1 ;;; mule-cmds.el --- commands for mulitilingual environment -*-coding: iso-2022-7bit -*-
3 ;; Copyright (C) 2000, 2001, 2002, 2003, 2004, 2005 Free Software Foundation, Inc.
4 ;; Copyright (C) 1995, 2003 Electrotechnical Laboratory, JAPAN.
5 ;; Licensed to the Free Software Foundation.
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.
30 (eval-when-compile (defvar dos-codepage))
32 ;;; MULE related key bindings and menus.
34 (defvar mule-keymap (make-sparse-keymap)
35 "Keymap for Mule (Multilingual environment) specific commands.")
37 ;; Keep "C-x C-m ..." for mule specific commands.
38 (define-key ctl-x-map "\C-m" mule-keymap)
40 (define-key mule-keymap "f" 'set-buffer-file-coding-system)
41 (define-key mule-keymap "r" 'revert-buffer-with-coding-system)
42 (define-key mule-keymap "F" 'set-file-name-coding-system)
43 (define-key mule-keymap "t" 'set-terminal-coding-system)
44 (define-key mule-keymap "k" 'set-keyboard-coding-system)
45 (define-key mule-keymap "p" 'set-buffer-process-coding-system)
46 (define-key mule-keymap "x" 'set-selection-coding-system)
47 (define-key mule-keymap "X" 'set-next-selection-coding-system)
48 (define-key mule-keymap "\C-\\" 'set-input-method)
49 (define-key mule-keymap "c" 'universal-coding-system-argument)
50 (define-key mule-keymap "l" 'set-language-environment)
52 (defvar mule-menu-keymap
53 (make-sparse-keymap "Mule (Multilingual Environment)")
54 "Keymap for Mule (Multilingual environment) menu specific commands.")
56 (defvar describe-language-environment-map
57 (make-sparse-keymap "Describe Language Environment"))
59 (defvar setup-language-environment-map
60 (make-sparse-keymap "Set Language Environment"))
62 (defvar set-coding-system-map
63 (make-sparse-keymap "Set Coding System"))
65 (define-key-after mule-menu-keymap [set-language-environment]
66 (list 'menu-item "Set Language Environment" setup-language-environment-map
67 :help "Multilingual environment suitable for a specific language"))
68 (define-key-after mule-menu-keymap [separator-mule]
71 (define-key-after mule-menu-keymap [toggle-input-method]
72 '(menu-item "Toggle Input Method" toggle-input-method)
74 (define-key-after mule-menu-keymap [set-input-method]
75 '(menu-item "Select Input Method..." set-input-method)
77 (define-key-after mule-menu-keymap [describe-input-method]
78 '(menu-item "Describe Input Method" describe-input-method))
79 (define-key-after mule-menu-keymap [separator-input-method]
82 (define-key-after mule-menu-keymap [set-various-coding-system]
83 (list 'menu-item "Set Coding Systems" set-coding-system-map
84 :enable 'default-enable-multibyte-characters))
85 (define-key-after mule-menu-keymap [view-hello-file]
86 '(menu-item "Show Multi-lingual Text" view-hello-file
87 :enable (file-readable-p
88 (expand-file-name "HELLO" data-directory))
89 :help "Display file which says HELLO in many languages")
91 (define-key-after mule-menu-keymap [separator-coding-system]
94 (define-key-after mule-menu-keymap [describe-language-environment]
95 (list 'menu-item "Describe Language Environment"
96 describe-language-environment-map
97 :help "Show multilingual settings for a specific language")
99 (define-key-after mule-menu-keymap [describe-input-method]
100 '(menu-item "Describe Input Method..." describe-input-method
101 :help "Keyboard layout for a specific input method")
103 (define-key-after mule-menu-keymap [describe-coding-system]
104 '(menu-item "Describe Coding System..." describe-coding-system)
106 (define-key-after mule-menu-keymap [list-character-sets]
107 '(menu-item "List Character Sets" list-character-sets
108 :help "Show table of available character sets"))
109 (define-key-after mule-menu-keymap [mule-diag]
110 '(menu-item "Show All of Mule Status" mule-diag
111 :help "Display multilingual environment settings")
114 (define-key-after set-coding-system-map [universal-coding-system-argument]
115 '(menu-item "For Next Command" universal-coding-system-argument
116 :help "Coding system to be used by next command")
118 (define-key-after set-coding-system-map [separator-1]
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 when saved")
125 (define-key-after set-coding-system-map [revert-buffer-with-coding-system]
126 '(menu-item "For Reverting This File Now" revert-buffer-with-coding-system
127 :enable buffer-file-name
128 :help "Revisit this file immediately using specified coding system")
130 (define-key-after set-coding-system-map [set-file-name-coding-system]
131 '(menu-item "For File Name" set-file-name-coding-system
132 :help "How to decode/encode file names")
134 (define-key-after set-coding-system-map [separator-2]
138 (define-key-after set-coding-system-map [set-keyboard-coding-system]
139 '(menu-item "For Keyboard" set-keyboard-coding-system
140 :help "How to decode keyboard input")
142 (define-key-after set-coding-system-map [set-terminal-coding-system]
143 '(menu-item "For Terminal" set-terminal-coding-system
144 :enable (null (memq initial-window-system '(x w32 mac)))
145 :help "How to encode terminal output")
147 (define-key-after set-coding-system-map [separator-3]
150 (define-key-after set-coding-system-map [set-selection-coding-system]
151 '(menu-item "For X Selections/Clipboard" set-selection-coding-system
152 :visible (display-selections-p)
153 :help "How to en/decode data to/from selection/clipboard")
155 (define-key-after set-coding-system-map [set-next-selection-coding-system]
156 '(menu-item "For Next X Selection" set-next-selection-coding-system
157 :visible (display-selections-p)
158 :help "How to en/decode next selection/clipboard operation")
160 (define-key-after set-coding-system-map [set-buffer-process-coding-system]
161 '(menu-item "For I/O with Subprocess" set-buffer-process-coding-system
162 :visible (fboundp 'start-process)
163 :enable (get-buffer-process (current-buffer))
164 :help "How to en/decode I/O from/to subprocess connected to this buffer")
168 (define-key setup-language-environment-map
169 [Default] '(menu-item "Default" setup-specified-language-environment))
171 (define-key describe-language-environment-map
172 [Default] '(menu-item "Default" describe-specified-language-support))
174 ;; This should be a single character key binding because users use it
175 ;; very frequently while editing multilingual text. Now we can use
176 ;; only two such keys: "\C-\\" and "\C-^", but the latter is not
177 ;; convenient because it requires shifting on most keyboards. An
178 ;; alternative is "\C-\]" which is now bound to `abort-recursive-edit'
179 ;; but it won't be used that frequently.
180 (define-key global-map "\C-\\" 'toggle-input-method)
182 ;; This is no good because people often type Shift-SPC
183 ;; meaning to type SPC. -- rms.
184 ;; ;; Here's an alternative key binding for X users (Shift-SPACE).
185 ;; (define-key global-map [?\S- ] 'toggle-input-method)
187 ;;; Mule related hyperlinks.
188 (defconst help-xref-mule-regexp-template
189 (purecopy (concat "\\(\\<\\("
190 "\\(coding system\\)\\|"
191 "\\(input method\\)\\|"
192 "\\(character set\\)\\|"
195 ;; Note starting with word-syntax character:
196 "`\\(\\sw\\(\\sw\\|\\s_\\)+\\)'")))
198 (defun coding-system-change-eol-conversion (coding-system eol-type)
199 "Return a coding system which differs from CODING-SYSTEM in eol conversion.
200 The returned coding system converts end-of-line by EOL-TYPE
201 but text as the same way as CODING-SYSTEM.
202 EOL-TYPE should be `unix', `dos', `mac', or nil.
203 If EOL-TYPE is nil, the returned coding system detects
204 how end-of-line is formatted automatically while decoding.
206 EOL-TYPE can be specified by an integer 0, 1, or 2.
207 They means `unix', `dos', and `mac' respectively."
208 (if (symbolp eol-type)
209 (setq eol-type (cond ((eq eol-type 'unix) 0)
210 ((eq eol-type 'dos) 1)
211 ((eq eol-type 'mac) 2)
213 (let ((orig-eol-type (coding-system-eol-type coding-system)))
214 (if (vectorp orig-eol-type)
217 (aref orig-eol-type eol-type))
218 (let ((base (coding-system-base coding-system)))
221 (if (= eol-type orig-eol-type)
223 (setq orig-eol-type (coding-system-eol-type base))
224 (if (vectorp orig-eol-type)
225 (aref orig-eol-type eol-type))))))))
227 (defun coding-system-change-text-conversion (coding-system coding)
228 "Return a coding system which differs from CODING-SYSTEM in text conversion.
229 The returned coding system converts text by CODING
230 but end-of-line as the same way as CODING-SYSTEM.
231 If CODING is nil, the returned coding system detects
232 how text is formatted automatically while decoding."
233 (let ((eol-type (coding-system-eol-type coding-system)))
234 (coding-system-change-eol-conversion
235 (if coding coding 'undecided)
236 (if (numberp eol-type) (aref [unix dos mac] eol-type)))))
238 (defun toggle-enable-multibyte-characters (&optional arg)
239 "Change whether this buffer uses multibyte characters.
240 With arg, use multibyte characters if the arg is positive.
242 Note that this command does not convert the byte contents of
243 the buffer; it only changes the way those bytes are interpreted.
244 In general, therefore, this command *changes* the sequence of
245 characters that the current buffer contains.
247 We suggest you avoid using use this command unless you know what you
248 are doing. If you use it by mistake, and the buffer is now displayed
249 wrong, use this command again to toggle back to the right mode."
252 (if (null arg) (null enable-multibyte-characters)
253 (> (prefix-numeric-value arg) 0))))
254 (set-buffer-multibyte new-flag))
255 (force-mode-line-update))
257 (defun view-hello-file ()
258 "Display the HELLO file which list up many languages and characters."
260 ;; We have to decode the file in any environment.
261 (let ((default-enable-multibyte-characters t)
262 (coding-system-for-read 'iso-2022-7bit))
263 (view-file (expand-file-name "HELLO" data-directory))))
265 (defun universal-coding-system-argument (coding-system)
266 "Execute an I/O command using the specified coding system."
268 (let ((default (and buffer-file-coding-system
269 (not (eq (coding-system-type buffer-file-coding-system)
271 buffer-file-coding-system)))
272 (list (read-coding-system
274 (format "Coding system for following command (default, %s): " default)
275 "Coding system for following command: ")
277 (let* ((keyseq (read-key-sequence
278 (format "Command to execute with %s:" coding-system)))
279 (cmd (key-binding keyseq))
282 (when (eq cmd 'universal-argument)
283 (call-interactively cmd)
285 ;; Process keys bound in `universal-argument-map'.
287 (setq keyseq (read-key-sequence nil t)
288 cmd (key-binding keyseq t))
289 (not (eq cmd 'universal-argument-other-key)))
290 (let ((current-prefix-arg prefix-arg)
291 ;; Have to bind `last-command-char' here so that
292 ;; `digit-argument', for instance, can compute the
294 (last-command-char (aref keyseq 0)))
295 (call-interactively cmd)))
297 ;; This is the final call to `universal-argument-other-key', which
298 ;; set's the final `prefix-arg.
299 (let ((current-prefix-arg prefix-arg))
300 (call-interactively cmd))
302 ;; Read the command to execute with the given prefix arg.
303 (setq prefix prefix-arg
304 keyseq (read-key-sequence nil t)
305 cmd (key-binding keyseq)))
307 (let ((coding-system-for-read coding-system)
308 (coding-system-for-write coding-system)
309 (coding-system-require-warning t)
310 (current-prefix-arg prefix))
312 (call-interactively cmd))))
314 (defun set-default-coding-systems (coding-system)
315 "Set default value of various coding systems to CODING-SYSTEM.
316 This sets the following coding systems:
317 o coding system of a newly created buffer
318 o default coding system for subprocess I/O
319 This also sets the following values:
320 o default value used as `file-name-coding-system' for converting file names.
321 o default value for the command `set-terminal-coding-system' (not on MSDOS)
322 o default value for the command `set-keyboard-coding-system'."
323 (check-coding-system coding-system)
324 (setq-default buffer-file-coding-system coding-system)
325 (if (fboundp 'ucs-set-table-for-input)
326 (dolist (buffer (buffer-list))
327 (or (local-variable-p 'buffer-file-coding-system buffer)
328 (ucs-set-table-for-input buffer))))
330 (if (and default-enable-multibyte-characters (not (eq system-type 'darwin)))
331 ;; The file-name coding system on Darwin systems is always utf-8.
332 (setq default-file-name-coding-system coding-system))
333 ;; If coding-system is nil, honor that on MS-DOS as well, so
334 ;; that they could reset the terminal coding system.
335 (unless (and (eq window-system 'pc) coding-system)
336 (setq default-terminal-coding-system coding-system))
337 (setq default-keyboard-coding-system coding-system)
338 ;; Preserve eol-type from existing default-process-coding-systems.
339 ;; On non-unix-like systems in particular, these may have been set
340 ;; carefully by the user, or by the startup code, to deal with the
341 ;; users shell appropriately, so should not be altered by changing
342 ;; language environment.
344 (coding-system-change-text-conversion
345 (car default-process-coding-system) coding-system))
347 (coding-system-change-text-conversion
348 (cdr default-process-coding-system) coding-system)))
349 (setq default-process-coding-system
350 (cons output-coding input-coding))))
352 (defalias 'update-iso-coding-systems 'update-coding-systems-internal)
353 (make-obsolete 'update-iso-coding-systems 'update-coding-systems-internal "20.3")
355 (defun prefer-coding-system (coding-system)
356 "Add CODING-SYSTEM at the front of the priority list for automatic detection.
357 This also sets the following coding systems:
358 o coding system of a newly created buffer
359 o default coding system for subprocess I/O
360 This also sets the following values:
361 o default value used as `file-name-coding-system' for converting file names.
362 o default value for the command `set-terminal-coding-system' (not on MSDOS)
363 o default value for the command `set-keyboard-coding-system'
365 If CODING-SYSTEM specifies a certain type of EOL conversion, the coding
366 systems set by this function will use that type of EOL conversion.
368 This command does not change the default value of terminal coding system
369 for MS-DOS terminal, because DOS terminals only support a single coding
370 system, and Emacs automatically sets the default to that coding system at
373 A coding system that requires automatic detection of text
374 encoding (e.g. undecided, unix) can't be preferred.
376 See also `coding-category-list' and `coding-system-category'."
377 (interactive "zPrefer coding system: ")
378 (if (not (and coding-system (coding-system-p coding-system)))
379 (error "Invalid coding system `%s'" coding-system))
380 (let ((coding-category (coding-system-category coding-system))
381 (base (coding-system-base coding-system))
382 (eol-type (coding-system-eol-type coding-system)))
383 (if (not coding-category)
384 ;; CODING-SYSTEM is no-conversion or undecided.
385 (error "Can't prefer the coding system `%s'" coding-system))
386 (set coding-category (or base coding-system))
387 ;; Changing the binding of a coding category requires this call.
388 (update-coding-systems-internal)
389 (or (eq coding-category (car coding-category-list))
390 ;; We must change the order.
391 (set-coding-priority (list coding-category)))
392 (if (and base (interactive-p))
393 (message "Highest priority is set to %s (base of %s)"
395 ;; If they asked for specific EOL conversion, honor that.
396 (if (memq eol-type '(0 1 2))
398 (coding-system-change-eol-conversion base eol-type))
399 (setq coding-system base))
400 (set-default-coding-systems coding-system)))
402 (defvar sort-coding-systems-predicate nil
403 "If non-nil, a predicate function to sort coding systems.
405 It is called with two coding systems, and should return t if the first
406 one is \"less\" than the second.
408 The function `sort-coding-systems' use it.")
410 (defun sort-coding-systems (codings)
411 "Sort coding system list CODINGS by a priority of each coding system.
413 If a coding system is most preferred, it has the highest priority.
414 Otherwise, a coding system corresponds to some MIME charset has higher
415 priorities. Among them, a coding system included in `coding-system'
416 key of the current language environment has higher priorities. See
417 also the documentation of `language-info-alist'.
419 If the variable `sort-coding-systems-predicate' (which see) is
420 non-nil, it is used to sort CODINGS in the different way than above."
421 (if sort-coding-systems-predicate
422 (sort codings sort-coding-systems-predicate)
423 (let* ((from-categories (mapcar #'(lambda (x) (symbol-value x))
424 coding-category-list))
425 (most-preferred (car from-categories))
426 (lang-preferred (get-language-info current-language-environment
430 (let ((base (coding-system-base x)))
431 ;; We calculate the priority number 0..255 by
432 ;; using the 8 bits PMMLCEII as this:
433 ;; P: 1 iff most preferred.
434 ;; MM: greater than 0 iff mime-charset.
435 ;; L: 1 iff one of the current lang. env.'s codings.
436 ;; C: 1 iff one of codings listed in the category list.
437 ;; E: 1 iff not XXX-with-esc
438 ;; II: if iso-2022 based, 0..3, else 1.
440 (lsh (if (eq base most-preferred) 1 0) 7)
442 (let ((mime (coding-system-get base 'mime-charset)))
443 ;; Prefer coding systems corresponding to a
446 ;; Lower utf-16 priority so that we
447 ;; normally prefer utf-8 to it, and put
448 ;; x-ctext below that.
449 (cond ((string-match "utf-16"
452 ((string-match "^x-" (symbol-name mime))
457 (lsh (if (memq base lang-preferred) 1 0) 4)
458 (lsh (if (memq base from-categories) 1 0) 3)
459 (lsh (if (string-match "-with-esc\\'"
462 (if (eq (coding-system-type base) 2)
463 ;; For ISO based coding systems, prefer
464 ;; one that doesn't use escape sequences.
465 (let ((flags (coding-system-flags base)))
466 (if (or (consp (aref flags 0))
467 (consp (aref flags 1))
468 (consp (aref flags 2))
469 (consp (aref flags 3)))
470 (if (or (aref flags 8) (aref flags 9))
475 (sort codings (function (lambda (x y)
476 (> (funcall func x) (funcall func y))))))))
478 (defun find-coding-systems-region (from to)
479 "Return a list of proper coding systems to encode a text between FROM and TO.
480 If FROM is a string, find coding systems in that instead of the buffer.
481 All coding systems in the list can safely encode any multibyte characters
484 If the text contains no multibyte characters, return a list of a single
485 element `undecided'."
486 (let ((codings (find-coding-systems-region-internal from to)))
488 ;; The text contains only ASCII characters. Any coding
491 ;; We need copy-sequence because sorting will alter the argument.
492 (sort-coding-systems (copy-sequence codings)))))
494 (defun find-coding-systems-string (string)
495 "Return a list of proper coding systems to encode STRING.
496 All coding systems in the list can safely encode any multibyte characters
499 If STRING contains no multibyte characters, return a list of a single
500 element `undecided'."
501 (find-coding-systems-region string nil))
503 (defun find-coding-systems-for-charsets (charsets)
504 "Return a list of proper coding systems to encode characters of CHARSETS.
505 CHARSETS is a list of character sets.
506 It actually checks at most the first 96 characters of each charset.
507 So, if a charset of dimension two is included in CHARSETS, the value may
508 contain a coding system that can't encode all characters of the charset."
509 (cond ((or (null charsets)
510 (and (= (length charsets) 1)
511 (eq 'ascii (car charsets))))
513 ((or (memq 'eight-bit-control charsets)
514 (memq 'eight-bit-graphic charsets))
515 '(raw-text emacs-mule))
519 (while (and codings charsets)
520 (setq charset (car charsets) charsets (cdr charsets))
521 (unless (eq charset 'ascii)
522 (setq str (make-string 96 32))
523 (if (= (charset-dimension charset) 1)
524 (if (= (charset-chars charset) 96)
526 (aset str i (make-char charset (+ i 32))))
528 (aset str i (make-char charset (+ i 33)))))
529 (if (= (charset-chars charset) 96)
531 (aset str i (make-char charset 32 (+ i 32))))
533 (aset str i (make-char charset 33 (+ i 33))))))
534 (setq l (find-coding-systems-string str))
538 (dolist (elt codings)
540 (setq ll (cons elt ll))))
541 (setq codings ll)))))
544 (defun find-multibyte-characters (from to &optional maxcount excludes)
545 "Find multibyte characters in the region specified by FROM and TO.
546 If FROM is a string, find multibyte characters in the string.
547 The return value is an alist of the following format:
548 ((CHARSET COUNT CHAR ...) ...)
550 CHARSET is a character set,
551 COUNT is a number of characters,
552 CHARs are the characters found from the character set.
553 Optional 3rd arg MAXCOUNT limits how many CHARs are put in the above list.
554 Optional 4th arg EXCLUDE is a list of character sets to be ignored.
556 For invalid characters, CHARs are actually strings."
561 (while (setq idx (string-match "[^\000-\177]" from idx))
562 (setq char (aref from idx)
563 charset (char-charset char))
564 (if (eq charset 'unknown)
565 (setq char (match-string 0)))
566 (if (or (memq charset '(unknown
567 eight-bit-control eight-bit-graphic))
568 (not (or (eq excludes t) (memq charset excludes))))
569 (let ((slot (assq charset chars)))
571 (if (not (memq char (nthcdr 2 slot)))
572 (let ((count (nth 1 slot)))
573 (setcar (cdr slot) (1+ count))
574 (if (or (not maxcount) (< count maxcount))
575 (nconc slot (list char)))))
576 (setq chars (cons (list charset 1 char) chars)))))
577 (setq idx (1+ idx))))
580 (while (re-search-forward "[^\000-\177]" to t)
581 (setq char (preceding-char)
582 charset (char-charset char))
583 (if (eq charset 'unknown)
584 (setq char (match-string 0)))
585 (if (or (memq charset '(unknown eight-bit-control eight-bit-graphic))
586 (not (or (eq excludes t) (memq charset excludes))))
587 (let ((slot (assq charset chars)))
589 (if (not (member char (nthcdr 2 slot)))
590 (let ((count (nth 1 slot)))
591 (setcar (cdr slot) (1+ count))
592 (if (or (not maxcount) (< count maxcount))
593 (nconc slot (list char)))))
594 (setq chars (cons (list charset 1 char) chars))))))))
598 (defun search-unencodable-char (coding-system)
599 "Search forward from point for a character that is not encodable.
600 It asks which coding system to check.
601 If such a character is found, set point after that character.
602 Otherwise, don't move point.
604 When called from a program, the value is a position of the found character,
605 or nil if all characters are encodable."
607 (list (let ((default (or buffer-file-coding-system 'us-ascii)))
609 (format "Coding-system (default, %s): " default)
611 (let ((pos (unencodable-char-position (point) (point-max) coding-system)))
614 (message "All following characters are encodable by %s" coding-system))
618 (defvar last-coding-system-specified nil
619 "Most recent coding system explicitly specified by the user when asked.
620 This variable is set whenever Emacs asks the user which coding system
621 to use in order to write a file. If you set it to nil explicitly,
622 then call `write-region', then afterward this variable will be non-nil
623 only if the user was explicitly asked and specified a coding system.")
625 (defvar select-safe-coding-system-accept-default-p nil
626 "If non-nil, a function to control the behaviour of coding system selection.
627 The meaning is the same as the argument ACCEPT-DEFAULT-P of the
628 function `select-safe-coding-system' (which see). This variable
629 overrides that argument.")
631 (defun select-safe-coding-system-interactively (from to codings unsafe
632 &optional rejected default)
633 "Select interactively a coding system for the region FROM ... TO.
634 FROM can be a string, as in `write-region'.
635 CODINGS is the list of base coding systems known to be safe for this region,
636 typically obtained with `find-coding-systems-region'.
637 UNSAFE is a list of coding systems known to be unsafe for this region.
638 REJECTED is a list of coding systems which were safe but for some reason
639 were not recommended in the particular context.
640 DEFAULT is the coding system to use by default in the query."
641 ;; At first, if some defaults are unsafe, record at most 11
642 ;; problematic characters and their positions for them by turning
645 ;; ((CODING (POS . CHAR) (POS . CHAR) ...) ...)
648 (mapcar #'(lambda (coding)
651 (mapcar #'(lambda (pos)
652 (cons pos (aref from pos)))
653 (unencodable-char-position
654 0 (length from) coding
656 (mapcar #'(lambda (pos)
657 (cons pos (char-after pos)))
658 (unencodable-char-position
659 from to coding 11)))))
662 ;; Change each safe coding system to the corresponding
663 ;; mime-charset name if it is also a coding system. Such a name
664 ;; is more friendly to users.
668 (setq mime-charset (coding-system-get (car l) 'mime-charset))
669 (if (and mime-charset (coding-system-p mime-charset))
670 (setcar l mime-charset))
673 ;; Don't offer variations with locking shift, which you
674 ;; basically never want.
676 (dolist (elt codings (setq codings (nreverse l)))
677 (unless (or (eq 'coding-category-iso-7-else
678 (coding-system-category elt))
679 (eq 'coding-category-iso-8-else
680 (coding-system-category elt)))
683 ;; Remove raw-text, emacs-mule and no-conversion unless nothing
684 ;; else is available.
688 (delq 'no-conversion codings)))
689 '(raw-text emacs-mule no-conversion)))
691 (let ((window-configuration (current-window-configuration))
692 (bufname (buffer-name))
695 ;; If some defaults are unsafe, make sure the offending
696 ;; buffer is displayed.
697 (when (and unsafe (not (stringp from)))
698 (pop-to-buffer bufname)
699 (goto-char (apply 'min (mapcar #'(lambda (x) (car (cadr x)))
701 ;; Then ask users to select one from CODINGS while showing
702 ;; the reason why none of the defaults are not used.
703 (with-output-to-temp-buffer "*Warning*"
704 (with-current-buffer standard-output
705 (if (and (null rejected) (null unsafe))
706 (insert "No default coding systems to try for "
708 (format "string \"%s\"." from)
709 (format "buffer `%s'." bufname)))
711 "These default coding systems were tried to encode"
713 (concat " \"" (if (> (length from) 10)
714 (concat (substring from 0 10) "...\"")
716 (format " text\nin the buffer `%s'" bufname))
720 (dolist (x (append rejected unsafe))
721 (princ " ") (princ (car x)))
723 (fill-region-as-paragraph pos (point)))
725 (insert "These safely encodes the target text,
726 but it is not recommended for encoding text in this context,
727 e.g., for sending an email message.\n ")
729 (princ " ") (princ x))
732 (insert (if rejected "And the others"
733 "However, each of them")
734 " encountered these problematic characters:\n")
735 (dolist (coding unsafe)
736 (insert (format " %s:" (car coding)))
739 #'(lambda (bufname pos)
740 (when (buffer-live-p (get-buffer bufname))
741 (pop-to-buffer bufname)
744 #'(lambda (bufname pos coding)
745 (when (buffer-live-p (get-buffer bufname))
746 (pop-to-buffer bufname)
750 (search-unencodable-char coding)
751 (forward-char -1))))))
752 (dolist (elt (cdr coding))
755 (insert (if (< i 10) (cdr elt) "..."))
761 "mouse-2, RET: jump to this character"
763 'help-args (list bufname (car elt)))
768 "mouse-2, RET: next unencodable character"
770 'help-args (list bufname (car elt)
775 The first problematic character is at point in the displayed buffer,\n"
776 (substitute-command-keys "\
777 and \\[universal-argument] \\[what-cursor-position] will give information about it.\n"))))
779 one of the following safe coding systems, or edit the buffer:\n")
783 (princ " ") (princ x))
785 (fill-region-as-paragraph pos (point)))
786 (insert "Or specify any other coding system
787 at the risk of losing the problematic characters.\n")))
789 ;; Read a coding system.
792 (format "Select coding system (default %s): " default)
794 (setq last-coding-system-specified coding-system))
796 (kill-buffer "*Warning*")
797 (set-window-configuration window-configuration)
800 (defun select-safe-coding-system (from to &optional default-coding-system
801 accept-default-p file)
802 "Ask a user to select a safe coding system from candidates.
803 The candidates of coding systems which can safely encode a text
804 between FROM and TO are shown in a popup window. Among them, the most
805 proper one is suggested as the default.
807 The list of `buffer-file-coding-system' of the current buffer,
808 the `default-buffer-file-coding-system', and the
809 most preferred coding system (if it corresponds to a MIME charset) is
810 treated as the default coding system list. Among them, the first one
811 that safely encodes the text is normally selected silently and
812 returned without any user interaction. See also the command
813 `prefer-coding-system'.
815 However, the user is queried if the chosen coding system is
816 inconsistent with what would be selected by `set-auto-coding' from
817 coding cookies &c. if the contents of the region were read from a
818 file. (That could lead to data corruption in a file subsequently
819 re-visited and edited.)
821 Optional 3rd arg DEFAULT-CODING-SYSTEM specifies a coding system or a
822 list of coding systems to be prepended to the default coding system
823 list. However, if DEFAULT-CODING-SYSTEM is a list and the first
824 element is t, the cdr part is used as the defualt coding system list,
825 i.e. `buffer-file-coding-system', `default-buffer-file-coding-system',
826 and the most preferred coding system are not used.
828 Optional 4th arg ACCEPT-DEFAULT-P, if non-nil, is a function to
829 determine the acceptability of the silently selected coding system.
830 It is called with that coding system, and should return nil if it
831 should not be silently selected and thus user interaction is required.
833 Optional 5th arg FILE is the file name to use for this purpose.
834 That is different from `buffer-file-name' when handling `write-region'
837 The variable `select-safe-coding-system-accept-default-p', if
838 non-nil, overrides ACCEPT-DEFAULT-P.
840 Kludgy feature: if FROM is a string, the string is the target text,
842 (if (and default-coding-system
843 (not (listp default-coding-system)))
844 (setq default-coding-system (list default-coding-system)))
846 (let ((no-other-defaults nil))
847 (if (eq (car default-coding-system) t)
848 (setq no-other-defaults t
849 default-coding-system (cdr default-coding-system)))
851 ;; Change elements of the list to (coding . base-coding).
852 (setq default-coding-system
853 (mapcar (function (lambda (x) (cons x (coding-system-base x))))
854 default-coding-system))
856 ;; From now on, the list of defaults is reversed.
857 (setq default-coding-system (nreverse default-coding-system))
859 (unless no-other-defaults
860 ;; If buffer-file-coding-system is not nil nor undecided, append it
862 (if buffer-file-coding-system
863 (let ((base (coding-system-base buffer-file-coding-system)))
864 (or (eq base 'undecided)
865 (rassq base default-coding-system)
866 (push (cons buffer-file-coding-system base)
867 default-coding-system))))
869 ;; If default-buffer-file-coding-system is not nil nor undecided,
870 ;; append it to the defaults.
871 (if default-buffer-file-coding-system
872 (let ((base (coding-system-base default-buffer-file-coding-system)))
873 (or (eq base 'undecided)
874 (rassq base default-coding-system)
875 (push (cons default-buffer-file-coding-system base)
876 default-coding-system))))
878 ;; If the most preferred coding system has the property mime-charset,
879 ;; append it to the defaults.
880 (let ((tail coding-category-list)
882 (while (and tail (not (setq preferred (symbol-value (car tail)))))
883 (setq tail (cdr tail)))
884 (and (coding-system-p preferred)
885 (setq base (coding-system-base preferred))
886 (coding-system-get preferred 'mime-charset)
887 (not (rassq base default-coding-system))
888 (push (cons preferred base)
889 default-coding-system)))))
891 (if select-safe-coding-system-accept-default-p
892 (setq accept-default-p select-safe-coding-system-accept-default-p))
894 (let ((codings (find-coding-systems-region from to))
896 safe rejected unsafe)
897 (if (eq (car codings) 'undecided)
898 ;; Any coding system is ok.
899 (setq coding-system t)
900 ;; Classify the defaults into safe, rejected, and unsafe.
901 (dolist (elt default-coding-system)
902 (if (memq (cdr elt) codings)
903 (if (and (functionp accept-default-p)
904 (not (funcall accept-default-p (cdr elt))))
905 (push (car elt) rejected)
906 (push (car elt) safe))
907 (push (car elt) unsafe)))
909 (setq coding-system (car safe))))
911 ;; If all the defaults failed, ask a user.
912 (when (not coding-system)
913 (setq coding-system (select-safe-coding-system-interactively
914 from to codings unsafe rejected (car codings))))
916 (if (vectorp (coding-system-eol-type coding-system))
917 (let ((eol (coding-system-eol-type buffer-file-coding-system)))
920 (coding-system-change-eol-conversion coding-system eol)))))
922 (if (eq coding-system t)
923 (setq coding-system buffer-file-coding-system))
924 ;; Check we're not inconsistent with what `coding:' spec &c would
925 ;; give when file is re-read.
926 ;; But don't do this if we explicitly ignored the cookie
927 ;; by using `find-file-literally'.
928 (unless (or (stringp from) find-file-literally)
929 (let ((auto-cs (save-excursion
932 (narrow-to-region from to)
933 (goto-char (point-min))
934 (set-auto-coding (or file buffer-file-name "")
936 ;; Merge coding-system and auto-cs as far as possible.
937 (if (not coding-system)
938 (setq coding-system auto-cs)
940 (setq auto-cs coding-system)
941 (let ((eol-type-1 (coding-system-eol-type coding-system))
942 (eol-type-2 (coding-system-eol-type auto-cs)))
943 (if (eq (coding-system-base coding-system) 'undecided)
944 (setq coding-system (coding-system-change-text-conversion
945 coding-system auto-cs))
946 (if (eq (coding-system-base auto-cs) 'undecided)
947 (setq auto-cs (coding-system-change-text-conversion
948 auto-cs coding-system))))
949 (if (vectorp eol-type-1)
950 (or (vectorp eol-type-2)
951 (setq coding-system (coding-system-change-eol-conversion
952 coding-system eol-type-2)))
953 (if (vectorp eol-type-2)
954 (setq auto-cs (coding-system-change-eol-conversion
955 auto-cs eol-type-1)))))))
958 ;; Don't barf if writing a compressed file, say.
959 ;; This check perhaps isn't ideal, but is probably
960 ;; the best thing to do.
961 (not (auto-coding-alist-lookup (or file buffer-file-name "")))
962 (not (coding-system-equal coding-system auto-cs)))
964 (format "Selected encoding %s disagrees with \
965 %s specified by file contents. Really save (else edit coding cookies \
966 and try again)? " coding-system auto-cs))
967 (error "Save aborted")))))
970 (setq select-safe-coding-system-function 'select-safe-coding-system)
972 (defun select-message-coding-system ()
973 "Return a coding system to encode the outgoing message of the current buffer.
974 It at first tries the first coding system found in these variables
976 (1) local value of `buffer-file-coding-system'
977 (2) value of `sendmail-coding-system'
978 (3) value of `default-sendmail-coding-system'
979 (4) value of `default-buffer-file-coding-system'
980 If the found coding system can't encode the current buffer,
981 or none of them are bound to a coding system,
982 it asks the user to select a proper coding system."
983 (let ((coding (or (and (local-variable-p 'buffer-file-coding-system)
984 buffer-file-coding-system)
985 sendmail-coding-system
986 default-sendmail-coding-system
987 default-buffer-file-coding-system)))
988 (if (eq coding 'no-conversion)
989 ;; We should never use no-conversion for outgoing mail.
991 (if (fboundp select-safe-coding-system-function)
992 (funcall select-safe-coding-system-function
993 (point-min) (point-max) coding
994 (function (lambda (x) (coding-system-get x 'mime-charset))))
997 ;;; Language support stuff.
999 (defvar language-info-alist nil
1000 "Alist of language environment definitions.
1001 Each element looks like:
1002 (LANGUAGE-NAME . ((KEY . INFO) ...))
1003 where LANGUAGE-NAME is a string, the name of the language environment,
1004 KEY is a symbol denoting the kind of information, and
1005 INFO is the data associated with KEY.
1006 Meaningful values for KEY include
1008 documentation value is documentation of what this language environment
1009 is meant for, and how to use it.
1010 charset value is a list of the character sets used by this
1011 language environment.
1012 sample-text value is an expression which is evalled to generate
1013 a line of text written using characters appropriate
1014 for this language environment.
1015 setup-function value is a function to call to switch to this
1016 language environment.
1017 exit-function value is a function to call to leave this
1018 language environment.
1019 coding-system value is a list of coding systems that are good
1020 for saving text written in this language environment.
1021 This list serves as suggestions to the user;
1022 in effect, as a kind of documentation.
1023 coding-priority value is a list of coding systems for this language
1024 environment, in order of decreasing priority.
1025 This is used to set up the coding system priority
1026 list when you switch to this language environment.
1027 nonascii-translation
1028 value is a translation table to be set in the
1029 variable `nonascii-translation-table' in this
1030 language environment, or a character set from
1031 which `nonascii-insert-offset' is calculated.
1032 input-method value is a default input method for this language
1034 features value is a list of features requested in this
1035 language environment.
1036 ctext-non-standard-encodings
1037 value is a list of non-standard encoding
1038 names used in extended segments of CTEXT.
1040 `ctext-non-standard-encodings' for more
1043 The following keys take effect only when multibyte characters are
1044 globally disabled, i.e. the value of `default-enable-multibyte-characters'
1047 unibyte-syntax value is a library name to load to set
1048 unibyte 8-bit character syntaxes for this
1049 language environment.
1051 unibyte-display value is a coding system to encode characters
1052 for the terminal. Characters in the range
1053 of 160 to 255 display not as octal escapes,
1054 but as non-ASCII characters in this language
1057 (defun get-language-info (lang-env key)
1058 "Return information listed under KEY for language environment LANG-ENV.
1059 KEY is a symbol denoting the kind of information.
1060 For a list of useful values for KEY and their meanings,
1061 see `language-info-alist'."
1062 (if (symbolp lang-env)
1063 (setq lang-env (symbol-name lang-env)))
1064 (let ((lang-slot (assoc-string lang-env language-info-alist t)))
1066 (cdr (assq key (cdr lang-slot))))))
1068 (defun set-language-info (lang-env key info)
1069 "Modify part of the definition of language environment LANG-ENV.
1070 Specifically, this stores the information INFO under KEY
1071 in the definition of this language environment.
1072 KEY is a symbol denoting the kind of information.
1073 INFO is the value for that information.
1075 For a list of useful values for KEY and their meanings,
1076 see `language-info-alist'."
1077 (if (symbolp lang-env)
1078 (setq lang-env (symbol-name lang-env)))
1079 (let (lang-slot key-slot)
1080 (setq lang-slot (assoc lang-env language-info-alist))
1081 (if (null lang-slot) ; If no slot for the language, add it.
1082 (setq lang-slot (list lang-env)
1083 language-info-alist (cons lang-slot language-info-alist)))
1084 (setq key-slot (assq key lang-slot))
1085 (if (null key-slot) ; If no slot for the key, add it.
1087 (setq key-slot (list key))
1088 (setcdr lang-slot (cons key-slot (cdr lang-slot)))))
1089 (setcdr key-slot (purecopy info))
1090 ;; Update the custom-type of `current-language-environment'.
1091 (put 'current-language-environment 'custom-type
1092 (cons 'choice (mapcar
1095 (sort (mapcar 'car language-info-alist) 'string<))))))
1097 (defun set-language-info-alist (lang-env alist &optional parents)
1098 "Store ALIST as the definition of language environment LANG-ENV.
1099 ALIST is an alist of KEY and INFO values. See the documentation of
1100 `language-info-alist' for the meanings of KEY and INFO.
1102 Optional arg PARENTS is a list of parent menu names; it specifies
1103 where to put this language environment in the
1104 Describe Language Environment and Set Language Environment menus.
1105 For example, (\"European\") means to put this language environment
1106 in the European submenu in each of those two menus."
1107 (if (symbolp lang-env)
1108 (setq lang-env (symbol-name lang-env)))
1109 (let ((describe-map describe-language-environment-map)
1110 (setup-map setup-language-environment-map))
1113 map parent-symbol parent prompt)
1115 (if (symbolp (setq parent-symbol (car l)))
1116 (setq parent (symbol-name parent))
1117 (setq parent parent-symbol parent-symbol (intern parent)))
1118 (setq map (lookup-key describe-map (vector parent-symbol)))
1119 ;; This prompt string is for define-prefix-command, so
1120 ;; that the map it creates will be suitable for a menu.
1121 (or map (setq prompt (format "%s Environment" parent)))
1124 (setq map (intern (format "describe-%s-environment-map"
1125 (downcase parent))))
1126 (define-prefix-command map nil prompt)
1127 (define-key-after describe-map (vector parent-symbol)
1128 (cons parent map) t)))
1129 (setq describe-map (symbol-value map))
1130 (setq map (lookup-key setup-map (vector parent-symbol)))
1133 (setq map (intern (format "setup-%s-environment-map"
1134 (downcase parent))))
1135 (define-prefix-command map nil prompt)
1136 (define-key-after setup-map (vector parent-symbol)
1137 (cons parent map) t)))
1138 (setq setup-map (symbol-value map))
1141 ;; Set up menu items for this language env.
1142 (let ((doc (assq 'documentation alist)))
1144 (define-key-after describe-map (vector (intern lang-env))
1145 (cons lang-env 'describe-specified-language-support) t)))
1146 (define-key-after setup-map (vector (intern lang-env))
1147 (cons lang-env 'setup-specified-language-environment) t)
1150 (set-language-info lang-env (car (car alist)) (cdr (car alist)))
1151 (setq alist (cdr alist)))))
1153 (defun read-language-name (key prompt &optional default)
1154 "Read a language environment name which has information for KEY.
1155 If KEY is nil, read any language environment.
1156 Prompt with PROMPT. DEFAULT is the default choice of language environment.
1157 This returns a language environment name as a string."
1158 (let* ((completion-ignore-case t)
1159 (name (completing-read prompt
1162 (function (lambda (elm) (and (listp elm) (assq key elm)))))
1163 t nil nil default)))
1164 (if (and (> (length name) 0)
1166 (get-language-info name key)))
1169 ;;; Multilingual input methods.
1171 "LEIM: Libraries of Emacs Input Methods."
1174 (defconst leim-list-file-name "leim-list.el"
1175 "Name of LEIM list file.
1176 This file contains a list of libraries of Emacs input methods (LEIM)
1177 in the format of Lisp expression for registering each input method.
1178 Emacs loads this file at startup time.")
1180 (defvar leim-list-header (format
1181 ";;; %s -- list of LEIM (Library of Emacs Input Method) -*-coding: iso-2022-7bit;-*-
1183 ;; This file contains a list of LEIM (Library of Emacs Input Method)
1184 ;; methods in the same directory as this file. Loading this file
1185 ;; registers all the input methods in Emacs.
1187 ;; Each entry has the form:
1188 ;; (register-input-method
1189 ;; INPUT-METHOD LANGUAGE-NAME ACTIVATE-FUNC
1190 ;; TITLE DESCRIPTION
1192 ;; See the function `register-input-method' for the meanings of the arguments.
1194 ;; If this directory is included in load-path, Emacs automatically
1195 ;; loads this file at startup time.
1198 leim-list-file-name)
1199 "Header to be inserted in LEIM list file.")
1201 (defvar leim-list-entry-regexp "^(register-input-method"
1202 "Regexp matching head of each entry in LEIM list file.
1203 See also the variable `leim-list-header'")
1205 (defvar update-leim-list-functions
1206 '(quail-update-leim-list-file)
1207 "List of functions to call to update LEIM list file.
1208 Each function is called with one arg, LEIM directory name.")
1210 (defun update-leim-list-file (&rest dirs)
1211 "Update LEIM list file in directories DIRS."
1212 (let ((functions update-leim-list-functions))
1214 (apply (car functions) dirs)
1215 (setq functions (cdr functions)))))
1217 (defvar current-input-method nil
1218 "The current input method for multilingual text.
1219 If nil, that means no input method is activated now.")
1220 (make-variable-buffer-local 'current-input-method)
1221 (put 'current-input-method 'permanent-local t)
1223 (defvar current-input-method-title nil
1224 "Title string of the current input method shown in mode line.")
1225 (make-variable-buffer-local 'current-input-method-title)
1226 (put 'current-input-method-title 'permanent-local t)
1228 (defcustom default-input-method nil
1229 "*Default input method for multilingual text (a string).
1230 This is the input method activated automatically by the command
1231 `toggle-input-method' (\\[toggle-input-method])."
1232 :link '(custom-manual "(emacs)Input Methods")
1234 :type '(choice (const nil) (string
1235 :completion-ignore-case t
1236 :complete-function widget-string-complete
1237 :completion-alist input-method-alist
1238 :prompt-history input-method-history))
1239 :set-after '(current-language-environment))
1241 (put 'input-method-function 'permanent-local t)
1243 (defvar input-method-history nil
1244 "History list for some commands that read input methods.")
1245 (make-variable-buffer-local 'input-method-history)
1246 (put 'input-method-history 'permanent-local t)
1248 (defvar inactivate-current-input-method-function nil
1249 "Function to call for inactivating the current input method.
1250 Every input method should set this to an appropriate value when activated.
1251 This function is called with no argument.
1253 This function should never change the value of `current-input-method'.
1254 It is set to nil by the function `inactivate-input-method'.")
1255 (make-variable-buffer-local 'inactivate-current-input-method-function)
1256 (put 'inactivate-current-input-method-function 'permanent-local t)
1258 (defvar describe-current-input-method-function nil
1259 "Function to call for describing the current input method.
1260 This function is called with no argument.")
1261 (make-variable-buffer-local 'describe-current-input-method-function)
1262 (put 'describe-current-input-method-function 'permanent-local t)
1264 (defvar input-method-alist nil
1265 "Alist of input method names vs how to use them.
1266 Each element has the form:
1267 (INPUT-METHOD LANGUAGE-ENV ACTIVATE-FUNC TITLE DESCRIPTION ARGS...)
1268 See the function `register-input-method' for the meanings of the elements.")
1270 (defun register-input-method (input-method lang-env &rest args)
1271 "Register INPUT-METHOD as an input method for language environment LANG-ENV.
1272 INPUT-METHOD and LANG-ENV are symbols or strings.
1274 The remaining arguments are:
1275 ACTIVATE-FUNC, TITLE, DESCRIPTION, and ARGS...
1276 ACTIVATE-FUNC is a function to call to activate this method.
1277 TITLE is a string to show in the mode line when this method is active.
1278 DESCRIPTION is a string describing this method and what it is good for.
1279 The ARGS, if any, are passed as arguments to ACTIVATE-FUNC.
1280 All told, the arguments to ACTIVATE-FUNC are INPUT-METHOD and the ARGS.
1282 This function is mainly used in the file \"leim-list.el\" which is
1283 created at Emacs build time, registering all Quail input methods
1284 contained in the Emacs distribution.
1286 In case you want to register a new Quail input method by yourself, be
1287 careful to use the same input method title as given in the third
1288 parameter of `quail-define-package'. (If the values are different, the
1289 string specified in this function takes precedence.)
1291 The commands `describe-input-method' and `list-input-methods' need
1292 these duplicated values to show some information about input methods
1293 without loading the relevant Quail packages."
1294 (if (symbolp lang-env)
1295 (setq lang-env (symbol-name lang-env)))
1296 (if (symbolp input-method)
1297 (setq input-method (symbol-name input-method)))
1298 (let ((info (cons lang-env args))
1299 (slot (assoc input-method input-method-alist)))
1302 (setq slot (cons input-method info))
1303 (setq input-method-alist (cons slot input-method-alist)))))
1305 (defun read-input-method-name (prompt &optional default inhibit-null)
1306 "Read a name of input method from a minibuffer prompting with PROMPT.
1307 If DEFAULT is non-nil, use that as the default,
1308 and substitute it into PROMPT at the first `%s'.
1309 If INHIBIT-NULL is non-nil, null input signals an error.
1311 The return value is a string."
1313 (setq prompt (format prompt default)))
1314 (let* ((completion-ignore-case t)
1315 ;; As it is quite normal to change input method in the
1316 ;; minibuffer, we must enable it even if
1317 ;; enable-recursive-minibuffers is currently nil.
1318 (enable-recursive-minibuffers t)
1319 ;; This binding is necessary because input-method-history is
1321 (input-method (completing-read prompt input-method-alist
1322 nil t nil 'input-method-history
1324 (if (and input-method (symbolp input-method))
1325 (setq input-method (symbol-name input-method)))
1326 (if (> (length input-method) 0)
1329 (error "No valid input method is specified")))))
1331 (defun activate-input-method (input-method)
1332 "Switch to input method INPUT-METHOD for the current buffer.
1333 If some other input method is already active, turn it off first.
1334 If INPUT-METHOD is nil, deactivate any current input method."
1335 (if (and input-method (symbolp input-method))
1336 (setq input-method (symbol-name input-method)))
1337 (if (and current-input-method
1338 (not (string= current-input-method input-method)))
1339 (inactivate-input-method))
1340 (unless (or current-input-method (null input-method))
1341 (let ((slot (assoc input-method input-method-alist)))
1343 (error "Can't activate input method `%s'" input-method))
1344 (setq current-input-method-title nil)
1345 (let ((func (nth 2 slot)))
1346 (if (functionp func)
1347 (apply (nth 2 slot) input-method (nthcdr 5 slot))
1348 (if (and (consp func) (symbolp (car func)) (symbolp (cdr func)))
1350 (require (cdr func))
1351 (apply (car func) input-method (nthcdr 5 slot)))
1352 (error "Can't activate input method `%s'" input-method))))
1353 (setq current-input-method input-method)
1354 (or (stringp current-input-method-title)
1355 (setq current-input-method-title (nth 3 slot)))
1357 (run-hooks 'input-method-activate-hook)
1358 (force-mode-line-update)))))
1360 (defun inactivate-input-method ()
1361 "Turn off the current input method."
1362 (when current-input-method
1363 (if input-method-history
1364 (unless (string= current-input-method (car input-method-history))
1365 (setq input-method-history
1366 (cons current-input-method
1367 (delete current-input-method input-method-history))))
1368 (setq input-method-history (list current-input-method)))
1370 (funcall inactivate-current-input-method-function)
1372 (run-hooks 'input-method-inactivate-hook)
1373 (setq current-input-method nil
1374 input-method-function nil
1375 current-input-method-title nil)
1376 (force-mode-line-update)))))
1378 (defun set-input-method (input-method &optional interactive)
1379 "Select and activate input method INPUT-METHOD for the current buffer.
1380 This also sets the default input method to the one you specify.
1381 If INPUT-METHOD is nil, this function turns off the input method, and
1382 also causes you to be prompted for a name of an input method the next
1383 time you invoke \\[toggle-input-method].
1384 When called interactively, the optional arg INTERACTIVE is non-nil,
1385 which marks the variable `default-input-method' as set for Custom buffers.
1387 To deactivate the input method interactively, use \\[toggle-input-method].
1388 To deactivate it programmatically, use \\[inactivate-input-method]."
1390 (let* ((default (or (car input-method-history) default-input-method)))
1391 (list (read-input-method-name
1392 (if default "Select input method (default %s): " "Select input method: ")
1395 (activate-input-method input-method)
1396 (setq default-input-method input-method)
1398 (customize-mark-as-set 'default-input-method))
1399 default-input-method)
1401 (defun toggle-input-method (&optional arg interactive)
1402 "Enable or disable multilingual text input method for the current buffer.
1403 Only one input method can be enabled at any time in a given buffer.
1405 The normal action is to enable an input method if none was
1406 enabled, and disable the current one otherwise. Which input method
1407 to enable can be determined in various ways--either the one most
1408 recently used, or the one specified by `default-input-method', or
1409 as a last resort by reading the name of an input method in the
1412 With a prefix argument, read an input method name with the minibuffer
1413 and enable that one. The default is the most recent input method specified
1414 \(not including the currently active input method, if any).
1416 When called interactively, the optional arg INTERACTIVE is non-nil,
1417 which marks the variable `default-input-method' as set for Custom buffers."
1419 (interactive "P\np")
1420 (if (and current-input-method (not arg))
1421 (inactivate-input-method)
1422 (let ((default (or (car input-method-history) default-input-method)))
1423 (if (and arg default (equal current-input-method default)
1424 (> (length input-method-history) 1))
1425 (setq default (nth 1 input-method-history)))
1426 (activate-input-method
1427 (if (or arg (not default))
1429 (read-input-method-name
1430 (if default "Input method (default %s): " "Input method: " )
1433 (unless default-input-method
1435 (setq default-input-method current-input-method)
1437 (customize-mark-as-set 'default-input-method)))))))
1439 (defun describe-input-method (input-method)
1440 "Describe input method INPUT-METHOD."
1442 (list (read-input-method-name
1443 "Describe input method (default, current choice): ")))
1444 (if (and input-method (symbolp input-method))
1445 (setq input-method (symbol-name input-method)))
1446 (help-setup-xref (list #'describe-input-method
1447 (or input-method current-input-method))
1450 (if (null input-method)
1451 (describe-current-input-method)
1452 (let ((current current-input-method))
1456 (activate-input-method input-method)
1457 (describe-current-input-method))
1458 (activate-input-method current))
1460 (activate-input-method current)
1461 (help-setup-xref (list #'describe-input-method input-method)
1463 (with-output-to-temp-buffer (help-buffer)
1464 (let ((elt (assoc input-method input-method-alist)))
1466 "Input method: %s (`%s' in mode line) for %s\n %s\n"
1467 input-method (nth 3 elt) (nth 1 elt) (nth 4 elt))))))))))
1469 (defun describe-current-input-method ()
1470 "Describe the input method currently in use.
1471 This is a subroutine for `describe-input-method'."
1472 (if current-input-method
1473 (if (and (symbolp describe-current-input-method-function)
1474 (fboundp describe-current-input-method-function))
1475 (funcall describe-current-input-method-function)
1476 (message "No way to describe the current input method `%s'"
1477 current-input-method)
1479 (error "No input method is activated now")))
1481 (defun read-multilingual-string (prompt &optional initial-input input-method)
1482 "Read a multilingual string from minibuffer, prompting with string PROMPT.
1483 The input method selected last time is activated in minibuffer.
1484 If optional second arg INITIAL-INPUT is non-nil, insert it in the minibuffer
1486 Optional 3rd argument INPUT-METHOD specifies the input method
1487 to be activated instead of the one selected last time. It is a symbol
1491 current-input-method
1492 default-input-method
1493 (read-input-method-name "Input method: " nil t)))
1494 (if (and input-method (symbolp input-method))
1495 (setq input-method (symbol-name input-method)))
1496 (let ((prev-input-method current-input-method))
1499 (activate-input-method input-method)
1500 (read-string prompt initial-input nil nil t))
1501 (activate-input-method prev-input-method))))
1503 ;; Variables to control behavior of input methods. All input methods
1504 ;; should react to these variables.
1506 (defcustom input-method-verbose-flag 'default
1507 "*A flag to control extra guidance given by input methods.
1508 The value should be nil, t, `complex-only', or `default'.
1510 The extra guidance is done by showing list of available keys in echo
1511 area. When you use the input method in the minibuffer, the guidance
1512 is shown at the bottom short window (split from the existing window).
1514 If the value is t, extra guidance is always given, if the value is
1515 nil, extra guidance is always suppressed.
1517 If the value is `complex-only', only complex input methods such as
1518 `chinese-py' and `japanese' give extra guidance.
1520 If the value is `default', complex input methods always give extra
1521 guidance, but simple input methods give it only when you are not in
1524 See also the variable `input-method-highlight-flag'."
1525 :type '(choice (const :tag "Always" t) (const :tag "Never" nil)
1526 (const complex-only) (const default))
1529 (defcustom input-method-highlight-flag t
1530 "*If this flag is non-nil, input methods highlight partially-entered text.
1531 For instance, while you are in the middle of a Quail input method sequence,
1532 the text inserted so far is temporarily underlined.
1533 The underlining goes away when you finish or abort the input method sequence.
1534 See also the variable `input-method-verbose-flag'."
1538 (defvar input-method-activate-hook nil
1539 "Normal hook run just after an input method is activated.
1541 The variable `current-input-method' keeps the input method name
1544 (defvar input-method-inactivate-hook nil
1545 "Normal hook run just after an input method is inactivated.
1547 The variable `current-input-method' still keeps the input method name
1550 (defvar input-method-after-insert-chunk-hook nil
1551 "Normal hook run just after an input method insert some chunk of text.")
1553 (defvar input-method-exit-on-first-char nil
1554 "This flag controls when an input method returns.
1555 Usually, the input method does not return while there's a possibility
1556 that it may find a different translation if a user types another key.
1557 But, it this flag is non-nil, the input method returns as soon as
1558 the current key sequence gets long enough to have some valid translation.")
1560 (defvar input-method-use-echo-area nil
1561 "This flag controls how an input method shows an intermediate key sequence.
1562 Usually, the input method inserts the intermediate key sequence,
1563 or candidate translations corresponding to the sequence,
1564 at point in the current buffer.
1565 But, if this flag is non-nil, it displays them in echo area instead.")
1567 (defvar input-method-exit-on-invalid-key nil
1568 "This flag controls the behaviour of an input method on invalid key input.
1569 Usually, when a user types a key which doesn't start any character
1570 handled by the input method, the key is handled by turning off the
1571 input method temporarily. After that key, the input method is re-enabled.
1572 But, if this flag is non-nil, the input method is never back on.")
1575 (defvar set-language-environment-hook nil
1576 "Normal hook run after some language environment is set.
1578 When you set some hook function here, that effect usually should not
1579 be inherited to another language environment. So, you had better set
1580 another function in `exit-language-environment-hook' (which see) to
1581 cancel the effect.")
1583 (defvar exit-language-environment-hook nil
1584 "Normal hook run after exiting from some language environment.
1585 When this hook is run, the variable `current-language-environment'
1586 is still bound to the language environment being exited.
1588 This hook is mainly used for canceling the effect of
1589 `set-language-environment-hook' (which-see).")
1591 (put 'setup-specified-language-environment 'apropos-inhibit t)
1593 (defun setup-specified-language-environment ()
1594 "Switch to a specified language environment."
1596 (let (language-name)
1597 (if (and (symbolp last-command-event)
1598 (or (not (eq last-command-event 'Default))
1599 (setq last-command-event 'English))
1600 (setq language-name (symbol-name last-command-event)))
1602 (set-language-environment language-name)
1603 (customize-mark-as-set 'current-language-environment))
1604 (error "Bogus calling sequence"))))
1606 (defcustom current-language-environment "English"
1607 "The last language environment specified with `set-language-environment'.
1608 This variable should be set only with \\[customize], which is equivalent
1609 to using the function `set-language-environment'."
1610 :link '(custom-manual "(emacs)Language Environments")
1611 :set (lambda (symbol value) (set-language-environment value))
1613 (or (car-safe (assoc-string
1614 (if (symbolp current-language-environment)
1615 (symbol-name current-language-environment)
1616 current-language-environment)
1617 language-info-alist t))
1619 ;; custom type will be updated with `set-language-info'.
1620 :type (if language-info-alist
1621 (cons 'choice (mapcar
1624 (sort (mapcar 'car language-info-alist) 'string<)))
1626 :initialize 'custom-initialize-default
1629 (defun reset-language-environment ()
1630 "Reset multilingual environment of Emacs to the default status.
1632 The default status is as follows:
1634 The default value of `buffer-file-coding-system' is nil.
1635 The default coding system for process I/O is nil.
1636 The default value for the command `set-terminal-coding-system' is nil.
1637 The default value for the command `set-keyboard-coding-system' is nil.
1639 The order of priorities of coding categories and the coding system
1640 bound to each category are as follows
1641 coding category coding system
1642 --------------------------------------------------
1643 coding-category-iso-8-1 iso-latin-1
1644 coding-category-iso-8-2 iso-latin-1
1645 coding-category-utf-8 mule-utf-8
1646 coding-category-utf-16-be mule-utf-16be-with-signature
1647 coding-category-utf-16-le mule-utf-16le-with-signature
1648 coding-category-iso-7-tight iso-2022-jp
1649 coding-category-iso-7 iso-2022-7bit
1650 coding-category-iso-7-else iso-2022-7bit-lock
1651 coding-category-iso-8-else iso-2022-8bit-ss2
1652 coding-category-emacs-mule emacs-mule
1653 coding-category-raw-text raw-text
1654 coding-category-sjis japanese-shift-jis
1655 coding-category-big5 chinese-big5
1656 coding-category-ccl nil
1657 coding-category-binary no-conversion"
1659 ;; This function formerly set default-enable-multibyte-characters to t,
1660 ;; but that is incorrect. It should not alter the unibyte/multibyte choice.
1662 (setq coding-category-iso-7-tight 'iso-2022-jp
1663 coding-category-iso-7 'iso-2022-7bit
1664 coding-category-iso-8-1 'iso-latin-1
1665 coding-category-iso-8-2 'iso-latin-1
1666 coding-category-iso-7-else 'iso-2022-7bit-lock
1667 coding-category-iso-8-else 'iso-2022-8bit-ss2
1668 coding-category-emacs-mule 'emacs-mule
1669 coding-category-raw-text 'raw-text
1670 coding-category-sjis 'japanese-shift-jis
1671 coding-category-big5 'chinese-big5
1672 coding-category-utf-16-be 'mule-utf-16be-with-signature
1673 coding-category-utf-16-le 'mule-utf-16le-with-signature
1674 coding-category-utf-8 'mule-utf-8
1675 coding-category-ccl nil
1676 coding-category-binary 'no-conversion)
1678 (set-coding-priority
1679 '(coding-category-iso-8-1
1680 coding-category-iso-8-2
1681 coding-category-utf-8
1682 coding-category-utf-16-be
1683 coding-category-utf-16-le
1684 coding-category-iso-7-tight
1685 coding-category-iso-7
1686 coding-category-iso-7-else
1687 coding-category-iso-8-else
1688 coding-category-emacs-mule
1689 coding-category-raw-text
1690 coding-category-sjis
1691 coding-category-big5
1693 coding-category-binary))
1695 ;; Changing the binding of a coding category requires this call.
1696 (update-coding-systems-internal)
1698 (set-default-coding-systems nil)
1699 (setq default-sendmail-coding-system 'iso-latin-1)
1700 ;; On Darwin systems, this should be utf-8, but when this file is loaded
1701 ;; utf-8 is not yet defined, so we set it in set-locale-environment instead.
1702 (setq default-file-name-coding-system 'iso-latin-1)
1703 ;; Preserve eol-type from existing default-process-coding-systems.
1704 ;; On non-unix-like systems in particular, these may have been set
1705 ;; carefully by the user, or by the startup code, to deal with the
1706 ;; users shell appropriately, so should not be altered by changing
1707 ;; language environment.
1708 (let ((output-coding
1709 ;; When bootstrapping, coding-systems are not defined yet, so
1710 ;; we need to catch the error from check-coding-system.
1712 (coding-system-change-text-conversion
1713 (car default-process-coding-system) 'undecided)
1714 (coding-system-error 'undecided)))
1717 (coding-system-change-text-conversion
1718 (cdr default-process-coding-system) 'iso-latin-1)
1719 (coding-system-error 'iso-latin-1))))
1720 (setq default-process-coding-system
1721 (cons output-coding input-coding)))
1723 ;; Don't alter the terminal and keyboard coding systems here.
1724 ;; The terminal still supports the same coding system
1725 ;; that it supported a minute ago.
1726 ;; (set-terminal-coding-system-internal nil)
1727 ;; (set-keyboard-coding-system-internal nil)
1729 (setq nonascii-translation-table nil
1730 nonascii-insert-offset 0)
1732 ;; Don't invoke fontset-related functions if fontsets aren't
1733 ;; supported in this build of Emacs.
1734 (and (fboundp 'fontset-list)
1735 (set-overriding-fontspec-internal nil)))
1737 (reset-language-environment)
1739 (defun set-display-table-and-terminal-coding-system (language-name &optional coding-system)
1740 "Set up the display table and terminal coding system for LANGUAGE-NAME."
1741 (let ((coding (get-language-info language-name 'unibyte-display)))
1743 (standard-display-european-internal)
1744 ;; The following 2 lines undo the 8-bit display that we set up
1745 ;; in standard-display-european-internal, which see. This is in
1746 ;; case the user has used standard-display-european earlier in
1747 ;; this session. (The MS-DOS port doesn't use that setup, so it
1748 ;; doesn't need to undo it.)
1749 (when standard-display-table
1751 (aset standard-display-table (+ i 128) nil))))
1752 (or (eq window-system 'pc)
1753 (set-terminal-coding-system (or coding-system coding)))))
1755 (defun set-language-environment (language-name)
1756 "Set up multi-lingual environment for using LANGUAGE-NAME.
1757 This sets the coding system priority and the default input method
1758 and sometimes other things. LANGUAGE-NAME should be a string
1759 which is the name of a language environment. For example, \"Latin-1\"
1760 specifies the character set for the major languages of Western Europe."
1761 (interactive (list (read-language-name
1763 "Set language environment (default, English): ")))
1765 (if (symbolp language-name)
1766 (setq language-name (symbol-name language-name)))
1767 (setq language-name "English"))
1768 (let ((slot (assoc-string language-name language-info-alist t)))
1770 (error "Language environment not defined: %S" language-name))
1771 (setq language-name (car slot)))
1772 (if current-language-environment
1773 (let ((func (get-language-info current-language-environment
1775 (run-hooks 'exit-language-environment-hook)
1776 (if (functionp func) (funcall func))))
1777 (let ((default-eol-type (coding-system-eol-type
1778 default-buffer-file-coding-system)))
1779 (reset-language-environment)
1781 ;; The features might set up coding systems.
1782 (let ((required-features (get-language-info language-name 'features)))
1783 (while required-features
1784 (require (car required-features))
1785 (setq required-features (cdr required-features))))
1787 (setq current-language-environment language-name)
1788 (set-language-environment-coding-systems language-name default-eol-type))
1789 (let ((input-method (get-language-info language-name 'input-method)))
1791 (setq default-input-method input-method)
1792 (if input-method-history
1793 (setq input-method-history
1795 (delete input-method input-method-history))))))
1796 (let ((nonascii (get-language-info language-name 'nonascii-translation))
1798 (if (eq window-system 'pc)
1800 (format "cp%d-nonascii-translation-table" dos-codepage)))))
1802 ((char-table-p nonascii)
1803 (setq nonascii-translation-table nonascii))
1804 ((and (eq window-system 'pc) (boundp dos-table))
1805 ;; DOS terminals' default is to use a special non-ASCII translation
1806 ;; table as appropriate for the installed codepage.
1807 (setq nonascii-translation-table (symbol-value dos-table)))
1808 ((charsetp nonascii)
1809 (setq nonascii-insert-offset (- (make-char nonascii) 128)))))
1811 ;; Unibyte setups if necessary.
1812 (unless default-enable-multibyte-characters
1813 ;; Syntax and case table.
1814 (let ((syntax (get-language-info language-name 'unibyte-syntax)))
1816 (let ((set-case-syntax-set-multibyte nil))
1817 (load syntax nil t))
1818 ;; No information for syntax and case. Reset to the defaults.
1819 (let ((syntax-table (standard-syntax-table))
1820 (case-table (standard-case-table))
1821 (ch (if (eq window-system 'pc) 128 160)))
1823 (modify-syntax-entry ch " " syntax-table)
1824 (aset case-table ch ch)
1826 (set-char-table-extra-slot case-table 0 nil)
1827 (set-char-table-extra-slot case-table 1 nil)
1828 (set-char-table-extra-slot case-table 2 nil))
1829 (set-standard-case-table (standard-case-table))
1830 (let ((list (buffer-list)))
1832 (with-current-buffer (car list)
1833 (set-case-table (standard-case-table)))
1834 (setq list (cdr list))))))
1835 (set-display-table-and-terminal-coding-system language-name))
1837 (let ((required-features (get-language-info language-name 'features)))
1838 (while required-features
1839 (require (car required-features))
1840 (setq required-features (cdr required-features))))
1842 ;; Don't invoke fontset-related functions if fontsets aren't
1843 ;; supported in this build of Emacs.
1844 (when (fboundp 'fontset-list)
1845 (let ((overriding-fontspec (get-language-info language-name
1846 'overriding-fontspec)))
1847 (if overriding-fontspec
1848 (set-overriding-fontspec-internal overriding-fontspec))))
1850 (let ((func (get-language-info language-name 'setup-function)))
1851 (if (functionp func)
1853 (if (and utf-translate-cjk-mode
1854 utf-translate-cjk-lang-env
1855 (not (eq utf-translate-cjk-lang-env language-name))
1857 (dolist (charset (get-language-info language-name 'charset))
1858 (if (memq charset utf-translate-cjk-charsets)
1861 (utf-translate-cjk-load-tables))
1862 (run-hooks 'set-language-environment-hook)
1863 (force-mode-line-update t))
1865 (defun standard-display-european-internal ()
1866 ;; Actually set up direct output of non-ASCII characters.
1867 (standard-display-8bit (if (eq window-system 'pc) 128 160) 255)
1868 ;; Unibyte Emacs on MS-DOS wants to display all 8-bit characters with
1869 ;; the native font, and codes 160 and 146 stand for something very
1871 (or (and (eq window-system 'pc) (not default-enable-multibyte-characters))
1873 ;; Most X fonts used to do the wrong thing for latin-1 code 160.
1874 (unless (and (eq window-system 'x)
1875 ;; XFree86 4 has fixed the fonts.
1876 (string= "The XFree86 Project, Inc" (x-server-vendor))
1877 (> (aref (number-to-string (nth 2 (x-server-version))) 0)
1879 ;; Make non-line-break space display as a plain space.
1880 (aset standard-display-table 160 [32]))
1881 ;; Most Windows programs send out apostrophes as \222. Most X fonts
1882 ;; don't contain a character at that position. Map it to the ASCII
1883 ;; apostrophe. [This is actually RIGHT SINGLE QUOTATION MARK,
1884 ;; U+2019, normally from the windows-1252 character set. XFree 4
1885 ;; fonts probably have the appropriate glyph at this position,
1886 ;; so they could use standard-display-8bit. It's better to use a
1887 ;; proper windows-1252 coding system. --fx]
1888 (aset standard-display-table 146 [39]))))
1890 (defun set-language-environment-coding-systems (language-name
1892 "Do various coding system setups for language environment LANGUAGE-NAME.
1894 The optional arg EOL-TYPE specifies the eol-type of the default value
1895 of `buffer-file-coding-system' set by this function."
1896 (let* ((priority (get-language-info language-name 'coding-priority))
1897 (default-coding (car priority)))
1899 (let ((categories (mapcar 'coding-system-category priority)))
1900 (set-default-coding-systems
1901 (if (memq eol-type '(0 1 2 unix dos mac))
1902 (coding-system-change-eol-conversion default-coding eol-type)
1904 (setq default-sendmail-coding-system default-coding)
1905 (set-coding-priority categories)
1907 (set (car categories) (car priority))
1908 (setq priority (cdr priority) categories (cdr categories)))
1909 ;; Changing the binding of a coding category requires this call.
1910 (update-coding-systems-internal)))))
1912 (defsubst princ-list (&rest args)
1913 "Print all arguments with `princ', then print \"\n\"."
1914 (while args (princ (car args)) (setq args (cdr args)))
1917 (put 'describe-specified-language-support 'apropos-inhibit t)
1919 ;; Print language-specific information such as input methods,
1920 ;; charsets, and coding systems. This function is intended to be
1921 ;; called from the menu:
1922 ;; [menu-bar mule describe-language-environment LANGUAGE]
1923 ;; and should not run it by `M-x describe-current-input-method-function'.
1924 (defun describe-specified-language-support ()
1925 "Describe how Emacs supports the specified language environment."
1927 (let (language-name)
1928 (if (not (and (symbolp last-command-event)
1929 (or (not (eq last-command-event 'Default))
1930 (setq last-command-event 'English))
1931 (setq language-name (symbol-name last-command-event))))
1932 (error "Bogus calling sequence"))
1933 (describe-language-environment language-name)))
1935 (defun describe-language-environment (language-name)
1936 "Describe how Emacs supports language environment LANGUAGE-NAME."
1938 (list (read-language-name
1940 "Describe language environment (default, current choice): ")))
1941 (if (null language-name)
1942 (setq language-name current-language-environment))
1943 (if (or (null language-name)
1944 (null (get-language-info language-name 'documentation)))
1945 (error "No documentation for the specified language"))
1946 (if (symbolp language-name)
1947 (setq language-name (symbol-name language-name)))
1948 (dolist (feature (get-language-info language-name 'features))
1950 (let ((doc (get-language-info language-name 'documentation)))
1951 (help-setup-xref (list #'describe-language-environment language-name)
1953 (with-output-to-temp-buffer (help-buffer)
1955 (set-buffer standard-output)
1956 (insert language-name " language environment\n\n")
1958 (insert doc "\n\n"))
1960 (let ((str (eval (get-language-info language-name 'sample-text))))
1962 (insert "Sample text:\n " str "\n\n")))
1964 (let ((input-method (get-language-info language-name 'input-method))
1965 (l (copy-sequence input-method-alist)))
1966 (insert "Input methods")
1968 (insert " (default, " input-method ")")
1969 (setq input-method (assoc input-method input-method-alist))
1970 (setq l (cons input-method (delete input-method l))))
1973 (when (string= language-name (nth 1 (car l)))
1974 (insert " " (car (car l)))
1975 (search-backward (car (car l)))
1976 (help-xref-button 0 'help-input-method (car (car l)))
1977 (goto-char (point-max))
1979 (if (stringp (nth 3 (car l)))
1981 (car (nth 3 (car l))))
1982 "\" in mode line)\n"))
1985 (insert "Character sets:\n")
1986 (let ((l (get-language-info language-name 'charset)))
1988 (insert " nothing specific to " language-name "\n")
1990 (insert " " (symbol-name (car l)))
1991 (search-backward (symbol-name (car l)))
1992 (help-xref-button 0 'help-character-set (car l))
1993 (goto-char (point-max))
1994 (insert ": " (charset-description (car l)) "\n")
1997 (insert "Coding systems:\n")
1998 (let ((l (get-language-info language-name 'coding-system)))
2000 (insert " nothing specific to " language-name "\n")
2002 (insert " " (symbol-name (car l)))
2003 (search-backward (symbol-name (car l)))
2004 (help-xref-button 0 'help-coding-system (car l))
2005 (goto-char (point-max))
2007 (coding-system-mnemonic (car l))
2008 "' in mode line):\n\t"
2009 (coding-system-doc-string (car l))
2011 (let ((aliases (coding-system-get (car l)
2012 'alias-coding-systems)))
2014 (insert "\t(alias:")
2016 (insert " " (symbol-name (car aliases)))
2017 (setq aliases (cdr aliases)))
2019 (setq l (cdr l)))))))))
2023 (defvar locale-translation-file-name nil
2024 "File name for the system's file of locale-name aliases, or nil if none.")
2026 ;; The following definitions might as well be marked as constants and
2027 ;; purecopied, since they're normally used on startup, and probably
2028 ;; should reflect the facilities of the base Emacs.
2029 (defconst locale-language-names
2032 ;; Locale names of the form LANGUAGE[_TERRITORY][.CODESET][@MODIFIER]
2033 ;; as specified in the Single Unix Spec, Version 2.
2034 ;; LANGUAGE is a language code taken from ISO 639:1988 (E/F)
2035 ;; with additions from ISO 639/RA Newsletter No.1/1989;
2036 ;; see Internet RFC 2165 (1997-06) and
2037 ;; http://www.evertype.com/standards/iso639/iso639-en.html
2038 ;; TERRITORY is a country code taken from ISO 3166
2039 ;; http://www.din.de/gremien/nas/nabd/iso3166ma/codlstp1/en_listp1.html.
2040 ;; CODESET and MODIFIER are implementation-dependent.
2042 ;; jasonr comments: MS Windows uses three letter codes for
2043 ;; languages instead of the two letter ISO codes that POSIX
2044 ;; uses. In most cases the first two letters are the same, so
2045 ;; most of the regexps in locale-language-names work. Japanese
2046 ;; and Chinese are exceptions, which are listed in the
2047 ;; non-standard section at the bottom of locale-language-names.
2049 ("aa_DJ" . "Latin-1") ; Afar
2052 ("af" . "Latin-1") ; Afrikaans
2053 ("am" "Ethiopic" utf-8) ; Amharic
2054 ("an" . "Latin-9") ; Aragonese
2055 ; ar Arabic glibc uses 8859-6
2058 ("az" . "UTF-8") ; Azerbaijani
2060 ("be" "Belarusian" cp1251) ; Belarusian [Byelorussian until early 1990s]
2061 ("bg" "Bulgarian" cp1251) ; Bulgarian
2064 ("bn" . "UTF-8") ; Bengali, Bangla
2066 ("br" . "Latin-1") ; Breton
2067 ("bs" . "Latin-2") ; Bosnian
2068 ("byn" . "UTF-8") ; Bilin; Blin
2069 ("ca" . "Latin-1") ; Catalan
2071 ("cs" "Czech" iso-8859-2)
2072 ("cy" "Welsh" iso-8859-14)
2073 ("da" . "Latin-1") ; Danish
2074 ("de" "German" iso-8859-1)
2076 ("el" "Greek" iso-8859-7)
2077 ;; Users who specify "en" explicitly typically want Latin-1, not ASCII.
2078 ;; That's actually what the GNU locales define, modulo things like
2080 ("en_IN" "English" utf-8) ; glibc uses utf-8 for English in India
2081 ("en" . "Latin-1") ; English
2082 ("eo" . "Latin-3") ; Esperanto
2083 ("es" "Spanish" iso-8859-1)
2084 ("et" . "Latin-1") ; Estonian
2085 ("eu" . "Latin-1") ; Basque
2086 ("fa" . "UTF-8") ; Persian
2087 ("fi" . "Latin-1") ; Finnish
2088 ("fj" . "Latin-1") ; Fiji
2089 ("fo" . "Latin-1") ; Faroese
2090 ("fr" "French" iso-8859-1) ; French
2091 ("fy" . "Latin-1") ; Frisian
2092 ("ga" . "Latin-1") ; Irish Gaelic (new orthography)
2093 ("gd" . "Latin-9") ; Scots Gaelic
2094 ("gez" "Ethiopic" utf-8) ; Geez
2095 ("gl" . "Latin-1") ; Gallegan; Galician
2097 ("gu" . "UTF-8") ; Gujarati
2098 ("gv" . "Latin-1") ; Manx Gaelic
2100 ("he" "Hebrew" iso-8859-8)
2101 ("hi" "Devanagari" utf-8) ; Hindi
2102 ("hr" "Croatian" iso-8859-2) ; Croatian
2103 ("hu" . "Latin-2") ; Hungarian
2106 ("id" . "Latin-1") ; Indonesian
2109 ("is" . "Latin-1") ; Icelandic
2110 ("it" "Italian" iso-8859-1) ; Italian
2112 ("iw" "Hebrew" iso-8859-8)
2113 ("ja" "Japanese" euc-jp)
2115 ("ka" "Georgian" georgian-ps) ; Georgian
2117 ("kl" . "Latin-1") ; Greenlandic
2119 ("kn" "Kannada" utf-8)
2120 ("ko" "Korean" euc-kr)
2123 ("kw" . "Latin-1") ; Cornish
2125 ("la" . "Latin-1") ; Latin
2126 ("lb" . "Latin-1") ; Luxemburgish
2127 ("lg" . "Laint-6") ; Ganda
2129 ("lo" "Lao" utf-8) ; Laothian
2130 ("lt" "Lithuanian" iso-8859-13)
2131 ("lv" . "Latvian") ; Latvian, Lettish
2133 ("mi" . "Latin-7") ; Maori
2134 ("mk" "Cyrillic-ISO" iso-8859-5) ; Macedonian
2135 ("ml" "Malayalam" utf-8)
2136 ("mn" . "UTF-8") ; Mongolian
2138 ("mr" "Devanagari" utf-8) ; Marathi
2139 ("ms" . "Latin-1") ; Malay
2140 ("mt" . "Latin-3") ; Maltese
2143 ("nb" . "Latin-1") ; Norwegian
2144 ("ne" "Devanagari" utf-8) ; Nepali
2145 ("nl" "Dutch" iso-8859-1)
2146 ("no" . "Latin-1") ; Norwegian
2147 ("oc" . "Latin-1") ; Occitan
2148 ("om_ET" . "UTF-8") ; (Afan) Oromo
2149 ("om" . "Latin-1") ; (Afan) Oromo
2151 ("pa" . "UTF-8") ; Punjabi
2152 ("pl" . "Latin-2") ; Polish
2154 ("pt" . "Latin-1") ; Portuguese
2156 ("rm" . "Latin-1") ; Rhaeto-Romanic
2158 ("ro" "Romanian" iso-8859-2)
2159 ("ru_RU" "Russian" iso-8859-5)
2160 ("ru_UA" "Russian" koi8-u)
2162 ("sa" . "Devanagari") ; Sanskrit
2164 ("se" . "UTF-8") ; Northern Sami
2166 ("sh" . "Latin-2") ; Serbo-Croatian
2168 ("sid" . "UTF-8") ; Sidamo
2169 ("sk" "Slovak" iso-8859-2)
2170 ("sl" "Slovenian" iso-8859-2)
2173 ("so_ET" "UTF-8") ; Somali
2174 ("so" "Latin-1") ; Somali
2175 ("sq" . "Latin-1") ; Albanian
2176 ("sr_YU@cyrillic" . "Cyrillic-ISO") ; Serbian (Cyrillic alphabet)
2177 ("sr" . "Latin-2") ; Serbian (Latin alphabet)
2179 ("st" . "Latin-1") ; Sesotho
2181 ("sv" "Swedish" iso-8859-1) ; Swedish
2182 ("sw" . "Latin-1") ; Swahili
2183 ("ta" "Tamil" utf-8)
2184 ("te" . "UTF-8") ; Telugu
2185 ("tg" "Tajik" koi8-t)
2186 ("th" "Thai" tis-620)
2187 ("ti" "Ethiopic" utf-8) ; Tigrinya
2188 ("tig_ER" . "UTF-8") ; Tigre
2190 ("tl" . "Latin-1") ; Tagalog
2193 ("tr" "Turkish" iso-8859-9)
2195 ("tt" . "UTF-8") ; Tatar
2198 ("uk" "Ukrainian" koi8-u)
2199 ("ur" . "UTF-8") ; Urdu
2200 ("uz_UZ@cyrillic" . "UTF-8"); Uzbek
2201 ("uz" . "Latin-1") ; Uzbek
2202 ("vi" "Vietnamese" utf-8)
2204 ("wa" . "Latin-1") ; Walloon
2206 ("xh" . "Latin-1") ; Xhosa
2207 ("yi" . "Windows-1255") ; Yiddish
2210 ("zh_HK" . "Chinese-Big5")
2211 ("zh_TW" . "Chinese-Big5")
2212 ("zh_CN" . "Chinese-GB")
2213 ("zh" . "Chinese-GB")
2214 ; zh_CN.GB18030/GB18030 \
2216 ; zh_HK/BIG5-HKSCS \
2217 ("zu" . "Latin-1") ; Zulu
2219 ;; ISO standard locales
2221 ("posix$" . "ASCII")
2223 ;; The "IPA" Emacs language environment does not correspond
2224 ;; to any ISO 639 code, so let it stand for itself.
2227 ;; Nonstandard or obsolete language codes
2228 ("cz" . "Czech") ; e.g. Solaris 2.6
2229 ("ee" . "Latin-4") ; Estonian, e.g. X11R6.4
2230 ("iw" . "Hebrew") ; e.g. X11R6.4
2231 ("sp" . "Cyrillic-ISO") ; Serbian (Cyrillic alphabet), e.g. X11R6.4
2232 ("su" . "Latin-1") ; Finnish, e.g. Solaris 2.6
2233 ("jp" . "Japanese") ; e.g. MS Windows
2234 ("chs" . "Chinese-GB") ; MS Windows Chinese Simplified
2235 ("cht" . "Chinese-BIG5") ; MS Windows Chinese Traditional
2237 "Alist of locale regexps vs the corresponding languages and coding systems.
2238 Each element has these form:
2239 \(LOCALE-REGEXP LANG-ENV CODING-SYSTEM)
2240 The first element whose LOCALE-REGEXP matches the start of a
2241 downcased locale specifies the LANG-ENV \(language environtment)
2242 and CODING-SYSTEM corresponding to that locale. If there is no
2243 appropriate language environment, the element may have this form:
2244 \(LOCALE-REGEXP . LANG-ENV)
2245 In this case, LANG-ENV is one of generic language environments for an
2246 specific encoding such as \"Latin-1\" and \"UTF-8\".")
2248 (defconst locale-charset-language-names
2250 '((".*8859[-_]?1\\>" . "Latin-1")
2251 (".*8859[-_]?2\\>" . "Latin-2")
2252 (".*8859[-_]?3\\>" . "Latin-3")
2253 (".*8859[-_]?4\\>" . "Latin-4")
2254 (".*8859[-_]?9\\>" . "Latin-5")
2255 (".*8859[-_]?14\\>" . "Latin-8")
2256 (".*8859[-_]?15\\>" . "Latin-9")
2257 (".*utf\\(?:-?8\\)?\\>" . "UTF-8")
2258 ;; utf-8@euro exists, so put this last. (@euro really specifies
2259 ;; the currency, rather than the charset.)
2260 (".*@euro\\>" . "Latin-9")))
2261 "List of pairs of locale regexps and charset language names.
2262 The first element whose locale regexp matches the start of a downcased locale
2263 specifies the language name whose charset corresponds to that locale.
2264 This language name is used if the locale is not listed in
2265 `locale-language-names'")
2267 (defconst locale-preferred-coding-systems
2269 '((".*8859[-_]?1\\>" . iso-8859-1)
2270 (".*8859[-_]?2\\>" . iso-8859-2)
2271 (".*8859[-_]?3\\>" . iso-8859-3)
2272 (".*8859[-_]?4\\>" . iso-8859-4)
2273 (".*8859[-_]?9\\>" . iso-8859-9)
2274 (".*8859[-_]?14\\>" . iso-8859-14)
2275 (".*8859[-_]?15\\>" . iso-8859-15)
2276 (".*utf\\(?:-?8\\)?" . utf-8)
2277 ;; utf-8@euro exists, so put this after utf-8. (@euro really
2278 ;; specifies the currency, rather than the charset.)
2279 (".*@euro" . iso-8859-15)
2280 ("koi8-?r" . koi8-r)
2281 ("koi8-?u" . koi8-u)
2284 ("euc-?tw" . euc-tw)
2285 ;; We don't support GBK, but as it is upper compatible with
2286 ;; GB-2312, we setup the default coding system to gb2312.
2288 ;; We don't support BIG5-HKSCS, but as it is upper compatible with
2289 ;; BIG5, we setup the default coding system to big5.
2290 ("big5hkscs" . big5)
2291 ("ja.*[._]euc" . japanese-iso-8bit)
2292 ("ja.*[._]jis7" . iso-2022-jp)
2293 ("ja.*[._]pck" . japanese-shift-jis)
2294 ("ja.*[._]sjis" . japanese-shift-jis)
2295 ("jpn" . japanese-shift-jis) ; MS-Windows uses this.
2297 "List of pairs of locale regexps and preferred coding systems.
2298 The first element whose locale regexp matches the start of a downcased locale
2299 specifies the coding system to prefer when using that locale.
2300 This coding system is used if the locale specifies a specific charset.")
2302 (defun locale-name-match (key alist)
2303 "Search for KEY in ALIST, which should be a list of regexp-value pairs.
2304 Return the value corresponding to the first regexp that matches the
2305 start of KEY, or nil if there is no match."
2307 (while (and alist (not element))
2308 (if (string-match (concat "\\`\\(?:" (car (car alist)) "\\)") key)
2309 (setq element (car alist)))
2310 (setq alist (cdr alist)))
2313 (defun locale-charset-match-p (charset1 charset2)
2314 "Whether charset names (strings) CHARSET1 and CHARSET2 are equivalent.
2315 Matching is done ignoring case and any hyphens and underscores in the
2316 names. E.g. `ISO_8859-1' and `iso88591' both match `iso-8859-1'."
2317 (setq charset1 (replace-regexp-in-string "[-_]" "" charset1))
2318 (setq charset2 (replace-regexp-in-string "[-_]" "" charset2))
2319 (eq t (compare-strings charset1 nil nil charset2 nil nil t)))
2321 (defvar locale-charset-alist nil
2322 "Coding system alist keyed on locale-style charset name.
2323 Used by `locale-charset-to-coding-system'.")
2325 (defun locale-charset-to-coding-system (charset)
2326 "Find coding system corresponding to CHARSET.
2327 CHARSET is any sort of non-Emacs charset name, such as might be used
2328 in a locale codeset, or elsewhere. It is matched to a coding system
2329 first by case-insensitive lookup in `locale-charset-alist'. Then
2330 matches are looked for in the coding system list, treating case and
2331 the characters `-' and `_' as insignificant. The coding system base
2332 is returned. Thus, for instance, if charset \"ISO8859-2\",
2333 `iso-latin-2' is returned."
2334 (or (car (assoc-string charset locale-charset-alist t))
2335 (let ((cs coding-system-alist)
2337 (while (and (not c) cs)
2338 (if (locale-charset-match-p charset (caar cs))
2339 (setq c (intern (caar cs)))
2341 (if c (coding-system-base c)))))
2343 ;; Fixme: This ought to deal with the territory part of the locale
2344 ;; too, for setting things such as calendar holidays, ps-print paper
2345 ;; size, spelling dictionary.
2347 (defun set-locale-environment (&optional locale-name)
2348 "Set up multi-lingual environment for using LOCALE-NAME.
2349 This sets the language environment, the coding system priority,
2350 the default input method and sometimes other things.
2352 LOCALE-NAME should be a string which is the name of a locale supported
2353 by the system. Often it is of the form xx_XX.CODE, where xx is a
2354 language, XX is a country, and CODE specifies a character set and
2355 coding system. For example, the locale name \"ja_JP.EUC\" might name
2356 a locale for Japanese in Japan using the `japanese-iso-8bit'
2357 coding-system. The name may also have a modifier suffix, e.g. `@euro'
2360 If LOCALE-NAME is nil, its value is taken from the environment
2361 variables LC_ALL, LC_CTYPE and LANG (the first one that is set).
2363 The locale names supported by your system can typically be found in a
2364 directory named `/usr/share/locale' or `/usr/lib/locale'. LOCALE-NAME
2365 will be translated according to the table specified by
2366 `locale-translation-file-name'.
2368 See also `locale-charset-language-names', `locale-language-names',
2369 `locale-preferred-coding-systems' and `locale-coding-system'."
2370 (interactive "sSet environment for locale: ")
2372 ;; Do this at runtime for the sake of binaries possibly transported
2373 ;; to a system without X.
2374 (setq locale-translation-file-name
2376 '("/usr/lib/X11/locale/locale.alias" ; e.g. X11R6.4
2377 "/usr/X11R6/lib/X11/locale/locale.alias" ; XFree86, e.g. RedHat 4.2
2378 "/usr/openwin/lib/locale/locale.alias" ; e.g. Solaris 2.6
2380 ;; The following name appears after the X-related names above,
2381 ;; since the X-related names are what X actually uses.
2382 "/usr/share/locale/locale.alias" ; GNU/Linux sans X
2384 (while (and files (not (file-exists-p (car files))))
2385 (setq files (cdr files)))
2388 (let ((locale locale-name))
2391 ;; Use the first of these three environment variables
2392 ;; that has a nonempty value.
2393 (let ((vars '("LC_ALL" "LC_CTYPE" "LANG")))
2395 (= 0 (length locale))) ; nil or empty string
2396 (setq locale (getenv (pop vars))))))
2398 (unless (or locale (not (fboundp 'mac-get-preference)))
2399 (setq locale (mac-get-preference "AppleLocale"))
2401 (let ((languages (mac-get-preference "AppleLanguages")))
2402 (unless (= (length languages) 0) ; nil or empty vector
2403 (setq locale (aref languages 0))))))
2404 (unless (or locale (not (boundp 'mac-system-locale)))
2405 (setq locale mac-system-locale))
2409 ;; Translate "swedish" into "sv_SE.ISO8859-1", and so on,
2410 ;; using the translation file that many systems have.
2411 (when locale-translation-file-name
2413 (insert-file-contents locale-translation-file-name)
2414 (when (re-search-forward
2415 (concat "^" (regexp-quote locale) ":?[ \t]+") nil t)
2416 (setq locale (buffer-substring (point) (line-end-position))))))
2418 ;; Leave the system locales alone if the caller did not specify
2419 ;; an explicit locale name, as their defaults are set from
2420 ;; LC_MESSAGES and LC_TIME, not LC_CTYPE, and the user might not
2421 ;; want to set them to the same value as LC_CTYPE.
2423 (setq system-messages-locale locale)
2424 (setq system-time-locale locale))
2426 (setq locale (downcase locale))
2428 (let ((language-name
2429 (locale-name-match locale locale-language-names))
2430 (charset-language-name
2431 (locale-name-match locale locale-charset-language-names))
2433 (or (locale-name-match locale locale-preferred-coding-systems)
2435 (if (string-match "\\.\\([^@]+\\)" locale)
2436 (locale-charset-to-coding-system
2437 (match-string 1 locale))))
2438 (and (eq system-type 'macos) mac-system-coding-system))))
2440 (if (consp language-name)
2441 ;; locale-language-names specify both lang-env and coding.
2442 ;; But, what specified in locale-preferred-coding-systems
2443 ;; has higher priority.
2444 (setq coding-system (or coding-system
2445 (nth 1 language-name))
2446 language-name (car language-name))
2447 ;; Otherwise, if locale is not listed in locale-language-names,
2448 ;; use what listed in locale-charset-language-names.
2449 (if (not language-name)
2450 (setq language-name charset-language-name)))
2454 ;; Set up for this character set. This is now the right way
2455 ;; to do it for both unibyte and multibyte modes.
2456 (set-language-environment language-name)
2458 ;; If default-enable-multibyte-characters is nil,
2459 ;; we are using single-byte characters,
2460 ;; so the display table and terminal coding system are irrelevant.
2461 (when default-enable-multibyte-characters
2462 (set-display-table-and-terminal-coding-system
2463 language-name coding-system))
2465 ;; Set the `keyboard-coding-system' if appropriate (tty
2466 ;; only). At least X and MS Windows can generate
2467 ;; multilingual input.
2468 (unless window-system
2469 (let ((kcs (or coding-system
2470 (car (get-language-info language-name
2472 (if kcs (set-keyboard-coding-system kcs))))
2474 (setq locale-coding-system
2475 (car (get-language-info language-name 'coding-priority))))
2477 (when (and coding-system
2478 (not (coding-system-equal coding-system
2479 locale-coding-system)))
2480 (prefer-coding-system coding-system)
2481 (setq locale-coding-system coding-system))))
2483 ;; On Windows, override locale-coding-system,
2484 ;; keyboard-coding-system with system codepage. Note:
2485 ;; selection-coding-system is already set in w32select.c.
2486 (when (boundp 'w32-ansi-code-page)
2487 (let ((code-page-coding (intern (format "cp%d" w32-ansi-code-page))))
2488 (when (coding-system-p code-page-coding)
2489 (setq locale-coding-system code-page-coding)
2490 (set-keyboard-coding-system code-page-coding)
2491 (set-terminal-coding-system code-page-coding))))
2493 (when (eq system-type 'darwin)
2494 ;; On Darwin, file names are always encoded in utf-8, no matter
2496 (setq default-file-name-coding-system 'utf-8)
2497 ;; Mac OS X's Terminal.app by default uses utf-8 regardless of
2499 (when (and (null window-system)
2500 (equal (getenv "TERM_PROGRAM") "Apple_Terminal"))
2501 (set-terminal-coding-system 'utf-8)
2502 (set-keyboard-coding-system 'utf-8)))
2504 ;; Default to A4 paper if we're not in a C, POSIX or US locale.
2505 ;; (See comments in Flocale_info.)
2506 (let ((locale locale)
2507 (paper (locale-info 'paper)))
2509 ;; This will always be null at the time of writing.
2511 ((equal paper '(216 279))
2512 (setq ps-paper-type 'letter))
2513 ((equal paper '(210 297))
2514 (setq ps-paper-type 'a4)))
2515 (let ((vars '("LC_ALL" "LC_PAPER" "LANG")))
2516 (while (and vars (= 0 (length locale)))
2517 (setq locale (getenv (pop vars)))))
2519 ;; As of glibc 2.2.5, these are the only US Letter locales,
2520 ;; and the rest are A4.
2522 (or (locale-name-match locale '(("c$" . letter)
2527 ("enu$" . letter) ; Windows
2534 ;;; Charset property
2536 (defun get-charset-property (charset propname)
2537 "Return the value of CHARSET's PROPNAME property.
2538 This is the last value stored with
2539 (put-charset-property CHARSET PROPNAME VALUE)."
2540 (and (not (eq charset 'composition))
2541 (plist-get (charset-plist charset) propname)))
2543 (defun put-charset-property (charset propname value)
2544 "Store CHARSETS's PROPNAME property with value VALUE.
2545 It can be retrieved with `(get-charset-property CHARSET PROPNAME)'."
2546 (or (eq charset 'composition)
2547 (set-charset-plist charset
2548 (plist-put (charset-plist charset) propname value))))
2550 ;;; Character code property
2551 (put 'char-code-property-table 'char-table-extra-slots 0)
2553 (defvar char-code-property-table
2554 (make-char-table 'char-code-property-table)
2555 "Char-table containing a property list of each character code.
2557 See also the documentation of `get-char-code-property' and
2558 `put-char-code-property'.")
2560 (defun get-char-code-property (char propname)
2561 "Return the value of CHAR's PROPNAME property in `char-code-property-table'."
2562 (let ((plist (aref char-code-property-table char)))
2564 (car (cdr (memq propname plist))))))
2566 (defun put-char-code-property (char propname value)
2567 "Store CHAR's PROPNAME property with VALUE in `char-code-property-table'.
2568 It can be retrieved with `(get-char-code-property CHAR PROPNAME)'."
2569 (let ((plist (aref char-code-property-table char)))
2571 (let ((slot (memq propname plist)))
2573 (setcar (cdr slot) value)
2574 (nconc plist (list propname value))))
2575 (aset char-code-property-table char (list propname value)))))
2578 ;; Pretty description of encoded string
2580 ;; Alist of ISO 2022 control code vs the corresponding mnemonic string.
2581 (defvar iso-2022-control-alist
2589 (defun encoded-string-description (str coding-system)
2590 "Return a pretty description of STR that is encoded by CODING-SYSTEM."
2591 (setq str (string-as-unibyte str))
2593 (if (and coding-system (eq (coding-system-type coding-system) 2))
2594 ;; Try to get a pretty description for ISO 2022 escape sequences.
2595 (function (lambda (x) (or (cdr (assq x iso-2022-control-alist))
2596 (format "0x%02X" x))))
2597 (function (lambda (x) (format "0x%02X" x))))
2600 (defun encode-coding-char (char coding-system)
2601 "Encode CHAR by CODING-SYSTEM and return the resulting string.
2602 If CODING-SYSTEM can't safely encode CHAR, return nil."
2603 (let ((str1 (string-as-multibyte (char-to-string char)))
2604 (str2 (string-as-multibyte (make-string 2 char)))
2605 (safe-chars (and coding-system
2606 (coding-system-get coding-system 'safe-chars)))
2607 (charset (char-charset char))
2609 (when (or (eq safe-chars t)
2611 (and safe-chars (aref safe-chars char)))
2612 ;; We must find the encoded string of CHAR. But, just encoding
2613 ;; CHAR will put extra control sequences (usually to designate
2614 ;; ASCII charset) at the tail if type of CODING is ISO 2022.
2615 ;; To exclude such tailing bytes, we at first encode one-char
2616 ;; string and two-char string, then check how many bytes at the
2617 ;; tail of both encoded strings are the same.
2619 (setq enc1 (encode-coding-string str1 coding-system)
2621 enc2 (encode-coding-string str2 coding-system)
2623 (while (and (> i1 0) (= (aref enc1 (1- i1)) (aref enc2 (1- i2))))
2624 (setq i1 (1- i1) i2 (1- i2)))
2626 ;; Now (substring enc1 i1) and (substring enc2 i2) are the same,
2627 ;; and they are the extra control sequences at the tail to
2629 (substring enc2 0 i2))))
2632 ;; arch-tag: b382c432-4b36-460e-bf4c-05efd0bb18dc
2633 ;;; mule-cmds.el ends here