| blob | a91c12bcaf20d27baec1381c388d112744ac1778 |
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)
14 ;; any later version.
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.
26 ;;; Commentary:
28 ;;; Code:
34 ;;; MULE related key bindings and menus.
39 ;; Keep "C-x C-m ..." for mule specific commands.
72 t)
75 t)
78 t)
83 t)
92 t)
95 t)
98 describe-language-environment-map
100 t)
104 t)
107 t)
114 t)
119 t)
122 t)
126 t)
129 :enable buffer-file-name
131 t)
135 t)
138 t)
143 t)
148 t)
151 t)
156 t)
161 t)
167 t)
171 [Default] '(menu-item "Default" setup-specified-language-environment))
173 (define-key describe-language-environment-map
176 ;; This should be a single character key binding because users use it
177 ;; very frequently while editing multilingual text. Now we can use
178 ;; only two such keys: "\C-\\" and "\C-^", but the latter is not
179 ;; convenient because it requires shifting on most keyboards. An
180 ;; alternative is "\C-\]" which is now bound to `abort-recursive-edit'
181 ;; but it won't be used that frequently.
184 ;; This is no good because people often type Shift-SPC
185 ;; meaning to type SPC. -- rms.
186 ;; ;; Here's an alternative key binding for X users (Shift-SPACE).
187 ;; (define-key global-map [?\S- ] 'toggle-input-method)
189 ;;; Mule related hyperlinks.
197 ;; Note starting with word-syntax character:
201 "Return a coding system which differs from CODING-SYSTEM in eol conversion.
202 The returned coding system converts end-of-line by EOL-TYPE
203 but text as the same way as CODING-SYSTEM.
204 EOL-TYPE should be `unix', `dos', `mac', or nil.
205 If EOL-TYPE is nil, the returned coding system detects
206 how end-of-line is formatted automatically while decoding.
208 EOL-TYPE can be specified by an integer 0, 1, or 2.
209 They means `unix', `dos', and `mac' respectively."
218 coding-system
222 base
224 coding-system
230 "Return a coding system which differs from CODING-SYSTEM in text conversion.
231 The returned coding system converts text by CODING
232 but end-of-line as the same way as CODING-SYSTEM.
233 If CODING is nil, the returned coding system detects
234 how text is formatted automatically while decoding."
241 "Change whether this buffer uses multibyte characters.
242 With arg, use multibyte characters if the arg is positive.
244 Note that this command does not convert the byte contents of
245 the buffer; it only changes the way those bytes are interpreted.
246 In general, therefore, this command *changes* the sequence of
247 characters that the current buffer contains.
249 We suggest you avoid using use this command unless you know what you
250 are doing. If you use it by mistake, and the buffer is now displayed
251 wrong, use this command again to toggle back to the right mode."
260 "Display the HELLO file which list up many languages and characters."
262 ;; We have to decode the file in any environment.
268 "Execute an I/O command using the specified coding system."
272 t))
273 buffer-file-coding-system)))
278 default))))
282 prefix)
287 ;; Process keys bound in `universal-argument-map'.
293 ;; Have to bind `last-command-char' here so that
294 ;; `digit-argument', for instance, can compute the
295 ;; prefix arg.
299 ;; This is the final call to `universal-argument-other-key', which
300 ;; set's the final `prefix-arg.
304 ;; Read the command to execute with the given prefix arg.
317 "Set default value of various coding systems to CODING-SYSTEM.
318 This sets the following coding systems:
319 o coding system of a newly created buffer
320 o default coding system for subprocess I/O
321 This also sets the following values:
322 o default value used as `file-name-coding-system' for converting file names.
323 o default value for the command `set-terminal-coding-system' (not on MSDOS)
324 o default value for the command `set-keyboard-coding-system'."
333 ;; The file-name coding system on Darwin systems is always utf-8.
335 ;; If coding-system is nil, honor that on MS-DOS as well, so
336 ;; that they could reset the terminal coding system.
340 ;; Preserve eol-type from existing default-process-coding-systems.
341 ;; On non-unix-like systems in particular, these may have been set
342 ;; carefully by the user, or by the startup code, to deal with the
343 ;; users shell appropriately, so should not be altered by changing
344 ;; language environment.
358 "Add CODING-SYSTEM at the front of the priority list for automatic detection.
359 This also sets the following coding systems:
360 o coding system of a newly created buffer
361 o default coding system for subprocess I/O
362 This also sets the following values:
363 o default value used as `file-name-coding-system' for converting file names.
364 o default value for the command `set-terminal-coding-system' (not on MSDOS)
365 o default value for the command `set-keyboard-coding-system'
367 If CODING-SYSTEM specifies a certain type of EOL conversion, the coding
368 systems set by this function will use that type of EOL conversion.
370 This command does not change the default value of terminal coding system
371 for MS-DOS terminal, because DOS terminals only support a single coding
372 system, and Emacs automatically sets the default to that coding system at
373 startup.
375 A coding system that requires automatic detection of text
376 encoding (e.g. undecided, unix) can't be preferred.
378 See also `coding-category-list' and `coding-system-category'."
386 ;; CODING-SYSTEM is no-conversion or undecided.
389 ;; Changing the binding of a coding category requires this call.
392 ;; We must change the order.
396 base coding-system))
397 ;; 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.
415 If a coding system is most preferred, it has the highest priority.
416 Otherwise, a coding system corresponds to some MIME charset has higher
417 priorities. Among them, a coding system included in `coding-system'
418 key of the current language environment has higher priorities. See
419 also the documentation of `language-info-alist'.
421 If the variable `sort-coding-systems-predicate' (which see) is
422 non-nil, it is used to sort CODINGS in the different way than above."
426 coding-category-list))
433 ;; We calculate the priority number 0..255 by
434 ;; using the 8 bits PMMLCEII as this:
435 ;; P: 1 iff most preferred.
436 ;; MM: greater than 0 iff mime-charset.
437 ;; L: 1 iff one of the current lang. env.'s codings.
438 ;; C: 1 iff one of codings listed in the category list.
439 ;; E: 1 iff 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.
465 ;; For ISO based coding systems, prefer
466 ;; one that doesn't use escape sequences.
473 0
481 "Return a list of proper coding systems to encode a text between FROM and TO.
482 If FROM is a string, find coding systems in that instead of the buffer.
483 All coding systems in the list can safely encode any multibyte characters
484 in the text.
486 If the text contains no multibyte characters, return a list of a single
487 element `undecided'."
490 ;; The text contains only ASCII characters. Any coding
491 ;; systems are safe.
493 ;; We need copy-sequence because sorting will alter the argument.
497 "Return a list of proper coding systems to encode STRING.
498 All coding systems in the list can safely encode any multibyte characters
499 in STRING.
501 If STRING contains no multibyte characters, return a list of a single
502 element `undecided'."
506 "Return a list of proper coding systems to encode characters of CHARSETS.
507 CHARSETS is a list of character sets.
508 It actually checks at most the first 96 characters of each charset.
509 So, if a charset of dimension two is included in CHARSETS, the value may
510 contain a coding system that can't encode all characters of the charset."
520 charset l str)
544 codings))))
547 "Find multibyte characters in the region specified by FROM and TO.
548 If FROM is a string, find multibyte characters in the string.
549 The return value is an alist of the following format:
550 ((CHARSET COUNT CHAR ...) ...)
551 where
552 CHARSET is a character set,
553 COUNT is a number of characters,
554 CHARs are the characters found from the character set.
555 Optional 3rd arg MAXCOUNT limits how many CHARs are put in the above list.
556 Optional 4th arg EXCLUDE is a list of character sets to be ignored.
558 For invalid characters, CHARs are actually strings."
560 charset char)
569 eight-bit-control eight-bit-graphic))
601 "Search forward from point for a character that is not encodable.
602 It asks which coding system to check.
603 If such a character is found, set point after that character.
604 Otherwise, don't move point.
606 When called from a program, the value is a position of the found character,
607 or nil if all characters are encodable."
612 default))))
617 pos))
621 "Most recent coding system explicitly specified by the user when asked.
622 This variable is set whenever Emacs asks the user which coding system
623 to use in order to write a file. If you set it to nil explicitly,
624 then call `write-region', then afterward this variable will be non-nil
628 "If non-nil, a function to control the behaviour of coding system selection.
629 The meaning is the same as the argument ACCEPT-DEFAULT-P of the
630 function `select-safe-coding-system' (which see). This variable
635 "Select interactively a coding system for the region FROM ... TO.
636 FROM can be a string, as in `write-region'.
637 CODINGS is the list of base coding systems known to be safe for this region,
638 typically obtained with `find-coding-systems-region'.
639 UNSAFE is a list of coding systems known to be unsafe for this region.
640 REJECTED is a list of coding systems which were safe but for some reason
641 were not recommended in the particular context.
642 DEFAULT is the coding system to use by default in the query."
643 ;; At first, if some defaults are unsafe, record at most 11
644 ;; problematic characters and their positions for them by turning
645 ;; (CODING ...)
646 ;; into
647 ;; ((CODING (POS . CHAR) (POS . CHAR) ...) ...)
662 unsafe)))
664 ;; Change each safe coding system to the corresponding
665 ;; mime-charset name if it is also a coding system. Such a name
666 ;; is more friendly to users.
668 mime-charset)
675 ;; Don't offer variations with locking shift, which you
676 ;; basically never want.
685 ;; Remove raw-text, emacs-mule and no-conversion unless nothing
686 ;; else is available.
695 coding-system)
697 ;; If some defaults are unsafe, make sure the offending
698 ;; buffer is displayed.
702 unsafe))))
703 ;; Then ask users to select one from CODINGS while showing
704 ;; the reason why none of the defaults are not used.
713 "These default coding systems were tried to encode"
728 but it is not recommended for encoding text in this context,
762 'help-echo
763 "mouse-2, RET: jump to this character"
764 'help-function func1
767 "..."
769 'help-echo
770 "mouse-2, RET: next unencodable character"
771 'help-function func2
791 ;; Read a coding system.
795 default))
800 coding-system))
803 accept-default-p file)
804 "Ask a user to select a safe coding system from candidates.
805 The candidates of coding systems which can safely encode a text
806 between FROM and TO are shown in a popup window. Among them, the most
807 proper one is suggested as the default.
809 The list of `buffer-file-coding-system' of the current buffer,
810 the `default-buffer-file-coding-system', and the
811 most preferred coding system (if it corresponds to a MIME charset) is
812 treated as the default coding system list. Among them, the first one
813 that safely encodes the text is normally selected silently and
814 returned without any user interaction. See also the command
815 `prefer-coding-system'.
817 However, the user is queried if the chosen coding system is
818 inconsistent with what would be selected by `set-auto-coding' from
819 coding cookies &c. if the contents of the region were read from a
820 file. (That could lead to data corruption in a file subsequently
821 re-visited and edited.)
823 Optional 3rd arg DEFAULT-CODING-SYSTEM specifies a coding system or a
824 list of coding systems to be prepended to the default coding system
825 list. However, if DEFAULT-CODING-SYSTEM is a list and the first
826 element is t, the cdr part is used as the defualt coding system list,
827 i.e. `buffer-file-coding-system', `default-buffer-file-coding-system',
828 and the most preferred coding system are not used.
830 Optional 4th arg ACCEPT-DEFAULT-P, if non-nil, is a function to
831 determine the acceptability of the silently selected coding system.
832 It is called with that coding system, and should return nil if it
833 should not be silently selected and thus user interaction is required.
835 Optional 5th arg FILE is the file name to use for this purpose.
836 That is different from `buffer-file-name' when handling `write-region'
837 \(for example).
839 The variable `select-safe-coding-system-accept-default-p', if
840 non-nil, overrides ACCEPT-DEFAULT-P.
842 Kludgy feature: if FROM is a string, the string is the target text,
843 and TO is ignored."
853 ;; Change elements of the list to (coding . base-coding).
856 default-coding-system))
858 ;; From now on, the list of defaults is reversed.
862 ;; If buffer-file-coding-system is not nil nor undecided, append it
863 ;; to the defaults.
869 default-coding-system))))
871 ;; If default-buffer-file-coding-system is not nil nor undecided,
872 ;; append it to the defaults.
878 default-coding-system))))
880 ;; If the most preferred coding system has the property mime-charset,
881 ;; append it to the defaults.
883 preferred base)
891 default-coding-system)))))
898 safe rejected unsafe)
900 ;; Any coding system is ok.
902 ;; Classify the defaults into safe, rejected, and unsafe.
913 ;; If all the defaults failed, ask a user.
926 ;; Check we're not inconsistent with what `coding:' spec &c would
927 ;; give when file is re-read.
928 ;; But don't do this if we explicitly ignored the cookie
929 ;; by using `find-file-literally'.
931 find-file-literally
941 ;; Merge coding-system and auto-cs as far as possible.
950 coding-system auto-cs))
953 auto-cs coding-system))))
957 coding-system eol-type-2)))
960 auto-cs eol-type-1)))))))
963 ;; Don't barf if writing a compressed file, say.
964 ;; This check perhaps isn't ideal, but is probably
965 ;; the best thing to do.
970 %s specified by file contents. Really save (else edit coding cookies \
973 coding-system))
978 "Return a coding system to encode the outgoing message of the current buffer.
979 It at first tries the first coding system found in these variables
980 in this order:
981 (1) local value of `buffer-file-coding-system'
982 (2) value of `sendmail-coding-system'
983 (3) value of `default-sendmail-coding-system'
984 (4) value of `default-buffer-file-coding-system'
985 If the found coding system can't encode the current buffer,
986 or none of them are bound to a coding system,
987 it asks the user to select a proper coding system."
989 buffer-file-coding-system)
990 sendmail-coding-system
991 default-sendmail-coding-system
992 default-buffer-file-coding-system)))
994 ;; We should never use no-conversion for outgoing mail.
1000 coding)))
1001 \f
1002 ;;; Language support stuff.
1005 "Alist of language environment definitions.
1006 Each element looks like:
1007 (LANGUAGE-NAME . ((KEY . INFO) ...))
1008 where LANGUAGE-NAME is a string, the name of the language environment,
1009 KEY is a symbol denoting the kind of information, and
1010 INFO is the data associated with KEY.
1011 Meaningful values for KEY include
1013 documentation value is documentation of what this language environment
1014 is meant for, and how to use it.
1015 charset value is a list of the character sets used by this
1016 language environment.
1017 sample-text value is an expression which is evalled to generate
1018 a line of text written using characters appropriate
1019 for this language environment.
1020 setup-function value is a function to call to switch to this
1021 language environment.
1022 exit-function value is a function to call to leave this
1023 language environment.
1024 coding-system value is a list of coding systems that are good
1025 for saving text written in this language environment.
1026 This list serves as suggestions to the user;
1027 in effect, as a kind of documentation.
1028 coding-priority value is a list of coding systems for this language
1029 environment, in order of decreasing priority.
1030 This is used to set up the coding system priority
1031 list when you switch to this language environment.
1032 nonascii-translation
1033 value is a translation table to be set in the
1034 variable `nonascii-translation-table' in this
1035 language environment, or a character set from
1036 which `nonascii-insert-offset' is calculated.
1037 input-method value is a default input method for this language
1038 environment.
1039 features value is a list of features requested in this
1040 language environment.
1041 ctext-non-standard-encodings
1042 value is a list of non-standard encoding
1043 names used in extended segments of CTEXT.
1044 See the variable
1045 `ctext-non-standard-encodings' for more
1046 detail.
1048 The following keys take effect only when multibyte characters are
1049 globally disabled, i.e. the value of `default-enable-multibyte-characters'
1050 is nil.
1052 unibyte-syntax value is a library name to load to set
1053 unibyte 8-bit character syntaxes for this
1054 language environment.
1056 unibyte-display value is a coding system to encode characters
1057 for the terminal. Characters in the range
1058 of 160 to 255 display not as octal escapes,
1059 but as non-ASCII characters in this language
1063 "Return information listed under KEY for language environment LANG-ENV.
1064 KEY is a symbol denoting the kind of information.
1065 For a list of useful values for KEY and their meanings,
1066 see `language-info-alist'."
1074 "Modify part of the definition of language environment LANG-ENV.
1075 Specifically, this stores the information INFO under KEY
1076 in the definition of this language environment.
1077 KEY is a symbol denoting the kind of information.
1078 INFO is the value for that information.
1080 For a list of useful values for KEY and their meanings,
1081 see `language-info-alist'."
1095 ;; Update the custom-type of `current-language-environment'.
1103 "Store ALIST as the definition of language environment LANG-ENV.
1104 ALIST is an alist of KEY and INFO values. See the documentation of
1105 `language-info-alist' for the meanings of KEY and INFO.
1107 Optional arg PARENTS is a list of parent menu names; it specifies
1108 where to put this language environment in the
1109 Describe Language Environment and Set Language Environment menus.
1111 in the European submenu in each of those two menus."
1118 map parent-symbol parent prompt)
1124 ;; This prompt string is for define-prefix-command, so
1125 ;; that the map it creates will be suitable for a menu.
1146 ;; Set up menu items for this language env.
1159 "Read a language environment name which has information for KEY.
1160 If KEY is nil, read any language environment.
1161 Prompt with PROMPT. DEFAULT is the default choice of language environment.
1162 This returns a language environment name as a string."
1165 language-info-alist
1168 t nil nil default)))
1172 name)))
1173 \f
1174 ;;; Multilingual input methods.
1176 "LEIM: Libraries of Emacs Input Methods."
1180 "Name of LEIM list file.
1181 This file contains a list of libraries of Emacs input methods (LEIM)
1182 in the format of Lisp expression for registering each input method.
1186 ";;; %s -- list of LEIM (Library of Emacs Input Method) -*-coding: iso-2022-7bit;-*-
1187 ;;
1188 ;; This file contains a list of LEIM (Library of Emacs Input Method)
1189 ;; methods in the same directory as this file. Loading this file
1190 ;; registers all the input methods in Emacs.
1191 ;;
1192 ;; Each entry has the form:
1193 ;; (register-input-method
1194 ;; INPUT-METHOD LANGUAGE-NAME ACTIVATE-FUNC
1195 ;; TITLE DESCRIPTION
1196 ;; ARG ...)
1197 ;; See the function `register-input-method' for the meanings of the arguments.
1198 ;;
1199 ;; If this directory is included in load-path, Emacs automatically
1200 ;; loads this file at startup time.
1202 "
1203 leim-list-file-name)
1207 "Regexp matching head of each entry in LEIM list file.
1212 "List of functions to call to update LEIM list file.
1216 "Update LEIM list file in directories DIRS."
1223 "The current input method for multilingual text.
1234 "*Default input method for multilingual text (a string).
1235 This is the input method activated automatically by the command
1240 :completion-ignore-case t
1241 :complete-function widget-string-complete
1242 :completion-alist input-method-alist
1254 "Function to call for inactivating the current input method.
1255 Every input method should set this to an appropriate value when activated.
1256 This function is called with no argument.
1258 This function should never change the value of `current-input-method'.
1264 "Function to call for describing the current input method.
1270 "Alist of input method names vs how to use them.
1271 Each element has the form:
1272 (INPUT-METHOD LANGUAGE-ENV ACTIVATE-FUNC TITLE DESCRIPTION ARGS...)
1276 "Register INPUT-METHOD as an input method for language environment LANG-ENV.
1277 INPUT-METHOD and LANG-ENV are symbols or strings.
1279 The remaining arguments are:
1280 ACTIVATE-FUNC, TITLE, DESCRIPTION, and ARGS...
1281 ACTIVATE-FUNC is a function to call to activate this method.
1282 TITLE is a string to show in the mode line when this method is active.
1283 DESCRIPTION is a string describing this method and what it is good for.
1284 The ARGS, if any, are passed as arguments to ACTIVATE-FUNC.
1285 All told, the arguments to ACTIVATE-FUNC are INPUT-METHOD and the ARGS.
1288 created at Emacs build time, registering all Quail input methods
1289 contained in the Emacs distribution.
1291 In case you want to register a new Quail input method by yourself, be
1292 careful to use the same input method title as given in the third
1293 parameter of `quail-define-package'. (If the values are different, the
1294 string specified in this function takes precedence.)
1296 The commands `describe-input-method' and `list-input-methods' need
1297 these duplicated values to show some information about input methods
1298 without loading the relevant Quail packages."
1311 "Read a name of input method from a minibuffer prompting with PROMPT.
1312 If DEFAULT is non-nil, use that as the default,
1313 and substitute it into PROMPT at the first `%s'.
1314 If INHIBIT-NULL is non-nil, null input signals an error.
1316 The return value is a string."
1320 ;; As it is quite normal to change input method in the
1321 ;; minibuffer, we must enable it even if
1322 ;; enable-recursive-minibuffers is currently nil.
1324 ;; This binding is necessary because input-method-history is
1325 ;; buffer local.
1327 nil t nil 'input-method-history
1328 default)))
1332 input-method
1337 "Switch to input method INPUT-METHOD for the current buffer.
1338 If some other input method is already active, turn it off first.
1339 If INPUT-METHOD is nil, deactivate any current input method."
1366 "Turn off the current input method."
1379 input-method-function nil
1380 current-input-method-title nil)
1384 "Select and activate input method INPUT-METHOD for the current buffer.
1385 This also sets the default input method to the one you specify.
1386 If INPUT-METHOD is nil, this function turns off the input method, and
1387 also causes you to be prompted for a name of an input method the next
1389 When called interactively, the optional arg INTERACTIVE is non-nil,
1390 which marks the variable `default-input-method' as set for Custom buffers.
1398 default t)
1399 t)))
1404 default-input-method)
1407 "Enable or disable multilingual text input method for the current buffer.
1408 Only one input method can be enabled at any time in a given buffer.
1410 The normal action is to enable an input method if none was
1411 enabled, and disable the current one otherwise. Which input method
1412 to enable can be determined in various ways--either the one most
1413 recently used, or the one specified by `default-input-method', or
1414 as a last resort by reading the name of an input method in the
1415 minibuffer.
1417 With a prefix argument, read an input method name with the minibuffer
1418 and enable that one. The default is the most recent input method specified
1419 \(not including the currently active input method, if any).
1421 When called interactively, the optional arg INTERACTIVE is non-nil,
1422 which marks the variable `default-input-method' as set for Custom buffers."
1436 default t))
1437 default))
1445 "Describe input method INPUT-METHOD."
1475 "Describe the input method currently in use.
1476 This is a subroutine for `describe-input-method'."
1482 current-input-method)
1487 "Read a multilingual string from minibuffer, prompting with string PROMPT.
1488 The input method selected last time is activated in minibuffer.
1489 If optional second arg INITIAL-INPUT is non-nil, insert it in the minibuffer
1490 initially.
1491 Optional 3rd argument INPUT-METHOD specifies the input method
1492 to be activated instead of the one selected last time. It is a symbol
1493 or a string."
1496 current-input-method
1497 default-input-method
1508 ;; Variables to control behavior of input methods. All input methods
1509 ;; should react to these variables.
1512 "*A flag to control extra guidance given by input methods.
1513 The value should be nil, t, `complex-only', or `default'.
1515 The extra guidance is done by showing list of available keys in echo
1516 area. When you use the input method in the minibuffer, the guidance
1517 is shown at the bottom short window (split from the existing window).
1519 If the value is t, extra guidance is always given, if the value is
1520 nil, extra guidance is always suppressed.
1522 If the value is `complex-only', only complex input methods such as
1523 `chinese-py' and `japanese' give extra guidance.
1525 If the value is `default', complex input methods always give extra
1526 guidance, but simple input methods give it only when you are not in
1527 the minibuffer.
1529 See also the variable `input-method-highlight-flag'."
1535 "*If this flag is non-nil, input methods highlight partially-entered text.
1536 For instance, while you are in the middle of a Quail input method sequence,
1537 the text inserted so far is temporarily underlined.
1538 The underlining goes away when you finish or abort the input method sequence.
1539 See also the variable `input-method-verbose-flag'."
1544 "Normal hook run just after an input method is activated.
1546 The variable `current-input-method' keeps the input method name
1550 "Normal hook run just after an input method is inactivated.
1552 The variable `current-input-method' still keeps the input method name
1559 "This flag controls when an input method returns.
1560 Usually, the input method does not return while there's a possibility
1561 that it may find a different translation if a user types another key.
1562 But, it this flag is non-nil, the input method returns as soon as
1566 "This flag controls how an input method shows an intermediate key sequence.
1567 Usually, the input method inserts the intermediate key sequence,
1568 or candidate translations corresponding to the sequence,
1569 at point in the current buffer.
1573 "This flag controls the behaviour of an input method on invalid key input.
1574 Usually, when a user types a key which doesn't start any character
1575 handled by the input method, the key is handled by turning off the
1576 input method temporarily. After that key, the input method is re-enabled.
1579 \f
1581 "Normal hook run after some language environment is set.
1583 When you set some hook function here, that effect usually should not
1584 be inherited to another language environment. So, you had better set
1585 another function in `exit-language-environment-hook' (which see) to
1589 "Normal hook run after exiting from some language environment.
1590 When this hook is run, the variable `current-language-environment'
1591 is still bound to the language environment being exited.
1593 This hook is mainly used for canceling the effect of
1599 "Switch to a specified language environment."
1612 "The last language environment specified with `set-language-environment'.
1614 to using the function `set-language-environment'."
1621 current-language-environment)
1622 language-info-alist t))
1624 ;; custom type will be updated with `set-language-info'.
1635 "Reset multilingual environment of Emacs to the default status.
1637 The default status is as follows:
1639 The default value of `buffer-file-coding-system' is nil.
1640 The default coding system for process I/O is nil.
1641 The default value for the command `set-terminal-coding-system' is nil.
1642 The default value for the command `set-keyboard-coding-system' is nil.
1644 The order of priorities of coding categories and the coding system
1645 bound to each category are as follows
1646 coding category coding system
1647 --------------------------------------------------
1648 coding-category-iso-8-1 iso-latin-1
1649 coding-category-iso-8-2 iso-latin-1
1650 coding-category-utf-8 mule-utf-8
1651 coding-category-utf-16-be mule-utf-16be-with-signature
1652 coding-category-utf-16-le mule-utf-16le-with-signature
1653 coding-category-iso-7-tight iso-2022-jp
1654 coding-category-iso-7 iso-2022-7bit
1655 coding-category-iso-7-else iso-2022-7bit-lock
1656 coding-category-iso-8-else iso-2022-8bit-ss2
1657 coding-category-emacs-mule emacs-mule
1658 coding-category-raw-text raw-text
1659 coding-category-sjis japanese-shift-jis
1660 coding-category-big5 chinese-big5
1661 coding-category-ccl nil
1662 coding-category-binary no-conversion"
1664 ;; This function formerly set default-enable-multibyte-characters to t,
1665 ;; but that is incorrect. It should not alter the unibyte/multibyte choice.
1668 coding-category-iso-7 'iso-2022-7bit
1669 coding-category-iso-8-1 'iso-latin-1
1670 coding-category-iso-8-2 'iso-latin-1
1671 coding-category-iso-7-else 'iso-2022-7bit-lock
1672 coding-category-iso-8-else 'iso-2022-8bit-ss2
1673 coding-category-emacs-mule 'emacs-mule
1674 coding-category-raw-text 'raw-text
1675 coding-category-sjis 'japanese-shift-jis
1676 coding-category-big5 'chinese-big5
1677 coding-category-utf-16-be 'mule-utf-16be-with-signature
1678 coding-category-utf-16-le 'mule-utf-16le-with-signature
1679 coding-category-utf-8 'mule-utf-8
1680 coding-category-ccl nil
1685 coding-category-iso-8-2
1686 coding-category-utf-8
1687 coding-category-utf-16-be
1688 coding-category-utf-16-le
1689 coding-category-iso-7-tight
1690 coding-category-iso-7
1691 coding-category-iso-7-else
1692 coding-category-iso-8-else
1693 coding-category-emacs-mule
1694 coding-category-raw-text
1695 coding-category-sjis
1696 coding-category-big5
1697 coding-category-ccl
1698 coding-category-binary))
1700 ;; Changing the binding of a coding category requires this call.
1705 ;; On Darwin systems, this should be utf-8, but when this file is loaded
1706 ;; utf-8 is not yet defined, so we set it in set-locale-environment instead.
1708 ;; Preserve eol-type from existing default-process-coding-systems.
1709 ;; On non-unix-like systems in particular, these may have been set
1710 ;; carefully by the user, or by the startup code, to deal with the
1711 ;; users shell appropriately, so should not be altered by changing
1712 ;; language environment.
1714 ;; When bootstrapping, coding-systems are not defined yet, so
1715 ;; we need to catch the error from check-coding-system.
1728 ;; Don't alter the terminal and keyboard coding systems here.
1729 ;; The terminal still supports the same coding system
1730 ;; that it supported a minute ago.
1731 ;; (set-terminal-coding-system-internal nil)
1732 ;; (set-keyboard-coding-system-internal nil)
1737 ;; Don't invoke fontset-related functions if fontsets aren't
1738 ;; supported in this build of Emacs.
1745 "Set up the display table and terminal coding system for LANGUAGE-NAME."
1749 ;; The following 2 lines undo the 8-bit display that we set up
1750 ;; in standard-display-european-internal, which see. This is in
1751 ;; case the user has used standard-display-european earlier in
1752 ;; this session. (The MS-DOS port doesn't use that setup, so it
1753 ;; doesn't need to undo it.)
1761 "Set up multi-lingual environment for using LANGUAGE-NAME.
1762 This sets the coding system priority and the default input method
1763 and sometimes other things. LANGUAGE-NAME should be a string
1765 specifies the character set for the major languages of Western Europe."
1767 nil
1783 default-buffer-file-coding-system)))
1786 ;; The features might set up coding systems.
1810 ;; DOS terminals' default is to use a special non-ASCII translation
1811 ;; table as appropriate for the installed codepage.
1816 ;; Unibyte setups if necessary.
1818 ;; Syntax and case table.
1823 ;; No information for syntax and case. Reset to the defaults.
1847 ;; Don't invoke fontset-related functions if fontsets aren't
1848 ;; supported in this build of Emacs.
1859 utf-translate-cjk-lang-env
1865 nil))
1871 ;; Actually set up direct output of non-ASCII characters.
1873 ;; Unibyte Emacs on MS-DOS wants to display all 8-bit characters with
1874 ;; the native font, and codes 160 and 146 stand for something very
1875 ;; different there.
1878 ;; Most X fonts used to do the wrong thing for latin-1 code 160.
1880 ;; XFree86 4 has fixed the fonts.
1884 ;; Make non-line-break space display as a plain space.
1887 ;; don't contain a character at that position. Map it to the ASCII
1888 ;; apostrophe. [This is actually RIGHT SINGLE QUOTATION MARK,
1889 ;; U+2019, normally from the windows-1252 character set. XFree 4
1890 ;; fonts probably have the appropriate glyph at this position,
1891 ;; so they could use standard-display-8bit. It's better to use a
1892 ;; proper windows-1252 coding system. --fx]
1893 (aset standard-display-table 146 [39]))))
1895 (defun set-language-environment-coding-systems (language-name
1896 &optional eol-type)
1897 "Do various coding system setups for language environment LANGUAGE-NAME.
1899 The optional arg EOL-TYPE specifies the eol-type of the default value
1900 of `buffer-file-coding-system' set by this function."
1901 (let* ((priority (get-language-info language-name 'coding-priority))
1902 (default-coding (car priority)))
1903 (if priority
1904 (let ((categories (mapcar 'coding-system-category priority)))
1905 (set-default-coding-systems
1906 (if (memq eol-type '(0 1 2 unix dos mac))
1907 (coding-system-change-eol-conversion default-coding eol-type)
1908 default-coding))
1909 (setq default-sendmail-coding-system default-coding)
1910 (set-coding-priority categories)
1911 (while priority
1912 (set (car categories) (car priority))
1913 (setq priority (cdr priority) categories (cdr categories)))
1914 ;; Changing the binding of a coding category requires this call.
1915 (update-coding-systems-internal)))))
1917 (defsubst princ-list (&rest args)
1919 (while args (princ (car args)) (setq args (cdr args)))
1922 (put 'describe-specified-language-support 'apropos-inhibit t)
1924 ;; Print language-specific information such as input methods,
1925 ;; charsets, and coding systems. This function is intended to be
1926 ;; called from the menu:
1927 ;; [menu-bar mule describe-language-environment LANGUAGE]
1928 ;; and should not run it by `M-x describe-current-input-method-function'.
1929 (defun describe-specified-language-support ()
1930 "Describe how Emacs supports the specified language environment."
1931 (interactive)
1932 (let (language-name)
1933 (if (not (and (symbolp last-command-event)
1934 (or (not (eq last-command-event 'Default))
1935 (setq last-command-event 'English))
1936 (setq language-name (symbol-name last-command-event))))
1937 (error "Bogus calling sequence"))
1938 (describe-language-environment language-name)))
1940 (defun describe-language-environment (language-name)
1941 "Describe how Emacs supports language environment LANGUAGE-NAME."
1942 (interactive
1943 (list (read-language-name
1944 'documentation
1945 "Describe language environment (default, current choice): ")))
1946 (if (null language-name)
1947 (setq language-name current-language-environment))
1948 (if (or (null language-name)
1949 (null (get-language-info language-name 'documentation)))
1950 (error "No documentation for the specified language"))
1951 (if (symbolp language-name)
1952 (setq language-name (symbol-name language-name)))
1953 (dolist (feature (get-language-info language-name 'features))
1954 (require feature))
1955 (let ((doc (get-language-info language-name 'documentation)))
1956 (help-setup-xref (list #'describe-language-environment language-name)
1957 (interactive-p))
1958 (with-output-to-temp-buffer (help-buffer)
1959 (save-excursion
1960 (set-buffer standard-output)
1962 (if (stringp doc)
1964 (condition-case nil
1965 (let ((str (eval (get-language-info language-name 'sample-text))))
1966 (if (stringp str)
1968 (error nil))
1969 (let ((input-method (get-language-info language-name 'input-method))
1970 (l (copy-sequence input-method-alist)))
1971 (insert "Input methods")
1972 (when input-method
1973 (insert " (default, " input-method ")")
1974 (setq input-method (assoc input-method input-method-alist))
1975 (setq l (cons input-method (delete input-method l))))
1977 (while l
1978 (when (string= language-name (nth 1 (car l)))
1979 (insert " " (car (car l)))
1980 (search-backward (car (car l)))
1981 (help-xref-button 0 'help-input-method (car (car l)))
1982 (goto-char (point-max))
1984 (if (stringp (nth 3 (car l)))
1985 (nth 3 (car l))
1986 (car (nth 3 (car l))))
1988 (setq l (cdr l)))
1991 (let ((l (get-language-info language-name 'charset)))
1992 (if (null l)
1994 (while l
1995 (insert " " (symbol-name (car l)))
1996 (search-backward (symbol-name (car l)))
1997 (help-xref-button 0 'help-character-set (car l))
1998 (goto-char (point-max))
2000 (setq l (cdr l)))))
2003 (let ((l (get-language-info language-name 'coding-system)))
2004 (if (null l)
2006 (while l
2007 (insert " " (symbol-name (car l)))
2008 (search-backward (symbol-name (car l)))
2009 (help-xref-button 0 'help-coding-system (car l))
2010 (goto-char (point-max))
2011 (insert " (`"
2012 (coding-system-mnemonic (car l))
2014 (coding-system-doc-string (car l))
2016 (let ((aliases (coding-system-get (car l)
2017 'alias-coding-systems)))
2018 (when aliases
2020 (while aliases
2021 (insert " " (symbol-name (car aliases)))
2022 (setq aliases (cdr aliases)))
2024 (setq l (cdr l)))))))))
2025 \f
2026 ;;; Locales.
2028 (defvar locale-translation-file-name nil
2029 "File name for the system's file of locale-name aliases, or nil if none.")
2031 ;; The following definitions might as well be marked as constants and
2032 ;; purecopied, since they're normally used on startup, and probably
2033 ;; should reflect the facilities of the base Emacs.
2034 (defconst locale-language-names
2035 (purecopy
2036 '(
2037 ;; Locale names of the form LANGUAGE[_TERRITORY][.CODESET][@MODIFIER]
2038 ;; as specified in the Single Unix Spec, Version 2.
2039 ;; LANGUAGE is a language code taken from ISO 639:1988 (E/F)
2040 ;; with additions from ISO 639/RA Newsletter No.1/1989;
2041 ;; see Internet RFC 2165 (1997-06) and
2042 ;; http://www.evertype.com/standards/iso639/iso639-en.html
2043 ;; TERRITORY is a country code taken from ISO 3166
2044 ;; http://www.din.de/gremien/nas/nabd/iso3166ma/codlstp1/en_listp1.html.
2045 ;; CODESET and MODIFIER are implementation-dependent.
2047 ;; jasonr comments: MS Windows uses three letter codes for
2048 ;; languages instead of the two letter ISO codes that POSIX
2049 ;; uses. In most cases the first two letters are the same, so
2050 ;; most of the regexps in locale-language-names work. Japanese
2051 ;; and Chinese are exceptions, which are listed in the
2052 ;; non-standard section at the bottom of locale-language-names.
2054 ("aa_DJ" . "Latin-1") ; Afar
2055 ("aa" . "UTF-8")
2056 ;; ab Abkhazian
2057 ("af" . "Latin-1") ; Afrikaans
2058 ("am" "Ethiopic" utf-8) ; Amharic
2059 ("an" . "Latin-9") ; Aragonese
2060 ; ar Arabic glibc uses 8859-6
2061 ; as Assamese
2062 ; ay Aymara
2063 ("az" . "UTF-8") ; Azerbaijani
2064 ; ba Bashkir
2065 ("be" "Belarusian" cp1251) ; Belarusian [Byelorussian until early 1990s]
2066 ("bg" "Bulgarian" cp1251) ; Bulgarian
2067 ; bh Bihari
2068 ; bi Bislama
2069 ("bn" . "UTF-8") ; Bengali, Bangla
2070 ("bo" . "Tibetan")
2071 ("br" . "Latin-1") ; Breton
2072 ("bs" . "Latin-2") ; Bosnian
2073 ("byn" . "UTF-8") ; Bilin; Blin
2074 ("ca" . "Latin-1") ; Catalan
2075 ; co Corsican
2076 ("cs" "Czech" iso-8859-2)
2077 ("cy" "Welsh" iso-8859-14)
2078 ("da" . "Latin-1") ; Danish
2079 ("de" "German" iso-8859-1)
2080 ; dz Bhutani
2081 ("el" "Greek" iso-8859-7)
2082 ;; Users who specify "en" explicitly typically want Latin-1, not ASCII.
2083 ;; That's actually what the GNU locales define, modulo things like
2084 ;; en_IN -- fx.
2085 ("en_IN" "English" utf-8) ; glibc uses utf-8 for English in India
2086 ("en" . "Latin-1") ; English
2087 ("eo" . "Latin-3") ; Esperanto
2088 ("es" "Spanish" iso-8859-1)
2089 ("et" . "Latin-1") ; Estonian
2090 ("eu" . "Latin-1") ; Basque
2091 ("fa" . "UTF-8") ; Persian
2092 ("fi" . "Latin-1") ; Finnish
2093 ("fj" . "Latin-1") ; Fiji
2094 ("fo" . "Latin-1") ; Faroese
2095 ("fr" "French" iso-8859-1) ; French
2096 ("fy" . "Latin-1") ; Frisian
2097 ("ga" . "Latin-1") ; Irish Gaelic (new orthography)
2098 ("gd" . "Latin-9") ; Scots Gaelic
2099 ("gez" "Ethiopic" utf-8) ; Geez
2100 ("gl" . "Latin-1") ; Gallegan; Galician
2101 ; gn Guarani
2102 ("gu" . "UTF-8") ; Gujarati
2103 ("gv" . "Latin-1") ; Manx Gaelic
2104 ; ha Hausa
2105 ("he" "Hebrew" iso-8859-8)
2106 ("hi" "Devanagari" utf-8) ; Hindi
2107 ("hr" "Croatian" iso-8859-2) ; Croatian
2108 ("hu" . "Latin-2") ; Hungarian
2109 ; hy Armenian
2110 ; ia Interlingua
2111 ("id" . "Latin-1") ; Indonesian
2112 ; ie Interlingue
2113 ; ik Inupiak
2114 ("is" . "Latin-1") ; Icelandic
2115 ("it" "Italian" iso-8859-1) ; Italian
2116 ; iu Inuktitut
2117 ("iw" "Hebrew" iso-8859-8)
2118 ("ja" "Japanese" euc-jp)
2119 ; jw Javanese
2120 ("ka" "Georgian" georgian-ps) ; Georgian
2121 ; kk Kazakh
2122 ("kl" . "Latin-1") ; Greenlandic
2123 ; km Cambodian
2124 ("kn" "Kannada" utf-8)
2125 ("ko" "Korean" euc-kr)
2126 ; ks Kashmiri
2127 ; ku Kurdish
2128 ("kw" . "Latin-1") ; Cornish
2129 ; ky Kirghiz
2130 ("la" . "Latin-1") ; Latin
2131 ("lb" . "Latin-1") ; Luxemburgish
2132 ("lg" . "Laint-6") ; Ganda
2133 ; ln Lingala
2134 ("lo" "Lao" utf-8) ; Laothian
2135 ("lt" "Lithuanian" iso-8859-13)
2136 ("lv" . "Latvian") ; Latvian, Lettish
2137 ; mg Malagasy
2138 ("mi" . "Latin-7") ; Maori
2139 ("mk" "Cyrillic-ISO" iso-8859-5) ; Macedonian
2140 ("ml" "Malayalam" utf-8)
2141 ("mn" . "UTF-8") ; Mongolian
2142 ; mo Moldavian
2143 ("mr" "Devanagari" utf-8) ; Marathi
2144 ("ms" . "Latin-1") ; Malay
2145 ("mt" . "Latin-3") ; Maltese
2146 ; my Burmese
2147 ; na Nauru
2148 ("nb" . "Latin-1") ; Norwegian
2149 ("ne" "Devanagari" utf-8) ; Nepali
2150 ("nl" "Dutch" iso-8859-1)
2151 ("no" . "Latin-1") ; Norwegian
2152 ("oc" . "Latin-1") ; Occitan
2153 ("om_ET" . "UTF-8") ; (Afan) Oromo
2154 ("om" . "Latin-1") ; (Afan) Oromo
2155 ; or Oriya
2156 ("pa" . "UTF-8") ; Punjabi
2157 ("pl" . "Latin-2") ; Polish
2158 ; ps Pashto, Pushto
2159 ("pt" . "Latin-1") ; Portuguese
2160 ; qu Quechua
2161 ("rm" . "Latin-1") ; Rhaeto-Romanic
2162 ; rn Kirundi
2163 ("ro" "Romanian" iso-8859-2)
2164 ("ru_RU" "Russian" iso-8859-5)
2165 ("ru_UA" "Russian" koi8-u)
2166 ; rw Kinyarwanda
2167 ("sa" . "Devanagari") ; Sanskrit
2168 ; sd Sindhi
2169 ("se" . "UTF-8") ; Northern Sami
2170 ; sg Sangho
2171 ("sh" . "Latin-2") ; Serbo-Croatian
2172 ; si Sinhalese
2173 ("sid" . "UTF-8") ; Sidamo
2174 ("sk" "Slovak" iso-8859-2)
2175 ("sl" "Slovenian" iso-8859-2)
2176 ; sm Samoan
2177 ; sn Shona
2178 ("so_ET" "UTF-8") ; Somali
2179 ("so" "Latin-1") ; Somali
2180 ("sq" . "Latin-1") ; Albanian
2181 ("sr_YU@cyrillic" . "Cyrillic-ISO") ; Serbian (Cyrillic alphabet)
2182 ("sr" . "Latin-2") ; Serbian (Latin alphabet)
2183 ; ss Siswati
2184 ("st" . "Latin-1") ; Sesotho
2185 ; su Sundanese
2186 ("sv" "Swedish" iso-8859-1) ; Swedish
2187 ("sw" . "Latin-1") ; Swahili
2188 ("ta" "Tamil" utf-8)
2189 ("te" . "UTF-8") ; Telugu
2190 ("tg" "Tajik" koi8-t)
2191 ("th" "Thai" tis-620)
2192 ("ti" "Ethiopic" utf-8) ; Tigrinya
2193 ("tig_ER" . "UTF-8") ; Tigre
2194 ; tk Turkmen
2195 ("tl" . "Latin-1") ; Tagalog
2196 ; tn Setswana
2197 ; to Tonga
2198 ("tr" "Turkish" iso-8859-9)
2199 ; ts Tsonga
2200 ("tt" . "UTF-8") ; Tatar
2201 ; tw Twi
2202 ; ug Uighur
2203 ("uk" "Ukrainian" koi8-u)
2204 ("ur" . "UTF-8") ; Urdu
2205 ("uz_UZ@cyrillic" . "UTF-8"); Uzbek
2206 ("uz" . "Latin-1") ; Uzbek
2207 ("vi" "Vietnamese" utf-8)
2208 ; vo Volapuk
2209 ("wa" . "Latin-1") ; Walloon
2210 ; wo Wolof
2211 ("xh" . "Latin-1") ; Xhosa
2212 ("yi" . "Windows-1255") ; Yiddish
2213 ; yo Yoruba
2214 ; za Zhuang
2215 ("zh_HK" . "Chinese-Big5")
2216 ("zh_TW" . "Chinese-Big5")
2217 ("zh_CN" . "Chinese-GB")
2218 ("zh" . "Chinese-GB")
2219 ; zh_CN.GB18030/GB18030 \
2220 ; zh_CN.GBK/GBK \
2221 ; zh_HK/BIG5-HKSCS \
2222 ("zu" . "Latin-1") ; Zulu
2224 ;; ISO standard locales
2225 ("c$" . "ASCII")
2226 ("posix$" . "ASCII")
2228 ;; The "IPA" Emacs language environment does not correspond
2229 ;; to any ISO 639 code, so let it stand for itself.
2230 ("ipa$" . "IPA")
2232 ;; Nonstandard or obsolete language codes
2233 ("cz" . "Czech") ; e.g. Solaris 2.6
2234 ("ee" . "Latin-4") ; Estonian, e.g. X11R6.4
2235 ("iw" . "Hebrew") ; e.g. X11R6.4
2236 ("sp" . "Cyrillic-ISO") ; Serbian (Cyrillic alphabet), e.g. X11R6.4
2237 ("su" . "Latin-1") ; Finnish, e.g. Solaris 2.6
2238 ("jp" . "Japanese") ; e.g. MS Windows
2239 ("chs" . "Chinese-GB") ; MS Windows Chinese Simplified
2240 ("cht" . "Chinese-BIG5") ; MS Windows Chinese Traditional
2241 ))
2242 "Alist of locale regexps vs the corresponding languages and coding systems.
2243 Each element has these form:
2244 \(LOCALE-REGEXP LANG-ENV CODING-SYSTEM)
2245 The first element whose LOCALE-REGEXP matches the start of a
2246 downcased locale specifies the LANG-ENV \(language environtment)
2247 and CODING-SYSTEM corresponding to that locale. If there is no
2248 appropriate language environment, the element may have this form:
2249 \(LOCALE-REGEXP . LANG-ENV)
2250 In this case, LANG-ENV is one of generic language environments for an
2253 (defconst locale-charset-language-names
2254 (purecopy
2263 ;; utf-8@euro exists, so put this last. (@euro really specifies
2264 ;; the currency, rather than the charset.)
2266 "List of pairs of locale regexps and charset language names.
2267 The first element whose locale regexp matches the start of a downcased locale
2268 specifies the language name whose charset corresponds to that locale.
2269 This language name is used if the locale is not listed in
2270 `locale-language-names'")
2272 (defconst locale-preferred-coding-systems
2273 (purecopy
2282 ;; utf-8@euro exists, so put this after utf-8. (@euro really
2283 ;; specifies the currency, rather than the charset.)
2284 (".*@euro" . iso-8859-15)
2285 ("koi8-?r" . koi8-r)
2286 ("koi8-?u" . koi8-u)
2287 ("tcvn" . tcvn)
2288 ("big5" . big5)
2289 ("euc-?tw" . euc-tw)
2290 ;; We don't support GBK, but as it is upper compatible with
2291 ;; GB-2312, we setup the default coding system to gb2312.
2292 ("gbk" . gb2312)
2293 ;; We don't support BIG5-HKSCS, but as it is upper compatible with
2294 ;; BIG5, we setup the default coding system to big5.
2295 ("big5hkscs" . big5)
2296 ("ja.*[._]euc" . japanese-iso-8bit)
2297 ("ja.*[._]jis7" . iso-2022-jp)
2298 ("ja.*[._]pck" . japanese-shift-jis)
2299 ("ja.*[._]sjis" . japanese-shift-jis)
2300 ("jpn" . japanese-shift-jis) ; MS-Windows uses this.
2301 ))
2302 "List of pairs of locale regexps and preferred coding systems.
2303 The first element whose locale regexp matches the start of a downcased locale
2304 specifies the coding system to prefer when using that locale.
2305 This coding system is used if the locale specifies a specific charset.")
2307 (defun locale-name-match (key alist)
2308 "Search for KEY in ALIST, which should be a list of regexp-value pairs.
2309 Return the value corresponding to the first regexp that matches the
2310 start of KEY, or nil if there is no match."
2311 (let (element)
2312 (while (and alist (not element))
2314 (setq element (car alist)))
2315 (setq alist (cdr alist)))
2316 (cdr element)))
2318 (defun locale-charset-match-p (charset1 charset2)
2319 "Whether charset names (strings) CHARSET1 and CHARSET2 are equivalent.
2320 Matching is done ignoring case and any hyphens and underscores in the
2321 names. E.g. `ISO_8859-1' and `iso88591' both match `iso-8859-1'."
2322 (setq charset1 (replace-regexp-in-string "[-_]" "" charset1))
2323 (setq charset2 (replace-regexp-in-string "[-_]" "" charset2))
2324 (eq t (compare-strings charset1 nil nil charset2 nil nil t)))
2326 (defvar locale-charset-alist nil
2327 "Coding system alist keyed on locale-style charset name.
2328 Used by `locale-charset-to-coding-system'.")
2330 (defun locale-charset-to-coding-system (charset)
2331 "Find coding system corresponding to CHARSET.
2332 CHARSET is any sort of non-Emacs charset name, such as might be used
2333 in a locale codeset, or elsewhere. It is matched to a coding system
2334 first by case-insensitive lookup in `locale-charset-alist'. Then
2335 matches are looked for in the coding system list, treating case and
2336 the characters `-' and `_' as insignificant. The coding system base
2338 `iso-latin-2' is returned."
2339 (or (car (assoc-string charset locale-charset-alist t))
2340 (let ((cs coding-system-alist)
2341 c)
2342 (while (and (not c) cs)
2343 (if (locale-charset-match-p charset (caar cs))
2344 (setq c (intern (caar cs)))
2345 (pop cs)))
2346 (if c (coding-system-base c)))))
2348 ;; Fixme: This ought to deal with the territory part of the locale
2349 ;; too, for setting things such as calendar holidays, ps-print paper
2350 ;; size, spelling dictionary.
2352 (defun set-locale-environment (&optional locale-name)
2353 "Set up multi-lingual environment for using LOCALE-NAME.
2354 This sets the language environment, the coding system priority,
2355 the default input method and sometimes other things.
2357 LOCALE-NAME should be a string which is the name of a locale supported
2358 by the system. Often it is of the form xx_XX.CODE, where xx is a
2359 language, XX is a country, and CODE specifies a character set and
2361 a locale for Japanese in Japan using the `japanese-iso-8bit'
2362 coding-system. The name may also have a modifier suffix, e.g. `@euro'
2363 or `@cyrillic'.
2365 If LOCALE-NAME is nil, its value is taken from the environment
2366 variables LC_ALL, LC_CTYPE and LANG (the first one that is set).
2368 The locale names supported by your system can typically be found in a
2369 directory named `/usr/share/locale' or `/usr/lib/locale'. LOCALE-NAME
2370 will be translated according to the table specified by
2371 `locale-translation-file-name'.
2373 See also `locale-charset-language-names', `locale-language-names',
2374 `locale-preferred-coding-systems' and `locale-coding-system'."
2375 (interactive "sSet environment for locale: ")
2377 ;; Do this at runtime for the sake of binaries possibly transported
2378 ;; to a system without X.
2379 (setq locale-translation-file-name
2380 (let ((files
2381 '("/usr/lib/X11/locale/locale.alias" ; e.g. X11R6.4
2382 "/usr/X11R6/lib/X11/locale/locale.alias" ; XFree86, e.g. RedHat 4.2
2383 "/usr/openwin/lib/locale/locale.alias" ; e.g. Solaris 2.6
2384 ;;
2385 ;; The following name appears after the X-related names above,
2386 ;; since the X-related names are what X actually uses.
2387 "/usr/share/locale/locale.alias" ; GNU/Linux sans X
2388 )))
2389 (while (and files (not (file-exists-p (car files))))
2390 (setq files (cdr files)))
2391 (car files)))
2393 (let ((locale locale-name))
2395 (unless locale
2396 ;; Use the first of these three environment variables
2397 ;; that has a nonempty value.
2398 (let ((vars '("LC_ALL" "LC_CTYPE" "LANG")))
2399 (while (and vars
2400 (= 0 (length locale))) ; nil or empty string
2401 (setq locale (getenv (pop vars))))))
2403 (unless (or locale (not (fboundp 'mac-get-preference)))
2404 (setq locale (mac-get-preference "AppleLocale"))
2405 (unless locale
2406 (let ((languages (mac-get-preference "AppleLanguages")))
2407 (unless (= (length languages) 0) ; nil or empty vector
2408 (setq locale (aref languages 0))))))
2409 (unless (or locale (not (boundp 'mac-system-locale)))
2410 (setq locale mac-system-locale))
2412 (when locale
2414 ;; Translate "swedish" into "sv_SE.ISO8859-1", and so on,
2415 ;; using the translation file that many systems have.
2416 (when locale-translation-file-name
2417 (with-temp-buffer
2418 (insert-file-contents locale-translation-file-name)
2419 (when (re-search-forward
2421 (setq locale (buffer-substring (point) (line-end-position))))))
2423 ;; Leave the system locales alone if the caller did not specify
2424 ;; an explicit locale name, as their defaults are set from
2425 ;; LC_MESSAGES and LC_TIME, not LC_CTYPE, and the user might not
2426 ;; want to set them to the same value as LC_CTYPE.
2427 (when locale-name
2428 (setq system-messages-locale locale)
2429 (setq system-time-locale locale))
2431 (setq locale (downcase locale))
2433 (let ((language-name
2434 (locale-name-match locale locale-language-names))
2435 (charset-language-name
2436 (locale-name-match locale locale-charset-language-names))
2437 (coding-system
2438 (or (locale-name-match locale locale-preferred-coding-systems)
2439 (when locale
2441 (locale-charset-to-coding-system
2442 (match-string 1 locale))))
2443 (and (eq system-type 'macos) mac-system-coding-system))))
2445 (if (consp language-name)
2446 ;; locale-language-names specify both lang-env and coding.
2447 ;; But, what specified in locale-preferred-coding-systems
2448 ;; has higher priority.
2449 (setq coding-system (or coding-system
2450 (nth 1 language-name))
2451 language-name (car language-name))
2452 ;; Otherwise, if locale is not listed in locale-language-names,
2453 ;; use what listed in locale-charset-language-names.
2454 (if (not language-name)
2455 (setq language-name charset-language-name)))
2457 (when language-name
2459 ;; Set up for this character set. This is now the right way
2460 ;; to do it for both unibyte and multibyte modes.
2461 (set-language-environment language-name)
2463 ;; If default-enable-multibyte-characters is nil,
2464 ;; we are using single-byte characters,
2465 ;; so the display table and terminal coding system are irrelevant.
2466 (when default-enable-multibyte-characters
2467 (set-display-table-and-terminal-coding-system
2468 language-name coding-system))
2470 ;; Set the `keyboard-coding-system' if appropriate (tty
2471 ;; only). At least X and MS Windows can generate
2472 ;; multilingual input.
2473 (unless window-system
2474 (let ((kcs (or coding-system
2475 (car (get-language-info language-name
2476 'coding-system)))))
2477 (if kcs (set-keyboard-coding-system kcs))))
2479 (setq locale-coding-system
2480 (car (get-language-info language-name 'coding-priority))))
2482 (when (and coding-system
2483 (not (coding-system-equal coding-system
2484 locale-coding-system)))
2485 (prefer-coding-system coding-system)
2486 (setq locale-coding-system coding-system))))
2488 ;; On Windows, override locale-coding-system,
2489 ;; keyboard-coding-system with system codepage. Note:
2490 ;; selection-coding-system is already set in w32select.c.
2491 (when (boundp 'w32-ansi-code-page)
2492 (let ((code-page-coding (intern (format "cp%d" w32-ansi-code-page))))
2493 (when (coding-system-p code-page-coding)
2494 (setq locale-coding-system code-page-coding)
2495 (set-keyboard-coding-system code-page-coding)
2496 (set-terminal-coding-system code-page-coding))))
2498 (when (eq system-type 'darwin)
2499 ;; On Darwin, file names are always encoded in utf-8, no matter
2500 ;; the locale.
2501 (setq default-file-name-coding-system 'utf-8)
2502 ;; Mac OS X's Terminal.app by default uses utf-8 regardless of
2503 ;; the locale.
2504 (when (and (null window-system)
2505 (equal (getenv "TERM_PROGRAM") "Apple_Terminal"))
2506 (set-terminal-coding-system 'utf-8)
2507 (set-keyboard-coding-system 'utf-8)))
2509 ;; Default to A4 paper if we're not in a C, POSIX or US locale.
2510 ;; (See comments in Flocale_info.)
2511 (let ((locale locale)
2512 (paper (locale-info 'paper)))
2513 (if paper
2514 ;; This will always be null at the time of writing.
2515 (cond
2516 ((equal paper '(216 279))
2517 (setq ps-paper-type 'letter))
2518 ((equal paper '(210 297))
2519 (setq ps-paper-type 'a4)))
2520 (let ((vars '("LC_ALL" "LC_PAPER" "LANG")))
2521 (while (and vars (= 0 (length locale)))
2522 (setq locale (getenv (pop vars)))))
2523 (when locale
2524 ;; As of glibc 2.2.5, these are the only US Letter locales,
2525 ;; and the rest are A4.
2526 (setq ps-paper-type
2527 (or (locale-name-match locale '(("c$" . letter)
2528 ("posix$" . letter)
2529 (".._us" . letter)
2530 (".._pr" . letter)
2531 (".._ca" . letter)
2532 ("enu$" . letter) ; Windows
2533 ("esu$" . letter)
2534 ("enc$" . letter)
2535 ("frc$" . letter)))
2536 'a4))))))
2537 nil)
2538 \f
2539 ;;; Charset property
2541 (defun get-charset-property (charset propname)
2542 "Return the value of CHARSET's PROPNAME property.
2543 This is the last value stored with
2544 (put-charset-property CHARSET PROPNAME VALUE)."
2545 (and (not (eq charset 'composition))
2546 (plist-get (charset-plist charset) propname)))
2548 (defun put-charset-property (charset propname value)
2549 "Store CHARSETS's PROPNAME property with value VALUE.
2550 It can be retrieved with `(get-charset-property CHARSET PROPNAME)'."
2551 (or (eq charset 'composition)
2552 (set-charset-plist charset
2553 (plist-put (charset-plist charset) propname value))))
2555 ;;; Character code property
2556 (put 'char-code-property-table 'char-table-extra-slots 0)
2558 (defvar char-code-property-table
2559 (make-char-table 'char-code-property-table)
2560 "Char-table containing a property list of each character code.
2562 See also the documentation of `get-char-code-property' and
2563 `put-char-code-property'.")
2565 (defun get-char-code-property (char propname)
2566 "Return the value of CHAR's PROPNAME property in `char-code-property-table'."
2567 (let ((plist (aref char-code-property-table char)))
2568 (if (listp plist)
2569 (car (cdr (memq propname plist))))))
2571 (defun put-char-code-property (char propname value)
2572 "Store CHAR's PROPNAME property with VALUE in `char-code-property-table'.
2573 It can be retrieved with `(get-char-code-property CHAR PROPNAME)'."
2574 (let ((plist (aref char-code-property-table char)))
2575 (if plist
2576 (let ((slot (memq propname plist)))
2577 (if slot
2578 (setcar (cdr slot) value)
2579 (nconc plist (list propname value))))
2580 (aset char-code-property-table char (list propname value)))))
2582 \f
2583 ;; Pretty description of encoded string
2585 ;; Alist of ISO 2022 control code vs the corresponding mnemonic string.
2586 (defvar iso-2022-control-alist
2594 (defun encoded-string-description (str coding-system)
2595 "Return a pretty description of STR that is encoded by CODING-SYSTEM."
2596 (setq str (string-as-unibyte str))
2597 (mapconcat
2598 (if (and coding-system (eq (coding-system-type coding-system) 2))
2599 ;; Try to get a pretty description for ISO 2022 escape sequences.
2600 (function (lambda (x) (or (cdr (assq x iso-2022-control-alist))
2601 (format "0x%02X" x))))
2602 (function (lambda (x) (format "0x%02X" x))))
2603 str " "))
2605 (defun encode-coding-char (char coding-system)
2606 "Encode CHAR by CODING-SYSTEM and return the resulting string.
2607 If CODING-SYSTEM can't safely encode CHAR, return nil."
2608 (let ((str1 (string-as-multibyte (char-to-string char)))
2609 (str2 (string-as-multibyte (make-string 2 char)))
2610 (safe-chars (and coding-system
2611 (coding-system-get coding-system 'safe-chars)))
2612 (charset (char-charset char))
2613 enc1 enc2 i1 i2)
2614 (when (or (eq safe-chars t)
2615 (eq charset 'ascii)
2616 (and safe-chars (aref safe-chars char)))
2617 ;; We must find the encoded string of CHAR. But, just encoding
2618 ;; CHAR will put extra control sequences (usually to designate
2619 ;; ASCII charset) at the tail if type of CODING is ISO 2022.
2620 ;; To exclude such tailing bytes, we at first encode one-char
2621 ;; string and two-char string, then check how many bytes at the
2622 ;; tail of both encoded strings are the same.
2624 (setq enc1 (encode-coding-string str1 coding-system)
2625 i1 (length enc1)
2626 enc2 (encode-coding-string str2 coding-system)
2627 i2 (length enc2))
2628 (while (and (> i1 0) (= (aref enc1 (1- i1)) (aref enc2 (1- i2))))
2629 (setq i1 (1- i1) i2 (1- i2)))
2631 ;; Now (substring enc1 i1) and (substring enc2 i2) are the same,
2632 ;; and they are the extra control sequences at the tail to
2633 ;; exclude.
2634 (substring enc2 0 i2))))
2637 ;; arch-tag: b382c432-4b36-460e-bf4c-05efd0bb18dc
2638 ;;; mule-cmds.el ends here