| blob | 9222b19c3845b7e98e7c7657fb58473d07dbdec5 |
1 ;;; mule-cmds.el --- commands for multilingual environment -*-coding: iso-2022-7bit -*-
3 ;; Copyright (C) 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006,
4 ;; 2007, 2008 Free Software Foundation, Inc.
5 ;; Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
6 ;; 2005, 2006, 2007, 2008
7 ;; National Institute of Advanced Industrial Science and Technology (AIST)
8 ;; Registration Number H14PRO021
9 ;; Copyright (C) 2003
10 ;; National Institute of Advanced Industrial Science and Technology (AIST)
11 ;; Registration Number H13PRO009
13 ;; Keywords: mule, i18n
15 ;; This file is part of GNU Emacs.
17 ;; GNU Emacs is free software: you can redistribute it and/or modify
18 ;; it under the terms of the GNU General Public License as published by
19 ;; the Free Software Foundation, either version 3 of the License, or
20 ;; (at your option) any later version.
22 ;; GNU Emacs is distributed in the hope that it will be useful,
23 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
24 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
25 ;; GNU General Public License for more details.
27 ;; You should have received a copy of the GNU General Public License
28 ;; along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>.
30 ;;; Commentary:
32 ;;; Code:
39 ;;; MULE related key bindings and menus.
44 ;; Keep "C-x C-m ..." for mule specific commands.
76 t)
79 t)
82 t)
87 t)
96 t)
99 t)
102 describe-language-environment-map
104 t)
108 t)
111 t)
118 t)
123 t)
126 t)
130 t)
133 :enable buffer-file-name
135 t)
139 t)
142 t)
147 t)
152 t)
155 t)
160 t)
165 t)
171 t)
175 [Default] '(menu-item "Default" setup-specified-language-environment))
177 (define-key describe-language-environment-map
180 ;; This should be a single character key binding because users use it
181 ;; very frequently while editing multilingual text. Now we can use
182 ;; only two such keys: "\C-\\" and "\C-^", but the latter is not
183 ;; convenient because it requires shifting on most keyboards. An
184 ;; alternative is "\C-\]" which is now bound to `abort-recursive-edit'
185 ;; but it won't be used that frequently.
188 ;; This is no good because people often type Shift-SPC
189 ;; meaning to type SPC. -- rms.
190 ;; ;; Here's an alternative key binding for X users (Shift-SPACE).
191 ;; (define-key global-map [?\S- ] 'toggle-input-method)
193 ;;; Mule related hyperlinks.
201 ;; Note starting with word-syntax character:
205 "Return a coding system which differs from CODING-SYSTEM in EOL conversion.
206 The returned coding system converts end-of-line by EOL-TYPE
207 but text as the same way as CODING-SYSTEM.
208 EOL-TYPE should be `unix', `dos', `mac', or nil.
209 If EOL-TYPE is nil, the returned coding system detects
210 how end-of-line is formatted automatically while decoding.
212 EOL-TYPE can be specified by an integer 0, 1, or 2.
213 They means `unix', `dos', and `mac' respectively."
219 ;; We call `coding-system-base' before `coding-system-eol-type',
220 ;; because the coding-system may not be initialized until then.
225 coding-system
228 base)
230 coding-system)
236 "Return a coding system which differs from CODING-SYSTEM in text conversion.
237 The returned coding system converts text by CODING
238 but end-of-line as the same way as CODING-SYSTEM.
239 If CODING is nil, the returned coding system detects
240 how text is formatted automatically while decoding."
247 "Change whether this buffer uses multibyte characters.
248 With ARG, use multibyte characters if the ARG is positive.
250 Note that this command does not convert the byte contents of
251 the buffer; it only changes the way those bytes are interpreted.
252 In general, therefore, this command *changes* the sequence of
253 characters that the current buffer contains.
255 We suggest you avoid using this command unless you know what you are
256 doing. If you use it by mistake, and the buffer is now displayed
257 wrong, use this command again to toggle back to the right mode."
266 "Display the HELLO file, which lists many languages and characters."
268 ;; We have to decode the file in any environment.
274 "Execute an I/O command using the specified coding system."
279 buffer-file-coding-system)))
284 default))))
288 prefix)
289 ;; read-key-sequence ignores quit, so make an explicit check.
290 ;; Like many places, this assumes quit == C-g, but it need not be.
296 ;; Process keys bound in `universal-argument-map'.
302 ;; Have to bind `last-command-char' here so that
303 ;; `digit-argument', for instance, can compute the
304 ;; prefix arg.
308 ;; This is the final call to `universal-argument-other-key', which
309 ;; set's the final `prefix-arg.
313 ;; Read the command to execute with the given prefix arg.
326 "Set default value of various coding systems to CODING-SYSTEM.
327 This sets the following coding systems:
328 o coding system of a newly created buffer
329 o default coding system for subprocess I/O
330 This also sets the following values:
331 o default value used as `file-name-coding-system' for converting file names
332 if CODING-SYSTEM is ASCII-compatible
333 o default value for the command `set-terminal-coding-system' (not on MS-DOS)
334 o default value for the command `set-keyboard-coding-system'
335 if CODING-SYSTEM is ASCII-compatible"
344 ;; The file-name coding system on Darwin systems is always utf-8.
352 ;; Preserve eol-type from existing default-process-coding-systems.
353 ;; On non-unix-like systems in particular, these may have been set
354 ;; carefully by the user, or by the startup code, to deal with the
355 ;; users shell appropriately, so should not be altered by changing
356 ;; language environment.
367 "Add CODING-SYSTEM at the front of the priority list for automatic detection.
368 This also sets the following coding systems:
369 o coding system of a newly created buffer
370 o default coding system for subprocess I/O
371 This also sets the following values:
372 o default value used as `file-name-coding-system' for converting file names
373 o default value for the command `set-terminal-coding-system' (not on MS-DOS)
374 o default value for the command `set-keyboard-coding-system'
376 If CODING-SYSTEM specifies a certain type of EOL conversion, the coding
377 systems set by this function will use that type of EOL conversion.
379 This command does not change the default value of terminal coding system
380 for MS-DOS terminal, because DOS terminals only support a single coding
381 system, and Emacs automatically sets the default to that coding system at
382 startup.
384 A coding system that requires automatic detection of text+encoding
385 \(e.g. undecided, unix) can't be preferred."
397 base coding-system)))
398 ;; If they asked for specific EOL conversion, honor that.
405 "If non-nil, a predicate function to sort coding systems.
407 It is called with two coding systems, and should return t if the first
413 "Sort coding system list CODINGS by a priority of each coding system.
414 Return the sorted list. CODINGS is modified by side effects.
416 If a coding system is most preferred, it has the highest priority.
417 Otherwise, coding systems that correspond to MIME charsets have
418 higher priorities. Among them, a coding system included in the
419 `coding-system' key of the current language environment has higher
420 priority. See also the documentation of `language-info-alist'.
422 If the variable `sort-coding-systems-predicate' (which see) is
423 non-nil, it is used to sort CODINGS instead."
433 ;; We calculate the priority number 0..255 by
434 ;; using the 8 bits PMMLCEII as this:
435 ;; P: 1 if most preferred.
436 ;; MM: greater than 0 if mime-charset.
437 ;; L: 1 if one of the current lang. env.'s codings.
438 ;; C: 1 if one of codings listed in the category list.
439 ;; E: 1 if not XXX-with-esc
440 ;; II: if iso-2022 based, 0..3, else 1.
445 ;; Prefer coding systems corresponding to a
446 ;; MIME charset.
448 ;; Lower utf-16 priority so that we
449 ;; normally prefer utf-8 to it, and put
450 ;; x-ctext below that.
466 ;; For ISO based coding systems, prefer
467 ;; one that doesn't use designation nor
468 ;; locking/single shifting.
479 ))))))
484 "Return a list of proper coding systems to encode a text between FROM and TO.
486 If FROM is a string, find coding systems in that instead of the buffer.
487 All coding systems in the list can safely encode any multibyte characters
488 in the text.
490 If the text contains no multibyte characters, return a list of a single
491 element `undecided'."
494 ;; The text contains only ASCII characters. Any coding
495 ;; systems are safe.
497 ;; We need copy-sequence because sorting will alter the argument.
501 "Return a list of proper coding systems to encode STRING.
502 All coding systems in the list can safely encode any multibyte characters
503 in STRING.
505 If STRING contains no multibyte characters, return a list of a single
506 element `undecided'."
510 "Return a list of proper coding systems to encode characters of CHARSETS.
511 CHARSETS is a list of character sets.
513 This only finds coding systems of type `charset', whose
514 `:charset-list' property includes all of CHARSETS (plus `ascii' for
515 ASCII-compatible coding systems). It was used in older versions of
516 Emacs, but is unlikely to be what you really want now."
517 ;; Deal with aliases.
520 charsets))
541 t))
546 "Find multibyte characters in the region specified by FROM and TO.
547 If FROM is a string, find multibyte characters in the string.
548 The return value is an alist of the following format:
549 ((CHARSET COUNT CHAR ...) ...)
550 where
551 CHARSET is a character set,
552 COUNT is a number of characters,
553 CHARs are the characters found from the character set.
554 Optional 3rd arg MAXCOUNT limits how many CHARs are put in the above list.
555 Optional 4th arg EXCLUDES is a list of character sets to be ignored."
557 charset char)
592 "Search forward from point for a character that is not encodable.
593 It asks which coding system to check.
594 If such a character is found, set point after that character.
595 Otherwise, don't move point.
597 When called from a program, the value is the position of the unencodable
598 character found, or nil if all characters are encodable."
603 default))))
608 pos))
611 "Most recent coding system explicitly specified by the user when asked.
612 This variable is set whenever Emacs asks the user which coding system
613 to use in order to write a file. If you set it to nil explicitly,
614 then call `write-region', then afterward this variable will be non-nil
618 "If non-nil, a function to control the behavior of coding system selection.
619 The meaning is the same as the argument ACCEPT-DEFAULT-P of the
620 function `select-safe-coding-system' (which see). This variable
625 "Select interactively a coding system for the region FROM ... TO.
626 FROM can be a string, as in `write-region'.
627 CODINGS is the list of base coding systems known to be safe for this region,
628 typically obtained with `find-coding-systems-region'.
629 UNSAFE is a list of coding systems known to be unsafe for this region.
630 REJECTED is a list of coding systems which were safe but for some reason
631 were not recommended in the particular context.
632 DEFAULT is the coding system to use by default in the query."
633 ;; At first, if some defaults are unsafe, record at most 11
634 ;; problematic characters and their positions for them by turning
635 ;; (CODING ...)
636 ;; into
637 ;; ((CODING (POS . CHAR) (POS . CHAR) ...) ...)
652 unsafe)))
654 ;; Change each safe coding system to the corresponding
655 ;; mime-charset name if it is also a coding system. Such a name
656 ;; is more friendly to users.
658 mime-charset)
666 ;; Don't offer variations with locking shift, which you
667 ;; basically never want.
676 ;; Remove raw-text, emacs-mule and no-conversion unless nothing
677 ;; else is available.
686 coding-system)
688 ;; If some defaults are unsafe, make sure the offending
689 ;; buffer is displayed.
693 unsafe))))
694 ;; Then ask users to select one from CODINGS while showing
695 ;; the reason why none of the defaults are not used.
704 "These default coding systems were tried to encode"
719 but are not recommended for encoding text in this context,
754 'help-echo
755 "mouse-2, RET: jump to this character"
756 'help-function func1
759 "..."
762 'help-echo
763 "mouse-2, RET: next unencodable character"
764 'help-function func2
787 ;; Read a coding system.
791 default))
796 coding-system))
799 accept-default-p file)
800 "Ask a user to select a safe coding system from candidates.
801 The candidates of coding systems which can safely encode a text
802 between FROM and TO are shown in a popup window. Among them, the most
803 proper one is suggested as the default.
805 The list of `buffer-file-coding-system' of the current buffer, the
806 `default-buffer-file-coding-system', and the most preferred coding
807 system (if it corresponds to a MIME charset) is treated as the
808 default coding system list. Among them, the first one that safely
809 encodes the text is normally selected silently and returned without
810 any user interaction. See also the command `prefer-coding-system'.
812 However, the user is queried if the chosen coding system is
813 inconsistent with what would be selected by `find-auto-coding' from
814 coding cookies &c. if the contents of the region were read from a
815 file. (That could lead to data corruption in a file subsequently
816 re-visited and edited.)
818 Optional 3rd arg DEFAULT-CODING-SYSTEM specifies a coding system or a
819 list of coding systems to be prepended to the default coding system
820 list. However, if DEFAULT-CODING-SYSTEM is a list and the first
821 element is t, the cdr part is used as the default coding system list,
822 i.e. `buffer-file-coding-system', `default-buffer-file-coding-system',
823 and the most preferred coding system are not used.
825 Optional 4th arg ACCEPT-DEFAULT-P, if non-nil, is a function to
826 determine the acceptability of the silently selected coding system.
827 It is called with that coding system, and should return nil if it
828 should not be silently selected and thus user interaction is required.
830 Optional 5th arg FILE is the file name to use for this purpose.
831 That is different from `buffer-file-name' when handling `write-region'
832 \(for example).
834 The variable `select-safe-coding-system-accept-default-p', if non-nil,
835 overrides ACCEPT-DEFAULT-P.
837 Kludgy feature: if FROM is a string, the string is the target text,
838 and TO is ignored."
844 auto-cs)
846 ;; Find an auto-coding that is specified for the the current
847 ;; buffer and file from the region FROM and TO.
858 :warning
860 Invalid coding system `%s' is specified
861 for the current buffer/file by the %s.
862 It is highly recommended to fix it before writing to a file."
874 ;; Change elements of the list to (coding . base-coding).
877 default-coding-system))
880 ;; If the file has a coding cookie, try to use it before anything
881 ;; else (i.e. before default-coding-system which will typically come
882 ;; from file-coding-system-alist).
889 ;; If buffer-file-coding-system is not nil nor undecided, append it
890 ;; to the defaults.
899 ;; If default-buffer-file-coding-system is not nil nor undecided,
900 ;; append it to the defaults.
908 base)))))))
910 ;; If the most preferred coding system has the property mime-charset,
911 ;; append it to the defaults.
913 base)
925 ;; Decide the eol-type from the top of the default codings,
926 ;; buffer-file-coding-system, or
927 ;; default-buffer-file-coding-system.
933 buffer-file-coding-system)))
936 default-buffer-file-coding-system)))
945 safe rejected unsafe)
947 ;; Any coding system is ok.
949 ;; Reverse the list so that elements are accumulated in safe,
950 ;; rejected, and unsafe in the correct order.
953 ;; Classify the defaults into safe, rejected, and unsafe.
965 ;; If all the defaults failed, ask a user.
970 ;; Check we're not inconsistent with what `coding:' spec &c would
971 ;; give when file is re-read.
972 ;; But don't do this if we explicitly ignored the cookie
973 ;; by using `find-file-literally'.
976 coding-system
978 ;; Merge coding-system and auto-cs as far as possible.
987 coding-system auto-cs))
990 auto-cs coding-system))))
994 coding-system eol-type-2)))
997 auto-cs eol-type-1)))))))
1000 ;; Don't barf if writing a compressed file, say.
1001 ;; This check perhaps isn't ideal, but is probably
1002 ;; the best thing to do.
1007 %s specified by file contents. Really save (else edit coding cookies \
1012 coding-system)))
1017 "Return a coding system to encode the outgoing message of the current buffer.
1018 It at first tries the first coding system found in these variables
1019 in this order:
1020 (1) local value of `buffer-file-coding-system'
1021 (2) value of `sendmail-coding-system'
1022 (3) value of `default-sendmail-coding-system'
1023 (4) value of `default-buffer-file-coding-system'
1024 If the found coding system can't encode the current buffer,
1025 or none of them are bound to a coding system,
1026 it asks the user to select a proper coding system."
1028 buffer-file-coding-system)
1029 sendmail-coding-system
1030 default-sendmail-coding-system
1031 default-buffer-file-coding-system)))
1033 ;; We should never use no-conversion for outgoing mail.
1039 coding)))
1040 \f
1041 ;;; Language support stuff.
1044 "Alist of language environment definitions.
1045 Each element looks like:
1046 (LANGUAGE-NAME . ((KEY . INFO) ...))
1047 where LANGUAGE-NAME is a string, the name of the language environment,
1048 KEY is a symbol denoting the kind of information, and
1049 INFO is the data associated with KEY.
1050 Meaningful values for KEY include
1052 documentation value is documentation of what this language environment
1053 is meant for, and how to use it.
1054 charset value is a list of the character sets mainly used
1055 by this language environment.
1056 sample-text value is an expression which is evalled to generate
1057 a line of text written using characters appropriate
1058 for this language environment.
1059 setup-function value is a function to call to switch to this
1060 language environment.
1061 exit-function value is a function to call to leave this
1062 language environment.
1063 coding-system value is a list of coding systems that are good for
1064 saving text written in this language environment.
1065 This list serves as suggestions to the user;
1066 in effect, as a kind of documentation.
1067 coding-priority value is a list of coding systems for this language
1068 environment, in order of decreasing priority.
1069 This is used to set up the coding system priority
1070 list when you switch to this language environment.
1071 nonascii-translation
1072 value is a charset of dimension one to use for
1073 converting a unibyte character to multibyte
1074 and vice versa.
1075 input-method value is a default input method for this language
1076 environment.
1077 features value is a list of features requested in this
1078 language environment.
1079 ctext-non-standard-encodings
1080 value is a list of non-standard encoding names used
1081 in extended segments of CTEXT. See the variable
1082 `ctext-non-standard-encodings' for more detail.
1084 The following keys take effect only when multibyte characters are
1085 globally disabled, i.e. the value of `default-enable-multibyte-characters'
1086 is nil.
1088 unibyte-display value is a coding system to encode characters for
1089 the terminal. Characters in the range of 160 to
1090 255 display not as octal escapes, but as non-ASCII
1094 "Return information listed under KEY for language environment LANG-ENV.
1095 KEY is a symbol denoting the kind of information.
1096 For a list of useful values for KEY and their meanings,
1097 see `language-info-alist'."
1105 "Modify part of the definition of language environment LANG-ENV.
1106 Specifically, this stores the information INFO under KEY
1107 in the definition of this language environment.
1108 KEY is a symbol denoting the kind of information.
1109 INFO is the value for that information.
1111 For a list of useful values for KEY and their meanings,
1112 see `language-info-alist'."
1131 "Internal use only.
1132 Arguments are the same as `set-language-info'."
1144 ;; Update the custom-type of `current-language-environment'.
1152 "Store ALIST as the definition of language environment LANG-ENV.
1153 ALIST is an alist of KEY and INFO values. See the documentation of
1154 `language-info-alist' for the meanings of KEY and INFO.
1156 Optional arg PARENTS is a list of parent menu names; it specifies
1157 where to put this language environment in the
1158 Describe Language Environment and Set Language Environment menus.
1160 in the European submenu in each of those two menus."
1167 map parent-symbol parent prompt)
1173 ;; This prompt string is for define-prefix-command, so
1174 ;; that the map it creates will be suitable for a menu.
1195 ;; Set up menu items for this language env.
1210 "Read a language environment name which has information for KEY.
1211 If KEY is nil, read any language environment.
1212 Prompt with PROMPT. DEFAULT is the default choice of language environment.
1213 This returns a language environment name as a string."
1216 language-info-alist
1219 t nil nil default)))
1223 name)))
1224 \f
1225 ;;; Multilingual input methods.
1227 "LEIM: Libraries of Emacs Input Methods."
1231 "Name of LEIM list file.
1232 This file contains a list of libraries of Emacs input methods (LEIM)
1233 in the format of Lisp expression for registering each input method.
1237 ";;; %s -- list of LEIM (Library of Emacs Input Method) -*-coding: iso-2022-7bit;-*-
1238 ;;
1239 ;; This file contains a list of LEIM (Library of Emacs Input Method)
1240 ;; methods in the same directory as this file. Loading this file
1241 ;; registers all the input methods in Emacs.
1242 ;;
1243 ;; Each entry has the form:
1244 ;; (register-input-method
1245 ;; INPUT-METHOD LANGUAGE-NAME ACTIVATE-FUNC
1246 ;; TITLE DESCRIPTION
1247 ;; ARG ...)
1248 ;; See the function `register-input-method' for the meanings of the arguments.
1249 ;;
1250 ;; If this directory is included in load-path, Emacs automatically
1251 ;; loads this file at startup time.
1253 "
1254 leim-list-file-name)
1258 "Regexp matching head of each entry in LEIM list file.
1263 "List of functions to call to update LEIM list file.
1267 "Update LEIM list file in directories DIRS."
1272 "The current input method for multilingual text.
1283 "Default input method for multilingual text (a string).
1284 This is the input method activated automatically by the command
1289 :completion-ignore-case t
1290 :complete-function widget-string-complete
1291 :completion-alist input-method-alist
1298 "History list of input methods read from the minibuffer.
1300 Maximum length of the history list is determined by the value
1306 "Function to call for inactivating the current input method.
1307 Every input method should set this to an appropriate value when activated.
1308 This function is called with no argument.
1310 This function should never change the value of `current-input-method'.
1316 "Function to call for describing the current input method.
1322 "Alist of input method names vs how to use them.
1323 Each element has the form:
1324 (INPUT-METHOD LANGUAGE-ENV ACTIVATE-FUNC TITLE DESCRIPTION ARGS...)
1328 "Register INPUT-METHOD as an input method for language environment LANG-ENV.
1330 INPUT-METHOD and LANG-ENV are symbols or strings.
1331 ACTIVATE-FUNC is a function to call to activate this method.
1332 TITLE is a string to show in the mode line when this method is active.
1333 DESCRIPTION is a string describing this method and what it is good for.
1334 The ARGS, if any, are passed as arguments to ACTIVATE-FUNC.
1335 All told, the arguments to ACTIVATE-FUNC are INPUT-METHOD and the ARGS.
1338 created at Emacs build time, registering all Quail input methods
1339 contained in the Emacs distribution.
1341 In case you want to register a new Quail input method by yourself, be
1342 careful to use the same input method title as given in the third
1343 parameter of `quail-define-package'. (If the values are different, the
1344 string specified in this function takes precedence.)
1346 The commands `describe-input-method' and `list-input-methods' need
1347 these duplicated values to show some information about input methods
1348 without loading the relevant Quail packages.
1362 "Read a name of input method from a minibuffer prompting with PROMPT.
1363 If DEFAULT is non-nil, use that as the default,
1364 and substitute it into PROMPT at the first `%s'.
1365 If INHIBIT-NULL is non-nil, null input signals an error.
1367 The return value is a string."
1371 ;; As it is quite normal to change input method in the
1372 ;; minibuffer, we must enable it even if
1373 ;; enable-recursive-minibuffers is currently nil.
1375 ;; This binding is necessary because input-method-history is
1376 ;; buffer local.
1378 nil t nil 'input-method-history
1379 default)))
1383 input-method
1388 "Switch to input method INPUT-METHOD for the current buffer.
1389 If some other input method is already active, turn it off first.
1390 If INPUT-METHOD is nil, deactivate any current input method."
1417 "Turn off the current input method."
1428 current-input-method-title nil)
1436 "Select and activate input method INPUT-METHOD for the current buffer.
1437 This also sets the default input method to the one you specify.
1438 If INPUT-METHOD is nil, this function turns off the input method, and
1439 also causes you to be prompted for a name of an input method the next
1441 When called interactively, the optional arg INTERACTIVE is non-nil,
1442 which marks the variable `default-input-method' as set for Custom buffers.
1445 To deactivate it programmatically, use `inactivate-input-method'."
1450 default t)
1451 t)))
1456 default-input-method)
1462 "Enable or disable multilingual text input method for the current buffer.
1463 Only one input method can be enabled at any time in a given buffer.
1465 The normal action is to enable an input method if none was enabled,
1466 and disable the current one otherwise. Which input method to enable
1467 can be determined in various ways--either the one most recently used,
1468 or the one specified by `default-input-method', or as a last resort
1469 by reading the name of an input method in the minibuffer.
1471 With a prefix argument ARG, read an input method name with the minibuffer
1472 and enable that one. The default is the most recent input method specified
1473 \(not including the currently active input method, if any).
1475 When called interactively, the optional argument INTERACTIVE is non-nil,
1476 which marks the variable `default-input-method' as set for Custom buffers."
1493 default t))
1494 default))
1504 "Describe input method INPUT-METHOD."
1534 "Describe the input method currently in use.
1535 This is a subroutine for `describe-input-method'."
1541 current-input-method)
1546 "Read a multilingual string from minibuffer, prompting with string PROMPT.
1547 The input method selected last time is activated in minibuffer.
1548 If optional second argument INITIAL-INPUT is non-nil, insert it in the
1549 minibuffer initially.
1550 Optional 3rd argument INPUT-METHOD specifies the input method to be activated
1551 instead of the one selected last time. It is a symbol or a string."
1554 current-input-method
1555 default-input-method
1566 ;; Variables to control behavior of input methods. All input methods
1567 ;; should react to these variables.
1570 "A flag to control extra guidance given by input methods.
1571 The value should be nil, t, `complex-only', or `default'.
1573 The extra guidance is done by showing list of available keys in echo
1574 area. When you use the input method in the minibuffer, the guidance
1575 is shown at the bottom short window (split from the existing window).
1577 If the value is t, extra guidance is always given, if the value is
1578 nil, extra guidance is always suppressed.
1580 If the value is `complex-only', only complex input methods such as
1581 `chinese-py' and `japanese' give extra guidance.
1583 If the value is `default', complex input methods always give extra
1584 guidance, but simple input methods give it only when you are not in
1585 the minibuffer.
1587 See also the variable `input-method-highlight-flag'."
1593 "If this flag is non-nil, input methods highlight partially-entered text.
1594 For instance, while you are in the middle of a Quail input method sequence,
1595 the text inserted so far is temporarily underlined.
1596 The underlining goes away when you finish or abort the input method sequence.
1597 See also the variable `input-method-verbose-flag'."
1602 "Normal hook run just after an input method is activated.
1604 The variable `current-input-method' keeps the input method name
1605 just activated."
1610 "Normal hook run just after an input method is inactivated.
1612 The variable `current-input-method' still keeps the input method name
1613 just inactivated."
1618 "Normal hook run just after an input method insert some chunk of text."
1623 "This flag controls when an input method returns.
1624 Usually, the input method does not return while there's a possibility
1625 that it may find a different translation if a user types another key.
1626 But, if this flag is non-nil, the input method returns as soon as the
1630 "This flag controls how an input method shows an intermediate key sequence.
1631 Usually, the input method inserts the intermediate key sequence,
1632 or candidate translations corresponding to the sequence,
1633 at point in the current buffer.
1634 But, if this flag is non-nil, it displays them in echo area instead."
1639 "This flag controls the behavior of an input method on invalid key input.
1640 Usually, when a user types a key which doesn't start any character
1641 handled by the input method, the key is handled by turning off the
1642 input method temporarily. After that key, the input method is re-enabled.
1645 \f
1647 "Normal hook run after some language environment is set.
1649 When you set some hook function here, that effect usually should not
1650 be inherited to another language environment. So, you had better set
1651 another function in `exit-language-environment-hook' (which see) to
1652 cancel the effect."
1657 "Normal hook run after exiting from some language environment.
1658 When this hook is run, the variable `current-language-environment'
1659 is still bound to the language environment being exited.
1661 This hook is mainly used for canceling the effect of
1662 `set-language-environment-hook' (which see)."
1669 "Switch to a specified language environment."
1682 "The last language environment specified with `set-language-environment'.
1684 to using the function `set-language-environment'."
1691 current-language-environment)
1692 language-info-alist t))
1694 ;; custom type will be updated with `set-language-info'.
1705 "Reset multilingual environment of Emacs to the default status.
1707 The default status is as follows:
1709 The default value of `buffer-file-coding-system' is nil.
1710 The default coding system for process I/O is nil.
1711 The default value for the command `set-terminal-coding-system' is nil.
1712 The default value for the command `set-keyboard-coding-system' is nil.
1714 The order of priorities of coding systems are as follows:
1715 utf-8
1716 iso-2022-7bit
1717 iso-latin-1
1718 iso-2022-7bit-lock
1719 iso-2022-8bit-ss2
1720 emacs-mule
1721 raw-text"
1723 ;; This function formerly set default-enable-multibyte-characters to t,
1724 ;; but that is incorrect. It should not alter the unibyte/multibyte choice.
1727 'utf-8
1728 'iso-2022-7bit
1729 'iso-latin-1
1730 'iso-2022-7bit-lock
1731 'iso-2022-8bit-ss2
1732 'emacs-mule
1737 ;; On Darwin systems, this should be utf-8, but when this file is loaded
1738 ;; utf-8 is not yet defined, so we set it in set-locale-environment instead.
1740 ;; Preserve eol-type from existing default-process-coding-systems.
1741 ;; On non-unix-like systems in particular, these may have been set
1742 ;; carefully by the user, or by the startup code, to deal with the
1743 ;; users shell appropriately, so should not be altered by changing
1744 ;; language environment.
1746 ;; When bootstrapping, coding-systems are not defined yet, so
1747 ;; we need to catch the error from check-coding-system.
1760 ;; Don't alter the terminal and keyboard coding systems here.
1761 ;; The terminal still supports the same coding system
1762 ;; that it supported a minute ago.
1763 ;; (set-terminal-coding-system-internal nil)
1764 ;; (set-keyboard-coding-system-internal nil)
1766 ;; Back in Emacs-20, it was necessary to provide some fallback implicit
1767 ;; conversion, because almost no packages handled coding-system issues.
1768 ;; Nowadays it'd just paper over bugs.
1769 ;; (set-unibyte-charset 'iso-8859-1)
1770 )
1774 (defun set-display-table-and-terminal-coding-system (language-name &optional coding-system display)
1775 "Set up the display table and terminal coding system for LANGUAGE-NAME."
1781 ;; The following 2 lines undo the 8-bit display that we set up
1782 ;; in standard-display-european-internal, which see. This is in
1783 ;; case the user has used standard-display-european earlier in
1784 ;; this session.
1791 "Set up multi-lingual environment for using LANGUAGE-NAME.
1792 This sets the coding system priority and the default input method
1793 and sometimes other things. LANGUAGE-NAME should be a string
1795 specifies the character set for the major languages of Western Europe."
1797 nil
1814 ;; The features might set up coding systems.
1826 ;; Unibyte setups if necessary.
1836 current-iso639-language))
1842 "An Emacs charset."
1847 :completion-ignore-case t
1853 widget))
1857 "Customizations of language environment parameters.
1858 Value is an alist with elements like those of `language-info-alist'.
1859 These are used to set values in `language-info-alist' which replace
1860 the defaults. A typical use is replacing the default input method for
1863 This option is intended for use at startup. Removing items doesn't
1864 remove them from the language info until you next restart Emacs.
1866 Setting this variable directly does not take effect.
1867 See `set-language-info-alist' for use in programs."
1872 ;; Can't do this before language environments are set up.
1874 ;; modify language-info-alist
1877 ;; re-set the environment in case its parameters changed
1881 :completion-ignore-case t
1882 :complete-function widget-string-complete
1884 :value-type
1896 :completion-ignore-case t
1897 :complete-function widget-string-complete
1898 :completion-alist input-method-alist
1907 ;; Actually set up direct output of non-ASCII characters.
1909 ;; Unibyte Emacs on MS-DOS wants to display all 8-bit characters with
1910 ;; the native font, and codes 160 and 146 stand for something very
1911 ;; different there.
1914 ;; Most X fonts used to do the wrong thing for latin-1 code 160.
1916 ;; XFree86 4 has fixed the fonts.
1920 ;; Make non-line-break space display as a plain space.
1923 ;; don't contain a character at that position. Map it to the ASCII
1924 ;; apostrophe. [This is actually RIGHT SINGLE QUOTATION MARK,
1925 ;; U+2019, normally from the windows-1252 character set. XFree 4
1926 ;; fonts probably have the appropriate glyph at this position,
1927 ;; so they could use standard-display-8bit. It's better to use a
1928 ;; proper windows-1252 coding system. --fx]
1929 (aset standard-display-table 146 [39]))))
1931 (defun set-language-environment-coding-systems (language-name)
1932 "Do various coding system setups for language environment LANGUAGE-NAME."
1933 (let* ((priority (get-language-info language-name 'coding-priority))
1934 (default-coding (car priority))
1935 (eol-type (coding-system-eol-type default-buffer-file-coding-system)))
1936 (when priority
1937 (set-default-coding-systems
1938 (if (memq eol-type '(0 1 2 unix dos mac))
1939 (coding-system-change-eol-conversion default-coding eol-type)
1940 default-coding))
1941 (setq default-sendmail-coding-system default-coding)
1942 (apply 'set-coding-system-priority priority))))
1944 (defun set-language-environment-input-method (language-name)
1945 "Do various input method setups for language environment LANGUAGE-NAME."
1946 (let ((input-method (get-language-info language-name 'input-method)))
1947 (when input-method
1948 (setq default-input-method input-method)
1949 (if input-method-history
1950 (setq input-method-history
1951 (cons input-method
1952 (delete input-method input-method-history)))))))
1954 (defun set-language-environment-nonascii-translation (language-name)
1955 "Do unibyte/multibyte translation setup for language environment LANGUAGE-NAME."
1956 ;; Note: For DOS, we assumed that the charset cpXXX is already
1957 ;; defined.
1958 (let ((nonascii (get-language-info language-name 'nonascii-translation)))
1959 (if (eq window-system 'pc)
1960 (setq nonascii (intern (format "cp%d" dos-codepage))))
1961 (or (and (charsetp nonascii)
1962 (get-charset-property nonascii :ascii-compatible-p))
1963 (setq nonascii 'iso-8859-1))
1964 ;; Back in Emacs-20, it was necessary to provide some fallback implicit
1965 ;; conversion, because almost no packages handled coding-system issues.
1966 ;; Nowadays it'd just paper over bugs.
1967 ;; (set-unibyte-charset nonascii)
1968 ))
1970 (defun set-language-environment-charset (language-name)
1971 "Do various charset setups for language environment LANGUAGE-NAME."
1972 ;; Put higher priorities to such charsets that are supported by the
1973 ;; coding systems of higher priorities in this environment.
1974 (let ((charsets (get-language-info language-name 'charset)))
1975 (dolist (coding (get-language-info language-name 'coding-priority))
1976 (setq charsets (append charsets (coding-system-charset-list coding))))
1977 (if charsets
1978 (apply 'set-charset-priority charsets))))
1980 (defun set-language-environment-unibyte (language-name)
1981 "Do various unibyte-mode setups for language environment LANGUAGE-NAME."
1982 (set-display-table-and-terminal-coding-system language-name))
1984 (defsubst princ-list (&rest args)
1986 (while args (princ (car args)) (setq args (cdr args)))
1989 (put 'describe-specified-language-support 'apropos-inhibit t)
1991 ;; Print language-specific information such as input methods,
1992 ;; charsets, and coding systems. This function is intended to be
1993 ;; called from the menu:
1994 ;; [menu-bar mule describe-language-environment LANGUAGE]
1995 ;; and should not run it by `M-x describe-current-input-method-function'.
1996 (defun describe-specified-language-support ()
1997 "Describe how Emacs supports the specified language environment."
1998 (interactive)
1999 (let (language-name)
2000 (if (not (and (symbolp last-command-event)
2001 (or (not (eq last-command-event 'Default))
2002 (setq last-command-event 'English))
2003 (setq language-name (symbol-name last-command-event))))
2004 (error "Bogus calling sequence"))
2005 (describe-language-environment language-name)))
2007 (defun describe-language-environment (language-name)
2008 "Describe how Emacs supports language environment LANGUAGE-NAME."
2009 (interactive
2010 (list (read-language-name
2011 'documentation
2012 "Describe language environment (default current choice): ")))
2013 (if (null language-name)
2014 (setq language-name current-language-environment))
2015 (if (or (null language-name)
2016 (null (get-language-info language-name 'documentation)))
2017 (error "No documentation for the specified language"))
2018 (if (symbolp language-name)
2019 (setq language-name (symbol-name language-name)))
2020 (dolist (feature (get-language-info language-name 'features))
2021 (require feature))
2022 (let ((doc (get-language-info language-name 'documentation)))
2023 (help-setup-xref (list #'describe-language-environment language-name)
2024 (interactive-p))
2025 (with-output-to-temp-buffer (help-buffer)
2026 (save-excursion
2027 (set-buffer standard-output)
2029 (if (stringp doc)
2031 (condition-case nil
2032 (let ((str (eval (get-language-info language-name 'sample-text))))
2033 (if (stringp str)
2037 (error nil))
2038 (let ((input-method (get-language-info language-name 'input-method))
2039 (l (copy-sequence input-method-alist))
2040 (first t))
2041 (when (and input-method
2042 (setq input-method (assoc input-method l)))
2044 (setq l (cons input-method (delete input-method l))
2045 first nil))
2046 (dolist (elt l)
2047 (when (or (eq input-method elt)
2048 (eq t (compare-strings language-name nil nil
2049 (nth 1 elt) nil nil t)))
2050 (when first
2052 (setq first nil))
2053 (insert " " (car elt))
2054 (search-backward (car elt))
2055 (help-xref-button 0 'help-input-method (car elt))
2056 (goto-char (point-max))
2058 (if (stringp (nth 3 elt)) (nth 3 elt) (car (nth 3 elt)))
2060 (or first
2063 (let ((l (get-language-info language-name 'charset)))
2064 (if (null l)
2066 (while l
2067 (insert " " (symbol-name (car l)))
2068 (search-backward (symbol-name (car l)))
2069 (help-xref-button 0 'help-character-set (car l))
2070 (goto-char (point-max))
2072 (setq l (cdr l)))))
2075 (let ((l (get-language-info language-name 'coding-system)))
2076 (if (null l)
2078 (while l
2079 (insert " " (symbol-name (car l)))
2080 (search-backward (symbol-name (car l)))
2081 (help-xref-button 0 'help-coding-system (car l))
2082 (goto-char (point-max))
2083 (insert " (`"
2084 (coding-system-mnemonic (car l))
2086 (coding-system-doc-string (car l))
2088 (let ((aliases (coding-system-aliases (car l))))
2089 (when aliases
2091 (while aliases
2092 (insert " " (symbol-name (car aliases)))
2093 (setq aliases (cdr aliases)))
2095 (setq l (cdr l)))))))))
2096 \f
2097 ;;; Locales.
2099 (defvar locale-translation-file-name nil
2100 "File name for the system's file of locale-name aliases, or nil if none.")
2102 ;; The following definitions might as well be marked as constants and
2103 ;; purecopied, since they're normally used on startup, and probably
2104 ;; should reflect the facilities of the base Emacs.
2105 (defconst locale-language-names
2106 (purecopy
2107 '(
2108 ;; Locale names of the form LANGUAGE[_TERRITORY][.CODESET][@MODIFIER]
2109 ;; as specified in the Single Unix Spec, Version 2.
2110 ;; LANGUAGE is a language code taken from ISO 639:1988 (E/F)
2111 ;; with additions from ISO 639/RA Newsletter No.1/1989;
2112 ;; see Internet RFC 2165 (1997-06) and
2113 ;; http://www.evertype.com/standards/iso639/iso639-en.html
2114 ;; TERRITORY is a country code taken from ISO 3166
2115 ;; http://www.din.de/gremien/nas/nabd/iso3166ma/codlstp1/en_listp1.html.
2116 ;; CODESET and MODIFIER are implementation-dependent.
2118 ;; jasonr comments: MS Windows uses three letter codes for
2119 ;; languages instead of the two letter ISO codes that POSIX
2120 ;; uses. In most cases the first two letters are the same, so
2121 ;; most of the regexps in locale-language-names work. Japanese
2122 ;; and Chinese are exceptions, which are listed in the
2123 ;; non-standard section at the bottom of locale-language-names.
2125 ("aa_DJ" . "Latin-1") ; Afar
2126 ("aa" . "UTF-8")
2127 ;; ab Abkhazian
2128 ("af" . "Latin-1") ; Afrikaans
2129 ("am" "Ethiopic" utf-8) ; Amharic
2130 ("an" . "Latin-9") ; Aragonese
2131 ; ar Arabic glibc uses 8859-6
2132 ; as Assamese
2133 ; ay Aymara
2134 ("az" . "UTF-8") ; Azerbaijani
2135 ; ba Bashkir
2136 ("be" "Belarusian" cp1251) ; Belarusian [Byelorussian until early 1990s]
2137 ("bg" "Bulgarian" cp1251) ; Bulgarian
2138 ; bh Bihari
2139 ; bi Bislama
2140 ("bn" . "UTF-8") ; Bengali, Bangla
2141 ("bo" . "Tibetan")
2142 ("br" . "Latin-1") ; Breton
2143 ("bs" . "Latin-2") ; Bosnian
2144 ("byn" . "UTF-8") ; Bilin; Blin
2145 ("ca" . "Latin-1") ; Catalan
2146 ; co Corsican
2147 ("cs" "Czech" iso-8859-2)
2148 ("cy" "Welsh" iso-8859-14)
2149 ("da" . "Latin-1") ; Danish
2150 ("de" "German" iso-8859-1)
2151 ; dv Divehi
2152 ; dz Bhutani
2153 ("el" "Greek" iso-8859-7)
2154 ;; Users who specify "en" explicitly typically want Latin-1, not ASCII.
2155 ;; That's actually what the GNU locales define, modulo things like
2156 ;; en_IN -- fx.
2157 ("en_IN" "English" utf-8) ; glibc uses utf-8 for English in India
2158 ("en" "English" iso-8859-1) ; English
2159 ("eo" . "Esperanto") ; Esperanto
2160 ("es" "Spanish" iso-8859-1)
2161 ("et" . "Latin-1") ; Estonian
2162 ("eu" . "Latin-1") ; Basque
2163 ("fa" . "UTF-8") ; Persian
2164 ("fi" . "Latin-1") ; Finnish
2165 ("fj" . "Latin-1") ; Fiji
2166 ("fo" . "Latin-1") ; Faroese
2167 ("fr" "French" iso-8859-1) ; French
2168 ("fy" . "Latin-1") ; Frisian
2169 ("ga" . "Latin-1") ; Irish Gaelic (new orthography)
2170 ("gd" . "Latin-9") ; Scots Gaelic
2171 ("gez" "Ethiopic" utf-8) ; Geez
2172 ("gl" . "Latin-1") ; Gallegan; Galician
2173 ; gn Guarani
2174 ("gu" . "UTF-8") ; Gujarati
2175 ("gv" . "Latin-1") ; Manx Gaelic
2176 ; ha Hausa
2177 ("he" "Hebrew" iso-8859-8)
2178 ("hi" "Devanagari" utf-8) ; Hindi
2179 ("hr" "Croatian" iso-8859-2) ; Croatian
2180 ("hu" . "Latin-2") ; Hungarian
2181 ; hy Armenian
2182 ; ia Interlingua
2183 ("id" . "Latin-1") ; Indonesian
2184 ; ie Interlingue
2185 ; ik Inupiak
2186 ("is" . "Latin-1") ; Icelandic
2187 ("it" "Italian" iso-8859-1) ; Italian
2188 ; iu Inuktitut
2189 ("iw" "Hebrew" iso-8859-8)
2190 ("ja" "Japanese" euc-jp)
2191 ; jw Javanese
2192 ("ka" "Georgian" georgian-ps) ; Georgian
2193 ; kk Kazakh
2194 ("kl" . "Latin-1") ; Greenlandic
2195 ; km Cambodian
2196 ("kn" "Kannada" utf-8)
2197 ("ko" "Korean" euc-kr)
2198 ; ks Kashmiri
2199 ; ku Kurdish
2200 ("kw" . "Latin-1") ; Cornish
2201 ; ky Kirghiz
2202 ("la" . "Latin-1") ; Latin
2203 ("lb" . "Latin-1") ; Luxemburgish
2204 ("lg" . "Laint-6") ; Ganda
2205 ; ln Lingala
2206 ("lo" "Lao" utf-8) ; Laothian
2207 ("lt" "Lithuanian" iso-8859-13)
2208 ("lv" . "Latvian") ; Latvian, Lettish
2209 ; mg Malagasy
2210 ("mi" . "Latin-7") ; Maori
2211 ("mk" "Cyrillic-ISO" iso-8859-5) ; Macedonian
2212 ("ml" "Malayalam" utf-8)
2213 ("mn" . "UTF-8") ; Mongolian
2214 ; mo Moldavian
2215 ("mr" "Devanagari" utf-8) ; Marathi
2216 ("ms" . "Latin-1") ; Malay
2217 ("mt" . "Latin-3") ; Maltese
2218 ; my Burmese
2219 ; na Nauru
2220 ("nb" . "Latin-1") ; Norwegian
2221 ("ne" "Devanagari" utf-8) ; Nepali
2222 ("nl" "Dutch" iso-8859-1)
2223 ("no" . "Latin-1") ; Norwegian
2224 ("oc" . "Latin-1") ; Occitan
2225 ("om_ET" . "UTF-8") ; (Afan) Oromo
2226 ("om" . "Latin-1") ; (Afan) Oromo
2227 ; or Oriya
2228 ("pa" . "UTF-8") ; Punjabi
2229 ("pl" . "Latin-2") ; Polish
2230 ; ps Pashto, Pushto
2231 ("pt" . "Latin-1") ; Portuguese
2232 ; qu Quechua
2233 ("rm" . "Latin-1") ; Rhaeto-Romanic
2234 ; rn Kirundi
2235 ("ro" "Romanian" iso-8859-2)
2236 ("ru_RU" "Russian" iso-8859-5)
2237 ("ru_UA" "Russian" koi8-u)
2238 ; rw Kinyarwanda
2239 ("sa" . "Devanagari") ; Sanskrit
2240 ; sd Sindhi
2241 ("se" . "UTF-8") ; Northern Sami
2242 ; sg Sangho
2243 ("sh" . "Latin-2") ; Serbo-Croatian
2244 ; si Sinhalese
2245 ("sid" . "UTF-8") ; Sidamo
2246 ("sk" "Slovak" iso-8859-2)
2247 ("sl" "Slovenian" iso-8859-2)
2248 ; sm Samoan
2249 ; sn Shona
2250 ("so_ET" "UTF-8") ; Somali
2251 ("so" "Latin-1") ; Somali
2252 ("sq" . "Latin-1") ; Albanian
2253 ("sr" . "Latin-2") ; Serbian (Latin alphabet)
2254 ; ss Siswati
2255 ("st" . "Latin-1") ; Sesotho
2256 ; su Sundanese
2257 ("sv" "Swedish" iso-8859-1) ; Swedish
2258 ("sw" . "Latin-1") ; Swahili
2259 ("ta" "Tamil" utf-8)
2260 ("te" . "UTF-8") ; Telugu
2261 ("tg" "Tajik" koi8-t)
2262 ("th" "Thai" tis-620)
2263 ("ti" "Ethiopic" utf-8) ; Tigrinya
2264 ("tig_ER" . "UTF-8") ; Tigre
2265 ; tk Turkmen
2266 ("tl" . "Latin-1") ; Tagalog
2267 ; tn Setswana
2268 ; to Tonga
2269 ("tr" "Turkish" iso-8859-9)
2270 ; ts Tsonga
2271 ("tt" . "UTF-8") ; Tatar
2272 ; tw Twi
2273 ; ug Uighur
2274 ("uk" "Ukrainian" koi8-u)
2275 ("ur" . "UTF-8") ; Urdu
2276 ("uz_UZ@cyrillic" . "UTF-8"); Uzbek
2277 ("uz" . "Latin-1") ; Uzbek
2278 ("vi" "Vietnamese" utf-8)
2279 ; vo Volapuk
2280 ("wa" . "Latin-1") ; Walloon
2281 ; wo Wolof
2282 ("xh" . "Latin-1") ; Xhosa
2283 ("yi" . "Windows-1255") ; Yiddish
2284 ; yo Yoruba
2285 ; za Zhuang
2286 ("zh_HK" . "Chinese-Big5")
2287 ; zh_HK/BIG5-HKSCS \
2288 ("zh_TW" . "Chinese-Big5")
2289 ("zh_CN.GB2312" "Chinese-GB")
2290 ("zh_CN.GBK" "Chinese-GBK")
2291 ("zh_CN.GB18030" "Chinese-GB18030")
2292 ("zh_CN.UTF-8" . "Chinese-GBK")
2293 ("zh_CN" . "Chinese-GB")
2294 ("zh" . "Chinese-GB")
2295 ("zu" . "Latin-1") ; Zulu
2297 ;; ISO standard locales
2298 ("c$" . "ASCII")
2299 ("posix$" . "ASCII")
2301 ;; The "IPA" Emacs language environment does not correspond
2302 ;; to any ISO 639 code, so let it stand for itself.
2303 ("ipa$" . "IPA")
2305 ;; Nonstandard or obsolete language codes
2306 ("cz" . "Czech") ; e.g. Solaris 2.6
2307 ("ee" . "Latin-4") ; Estonian, e.g. X11R6.4
2308 ("iw" . "Hebrew") ; e.g. X11R6.4
2309 ("sp" . "Cyrillic-ISO") ; Serbian (Cyrillic alphabet), e.g. X11R6.4
2310 ("su" . "Latin-1") ; Finnish, e.g. Solaris 2.6
2311 ("jp" . "Japanese") ; e.g. MS Windows
2312 ("chs" . "Chinese-GBK") ; MS Windows Chinese Simplified
2313 ("cht" . "Chinese-BIG5") ; MS Windows Chinese Traditional
2314 ("gbz" . "UTF-8") ; MS Windows Dari Persian
2315 ("div" . "UTF-8") ; MS Windows Divehi (Maldives)
2316 ("wee" . "Latin-2") ; MS Windows Lower Sorbian
2317 ("wen" . "Latin-2") ; MS Windows Upper Sorbian
2318 ))
2319 "Alist of locale regexps vs the corresponding languages and coding systems.
2320 Each element has this form:
2321 \(LOCALE-REGEXP LANG-ENV CODING-SYSTEM)
2322 The first element whose LOCALE-REGEXP matches the start of a
2323 downcased locale specifies the LANG-ENV \(language environment)
2324 and CODING-SYSTEM corresponding to that locale. If there is no
2325 appropriate language environment, the element may have this form:
2326 \(LOCALE-REGEXP . LANG-ENV)
2327 In this case, LANG-ENV is one of generic language environments for an
2330 (defconst locale-charset-language-names
2331 (purecopy
2340 ;; utf-8@euro exists, so put this last. (@euro really specifies
2341 ;; the currency, rather than the charset.)
2343 "List of pairs of locale regexps and charset language names.
2344 The first element whose locale regexp matches the start of a downcased locale
2345 specifies the language name whose charset corresponds to that locale.
2346 This language name is used if the locale is not listed in
2347 `locale-language-names'.")
2349 (defconst locale-preferred-coding-systems
2350 (purecopy
2359 ;; utf-8@euro exists, so put this after utf-8. (@euro really
2360 ;; specifies the currency, rather than the charset.)
2361 (".*@euro" . iso-8859-15)
2362 ("koi8-?r" . koi8-r)
2363 ("koi8-?u" . koi8-u)
2364 ("tcvn" . tcvn)
2365 ("big5[-_]?hkscs" . big5-hkscs)
2366 ("big5" . big5)
2367 ("euc-?tw" . euc-tw)
2368 ("euc-?cn" . euc-cn)
2369 ("gb2312" . gb2312)
2370 ("gbk" . gbk)
2371 ("gb18030" . gb18030)
2372 ("ja.*[._]euc" . japanese-iso-8bit)
2373 ("ja.*[._]jis7" . iso-2022-jp)
2374 ("ja.*[._]pck" . japanese-shift-jis)
2375 ("ja.*[._]sjis" . japanese-shift-jis)
2376 ("jpn" . japanese-shift-jis) ; MS-Windows uses this.
2377 ))
2378 "List of pairs of locale regexps and preferred coding systems.
2379 The first element whose locale regexp matches the start of a downcased locale
2380 specifies the coding system to prefer when using that locale.
2381 This coding system is used if the locale specifies a specific charset.")
2383 (defun locale-name-match (key alist)
2384 "Search for KEY in ALIST, which should be a list of regexp-value pairs.
2385 Return the value corresponding to the first regexp that matches the
2386 start of KEY, or nil if there is no match."
2387 (let (element)
2388 (while (and alist (not element))
2390 (setq element (car alist)))
2391 (setq alist (cdr alist)))
2392 (cdr element)))
2394 (defun locale-charset-match-p (charset1 charset2)
2395 "Whether charset names (strings) CHARSET1 and CHARSET2 are equivalent.
2396 Matching is done ignoring case and any hyphens and underscores in the
2397 names. E.g. `ISO_8859-1' and `iso88591' both match `iso-8859-1'."
2398 (setq charset1 (replace-regexp-in-string "[-_]" "" charset1))
2399 (setq charset2 (replace-regexp-in-string "[-_]" "" charset2))
2400 (eq t (compare-strings charset1 nil nil charset2 nil nil t)))
2402 (defvar locale-charset-alist nil
2403 "Coding system alist keyed on locale-style charset name.
2404 Used by `locale-charset-to-coding-system'.")
2406 (defun locale-charset-to-coding-system (charset)
2407 "Find coding system corresponding to CHARSET.
2408 CHARSET is any sort of non-Emacs charset name, such as might be used
2409 in a locale codeset, or elsewhere. It is matched to a coding system
2410 first by case-insensitive lookup in `locale-charset-alist'. Then
2411 matches are looked for in the coding system list, treating case and
2412 the characters `-' and `_' as insignificant. The coding system base
2414 `iso-latin-2' is returned."
2415 (or (car (assoc-string charset locale-charset-alist t))
2416 (let ((cs coding-system-alist)
2417 c)
2418 (while (and (not c) cs)
2419 (if (locale-charset-match-p charset (caar cs))
2420 (setq c (intern (caar cs)))
2421 (pop cs)))
2422 (if c (coding-system-base c)))))
2424 ;; Fixme: This ought to deal with the territory part of the locale
2425 ;; too, for setting things such as calendar holidays, ps-print paper
2426 ;; size, spelling dictionary.
2428 (defun locale-translate (locale)
2429 "Expand LOCALE according to `locale-translation-file-name', if possible.
2431 (if locale-translation-file-name
2432 (with-temp-buffer
2433 (set-buffer-multibyte nil)
2434 (insert-file-contents locale-translation-file-name)
2435 (if (re-search-forward
2437 (buffer-substring (point) (line-end-position))
2438 locale))
2439 locale))
2441 (defun set-locale-environment (&optional locale-name frame)
2442 "Set up multi-lingual environment for using LOCALE-NAME.
2443 This sets the language environment, the coding system priority,
2444 the default input method and sometimes other things.
2446 LOCALE-NAME should be a string which is the name of a locale supported
2447 by the system. Often it is of the form xx_XX.CODE, where xx is a
2448 language, XX is a country, and CODE specifies a character set and
2450 a locale for Japanese in Japan using the `japanese-iso-8bit'
2451 coding-system. The name may also have a modifier suffix, e.g. `@euro'
2452 or `@cyrillic'.
2454 If LOCALE-NAME is nil, its value is taken from the environment
2455 variables LC_ALL, LC_CTYPE and LANG (the first one that is set).
2457 The locale names supported by your system can typically be found in a
2458 directory named `/usr/share/locale' or `/usr/lib/locale'. LOCALE-NAME
2459 will be translated according to the table specified by
2460 `locale-translation-file-name'.
2462 If FRAME is non-nil, only set the keyboard coding system and the
2463 terminal coding system for the terminal of that frame, and don't
2464 touch session-global parameters like the language environment.
2466 See also `locale-charset-language-names', `locale-language-names',
2467 `locale-preferred-coding-systems' and `locale-coding-system'."
2468 (interactive "sSet environment for locale: ")
2470 ;; Do this at runtime for the sake of binaries possibly transported
2471 ;; to a system without X.
2472 (setq locale-translation-file-name
2473 (let ((files
2474 '("/usr/share/X11/locale/locale.alias" ; e.g. X11R7
2475 "/usr/lib/X11/locale/locale.alias" ; e.g. X11R6.4
2476 "/usr/X11R6/lib/X11/locale/locale.alias" ; XFree86, e.g. RedHat 4.2
2477 "/usr/openwin/lib/locale/locale.alias" ; e.g. Solaris 2.6
2478 ;;
2479 ;; The following name appears after the X-related names above,
2480 ;; since the X-related names are what X actually uses.
2481 "/usr/share/locale/locale.alias" ; GNU/Linux sans X
2482 )))
2483 (while (and files (not (file-exists-p (car files))))
2484 (setq files (cdr files)))
2485 (car files)))
2487 (let ((locale locale-name))
2489 (unless locale
2490 ;; Use the first of these three environment variables
2491 ;; that has a nonempty value.
2492 (let ((vars '("LC_ALL" "LC_CTYPE" "LANG")))
2493 (while (and vars
2494 (= 0 (length locale))) ; nil or empty string
2495 (setq locale (getenv (pop vars) frame)))))
2497 (when locale
2498 (setq locale (locale-translate locale))
2500 ;; Leave the system locales alone if the caller did not specify
2501 ;; an explicit locale name, as their defaults are set from
2502 ;; LC_MESSAGES and LC_TIME, not LC_CTYPE, and the user might not
2503 ;; want to set them to the same value as LC_CTYPE.
2504 (when locale-name
2505 (setq system-messages-locale locale)
2506 (setq system-time-locale locale))
2508 (if (string-match "^[a-z][a-z]" locale)
2509 (setq current-iso639-language (intern (match-string 0 locale)))))
2511 (setq woman-locale
2512 (or system-messages-locale
2513 (let ((msglocale (getenv "LC_MESSAGES" frame)))
2514 (if (zerop (length msglocale))
2515 locale
2516 (locale-translate msglocale)))))
2518 (when locale
2519 (setq locale (downcase locale))
2521 (let ((language-name
2522 (locale-name-match locale locale-language-names))
2523 (charset-language-name
2524 (locale-name-match locale locale-charset-language-names))
2525 (default-eol-type (coding-system-eol-type
2526 default-buffer-file-coding-system))
2527 (coding-system
2528 (or (locale-name-match locale locale-preferred-coding-systems)
2529 (when locale
2531 (locale-charset-to-coding-system
2532 (match-string 1 locale)))))))
2534 (if (consp language-name)
2535 ;; locale-language-names specify both lang-env and coding.
2536 ;; But, what specified in locale-preferred-coding-systems
2537 ;; has higher priority.
2538 (setq coding-system (or coding-system
2539 (nth 1 language-name))
2540 language-name (car language-name))
2541 ;; Otherwise, if locale is not listed in locale-language-names,
2542 ;; use what listed in locale-charset-language-names.
2543 (if (not language-name)
2544 (setq language-name charset-language-name)))
2546 ;; If a specific EOL conversion was specified in the default
2547 ;; buffer-file-coding-system, preserve it in the coding system
2548 ;; we will be using from now on.
2549 (if (and (memq default-eol-type '(0 1 2 unix dos mac))
2550 coding-system
2551 (coding-system-p coding-system))
2552 (setq coding-system (coding-system-change-eol-conversion
2553 coding-system default-eol-type)))
2555 (when language-name
2557 ;; Set up for this character set. This is now the right way
2558 ;; to do it for both unibyte and multibyte modes.
2559 (unless frame
2560 (set-language-environment language-name))
2562 ;; If default-enable-multibyte-characters is nil,
2563 ;; we are using single-byte characters,
2564 ;; so the display table and terminal coding system are irrelevant.
2565 (when default-enable-multibyte-characters
2566 (set-display-table-and-terminal-coding-system
2567 language-name coding-system frame))
2569 ;; Set the `keyboard-coding-system' if appropriate (tty
2570 ;; only). At least X and MS Windows can generate
2571 ;; multilingual input.
2572 ;; XXX This was disabled unless `window-system', but that
2573 ;; leads to buggy behavior when a tty frame is opened
2574 ;; later. Setting the keyboard coding system has no adverse
2575 ;; effect on X, so let's do it anyway. -- Lorentey
2576 (let ((kcs (or coding-system
2577 (car (get-language-info language-name
2578 'coding-system)))))
2579 (if kcs (set-keyboard-coding-system kcs frame)))
2581 (unless frame
2582 (setq locale-coding-system
2583 (car (get-language-info language-name 'coding-priority)))))
2585 (when (and (not frame)
2586 coding-system
2587 (not (coding-system-equal coding-system
2588 locale-coding-system)))
2589 (prefer-coding-system coding-system)
2590 ;; Fixme: perhaps prefer-coding-system should set this too.
2591 ;; But it's not the time to do such a fundamental change.
2592 (setq default-sendmail-coding-system coding-system)
2593 (setq locale-coding-system coding-system))))
2595 ;; On Windows, override locale-coding-system,
2596 ;; default-file-name-coding-system, keyboard-coding-system,
2597 ;; terminal-coding-system with system codepage.
2598 (when (boundp 'w32-ansi-code-page)
2599 (let ((code-page-coding (intern (format "cp%d" w32-ansi-code-page))))
2600 (when (coding-system-p code-page-coding)
2601 (unless frame (setq locale-coding-system code-page-coding))
2602 (set-keyboard-coding-system code-page-coding frame)
2603 (set-terminal-coding-system code-page-coding frame)
2604 ;; Set default-file-name-coding-system last, so that Emacs
2605 ;; doesn't try to use cpNNNN when it defines keyboard and
2606 ;; terminal encoding. That's because the above two lines
2607 ;; will want to load code-pages.el, where cpNNNN are
2608 ;; defined; if default-file-name-coding-system were set to
2609 ;; cpNNNN while these two lines run, Emacs will want to use
2610 ;; it for encoding the file name it wants to load. And that
2611 ;; will fail, since cpNNNN is not yet usable until
2612 ;; code-pages.el finishes loading.
2613 (setq default-file-name-coding-system code-page-coding))))
2615 (when (eq system-type 'darwin)
2616 ;; On Darwin, file names are always encoded in utf-8, no matter
2617 ;; the locale.
2618 (setq default-file-name-coding-system 'utf-8)
2619 ;; Mac OS X's Terminal.app by default uses utf-8 regardless of
2620 ;; the locale.
2621 (when (and (null window-system)
2622 (equal (getenv "TERM_PROGRAM" frame) "Apple_Terminal"))
2623 (set-terminal-coding-system 'utf-8)
2624 (set-keyboard-coding-system 'utf-8)))
2626 ;; Default to A4 paper if we're not in a C, POSIX or US locale.
2627 ;; (See comments in Flocale_info.)
2628 (unless frame
2629 (let ((locale locale)
2630 (paper (locale-info 'paper)))
2631 (if paper
2632 ;; This will always be null at the time of writing.
2633 (cond
2634 ((equal paper '(216 279))
2635 (setq ps-paper-type 'letter))
2636 ((equal paper '(210 297))
2637 (setq ps-paper-type 'a4)))
2638 (let ((vars '("LC_ALL" "LC_PAPER" "LANG")))
2639 (while (and vars (= 0 (length locale)))
2640 (setq locale (getenv (pop vars) frame))))
2641 (when locale
2642 ;; As of glibc 2.2.5, these are the only US Letter locales,
2643 ;; and the rest are A4.
2644 (setq ps-paper-type
2645 (or (locale-name-match locale '(("c$" . letter)
2646 ("posix$" . letter)
2647 (".._us" . letter)
2648 (".._pr" . letter)
2649 (".._ca" . letter)
2650 ("enu$" . letter) ; Windows
2651 ("esu$" . letter)
2652 ("enc$" . letter)
2653 ("frc$" . letter)))
2654 'a4)))))))
2655 nil)
2656 \f
2657 ;;; Character property
2659 ;; Each element has the form (PROP . TABLE).
2660 ;; PROP is a symbol representing a character property.
2661 ;; TABLE is a char-table containing the property value for each character.
2662 ;; TABLE may be a name of file to load to build a char-table.
2663 ;; Don't modify this variable directly but use `define-char-code-property'.
2665 (defvar char-code-property-alist nil
2666 "Alist of character property name vs char-table containing property values.
2667 Internal use only.")
2669 (put 'char-code-property-table 'char-table-extra-slots 5)
2671 (defun define-char-code-property (name table &optional docstring)
2672 "Define NAME as a character code property given by TABLE.
2673 TABLE is a char-table of purpose `char-code-property-table' with
2674 these extra slots:
2675 1st: NAME.
2676 2nd: Function to call to get a property value of a character.
2677 It is called with three arguments CHAR, VAL, and TABLE, where
2678 CHAR is a character, VAL is the value of (aref TABLE CHAR).
2679 3rd: Function to call to put a property value of a character.
2680 It is called with the same arguments as above.
2681 4th: Function to call to get a description string of a property value.
2682 It is called with one argument VALUE, a property value.
2683 5th: Data used by the above functions.
2685 TABLE may be a name of file to load to build a char-table. The
2686 file should contain a call of `define-char-code-property' with a
2687 char-table of the above format as the argument TABLE.
2689 TABLE may also be nil, in which case no property value is pre-assigned.
2691 Optional 3rd argument DOCSTRING is a documentation string of the property.
2693 See also the documentation of `get-char-code-property' and
2694 `put-char-code-property'."
2695 (or (symbolp name)
2696 (error "Not a symbol: %s" name))
2697 (if (char-table-p table)
2698 (or (and (eq (char-table-subtype table) 'char-code-property-table)
2699 (eq (char-table-extra-slot table 0) name))
2700 (error "Invalid char-table: %s" table))
2701 (or (stringp table)
2702 (error "Not a char-table nor a file name: %s" table)))
2703 (let ((slot (assq name char-code-property-alist)))
2704 (if slot
2705 (setcdr slot table)
2706 (setq char-code-property-alist
2707 (cons (cons name table) char-code-property-alist))))
2708 (put name 'char-code-property-documentation docstring))
2710 (defvar char-code-property-table
2711 (make-char-table 'char-code-property-table)
2712 "Char-table containing a property list of each character code.
2713 This table is used for properties not listed in `char-code-property-alist'.
2714 See also the documentation of `get-char-code-property' and
2715 `put-char-code-property'.")
2717 (defun get-char-code-property (char propname)
2718 "Return the value of CHAR's PROPNAME property."
2719 (let ((slot (assq propname char-code-property-alist)))
2720 (if slot
2721 (let (table value func)
2722 (if (stringp (cdr slot))
2723 (load (cdr slot) nil t))
2724 (setq table (cdr slot)
2725 value (aref table char)
2726 func (char-table-extra-slot table 1))
2727 (if (functionp func)
2728 (setq value (funcall func char value table)))
2729 value)
2730 (plist-get (aref char-code-property-table char) propname))))
2732 (defun put-char-code-property (char propname value)
2733 "Store CHAR's PROPNAME property with VALUE.
2734 It can be retrieved with `(get-char-code-property CHAR PROPNAME)'."
2735 (let ((slot (assq propname char-code-property-alist)))
2736 (if slot
2737 (let (table func)
2738 (if (stringp (cdr slot))
2739 (load (cdr slot) nil t))
2740 (setq table (cdr slot)
2741 func (char-table-extra-slot table 2))
2742 (if (functionp func)
2743 (funcall func char value table)
2744 (aset table char value)))
2745 (let* ((plist (aref char-code-property-table char))
2746 (x (plist-put plist propname value)))
2747 (or (eq x plist)
2748 (aset char-code-property-table char x))))
2749 value))
2751 (defun char-code-property-description (prop value)
2752 "Return a description string of character property PROP's value VALUE.
2753 If there's no description string for VALUE, return nil."
2754 (let ((slot (assq prop char-code-property-alist)))
2755 (if slot
2756 (let (table func)
2757 (if (stringp (cdr slot))
2758 (load (cdr slot) nil t))
2759 (setq table (cdr slot)
2760 func (char-table-extra-slot table 3))
2761 (if (functionp func)
2762 (funcall func value))))))
2764 \f
2765 ;; Pretty description of encoded string
2767 ;; Alist of ISO 2022 control code vs the corresponding mnemonic string.
2768 (defvar iso-2022-control-alist
2776 (defun encoded-string-description (str coding-system)
2777 "Return a pretty description of STR that is encoded by CODING-SYSTEM."
2778 (setq str (string-as-unibyte str))
2779 (mapconcat
2780 (if (and coding-system (eq (coding-system-type coding-system) 'iso-2022))
2781 ;; Try to get a pretty description for ISO 2022 escape sequences.
2782 (function (lambda (x) (or (cdr (assq x iso-2022-control-alist))
2783 (format "#x%02X" x))))
2784 (function (lambda (x) (format "#x%02X" x))))
2785 str " "))
2787 (defun encode-coding-char (char coding-system &optional charset)
2788 "Encode CHAR by CODING-SYSTEM and return the resulting string.
2789 If CODING-SYSTEM can't safely encode CHAR, return nil.
2790 The 3rd optional argument CHARSET, if non-nil, is a charset preferred
2791 on encoding."
2792 (let* ((str1 (string-as-multibyte (string char)))
2793 (str2 (string-as-multibyte (string char char)))
2794 (found (find-coding-systems-string str1))
2795 enc1 enc2 i1 i2)
2796 (if (and (consp found)
2797 (eq (car found) 'undecided))
2798 str1
2799 (when (memq (coding-system-base coding-system) found)
2800 ;; We must find the encoded string of CHAR. But, just encoding
2801 ;; CHAR will put extra control sequences (usually to designate
2802 ;; ASCII charset) at the tail if type of CODING is ISO 2022.
2803 ;; To exclude such tailing bytes, we at first encode one-char
2804 ;; string and two-char string, then check how many bytes at the
2805 ;; tail of both encoded strings are the same.
2807 (when charset
2808 (put-text-property 0 1 'charset charset str1)
2809 (put-text-property 0 2 'charset charset str2))
2810 (setq enc1 (encode-coding-string str1 coding-system)
2811 i1 (length enc1)
2812 enc2 (encode-coding-string str2 coding-system)
2813 i2 (length enc2))
2814 (while (and (> i1 0) (= (aref enc1 (1- i1)) (aref enc2 (1- i2))))
2815 (setq i1 (1- i1) i2 (1- i2)))
2817 ;; Now (substring enc1 i1) and (substring enc2 i2) are the same,
2818 ;; and they are the extra control sequences at the tail to
2819 ;; exclude.
2820 (substring enc2 0 i2)))))
2822 ;; Backwards compatibility. These might be better with :init-value t,
2823 ;; but that breaks loadup.
2824 (define-minor-mode unify-8859-on-encoding-mode
2825 "Obsolete."
2826 :group 'mule
2827 :global t)
2828 (define-minor-mode unify-8859-on-decoding-mode
2829 "Obsolete."
2830 :group 'mule
2831 :global t)
2833 (defvar nonascii-insert-offset 0 "This variable is obsolete.")
2834 (defvar nonascii-translation-table nil "This variable is obsolete.")
2836 (defvar ucs-names nil
2837 "Alist of cached (CHAR-NAME . CHAR-CODE) pairs.")
2839 (defun ucs-names ()
2840 "Return alist of (CHAR-NAME . CHAR-CODE) pairs cached in `ucs-names'."
2841 (or ucs-names
2842 (setq ucs-names
2843 (let (name names)
2844 (dotimes-with-progress-reporter (c #xEFFFF)
2845 "Loading Unicode character names..."
2846 (unless (or
2847 (and (>= c #x3400 ) (<= c #x4dbf )) ; CJK Ideograph Extension A
2848 (and (>= c #x4e00 ) (<= c #x9fff )) ; CJK Ideograph
2849 (and (>= c #xd800 ) (<= c #xfaff )) ; Private/Surrogate
2850 (and (>= c #x20000) (<= c #x2ffff)) ; CJK Ideograph Extension B
2851 )
2852 (if (setq name (get-char-code-property c 'name))
2853 (setq names (cons (cons name c) names)))
2854 (if (setq name (get-char-code-property c 'old-name))
2855 (setq names (cons (cons name c) names)))))
2856 names))))
2858 (defvar ucs-completions (lazy-completion-table ucs-completions ucs-names)
2859 "Lazy completion table for completing on Unicode character names.")
2860 (put 'ucs-completions 'risky-local-variable t)
2862 (defun read-char-by-name (prompt)
2863 "Read a character by its Unicode name or hex number string.
2864 Display PROMPT and read a string that represents a character by its
2865 Unicode property `name' or `old-name'. You can type a few of first
2866 letters of the Unicode name and use completion. This function also
2867 accepts a hexadecimal number of Unicode code point or a number in
2868 hash notation, e.g. #o21430 for octal, #x2318 for hex, or #10r8984
2869 for decimal. Returns a character as a number."
2870 (let* ((completion-ignore-case t)
2871 (input (completing-read prompt ucs-completions)))
2872 (cond
2873 ((string-match-p "^[0-9a-fA-F]+$" input)
2874 (string-to-number input 16))
2875 ((string-match-p "^#" input)
2876 (read input))
2877 (t
2878 (cdr (assoc-string input (ucs-names) t))))))
2880 (defun ucs-insert (arg)
2881 "Insert a character of the given Unicode code point.
2882 Interactively, prompts for a Unicode character name or a hex number
2883 using `read-char-by-name'."
2884 (interactive (list (read-char-by-name "Unicode (name or hex): ")))
2885 (if (stringp arg)
2886 (setq arg (string-to-number arg 16)))
2887 (cond
2888 ((not (integerp arg))
2889 (error "Not a Unicode character code: %S" arg))
2890 ((or (< arg 0) (> arg #x10FFFF))
2891 (error "Not a Unicode character code: 0x%X" arg)))
2892 (insert-and-inherit arg))
2896 ;; arch-tag: b382c432-4b36-460e-bf4c-05efd0bb18dc
2897 ;;; mule-cmds.el ends here