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 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 (update-coding-systems-internal)
388 (or (eq coding-category
(car coding-category-list
))
389 ;; We must change the order.
390 (set-coding-priority (list coding-category
)))
391 (if (and base
(interactive-p))
392 (message "Highest priority is set to %s (base of %s)"
394 ;; If they asked for specific EOL conversion, honor that.
395 (if (memq eol-type
'(0 1 2))
397 (coding-system-change-eol-conversion base eol-type
))
398 (setq coding-system base
))
399 (set-default-coding-systems coding-system
)))
401 (defvar sort-coding-systems-predicate nil
402 "If non-nil, a predicate function to sort coding systems.
404 It is called with two coding systems, and should return t if the first
405 one is \"less\" than the second.
407 The function `sort-coding-systems' use it.")
409 (defun sort-coding-systems (codings)
410 "Sort coding system list CODINGS by a priority of each coding system.
412 If a coding system is most preferred, it has the highest priority.
413 Otherwise, a coding system corresponds to some MIME charset has higher
414 priorities. Among them, a coding system included in `coding-system'
415 key of the current language environment has higher priorities. See
416 also the documentation of `language-info-alist'.
418 If the variable `sort-coding-systems-predicate' (which see) is
419 non-nil, it is used to sort CODINGS in the different way than above."
420 (if sort-coding-systems-predicate
421 (sort codings sort-coding-systems-predicate
)
422 (let* ((from-categories (mapcar #'(lambda (x) (symbol-value x
))
423 coding-category-list
))
424 (most-preferred (car from-categories
))
425 (lang-preferred (get-language-info current-language-environment
429 (let ((base (coding-system-base x
)))
430 ;; We calculate the priority number 0..255 by
431 ;; using the 8 bits PMMLCEII as this:
432 ;; P: 1 iff most preferred.
433 ;; MM: greater than 0 iff mime-charset.
434 ;; L: 1 iff one of the current lang. env.'s codings.
435 ;; C: 1 iff one of codings listed in the category list.
436 ;; E: 1 iff not XXX-with-esc
437 ;; II: if iso-2022 based, 0..3, else 1.
439 (lsh (if (eq base most-preferred
) 1 0) 7)
441 (let ((mime (coding-system-get base
'mime-charset
)))
442 ;; Prefer coding systems corresponding to a
445 ;; Lower utf-16 priority so that we
446 ;; normally prefer utf-8 to it, and put
447 ;; x-ctext below that.
448 (cond ((string-match "utf-16"
451 ((string-match "^x-" (symbol-name mime
))
456 (lsh (if (memq base lang-preferred
) 1 0) 4)
457 (lsh (if (memq base from-categories
) 1 0) 3)
458 (lsh (if (string-match "-with-esc\\'"
461 (if (eq (coding-system-type base
) 2)
462 ;; For ISO based coding systems, prefer
463 ;; one that doesn't use escape sequences.
464 (let ((flags (coding-system-flags base
)))
465 (if (or (consp (aref flags
0))
466 (consp (aref flags
1))
467 (consp (aref flags
2))
468 (consp (aref flags
3)))
469 (if (or (aref flags
8) (aref flags
9))
474 (sort codings
(function (lambda (x y
)
475 (> (funcall func x
) (funcall func y
))))))))
477 (defun find-coding-systems-region (from to
)
478 "Return a list of proper coding systems to encode a text between FROM and TO.
479 If FROM is a string, find coding systems in that instead of the buffer.
480 All coding systems in the list can safely encode any multibyte characters
483 If the text contains no multibyte characters, return a list of a single
484 element `undecided'."
485 (let ((codings (find-coding-systems-region-internal from to
)))
487 ;; The text contains only ASCII characters. Any coding
490 ;; We need copy-sequence because sorting will alter the argument.
491 (sort-coding-systems (copy-sequence codings
)))))
493 (defun find-coding-systems-string (string)
494 "Return a list of proper coding systems to encode STRING.
495 All coding systems in the list can safely encode any multibyte characters
498 If STRING contains no multibyte characters, return a list of a single
499 element `undecided'."
500 (find-coding-systems-region string nil
))
502 (defun find-coding-systems-for-charsets (charsets)
503 "Return a list of proper coding systems to encode characters of CHARSETS.
504 CHARSETS is a list of character sets.
505 It actually checks at most the first 96 characters of each charset.
506 So, if a charset of dimension two is included in CHARSETS, the value may
507 contain a coding system that can't encode all characters of the charset."
508 (cond ((or (null charsets
)
509 (and (= (length charsets
) 1)
510 (eq 'ascii
(car charsets
))))
512 ((or (memq 'eight-bit-control charsets
)
513 (memq 'eight-bit-graphic charsets
))
514 '(raw-text emacs-mule
))
518 (while (and codings charsets
)
519 (setq charset
(car charsets
) charsets
(cdr charsets
))
520 (unless (eq charset
'ascii
)
521 (setq str
(make-string 96 32))
522 (if (= (charset-dimension charset
) 1)
523 (if (= (charset-chars charset
) 96)
525 (aset str i
(make-char charset
(+ i
32))))
527 (aset str i
(make-char charset
(+ i
33)))))
528 (if (= (charset-chars charset
) 96)
530 (aset str i
(make-char charset
32 (+ i
32))))
532 (aset str i
(make-char charset
33 (+ i
33))))))
533 (setq l
(find-coding-systems-string str
))
537 (dolist (elt codings
)
539 (setq ll
(cons elt ll
))))
540 (setq codings ll
)))))
543 (defun find-multibyte-characters (from to
&optional maxcount excludes
)
544 "Find multibyte characters in the region specified by FROM and TO.
545 If FROM is a string, find multibyte characters in the string.
546 The return value is an alist of the following format:
547 ((CHARSET COUNT CHAR ...) ...)
549 CHARSET is a character set,
550 COUNT is a number of characters,
551 CHARs are the characters found from the character set.
552 Optional 3rd arg MAXCOUNT limits how many CHARs are put in the above list.
553 Optional 4th arg EXCLUDE is a list of character sets to be ignored.
555 For invalid characters, CHARs are actually strings."
560 (while (setq idx
(string-match "[^\000-\177]" from idx
))
561 (setq char
(aref from idx
)
562 charset
(char-charset char
))
563 (if (eq charset
'unknown
)
564 (setq char
(match-string 0)))
565 (if (or (memq charset
'(unknown
566 eight-bit-control eight-bit-graphic
))
567 (not (or (eq excludes t
) (memq charset excludes
))))
568 (let ((slot (assq charset chars
)))
570 (if (not (memq char
(nthcdr 2 slot
)))
571 (let ((count (nth 1 slot
)))
572 (setcar (cdr slot
) (1+ count
))
573 (if (or (not maxcount
) (< count maxcount
))
574 (nconc slot
(list char
)))))
575 (setq chars
(cons (list charset
1 char
) chars
)))))
576 (setq idx
(1+ idx
))))
579 (while (re-search-forward "[^\000-\177]" to t
)
580 (setq char
(preceding-char)
581 charset
(char-charset char
))
582 (if (eq charset
'unknown
)
583 (setq char
(match-string 0)))
584 (if (or (memq charset
'(unknown eight-bit-control eight-bit-graphic
))
585 (not (or (eq excludes t
) (memq charset excludes
))))
586 (let ((slot (assq charset chars
)))
588 (if (not (member char
(nthcdr 2 slot
)))
589 (let ((count (nth 1 slot
)))
590 (setcar (cdr slot
) (1+ count
))
591 (if (or (not maxcount
) (< count maxcount
))
592 (nconc slot
(list char
)))))
593 (setq chars
(cons (list charset
1 char
) chars
))))))))
597 (defun search-unencodable-char (coding-system)
598 "Search forward from point for a character that is not encodable.
599 It asks which coding system to check.
600 If such a character is found, set point after that character.
601 Otherwise, don't move point.
603 When called from a program, the value is a position of the found character,
604 or nil if all characters are encodable."
606 (list (let ((default (or buffer-file-coding-system
'us-ascii
)))
608 (format "Coding-system (default, %s): " default
)
610 (let ((pos (unencodable-char-position (point) (point-max) coding-system
)))
613 (message "All following characters are encodable by %s" coding-system
))
617 (defvar last-coding-system-specified nil
618 "Most recent coding system explicitly specified by the user when asked.
619 This variable is set whenever Emacs asks the user which coding system
620 to use in order to write a file. If you set it to nil explicitly,
621 then call `write-region', then afterward this variable will be non-nil
622 only if the user was explicitly asked and specified a coding system.")
624 (defvar select-safe-coding-system-accept-default-p nil
625 "If non-nil, a function to control the behaviour of coding system selection.
626 The meaning is the same as the argument ACCEPT-DEFAULT-P of the
627 function `select-safe-coding-system' (which see). This variable
628 overrides that argument.")
630 (defun select-safe-coding-system-interactively (from to codings unsafe
631 &optional rejected default
)
632 "Select interactively a coding system for the region FROM ... TO.
633 FROM can be a string, as in `write-region'.
634 CODINGS is the list of base coding systems known to be safe for this region,
635 typically obtained with `find-coding-systems-region'.
636 UNSAFE is a list of coding systems known to be unsafe for this region.
637 REJECTED is a list of coding systems which were safe but for some reason
638 were not recommended in the particular context.
639 DEFAULT is the coding system to use by default in the query."
640 ;; At first, if some defaults are unsafe, record at most 11
641 ;; problematic characters and their positions for them by turning
644 ;; ((CODING (POS . CHAR) (POS . CHAR) ...) ...)
647 (mapcar #'(lambda (coding)
650 (mapcar #'(lambda (pos)
651 (cons pos
(aref from pos
)))
652 (unencodable-char-position
653 0 (length from
) coding
655 (mapcar #'(lambda (pos)
656 (cons pos
(char-after pos
)))
657 (unencodable-char-position
658 from to coding
11)))))
661 ;; Change each safe coding system to the corresponding
662 ;; mime-charset name if it is also a coding system. Such a name
663 ;; is more friendly to users.
667 (setq mime-charset
(coding-system-get (car l
) 'mime-charset
))
668 (if (and mime-charset
(coding-system-p mime-charset
))
669 (setcar l mime-charset
))
672 ;; Don't offer variations with locking shift, which you
673 ;; basically never want.
675 (dolist (elt codings
(setq codings
(nreverse l
)))
676 (unless (or (eq 'coding-category-iso-7-else
677 (coding-system-category elt
))
678 (eq 'coding-category-iso-8-else
679 (coding-system-category elt
)))
682 ;; Remove raw-text, emacs-mule and no-conversion unless nothing
683 ;; else is available.
687 (delq 'no-conversion codings
)))
688 '(raw-text emacs-mule no-conversion
)))
690 (let ((window-configuration (current-window-configuration))
691 (bufname (buffer-name))
694 ;; If some defaults are unsafe, make sure the offending
695 ;; buffer is displayed.
696 (when (and unsafe
(not (stringp from
)))
697 (pop-to-buffer bufname
)
698 (goto-char (apply 'min
(mapcar #'(lambda (x) (car (cadr x
)))
700 ;; Then ask users to select one from CODINGS while showing
701 ;; the reason why none of the defaults are not used.
702 (with-output-to-temp-buffer "*Warning*"
703 (with-current-buffer standard-output
704 (if (and (null rejected
) (null unsafe
))
705 (insert "No default coding systems to try for "
707 (format "string \"%s\"." from
)
708 (format "buffer `%s'." bufname
)))
710 "These default coding systems were tried to encode"
712 (concat " \"" (if (> (length from
) 10)
713 (concat (substring from
0 10) "...\"")
715 (format " text\nin the buffer `%s'" bufname
))
719 (dolist (x (append rejected unsafe
))
720 (princ " ") (princ (car x
)))
722 (fill-region-as-paragraph pos
(point)))
724 (insert "These safely encodes the target text,
725 but it is not recommended for encoding text in this context,
726 e.g., for sending an email message.\n ")
728 (princ " ") (princ x
))
731 (insert (if rejected
"And the others"
732 "However, each of them")
733 " encountered these problematic characters:\n")
734 (dolist (coding unsafe
)
735 (insert (format " %s:" (car coding
)))
738 #'(lambda (bufname pos
)
739 (when (buffer-live-p (get-buffer bufname
))
740 (pop-to-buffer bufname
)
743 #'(lambda (bufname pos coding
)
744 (when (buffer-live-p (get-buffer bufname
))
745 (pop-to-buffer bufname
)
749 (search-unencodable-char coding
)
750 (forward-char -
1))))))
751 (dolist (elt (cdr coding
))
754 (insert (if (< i
10) (cdr elt
) "..."))
760 "mouse-2, RET: jump to this character"
762 'help-args
(list bufname
(car elt
)))
767 "mouse-2, RET: next unencodable character"
769 'help-args
(list bufname
(car elt
)
774 The first problematic character is at point in the displayed buffer,\n"
775 (substitute-command-keys "\
776 and \\[universal-argument] \\[what-cursor-position] will give information about it.\n"))))
778 one of the following safe coding systems, or edit the buffer:\n")
782 (princ " ") (princ x
))
784 (fill-region-as-paragraph pos
(point)))
785 (insert "Or specify any other coding system
786 at the risk of losing the problematic characters.\n")))
788 ;; Read a coding system.
791 (format "Select coding system (default %s): " default
)
793 (setq last-coding-system-specified coding-system
))
795 (kill-buffer "*Warning*")
796 (set-window-configuration window-configuration
)
799 (defun select-safe-coding-system (from to
&optional default-coding-system
800 accept-default-p file
)
801 "Ask a user to select a safe coding system from candidates.
802 The candidates of coding systems which can safely encode a text
803 between FROM and TO are shown in a popup window. Among them, the most
804 proper one is suggested as the default.
806 The list of `buffer-file-coding-system' of the current buffer,
807 the `default-buffer-file-coding-system', and the
808 most preferred coding system (if it corresponds to a MIME charset) is
809 treated as the default coding system list. Among them, the first one
810 that safely encodes the text is normally selected silently and
811 returned without any user interaction. See also the command
812 `prefer-coding-system'.
814 However, the user is queried if the chosen coding system is
815 inconsistent with what would be selected by `set-auto-coding' from
816 coding cookies &c. if the contents of the region were read from a
817 file. (That could lead to data corruption in a file subsequently
818 re-visited and edited.)
820 Optional 3rd arg DEFAULT-CODING-SYSTEM specifies a coding system or a
821 list of coding systems to be prepended to the default coding system
822 list. However, if DEFAULT-CODING-SYSTEM is a list and the first
823 element is t, the cdr part is used as the defualt coding system list,
824 i.e. `buffer-file-coding-system', `default-buffer-file-coding-system',
825 and the most preferred coding system are not used.
827 Optional 4th arg ACCEPT-DEFAULT-P, if non-nil, is a function to
828 determine the acceptability of the silently selected coding system.
829 It is called with that coding system, and should return nil if it
830 should not be silently selected and thus user interaction is required.
832 Optional 5th arg FILE is the file name to use for this purpose.
833 That is different from `buffer-file-name' when handling `write-region'
836 The variable `select-safe-coding-system-accept-default-p', if
837 non-nil, overrides ACCEPT-DEFAULT-P.
839 Kludgy feature: if FROM is a string, the string is the target text,
841 (if (and default-coding-system
842 (not (listp default-coding-system
)))
843 (setq default-coding-system
(list default-coding-system
)))
845 (let ((no-other-defaults nil
))
846 (if (eq (car default-coding-system
) t
)
847 (setq no-other-defaults t
848 default-coding-system
(cdr default-coding-system
)))
850 ;; Change elements of the list to (coding . base-coding).
851 (setq default-coding-system
852 (mapcar (function (lambda (x) (cons x
(coding-system-base x
))))
853 default-coding-system
))
855 ;; From now on, the list of defaults is reversed.
856 (setq default-coding-system
(nreverse default-coding-system
))
858 (unless no-other-defaults
859 ;; If buffer-file-coding-system is not nil nor undecided, append it
861 (if buffer-file-coding-system
862 (let ((base (coding-system-base buffer-file-coding-system
)))
863 (or (eq base
'undecided
)
864 (rassq base default-coding-system
)
865 (push (cons buffer-file-coding-system base
)
866 default-coding-system
))))
868 ;; If default-buffer-file-coding-system is not nil nor undecided,
869 ;; append it to the defaults.
870 (if default-buffer-file-coding-system
871 (let ((base (coding-system-base default-buffer-file-coding-system
)))
872 (or (eq base
'undecided
)
873 (rassq base default-coding-system
)
874 (push (cons default-buffer-file-coding-system base
)
875 default-coding-system
))))
877 ;; If the most preferred coding system has the property mime-charset,
878 ;; append it to the defaults.
879 (let ((tail coding-category-list
)
881 (while (and tail
(not (setq preferred
(symbol-value (car tail
)))))
882 (setq tail
(cdr tail
)))
883 (and (coding-system-p preferred
)
884 (setq base
(coding-system-base preferred
))
885 (coding-system-get preferred
'mime-charset
)
886 (not (rassq base default-coding-system
))
887 (push (cons preferred base
)
888 default-coding-system
)))))
890 (if select-safe-coding-system-accept-default-p
891 (setq accept-default-p select-safe-coding-system-accept-default-p
))
893 (let ((codings (find-coding-systems-region from to
))
895 safe rejected unsafe
)
896 (if (eq (car codings
) 'undecided
)
897 ;; Any coding system is ok.
898 (setq coding-system t
)
899 ;; Classify the defaults into safe, rejected, and unsafe.
900 (dolist (elt default-coding-system
)
901 (if (memq (cdr elt
) codings
)
902 (if (and (functionp accept-default-p
)
903 (not (funcall accept-default-p
(cdr elt
))))
904 (push (car elt
) rejected
)
905 (push (car elt
) safe
))
906 (push (car elt
) unsafe
)))
908 (setq coding-system
(car safe
))))
910 ;; If all the defaults failed, ask a user.
911 (when (not coding-system
)
912 (setq coding-system
(select-safe-coding-system-interactively
913 from to codings unsafe rejected
(car codings
))))
915 (if (vectorp (coding-system-eol-type coding-system
))
916 (let ((eol (coding-system-eol-type buffer-file-coding-system
)))
919 (coding-system-change-eol-conversion coding-system eol
)))))
921 (if (eq coding-system t
)
922 (setq coding-system buffer-file-coding-system
))
923 ;; Check we're not inconsistent with what `coding:' spec &c would
924 ;; give when file is re-read.
925 ;; But don't do this if we explicitly ignored the cookie
926 ;; by using `find-file-literally'.
927 (unless (or (stringp from
) find-file-literally
)
928 (let ((auto-cs (save-excursion
931 (narrow-to-region from to
)
932 (goto-char (point-min))
933 (set-auto-coding (or file buffer-file-name
"")
935 ;; Merge coding-system and auto-cs as far as possible.
936 (if (not coding-system
)
937 (setq coding-system auto-cs
)
939 (setq auto-cs coding-system
)
940 (let ((eol-type-1 (coding-system-eol-type coding-system
))
941 (eol-type-2 (coding-system-eol-type auto-cs
)))
942 (if (eq (coding-system-base coding-system
) 'undecided
)
943 (setq coding-system
(coding-system-change-text-conversion
944 coding-system auto-cs
))
945 (if (eq (coding-system-base auto-cs
) 'undecided
)
946 (setq auto-cs
(coding-system-change-text-conversion
947 auto-cs coding-system
))))
948 (if (vectorp eol-type-1
)
949 (or (vectorp eol-type-2
)
950 (setq coding-system
(coding-system-change-eol-conversion
951 coding-system eol-type-2
)))
952 (if (vectorp eol-type-2
)
953 (setq auto-cs
(coding-system-change-eol-conversion
954 auto-cs eol-type-1
)))))))
957 ;; Don't barf if writing a compressed file, say.
958 ;; This check perhaps isn't ideal, but is probably
959 ;; the best thing to do.
960 (not (auto-coding-alist-lookup (or file buffer-file-name
"")))
961 (not (coding-system-equal coding-system auto-cs
)))
963 (format "Selected encoding %s disagrees with \
964 %s specified by file contents. Really save (else edit coding cookies \
965 and try again)? " coding-system auto-cs
))
966 (error "Save aborted")))))
969 (setq select-safe-coding-system-function
'select-safe-coding-system
)
971 (defun select-message-coding-system ()
972 "Return a coding system to encode the outgoing message of the current buffer.
973 It at first tries the first coding system found in these variables
975 (1) local value of `buffer-file-coding-system'
976 (2) value of `sendmail-coding-system'
977 (3) value of `default-sendmail-coding-system'
978 (4) value of `default-buffer-file-coding-system'
979 If the found coding system can't encode the current buffer,
980 or none of them are bound to a coding system,
981 it asks the user to select a proper coding system."
982 (let ((coding (or (and (local-variable-p 'buffer-file-coding-system
)
983 buffer-file-coding-system
)
984 sendmail-coding-system
985 default-sendmail-coding-system
986 default-buffer-file-coding-system
)))
987 (if (eq coding
'no-conversion
)
988 ;; We should never use no-conversion for outgoing mail.
990 (if (fboundp select-safe-coding-system-function
)
991 (funcall select-safe-coding-system-function
992 (point-min) (point-max) coding
993 (function (lambda (x) (coding-system-get x
'mime-charset
))))
996 ;;; Language support stuff.
998 (defvar language-info-alist nil
999 "Alist of language environment definitions.
1000 Each element looks like:
1001 (LANGUAGE-NAME . ((KEY . INFO) ...))
1002 where LANGUAGE-NAME is a string, the name of the language environment,
1003 KEY is a symbol denoting the kind of information, and
1004 INFO is the data associated with KEY.
1005 Meaningful values for KEY include
1007 documentation value is documentation of what this language environment
1008 is meant for, and how to use it.
1009 charset value is a list of the character sets used by this
1010 language environment.
1011 sample-text value is an expression which is evalled to generate
1012 a line of text written using characters appropriate
1013 for this language environment.
1014 setup-function value is a function to call to switch to this
1015 language environment.
1016 exit-function value is a function to call to leave this
1017 language environment.
1018 coding-system value is a list of coding systems that are good
1019 for saving text written in this language environment.
1020 This list serves as suggestions to the user;
1021 in effect, as a kind of documentation.
1022 coding-priority value is a list of coding systems for this language
1023 environment, in order of decreasing priority.
1024 This is used to set up the coding system priority
1025 list when you switch to this language environment.
1026 nonascii-translation
1027 value is a translation table to be set in the
1028 variable `nonascii-translation-table' in this
1029 language environment, or a character set from
1030 which `nonascii-insert-offset' is calculated.
1031 input-method value is a default input method for this language
1033 features value is a list of features requested in this
1034 language environment.
1035 ctext-non-standard-encodings
1036 value is a list of non-standard encoding
1037 names used in extended segments of CTEXT.
1039 `ctext-non-standard-encodings' for more
1042 The following keys take effect only when multibyte characters are
1043 globally disabled, i.e. the value of `default-enable-multibyte-characters'
1046 unibyte-syntax value is a library name to load to set
1047 unibyte 8-bit character syntaxes for this
1048 language environment.
1050 unibyte-display value is a coding system to encode characters
1051 for the terminal. Characters in the range
1052 of 160 to 255 display not as octal escapes,
1053 but as non-ASCII characters in this language
1056 (defun get-language-info (lang-env key
)
1057 "Return information listed under KEY for language environment LANG-ENV.
1058 KEY is a symbol denoting the kind of information.
1059 For a list of useful values for KEY and their meanings,
1060 see `language-info-alist'."
1061 (if (symbolp lang-env
)
1062 (setq lang-env
(symbol-name lang-env
)))
1063 (let ((lang-slot (assoc-string lang-env language-info-alist t
)))
1065 (cdr (assq key
(cdr lang-slot
))))))
1067 (defun set-language-info (lang-env key info
)
1068 "Modify part of the definition of language environment LANG-ENV.
1069 Specifically, this stores the information INFO under KEY
1070 in the definition of this language environment.
1071 KEY is a symbol denoting the kind of information.
1072 INFO is the value for that information.
1074 For a list of useful values for KEY and their meanings,
1075 see `language-info-alist'."
1076 (if (symbolp lang-env
)
1077 (setq lang-env
(symbol-name lang-env
)))
1078 (let (lang-slot key-slot
)
1079 (setq lang-slot
(assoc lang-env language-info-alist
))
1080 (if (null lang-slot
) ; If no slot for the language, add it.
1081 (setq lang-slot
(list lang-env
)
1082 language-info-alist
(cons lang-slot language-info-alist
)))
1083 (setq key-slot
(assq key lang-slot
))
1084 (if (null key-slot
) ; If no slot for the key, add it.
1086 (setq key-slot
(list key
))
1087 (setcdr lang-slot
(cons key-slot
(cdr lang-slot
)))))
1088 (setcdr key-slot
(purecopy info
))
1089 ;; Update the custom-type of `current-language-environment'.
1090 (put 'current-language-environment
'custom-type
1091 (cons 'choice
(mapcar
1094 (sort (mapcar 'car language-info-alist
) 'string
<))))))
1096 (defun set-language-info-alist (lang-env alist
&optional parents
)
1097 "Store ALIST as the definition of language environment LANG-ENV.
1098 ALIST is an alist of KEY and INFO values. See the documentation of
1099 `language-info-alist' for the meanings of KEY and INFO.
1101 Optional arg PARENTS is a list of parent menu names; it specifies
1102 where to put this language environment in the
1103 Describe Language Environment and Set Language Environment menus.
1104 For example, (\"European\") means to put this language environment
1105 in the European submenu in each of those two menus."
1106 (if (symbolp lang-env
)
1107 (setq lang-env
(symbol-name lang-env
)))
1108 (let ((describe-map describe-language-environment-map
)
1109 (setup-map setup-language-environment-map
))
1112 map parent-symbol parent prompt
)
1114 (if (symbolp (setq parent-symbol
(car l
)))
1115 (setq parent
(symbol-name parent
))
1116 (setq parent parent-symbol parent-symbol
(intern parent
)))
1117 (setq map
(lookup-key describe-map
(vector parent-symbol
)))
1118 ;; This prompt string is for define-prefix-command, so
1119 ;; that the map it creates will be suitable for a menu.
1120 (or map
(setq prompt
(format "%s Environment" parent
)))
1123 (setq map
(intern (format "describe-%s-environment-map"
1124 (downcase parent
))))
1125 (define-prefix-command map nil prompt
)
1126 (define-key-after describe-map
(vector parent-symbol
)
1127 (cons parent map
) t
)))
1128 (setq describe-map
(symbol-value map
))
1129 (setq map
(lookup-key setup-map
(vector parent-symbol
)))
1132 (setq map
(intern (format "setup-%s-environment-map"
1133 (downcase parent
))))
1134 (define-prefix-command map nil prompt
)
1135 (define-key-after setup-map
(vector parent-symbol
)
1136 (cons parent map
) t
)))
1137 (setq setup-map
(symbol-value map
))
1140 ;; Set up menu items for this language env.
1141 (let ((doc (assq 'documentation alist
)))
1143 (define-key-after describe-map
(vector (intern lang-env
))
1144 (cons lang-env
'describe-specified-language-support
) t
)))
1145 (define-key-after setup-map
(vector (intern lang-env
))
1146 (cons lang-env
'setup-specified-language-environment
) t
)
1149 (set-language-info lang-env
(car (car alist
)) (cdr (car alist
)))
1150 (setq alist
(cdr alist
)))))
1152 (defun read-language-name (key prompt
&optional default
)
1153 "Read a language environment name which has information for KEY.
1154 If KEY is nil, read any language environment.
1155 Prompt with PROMPT. DEFAULT is the default choice of language environment.
1156 This returns a language environment name as a string."
1157 (let* ((completion-ignore-case t
)
1158 (name (completing-read prompt
1161 (function (lambda (elm) (and (listp elm
) (assq key elm
)))))
1162 t nil nil default
)))
1163 (if (and (> (length name
) 0)
1165 (get-language-info name key
)))
1168 ;;; Multilingual input methods.
1170 "LEIM: Libraries of Emacs Input Methods."
1173 (defconst leim-list-file-name
"leim-list.el"
1174 "Name of LEIM list file.
1175 This file contains a list of libraries of Emacs input methods (LEIM)
1176 in the format of Lisp expression for registering each input method.
1177 Emacs loads this file at startup time.")
1179 (defvar leim-list-header
(format
1180 ";;; %s -- list of LEIM (Library of Emacs Input Method) -*-coding: iso-2022-7bit;-*-
1182 ;; This file contains a list of LEIM (Library of Emacs Input Method)
1183 ;; methods in the same directory as this file. Loading this file
1184 ;; registers all the input methods in Emacs.
1186 ;; Each entry has the form:
1187 ;; (register-input-method
1188 ;; INPUT-METHOD LANGUAGE-NAME ACTIVATE-FUNC
1189 ;; TITLE DESCRIPTION
1191 ;; See the function `register-input-method' for the meanings of the arguments.
1193 ;; If this directory is included in load-path, Emacs automatically
1194 ;; loads this file at startup time.
1197 leim-list-file-name
)
1198 "Header to be inserted in LEIM list file.")
1200 (defvar leim-list-entry-regexp
"^(register-input-method"
1201 "Regexp matching head of each entry in LEIM list file.
1202 See also the variable `leim-list-header'")
1204 (defvar update-leim-list-functions
1205 '(quail-update-leim-list-file)
1206 "List of functions to call to update LEIM list file.
1207 Each function is called with one arg, LEIM directory name.")
1209 (defun update-leim-list-file (&rest dirs
)
1210 "Update LEIM list file in directories DIRS."
1211 (let ((functions update-leim-list-functions
))
1213 (apply (car functions
) dirs
)
1214 (setq functions
(cdr functions
)))))
1216 (defvar current-input-method nil
1217 "The current input method for multilingual text.
1218 If nil, that means no input method is activated now.")
1219 (make-variable-buffer-local 'current-input-method
)
1220 (put 'current-input-method
'permanent-local t
)
1222 (defvar current-input-method-title nil
1223 "Title string of the current input method shown in mode line.")
1224 (make-variable-buffer-local 'current-input-method-title
)
1225 (put 'current-input-method-title
'permanent-local t
)
1227 (defcustom default-input-method nil
1228 "*Default input method for multilingual text (a string).
1229 This is the input method activated automatically by the command
1230 `toggle-input-method' (\\[toggle-input-method])."
1231 :link
'(custom-manual "(emacs)Input Methods")
1233 :type
'(choice (const nil
) (string
1234 :completion-ignore-case t
1235 :complete-function widget-string-complete
1236 :completion-alist input-method-alist
1237 :prompt-history input-method-history
))
1238 :set-after
'(current-language-environment))
1240 (put 'input-method-function
'permanent-local t
)
1242 (defvar input-method-history nil
1243 "History list for some commands that read input methods.")
1244 (make-variable-buffer-local 'input-method-history
)
1245 (put 'input-method-history
'permanent-local t
)
1247 (defvar inactivate-current-input-method-function nil
1248 "Function to call for inactivating the current input method.
1249 Every input method should set this to an appropriate value when activated.
1250 This function is called with no argument.
1252 This function should never change the value of `current-input-method'.
1253 It is set to nil by the function `inactivate-input-method'.")
1254 (make-variable-buffer-local 'inactivate-current-input-method-function
)
1255 (put 'inactivate-current-input-method-function
'permanent-local t
)
1257 (defvar describe-current-input-method-function nil
1258 "Function to call for describing the current input method.
1259 This function is called with no argument.")
1260 (make-variable-buffer-local 'describe-current-input-method-function
)
1261 (put 'describe-current-input-method-function
'permanent-local t
)
1263 (defvar input-method-alist nil
1264 "Alist of input method names vs how to use them.
1265 Each element has the form:
1266 (INPUT-METHOD LANGUAGE-ENV ACTIVATE-FUNC TITLE DESCRIPTION ARGS...)
1267 See the function `register-input-method' for the meanings of the elements.")
1269 (defun register-input-method (input-method lang-env
&rest args
)
1270 "Register INPUT-METHOD as an input method for language environment LANG-ENV.
1271 INPUT-METHOD and LANG-ENV are symbols or strings.
1273 The remaining arguments are:
1274 ACTIVATE-FUNC, TITLE, DESCRIPTION, and ARGS...
1275 ACTIVATE-FUNC is a function to call to activate this method.
1276 TITLE is a string to show in the mode line when this method is active.
1277 DESCRIPTION is a string describing this method and what it is good for.
1278 The ARGS, if any, are passed as arguments to ACTIVATE-FUNC.
1279 All told, the arguments to ACTIVATE-FUNC are INPUT-METHOD and the ARGS.
1281 This function is mainly used in the file \"leim-list.el\" which is
1282 created at Emacs build time, registering all Quail input methods
1283 contained in the Emacs distribution.
1285 In case you want to register a new Quail input method by yourself, be
1286 careful to use the same input method title as given in the third
1287 parameter of `quail-define-package'. (If the values are different, the
1288 string specified in this function takes precedence.)
1290 The commands `describe-input-method' and `list-input-methods' need
1291 these duplicated values to show some information about input methods
1292 without loading the relevant Quail packages."
1293 (if (symbolp lang-env
)
1294 (setq lang-env
(symbol-name lang-env
)))
1295 (if (symbolp input-method
)
1296 (setq input-method
(symbol-name input-method
)))
1297 (let ((info (cons lang-env args
))
1298 (slot (assoc input-method input-method-alist
)))
1301 (setq slot
(cons input-method info
))
1302 (setq input-method-alist
(cons slot input-method-alist
)))))
1304 (defun read-input-method-name (prompt &optional default inhibit-null
)
1305 "Read a name of input method from a minibuffer prompting with PROMPT.
1306 If DEFAULT is non-nil, use that as the default,
1307 and substitute it into PROMPT at the first `%s'.
1308 If INHIBIT-NULL is non-nil, null input signals an error.
1310 The return value is a string."
1312 (setq prompt
(format prompt default
)))
1313 (let* ((completion-ignore-case t
)
1314 ;; As it is quite normal to change input method in the
1315 ;; minibuffer, we must enable it even if
1316 ;; enable-recursive-minibuffers is currently nil.
1317 (enable-recursive-minibuffers t
)
1318 ;; This binding is necessary because input-method-history is
1320 (input-method (completing-read prompt input-method-alist
1321 nil t nil
'input-method-history
1323 (if (and input-method
(symbolp input-method
))
1324 (setq input-method
(symbol-name input-method
)))
1325 (if (> (length input-method
) 0)
1328 (error "No valid input method is specified")))))
1330 (defun activate-input-method (input-method)
1331 "Switch to input method INPUT-METHOD for the current buffer.
1332 If some other input method is already active, turn it off first.
1333 If INPUT-METHOD is nil, deactivate any current input method."
1334 (if (and input-method
(symbolp input-method
))
1335 (setq input-method
(symbol-name input-method
)))
1336 (if (and current-input-method
1337 (not (string= current-input-method input-method
)))
1338 (inactivate-input-method))
1339 (unless (or current-input-method
(null input-method
))
1340 (let ((slot (assoc input-method input-method-alist
)))
1342 (error "Can't activate input method `%s'" input-method
))
1343 (setq current-input-method-title nil
)
1344 (let ((func (nth 2 slot
)))
1345 (if (functionp func
)
1346 (apply (nth 2 slot
) input-method
(nthcdr 5 slot
))
1347 (if (and (consp func
) (symbolp (car func
)) (symbolp (cdr func
)))
1349 (require (cdr func
))
1350 (apply (car func
) input-method
(nthcdr 5 slot
)))
1351 (error "Can't activate input method `%s'" input-method
))))
1352 (setq current-input-method input-method
)
1353 (or (stringp current-input-method-title
)
1354 (setq current-input-method-title
(nth 3 slot
)))
1356 (run-hooks 'input-method-activate-hook
)
1357 (force-mode-line-update)))))
1359 (defun inactivate-input-method ()
1360 "Turn off the current input method."
1361 (when current-input-method
1362 (if input-method-history
1363 (unless (string= current-input-method
(car input-method-history
))
1364 (setq input-method-history
1365 (cons current-input-method
1366 (delete current-input-method input-method-history
))))
1367 (setq input-method-history
(list current-input-method
)))
1369 (funcall inactivate-current-input-method-function
)
1371 (run-hooks 'input-method-inactivate-hook
)
1372 (setq current-input-method nil
1373 input-method-function nil
1374 current-input-method-title nil
)
1375 (force-mode-line-update)))))
1377 (defun set-input-method (input-method &optional interactive
)
1378 "Select and activate input method INPUT-METHOD for the current buffer.
1379 This also sets the default input method to the one you specify.
1380 If INPUT-METHOD is nil, this function turns off the input method, and
1381 also causes you to be prompted for a name of an input method the next
1382 time you invoke \\[toggle-input-method].
1383 When called interactively, the optional arg INTERACTIVE is non-nil,
1384 which marks the variable `default-input-method' as set for Custom buffers.
1386 To deactivate the input method interactively, use \\[toggle-input-method].
1387 To deactivate it programmatically, use \\[inactivate-input-method]."
1389 (let* ((default (or (car input-method-history
) default-input-method
)))
1390 (list (read-input-method-name
1391 (if default
"Select input method (default %s): " "Select input method: ")
1394 (activate-input-method input-method
)
1395 (setq default-input-method input-method
)
1397 (customize-mark-as-set 'default-input-method
))
1398 default-input-method
)
1400 (defun toggle-input-method (&optional arg interactive
)
1401 "Enable or disable multilingual text input method for the current buffer.
1402 Only one input method can be enabled at any time in a given buffer.
1404 The normal action is to enable an input method if none was
1405 enabled, and disable the current one otherwise. Which input method
1406 to enable can be determined in various ways--either the one most
1407 recently used, or the one specified by `default-input-method', or
1408 as a last resort by reading the name of an input method in the
1411 With a prefix argument, read an input method name with the minibuffer
1412 and enable that one. The default is the most recent input method specified
1413 \(not including the currently active input method, if any).
1415 When called interactively, the optional arg INTERACTIVE is non-nil,
1416 which marks the variable `default-input-method' as set for Custom buffers."
1418 (interactive "P\np")
1419 (if (and current-input-method
(not arg
))
1420 (inactivate-input-method)
1421 (let ((default (or (car input-method-history
) default-input-method
)))
1422 (if (and arg default
(equal current-input-method default
)
1423 (> (length input-method-history
) 1))
1424 (setq default
(nth 1 input-method-history
)))
1425 (activate-input-method
1426 (if (or arg
(not default
))
1428 (read-input-method-name
1429 (if default
"Input method (default %s): " "Input method: " )
1432 (unless default-input-method
1434 (setq default-input-method current-input-method
)
1436 (customize-mark-as-set 'default-input-method
)))))))
1438 (defun describe-input-method (input-method)
1439 "Describe input method INPUT-METHOD."
1441 (list (read-input-method-name
1442 "Describe input method (default, current choice): ")))
1443 (if (and input-method
(symbolp input-method
))
1444 (setq input-method
(symbol-name input-method
)))
1445 (help-setup-xref (list #'describe-input-method
1446 (or input-method current-input-method
))
1449 (if (null input-method
)
1450 (describe-current-input-method)
1451 (let ((current current-input-method
))
1455 (activate-input-method input-method
)
1456 (describe-current-input-method))
1457 (activate-input-method current
))
1459 (activate-input-method current
)
1460 (help-setup-xref (list #'describe-input-method input-method
)
1462 (with-output-to-temp-buffer (help-buffer)
1463 (let ((elt (assoc input-method input-method-alist
)))
1465 "Input method: %s (`%s' in mode line) for %s\n %s\n"
1466 input-method
(nth 3 elt
) (nth 1 elt
) (nth 4 elt
))))))))))
1468 (defun describe-current-input-method ()
1469 "Describe the input method currently in use.
1470 This is a subroutine for `describe-input-method'."
1471 (if current-input-method
1472 (if (and (symbolp describe-current-input-method-function
)
1473 (fboundp describe-current-input-method-function
))
1474 (funcall describe-current-input-method-function
)
1475 (message "No way to describe the current input method `%s'"
1476 current-input-method
)
1478 (error "No input method is activated now")))
1480 (defun read-multilingual-string (prompt &optional initial-input input-method
)
1481 "Read a multilingual string from minibuffer, prompting with string PROMPT.
1482 The input method selected last time is activated in minibuffer.
1483 If optional second arg INITIAL-INPUT is non-nil, insert it in the minibuffer
1485 Optional 3rd argument INPUT-METHOD specifies the input method
1486 to be activated instead of the one selected last time. It is a symbol
1490 current-input-method
1491 default-input-method
1492 (read-input-method-name "Input method: " nil t
)))
1493 (if (and input-method
(symbolp input-method
))
1494 (setq input-method
(symbol-name input-method
)))
1495 (let ((prev-input-method current-input-method
))
1498 (activate-input-method input-method
)
1499 (read-string prompt initial-input nil nil t
))
1500 (activate-input-method prev-input-method
))))
1502 ;; Variables to control behavior of input methods. All input methods
1503 ;; should react to these variables.
1505 (defcustom input-method-verbose-flag
'default
1506 "*A flag to control extra guidance given by input methods.
1507 The value should be nil, t, `complex-only', or `default'.
1509 The extra guidance is done by showing list of available keys in echo
1510 area. When you use the input method in the minibuffer, the guidance
1511 is shown at the bottom short window (split from the existing window).
1513 If the value is t, extra guidance is always given, if the value is
1514 nil, extra guidance is always suppressed.
1516 If the value is `complex-only', only complex input methods such as
1517 `chinese-py' and `japanese' give extra guidance.
1519 If the value is `default', complex input methods always give extra
1520 guidance, but simple input methods give it only when you are not in
1523 See also the variable `input-method-highlight-flag'."
1524 :type
'(choice (const :tag
"Always" t
) (const :tag
"Never" nil
)
1525 (const complex-only
) (const default
))
1528 (defcustom input-method-highlight-flag t
1529 "*If this flag is non-nil, input methods highlight partially-entered text.
1530 For instance, while you are in the middle of a Quail input method sequence,
1531 the text inserted so far is temporarily underlined.
1532 The underlining goes away when you finish or abort the input method sequence.
1533 See also the variable `input-method-verbose-flag'."
1537 (defvar input-method-activate-hook nil
1538 "Normal hook run just after an input method is activated.
1540 The variable `current-input-method' keeps the input method name
1543 (defvar input-method-inactivate-hook nil
1544 "Normal hook run just after an input method is inactivated.
1546 The variable `current-input-method' still keeps the input method name
1549 (defvar input-method-after-insert-chunk-hook nil
1550 "Normal hook run just after an input method insert some chunk of text.")
1552 (defvar input-method-exit-on-first-char nil
1553 "This flag controls when an input method returns.
1554 Usually, the input method does not return while there's a possibility
1555 that it may find a different translation if a user types another key.
1556 But, it this flag is non-nil, the input method returns as soon as
1557 the current key sequence gets long enough to have some valid translation.")
1559 (defvar input-method-use-echo-area nil
1560 "This flag controls how an input method shows an intermediate key sequence.
1561 Usually, the input method inserts the intermediate key sequence,
1562 or candidate translations corresponding to the sequence,
1563 at point in the current buffer.
1564 But, if this flag is non-nil, it displays them in echo area instead.")
1566 (defvar input-method-exit-on-invalid-key nil
1567 "This flag controls the behaviour of an input method on invalid key input.
1568 Usually, when a user types a key which doesn't start any character
1569 handled by the input method, the key is handled by turning off the
1570 input method temporarily. After that key, the input method is re-enabled.
1571 But, if this flag is non-nil, the input method is never back on.")
1574 (defvar set-language-environment-hook nil
1575 "Normal hook run after some language environment is set.
1577 When you set some hook function here, that effect usually should not
1578 be inherited to another language environment. So, you had better set
1579 another function in `exit-language-environment-hook' (which see) to
1580 cancel the effect.")
1582 (defvar exit-language-environment-hook nil
1583 "Normal hook run after exiting from some language environment.
1584 When this hook is run, the variable `current-language-environment'
1585 is still bound to the language environment being exited.
1587 This hook is mainly used for canceling the effect of
1588 `set-language-environment-hook' (which-see).")
1590 (put 'setup-specified-language-environment
'apropos-inhibit t
)
1592 (defun setup-specified-language-environment ()
1593 "Switch to a specified language environment."
1595 (let (language-name)
1596 (if (and (symbolp last-command-event
)
1597 (or (not (eq last-command-event
'Default
))
1598 (setq last-command-event
'English
))
1599 (setq language-name
(symbol-name last-command-event
)))
1601 (set-language-environment language-name
)
1602 (customize-mark-as-set 'current-language-environment
))
1603 (error "Bogus calling sequence"))))
1605 (defcustom current-language-environment
"English"
1606 "The last language environment specified with `set-language-environment'.
1607 This variable should be set only with \\[customize], which is equivalent
1608 to using the function `set-language-environment'."
1609 :link
'(custom-manual "(emacs)Language Environments")
1610 :set
(lambda (symbol value
) (set-language-environment value
))
1612 (or (car-safe (assoc-string
1613 (if (symbolp current-language-environment
)
1614 (symbol-name current-language-environment
)
1615 current-language-environment
)
1616 language-info-alist t
))
1618 ;; custom type will be updated with `set-language-info'.
1619 :type
(if language-info-alist
1620 (cons 'choice
(mapcar
1623 (sort (mapcar 'car language-info-alist
) 'string
<)))
1625 :initialize
'custom-initialize-default
1628 (defun reset-language-environment ()
1629 "Reset multilingual environment of Emacs to the default status.
1631 The default status is as follows:
1633 The default value of `buffer-file-coding-system' is nil.
1634 The default coding system for process I/O is nil.
1635 The default value for the command `set-terminal-coding-system' is nil.
1636 The default value for the command `set-keyboard-coding-system' is nil.
1638 The order of priorities of coding categories and the coding system
1639 bound to each category are as follows
1640 coding category coding system
1641 --------------------------------------------------
1642 coding-category-iso-8-1 iso-latin-1
1643 coding-category-iso-8-2 iso-latin-1
1644 coding-category-utf-8 mule-utf-8
1645 coding-category-utf-16-be mule-utf-16be-with-signature
1646 coding-category-utf-16-le mule-utf-16le-with-signature
1647 coding-category-iso-7-tight iso-2022-jp
1648 coding-category-iso-7 iso-2022-7bit
1649 coding-category-iso-7-else iso-2022-7bit-lock
1650 coding-category-iso-8-else iso-2022-8bit-ss2
1651 coding-category-emacs-mule emacs-mule
1652 coding-category-raw-text raw-text
1653 coding-category-sjis japanese-shift-jis
1654 coding-category-big5 chinese-big5
1655 coding-category-ccl nil
1656 coding-category-binary no-conversion"
1658 ;; This function formerly set default-enable-multibyte-characters to t,
1659 ;; but that is incorrect. It should not alter the unibyte/multibyte choice.
1661 (setq coding-category-iso-7-tight
'iso-2022-jp
1662 coding-category-iso-7
'iso-2022-7bit
1663 coding-category-iso-8-1
'iso-latin-1
1664 coding-category-iso-8-2
'iso-latin-1
1665 coding-category-iso-7-else
'iso-2022-7bit-lock
1666 coding-category-iso-8-else
'iso-2022-8bit-ss2
1667 coding-category-emacs-mule
'emacs-mule
1668 coding-category-raw-text
'raw-text
1669 coding-category-sjis
'japanese-shift-jis
1670 coding-category-big5
'chinese-big5
1671 coding-category-utf-16-be
'mule-utf-16be-with-signature
1672 coding-category-utf-16-le
'mule-utf-16le-with-signature
1673 coding-category-utf-8
'mule-utf-8
1674 coding-category-ccl nil
1675 coding-category-binary
'no-conversion
)
1677 (set-coding-priority
1678 '(coding-category-iso-8-1
1679 coding-category-iso-8-2
1680 coding-category-utf-8
1681 coding-category-utf-16-be
1682 coding-category-utf-16-le
1683 coding-category-iso-7-tight
1684 coding-category-iso-7
1685 coding-category-iso-7-else
1686 coding-category-iso-8-else
1687 coding-category-emacs-mule
1688 coding-category-raw-text
1689 coding-category-sjis
1690 coding-category-big5
1692 coding-category-binary
))
1694 (update-coding-systems-internal)
1696 (set-default-coding-systems nil
)
1697 (setq default-sendmail-coding-system
'iso-latin-1
)
1698 ;; On Darwin systems, this should be utf-8, but when this file is loaded
1699 ;; utf-8 is not yet defined, so we set it in set-locale-environment instead.
1700 (setq default-file-name-coding-system
'iso-latin-1
)
1701 ;; Preserve eol-type from existing default-process-coding-systems.
1702 ;; On non-unix-like systems in particular, these may have been set
1703 ;; carefully by the user, or by the startup code, to deal with the
1704 ;; users shell appropriately, so should not be altered by changing
1705 ;; language environment.
1706 (let ((output-coding
1707 ;; When bootstrapping, coding-systems are not defined yet, so
1708 ;; we need to catch the error from check-coding-system.
1710 (coding-system-change-text-conversion
1711 (car default-process-coding-system
) 'undecided
)
1712 (coding-system-error 'undecided
)))
1715 (coding-system-change-text-conversion
1716 (cdr default-process-coding-system
) 'iso-latin-1
)
1717 (coding-system-error 'iso-latin-1
))))
1718 (setq default-process-coding-system
1719 (cons output-coding input-coding
)))
1721 ;; Don't alter the terminal and keyboard coding systems here.
1722 ;; The terminal still supports the same coding system
1723 ;; that it supported a minute ago.
1724 ;; (set-terminal-coding-system-internal nil)
1725 ;; (set-keyboard-coding-system-internal nil)
1727 (setq nonascii-translation-table nil
1728 nonascii-insert-offset
0)
1730 ;; Don't invoke fontset-related functions if fontsets aren't
1731 ;; supported in this build of Emacs.
1732 (and (fboundp 'fontset-list
)
1733 (set-overriding-fontspec-internal nil
)))
1735 (reset-language-environment)
1737 (defun set-display-table-and-terminal-coding-system (language-name)
1738 "Set up the display table and terminal coding system for LANGUAGE-NAME."
1739 (let ((coding (get-language-info language-name
'unibyte-display
)))
1741 (standard-display-european-internal)
1742 ;; The following 2 lines undo the 8-bit display that we set up
1743 ;; in standard-display-european-internal, which see. This is in
1744 ;; case the user has used standard-display-european earlier in
1745 ;; this session. (The MS-DOS port doesn't use that setup, so it
1746 ;; doesn't need to undo it.)
1747 (when standard-display-table
1749 (aset standard-display-table
(+ i
128) nil
))))
1750 (or (eq window-system
'pc
)
1751 (set-terminal-coding-system coding
))))
1753 (defun set-language-environment (language-name)
1754 "Set up multi-lingual environment for using LANGUAGE-NAME.
1755 This sets the coding system priority and the default input method
1756 and sometimes other things. LANGUAGE-NAME should be a string
1757 which is the name of a language environment. For example, \"Latin-1\"
1758 specifies the character set for the major languages of Western Europe."
1759 (interactive (list (read-language-name
1761 "Set language environment (default, English): ")))
1763 (if (symbolp language-name
)
1764 (setq language-name
(symbol-name language-name
)))
1765 (setq language-name
"English"))
1766 (let ((slot (assoc-string language-name language-info-alist t
)))
1768 (error "Language environment not defined: %S" language-name
))
1769 (setq language-name
(car slot
)))
1770 (if current-language-environment
1771 (let ((func (get-language-info current-language-environment
1773 (run-hooks 'exit-language-environment-hook
)
1774 (if (functionp func
) (funcall func
))))
1775 (let ((default-eol-type (coding-system-eol-type
1776 default-buffer-file-coding-system
)))
1777 (reset-language-environment)
1779 ;; The features might set up coding systems.
1780 (let ((required-features (get-language-info language-name
'features
)))
1781 (while required-features
1782 (require (car required-features
))
1783 (setq required-features
(cdr required-features
))))
1785 (setq current-language-environment language-name
)
1786 (set-language-environment-coding-systems language-name default-eol-type
))
1787 (let ((input-method (get-language-info language-name
'input-method
)))
1789 (setq default-input-method input-method
)
1790 (if input-method-history
1791 (setq input-method-history
1793 (delete input-method input-method-history
))))))
1794 (let ((nonascii (get-language-info language-name
'nonascii-translation
))
1796 (if (eq window-system
'pc
)
1798 (format "cp%d-nonascii-translation-table" dos-codepage
)))))
1800 ((char-table-p nonascii
)
1801 (setq nonascii-translation-table nonascii
))
1802 ((and (eq window-system
'pc
) (boundp dos-table
))
1803 ;; DOS terminals' default is to use a special non-ASCII translation
1804 ;; table as appropriate for the installed codepage.
1805 (setq nonascii-translation-table
(symbol-value dos-table
)))
1806 ((charsetp nonascii
)
1807 (setq nonascii-insert-offset
(- (make-char nonascii
) 128)))))
1809 ;; Unibyte setups if necessary.
1810 (unless default-enable-multibyte-characters
1811 ;; Syntax and case table.
1812 (let ((syntax (get-language-info language-name
'unibyte-syntax
)))
1814 (let ((set-case-syntax-set-multibyte nil
))
1815 (load syntax nil t
))
1816 ;; No information for syntax and case. Reset to the defaults.
1817 (let ((syntax-table (standard-syntax-table))
1818 (case-table (standard-case-table))
1819 (ch (if (eq window-system
'pc
) 128 160)))
1821 (modify-syntax-entry ch
" " syntax-table
)
1822 (aset case-table ch ch
)
1824 (set-char-table-extra-slot case-table
0 nil
)
1825 (set-char-table-extra-slot case-table
1 nil
)
1826 (set-char-table-extra-slot case-table
2 nil
))
1827 (set-standard-case-table (standard-case-table))
1828 (let ((list (buffer-list)))
1830 (with-current-buffer (car list
)
1831 (set-case-table (standard-case-table)))
1832 (setq list
(cdr list
))))))
1833 (set-display-table-and-terminal-coding-system language-name
))
1835 (let ((required-features (get-language-info language-name
'features
)))
1836 (while required-features
1837 (require (car required-features
))
1838 (setq required-features
(cdr required-features
))))
1840 ;; Don't invoke fontset-related functions if fontsets aren't
1841 ;; supported in this build of Emacs.
1842 (when (fboundp 'fontset-list
)
1843 (let ((overriding-fontspec (get-language-info language-name
1844 'overriding-fontspec
)))
1845 (if overriding-fontspec
1846 (set-overriding-fontspec-internal overriding-fontspec
))))
1848 (let ((func (get-language-info language-name
'setup-function
)))
1849 (if (functionp func
)
1851 (if (and utf-translate-cjk-mode
1852 utf-translate-cjk-lang-env
1853 (not (eq utf-translate-cjk-lang-env language-name
))
1855 (dolist (charset (get-language-info language-name
'charset
))
1856 (if (memq charset utf-translate-cjk-charsets
)
1859 (utf-translate-cjk-load-tables))
1860 (run-hooks 'set-language-environment-hook
)
1861 (force-mode-line-update t
))
1863 (defun standard-display-european-internal ()
1864 ;; Actually set up direct output of non-ASCII characters.
1865 (standard-display-8bit (if (eq window-system
'pc
) 128 160) 255)
1866 ;; Unibyte Emacs on MS-DOS wants to display all 8-bit characters with
1867 ;; the native font, and codes 160 and 146 stand for something very
1869 (or (and (eq window-system
'pc
) (not default-enable-multibyte-characters
))
1871 ;; Most X fonts used to do the wrong thing for latin-1 code 160.
1872 (unless (and (eq window-system
'x
)
1873 ;; XFree86 4 has fixed the fonts.
1874 (string= "The XFree86 Project, Inc" (x-server-vendor))
1875 (> (aref (number-to-string (nth 2 (x-server-version))) 0)
1877 ;; Make non-line-break space display as a plain space.
1878 (aset standard-display-table
160 [32]))
1879 ;; Most Windows programs send out apostrophes as \222. Most X fonts
1880 ;; don't contain a character at that position. Map it to the ASCII
1881 ;; apostrophe. [This is actually RIGHT SINGLE QUOTATION MARK,
1882 ;; U+2019, normally from the windows-1252 character set. XFree 4
1883 ;; fonts probably have the appropriate glyph at this position,
1884 ;; so they could use standard-display-8bit. It's better to use a
1885 ;; proper windows-1252 coding system. --fx]
1886 (aset standard-display-table 146 [39]))))
1888 (defun set-language-environment-coding-systems (language-name
1890 "Do various coding system setups for language environment LANGUAGE-NAME.
1892 The optional arg EOL-TYPE specifies the eol-type of the default value
1893 of `buffer-file-coding-system' set by this function."
1894 (let* ((priority (get-language-info language-name 'coding-priority))
1895 (default-coding (car priority)))
1897 (let ((categories (mapcar 'coding-system-category priority)))
1898 (set-default-coding-systems
1899 (if (memq eol-type '(0 1 2 unix dos mac))
1900 (coding-system-change-eol-conversion default-coding eol-type)
1902 (setq default-sendmail-coding-system default-coding)
1903 (set-coding-priority categories)
1905 (set (car categories) (car priority))
1906 (setq priority (cdr priority) categories (cdr categories)))
1907 (update-coding-systems-internal)))))
1909 (defsubst princ-list (&rest args)
1910 "Print all arguments with `princ', then print \"\n\"."
1911 (while args (princ (car args)) (setq args (cdr args)))
1914 (put 'describe-specified-language-support 'apropos-inhibit t)
1916 ;; Print language-specific information such as input methods,
1917 ;; charsets, and coding systems. This function is intended to be
1918 ;; called from the menu:
1919 ;; [menu-bar mule describe-language-environment LANGUAGE]
1920 ;; and should not run it by `M-x describe-current-input-method-function'.
1921 (defun describe-specified-language-support ()
1922 "Describe how Emacs supports the specified language environment."
1924 (let (language-name)
1925 (if (not (and (symbolp last-command-event)
1926 (or (not (eq last-command-event 'Default))
1927 (setq last-command-event 'English))
1928 (setq language-name (symbol-name last-command-event))))
1929 (error "Bogus calling sequence"))
1930 (describe-language-environment language-name)))
1932 (defun describe-language-environment (language-name)
1933 "Describe how Emacs supports language environment LANGUAGE-NAME."
1935 (list (read-language-name
1937 "Describe language environment (default, current choice): ")))
1938 (if (null language-name)
1939 (setq language-name current-language-environment))
1940 (if (or (null language-name)
1941 (null (get-language-info language-name 'documentation)))
1942 (error "No documentation for the specified language"))
1943 (if (symbolp language-name)
1944 (setq language-name (symbol-name language-name)))
1945 (dolist (feature (get-language-info language-name 'features))
1947 (let ((doc (get-language-info language-name 'documentation)))
1948 (help-setup-xref (list #'describe-language-environment language-name)
1950 (with-output-to-temp-buffer (help-buffer)
1952 (set-buffer standard-output)
1953 (insert language-name " language environment\n\n")
1955 (insert doc "\n\n"))
1957 (let ((str (eval (get-language-info language-name 'sample-text))))
1959 (insert "Sample text:\n " str "\n\n")))
1961 (let ((input-method (get-language-info language-name 'input-method))
1962 (l (copy-sequence input-method-alist)))
1963 (insert "Input methods")
1965 (insert " (default, " input-method ")")
1966 (setq input-method (assoc input-method input-method-alist))
1967 (setq l (cons input-method (delete input-method l))))
1970 (when (string= language-name (nth 1 (car l)))
1971 (insert " " (car (car l)))
1972 (search-backward (car (car l)))
1973 (help-xref-button 0 'help-input-method (car (car l)))
1974 (goto-char (point-max))
1976 (if (stringp (nth 3 (car l)))
1978 (car (nth 3 (car l))))
1979 "\" in mode line)\n"))
1982 (insert "Character sets:\n")
1983 (let ((l (get-language-info language-name 'charset)))
1985 (insert " nothing specific to " language-name "\n")
1987 (insert " " (symbol-name (car l)))
1988 (search-backward (symbol-name (car l)))
1989 (help-xref-button 0 'help-character-set (car l))
1990 (goto-char (point-max))
1991 (insert ": " (charset-description (car l)) "\n")
1994 (insert "Coding systems:\n")
1995 (let ((l (get-language-info language-name 'coding-system)))
1997 (insert " nothing specific to " language-name "\n")
1999 (insert " " (symbol-name (car l)))
2000 (search-backward (symbol-name (car l)))
2001 (help-xref-button 0 'help-coding-system (car l))
2002 (goto-char (point-max))
2004 (coding-system-mnemonic (car l))
2005 "' in mode line):\n\t"
2006 (coding-system-doc-string (car l))
2008 (let ((aliases (coding-system-get (car l)
2009 'alias-coding-systems)))
2011 (insert "\t(alias:")
2013 (insert " " (symbol-name (car aliases)))
2014 (setq aliases (cdr aliases)))
2016 (setq l (cdr l)))))))))
2020 (defvar locale-translation-file-name nil
2021 "File name for the system's file of locale-name aliases, or nil if none.")
2023 ;; The following definitions might as well be marked as constants and
2024 ;; purecopied, since they're normally used on startup, and probably
2025 ;; should reflect the facilities of the base Emacs.
2026 (defconst locale-language-names
2029 ;; Locale names of the form LANGUAGE[_TERRITORY][.CODESET][@MODIFIER]
2030 ;; as specified in the Single Unix Spec, Version 2.
2031 ;; LANGUAGE is a language code taken from ISO 639:1988 (E/F)
2032 ;; with additions from ISO 639/RA Newsletter No.1/1989;
2033 ;; see Internet RFC 2165 (1997-06) and
2034 ;; http://www.evertype.com/standards/iso639/iso639-en.html
2035 ;; TERRITORY is a country code taken from ISO 3166
2036 ;; http://www.din.de/gremien/nas/nabd/iso3166ma/codlstp1/en_listp1.html.
2037 ;; CODESET and MODIFIER are implementation-dependent.
2039 ;; jasonr comments: MS Windows uses three letter codes for
2040 ;; languages instead of the two letter ISO codes that POSIX
2041 ;; uses. In most cases the first two letters are the same, so
2042 ;; most of the regexps in locale-language-names work. Japanese
2043 ;; and Chinese are exceptions, which are listed in the
2044 ;; non-standard section at the bottom of locale-language-names.
2048 ("af" . "Latin-1") ; Afrikaans
2049 ("am" . "Ethiopic") ; Amharic
2050 ; ar Arabic glibc uses 8859-6
2055 ("be" . "Belarusian") ; Belarusian [Byelorussian until early 1990s]
2056 ("bg" . "Bulgarian") ; Bulgarian
2059 ; bn Bengali, Bangla
2061 ("br" . "Latin-1") ; Breton
2062 ("bs" . "Latin-2") ; Bosnian
2063 ("ca" . "Latin-1") ; Catalan
2066 ("cy" . "Welsh") ; Welsh [glibc uses Latin-8. Did this change?]
2067 ("da" . "Latin-1") ; Danish
2071 ;; Users who specify "en" explicitly typically want Latin-1, not ASCII.
2072 ;; That's actually what the GNU locales define, modulo things like
2074 ("en" . "Latin-1") ; English
2075 ("eo" . "Latin-3") ; Esperanto
2077 ("et" . "Latin-4") ; Estonian
2078 ("eu" . "Latin-1") ; Basque
2079 ; fa Persian glibc uses utf-8
2080 ("fi" . "Latin-1") ; Finnish
2082 ("fo" . "Latin-1") ; Faroese
2083 ("fr" . "French") ; French
2084 ("fy" . "Latin-1") ; Frisian
2085 ("ga" . "Latin-1") ; Irish Gaelic (new orthography)
2086 ("gd" . "Latin-1") ; Scots Gaelic
2087 ("gl" . "Latin-1") ; Galician
2090 ("gv" . "Latin-8") ; Manx Gaelic glibc uses 8859-1
2093 ("hi" . "Devanagari") ; Hindi glibc uses utf-8
2094 ("hr" . "Croatian") ; Croatian
2095 ("hu" . "Latin-2") ; Hungarian
2098 ("id" . "Latin-1") ; Indonesian
2101 ("is" . "Latin-1") ; Icelandic
2102 ("it" . "Italian") ; Italian
2106 ("ka" . "Georgian") ; Georgian
2108 ("kl" . "Latin-1") ; Greenlandic
2114 ("kw" . "Latin-1") ; Cornish
2116 ("la" . "Latin-1") ; Latin
2117 ("lb" . "Latin-1") ; Luxemburgish
2119 ("lo" . "Lao") ; Laothian
2120 ("lt" . "Lithuanian")
2121 ("lv" . "Latvian") ; Latvian, Lettish
2123 ("mi" . "Latin-7") ; Maori
2124 ("mk" . "Cyrillic-ISO") ; Macedonian
2128 ("mr" . "Devanagari") ; Marathi glibc uses utf-8
2129 ("ms" . "Latin-1") ; Malay
2130 ("mt" . "Latin-3") ; Maltese
2133 ("ne" . "Devanagari") ; Nepali
2135 ("no" . "Latin-1") ; Norwegian
2136 ("oc" . "Latin-1") ; Occitan
2140 ("pl" . "Latin-2") ; Polish
2142 ("pt" . "Latin-1") ; Portuguese
2144 ("rm" . "Latin-1") ; Rhaeto-Romanic
2147 ("ru.*[_.]koi8" . "Russian")
2148 ("ru" . "Cyrillic-ISO") ; Russian
2150 ("sa" . "Devanagari") ; Sanskrit
2154 ("sh" . "Latin-2") ; Serbo-Croatian
2157 ("sl" . "Slovenian")
2161 ("sq" . "Latin-1") ; Albanian
2162 ("sr" . "Latin-2") ; Serbian (Latin alphabet)
2163 ("sr_YU@cyrillic" . "Cyrillic-ISO") ; per glibc
2167 ("sv" . "Swedish") ; Swedish
2168 ("sw" . "Latin-1") ; Swahili
2169 ; ta Tamil glibc uses utf-8
2170 ; te Telugu glibc uses utf-8
2175 ("tl" . "Latin-1") ; Tagalog
2183 ("uk" . "Ukrainian") ; Ukrainian
2184 ; ur Urdu glibc uses utf-8
2185 ("uz" . "Latin-1") ; Uzbek
2186 ("vi" . "Vietnamese") ; glibc uses utf-8
2188 ("wa" . "Latin-1") ; Walloon
2191 ("yi" . "Windows-1255") ; Yiddish
2196 ; zh_CN.GB18030/GB18030 \
2198 ; zh_HK/BIG5-HKSCS \
2200 ("zh.*[._]big5" . "Chinese-BIG5")
2201 ("zh.*[._]gbk" . nil) ; Solaris 2.7; has gbk-0 as well as GB 2312.1980-0
2202 ("zh_tw" . "Chinese-CNS") ; glibc uses big5
2203 ("zh_tw[._]euc-tw" . "Chinese-EUC-TW")
2204 ("zh" . "Chinese-GB")
2207 ;; ISO standard locales
2209 ("posix$" . "ASCII")
2211 ;; The "IPA" Emacs language environment does not correspond
2212 ;; to any ISO 639 code, so let it stand for itself.
2215 ;; Nonstandard or obsolete language codes
2216 ("cz" . "Czech") ; e.g. Solaris 2.6
2217 ("ee" . "Latin-4") ; Estonian, e.g. X11R6.4
2218 ("iw" . "Hebrew") ; e.g. X11R6.4
2219 ("sp" . "Cyrillic-ISO") ; Serbian (Cyrillic alphabet), e.g. X11R6.4
2220 ("su" . "Latin-1") ; Finnish, e.g. Solaris 2.6
2221 ("jp" . "Japanese") ; e.g. MS Windows
2222 ("chs" . "Chinese-GB") ; MS Windows Chinese Simplified
2223 ("cht" . "Chinese-BIG5") ; MS Windows Chinese Traditional
2225 "List of pairs of locale regexps and language names.
2226 The first element whose locale regexp matches the start of a downcased locale
2227 specifies the language name corresponding to that locale.
2228 If the language name is nil, there is no corresponding language environment.")
2230 (defconst locale-charset-language-names
2232 '((".*8859[-_]?1\\>" . "Latin-1")
2233 (".*8859[-_]?2\\>" . "Latin-2")
2234 (".*8859[-_]?3\\>" . "Latin-3")
2235 (".*8859[-_]?4\\>" . "Latin-4")
2236 (".*8859[-_]?9\\>" . "Latin-5")
2237 (".*8859[-_]?14\\>" . "Latin-8")
2238 (".*8859[-_]?15\\>" . "Latin-9")
2239 (".*utf\\(?:-?8\\)?\\>" . "UTF-8")
2240 ;; utf-8@euro exists, so put this last. (@euro really specifies
2241 ;; the currency, rather than the charset.)
2242 (".*@euro\\>" . "Latin-9")))
2243 "List of pairs of locale regexps and charset language names.
2244 The first element whose locale regexp matches the start of a downcased locale
2245 specifies the language name whose charset corresponds to that locale.
2246 This language name is used if its charsets disagree with the charsets of
2247 the language name that would otherwise be used for this locale.")
2249 (defconst locale-preferred-coding-systems
2251 '(("ja.*[._]euc" . japanese-iso-8bit)
2252 ("ja.*[._]jis7" . iso-2022-jp)
2253 ("ja.*[._]pck" . japanese-shift-jis)
2254 ("ja.*[._]sjis" . japanese-shift-jis)
2255 ("jpn" . japanese-shift-jis) ; MS-Windows uses this.
2256 (".*[._]utf" . utf-8)))
2257 "List of pairs of locale regexps and preferred coding systems.
2258 The first element whose locale regexp matches the start of a downcased locale
2259 specifies the coding system to prefer when using that locale.")
2261 (defun locale-name-match (key alist)
2262 "Search for KEY in ALIST, which should be a list of regexp-value pairs.
2263 Return the value corresponding to the first regexp that matches the
2264 start of KEY, or nil if there is no match."
2266 (while (and alist (not element))
2267 (if (string-match (concat "\\`\\(?:" (car (car alist)) "\\)") key)
2268 (setq element (car alist)))
2269 (setq alist (cdr alist)))
2272 (defun locale-charset-match-p (charset1 charset2)
2273 "Whether charset names (strings) CHARSET1 and CHARSET2 are equivalent.
2274 Matching is done ignoring case and any hyphens and underscores in the
2275 names. E.g. `ISO_8859-1' and `iso88591' both match `iso-8859-1'."
2276 (setq charset1 (replace-regexp-in-string "[-_]" "" charset1))
2277 (setq charset2 (replace-regexp-in-string "[-_]" "" charset2))
2278 (eq t (compare-strings charset1 nil nil charset2 nil nil t)))
2280 (defvar locale-charset-alist nil
2281 "Coding system alist keyed on locale-style charset name.
2282 Used by `locale-charset-to-coding-system'.")
2284 (defun locale-charset-to-coding-system (charset)
2285 "Find coding system corresponding to CHARSET.
2286 CHARSET is any sort of non-Emacs charset name, such as might be used
2287 in a locale codeset, or elsewhere. It is matched to a coding system
2288 first by case-insensitive lookup in `locale-charset-alist'. Then
2289 matches are looked for in the coding system list, treating case and
2290 the characters `-' and `_' as insignificant. The coding system base
2291 is returned. Thus, for instance, if charset \"ISO8859-2\",
2292 `iso-latin-2' is returned."
2293 (or (car (assoc-string charset locale-charset-alist t))
2294 (let ((cs coding-system-alist)
2296 (while (and (not c) cs)
2297 (if (locale-charset-match-p charset (caar cs))
2298 (setq c (intern (caar cs)))
2300 (if c (coding-system-base c)))))
2302 ;; Fixme: This ought to deal with the territory part of the locale
2303 ;; too, for setting things such as calendar holidays, ps-print paper
2304 ;; size, spelling dictionary.
2306 (defun set-locale-environment (&optional locale-name)
2307 "Set up multi-lingual environment for using LOCALE-NAME.
2308 This sets the language environment, the coding system priority,
2309 the default input method and sometimes other things.
2311 LOCALE-NAME should be a string which is the name of a locale supported
2312 by the system. Often it is of the form xx_XX.CODE, where xx is a
2313 language, XX is a country, and CODE specifies a character set and
2314 coding system. For example, the locale name \"ja_JP.EUC\" might name
2315 a locale for Japanese in Japan using the `japanese-iso-8bit'
2316 coding-system. The name may also have a modifier suffix, e.g. `@euro'
2319 If LOCALE-NAME is nil, its value is taken from the environment
2320 variables LC_ALL, LC_CTYPE and LANG (the first one that is set).
2322 The locale names supported by your system can typically be found in a
2323 directory named `/usr/share/locale' or `/usr/lib/locale'. LOCALE-NAME
2324 will be translated according to the table specified by
2325 `locale-translation-file-name'.
2327 See also `locale-charset-language-names', `locale-language-names',
2328 `locale-preferred-coding-systems' and `locale-coding-system'."
2329 (interactive "sSet environment for locale: ")
2331 ;; Do this at runtime for the sake of binaries possibly transported
2332 ;; to a system without X.
2333 (setq locale-translation-file-name
2335 '("/usr/lib/X11/locale/locale.alias" ; e.g. X11R6.4
2336 "/usr/X11R6/lib/X11/locale/locale.alias" ; XFree86, e.g. RedHat 4.2
2337 "/usr/openwin/lib/locale/locale.alias" ; e.g. Solaris 2.6
2339 ;; The following name appears after the X-related names above,
2340 ;; since the X-related names are what X actually uses.
2341 "/usr/share/locale/locale.alias" ; GNU/Linux sans X
2343 (while (and files (not (file-exists-p (car files))))
2344 (setq files (cdr files)))
2347 (let ((locale locale-name))
2350 ;; Use the first of these three environment variables
2351 ;; that has a nonempty value.
2352 (let ((vars '("LC_ALL" "LC_CTYPE" "LANG")))
2354 (= 0 (length locale))) ; nil or empty string
2355 (setq locale (getenv (pop vars))))))
2359 ;; Translate "swedish" into "sv_SE.ISO8859-1", and so on,
2360 ;; using the translation file that many systems have.
2361 (when locale-translation-file-name
2363 (insert-file-contents locale-translation-file-name)
2364 (when (re-search-forward
2365 (concat "^" (regexp-quote locale) ":?[ \t]+") nil t)
2366 (setq locale (buffer-substring (point) (line-end-position))))))
2368 ;; Leave the system locales alone if the caller did not specify
2369 ;; an explicit locale name, as their defaults are set from
2370 ;; LC_MESSAGES and LC_TIME, not LC_CTYPE, and the user might not
2371 ;; want to set them to the same value as LC_CTYPE.
2373 (setq system-messages-locale locale)
2374 (setq system-time-locale locale))
2376 (setq locale (downcase locale))
2378 (let ((language-name
2379 (locale-name-match locale locale-language-names))
2380 (charset-language-name
2381 (locale-name-match locale locale-charset-language-names))
2383 (or (locale-name-match locale locale-preferred-coding-systems)
2385 (if (string-match "\\.\\([^@]+\\)" locale)
2386 (locale-charset-to-coding-system
2387 (match-string 1 locale)))))))
2389 ;; Give preference to charset-language-name over language-name.
2390 (if (and charset-language-name
2392 (equal (get-language-info language-name 'charset)
2393 (get-language-info charset-language-name 'charset))))
2394 (setq language-name charset-language-name))
2398 ;; Set up for this character set. This is now the right way
2399 ;; to do it for both unibyte and multibyte modes.
2400 (set-language-environment language-name)
2402 ;; If default-enable-multibyte-characters is nil,
2403 ;; we are using single-byte characters,
2404 ;; so the display table and terminal coding system are irrelevant.
2405 (when default-enable-multibyte-characters
2406 (set-display-table-and-terminal-coding-system language-name))
2408 ;; Set the `keyboard-coding-system' if appropriate (tty
2409 ;; only). At least X and MS Windows can generate
2410 ;; multilingual input.
2411 (unless window-system
2412 (let ((kcs (or coding-system
2413 (car (get-language-info language-name
2415 (if kcs (set-keyboard-coding-system kcs))))
2417 (setq locale-coding-system
2418 (car (get-language-info language-name 'coding-priority))))
2421 (prefer-coding-system coding-system)
2422 (setq locale-coding-system coding-system))))
2424 ;; On Windows, override locale-coding-system,
2425 ;; keyboard-coding-system with system codepage. Note:
2426 ;; selection-coding-system is already set in w32select.c.
2427 (when (boundp 'w32-ansi-code-page)
2428 (let ((code-page-coding (intern (format "cp%d" w32-ansi-code-page))))
2429 (when (coding-system-p code-page-coding)
2430 (setq locale-coding-system code-page-coding)
2431 (set-keyboard-coding-system code-page-coding)
2432 (set-terminal-coding-system code-page-coding))))
2434 ;; On Darwin, file names are always encoded in utf-8, no matter the locale.
2435 (when (eq system-type 'darwin)
2436 (setq default-file-name-coding-system 'utf-8))
2438 ;; Default to A4 paper if we're not in a C, POSIX or US locale.
2439 ;; (See comments in Flocale_info.)
2440 (let ((locale locale)
2441 (paper (locale-info 'paper)))
2443 ;; This will always be null at the time of writing.
2445 ((equal paper '(216 279))
2446 (setq ps-paper-type 'letter))
2447 ((equal paper '(210 297))
2448 (setq ps-paper-type 'a4)))
2449 (let ((vars '("LC_ALL" "LC_PAPER" "LANG")))
2450 (while (and vars (= 0 (length locale)))
2451 (setq locale (getenv (pop vars)))))
2453 ;; As of glibc 2.2.5, these are the only US Letter locales,
2454 ;; and the rest are A4.
2456 (or (locale-name-match locale '(("c$" . letter)
2461 ("enu$" . letter) ; Windows
2468 ;;; Charset property
2470 (defun get-charset-property (charset propname)
2471 "Return the value of CHARSET's PROPNAME property.
2472 This is the last value stored with
2473 (put-charset-property CHARSET PROPNAME VALUE)."
2474 (and (not (eq charset 'composition))
2475 (plist-get (charset-plist charset) propname)))
2477 (defun put-charset-property (charset propname value)
2478 "Store CHARSETS's PROPNAME property with value VALUE.
2479 It can be retrieved with `(get-charset-property CHARSET PROPNAME)'."
2480 (or (eq charset 'composition)
2481 (set-charset-plist charset
2482 (plist-put (charset-plist charset) propname value))))
2484 ;;; Character code property
2485 (put 'char-code-property-table 'char-table-extra-slots 0)
2487 (defvar char-code-property-table
2488 (make-char-table 'char-code-property-table)
2489 "Char-table containing a property list of each character code.
2491 See also the documentation of `get-char-code-property' and
2492 `put-char-code-property'.")
2494 (defun get-char-code-property (char propname)
2495 "Return the value of CHAR's PROPNAME property in `char-code-property-table'."
2496 (let ((plist (aref char-code-property-table char)))
2498 (car (cdr (memq propname plist))))))
2500 (defun put-char-code-property (char propname value)
2501 "Store CHAR's PROPNAME property with VALUE in `char-code-property-table'.
2502 It can be retrieved with `(get-char-code-property CHAR PROPNAME)'."
2503 (let ((plist (aref char-code-property-table char)))
2505 (let ((slot (memq propname plist)))
2507 (setcar (cdr slot) value)
2508 (nconc plist (list propname value))))
2509 (aset char-code-property-table char (list propname value)))))
2512 ;; Pretty description of encoded string
2514 ;; Alist of ISO 2022 control code vs the corresponding mnemonic string.
2515 (defvar iso-2022-control-alist
2523 (defun encoded-string-description (str coding-system)
2524 "Return a pretty description of STR that is encoded by CODING-SYSTEM."
2525 (setq str (string-as-unibyte str))
2527 (if (and coding-system (eq (coding-system-type coding-system) 2))
2528 ;; Try to get a pretty description for ISO 2022 escape sequences.
2529 (function (lambda (x) (or (cdr (assq x iso-2022-control-alist))
2530 (format "0x%02X" x))))
2531 (function (lambda (x) (format "0x%02X" x))))
2534 (defun encode-coding-char (char coding-system)
2535 "Encode CHAR by CODING-SYSTEM and return the resulting string.
2536 If CODING-SYSTEM can't safely encode CHAR, return nil."
2537 (let ((str1 (string-as-multibyte (char-to-string char)))
2538 (str2 (string-as-multibyte (make-string 2 char)))
2539 (safe-chars (and coding-system
2540 (coding-system-get coding-system 'safe-chars)))
2541 (charset (char-charset char))
2543 (when (or (eq safe-chars t)
2545 (and safe-chars (aref safe-chars char)))
2546 ;; We must find the encoded string of CHAR. But, just encoding
2547 ;; CHAR will put extra control sequences (usually to designate
2548 ;; ASCII charset) at the tail if type of CODING is ISO 2022.
2549 ;; To exclude such tailing bytes, we at first encode one-char
2550 ;; string and two-char string, then check how many bytes at the
2551 ;; tail of both encoded strings are the same.
2553 (setq enc1 (encode-coding-string str1 coding-system)
2555 enc2 (encode-coding-string str2 coding-system)
2557 (while (and (> i1 0) (= (aref enc1 (1- i1)) (aref enc2 (1- i2))))
2558 (setq i1 (1- i1) i2 (1- i2)))
2560 ;; Now (substring enc1 i1) and (substring enc2 i2) are the same,
2561 ;; and they are the extra control sequences at the tail to
2563 (substring enc2 0 i2))))
2566 ;; arch-tag: b382c432-4b36-460e-bf4c-05efd0bb18dc
2567 ;;; mule-cmds.el ends here