacs on MS-DOS wants to display all 8-bit characters with
1404 ;; the native font, and codes 160 and 146 stand for something very
1406 (or (and (eq window-system 'pc) (not default-enable-multibyte-characters))
1408 ;; Make non-line-break space display as a plain space.
1409 ;; Most X fonts do the wrong thing for code 160.
1410 (aset standard-display-table 160 [32])
1411 ;; Most Windows programs send out apostrophe's as \222. Most X fonts
1412 ;; don't contain a character at that position. Map it to the ASCII
1414 (aset standard-display-table 146 [39]))))
1416 (defun set-language-environment-coding-systems (language-name
1418 "Do various coding system setups for language environment LANGUAGE-NAME.
1420 The optional arg EOL-TYPE specifies the eol-type of the default value
1421 of buffer-file-coding-system set by this function."
1422 (let* ((priority (get-language-info language-name 'coding-priority))
1423 (default-coding (car priority)))
1425 (let ((categories (mapcar 'coding-system-category priority)))
1426 (set-default-coding-systems
1427 (if (memq eol-type '(0 1 2 unix dos mac))
1428 (coding-system-change-eol-conversion default-coding eol-type)
1430 (setq default-sendmail-coding-system default-coding)
1431 (set-coding-priority categories)
1433 (set (car categories) (car priority))
1434 (setq priority (cdr priority) categories (cdr categories)))
1435 (update-coding-systems-internal)))))
1437 ;; Print all arguments with `princ', then print "\n".
1438 (defsubst princ-list (&rest args)
1439 (while args (princ (car args)) (setq args (cdr args)))
1442 (put 'describe-specified-language-support 'apropos-inhibit t)
1444 ;; Print a language specific information such as input methods,
1445 ;; charsets, and coding systems. This function is intended to be
1446 ;; called from the menu:
1447 ;; [menu-bar mule describe-language-environment LANGUAGE]
1448 ;; and should not run it by `M-x describe-current-input-method-function'.
1449 (defun describe-specified-language-support ()
1450 "Describe how Emacs supports the specified language environment."
1452 (let (language-name)
1453 (if (not (and (symbolp last-command-event)
1454 (or (not (eq last-command-event 'Default))
1455 (setq last-command-event 'English))
1456 (setq language-name (symbol-name last-command-event))))
1457 (error "Bogus calling sequence"))
1458 (describe-language-environment language-name)))
1460 (defun describe-language-environment (language-name)
1461 "Describe how Emacs supports language environment LANGUAGE-NAME."
1463 (list (read-language-name
1465 "Describe language environment (default, current choice): ")))
1466 (if (null language-name)
1467 (setq language-name current-language-environment))
1468 (if (or (null language-name)
1469 (null (get-language-info language-name 'documentation)))
1470 (error "No documentation for the specified language"))
1471 (if (symbolp language-name)
1472 (setq language-name (symbol-name language-name)))
1473 (let ((doc (get-language-info language-name 'documentation)))
1474 (with-output-to-temp-buffer "*Help*"
1475 (princ-list language-name " language environment" "\n")
1480 (let ((str (get-language-info language-name 'sample-text)))
1483 (princ "Sample text:\n")
1484 (princ-list " " str)
1486 (let ((input-method (get-language-info language-name 'input-method))
1487 (l (copy-sequence input-method-alist)))
1488 (princ "Input methods")
1490 (princ (format " (default, %s)" input-method))
1491 (setq input-method (assoc input-method input-method-alist))
1492 (setq l (cons input-method (delete input-method l))))
1495 (if (string= language-name (nth 1 (car l)))
1496 (princ-list " " (car (car l))
1497 (format " (`%s' in mode line)" (nth 3 (car l)))))
1500 (princ "Character sets:\n")
1501 (let ((l (get-language-info language-name 'charset)))
1503 (princ-list " nothing specific to " language-name)
1505 (princ-list " " (car l) ": "
1506 (charset-description (car l)))
1509 (princ "Coding systems:\n")
1510 (let ((l (get-language-info language-name 'coding-system)))
1512 (princ-list " nothing specific to " language-name)
1514 (princ (format " %s (`%c' in mode line):\n\t%s\n"
1516 (coding-system-mnemonic (car l))
1517 (coding-system-doc-string (car l))))
1518 (let ((aliases (coding-system-get (car l) 'alias-coding-systems)))
1521 (princ (cons 'alias: (cdr aliases)))
1523 (setq l (cdr l))))))))
1527 (defvar locale-translation-file-name nil
1528 "File name for the system's file of locale-name aliases, or nil if none.")
1530 (defvar locale-language-names
1532 ;; UTF-8 is not yet implemented.
1533 ;; Put this first, so that e.g. "ko.UTF-8" does not match "ko" below.
1536 ;; Locale names of the form LANGUAGE[_TERRITORY][.CODESET][@MODIFIER]
1537 ;; as specified in the Single Unix Spec, Version 2.
1538 ;; LANGUAGE is a language code taken from ISO 639:1988 (E/F)
1539 ;; with additions from ISO 639/RA Newsletter No.1/1989;
1540 ;; see Internet RFC 2165 (1997-06).
1541 ;; TERRITORY is a country code taken from ISO 3166.
1542 ;; CODESET and MODIFIER are implementation-dependent.
1546 ("af" . "Latin-1") ; Afrikaans
1547 ("am" . "Ethiopic") ; Amharic
1553 ("be" . "Latin-5") ; Byelorussian
1554 ("bg" . "Latin-5") ; Bulgarian
1557 ; bn Bengali, Bangla
1559 ("br" . "Latin-1") ; Breton
1560 ("ca" . "Latin-1") ; Catalan
1563 ("cy" . "Latin-8") ; Welsh
1564 ("da" . "Latin-1") ; Danish
1568 ;; Users who specify "en" explicitly typically want Latin-1, not ASCII.
1569 ("en" . "Latin-1") ; English
1570 ("eo" . "Latin-3") ; Esperanto
1571 ("es" . "Latin-1") ; Spanish
1572 ("et" . "Latin-4") ; Estonian
1573 ("eu" . "Latin-1") ; Basque
1575 ("fi" . "Latin-1") ; Finnish
1577 ("fo" . "Latin-1") ; Faroese
1578 ("fr" . "Latin-1") ; French
1579 ("fy" . "Latin-1") ; Frisian
1580 ("ga" . "Latin-1") ; Irish Gaelic (new orthography)
1581 ("gd" . "Latin-1") ; Scots Gaelic
1582 ("gl" . "Latin-1") ; Galician
1587 ("hi" . "Devanagari") ; Hindi
1588 ("hr" . "Latin-2") ; Croatian
1589 ("hu" . "Latin-2") ; Hungarian
1592 ("id" . "Latin-1") ; Indonesian
1595 ("is" . "Latin-1") ; Icelandic
1596 ("it" . "Latin-1") ; Italian
1602 ("kl" . "Latin-1") ; Greenlandic
1609 ("la" . "Latin-1") ; Latin
1611 ("lo" . "Lao") ; Laothian
1612 ("lt" . "Latin-4") ; Lithuanian
1613 ("lv" . "Latin-4") ; Latvian, Lettish
1616 ("mk" . "Latin-5") ; Macedonian
1620 ("mr" . "Devanagari") ; Marathi
1622 ("mt" . "Latin-3") ; Maltese
1625 ("ne" . "Devanagari") ; Nepali
1626 ("nl" . "Latin-1") ; Dutch
1627 ("no" . "Latin-1") ; Norwegian
1632 ("pl" . "Latin-2") ; Polish
1634 ("pt" . "Latin-1") ; Portuguese
1636 ("rm" . "Latin-1") ; Rhaeto-Romanic
1639 ("ru.*[_.]koi8" . "Cyrillic-KOI8") ; Russian
1640 ("ru" . "Latin-5") ; Russian
1642 ("sa" . "Devanagari") ; Sanskrit
1645 ("sh" . "Latin-2") ; Serbo-Croatian
1648 ("sl" . "Slovenian")
1652 ("sq" . "Latin-1") ; Albanian
1653 ("sr" . "Latin-2") ; Serbian (Latin alphabet)
1657 ("sv" . "Latin-1") ; Swedish
1658 ("sw" . "Latin-1") ; Swahili
1665 ("tl" . "Latin-1") ; Tagalog
1668 ("tr" . "Latin-5") ; Turkish
1673 ("uk" . "Latin-5") ; Ukrainian
1676 ("vi" . "Vietnamese")
1683 ("zh.*[._]big5" . "Chinese-BIG5")
1684 ("zh.*[._]gbk" . nil) ; Solaris 2.7; has gbk-0 as well as GB 2312.1980-0
1685 ("zh_tw" . "Chinese-CNS")
1686 ("zh" . "Chinese-GB")
1689 ;; ISO standard locales
1691 ("posix$" . "ASCII")
1693 ;; The "IPA" Emacs language environment does not correspond
1694 ;; to any ISO 639 code, so let it stand for itself.
1697 ;; Nonstandard or obsolete language codes
1698 ("cz" . "Czech") ; e.g. Solaris 2.6
1699 ("ee" . "Latin-4") ; Estonian, e.g. X11R6.4
1700 ("iw" . "Hebrew") ; e.g. X11R6.4
1701 ("sp" . "Latin-5") ; Serbian (Cyrillic alphabet), e.g. X11R6.4
1702 ("su" . "Latin-1") ; Finnish, e.g. Solaris 2.6
1704 "List of pairs of locale regexps and language names.
1705 The first element whose locale regexp matches the start of a downcased locale
1706 specifies the language name corresponding to that locale.
1707 If the language name is nil, there is no corresponding language environment.")
1709 (defvar locale-charset-language-names
1710 '((".*8859[-_]?1\\>" . "Latin-1")
1711 (".*8859[-_]?2\\>" . "Latin-2")
1712 (".*8859[-_]?3\\>" . "Latin-3")
1713 (".*8859[-_]?4\\>" . "Latin-4")
1714 (".*8859[-_]?9\\>" . "Latin-5")
1715 (".*8859[-_]?14\\>" . "Latin-8")
1716 (".*8859[-_]?15\\>" . "Latin-9")
1718 "List of pairs of locale regexps and charset language names.
1719 The first element whose locale regexp matches the start of a downcased locale
1720 specifies the language name whose charsets corresponds to that locale.
1721 This language name is used if its charsets disagree with the charsets of
1722 the language name that would otherwise be used for this locale.")
1724 (defvar locale-preferred-coding-systems
1725 '(("ja.*[._]euc" . japanese-iso-8bit)
1726 ("ja.*[._]jis7" . iso-2022-jp)
1727 ("ja.*[._]pck" . japanese-shift-jis)
1728 ("ja.*[._]sjis" . japanese-shift-jis)
1730 "List of pairs of locale regexps and preferred coding systems.
1731 The first element whose locale regexp matches the start of a downcased locale
1732 specifies the coding system to prefer when using that locale.")
1734 (defun locale-name-match (key alist)
1735 "Search for KEY in ALIST, which should be a list of regexp-value pairs.
1736 Return the value corresponding to the first regexp that matches the
1737 start of KEY, or nil if there is no match."
1739 (while (and alist (not element))
1740 (if (string-match (concat "^\\(" (car (car alist)) "\\)") key)
1741 (setq element (car alist)))
1742 (setq alist (cdr alist)))
1745 (defun set-locale-environment (locale-name)
1746 "Set up multi-lingual environment for using LOCALE-NAME.
1747 This sets the coding system priority and the default input method
1748 and sometimes other things. LOCALE-NAME should be a string
1749 which is the name of a locale supported by the system;
1750 often it is of the form xx_XX.CODE, where xx is a language,
1751 XX is a country, and CODE specifies a character set and coding system.
1752 For example, the locale name \"ja_JP.EUC\" might name a locale
1753 for Japanese in Japan using the `japanese-iso-8bit' coding-system.
1755 If LOCALE-NAME is nil, its value is taken from the environment.
1757 The locale names supported by your system can typically be found in a
1758 directory named `/usr/share/locale' or `/usr/lib/locale'."
1760 ;; Do this at runtime for the sake of binaries possibly transported
1761 ;; to a system without X.
1762 (setq locale-translation-file-name
1764 '("/usr/lib/X11/locale/locale.alias" ; e.g. X11R6.4
1765 "/usr/X11R6/lib/X11/locale/locale.alias" ; e.g. RedHat 4.2
1766 "/usr/openwin/lib/locale/locale.alias" ; e.g. Solaris 2.6
1768 ;; The following name appears after the X-related names above,
1769 ;; since the X-related names are what X actually uses.
1770 "/usr/share/locale/locale.alias" ; GNU/Linux sans X
1772 (while (and files (not (file-exists-p (car files))))
1773 (setq files (cdr files)))
1777 ;; Use the first of these three environment variables
1778 ;; that has a nonempty value.
1779 (let ((vars '("LC_ALL" "LC_CTYPE" "LANG")))
1780 (while (and vars (not (setq locale-name (getenv (car vars)))))
1781 (setq vars (cdr vars)))))
1785 ;; Translate "swedish" into "sv_SE.ISO8859-1", and so on,
1786 ;; using the translation file that many systems have.
1787 (when locale-translation-file-name
1789 (insert-file-contents locale-translation-file-name)
1790 (when (re-search-forward
1791 (concat "^" (regexp-quote locale-name) ":?[ \t]+") nil t)
1792 (setq locale-name (buffer-substring (point) (line-end-position))))))
1794 (setq locale-name (downcase locale-name))
1796 (let ((language-name
1797 (locale-name-match locale-name locale-language-names))
1798 (charset-language-name
1799 (locale-name-match locale-name locale-charset-language-names))
1801 (locale-name-match locale-name locale-preferred-coding-systems)))
1803 (if (and charset-language-name
1805 (equal (get-language-info language-name 'charset)
1806 (get-language-info charset-language-name 'charset))))
1807 (setq language-name charset-language-name))
1811 ;; Set up for this character set. This is now the right way
1812 ;; to do it for both unibyte and multibyte modes.
1813 (set-language-environment language-name)
1815 ;; If default-enable-multibyte-characters is nil,
1816 ;; we are using single-byte characters,
1817 ;; so the display table and terminal coding system are irrelevant.
1818 (when default-enable-multibyte-characters
1819 (set-display-table-and-terminal-coding-system language-name))
1821 (setq locale-coding-system
1822 (car (get-language-info language-name 'coding-priority))))
1825 (prefer-coding-system coding-system)
1826 (setq locale-coding-system coding-system)))))
1828 ;;; Charset property
1830 (defun get-charset-property (charset propname)
1831 "Return the value of CHARSET's PROPNAME property.
1832 This is the last value stored with
1833 (put-charset-property CHARSET PROPNAME VALUE)."
1834 (and (not (eq charset 'composition))
1835 (plist-get (charset-plist charset) propname)))
1837 (defun put-charset-property (charset propname value)
1838 "Store CHARSETS's PROPNAME property with value VALUE.
1839 It can be retrieved with `(get-charset-property CHARSET PROPNAME)'."
1840 (or (eq charset 'composition)
1841 (set-charset-plist charset
1842 (plist-put (charset-plist charset) propname value))))
1844 ;;; Character code property
1845 (put 'char-code-property-table 'char-table-extra-slots 0)
1847 (defvar char-code-property-table
1848 (make-char-table 'char-code-property-table)
1849 "Char-table containing a property list of each character code.
1851 See also the documentation of `get-char-code-property' and
1852 `put-char-code-property'.")
1854 (defun get-char-code-property (char propname)
1855 "Return the value of CHAR's PROPNAME property in `char-code-property-table'."
1856 (let ((plist (aref char-code-property-table char)))
1858 (car (cdr (memq propname plist))))))
1860 (defun put-char-code-property (char propname value)
1861 "Store CHAR's PROPNAME property with VALUE in `char-code-property-table'.
1862 It can be retrieved with `(get-char-code-property CHAR PROPNAME)'."
1863 (let ((plist (aref char-code-property-table char)))
1865 (let ((slot (memq propname plist)))
1867 (setcar (cdr slot) value)
1868 (nconc plist (list propname value))))
1869 (aset char-code-property-table char (list propname value)))))
1872 ;; Pretty description of encoded string
1874 ;; Alist of ISO 2022 control code vs the corresponding mnemonic string.
1875 (defvar iso-2022-control-alist
1883 (defun encoded-string-description (str coding-system)
1884 "Return a pretty description of STR that is encoded by CODING-SYSTEM."
1885 (setq str (string-as-unibyte str))
1887 (if (and coding-system (eq (coding-system-type coding-system) 2))
1888 ;; Try to get a pretty description for ISO 2022 escape sequences.
1889 (function (lambda (x) (or (cdr (assq x iso-2022-control-alist))
1890 (format "%02X" x))))
1891 (function (lambda (x) (format "0x%02X" x))))
1894 (defun encode-coding-char (char coding-system)
1895 "Encode CHAR by CODING-SYSTEM and return the resulting string.
1896 If CODING-SYSTEM can't safely encode CHAR, return nil."
1897 (let ((str1 (string-as-multibyte (char-to-string char)))
1898 (str2 (string-as-multibyte (make-string 2 char)))
1899 (safe-charsets (and coding-system
1900 (coding-system-get coding-system 'safe-charsets)))
1901 (charset (char-charset char))
1903 (when (or (eq safe-charsets t)
1905 (memq charset safe-charsets))
1906 ;; We must find the encoded string of CHAR. But, just encoding
1907 ;; CHAR will put extra control sequences (usually to designate
1908 ;; ASCII charaset) at the tail if type of CODING is ISO 2022.
1909 ;; To exclude such tailing bytes, we at first encode one-char
1910 ;; string and two-char string, then check how many bytes at the
1911 ;; tail of both encoded strings are the same.
1913 (setq enc1 (encode-coding-string str1 coding-system)
1915 enc2 (encode-coding-string str2 coding-system)
1917 (while (and (> i1 0) (= (aref enc1 (1- i1)) (aref enc2 (1- i2))))
1918 (setq i1 (1- i1) i2 (1- i2)))
1920 ;; Now (substring enc1 i1) and (substring enc2 i2) are the same,
1921 ;; and they are the extra control sequences at the tail to
1923 (substring enc2 0 i2))))
1926 ;;; mule-cmds.el ends here