| blob | f52e0e85b59c81ed1195ecbfffebbc7634b230c7 |
1 ;;; mule-cmds.el --- commands for mulitilingual environment -*-coding: iso-2022-7bit -*-
3 ;; Copyright (C) 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005
4 ;; Free Software Foundation, Inc.
5 ;; Copyright (C) 1995, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005
6 ;; National Institute of Advanced Industrial Science and Technology (AIST)
7 ;; Registration Number H14PRO021
9 ;; Keywords: mule, multilingual
11 ;; This file is part of GNU Emacs.
13 ;; GNU Emacs is free software; you can redistribute it and/or modify
14 ;; it under the terms of the GNU General Public License as published by
15 ;; the Free Software Foundation; either version 2, or (at your option)
16 ;; any later version.
18 ;; GNU Emacs is distributed in the hope that it will be useful,
19 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
20 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
21 ;; GNU General Public License for more details.
23 ;; You should have received a copy of the GNU General Public License
24 ;; along with GNU Emacs; see the file COPYING. If not, write to the
25 ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330,
26 ;; Boston, MA 02111-1307, USA.
28 ;;; Commentary:
30 ;;; Code:
36 ;;; MULE related key bindings and menus.
41 ;; Keep "C-x C-m ..." for mule specific commands.
74 t)
77 t)
80 t)
85 t)
94 t)
97 t)
100 describe-language-environment-map
102 t)
106 t)
109 t)
116 t)
121 t)
124 t)
128 t)
131 :enable buffer-file-name
133 t)
137 t)
140 t)
145 t)
150 t)
153 t)
158 t)
163 t)
169 t)
173 [Default] '(menu-item "Default" setup-specified-language-environment))
175 (define-key describe-language-environment-map
178 ;; This should be a single character key binding because users use it
179 ;; very frequently while editing multilingual text. Now we can use
180 ;; only two such keys: "\C-\\" and "\C-^", but the latter is not
181 ;; convenient because it requires shifting on most keyboards. An
182 ;; alternative is "\C-\]" which is now bound to `abort-recursive-edit'
183 ;; but it won't be used that frequently.
186 ;; This is no good because people often type Shift-SPC
187 ;; meaning to type SPC. -- rms.
188 ;; ;; Here's an alternative key binding for X users (Shift-SPACE).
189 ;; (define-key global-map [?\S- ] 'toggle-input-method)
191 ;;; Mule related hyperlinks.
199 ;; Note starting with word-syntax character:
203 "Return a coding system which differs from CODING-SYSTEM in eol conversion.
204 The returned coding system converts end-of-line by EOL-TYPE
205 but text as the same way as CODING-SYSTEM.
206 EOL-TYPE should be `unix', `dos', `mac', or nil.
207 If EOL-TYPE is nil, the returned coding system detects
208 how end-of-line is formatted automatically while decoding.
210 EOL-TYPE can be specified by an integer 0, 1, or 2.
211 They means `unix', `dos', and `mac' respectively."
220 coding-system
224 base
226 coding-system
232 "Return a coding system which differs from CODING-SYSTEM in text conversion.
233 The returned coding system converts text by CODING
234 but end-of-line as the same way as CODING-SYSTEM.
235 If CODING is nil, the returned coding system detects
236 how text is formatted automatically while decoding."
243 "Change whether this buffer uses multibyte characters.
244 With arg, use multibyte characters if the arg is positive.
246 Note that this command does not convert the byte contents of
247 the buffer; it only changes the way those bytes are interpreted.
248 In general, therefore, this command *changes* the sequence of
249 characters that the current buffer contains.
251 We suggest you avoid using use this command unless you know what you
252 are doing. If you use it by mistake, and the buffer is now displayed
253 wrong, use this command again to toggle back to the right mode."
262 "Display the HELLO file which list up many languages and characters."
264 ;; We have to decode the file in any environment.
270 "Execute an I/O command using the specified coding system."
274 t))
275 buffer-file-coding-system)))
280 default))))
284 prefix)
289 ;; Process keys bound in `universal-argument-map'.
295 ;; Have to bind `last-command-char' here so that
296 ;; `digit-argument', for instance, can compute the
297 ;; prefix arg.
301 ;; This is the final call to `universal-argument-other-key', which
302 ;; set's the final `prefix-arg.
306 ;; Read the command to execute with the given prefix arg.
319 "Set default value of various coding systems to CODING-SYSTEM.
320 This sets the following coding systems:
321 o coding system of a newly created buffer
322 o default coding system for subprocess I/O
323 This also sets the following values:
324 o default value used as `file-name-coding-system' for converting file names.
325 o default value for the command `set-terminal-coding-system' (not on MSDOS)
326 o default value for the command `set-keyboard-coding-system'."
335 ;; The file-name coding system on Darwin systems is always utf-8.
337 ;; If coding-system is nil, honor that on MS-DOS as well, so
338 ;; that they could reset the terminal coding system.
342 ;; Preserve eol-type from existing default-process-coding-systems.
343 ;; On non-unix-like systems in particular, these may have been set
344 ;; carefully by the user, or by the startup code, to deal with the
345 ;; users shell appropriately, so should not be altered by changing
346 ;; language environment.
357 "Add CODING-SYSTEM at the front of the priority list for automatic detection.
358 This also sets the following coding systems:
359 o coding system of a newly created buffer
360 o default coding system for subprocess I/O
361 This also sets the following values:
362 o default value used as `file-name-coding-system' for converting file names.
363 o default value for the command `set-terminal-coding-system' (not on MSDOS)
364 o default value for the command `set-keyboard-coding-system'
366 If CODING-SYSTEM specifies a certain type of EOL conversion, the coding
367 systems set by this function will use that type of EOL conversion.
369 This command does not change the default value of terminal coding system
370 for MS-DOS terminal, because DOS terminals only support a single coding
371 system, and Emacs automatically sets the default to that coding system at
372 startup.
374 A coding system that requires automatic detection of text
375 encoding (e.g. undecided, unix) can't be preferred.
377 See also `coding-category-list' and `coding-system-category'."
385 ;; CODING-SYSTEM is no-conversion or undecided.
388 ;; Changing the binding of a coding category requires this call.
391 ;; We must change the order.
395 base coding-system))
396 ;; If they asked for specific EOL conversion, honor that.
404 "If non-nil, a predicate function to sort coding systems.
406 It is called with two coding systems, and should return t if the first
412 "Sort coding system list CODINGS by a priority of each coding system.
414 If a coding system is most preferred, it has the highest priority.
415 Otherwise, a coding system corresponds to some MIME charset has higher
416 priorities. Among them, a coding system included in `coding-system'
417 key of the current language environment has higher priorities. See
418 also the documentation of `language-info-alist'.
420 If the variable `sort-coding-systems-predicate' (which see) is
421 non-nil, it is used to sort CODINGS in the different way than above."
425 coding-category-list))
432 ;; We calculate the priority number 0..255 by
433 ;; using the 8 bits PMMLCEII as this:
434 ;; P: 1 iff most preferred.
435 ;; MM: greater than 0 iff mime-charset.
436 ;; L: 1 iff one of the current lang. env.'s codings.
437 ;; C: 1 iff one of codings listed in the category list.
438 ;; E: 1 iff not XXX-with-esc
439 ;; II: if iso-2022 based, 0..3, else 1.
444 ;; Prefer coding systems corresponding to a
445 ;; MIME charset.
447 ;; Lower utf-16 priority so that we
448 ;; normally prefer utf-8 to it, and put
449 ;; x-ctext below that.
464 ;; For ISO based coding systems, prefer
465 ;; one that doesn't use escape sequences.
472 0
480 "Return a list of proper coding systems to encode a text between FROM and TO.
481 If FROM is a string, find coding systems in that instead of the buffer.
482 All coding systems in the list can safely encode any multibyte characters
483 in the text.
485 If the text contains no multibyte characters, return a list of a single
486 element `undecided'."
489 ;; The text contains only ASCII characters. Any coding
490 ;; systems are safe.
492 ;; We need copy-sequence because sorting will alter the argument.
496 "Return a list of proper coding systems to encode STRING.
497 All coding systems in the list can safely encode any multibyte characters
498 in STRING.
500 If STRING contains no multibyte characters, return a list of a single
501 element `undecided'."
505 "Return a list of proper coding systems to encode characters of CHARSETS.
506 CHARSETS is a list of character sets.
507 It actually checks at most the first 96 characters of each charset.
508 So, if a charset of dimension two is included in CHARSETS, the value may
509 contain a coding system that can't encode all characters of the charset."
519 charset l str)
543 codings))))
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 EXCLUDE is a list of character sets to be ignored.
557 For invalid characters, CHARs are actually strings."
559 charset char)
568 eight-bit-control eight-bit-graphic))
600 "Search forward from point for a character that is not encodable.
601 It asks which coding system to check.
602 If such a character is found, set point after that character.
603 Otherwise, don't move point.
605 When called from a program, the value is a position of the found character,
606 or nil if all characters are encodable."
611 default))))
616 pos))
620 "Most recent coding system explicitly specified by the user when asked.
621 This variable is set whenever Emacs asks the user which coding system
622 to use in order to write a file. If you set it to nil explicitly,
623 then call `write-region', then afterward this variable will be non-nil
627 "If non-nil, a function to control the behavior of coding system selection.
628 The meaning is the same as the argument ACCEPT-DEFAULT-P of the
629 function `select-safe-coding-system' (which see). This variable
634 "Select interactively a coding system for the region FROM ... TO.
635 FROM can be a string, as in `write-region'.
636 CODINGS is the list of base coding systems known to be safe for this region,
637 typically obtained with `find-coding-systems-region'.
638 UNSAFE is a list of coding systems known to be unsafe for this region.
639 REJECTED is a list of coding systems which were safe but for some reason
640 were not recommended in the particular context.
641 DEFAULT is the coding system to use by default in the query."
642 ;; At first, if some defaults are unsafe, record at most 11
643 ;; problematic characters and their positions for them by turning
644 ;; (CODING ...)
645 ;; into
646 ;; ((CODING (POS . CHAR) (POS . CHAR) ...) ...)
661 unsafe)))
663 ;; Change each safe coding system to the corresponding
664 ;; mime-charset name if it is also a coding system. Such a name
665 ;; is more friendly to users.
667 mime-charset)
674 ;; Don't offer variations with locking shift, which you
675 ;; basically never want.
684 ;; Remove raw-text, emacs-mule and no-conversion unless nothing
685 ;; else is available.
694 coding-system)
696 ;; If some defaults are unsafe, make sure the offending
697 ;; buffer is displayed.
701 unsafe))))
702 ;; Then ask users to select one from CODINGS while showing
703 ;; the reason why none of the defaults are not used.
712 "These default coding systems were tried to encode"
727 but it is not recommended for encoding text in this context,
761 'help-echo
762 "mouse-2, RET: jump to this character"
763 'help-function func1
766 "..."
768 'help-echo
769 "mouse-2, RET: next unencodable character"
770 'help-function func2
790 ;; Read a coding system.
794 default))
799 coding-system))
802 accept-default-p file)
803 "Ask a user to select a safe coding system from candidates.
804 The candidates of coding systems which can safely encode a text
805 between FROM and TO are shown in a popup window. Among them, the most
806 proper one is suggested as the default.
808 The list of `buffer-file-coding-system' of the current buffer,
809 the `default-buffer-file-coding-system', and the
810 most preferred coding system (if it corresponds to a MIME charset) is
811 treated as the default coding system list. Among them, the first one
812 that safely encodes the text is normally selected silently and
813 returned without any user interaction. See also the command
814 `prefer-coding-system'.
816 However, the user is queried if the chosen coding system is
817 inconsistent with what would be selected by `set-auto-coding' from
818 coding cookies &c. if the contents of the region were read from a
819 file. (That could lead to data corruption in a file subsequently
820 re-visited and edited.)
822 Optional 3rd arg DEFAULT-CODING-SYSTEM specifies a coding system or a
823 list of coding systems to be prepended to the default coding system
824 list. However, if DEFAULT-CODING-SYSTEM is a list and the first
825 element is t, the cdr part is used as the defualt coding system list,
826 i.e. `buffer-file-coding-system', `default-buffer-file-coding-system',
827 and the most preferred coding system are not used.
829 Optional 4th arg ACCEPT-DEFAULT-P, if non-nil, is a function to
830 determine the acceptability of the silently selected coding system.
831 It is called with that coding system, and should return nil if it
832 should not be silently selected and thus user interaction is required.
834 Optional 5th arg FILE is the file name to use for this purpose.
835 That is different from `buffer-file-name' when handling `write-region'
836 \(for example).
838 The variable `select-safe-coding-system-accept-default-p', if
839 non-nil, overrides ACCEPT-DEFAULT-P.
841 Kludgy feature: if FROM is a string, the string is the target text,
842 and TO is ignored."
852 ;; Change elements of the list to (coding . base-coding).
855 default-coding-system))
857 ;; From now on, the list of defaults is reversed.
861 ;; If buffer-file-coding-system is not nil nor undecided, append it
862 ;; to the defaults.
868 default-coding-system))))
870 ;; If default-buffer-file-coding-system is not nil nor undecided,
871 ;; append it to the defaults.
877 default-coding-system))))
879 ;; If the most preferred coding system has the property mime-charset,
880 ;; append it to the defaults.
882 preferred base)
890 default-coding-system)))))
897 safe rejected unsafe)
899 ;; Any coding system is ok.
901 ;; Classify the defaults into safe, rejected, and unsafe.
912 ;; If all the defaults failed, ask a user.
925 ;; Check we're not inconsistent with what `coding:' spec &c would
926 ;; give when file is re-read.
927 ;; But don't do this if we explicitly ignored the cookie
928 ;; by using `find-file-literally'.
930 find-file-literally
940 ;; Merge coding-system and auto-cs as far as possible.
949 coding-system auto-cs))
952 auto-cs coding-system))))
956 coding-system eol-type-2)))
959 auto-cs eol-type-1)))))))
962 ;; Don't barf if writing a compressed file, say.
963 ;; This check perhaps isn't ideal, but is probably
964 ;; the best thing to do.
969 %s specified by file contents. Really save (else edit coding cookies \
972 coding-system))
977 "Return a coding system to encode the outgoing message of the current buffer.
978 It at first tries the first coding system found in these variables
979 in this order:
980 (1) local value of `buffer-file-coding-system'
981 (2) value of `sendmail-coding-system'
982 (3) value of `default-sendmail-coding-system'
983 (4) value of `default-buffer-file-coding-system'
984 If the found coding system can't encode the current buffer,
985 or none of them are bound to a coding system,
986 it asks the user to select a proper coding system."
988 buffer-file-coding-system)
989 sendmail-coding-system
990 default-sendmail-coding-system
991 default-buffer-file-coding-system)))
993 ;; We should never use no-conversion for outgoing mail.
999 coding)))
1000 \f
1001 ;;; Language support stuff.
1004 "Alist of language environment definitions.
1005 Each element looks like:
1006 (LANGUAGE-NAME . ((KEY . INFO) ...))
1007 where LANGUAGE-NAME is a string, the name of the language environment,
1008 KEY is a symbol denoting the kind of information, and
1009 INFO is the data associated with KEY.
1010 Meaningful values for KEY include
1012 documentation value is documentation of what this language environment
1013 is meant for, and how to use it.
1014 charset value is a list of the character sets used by this
1015 language environment.
1016 sample-text value is an expression which is evalled to generate
1017 a line of text written using characters appropriate
1018 for this language environment.
1019 setup-function value is a function to call to switch to this
1020 language environment.
1021 exit-function value is a function to call to leave this
1022 language environment.
1023 coding-system value is a list of coding systems that are good
1024 for saving text written in this language environment.
1025 This list serves as suggestions to the user;
1026 in effect, as a kind of documentation.
1027 coding-priority value is a list of coding systems for this language
1028 environment, in order of decreasing priority.
1029 This is used to set up the coding system priority
1030 list when you switch to this language environment.
1031 nonascii-translation
1032 value is a translation table to be set in the
1033 variable `nonascii-translation-table' in this
1034 language environment, or a character set from
1035 which `nonascii-insert-offset' is calculated.
1036 input-method value is a default input method for this language
1037 environment.
1038 features value is a list of features requested in this
1039 language environment.
1040 ctext-non-standard-encodings
1041 value is a list of non-standard encoding
1042 names used in extended segments of CTEXT.
1043 See the variable
1044 `ctext-non-standard-encodings' for more
1045 detail.
1047 The following keys take effect only when multibyte characters are
1048 globally disabled, i.e. the value of `default-enable-multibyte-characters'
1049 is nil.
1051 unibyte-syntax value is a library name to load to set
1052 unibyte 8-bit character syntaxes for this
1053 language environment.
1055 unibyte-display value is a coding system to encode characters
1056 for the terminal. Characters in the range
1057 of 160 to 255 display not as octal escapes,
1058 but as non-ASCII characters in this language
1062 "Return information listed under KEY for language environment LANG-ENV.
1063 KEY is a symbol denoting the kind of information.
1064 For a list of useful values for KEY and their meanings,
1065 see `language-info-alist'."
1073 "Modify part of the definition of language environment LANG-ENV.
1074 Specifically, this stores the information INFO under KEY
1075 in the definition of this language environment.
1076 KEY is a symbol denoting the kind of information.
1077 INFO is the value for that information.
1079 For a list of useful values for KEY and their meanings,
1080 see `language-info-alist'."
1094 ;; Update the custom-type of `current-language-environment'.
1102 "Store ALIST as the definition of language environment LANG-ENV.
1103 ALIST is an alist of KEY and INFO values. See the documentation of
1104 `language-info-alist' for the meanings of KEY and INFO.
1106 Optional arg PARENTS is a list of parent menu names; it specifies
1107 where to put this language environment in the
1108 Describe Language Environment and Set Language Environment menus.
1110 in the European submenu in each of those two menus."
1117 map parent-symbol parent prompt)
1123 ;; This prompt string is for define-prefix-command, so
1124 ;; that the map it creates will be suitable for a menu.
1145 ;; Set up menu items for this language env.
1158 "Read a language environment name which has information for KEY.
1159 If KEY is nil, read any language environment.
1160 Prompt with PROMPT. DEFAULT is the default choice of language environment.
1161 This returns a language environment name as a string."
1164 language-info-alist
1167 t nil nil default)))
1171 name)))
1172 \f
1173 ;;; Multilingual input methods.
1175 "LEIM: Libraries of Emacs Input Methods."
1179 "Name of LEIM list file.
1180 This file contains a list of libraries of Emacs input methods (LEIM)
1181 in the format of Lisp expression for registering each input method.
1185 ";;; %s -- list of LEIM (Library of Emacs Input Method) -*-coding: iso-2022-7bit;-*-
1186 ;;
1187 ;; This file contains a list of LEIM (Library of Emacs Input Method)
1188 ;; methods in the same directory as this file. Loading this file
1189 ;; registers all the input methods in Emacs.
1190 ;;
1191 ;; Each entry has the form:
1192 ;; (register-input-method
1193 ;; INPUT-METHOD LANGUAGE-NAME ACTIVATE-FUNC
1194 ;; TITLE DESCRIPTION
1195 ;; ARG ...)
1196 ;; See the function `register-input-method' for the meanings of the arguments.
1197 ;;
1198 ;; If this directory is included in load-path, Emacs automatically
1199 ;; loads this file at startup time.
1201 "
1202 leim-list-file-name)
1206 "Regexp matching head of each entry in LEIM list file.
1211 "List of functions to call to update LEIM list file.
1215 "Update LEIM list file in directories DIRS."
1222 "The current input method for multilingual text.
1233 "*Default input method for multilingual text (a string).
1234 This is the input method activated automatically by the command
1239 :completion-ignore-case t
1240 :complete-function widget-string-complete
1241 :completion-alist input-method-alist
1253 "Function to call for inactivating the current input method.
1254 Every input method should set this to an appropriate value when activated.
1255 This function is called with no argument.
1257 This function should never change the value of `current-input-method'.
1263 "Function to call for describing the current input method.
1269 "Alist of input method names vs how to use them.
1270 Each element has the form:
1271 (INPUT-METHOD LANGUAGE-ENV ACTIVATE-FUNC TITLE DESCRIPTION ARGS...)
1275 "Register INPUT-METHOD as an input method for language environment LANG-ENV.
1276 INPUT-METHOD and LANG-ENV are symbols or strings.
1278 The remaining arguments are:
1279 ACTIVATE-FUNC, TITLE, DESCRIPTION, and ARGS...
1280 ACTIVATE-FUNC is a function to call to activate this method.
1281 TITLE is a string to show in the mode line when this method is active.
1282 DESCRIPTION is a string describing this method and what it is good for.
1283 The ARGS, if any, are passed as arguments to ACTIVATE-FUNC.
1284 All told, the arguments to ACTIVATE-FUNC are INPUT-METHOD and the ARGS.
1287 created at Emacs build time, registering all Quail input methods
1288 contained in the Emacs distribution.
1290 In case you want to register a new Quail input method by yourself, be
1291 careful to use the same input method title as given in the third
1292 parameter of `quail-define-package'. (If the values are different, the
1293 string specified in this function takes precedence.)
1295 The commands `describe-input-method' and `list-input-methods' need
1296 these duplicated values to show some information about input methods
1297 without loading the relevant Quail packages."
1310 "Read a name of input method from a minibuffer prompting with PROMPT.
1311 If DEFAULT is non-nil, use that as the default,
1312 and substitute it into PROMPT at the first `%s'.
1313 If INHIBIT-NULL is non-nil, null input signals an error.
1315 The return value is a string."
1319 ;; As it is quite normal to change input method in the
1320 ;; minibuffer, we must enable it even if
1321 ;; enable-recursive-minibuffers is currently nil.
1323 ;; This binding is necessary because input-method-history is
1324 ;; buffer local.
1326 nil t nil 'input-method-history
1327 default)))
1331 input-method
1336 "Switch to input method INPUT-METHOD for the current buffer.
1337 If some other input method is already active, turn it off first.
1338 If INPUT-METHOD is nil, deactivate any current input method."
1365 "Turn off the current input method."
1378 input-method-function nil
1379 current-input-method-title nil)
1383 "Select and activate input method INPUT-METHOD for the current buffer.
1384 This also sets the default input method to the one you specify.
1385 If INPUT-METHOD is nil, this function turns off the input method, and
1386 also causes you to be prompted for a name of an input method the next
1388 When called interactively, the optional arg INTERACTIVE is non-nil,
1389 which marks the variable `default-input-method' as set for Custom buffers.
1397 default t)
1398 t)))
1403 default-input-method)
1406 "Enable or disable multilingual text input method for the current buffer.
1407 Only one input method can be enabled at any time in a given buffer.
1409 The normal action is to enable an input method if none was
1410 enabled, and disable the current one otherwise. Which input method
1411 to enable can be determined in various ways--either the one most
1412 recently used, or the one specified by `default-input-method', or
1413 as a last resort by reading the name of an input method in the
1414 minibuffer.
1416 With a prefix argument, read an input method name with the minibuffer
1417 and enable that one. The default is the most recent input method specified
1418 \(not including the currently active input method, if any).
1420 When called interactively, the optional arg INTERACTIVE is non-nil,
1421 which marks the variable `default-input-method' as set for Custom buffers."
1435 default t))
1436 default))
1444 "Describe input method INPUT-METHOD."
1474 "Describe the input method currently in use.
1475 This is a subroutine for `describe-input-method'."
1481 current-input-method)
1486 "Read a multilingual string from minibuffer, prompting with string PROMPT.
1487 The input method selected last time is activated in minibuffer.
1488 If optional second arg INITIAL-INPUT is non-nil, insert it in the minibuffer
1489 initially.
1490 Optional 3rd argument INPUT-METHOD specifies the input method
1491 to be activated instead of the one selected last time. It is a symbol
1492 or a string."
1495 current-input-method
1496 default-input-method
1507 ;; Variables to control behavior of input methods. All input methods
1508 ;; should react to these variables.
1511 "*A flag to control extra guidance given by input methods.
1512 The value should be nil, t, `complex-only', or `default'.
1514 The extra guidance is done by showing list of available keys in echo
1515 area. When you use the input method in the minibuffer, the guidance
1516 is shown at the bottom short window (split from the existing window).
1518 If the value is t, extra guidance is always given, if the value is
1519 nil, extra guidance is always suppressed.
1521 If the value is `complex-only', only complex input methods such as
1522 `chinese-py' and `japanese' give extra guidance.
1524 If the value is `default', complex input methods always give extra
1525 guidance, but simple input methods give it only when you are not in
1526 the minibuffer.
1528 See also the variable `input-method-highlight-flag'."
1534 "*If this flag is non-nil, input methods highlight partially-entered text.
1535 For instance, while you are in the middle of a Quail input method sequence,
1536 the text inserted so far is temporarily underlined.
1537 The underlining goes away when you finish or abort the input method sequence.
1538 See also the variable `input-method-verbose-flag'."
1543 "Normal hook run just after an input method is activated.
1545 The variable `current-input-method' keeps the input method name
1549 "Normal hook run just after an input method is inactivated.
1551 The variable `current-input-method' still keeps the input method name
1558 "This flag controls when an input method returns.
1559 Usually, the input method does not return while there's a possibility
1560 that it may find a different translation if a user types another key.
1561 But, it this flag is non-nil, the input method returns as soon as
1565 "This flag controls how an input method shows an intermediate key sequence.
1566 Usually, the input method inserts the intermediate key sequence,
1567 or candidate translations corresponding to the sequence,
1568 at point in the current buffer.
1572 "This flag controls the behavior of an input method on invalid key input.
1573 Usually, when a user types a key which doesn't start any character
1574 handled by the input method, the key is handled by turning off the
1575 input method temporarily. After that key, the input method is re-enabled.
1578 \f
1580 "Normal hook run after some language environment is set.
1582 When you set some hook function here, that effect usually should not
1583 be inherited to another language environment. So, you had better set
1584 another function in `exit-language-environment-hook' (which see) to
1588 "Normal hook run after exiting from some language environment.
1589 When this hook is run, the variable `current-language-environment'
1590 is still bound to the language environment being exited.
1592 This hook is mainly used for canceling the effect of
1598 "Switch to a specified language environment."
1611 "The last language environment specified with `set-language-environment'.
1613 to using the function `set-language-environment'."
1620 current-language-environment)
1621 language-info-alist t))
1623 ;; custom type will be updated with `set-language-info'.
1634 "Reset multilingual environment of Emacs to the default status.
1636 The default status is as follows:
1638 The default value of `buffer-file-coding-system' is nil.
1639 The default coding system for process I/O is nil.
1640 The default value for the command `set-terminal-coding-system' is nil.
1641 The default value for the command `set-keyboard-coding-system' is nil.
1643 The order of priorities of coding categories and the coding system
1644 bound to each category are as follows
1645 coding category coding system
1646 --------------------------------------------------
1647 coding-category-iso-8-1 iso-latin-1
1648 coding-category-iso-8-2 iso-latin-1
1649 coding-category-utf-8 mule-utf-8
1650 coding-category-utf-16-be mule-utf-16be-with-signature
1651 coding-category-utf-16-le mule-utf-16le-with-signature
1652 coding-category-iso-7-tight iso-2022-jp
1653 coding-category-iso-7 iso-2022-7bit
1654 coding-category-iso-7-else iso-2022-7bit-lock
1655 coding-category-iso-8-else iso-2022-8bit-ss2
1656 coding-category-emacs-mule emacs-mule
1657 coding-category-raw-text raw-text
1658 coding-category-sjis japanese-shift-jis
1659 coding-category-big5 chinese-big5
1660 coding-category-ccl nil
1661 coding-category-binary no-conversion"
1663 ;; This function formerly set default-enable-multibyte-characters to t,
1664 ;; but that is incorrect. It should not alter the unibyte/multibyte choice.
1667 coding-category-iso-7 'iso-2022-7bit
1668 coding-category-iso-8-1 'iso-latin-1
1669 coding-category-iso-8-2 'iso-latin-1
1670 coding-category-iso-7-else 'iso-2022-7bit-lock
1671 coding-category-iso-8-else 'iso-2022-8bit-ss2
1672 coding-category-emacs-mule 'emacs-mule
1673 coding-category-raw-text 'raw-text
1674 coding-category-sjis 'japanese-shift-jis
1675 coding-category-big5 'chinese-big5
1676 coding-category-utf-16-be 'mule-utf-16be-with-signature
1677 coding-category-utf-16-le 'mule-utf-16le-with-signature
1678 coding-category-utf-8 'mule-utf-8
1679 coding-category-ccl nil
1684 coding-category-iso-8-2
1685 coding-category-utf-8
1686 coding-category-utf-16-be
1687 coding-category-utf-16-le
1688 coding-category-iso-7-tight
1689 coding-category-iso-7
1690 coding-category-iso-7-else
1691 coding-category-iso-8-else
1692 coding-category-emacs-mule
1693 coding-category-raw-text
1694 coding-category-sjis
1695 coding-category-big5
1696 coding-category-ccl
1697 coding-category-binary))
1699 ;; Changing the binding of a coding category requires this call.
1704 ;; On Darwin systems, this should be utf-8, but when this file is loaded
1705 ;; utf-8 is not yet defined, so we set it in set-locale-environment instead.
1707 ;; Preserve eol-type from existing default-process-coding-systems.
1708 ;; On non-unix-like systems in particular, these may have been set
1709 ;; carefully by the user, or by the startup code, to deal with the
1710 ;; users shell appropriately, so should not be altered by changing
1711 ;; language environment.
1713 ;; When bootstrapping, coding-systems are not defined yet, so
1714 ;; we need to catch the error from check-coding-system.
1727 ;; Don't alter the terminal and keyboard coding systems here.
1728 ;; The terminal still supports the same coding system
1729 ;; that it supported a minute ago.
1730 ;; (set-terminal-coding-system-internal nil)
1731 ;; (set-keyboard-coding-system-internal nil)
1736 ;; Don't invoke fontset-related functions if fontsets aren't
1737 ;; supported in this build of Emacs.
1744 "Set up the display table and terminal coding system for LANGUAGE-NAME."
1748 ;; The following 2 lines undo the 8-bit display that we set up
1749 ;; in standard-display-european-internal, which see. This is in
1750 ;; case the user has used standard-display-european earlier in
1751 ;; this session. (The MS-DOS port doesn't use that setup, so it
1752 ;; doesn't need to undo it.)
1760 "Set up multi-lingual environment for using LANGUAGE-NAME.
1761 This sets the coding system priority and the default input method
1762 and sometimes other things. LANGUAGE-NAME should be a string
1764 specifies the character set for the major languages of Western Europe."
1766 nil
1782 default-buffer-file-coding-system)))
1785 ;; The features might set up coding systems.
1809 ;; DOS terminals' default is to use a special non-ASCII translation
1810 ;; table as appropriate for the installed codepage.
1815 ;; Unibyte setups if necessary.
1817 ;; Syntax and case table.
1822 ;; No information for syntax and case. Reset to the defaults.
1846 ;; Don't invoke fontset-related functions if fontsets aren't
1847 ;; supported in this build of Emacs.
1858 utf-translate-cjk-lang-env
1864 nil))
1870 ;; Actually set up direct output of non-ASCII characters.
1872 ;; Unibyte Emacs on MS-DOS wants to display all 8-bit characters with
1873 ;; the native font, and codes 160 and 146 stand for something very
1874 ;; different there.
1877 ;; Most X fonts used to do the wrong thing for latin-1 code 160.
1879 ;; XFree86 4 has fixed the fonts.
1883 ;; Make non-line-break space display as a plain space.
1886 ;; don't contain a character at that position. Map it to the ASCII
1887 ;; apostrophe. [This is actually RIGHT SINGLE QUOTATION MARK,
1888 ;; U+2019, normally from the windows-1252 character set. XFree 4
1889 ;; fonts probably have the appropriate glyph at this position,
1890 ;; so they could use standard-display-8bit. It's better to use a
1891 ;; proper windows-1252 coding system. --fx]
1892 (aset standard-display-table 146 [39]))))
1894 (defun set-language-environment-coding-systems (language-name
1895 &optional eol-type)
1896 "Do various coding system setups for language environment LANGUAGE-NAME.
1898 The optional arg EOL-TYPE specifies the eol-type of the default value
1899 of `buffer-file-coding-system' set by this function."
1900 (let* ((priority (get-language-info language-name 'coding-priority))
1901 (default-coding (car priority)))
1902 (if priority
1903 (let ((categories (mapcar 'coding-system-category priority)))
1904 (set-default-coding-systems
1905 (if (memq eol-type '(0 1 2 unix dos mac))
1906 (coding-system-change-eol-conversion default-coding eol-type)
1907 default-coding))
1908 (setq default-sendmail-coding-system default-coding)
1909 (set-coding-priority categories)
1910 (while priority
1911 (set (car categories) (car priority))
1912 (setq priority (cdr priority) categories (cdr categories)))
1913 ;; Changing the binding of a coding category requires this call.
1914 (update-coding-systems-internal)))))
1916 (defsubst princ-list (&rest args)
1918 (while args (princ (car args)) (setq args (cdr args)))
1921 (put 'describe-specified-language-support 'apropos-inhibit t)
1923 ;; Print language-specific information such as input methods,
1924 ;; charsets, and coding systems. This function is intended to be
1925 ;; called from the menu:
1926 ;; [menu-bar mule describe-language-environment LANGUAGE]
1927 ;; and should not run it by `M-x describe-current-input-method-function'.
1928 (defun describe-specified-language-support ()
1929 "Describe how Emacs supports the specified language environment."
1930 (interactive)
1931 (let (language-name)
1932 (if (not (and (symbolp last-command-event)
1933 (or (not (eq last-command-event 'Default))
1934 (setq last-command-event 'English))
1935 (setq language-name (symbol-name last-command-event))))
1936 (error "Bogus calling sequence"))
1937 (describe-language-environment language-name)))
1939 (defun describe-language-environment (language-name)
1940 "Describe how Emacs supports language environment LANGUAGE-NAME."
1941 (interactive
1942 (list (read-language-name
1943 'documentation
1944 "Describe language environment (default, current choice): ")))
1945 (if (null language-name)
1946 (setq language-name current-language-environment))
1947 (if (or (null language-name)
1948 (null (get-language-info language-name 'documentation)))
1949 (error "No documentation for the specified language"))
1950 (if (symbolp language-name)
1951 (setq language-name (symbol-name language-name)))
1952 (dolist (feature (get-language-info language-name 'features))
1953 (require feature))
1954 (let ((doc (get-language-info language-name 'documentation)))
1955 (help-setup-xref (list #'describe-language-environment language-name)
1956 (interactive-p))
1957 (with-output-to-temp-buffer (help-buffer)
1958 (save-excursion
1959 (set-buffer standard-output)
1961 (if (stringp doc)
1963 (condition-case nil
1964 (let ((str (eval (get-language-info language-name 'sample-text))))
1965 (if (stringp str)
1967 (error nil))
1968 (let ((input-method (get-language-info language-name 'input-method))
1969 (l (copy-sequence input-method-alist)))
1970 (insert "Input methods")
1971 (when input-method
1972 (insert " (default, " input-method ")")
1973 (setq input-method (assoc input-method input-method-alist))
1974 (setq l (cons input-method (delete input-method l))))
1976 (while l
1977 (when (string= language-name (nth 1 (car l)))
1978 (insert " " (car (car l)))
1979 (search-backward (car (car l)))
1980 (help-xref-button 0 'help-input-method (car (car l)))
1981 (goto-char (point-max))
1983 (if (stringp (nth 3 (car l)))
1984 (nth 3 (car l))
1985 (car (nth 3 (car l))))
1987 (setq l (cdr l)))
1990 (let ((l (get-language-info language-name 'charset)))
1991 (if (null l)
1993 (while l
1994 (insert " " (symbol-name (car l)))
1995 (search-backward (symbol-name (car l)))
1996 (help-xref-button 0 'help-character-set (car l))
1997 (goto-char (point-max))
1999 (setq l (cdr l)))))
2002 (let ((l (get-language-info language-name 'coding-system)))
2003 (if (null l)
2005 (while l
2006 (insert " " (symbol-name (car l)))
2007 (search-backward (symbol-name (car l)))
2008 (help-xref-button 0 'help-coding-system (car l))
2009 (goto-char (point-max))
2010 (insert " (`"
2011 (coding-system-mnemonic (car l))
2013 (coding-system-doc-string (car l))
2015 (let ((aliases (coding-system-get (car l)
2016 'alias-coding-systems)))
2017 (when aliases
2019 (while aliases
2020 (insert " " (symbol-name (car aliases)))
2021 (setq aliases (cdr aliases)))
2023 (setq l (cdr l)))))))))
2024 \f
2025 ;;; Locales.
2027 (defvar locale-translation-file-name nil
2028 "File name for the system's file of locale-name aliases, or nil if none.")
2030 ;; The following definitions might as well be marked as constants and
2031 ;; purecopied, since they're normally used on startup, and probably
2032 ;; should reflect the facilities of the base Emacs.
2033 (defconst locale-language-names
2034 (purecopy
2035 '(
2036 ;; Locale names of the form LANGUAGE[_TERRITORY][.CODESET][@MODIFIER]
2037 ;; as specified in the Single Unix Spec, Version 2.
2038 ;; LANGUAGE is a language code taken from ISO 639:1988 (E/F)
2039 ;; with additions from ISO 639/RA Newsletter No.1/1989;
2040 ;; see Internet RFC 2165 (1997-06) and
2041 ;; http://www.evertype.com/standards/iso639/iso639-en.html
2042 ;; TERRITORY is a country code taken from ISO 3166
2043 ;; http://www.din.de/gremien/nas/nabd/iso3166ma/codlstp1/en_listp1.html.
2044 ;; CODESET and MODIFIER are implementation-dependent.
2046 ;; jasonr comments: MS Windows uses three letter codes for
2047 ;; languages instead of the two letter ISO codes that POSIX
2048 ;; uses. In most cases the first two letters are the same, so
2049 ;; most of the regexps in locale-language-names work. Japanese
2050 ;; and Chinese are exceptions, which are listed in the
2051 ;; non-standard section at the bottom of locale-language-names.
2053 ("aa_DJ" . "Latin-1") ; Afar
2054 ("aa" . "UTF-8")
2055 ;; ab Abkhazian
2056 ("af" . "Latin-1") ; Afrikaans
2057 ("am" "Ethiopic" utf-8) ; Amharic
2058 ("an" . "Latin-9") ; Aragonese
2059 ; ar Arabic glibc uses 8859-6
2060 ; as Assamese
2061 ; ay Aymara
2062 ("az" . "UTF-8") ; Azerbaijani
2063 ; ba Bashkir
2064 ("be" "Belarusian" cp1251) ; Belarusian [Byelorussian until early 1990s]
2065 ("bg" "Bulgarian" cp1251) ; Bulgarian
2066 ; bh Bihari
2067 ; bi Bislama
2068 ("bn" . "UTF-8") ; Bengali, Bangla
2069 ("bo" . "Tibetan")
2070 ("br" . "Latin-1") ; Breton
2071 ("bs" . "Latin-2") ; Bosnian
2072 ("byn" . "UTF-8") ; Bilin; Blin
2073 ("ca" . "Latin-1") ; Catalan
2074 ; co Corsican
2075 ("cs" "Czech" iso-8859-2)
2076 ("cy" "Welsh" iso-8859-14)
2077 ("da" . "Latin-1") ; Danish
2078 ("de" "German" iso-8859-1)
2079 ; dz Bhutani
2080 ("el" "Greek" iso-8859-7)
2081 ;; Users who specify "en" explicitly typically want Latin-1, not ASCII.
2082 ;; That's actually what the GNU locales define, modulo things like
2083 ;; en_IN -- fx.
2084 ("en_IN" "English" utf-8) ; glibc uses utf-8 for English in India
2085 ("en" . "Latin-1") ; English
2086 ("eo" . "Latin-3") ; Esperanto
2087 ("es" "Spanish" iso-8859-1)
2088 ("et" . "Latin-1") ; Estonian
2089 ("eu" . "Latin-1") ; Basque
2090 ("fa" . "UTF-8") ; Persian
2091 ("fi" . "Latin-1") ; Finnish
2092 ("fj" . "Latin-1") ; Fiji
2093 ("fo" . "Latin-1") ; Faroese
2094 ("fr" "French" iso-8859-1) ; French
2095 ("fy" . "Latin-1") ; Frisian
2096 ("ga" . "Latin-1") ; Irish Gaelic (new orthography)
2097 ("gd" . "Latin-9") ; Scots Gaelic
2098 ("gez" "Ethiopic" utf-8) ; Geez
2099 ("gl" . "Latin-1") ; Gallegan; Galician
2100 ; gn Guarani
2101 ("gu" . "UTF-8") ; Gujarati
2102 ("gv" . "Latin-1") ; Manx Gaelic
2103 ; ha Hausa
2104 ("he" "Hebrew" iso-8859-8)
2105 ("hi" "Devanagari" utf-8) ; Hindi
2106 ("hr" "Croatian" iso-8859-2) ; Croatian
2107 ("hu" . "Latin-2") ; Hungarian
2108 ; hy Armenian
2109 ; ia Interlingua
2110 ("id" . "Latin-1") ; Indonesian
2111 ; ie Interlingue
2112 ; ik Inupiak
2113 ("is" . "Latin-1") ; Icelandic
2114 ("it" "Italian" iso-8859-1) ; Italian
2115 ; iu Inuktitut
2116 ("iw" "Hebrew" iso-8859-8)
2117 ("ja" "Japanese" euc-jp)
2118 ; jw Javanese
2119 ("ka" "Georgian" georgian-ps) ; Georgian
2120 ; kk Kazakh
2121 ("kl" . "Latin-1") ; Greenlandic
2122 ; km Cambodian
2123 ("kn" "Kannada" utf-8)
2124 ("ko" "Korean" euc-kr)
2125 ; ks Kashmiri
2126 ; ku Kurdish
2127 ("kw" . "Latin-1") ; Cornish
2128 ; ky Kirghiz
2129 ("la" . "Latin-1") ; Latin
2130 ("lb" . "Latin-1") ; Luxemburgish
2131 ("lg" . "Laint-6") ; Ganda
2132 ; ln Lingala
2133 ("lo" "Lao" utf-8) ; Laothian
2134 ("lt" "Lithuanian" iso-8859-13)
2135 ("lv" . "Latvian") ; Latvian, Lettish
2136 ; mg Malagasy
2137 ("mi" . "Latin-7") ; Maori
2138 ("mk" "Cyrillic-ISO" iso-8859-5) ; Macedonian
2139 ("ml" "Malayalam" utf-8)
2140 ("mn" . "UTF-8") ; Mongolian
2141 ; mo Moldavian
2142 ("mr" "Devanagari" utf-8) ; Marathi
2143 ("ms" . "Latin-1") ; Malay
2144 ("mt" . "Latin-3") ; Maltese
2145 ; my Burmese
2146 ; na Nauru
2147 ("nb" . "Latin-1") ; Norwegian
2148 ("ne" "Devanagari" utf-8) ; Nepali
2149 ("nl" "Dutch" iso-8859-1)
2150 ("no" . "Latin-1") ; Norwegian
2151 ("oc" . "Latin-1") ; Occitan
2152 ("om_ET" . "UTF-8") ; (Afan) Oromo
2153 ("om" . "Latin-1") ; (Afan) Oromo
2154 ; or Oriya
2155 ("pa" . "UTF-8") ; Punjabi
2156 ("pl" . "Latin-2") ; Polish
2157 ; ps Pashto, Pushto
2158 ("pt" . "Latin-1") ; Portuguese
2159 ; qu Quechua
2160 ("rm" . "Latin-1") ; Rhaeto-Romanic
2161 ; rn Kirundi
2162 ("ro" "Romanian" iso-8859-2)
2163 ("ru_RU" "Russian" iso-8859-5)
2164 ("ru_UA" "Russian" koi8-u)
2165 ; rw Kinyarwanda
2166 ("sa" . "Devanagari") ; Sanskrit
2167 ; sd Sindhi
2168 ("se" . "UTF-8") ; Northern Sami
2169 ; sg Sangho
2170 ("sh" . "Latin-2") ; Serbo-Croatian
2171 ; si Sinhalese
2172 ("sid" . "UTF-8") ; Sidamo
2173 ("sk" "Slovak" iso-8859-2)
2174 ("sl" "Slovenian" iso-8859-2)
2175 ; sm Samoan
2176 ; sn Shona
2177 ("so_ET" "UTF-8") ; Somali
2178 ("so" "Latin-1") ; Somali
2179 ("sq" . "Latin-1") ; Albanian
2180 ("sr_YU@cyrillic" . "Cyrillic-ISO") ; Serbian (Cyrillic alphabet)
2181 ("sr" . "Latin-2") ; Serbian (Latin alphabet)
2182 ; ss Siswati
2183 ("st" . "Latin-1") ; Sesotho
2184 ; su Sundanese
2185 ("sv" "Swedish" iso-8859-1) ; Swedish
2186 ("sw" . "Latin-1") ; Swahili
2187 ("ta" "Tamil" utf-8)
2188 ("te" . "UTF-8") ; Telugu
2189 ("tg" "Tajik" koi8-t)
2190 ("th" "Thai" tis-620)
2191 ("ti" "Ethiopic" utf-8) ; Tigrinya
2192 ("tig_ER" . "UTF-8") ; Tigre
2193 ; tk Turkmen
2194 ("tl" . "Latin-1") ; Tagalog
2195 ; tn Setswana
2196 ; to Tonga
2197 ("tr" "Turkish" iso-8859-9)
2198 ; ts Tsonga
2199 ("tt" . "UTF-8") ; Tatar
2200 ; tw Twi
2201 ; ug Uighur
2202 ("uk" "Ukrainian" koi8-u)
2203 ("ur" . "UTF-8") ; Urdu
2204 ("uz_UZ@cyrillic" . "UTF-8"); Uzbek
2205 ("uz" . "Latin-1") ; Uzbek
2206 ("vi" "Vietnamese" utf-8)
2207 ; vo Volapuk
2208 ("wa" . "Latin-1") ; Walloon
2209 ; wo Wolof
2210 ("xh" . "Latin-1") ; Xhosa
2211 ("yi" . "Windows-1255") ; Yiddish
2212 ; yo Yoruba
2213 ; za Zhuang
2214 ("zh_HK" . "Chinese-Big5")
2215 ("zh_TW" . "Chinese-Big5")
2216 ("zh_CN" . "Chinese-GB")
2217 ("zh" . "Chinese-GB")
2218 ; zh_CN.GB18030/GB18030 \
2219 ; zh_CN.GBK/GBK \
2220 ; zh_HK/BIG5-HKSCS \
2221 ("zu" . "Latin-1") ; Zulu
2223 ;; ISO standard locales
2224 ("c$" . "ASCII")
2225 ("posix$" . "ASCII")
2227 ;; The "IPA" Emacs language environment does not correspond
2228 ;; to any ISO 639 code, so let it stand for itself.
2229 ("ipa$" . "IPA")
2231 ;; Nonstandard or obsolete language codes
2232 ("cz" . "Czech") ; e.g. Solaris 2.6
2233 ("ee" . "Latin-4") ; Estonian, e.g. X11R6.4
2234 ("iw" . "Hebrew") ; e.g. X11R6.4
2235 ("sp" . "Cyrillic-ISO") ; Serbian (Cyrillic alphabet), e.g. X11R6.4
2236 ("su" . "Latin-1") ; Finnish, e.g. Solaris 2.6
2237 ("jp" . "Japanese") ; e.g. MS Windows
2238 ("chs" . "Chinese-GB") ; MS Windows Chinese Simplified
2239 ("cht" . "Chinese-BIG5") ; MS Windows Chinese Traditional
2240 ))
2241 "Alist of locale regexps vs the corresponding languages and coding systems.
2242 Each element has these form:
2243 \(LOCALE-REGEXP LANG-ENV CODING-SYSTEM)
2244 The first element whose LOCALE-REGEXP matches the start of a
2245 downcased locale specifies the LANG-ENV \(language environtment)
2246 and CODING-SYSTEM corresponding to that locale. If there is no
2247 appropriate language environment, the element may have this form:
2248 \(LOCALE-REGEXP . LANG-ENV)
2249 In this case, LANG-ENV is one of generic language environments for an
2252 (defconst locale-charset-language-names
2253 (purecopy
2262 ;; utf-8@euro exists, so put this last. (@euro really specifies
2263 ;; the currency, rather than the charset.)
2265 "List of pairs of locale regexps and charset language names.
2266 The first element whose locale regexp matches the start of a downcased locale
2267 specifies the language name whose charset corresponds to that locale.
2268 This language name is used if the locale is not listed in
2269 `locale-language-names'")
2271 (defconst locale-preferred-coding-systems
2272 (purecopy
2281 ;; utf-8@euro exists, so put this after utf-8. (@euro really
2282 ;; specifies the currency, rather than the charset.)
2283 (".*@euro" . iso-8859-15)
2284 ("koi8-?r" . koi8-r)
2285 ("koi8-?u" . koi8-u)
2286 ("tcvn" . tcvn)
2287 ("big5" . big5)
2288 ("euc-?tw" . euc-tw)
2289 ;; We don't support GBK, but as it is upper compatible with
2290 ;; GB-2312, we setup the default coding system to gb2312.
2291 ("gbk" . gb2312)
2292 ;; We don't support BIG5-HKSCS, but as it is upper compatible with
2293 ;; BIG5, we setup the default coding system to big5.
2294 ("big5hkscs" . big5)
2295 ("ja.*[._]euc" . japanese-iso-8bit)
2296 ("ja.*[._]jis7" . iso-2022-jp)
2297 ("ja.*[._]pck" . japanese-shift-jis)
2298 ("ja.*[._]sjis" . japanese-shift-jis)
2299 ("jpn" . japanese-shift-jis) ; MS-Windows uses this.
2300 ))
2301 "List of pairs of locale regexps and preferred coding systems.
2302 The first element whose locale regexp matches the start of a downcased locale
2303 specifies the coding system to prefer when using that locale.
2304 This coding system is used if the locale specifies a specific charset.")
2306 (defun locale-name-match (key alist)
2307 "Search for KEY in ALIST, which should be a list of regexp-value pairs.
2308 Return the value corresponding to the first regexp that matches the
2309 start of KEY, or nil if there is no match."
2310 (let (element)
2311 (while (and alist (not element))
2313 (setq element (car alist)))
2314 (setq alist (cdr alist)))
2315 (cdr element)))
2317 (defun locale-charset-match-p (charset1 charset2)
2318 "Whether charset names (strings) CHARSET1 and CHARSET2 are equivalent.
2319 Matching is done ignoring case and any hyphens and underscores in the
2320 names. E.g. `ISO_8859-1' and `iso88591' both match `iso-8859-1'."
2321 (setq charset1 (replace-regexp-in-string "[-_]" "" charset1))
2322 (setq charset2 (replace-regexp-in-string "[-_]" "" charset2))
2323 (eq t (compare-strings charset1 nil nil charset2 nil nil t)))
2325 (defvar locale-charset-alist nil
2326 "Coding system alist keyed on locale-style charset name.
2327 Used by `locale-charset-to-coding-system'.")
2329 (defun locale-charset-to-coding-system (charset)
2330 "Find coding system corresponding to CHARSET.
2331 CHARSET is any sort of non-Emacs charset name, such as might be used
2332 in a locale codeset, or elsewhere. It is matched to a coding system
2333 first by case-insensitive lookup in `locale-charset-alist'. Then
2334 matches are looked for in the coding system list, treating case and
2335 the characters `-' and `_' as insignificant. The coding system base
2337 `iso-latin-2' is returned."
2338 (or (car (assoc-string charset locale-charset-alist t))
2339 (let ((cs coding-system-alist)
2340 c)
2341 (while (and (not c) cs)
2342 (if (locale-charset-match-p charset (caar cs))
2343 (setq c (intern (caar cs)))
2344 (pop cs)))
2345 (if c (coding-system-base c)))))
2347 ;; Fixme: This ought to deal with the territory part of the locale
2348 ;; too, for setting things such as calendar holidays, ps-print paper
2349 ;; size, spelling dictionary.
2351 (defun set-locale-environment (&optional locale-name)
2352 "Set up multi-lingual environment for using LOCALE-NAME.
2353 This sets the language environment, the coding system priority,
2354 the default input method and sometimes other things.
2356 LOCALE-NAME should be a string which is the name of a locale supported
2357 by the system. Often it is of the form xx_XX.CODE, where xx is a
2358 language, XX is a country, and CODE specifies a character set and
2360 a locale for Japanese in Japan using the `japanese-iso-8bit'
2361 coding-system. The name may also have a modifier suffix, e.g. `@euro'
2362 or `@cyrillic'.
2364 If LOCALE-NAME is nil, its value is taken from the environment
2365 variables LC_ALL, LC_CTYPE and LANG (the first one that is set).
2367 The locale names supported by your system can typically be found in a
2368 directory named `/usr/share/locale' or `/usr/lib/locale'. LOCALE-NAME
2369 will be translated according to the table specified by
2370 `locale-translation-file-name'.
2372 See also `locale-charset-language-names', `locale-language-names',
2373 `locale-preferred-coding-systems' and `locale-coding-system'."
2374 (interactive "sSet environment for locale: ")
2376 ;; Do this at runtime for the sake of binaries possibly transported
2377 ;; to a system without X.
2378 (setq locale-translation-file-name
2379 (let ((files
2380 '("/usr/lib/X11/locale/locale.alias" ; e.g. X11R6.4
2381 "/usr/X11R6/lib/X11/locale/locale.alias" ; XFree86, e.g. RedHat 4.2
2382 "/usr/openwin/lib/locale/locale.alias" ; e.g. Solaris 2.6
2383 ;;
2384 ;; The following name appears after the X-related names above,
2385 ;; since the X-related names are what X actually uses.
2386 "/usr/share/locale/locale.alias" ; GNU/Linux sans X
2387 )))
2388 (while (and files (not (file-exists-p (car files))))
2389 (setq files (cdr files)))
2390 (car files)))
2392 (let ((locale locale-name))
2394 (unless locale
2395 ;; Use the first of these three environment variables
2396 ;; that has a nonempty value.
2397 (let ((vars '("LC_ALL" "LC_CTYPE" "LANG")))
2398 (while (and vars
2399 (= 0 (length locale))) ; nil or empty string
2400 (setq locale (getenv (pop vars))))))
2402 (unless (or locale (not (fboundp 'mac-get-preference)))
2403 (setq locale (mac-get-preference "AppleLocale"))
2404 (unless locale
2405 (let ((languages (mac-get-preference "AppleLanguages")))
2406 (unless (= (length languages) 0) ; nil or empty vector
2407 (setq locale (aref languages 0))))))
2408 (unless (or locale (not (boundp 'mac-system-locale)))
2409 (setq locale mac-system-locale))
2411 (when locale
2413 ;; Translate "swedish" into "sv_SE.ISO8859-1", and so on,
2414 ;; using the translation file that many systems have.
2415 (when locale-translation-file-name
2416 (with-temp-buffer
2417 (insert-file-contents locale-translation-file-name)
2418 (when (re-search-forward
2420 (setq locale (buffer-substring (point) (line-end-position))))))
2422 ;; Leave the system locales alone if the caller did not specify
2423 ;; an explicit locale name, as their defaults are set from
2424 ;; LC_MESSAGES and LC_TIME, not LC_CTYPE, and the user might not
2425 ;; want to set them to the same value as LC_CTYPE.
2426 (when locale-name
2427 (setq system-messages-locale locale)
2428 (setq system-time-locale locale))
2430 (setq locale (downcase locale))
2432 (let ((language-name
2433 (locale-name-match locale locale-language-names))
2434 (charset-language-name
2435 (locale-name-match locale locale-charset-language-names))
2436 (coding-system
2437 (or (locale-name-match locale locale-preferred-coding-systems)
2438 (when locale
2440 (locale-charset-to-coding-system
2441 (match-string 1 locale))))
2442 (and (eq system-type 'macos) mac-system-coding-system))))
2444 (if (consp language-name)
2445 ;; locale-language-names specify both lang-env and coding.
2446 ;; But, what specified in locale-preferred-coding-systems
2447 ;; has higher priority.
2448 (setq coding-system (or coding-system
2449 (nth 1 language-name))
2450 language-name (car language-name))
2451 ;; Otherwise, if locale is not listed in locale-language-names,
2452 ;; use what listed in locale-charset-language-names.
2453 (if (not language-name)
2454 (setq language-name charset-language-name)))
2456 (when language-name
2458 ;; Set up for this character set. This is now the right way
2459 ;; to do it for both unibyte and multibyte modes.
2460 (set-language-environment language-name)
2462 ;; If default-enable-multibyte-characters is nil,
2463 ;; we are using single-byte characters,
2464 ;; so the display table and terminal coding system are irrelevant.
2465 (when default-enable-multibyte-characters
2466 (set-display-table-and-terminal-coding-system
2467 language-name coding-system))
2469 ;; Set the `keyboard-coding-system' if appropriate (tty
2470 ;; only). At least X and MS Windows can generate
2471 ;; multilingual input.
2472 (unless window-system
2473 (let ((kcs (or coding-system
2474 (car (get-language-info language-name
2475 'coding-system)))))
2476 (if kcs (set-keyboard-coding-system kcs))))
2478 (setq locale-coding-system
2479 (car (get-language-info language-name 'coding-priority))))
2481 (when (and coding-system
2482 (not (coding-system-equal coding-system
2483 locale-coding-system)))
2484 (prefer-coding-system coding-system)
2485 (setq locale-coding-system coding-system))))
2487 ;; On Windows, override locale-coding-system,
2488 ;; keyboard-coding-system with system codepage. Note:
2489 ;; selection-coding-system is already set in w32select.c.
2490 (when (boundp 'w32-ansi-code-page)
2491 (let ((code-page-coding (intern (format "cp%d" w32-ansi-code-page))))
2492 (when (coding-system-p code-page-coding)
2493 (setq locale-coding-system code-page-coding)
2494 (set-keyboard-coding-system code-page-coding)
2495 (set-terminal-coding-system code-page-coding))))
2497 (when (eq system-type 'darwin)
2498 ;; On Darwin, file names are always encoded in utf-8, no matter
2499 ;; the locale.
2500 (setq default-file-name-coding-system 'utf-8)
2501 ;; Mac OS X's Terminal.app by default uses utf-8 regardless of
2502 ;; the locale.
2503 (when (and (null window-system)
2504 (equal (getenv "TERM_PROGRAM") "Apple_Terminal"))
2505 (set-terminal-coding-system 'utf-8)
2506 (set-keyboard-coding-system 'utf-8)))
2508 ;; Default to A4 paper if we're not in a C, POSIX or US locale.
2509 ;; (See comments in Flocale_info.)
2510 (let ((locale locale)
2511 (paper (locale-info 'paper)))
2512 (if paper
2513 ;; This will always be null at the time of writing.
2514 (cond
2515 ((equal paper '(216 279))
2516 (setq ps-paper-type 'letter))
2517 ((equal paper '(210 297))
2518 (setq ps-paper-type 'a4)))
2519 (let ((vars '("LC_ALL" "LC_PAPER" "LANG")))
2520 (while (and vars (= 0 (length locale)))
2521 (setq locale (getenv (pop vars)))))
2522 (when locale
2523 ;; As of glibc 2.2.5, these are the only US Letter locales,
2524 ;; and the rest are A4.
2525 (setq ps-paper-type
2526 (or (locale-name-match locale '(("c$" . letter)
2527 ("posix$" . letter)
2528 (".._us" . letter)
2529 (".._pr" . letter)
2530 (".._ca" . letter)
2531 ("enu$" . letter) ; Windows
2532 ("esu$" . letter)
2533 ("enc$" . letter)
2534 ("frc$" . letter)))
2535 'a4))))))
2536 nil)
2537 \f
2538 ;;; Charset property
2540 (defun get-charset-property (charset propname)
2541 "Return the value of CHARSET's PROPNAME property.
2542 This is the last value stored with
2543 (put-charset-property CHARSET PROPNAME VALUE)."
2544 (and (not (eq charset 'composition))
2545 (plist-get (charset-plist charset) propname)))
2547 (defun put-charset-property (charset propname value)
2548 "Store CHARSETS's PROPNAME property with value VALUE.
2549 It can be retrieved with `(get-charset-property CHARSET PROPNAME)'."
2550 (or (eq charset 'composition)
2551 (set-charset-plist charset
2552 (plist-put (charset-plist charset) propname value))))
2554 ;;; Character code property
2555 (put 'char-code-property-table 'char-table-extra-slots 0)
2557 (defvar char-code-property-table
2558 (make-char-table 'char-code-property-table)
2559 "Char-table containing a property list of each character code.
2561 See also the documentation of `get-char-code-property' and
2562 `put-char-code-property'.")
2564 (defun get-char-code-property (char propname)
2565 "Return the value of CHAR's PROPNAME property in `char-code-property-table'."
2566 (let ((plist (aref char-code-property-table char)))
2567 (if (listp plist)
2568 (car (cdr (memq propname plist))))))
2570 (defun put-char-code-property (char propname value)
2571 "Store CHAR's PROPNAME property with VALUE in `char-code-property-table'.
2572 It can be retrieved with `(get-char-code-property CHAR PROPNAME)'."
2573 (let ((plist (aref char-code-property-table char)))
2574 (if plist
2575 (let ((slot (memq propname plist)))
2576 (if slot
2577 (setcar (cdr slot) value)
2578 (nconc plist (list propname value))))
2579 (aset char-code-property-table char (list propname value)))))
2581 \f
2582 ;; Pretty description of encoded string
2584 ;; Alist of ISO 2022 control code vs the corresponding mnemonic string.
2585 (defvar iso-2022-control-alist
2593 (defun encoded-string-description (str coding-system)
2594 "Return a pretty description of STR that is encoded by CODING-SYSTEM."
2595 (setq str (string-as-unibyte str))
2596 (mapconcat
2597 (if (and coding-system (eq (coding-system-type coding-system) 2))
2598 ;; Try to get a pretty description for ISO 2022 escape sequences.
2599 (function (lambda (x) (or (cdr (assq x iso-2022-control-alist))
2600 (format "0x%02X" x))))
2601 (function (lambda (x) (format "0x%02X" x))))
2602 str " "))
2604 (defun encode-coding-char (char coding-system)
2605 "Encode CHAR by CODING-SYSTEM and return the resulting string.
2606 If CODING-SYSTEM can't safely encode CHAR, return nil."
2607 (let ((str1 (string-as-multibyte (char-to-string char)))
2608 (str2 (string-as-multibyte (make-string 2 char)))
2609 (safe-chars (and coding-system
2610 (coding-system-get coding-system 'safe-chars)))
2611 (charset (char-charset char))
2612 enc1 enc2 i1 i2)
2613 (when (or (eq safe-chars t)
2614 (eq charset 'ascii)
2615 (and safe-chars (aref safe-chars char)))
2616 ;; We must find the encoded string of CHAR. But, just encoding
2617 ;; CHAR will put extra control sequences (usually to designate
2618 ;; ASCII charset) at the tail if type of CODING is ISO 2022.
2619 ;; To exclude such tailing bytes, we at first encode one-char
2620 ;; string and two-char string, then check how many bytes at the
2621 ;; tail of both encoded strings are the same.
2623 (setq enc1 (encode-coding-string str1 coding-system)
2624 i1 (length enc1)
2625 enc2 (encode-coding-string str2 coding-system)
2626 i2 (length enc2))
2627 (while (and (> i1 0) (= (aref enc1 (1- i1)) (aref enc2 (1- i2))))
2628 (setq i1 (1- i1) i2 (1- i2)))
2630 ;; Now (substring enc1 i1) and (substring enc2 i2) are the same,
2631 ;; and they are the extra control sequences at the tail to
2632 ;; exclude.
2633 (substring enc2 0 i2))))
2636 ;; arch-tag: b382c432-4b36-460e-bf4c-05efd0bb18dc
2637 ;;; mule-cmds.el ends here