| blob | a596411eb7881e24b58d93243993816e6503ef6f |
1 ;;; mule-cmds.el --- commands for multilingual environment -*- lexical-binding:t -*-
3 ;; Copyright (C) 1997-2017 Free Software Foundation, Inc.
4 ;; Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
5 ;; 2005, 2006, 2007, 2008, 2009, 2010, 2011
6 ;; National Institute of Advanced Industrial Science and Technology (AIST)
7 ;; Registration Number H14PRO021
8 ;; Copyright (C) 2003
9 ;; National Institute of Advanced Industrial Science and Technology (AIST)
10 ;; Registration Number H13PRO009
12 ;; Keywords: mule, i18n
14 ;; This file is part of GNU Emacs.
16 ;; GNU Emacs is free software: you can redistribute it and/or modify
17 ;; it under the terms of the GNU General Public License as published by
18 ;; the Free Software Foundation, either version 3 of the License, or
19 ;; (at your option) any later version.
21 ;; GNU Emacs is distributed in the hope that it will be useful,
22 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
23 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
24 ;; GNU General Public License for more details.
26 ;; You should have received a copy of the GNU General Public License
27 ;; along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>.
29 ;;; Commentary:
31 ;;; Code:
38 ;;; MULE related key bindings and menus.
53 map)
56 ;; Keep "C-x C-m ..." for mule specific commands.
62 [Default] '(menu-item "Default" describe-specified-language-support))
63 map))
65 (defvar setup-language-environment-map
66 (let ((map (make-sparse-keymap "Set Language Environment")))
67 (bindings--define-key map
69 map))
102 revert-buffer-with-coding-system
103 :enable buffer-file-name
112 map))
129 ,describe-language-environment-map
153 map)
156 ;; This should be a single character key binding because users use it
157 ;; very frequently while editing multilingual text. Now we can use
158 ;; only two such keys: "\C-\\" and "\C-^", but the latter is not
159 ;; convenient because it requires shifting on most keyboards. An
160 ;; alternative is "\C-]" which is now bound to `abort-recursive-edit'
161 ;; but it won't be used that frequently.
164 ;; This is no good because people often type Shift-SPC
165 ;; meaning to type SPC. -- rms.
166 ;; ;; Here's an alternative key binding for X users (Shift-SPACE).
167 ;; (define-key global-map [?\S- ] 'toggle-input-method)
169 ;;; Mule related hyperlinks.
177 ;; Note starting with word-syntax character:
181 "Return a coding system which differs from CODING-SYSTEM in EOL conversion.
182 The returned coding system converts end-of-line by EOL-TYPE
183 but text as the same way as CODING-SYSTEM.
184 EOL-TYPE should be `unix', `dos', `mac', or nil.
185 If EOL-TYPE is nil, the returned coding system detects
186 how end-of-line is formatted automatically while decoding.
188 EOL-TYPE can be specified by an integer 0, 1, or 2.
189 They means `unix', `dos', and `mac' respectively."
195 ;; We call `coding-system-base' before `coding-system-eol-type',
196 ;; because the coding-system may not be initialized until then.
201 coding-system
204 base)
206 coding-system)
212 "Return a coding system which differs from CODING-SYSTEM in text conversion.
213 The returned coding system converts text by CODING
214 but end-of-line as the same way as CODING-SYSTEM.
215 If CODING is nil, the returned coding system detects
216 how text is formatted automatically while decoding."
222 ;; Canonicalize the coding system name NAME by removing some prefixes
223 ;; and delimiter characters. Support function of
224 ;; coding-system-from-name.
227 ;; "ms950", "ibm950", "windows-950" -> "cp950"
230 ;; "iso-8859-1" -> "8859-1", "iso-2022-jp" ->"2022-jp"
233 ;; Delete "-", "_", " ", "/" but do distinguish "16-be" and "16be".
241 name)))
244 "Return a coding system whose name matches with NAME (string or symbol)."
249 sym
258 name)
260 elt eol-type)
261 elt)))))))))
264 "Change whether this buffer uses multibyte characters.
265 With ARG, use multibyte characters if the ARG is positive.
267 Note that this command does not convert the byte contents of
268 the buffer; it only changes the way those bytes are interpreted.
269 In general, therefore, this command *changes* the sequence of
270 characters that the current buffer contains.
272 We suggest you avoid using this command unless you know what you are
273 doing. If you use it by mistake, and the buffer is now displayed
274 wrong, use this command again to toggle back to the right mode."
283 "Display the HELLO file, which lists many languages and characters."
285 ;; We have to decode the file in any environment.
290 "Execute an I/O command using the specified coding system."
295 buffer-file-coding-system)))
300 default))))
304 prefix)
305 ;; read-key-sequence ignores quit, so make an explicit check.
306 ;; Like many places, this assumes quit == C-g, but it need not be.
312 ;; Process keys bound in `universal-argument-map'.
318 ;; Have to bind `last-command-event' here so that
319 ;; `digit-argument', for instance, can compute the
320 ;; prefix arg.
324 ;; This is the final call to `universal-argument-other-key', which
325 ;; set's the final `prefix-arg.
329 ;; Read the command to execute with the given prefix arg.
342 "Set default value of various coding systems to CODING-SYSTEM.
343 This sets the following coding systems:
344 o coding system of a newly created buffer
345 o default coding system for subprocess I/O
346 This also sets the following values:
347 o default value used as `file-name-coding-system' for converting file names
348 if CODING-SYSTEM is ASCII-compatible
349 o default value for the command `set-terminal-coding-system'
350 o default value for the command `set-keyboard-coding-system'
351 if CODING-SYSTEM is ASCII-compatible"
356 ;; The file-name coding system on Darwin systems is always utf-8.
364 ;; Prevent default-terminal-coding-system from converting ^M to ^J.
367 ;; Preserve eol-type from existing default-process-coding-systems.
368 ;; On non-unix-like systems in particular, these may have been set
369 ;; carefully by the user, or by the startup code, to deal with the
370 ;; users shell appropriately, so should not be altered by changing
371 ;; language environment.
382 "Add CODING-SYSTEM at the front of the priority list for automatic detection.
383 This also sets the following coding systems:
384 o coding system of a newly created buffer
385 o default coding system for subprocess I/O
386 This also sets the following values:
387 o default value used as `file-name-coding-system' for converting file names
388 o default value for the command `set-terminal-coding-system'
389 o default value for the command `set-keyboard-coding-system'
391 If CODING-SYSTEM specifies a certain type of EOL conversion, the coding
392 systems set by this function will use that type of EOL conversion.
394 A coding system that requires automatic detection of text+encoding
395 \(e.g. undecided, unix) can't be preferred.
397 To prefer, for instance, utf-8, say the following:
411 base coding-system)))
412 ;; If they asked for specific EOL conversion, honor that.
419 (message "The default value of `file-name-coding-system' was not changed because the specified coding system is not suitable for file names.")))))
422 "If non-nil, a predicate function to sort coding systems.
424 It is called with two coding systems, and should return t if the first
430 "Sort coding system list CODINGS by a priority of each coding system.
431 Return the sorted list. CODINGS is modified by side effects.
433 If a coding system is most preferred, it has the highest priority.
434 Otherwise, coding systems that correspond to MIME charsets have
435 higher priorities. Among them, a coding system included in the
436 `coding-system' key of the current language environment has higher
437 priority. See also the documentation of `language-info-alist'.
439 If the variable `sort-coding-systems-predicate' (which see) is
440 non-nil, it is used to sort CODINGS instead."
450 ;; We calculate the priority number 0..255 by
451 ;; using the 8 bits PMMLCEII as this:
452 ;; P: 1 if most preferred.
453 ;; MM: greater than 0 if mime-charset.
454 ;; L: 1 if one of the current lang. env.'s codings.
455 ;; C: 1 if one of codings listed in the category list.
456 ;; E: 1 if not XXX-with-esc
457 ;; II: if iso-2022 based, 0..3, else 1.
462 ;; Prefer coding systems corresponding to a
463 ;; MIME charset.
465 ;; Lower utf-16 priority so that we
466 ;; normally prefer utf-8 to it, and put
467 ;; x-ctext below that.
483 ;; For ISO based coding systems, prefer
484 ;; one that doesn't use designation nor
485 ;; locking/single shifting.
496 ))))))
501 "Return a list of proper coding systems to encode a text between FROM and TO.
503 If FROM is a string, find coding systems in that instead of the buffer.
504 All coding systems in the list can safely encode any multibyte characters
505 in the text.
507 If the text contains no multibyte characters, return a list of a single
508 element `undecided'."
511 ;; The text contains only ASCII characters. Any coding
512 ;; systems are safe.
514 ;; We need copy-sequence because sorting will alter the argument.
518 "Return a list of proper coding systems to encode STRING.
519 All coding systems in the list can safely encode any multibyte characters
520 in STRING.
522 If STRING contains no multibyte characters, return a list of a single
523 element `undecided'."
527 "Return a list of proper coding systems to encode characters of CHARSETS.
528 CHARSETS is a list of character sets.
530 This only finds coding systems of type `charset', whose
531 `:charset-list' property includes all of CHARSETS (plus `ascii' for
532 ASCII-compatible coding systems). It was used in older versions of
533 Emacs, but is unlikely to be what you really want now."
534 ;; Deal with aliases.
537 charsets))
558 t))
563 "Find multibyte characters in the region specified by FROM and TO.
564 If FROM is a string, find multibyte characters in the string.
565 The return value is an alist of the following format:
566 ((CHARSET COUNT CHAR ...) ...)
567 where
568 CHARSET is a character set,
569 COUNT is a number of characters,
570 CHARs are the characters found from the character set.
571 Optional 3rd arg MAXCOUNT limits how many CHARs are put in the above list.
572 Optional 4th arg EXCLUDES is a list of character sets to be ignored."
574 charset char)
609 "Search forward from point for a character that is not encodable.
610 It asks which coding system to check.
611 If such a character is found, set point after that character.
612 Otherwise, don't move point.
614 When called from a program, the value is the position of the unencodable
615 character found, or nil if all characters are encodable."
620 default))))
625 pos))
628 "Most recent coding system explicitly specified by the user when asked.
629 This variable is set whenever Emacs asks the user which coding system
630 to use in order to write a file. If you set it to nil explicitly,
631 then call `write-region', then afterward this variable will be non-nil
635 "If non-nil, a function to control the behavior of coding system selection.
636 The meaning is the same as the argument ACCEPT-DEFAULT-P of the
637 function `select-safe-coding-system' (which see). This variable
641 "Return a list of coding systems presumably more user-friendly than CODINGS."
642 ;; Change each safe coding system to the corresponding
643 ;; mime-charset name if it is also a coding system. Such a name
644 ;; is more friendly to users.
650 mime-charset cs)))
651 codings))
653 ;; Don't offer variations with locking shift, which you
654 ;; basically never want.
663 ;; Remove raw-text, emacs-mule and no-conversion unless nothing
664 ;; else is available.
668 codings))
672 "Select interactively a coding system for the region FROM ... TO.
673 FROM can be a string, as in `write-region'.
674 CODINGS is the list of base coding systems known to be safe for this region,
675 typically obtained with `find-coding-systems-region'.
676 UNSAFE is a list of coding systems known to be unsafe for this region.
677 REJECTED is a list of coding systems which were safe but for some reason
678 were not recommended in the particular context.
679 DEFAULT is the coding system to use by default in the query."
680 ;; At first, if some defaults are unsafe, record at most 11
681 ;; problematic characters and their positions for them by turning
682 ;; (CODING ...)
683 ;; into
684 ;; ((CODING (POS . CHAR) (POS . CHAR) ...) ...)
699 unsafe)))
705 coding-system)
707 ;; If some defaults are unsafe, make sure the offending
708 ;; buffer is displayed.
712 unsafe))))
713 ;; Then ask users to select one from CODINGS while showing
714 ;; the reason why none of the defaults are not used.
723 "These default coding systems were tried to encode"
738 but are not recommended for encoding text in this context,
774 'help-echo
775 "mouse-2, RET: jump to this character"
776 'help-function func1
779 "..."
782 'help-echo
783 "mouse-2, RET: next unencodable character"
784 'help-function func2
807 ;; Read a coding system.
811 default))
816 coding-system))
819 accept-default-p file)
820 "Ask a user to select a safe coding system from candidates.
821 The candidates of coding systems which can safely encode a text
822 between FROM and TO are shown in a popup window. Among them, the most
823 proper one is suggested as the default.
825 The list of `buffer-file-coding-system' of the current buffer, the
826 default `buffer-file-coding-system', and the most preferred coding
827 system (if it corresponds to a MIME charset) is treated as the
828 default coding system list. Among them, the first one that safely
829 encodes the text is normally selected silently and returned without
830 any user interaction. See also the command `prefer-coding-system'.
832 However, the user is queried if the chosen coding system is
833 inconsistent with what would be selected by `find-auto-coding' from
834 coding cookies &c. if the contents of the region were read from a
835 file. (That could lead to data corruption in a file subsequently
836 re-visited and edited.)
838 Optional 3rd arg DEFAULT-CODING-SYSTEM specifies a coding system or a
839 list of coding systems to be prepended to the default coding system
840 list. However, if DEFAULT-CODING-SYSTEM is a list and the first
841 element is t, the cdr part is used as the default coding system list,
842 i.e. current `buffer-file-coding-system', default `buffer-file-coding-system',
843 and the most preferred coding system are not used.
845 Optional 4th arg ACCEPT-DEFAULT-P, if non-nil, is a function to
846 determine the acceptability of the silently selected coding system.
847 It is called with that coding system, and should return nil if it
848 should not be silently selected and thus user interaction is required.
850 Optional 5th arg FILE is the file name to use for this purpose.
851 That is different from `buffer-file-name' when handling `write-region'
852 \(for example).
854 The variable `select-safe-coding-system-accept-default-p', if non-nil,
855 overrides ACCEPT-DEFAULT-P.
857 Kludgy feature: if FROM is a string, the string is the target text,
858 and TO is ignored."
863 auto-cs)
865 ;; Find an auto-coding that is specified for the current
866 ;; buffer and file from the region FROM and TO.
877 'mule
879 Invalid coding system `%s' is specified
880 for the current buffer/file by the %s.
881 It is highly recommended to fix it before writing to a file."
894 ;; Change elements of the list to (coding . base-coding).
897 default-coding-system))
900 ;; If the file has a coding cookie, use it regardless of any
901 ;; other setting.
908 ;; If buffer-file-coding-system is not nil nor undecided, append it
909 ;; to the defaults.
920 ;; If default buffer-file-coding-system is not nil nor undecided,
921 ;; append it to the defaults.
931 base)))))))
933 ;; If the most preferred coding system has the property mime-charset,
934 ;; append it to the defaults.
936 base)
948 ;; Decide the eol-type from the top of the default codings,
949 ;; current buffer-file-coding-system, or default buffer-file-coding-system.
955 buffer-file-coding-system)))
969 safe rejected unsafe)
971 ;; Any coding system is ok.
973 ;; Reverse the list so that elements are accumulated in safe,
974 ;; rejected, and unsafe in the correct order.
977 ;; Classify the defaults into safe, rejected, and unsafe.
980 ;; This is safe. Is it acceptable?
983 ;; No, not acceptable.
985 ;; Yes, acceptable.
987 ;; This is not safe.
989 ;; If there are safe ones, the first one is what we want.
993 ;; If all the defaults failed, ask a user.
998 ;; Check we're not inconsistent with what `coding:' spec &c would
999 ;; give when file is re-read.
1000 ;; But don't do this if we explicitly ignored the cookie
1001 ;; by using `find-file-literally'.
1004 coding-system
1006 ;; Merge coding-system and auto-cs as far as possible.
1015 coding-system auto-cs))
1018 auto-cs coding-system))))
1022 coding-system eol-type-2)))
1025 auto-cs eol-type-1)))))))
1028 ;; Don't barf if writing a compressed file, say.
1029 ;; This check perhaps isn't ideal, but is probably
1030 ;; the best thing to do.
1035 %s specified by file contents. Really save (else edit coding cookies \
1048 coding-system)))
1053 "Return a coding system to encode the outgoing message of the current buffer.
1054 It at first tries the first coding system found in these variables
1055 in this order:
1056 (1) local value of `buffer-file-coding-system'
1057 (2) value of `sendmail-coding-system'
1058 (3) value of `default-sendmail-coding-system'
1059 (4) default value of `buffer-file-coding-system'
1060 If the found coding system can't encode the current buffer,
1061 or none of them are bound to a coding system,
1062 it asks the user to select a proper coding system."
1064 buffer-file-coding-system)
1065 sendmail-coding-system
1066 default-sendmail-coding-system
1069 ;; We should never use no-conversion for outgoing mail.
1075 coding)))
1076 \f
1077 ;;; Language support stuff.
1080 "Alist of language environment definitions.
1081 Each element looks like:
1082 (LANGUAGE-NAME . ((KEY . INFO) ...))
1083 where LANGUAGE-NAME is a string, the name of the language environment,
1084 KEY is a symbol denoting the kind of information, and
1085 INFO is the data associated with KEY.
1086 Meaningful values for KEY include
1088 documentation value is documentation of what this language environment
1089 is meant for, and how to use it.
1090 charset value is a list of the character sets mainly used
1091 by this language environment.
1092 sample-text value is an expression which is evalled to generate
1093 a line of text written using characters appropriate
1094 for this language environment.
1095 setup-function value is a function to call to switch to this
1096 language environment.
1097 exit-function value is a function to call to leave this
1098 language environment.
1099 coding-system value is a list of coding systems that are good for
1100 saving text written in this language environment.
1101 This list serves as suggestions to the user;
1102 in effect, as a kind of documentation.
1103 coding-priority value is a list of coding systems for this language
1104 environment, in order of decreasing priority.
1105 This is used to set up the coding system priority
1106 list when you switch to this language environment.
1107 nonascii-translation
1108 value is a charset of dimension one to use for
1109 converting a unibyte character to multibyte
1110 and vice versa.
1111 input-method value is a default input method for this language
1112 environment.
1113 features value is a list of features requested in this
1114 language environment.
1115 ctext-non-standard-encodings
1116 value is a list of non-standard encoding names used
1117 in extended segments of CTEXT. See the variable
1118 `ctext-non-standard-encodings' for more detail.
1120 The following key takes effect only when multibyte characters are
1121 globally disabled, i.e. the default value of `enable-multibyte-characters'
1122 is nil (which is an obsolete and deprecated use):
1124 unibyte-display value is a coding system to encode characters for
1125 the terminal. Characters in the range of 160 to
1126 255 display not as octal escapes, but as non-ASCII
1130 "Return information listed under KEY for language environment LANG-ENV.
1131 KEY is a symbol denoting the kind of information.
1132 For a list of useful values for KEY and their meanings,
1133 see `language-info-alist'."
1141 "Modify part of the definition of language environment LANG-ENV.
1142 Specifically, this stores the information INFO under KEY
1143 in the definition of this language environment.
1144 KEY is a symbol denoting the kind of information.
1145 INFO is the value for that information.
1147 For a list of useful values for KEY and their meanings,
1148 see `language-info-alist'."
1167 "Internal use only.
1168 Arguments are the same as `set-language-info'."
1180 ;; Update the custom-type of `current-language-environment'.
1188 "Store ALIST as the definition of language environment LANG-ENV.
1189 ALIST is an alist of KEY and INFO values. See the documentation of
1190 `language-info-alist' for the meanings of KEY and INFO.
1192 Optional arg PARENTS is a list of parent menu names; it specifies
1193 where to put this language environment in the
1194 Describe Language Environment and Set Language Environment menus.
1196 in the European submenu in each of those two menus."
1205 map parent-symbol parent prompt)
1211 ;; This prompt string is for define-prefix-command, so
1212 ;; that the map it creates will be suitable for a menu.
1233 ;; Set up menu items for this language env.
1248 "Read a language environment name which has information for KEY.
1249 If KEY is nil, read any language environment.
1250 Prompt with PROMPT. DEFAULT is the default choice of language environment.
1251 This returns a language environment name as a string."
1254 language-info-alist
1257 t nil nil default)))
1261 name)))
1262 \f
1263 ;;; Multilingual input methods.
1265 "LEIM: Libraries of Emacs Input Methods."
1269 "Name of LEIM list file.
1270 This file contains a list of libraries of Emacs input methods (LEIM)
1271 in the format of Lisp expression for registering each input method.
1275 ";;; %s -- list of LEIM (Library of Emacs Input Method) -*-coding: utf-8;-*-
1276 ;;
1277 ;; This file is automatically generated.
1278 ;;
1279 ;; This file contains a list of LEIM (Library of Emacs Input Method)
1280 ;; methods in the same directory as this file. Loading this file
1281 ;; registers all the input methods in Emacs.
1282 ;;
1283 ;; Each entry has the form:
1284 ;; (register-input-method
1285 ;; INPUT-METHOD LANGUAGE-NAME ACTIVATE-FUNC
1286 ;; TITLE DESCRIPTION
1287 ;; ARG ...)
1288 ;; See the function `register-input-method' for the meanings of the arguments.
1289 ;;
1290 ;; If this directory is included in `load-path', Emacs automatically
1291 ;; loads this file at startup time.
1293 "
1294 leim-list-file-name)
1298 "Regexp matching head of each entry in LEIM list file.
1303 "List of functions to call to update LEIM list file.
1307 "Update LEIM list file in directories DIRS."
1312 "The current input method for multilingual text.
1323 "String widget with completion for input method."
1324 :completions
1331 "Default input method for multilingual text (a string).
1332 This is the input method activated automatically by the command
1337 mule-input-method-string)
1343 "History list of input methods read from the minibuffer.
1345 Maximum length of the history list is determined by the value
1351 'inactivate-current-input-method-function
1354 "Function to call for deactivating the current input method.
1355 Every input method should set this to an appropriate value when activated.
1356 This function is called with no argument.
1358 This function should never change the value of `current-input-method'.
1364 "Function to call for describing the current input method.
1370 "Alist of input method names vs how to use them.
1371 Each element has the form:
1372 (INPUT-METHOD LANGUAGE-ENV ACTIVATE-FUNC TITLE DESCRIPTION ARGS...)
1374 ;;;###autoload
1378 "Register INPUT-METHOD as an input method for language environment LANG-ENV.
1380 INPUT-METHOD and LANG-ENV are symbols or strings.
1381 ACTIVATE-FUNC is a function to call to activate this method.
1382 TITLE is a string to show in the mode line when this method is active.
1383 DESCRIPTION is a string describing this method and what it is good for.
1384 The ARGS, if any, are passed as arguments to ACTIVATE-FUNC.
1385 All told, the arguments to ACTIVATE-FUNC are INPUT-METHOD and the ARGS.
1388 created at Emacs build time, registering all Quail input methods
1389 contained in the Emacs distribution.
1391 In case you want to register a new Quail input method by yourself, be
1392 careful to use the same input method title as given in the third
1393 parameter of `quail-define-package'. (If the values are different, the
1394 string specified in this function takes precedence.)
1396 The commands `describe-input-method' and `list-input-methods' need
1397 these duplicated values to show some information about input methods
1398 without loading the relevant Quail packages.
1415 "Read a name of input method from a minibuffer prompting with PROMPT.
1416 If DEFAULT is non-nil, use that as the default,
1417 and substitute it into PROMPT at the first `%s'.
1418 If INHIBIT-NULL is non-nil, null input signals an error.
1420 The return value is a string."
1424 ;; As it is quite normal to change input method in the
1425 ;; minibuffer, we must enable it even if
1426 ;; enable-recursive-minibuffers is currently nil.
1428 ;; This binding is necessary because input-method-history is
1429 ;; buffer local.
1431 nil t nil 'input-method-history
1434 default))))
1438 input-method
1443 "Switch to input method INPUT-METHOD for the current buffer.
1444 If some other input method is already active, turn it off first.
1445 If INPUT-METHOD is nil, deactivate any current input method."
1472 "Turn off the current input method."
1483 current-input-method-title nil)
1491 'inactivate-input-method
1495 "Select and activate input method INPUT-METHOD for the current buffer.
1496 This also sets the default input method to the one you specify.
1497 If INPUT-METHOD is nil, this function turns off the input method, and
1498 also causes you to be prompted for a name of an input method the next
1500 When called interactively, the optional arg INTERACTIVE is non-nil,
1501 which marks the variable `default-input-method' as set for Custom buffers.
1504 To deactivate it programmatically, use `deactivate-input-method'."
1509 default t)
1510 t)))
1515 default-input-method)
1521 "Enable or disable multilingual text input method for the current buffer.
1522 Only one input method can be enabled at any time in a given buffer.
1524 The normal action is to enable an input method if none was enabled,
1525 and disable the current one otherwise. Which input method to enable
1526 can be determined in various ways--either the one most recently used,
1527 or the one specified by `default-input-method', or as a last resort
1528 by reading the name of an input method in the minibuffer.
1530 With a prefix argument ARG, read an input method name with the minibuffer
1531 and enable that one. The default is the most recent input method specified
1532 \(not including the currently active input method, if any).
1534 When called interactively, the optional argument INTERACTIVE is non-nil,
1535 which marks the variable `default-input-method' as set for Custom buffers."
1552 default t))
1553 default))
1563 "Describe input method INPUT-METHOD."
1593 "Describe the input method currently in use.
1594 This is a subroutine for `describe-input-method'."
1600 current-input-method)
1605 "Read a multilingual string from minibuffer, prompting with string PROMPT.
1606 The input method selected last time is activated in minibuffer.
1607 If optional second argument INITIAL-INPUT is non-nil, insert it in the
1608 minibuffer initially.
1609 Optional 3rd argument INPUT-METHOD specifies the input method to be activated
1610 instead of the one selected last time. It is a symbol or a string."
1613 current-input-method
1614 default-input-method
1625 ;; Variables to control behavior of input methods. All input methods
1626 ;; should react to these variables.
1629 "A flag to control extra guidance given by input methods.
1630 The value should be nil, t, `complex-only', or `default'.
1632 The extra guidance is done by showing list of available keys in echo
1633 area. When you use the input method in the minibuffer, the guidance
1634 is shown at the bottom short window (split from the existing window).
1636 If the value is t, extra guidance is always given, if the value is
1637 nil, extra guidance is always suppressed.
1639 If the value is `complex-only', only complex input methods such as
1640 `chinese-py' and `japanese' give extra guidance.
1642 If the value is `default', complex input methods always give extra
1643 guidance, but simple input methods give it only when you are not in
1644 the minibuffer.
1646 See also the variable `input-method-highlight-flag'."
1652 "If this flag is non-nil, input methods highlight partially-entered text.
1653 For instance, while you are in the middle of a Quail input method sequence,
1654 the text inserted so far is temporarily underlined.
1655 The underlining goes away when you finish or abort the input method sequence.
1656 See also the variable `input-method-verbose-flag'."
1661 "Normal hook run just after an input method is activated.
1663 The variable `current-input-method' keeps the input method name
1664 just activated."
1669 'input-method-inactivate-hook
1673 "Normal hook run just after an input method is deactivated.
1675 The variable `current-input-method' still keeps the input method name
1676 just deactivated."
1682 "Normal hook run just after an input method insert some chunk of text."
1687 "This flag controls when an input method returns.
1688 Usually, the input method does not return while there's a possibility
1689 that it may find a different translation if a user types another key.
1690 But, if this flag is non-nil, the input method returns as soon as the
1694 "This flag controls how an input method shows an intermediate key sequence.
1695 Usually, the input method inserts the intermediate key sequence,
1696 or candidate translations corresponding to the sequence,
1697 at point in the current buffer.
1698 But, if this flag is non-nil, it displays them in echo area instead."
1703 "This flag controls the behavior of an input method on invalid key input.
1704 Usually, when a user types a key which doesn't start any character
1705 handled by the input method, the key is handled by turning off the
1706 input method temporarily. After that key, the input method is re-enabled.
1709 \f
1711 "Normal hook run after some language environment is set.
1713 When you set some hook function here, that effect usually should not
1714 be inherited to another language environment. So, you had better set
1715 another function in `exit-language-environment-hook' (which see) to
1716 cancel the effect."
1721 "Normal hook run after exiting from some language environment.
1722 When this hook is run, the variable `current-language-environment'
1723 is still bound to the language environment being exited.
1725 This hook is mainly used for canceling the effect of
1726 `set-language-environment-hook' (which see)."
1733 "Switch to a specified language environment."
1746 "The last language environment specified with `set-language-environment'.
1748 to using the function `set-language-environment'."
1755 current-language-environment)
1756 language-info-alist t))
1758 ;; custom type will be updated with `set-language-info'.
1769 "Reset multilingual environment of Emacs to the default status.
1771 The default status is as follows:
1773 The default value of `buffer-file-coding-system' is nil.
1774 The default coding system for process I/O is nil.
1775 The default value for the command `set-terminal-coding-system' is nil.
1776 The default value for the command `set-keyboard-coding-system' is nil.
1778 The order of priorities of coding systems are as follows:
1779 utf-8
1780 iso-2022-7bit
1781 iso-latin-1
1782 iso-2022-7bit-lock
1783 iso-2022-8bit-ss2
1784 emacs-mule
1785 raw-text"
1787 ;; This function formerly set default-enable-multibyte-characters to t,
1788 ;; but that is incorrect. It should not alter the unibyte/multibyte choice.
1791 'utf-8
1792 'iso-2022-7bit
1793 'iso-latin-1
1794 'iso-2022-7bit-lock
1795 'iso-2022-8bit-ss2
1796 'emacs-mule
1801 ;; On Darwin systems, this should be utf-8-unix, but when this file is loaded
1802 ;; that is not yet defined, so we set it in set-locale-environment instead.
1804 ;; Preserve eol-type from existing default-process-coding-systems.
1805 ;; On non-unix-like systems in particular, these may have been set
1806 ;; carefully by the user, or by the startup code, to deal with the
1807 ;; users shell appropriately, so should not be altered by changing
1808 ;; language environment.
1810 ;; When bootstrapping, coding-systems are not defined yet, so
1811 ;; we need to catch the error from check-coding-system.
1824 ;; Put the highest priority to the charset iso-8859-1 to prefer the
1825 ;; registry iso8859-1 over iso8859-2 in font selection. It also
1826 ;; makes unibyte-display-via-language-environment to use iso-8859-1
1827 ;; as the unibyte charset.
1830 ;; Don't alter the terminal and keyboard coding systems here.
1831 ;; The terminal still supports the same coding system
1832 ;; that it supported a minute ago.
1833 ;; (set-terminal-coding-system-internal nil)
1834 ;; (set-keyboard-coding-system-internal nil)
1836 ;; Back in Emacs-20, it was necessary to provide some fallback implicit
1837 ;; conversion, because almost no packages handled coding-system issues.
1838 ;; Nowadays it'd just paper over bugs.
1839 ;; (set-unibyte-charset 'iso-8859-1)
1840 )
1844 (defun set-display-table-and-terminal-coding-system (language-name &optional coding-system display)
1845 "Set up the display table and terminal coding system for LANGUAGE-NAME."
1851 ;; The following 2 lines undo the 8-bit display that we set up
1852 ;; in standard-display-european-internal, which see. This is in
1853 ;; case the user has used standard-display-european earlier in
1854 ;; this session.
1861 "Set up multilingual environment for using LANGUAGE-NAME.
1862 This sets the coding system priority and the default input method
1863 and sometimes other things. LANGUAGE-NAME should be a string
1865 specifies the character set for the major languages of Western Europe.
1867 If there is a prior value for `current-language-environment', this
1868 runs the hook `exit-language-environment-hook'. After setting up
1869 the new language environment, it runs `set-language-environment-hook'."
1871 nil
1888 ;; The features might set up coding systems.
1900 ;; Unibyte setups if necessary.
1910 current-iso639-language))
1916 "An Emacs charset."
1918 :completions
1928 widget))
1932 "Customizations of language environment parameters.
1933 Value is an alist with elements like those of `language-info-alist'.
1934 These are used to set values in `language-info-alist' which replace
1935 the defaults. A typical use is replacing the default input method for
1938 This option is intended for use at startup. Removing items doesn't
1939 remove them from the language info until you next restart Emacs.
1941 Setting this variable directly does not take effect.
1942 See `set-language-info-alist' for use in programs."
1947 ;; Can't do this before language environments are set up.
1949 ;; modify language-info-alist
1952 ;; re-set the environment in case its parameters changed
1956 :completions
1960 action language-info-alist string pred))))
1961 :value-type
1979 ;; Actually set up direct output of non-ASCII characters.
1981 ;; Unibyte Emacs on MS-DOS wants to display all 8-bit characters with
1982 ;; the native font, and codes 160 and 146 stand for something very
1983 ;; different there.
1987 ;; Most X fonts used to do the wrong thing for latin-1 code 160.
1989 ;; XFree86 4 has fixed the fonts.
1993 ;; Make non-line-break space display as a plain space.
1996 ;; don't contain a character at that position. Map it to the ASCII
1997 ;; apostrophe. [This is actually RIGHT SINGLE QUOTATION MARK,
1998 ;; U+2019, normally from the windows-1252 character set. XFree 4
1999 ;; fonts probably have the appropriate glyph at this position,
2000 ;; so they could use standard-display-8bit. It's better to use a
2001 ;; proper windows-1252 coding system. --fx]
2002 (aset standard-display-table (unibyte-char-to-multibyte 146) [39]))))
2004 (defun set-language-environment-coding-systems (language-name)
2005 "Do various coding system setups for language environment LANGUAGE-NAME."
2006 (let* ((priority (get-language-info language-name 'coding-priority))
2007 (default-coding (car priority))
2008 ;; If the default buffer-file-coding-system is nil, don't use
2009 ;; coding-system-eol-type, because it treats nil as
2010 ;; `no-conversion'. The default buffer-file-coding-system is set
2011 ;; to nil by reset-language-environment, and in that case we
2012 ;; want to have here the native EOL type for each platform.
2013 ;; FIXME: there should be a common code that runs both on
2014 ;; startup and here to set the default EOL type correctly.
2015 ;; Right now, DOS/Windows platforms set this on dos-w32.el,
2016 ;; which works only as long as the order of loading files at
2017 ;; dump time and calling functions at startup is not modified
2018 ;; significantly, i.e. as long as this function is called
2019 ;; _after_ the default buffer-file-coding-system was set by
2020 ;; dos-w32.el.
2021 (eol-type
2022 (coding-system-eol-type
2023 (or (default-value 'buffer-file-coding-system)
2024 (if (memq system-type '(windows-nt ms-dos)) 'dos 'unix)))))
2025 (when priority
2026 (set-default-coding-systems
2027 (if (memq eol-type '(0 1 2 unix dos mac))
2028 (coding-system-change-eol-conversion default-coding eol-type)
2029 default-coding))
2030 (setq default-sendmail-coding-system default-coding)
2031 (apply 'set-coding-system-priority priority))))
2033 (defun set-language-environment-input-method (language-name)
2034 "Do various input method setups for language environment LANGUAGE-NAME."
2035 (let ((input-method (get-language-info language-name 'input-method)))
2036 (when input-method
2037 (setq default-input-method input-method)
2038 (if input-method-history
2039 (setq input-method-history
2040 (cons input-method
2041 (delete input-method input-method-history)))))))
2043 (defun set-language-environment-nonascii-translation (language-name)
2044 "Do unibyte/multibyte translation setup for language environment LANGUAGE-NAME."
2045 ;; Note: For DOS, we assumed that the charset cpXXX is already
2046 ;; defined.
2047 (let ((nonascii (get-language-info language-name 'nonascii-translation)))
2048 (if (eq window-system 'pc)
2049 (setq nonascii (intern (format "cp%d" dos-codepage))))
2050 (or (and (charsetp nonascii)
2051 (get-charset-property nonascii :ascii-compatible-p))
2052 (setq nonascii 'iso-8859-1))
2053 ;; Back in Emacs-20, it was necessary to provide some fallback implicit
2054 ;; conversion, because almost no packages handled coding-system issues.
2055 ;; Nowadays it'd just paper over bugs.
2056 ;; (set-unibyte-charset nonascii)
2057 ))
2059 (defun set-language-environment-charset (language-name)
2060 "Do various charset setups for language environment LANGUAGE-NAME."
2061 ;; Put higher priorities to such charsets that are supported by the
2062 ;; coding systems of higher priorities in this environment.
2063 (let ((charsets (get-language-info language-name 'charset)))
2064 (dolist (coding (get-language-info language-name 'coding-priority))
2065 (let ((list (coding-system-charset-list coding)))
2066 (if (consp list)
2067 (setq charsets (append charsets list)))))
2068 (if charsets
2069 (apply 'set-charset-priority charsets))))
2071 (defun set-language-environment-unibyte (language-name)
2072 "Do various unibyte-mode setups for language environment LANGUAGE-NAME."
2073 (set-display-table-and-terminal-coding-system language-name))
2075 (defun princ-list (&rest args)
2077 (declare (obsolete "use mapc and princ instead." "23.3"))
2078 (mapc #'princ args)
2081 (put 'describe-specified-language-support 'apropos-inhibit t)
2083 ;; Print language-specific information such as input methods,
2084 ;; charsets, and coding systems. This function is intended to be
2085 ;; called from the menu:
2086 ;; [menu-bar mule describe-language-environment LANGUAGE]
2087 ;; and should not run it by `M-x describe-current-input-method-function'.
2088 (defun describe-specified-language-support ()
2089 "Describe how Emacs supports the specified language environment."
2090 (interactive)
2091 (let (language-name)
2092 (if (not (and (symbolp last-command-event)
2093 (or (not (eq last-command-event 'Default))
2094 (setq last-command-event 'English))
2095 (setq language-name (symbol-name last-command-event))))
2096 (error "This command should only be called from the menu bar"))
2097 (describe-language-environment language-name)))
2099 (defun describe-language-environment (language-name)
2100 "Describe how Emacs supports language environment LANGUAGE-NAME."
2101 (interactive
2102 (list (read-language-name
2103 'documentation
2104 "Describe language environment (default current choice): ")))
2105 (if (null language-name)
2106 (setq language-name current-language-environment))
2107 (if (or (null language-name)
2108 (null (get-language-info language-name 'documentation)))
2109 (error "No documentation for the specified language"))
2110 (if (symbolp language-name)
2111 (setq language-name (symbol-name language-name)))
2112 (dolist (feature (get-language-info language-name 'features))
2113 (require feature))
2114 (let ((doc (get-language-info language-name 'documentation)))
2115 (help-setup-xref (list #'describe-language-environment language-name)
2116 (called-interactively-p 'interactive))
2117 (with-output-to-temp-buffer (help-buffer)
2118 (with-current-buffer standard-output
2120 (if (stringp doc)
2122 (condition-case nil
2123 (let ((str (eval (get-language-info language-name 'sample-text))))
2124 (if (stringp str)
2128 (error nil))
2129 (let ((input-method (get-language-info language-name 'input-method))
2130 (l (copy-sequence input-method-alist))
2131 (first t))
2132 (when (and input-method
2133 (setq input-method (assoc input-method l)))
2135 (setq l (cons input-method (delete input-method l))
2136 first nil))
2137 (dolist (elt l)
2138 (when (or (eq input-method elt)
2139 (eq t (compare-strings language-name nil nil
2140 (nth 1 elt) nil nil t)))
2141 (when first
2143 (setq first nil))
2144 (insert " " (car elt))
2145 (search-backward (car elt))
2146 (help-xref-button 0 'help-input-method (car elt))
2147 (goto-char (point-max))
2149 (if (stringp (nth 3 elt)) (nth 3 elt) (car (nth 3 elt)))
2151 (or first
2154 (let ((l (get-language-info language-name 'charset)))
2155 (if (null l)
2157 (while l
2158 (insert " " (symbol-name (car l)))
2159 (search-backward (symbol-name (car l)))
2160 (help-xref-button 0 'help-character-set (car l))
2161 (goto-char (point-max))
2163 (setq l (cdr l)))))
2166 (let ((l (get-language-info language-name 'coding-system)))
2167 (if (null l)
2169 (while l
2170 (insert " " (symbol-name (car l)))
2171 (search-backward (symbol-name (car l)))
2172 (help-xref-button 0 'help-coding-system (car l))
2173 (goto-char (point-max))
2174 (insert (substitute-command-keys " (`")
2175 (coding-system-mnemonic (car l))
2177 (substitute-command-keys
2178 (coding-system-doc-string (car l)))
2180 (let ((aliases (coding-system-aliases (car l))))
2181 (when aliases
2183 (while aliases
2184 (insert " " (symbol-name (car aliases)))
2185 (setq aliases (cdr aliases)))
2187 (setq l (cdr l)))))))))
2188 \f
2189 ;;; Locales.
2191 (defvar locale-translation-file-name nil
2192 "File name for the system's file of locale-name aliases, or nil if none.")
2194 ;; The following definitions might as well be marked as constants and
2195 ;; purecopied, since they're normally used on startup, and probably
2196 ;; should reflect the facilities of the base Emacs.
2197 (defconst locale-language-names
2198 (purecopy
2199 '(
2200 ;; Locale names of the form LANGUAGE[_TERRITORY][.CODESET][@MODIFIER]
2201 ;; as specified in the Single Unix Spec, Version 2.
2202 ;; LANGUAGE is a language code taken from ISO 639:1988 (E/F)
2203 ;; with additions from ISO 639/RA Newsletter No.1/1989;
2204 ;; see Internet RFC 2165 (1997-06) and
2205 ;; http://www.evertype.com/standards/iso639/iso639-en.html
2206 ;; TERRITORY is a country code taken from ISO 3166
2207 ;; http://www.din.de/gremien/nas/nabd/iso3166ma/codlstp1/en_listp1.html.
2208 ;; CODESET and MODIFIER are implementation-dependent.
2210 ;; jasonr comments: MS Windows uses three letter codes for
2211 ;; languages instead of the two letter ISO codes that POSIX
2212 ;; uses. In most cases the first two letters are the same, so
2213 ;; most of the regexps in locale-language-names work. Japanese
2214 ;; and Chinese are exceptions, which are listed in the
2215 ;; non-standard section at the bottom of locale-language-names.
2217 ("aa_DJ" . "Latin-1") ; Afar
2218 ("aa" . "UTF-8")
2219 ;; ab Abkhazian
2220 ("af" . "Latin-1") ; Afrikaans
2221 ("am" "Ethiopic" utf-8) ; Amharic
2222 ("an" . "Latin-9") ; Aragonese
2223 ("ar" . "Arabic")
2224 ; as Assamese
2225 ; ay Aymara
2226 ("az" . "UTF-8") ; Azerbaijani
2227 ; ba Bashkir
2228 ("be" "Belarusian" cp1251) ; Belarusian [Byelorussian until early 1990s]
2229 ("bg" "Bulgarian" cp1251) ; Bulgarian
2230 ; bh Bihari
2231 ; bi Bislama
2232 ("bn" . "UTF-8") ; Bengali, Bangla
2233 ("bo" . "Tibetan")
2234 ("br" . "Latin-1") ; Breton
2235 ("bs" . "Latin-2") ; Bosnian
2236 ("byn" . "UTF-8") ; Bilin; Blin
2237 ("ca" "Catalan" iso-8859-1) ; Catalan
2238 ; co Corsican
2239 ("cs" "Czech" iso-8859-2)
2240 ("cy" "Welsh" iso-8859-14)
2241 ("da" . "Latin-1") ; Danish
2242 ("de" "German" iso-8859-1)
2243 ; dv Divehi
2244 ; dz Bhutani
2245 ("el" "Greek" iso-8859-7)
2246 ;; Users who specify "en" explicitly typically want Latin-1, not ASCII.
2247 ;; That's actually what the GNU locales define, modulo things like
2248 ;; en_IN -- fx.
2249 ("en_IN" "English" utf-8) ; glibc uses utf-8 for English in India
2250 ("en" "English" iso-8859-1) ; English
2251 ("eo" . "Esperanto") ; Esperanto
2252 ("es" "Spanish" iso-8859-1)
2253 ("et" . "Latin-1") ; Estonian
2254 ("eu" . "Latin-1") ; Basque
2255 ("fa" . "UTF-8") ; Persian
2256 ("fi" . "Latin-1") ; Finnish
2257 ("fj" . "Latin-1") ; Fiji
2258 ("fo" . "Latin-1") ; Faroese
2259 ("fr" "French" iso-8859-1) ; French
2260 ("fy" . "Latin-1") ; Frisian
2261 ("ga" . "Latin-1") ; Irish Gaelic (new orthography)
2262 ("gd" . "Latin-9") ; Scots Gaelic
2263 ("gez" "Ethiopic" utf-8) ; Geez
2264 ("gl" . "Latin-1") ; Gallegan; Galician
2265 ; gn Guarani
2266 ("gu" . "UTF-8") ; Gujarati
2267 ("gv" . "Latin-1") ; Manx Gaelic
2268 ; ha Hausa
2269 ("he" "Hebrew" iso-8859-8)
2270 ("hi" "Devanagari" utf-8) ; Hindi
2271 ("hr" "Croatian" iso-8859-2) ; Croatian
2272 ("hu" . "Latin-2") ; Hungarian
2273 ; hy Armenian
2274 ; ia Interlingua
2275 ("id" . "Latin-1") ; Indonesian
2276 ; ie Interlingue
2277 ; ik Inupiak
2278 ("is" . "Latin-1") ; Icelandic
2279 ("it" "Italian" iso-8859-1) ; Italian
2280 ; iu Inuktitut
2281 ("iw" "Hebrew" iso-8859-8)
2282 ("ja" "Japanese" euc-jp)
2283 ; jw Javanese
2284 ("ka" "Georgian" georgian-ps) ; Georgian
2285 ; kk Kazakh
2286 ("kl" . "Latin-1") ; Greenlandic
2287 ; km Cambodian
2288 ("kn" "Kannada" utf-8)
2289 ("ko" "Korean" euc-kr)
2290 ; ks Kashmiri
2291 ; ku Kurdish
2292 ("kw" . "Latin-1") ; Cornish
2293 ; ky Kirghiz
2294 ("la" . "Latin-1") ; Latin
2295 ("lb" . "Latin-1") ; Luxemburgish
2296 ("lg" . "Laint-6") ; Ganda
2297 ; ln Lingala
2298 ("lo" "Lao" utf-8) ; Laothian
2299 ("lt" "Lithuanian" iso-8859-13)
2300 ("lv" . "Latvian") ; Latvian, Lettish
2301 ; mg Malagasy
2302 ("mi" . "Latin-7") ; Maori
2303 ("mk" "Cyrillic-ISO" iso-8859-5) ; Macedonian
2304 ("ml" "Malayalam" utf-8)
2305 ("mn" . "UTF-8") ; Mongolian
2306 ; mo Moldavian
2307 ("mr" "Devanagari" utf-8) ; Marathi
2308 ("ms" . "Latin-1") ; Malay
2309 ("mt" . "Latin-3") ; Maltese
2310 ; my Burmese
2311 ; na Nauru
2312 ("nb" . "Latin-1") ; Norwegian
2313 ("ne" "Devanagari" utf-8) ; Nepali
2314 ("nl" "Dutch" iso-8859-1)
2315 ("no" . "Latin-1") ; Norwegian
2316 ("oc" . "Latin-1") ; Occitan
2317 ("om_ET" . "UTF-8") ; (Afan) Oromo
2318 ("om" . "Latin-1") ; (Afan) Oromo
2319 ; or Oriya
2320 ("pa" . "UTF-8") ; Punjabi
2321 ("pl" . "Latin-2") ; Polish
2322 ; ps Pashto, Pushto
2323 ("pt" . "Latin-1") ; Portuguese
2324 ; qu Quechua
2325 ("rm" . "Latin-1") ; Rhaeto-Romanic
2326 ; rn Kirundi
2327 ("ro" "Romanian" iso-8859-2)
2328 ("ru_RU" "Russian" iso-8859-5)
2329 ("ru_UA" "Russian" koi8-u)
2330 ; rw Kinyarwanda
2331 ("sa" . "Devanagari") ; Sanskrit
2332 ; sd Sindhi
2333 ("se" . "UTF-8") ; Northern Sami
2334 ; sg Sangho
2335 ("sh" . "Latin-2") ; Serbo-Croatian
2336 ; si Sinhalese
2337 ("sid" . "UTF-8") ; Sidamo
2338 ("sk" "Slovak" iso-8859-2)
2339 ("sl" "Slovenian" iso-8859-2)
2340 ; sm Samoan
2341 ; sn Shona
2342 ("so_ET" "UTF-8") ; Somali
2343 ("so" "Latin-1") ; Somali
2344 ("sq" . "Latin-1") ; Albanian
2345 ("sr" . "Latin-2") ; Serbian (Latin alphabet)
2346 ; ss Siswati
2347 ("st" . "Latin-1") ; Sesotho
2348 ; su Sundanese
2349 ("sv" "Swedish" iso-8859-1) ; Swedish
2350 ("sw" . "Latin-1") ; Swahili
2351 ("ta" "Tamil" utf-8)
2352 ("te" . "UTF-8") ; Telugu
2353 ("tg" "Tajik" koi8-t)
2354 ("th" "Thai" tis-620)
2355 ("ti" "Ethiopic" utf-8) ; Tigrinya
2356 ("tig_ER" . "UTF-8") ; Tigre
2357 ; tk Turkmen
2358 ("tl" . "Latin-1") ; Tagalog
2359 ; tn Setswana
2360 ; to Tonga
2361 ("tr" "Turkish" iso-8859-9)
2362 ; ts Tsonga
2363 ("tt" . "UTF-8") ; Tatar
2364 ; tw Twi
2365 ; ug Uighur
2366 ("uk" "Ukrainian" koi8-u)
2367 ("ur" . "UTF-8") ; Urdu
2368 ("uz_UZ@cyrillic" . "UTF-8"); Uzbek
2369 ("uz" . "Latin-1") ; Uzbek
2370 ("vi" "Vietnamese" utf-8)
2371 ; vo Volapuk
2372 ("wa" . "Latin-1") ; Walloon
2373 ; wo Wolof
2374 ("xh" . "Latin-1") ; Xhosa
2375 ("yi" . "Windows-1255") ; Yiddish
2376 ; yo Yoruba
2377 ; za Zhuang
2378 ("zh_HK" . "Chinese-Big5")
2379 ; zh_HK/BIG5-HKSCS \
2380 ("zh_TW" . "Chinese-Big5")
2381 ("zh_CN.GB2312" "Chinese-GB")
2382 ("zh_CN.GBK" "Chinese-GBK")
2383 ("zh_CN.GB18030" "Chinese-GB18030")
2384 ("zh_CN.UTF-8" . "Chinese-GBK")
2385 ("zh_CN" . "Chinese-GB")
2386 ("zh" . "Chinese-GB")
2387 ("zu" . "Latin-1") ; Zulu
2389 ;; ISO standard locales
2390 ("c$" . "ASCII")
2391 ("posix$" . "ASCII")
2393 ;; The "IPA" Emacs language environment does not correspond
2394 ;; to any ISO 639 code, so let it stand for itself.
2395 ("ipa$" . "IPA")
2397 ;; Nonstandard or obsolete language codes
2398 ("cz" . "Czech") ; e.g. Solaris 2.6
2399 ("ee" . "Latin-4") ; Estonian, e.g. X11R6.4
2400 ("iw" . "Hebrew") ; e.g. X11R6.4
2401 ("sp" . "Cyrillic-ISO") ; Serbian (Cyrillic alphabet), e.g. X11R6.4
2402 ("su" . "Latin-1") ; Finnish, e.g. Solaris 2.6
2403 ("jp" . "Japanese") ; e.g. MS Windows
2404 ("chs" . "Chinese-GBK") ; MS Windows Chinese Simplified
2405 ("cht" . "Chinese-BIG5") ; MS Windows Chinese Traditional
2406 ("gbz" . "UTF-8") ; MS Windows Dari Persian
2407 ("div" . "UTF-8") ; MS Windows Divehi (Maldives)
2408 ("wee" . "Latin-2") ; MS Windows Lower Sorbian
2409 ("wen" . "Latin-2") ; MS Windows Upper Sorbian
2410 ))
2411 "Alist of locale regexps vs the corresponding languages and coding systems.
2412 Each element has this form:
2413 (LOCALE-REGEXP LANG-ENV CODING-SYSTEM)
2414 The first element whose LOCALE-REGEXP matches the start of a
2415 downcased locale specifies the LANG-ENV \(language environment)
2416 and CODING-SYSTEM corresponding to that locale. If there is no
2417 appropriate language environment, the element may have this form:
2418 (LOCALE-REGEXP . LANG-ENV)
2419 In this case, LANG-ENV is one of generic language environments for an
2422 (defconst locale-charset-language-names
2423 (purecopy
2432 ;; utf-8@euro exists, so put this last. (@euro really specifies
2433 ;; the currency, rather than the charset.)
2435 "List of pairs of locale regexps and charset language names.
2436 The first element whose locale regexp matches the start of a downcased locale
2437 specifies the language name whose charset corresponds to that locale.
2438 This language name is used if the locale is not listed in
2439 `locale-language-names'.")
2441 (defconst locale-preferred-coding-systems
2442 (purecopy
2451 ;; utf-8@euro exists, so put this after utf-8. (@euro really
2452 ;; specifies the currency, rather than the charset.)
2453 (".*@euro" . iso-8859-15)
2454 ("koi8-?r" . koi8-r)
2455 ("koi8-?u" . koi8-u)
2456 ("tcvn" . tcvn)
2457 ("big5[-_]?hkscs" . big5-hkscs)
2458 ("big5" . big5)
2459 ("euc-?tw" . euc-tw)
2460 ("euc-?cn" . euc-cn)
2461 ("gb2312" . gb2312)
2462 ("gbk" . gbk)
2463 ("gb18030" . gb18030)
2464 ("ja.*[._]euc" . japanese-iso-8bit)
2465 ("ja.*[._]jis7" . iso-2022-jp)
2466 ("ja.*[._]pck" . japanese-shift-jis)
2467 ("ja.*[._]sjis" . japanese-shift-jis)
2468 ("jpn" . japanese-shift-jis) ; MS-Windows uses this.
2469 ))
2470 "List of pairs of locale regexps and preferred coding systems.
2471 The first element whose locale regexp matches the start of a downcased locale
2472 specifies the coding system to prefer when using that locale.
2473 This coding system is used if the locale specifies a specific charset.")
2475 (defun locale-name-match (key alist)
2476 "Search for KEY in ALIST, which should be a list of regexp-value pairs.
2477 Return the value corresponding to the first regexp that matches the
2478 start of KEY, or nil if there is no match."
2479 (let (element)
2480 (while (and alist (not element))
2482 (setq element (car alist)))
2483 (setq alist (cdr alist)))
2484 (cdr element)))
2486 (defun locale-charset-match-p (charset1 charset2)
2487 "Whether charset names (strings) CHARSET1 and CHARSET2 are equivalent.
2488 Matching is done ignoring case and any hyphens and underscores in the
2489 names. E.g. `ISO_8859-1' and `iso88591' both match `iso-8859-1'."
2490 (setq charset1 (replace-regexp-in-string "[-_]" "" charset1))
2491 (setq charset2 (replace-regexp-in-string "[-_]" "" charset2))
2492 (eq t (compare-strings charset1 nil nil charset2 nil nil t)))
2494 (defvar locale-charset-alist nil
2495 "Coding system alist keyed on locale-style charset name.
2496 Used by `locale-charset-to-coding-system'.")
2498 (defun locale-charset-to-coding-system (charset)
2499 "Find coding system corresponding to CHARSET.
2500 CHARSET is any sort of non-Emacs charset name, such as might be used
2501 in a locale codeset, or elsewhere. It is matched to a coding system
2502 first by case-insensitive lookup in `locale-charset-alist'. Then
2503 matches are looked for in the coding system list, treating case and
2504 the characters `-' and `_' as insignificant. The coding system base
2506 `iso-latin-2' is returned."
2507 (or (car (assoc-string charset locale-charset-alist t))
2508 (let ((cs coding-system-alist)
2509 c)
2510 (while (and (not c) cs)
2511 (if (locale-charset-match-p charset (caar cs))
2512 (setq c (intern (caar cs)))
2513 (pop cs)))
2514 (if c (coding-system-base c)))))
2516 ;; Fixme: This ought to deal with the territory part of the locale
2517 ;; too, for setting things such as calendar holidays, ps-print paper
2518 ;; size, spelling dictionary.
2520 (declare-function w32-get-console-codepage "w32proc.c" ())
2521 (declare-function w32-get-console-output-codepage "w32proc.c" ())
2523 (defun locale-translate (locale)
2524 "Expand LOCALE according to `locale-translation-file-name', if possible.
2526 (if locale-translation-file-name
2527 (with-temp-buffer
2528 (set-buffer-multibyte nil)
2529 (insert-file-contents locale-translation-file-name)
2530 (if (re-search-forward
2532 (buffer-substring (point) (line-end-position))
2533 locale))
2534 locale))
2536 (defun set-locale-environment (&optional locale-name frame)
2537 "Set up multilingual environment for using LOCALE-NAME.
2538 This sets the language environment, the coding system priority,
2539 the default input method and sometimes other things.
2541 LOCALE-NAME should be a string which is the name of a locale supported
2542 by the system. Often it is of the form xx_XX.CODE, where xx is a
2543 language, XX is a country, and CODE specifies a character set and
2545 a locale for Japanese in Japan using the `japanese-iso-8bit'
2546 coding-system. The name may also have a modifier suffix, e.g. `@euro'
2547 or `@cyrillic'.
2549 If LOCALE-NAME is nil, its value is taken from the environment
2550 variables LC_ALL, LC_CTYPE and LANG (the first one that is set).
2552 The locale names supported by your system can typically be found in a
2553 directory named `/usr/share/locale' or `/usr/lib/locale'. LOCALE-NAME
2554 will be translated according to the table specified by
2555 `locale-translation-file-name'.
2557 If FRAME is non-nil, only set the keyboard coding system and the
2558 terminal coding system for the terminal of that frame, and don't
2559 touch session-global parameters like the language environment.
2561 See also `locale-charset-language-names', `locale-language-names',
2562 `locale-preferred-coding-systems' and `locale-coding-system'."
2563 (interactive "sSet environment for locale: ")
2565 ;; Do this at runtime for the sake of binaries possibly transported
2566 ;; to a system without X.
2567 (setq locale-translation-file-name
2568 (let ((files
2569 '("/usr/share/X11/locale/locale.alias" ; e.g. X11R7
2570 "/usr/lib/X11/locale/locale.alias" ; e.g. X11R6.4
2571 "/usr/X11R6/lib/X11/locale/locale.alias" ; XFree86, e.g. RedHat 4.2
2572 "/usr/openwin/lib/locale/locale.alias" ; e.g. Solaris 2.6
2573 ;;
2574 ;; The following name appears after the X-related names above,
2575 ;; since the X-related names are what X actually uses.
2576 "/usr/share/locale/locale.alias" ; GNU/Linux sans X
2577 )))
2578 (while (and files (not (file-exists-p (car files))))
2579 (setq files (cdr files)))
2580 (car files)))
2582 (let ((locale locale-name))
2584 (unless locale
2585 ;; Use the first of these three environment variables
2586 ;; that has a nonempty value.
2587 (let ((vars '("LC_ALL" "LC_CTYPE" "LANG")))
2588 (while (and vars
2589 (= 0 (length locale))) ; nil or empty string
2590 (setq locale (getenv (pop vars) frame)))))
2592 (when locale
2593 (setq locale (locale-translate locale))
2595 ;; Leave the system locales alone if the caller did not specify
2596 ;; an explicit locale name, as their defaults are set from
2597 ;; LC_MESSAGES and LC_TIME, not LC_CTYPE, and the user might not
2598 ;; want to set them to the same value as LC_CTYPE.
2599 (when locale-name
2600 (setq system-messages-locale locale)
2601 (setq system-time-locale locale))
2603 (if (string-match "^[a-z][a-z]" locale)
2604 ;; The value of 'current-iso639-language' is matched against
2605 ;; the ':lang' property of font-spec objects when selecting
2606 ;; and prioritizing available fonts for displaying
2607 ;; characters; see fontset.c.
2608 (setq current-iso639-language
2609 ;; The call to 'downcase' is for w32, where the
2610 ;; MS-Windows locale names are in caps, as in "ENU",
2611 ;; the equivalent of the Posix "en_US". Since the
2612 ;; match mentioned above uses memq, and ':lang'
2613 ;; properties have lower-case values, the letter-case
2614 ;; must match exactly.
2615 (intern (downcase (match-string 0 locale))))))
2617 (setq woman-locale
2618 (or system-messages-locale
2619 (let ((msglocale (getenv "LC_MESSAGES" frame)))
2620 (if (zerop (length msglocale))
2621 locale
2622 (locale-translate msglocale)))))
2624 (when locale
2625 (setq locale (downcase locale))
2627 (let ((language-name
2628 (locale-name-match locale locale-language-names))
2629 (charset-language-name
2630 (locale-name-match locale locale-charset-language-names))
2631 (default-eol-type (coding-system-eol-type
2632 (default-value 'buffer-file-coding-system)))
2633 (coding-system
2634 (or (locale-name-match locale locale-preferred-coding-systems)
2635 (when locale
2637 (locale-charset-to-coding-system
2638 (match-string 1 locale)))))))
2640 (if (consp language-name)
2641 ;; locale-language-names specify both lang-env and coding.
2642 ;; But, what specified in locale-preferred-coding-systems
2643 ;; has higher priority.
2644 (setq coding-system (or coding-system
2645 (nth 1 language-name))
2646 language-name (car language-name))
2647 ;; Otherwise, if locale is not listed in locale-language-names,
2648 ;; use what listed in locale-charset-language-names.
2649 (if (not language-name)
2650 (setq language-name charset-language-name)))
2652 ;; If a specific EOL conversion was specified in the default
2653 ;; buffer-file-coding-system, preserve it in the coding system
2654 ;; we will be using from now on.
2655 (if (and (memq default-eol-type '(0 1 2 unix dos mac))
2656 coding-system
2657 (coding-system-p coding-system))
2658 (setq coding-system (coding-system-change-eol-conversion
2659 coding-system default-eol-type)))
2661 (when language-name
2663 ;; Set up for this character set. This is now the right way
2664 ;; to do it for both unibyte and multibyte modes.
2665 (unless frame
2666 (set-language-environment language-name))
2668 ;; If the default enable-multibyte-characters is nil,
2669 ;; we are using single-byte characters,
2670 ;; so the display table and terminal coding system are irrelevant.
2671 (when (default-value 'enable-multibyte-characters)
2672 (set-display-table-and-terminal-coding-system
2673 language-name coding-system frame))
2675 ;; Set the `keyboard-coding-system' if appropriate (tty
2676 ;; only). At least X and MS Windows can generate
2677 ;; multilingual input.
2678 ;; XXX This was disabled unless `window-system', but that
2679 ;; leads to buggy behavior when a tty frame is opened
2680 ;; later. Setting the keyboard coding system has no adverse
2681 ;; effect on X, so let's do it anyway. -- Lorentey
2682 (let ((kcs (or coding-system
2683 (car (get-language-info language-name
2684 'coding-system)))))
2685 (if kcs (set-keyboard-coding-system kcs frame)))
2687 (unless frame
2688 (setq locale-coding-system
2689 (car (get-language-info language-name 'coding-priority)))))
2691 (when (and (not frame)
2692 coding-system
2693 (not (coding-system-equal coding-system
2694 locale-coding-system)))
2695 (prefer-coding-system coding-system)
2696 ;; Fixme: perhaps prefer-coding-system should set this too.
2697 ;; But it's not the time to do such a fundamental change.
2698 (setq default-sendmail-coding-system coding-system)
2699 (setq locale-coding-system coding-system))))
2701 ;; On Windows, override locale-coding-system,
2702 ;; default-file-name-coding-system, keyboard-coding-system,
2703 ;; terminal-coding-system with the ANSI or console codepage.
2704 (when (and (eq system-type 'windows-nt)
2705 (boundp 'w32-ansi-code-page))
2706 (let* ((ansi-code-page-coding
2707 (intern (format "cp%d" w32-ansi-code-page)))
2708 (code-page-coding
2709 (if noninteractive
2710 (intern (format "cp%d" (w32-get-console-codepage)))
2711 ansi-code-page-coding))
2712 (output-coding
2713 (if noninteractive
2714 (intern (format "cp%d" (w32-get-console-output-codepage)))
2715 code-page-coding)))
2716 (when (coding-system-p code-page-coding)
2717 (or output-coding (setq output-coding code-page-coding))
2718 (unless frame (setq locale-coding-system code-page-coding))
2719 (set-keyboard-coding-system code-page-coding frame)
2720 (set-terminal-coding-system output-coding frame)
2721 (setq default-file-name-coding-system ansi-code-page-coding))))
2723 (when (eq system-type 'darwin)
2724 ;; On Darwin, file names are always encoded in utf-8, no matter
2725 ;; the locale.
2726 (setq default-file-name-coding-system 'utf-8-unix)
2727 ;; macOS's Terminal.app by default uses utf-8 regardless of
2728 ;; the locale.
2729 (when (and (null window-system)
2730 (equal (getenv "TERM_PROGRAM" frame) "Apple_Terminal"))
2731 (set-terminal-coding-system 'utf-8)
2732 (set-keyboard-coding-system 'utf-8)))
2734 ;; Default to A4 paper if we're not in a C, POSIX or US locale.
2735 ;; (See comments in Flocale_info.)
2736 (unless frame
2737 (let ((paper (locale-info 'paper))
2738 locale)
2739 (if paper
2740 ;; This will always be null at the time of writing.
2741 (cond
2742 ((equal paper '(216 279))
2743 (setq ps-paper-type 'letter))
2744 ((equal paper '(210 297))
2745 (setq ps-paper-type 'a4)))
2746 (let ((vars '("LC_ALL" "LC_PAPER" "LANG")))
2747 (while (and vars (= 0 (length locale)))
2748 (setq locale (getenv (pop vars) frame))))
2749 (when locale
2750 ;; As of glibc 2.2.5, these are the only US Letter locales,
2751 ;; and the rest are A4.
2752 (setq ps-paper-type
2753 (or (locale-name-match locale '(("c$" . letter)
2754 ("posix$" . letter)
2755 (".._us" . letter)
2756 (".._pr" . letter)
2757 (".._ca" . letter)
2758 ("enu$" . letter) ; Windows
2759 ("esu$" . letter)
2760 ("enc$" . letter)
2761 ("frc$" . letter)))
2762 'a4)))))))
2763 nil)
2764 \f
2765 ;;; Character property
2767 (put 'char-code-property-table 'char-table-extra-slots 5)
2769 (defun define-char-code-property (name table &optional docstring)
2770 "Define NAME as a character code property given by TABLE.
2771 TABLE is a char-table of purpose `char-code-property-table' with
2772 these extra slots:
2773 1st: NAME.
2774 2nd: Function to call to get a property value of a character.
2775 It is called with three arguments CHAR, VAL, and TABLE, where
2776 CHAR is a character, VAL is the value of (aref TABLE CHAR).
2777 3rd: Function to call to put a property value of a character.
2778 It is called with the same arguments as above.
2779 4th: Function to call to get a description string of a property value.
2780 It is called with one argument VALUE, a property value.
2781 5th: Data used by the above functions.
2783 TABLE may be a name of file to load to build a char-table. The
2784 file should contain a call of `define-char-code-property' with a
2785 char-table of the above format as the argument TABLE.
2787 TABLE may also be nil, in which case no property value is pre-assigned.
2789 Optional 3rd argument DOCSTRING is a documentation string of the property.
2791 See also the documentation of `get-char-code-property' and
2792 `put-char-code-property'."
2793 (or (symbolp name)
2794 (error "Not a symbol: %s" name))
2795 (if (char-table-p table)
2796 (or (and (eq (char-table-subtype table) 'char-code-property-table)
2797 (eq (char-table-extra-slot table 0) name))
2798 (error "Invalid char-table: %s" table))
2799 (or (stringp table)
2800 (error "Not a char-table nor a file name: %s" table)))
2801 (if (stringp table) (setq table (purecopy table)))
2802 (setf (alist-get name char-code-property-alist) table)
2803 (put name 'char-code-property-documentation (purecopy docstring)))
2805 (defvar char-code-property-table
2806 (make-char-table 'char-code-property-table)
2807 "Char-table containing a property list of each character code.
2808 This table is used for properties not listed in `char-code-property-alist'.
2809 See also the documentation of `get-char-code-property' and
2810 `put-char-code-property'.")
2812 (defun get-char-code-property (char propname)
2813 "Return the value of CHAR's PROPNAME property."
2814 (let ((table (unicode-property-table-internal propname)))
2815 (if table
2816 (let ((func (char-table-extra-slot table 1)))
2817 (if (functionp func)
2818 (funcall func char (aref table char) table)
2819 (get-unicode-property-internal table char)))
2820 (plist-get (aref char-code-property-table char) propname))))
2822 (defun put-char-code-property (char propname value)
2823 "Store CHAR's PROPNAME property with VALUE.
2824 It can be retrieved with `(get-char-code-property CHAR PROPNAME)'."
2825 (let ((table (unicode-property-table-internal propname)))
2826 (if table
2827 (let ((func (char-table-extra-slot table 2)))
2828 (if (functionp func)
2829 (funcall func char value table)
2830 (put-unicode-property-internal table char value)))
2831 (let* ((plist (aref char-code-property-table char))
2832 (x (plist-put plist propname value)))
2833 (or (eq x plist)
2834 (aset char-code-property-table char x))))
2835 value))
2837 (defun char-code-property-description (prop value)
2838 "Return a description string of character property PROP's value VALUE.
2839 If there's no description string for VALUE, return nil."
2840 (let ((table (unicode-property-table-internal prop)))
2841 (if table
2842 (let ((func (char-table-extra-slot table 3)))
2843 (if (functionp func)
2844 (funcall func value))))))
2846 \f
2847 ;; Pretty description of encoded string
2849 ;; Alist of ISO 2022 control code vs the corresponding mnemonic string.
2850 (defconst iso-2022-control-alist
2858 (defun encoded-string-description (str coding-system)
2859 "Return a pretty description of STR that is encoded by CODING-SYSTEM."
2860 (setq str (string-as-unibyte str))
2861 (mapconcat
2862 (if (and coding-system (eq (coding-system-type coding-system) 'iso-2022))
2863 ;; Try to get a pretty description for ISO 2022 escape sequences.
2864 (function (lambda (x) (or (cdr (assq x iso-2022-control-alist))
2865 (format "#x%02X" x))))
2866 (function (lambda (x) (format "#x%02X" x))))
2867 str " "))
2869 (defun encode-coding-char (char coding-system &optional charset)
2870 "Encode CHAR by CODING-SYSTEM and return the resulting string.
2871 If CODING-SYSTEM can't safely encode CHAR, return nil.
2872 The 3rd optional argument CHARSET, if non-nil, is a charset preferred
2873 on encoding."
2874 (let* ((str1 (string-as-multibyte (string char)))
2875 (str2 (string-as-multibyte (string char char)))
2876 (found (find-coding-systems-string str1))
2877 enc1 enc2 i1 i2)
2878 (if (and (consp found)
2879 (eq (car found) 'undecided))
2880 str1
2881 (when (memq (coding-system-base coding-system) found)
2882 ;; We must find the encoded string of CHAR. But, just encoding
2883 ;; CHAR will put extra control sequences (usually to designate
2884 ;; ASCII charset) at the tail if type of CODING is ISO 2022.
2885 ;; To exclude such tailing bytes, we at first encode one-char
2886 ;; string and two-char string, then check how many bytes at the
2887 ;; tail of both encoded strings are the same.
2889 (when charset
2890 (put-text-property 0 1 'charset charset str1)
2891 (put-text-property 0 2 'charset charset str2))
2892 (setq enc1 (encode-coding-string str1 coding-system)
2893 i1 (length enc1)
2894 enc2 (encode-coding-string str2 coding-system)
2895 i2 (length enc2))
2896 (while (and (> i1 0) (= (aref enc1 (1- i1)) (aref enc2 (1- i2))))
2897 (setq i1 (1- i1) i2 (1- i2)))
2899 ;; Now (substring enc1 i1) and (substring enc2 i2) are the same,
2900 ;; and they are the extra control sequences at the tail to
2901 ;; exclude.
2902 (substring enc2 0 i2)))))
2904 ;; Backwards compatibility. These might be better with :init-value t,
2905 ;; but that breaks loadup.
2906 (define-minor-mode unify-8859-on-encoding-mode
2907 "Exists only for backwards compatibility."
2908 :group 'mule
2909 :global t)
2910 ;; Doc said "obsolete" in 23.1, this statement only added in 24.1.
2911 (make-obsolete 'unify-8859-on-encoding-mode "don't use it." "23.1")
2913 (define-minor-mode unify-8859-on-decoding-mode
2914 "Exists only for backwards compatibility."
2915 :group 'mule
2916 :global t)
2917 ;; Doc said "obsolete" in 23.1, this statement only added in 24.1.
2918 (make-obsolete 'unify-8859-on-decoding-mode "don't use it." "23.1")
2920 (defvar nonascii-insert-offset 0)
2921 (make-obsolete-variable 'nonascii-insert-offset "do not use it." "23.1")
2922 (defvar nonascii-translation-table nil)
2923 (make-obsolete-variable 'nonascii-translation-table "do not use it." "23.1")
2925 (defvar ucs-names nil
2926 "Hash table of cached CHAR-NAME keys to CHAR-CODE values.")
2928 (defun ucs-names ()
2929 "Return table of CHAR-NAME keys and CHAR-CODE values cached in `ucs-names'."
2930 (or ucs-names
2931 (let ((ranges
2932 '((#x0000 . #x33FF)
2933 ;; (#x3400 . #x4DBF) CJK Ideographs Extension A
2934 (#x4DC0 . #x4DFF)
2935 ;; (#x4E00 . #x9FFF) CJK Unified Ideographs
2936 (#xA000 . #xD7FF)
2937 ;; (#xD800 . #xFAFF) Surrogate/Private
2938 (#xFB00 . #x134FF)
2939 ;; (#x13500 . #x143FF) unused
2940 (#x14400 . #x14646)
2941 ;; (#x14647 . #x167FF) unused
2942 (#x16800 . #x16F9F)
2943 (#x16FE0 . #x16FE0)
2944 ;; (#x17000 . #x187FF) Tangut Ideographs
2945 ;; (#x18800 . #x18AFF) Tangut Components
2946 ;; (#x18B00 . #x1AFFF) unused
2947 (#x1B000 . #x1B12F)
2948 ;; (#x1B130 . #x1B16F) unused
2949 (#x1B170 . #x1B2FF)
2950 ;; (#x1B300 . #x1BBFF) unused
2951 (#x1BC00 . #x1BCAF)
2952 ;; (#x1BCB0 . #x1CFFF) unused
2953 (#x1D000 . #x1FFFF)
2954 ;; (#x20000 . #xDFFFF) CJK Ideograph Extension A, B, etc, unused
2955 (#xE0000 . #xE01FF)))
2956 (gc-cons-threshold 10000000)
2957 (names (make-hash-table :size 42943 :test #'equal)))
2958 (dolist (range ranges)
2959 (let ((c (car range))
2960 (end (cdr range)))
2961 (while (<= c end)
2962 (let ((new-name (get-char-code-property c 'name))
2963 (old-name (get-char-code-property c 'old-name)))
2964 ;; In theory this code could end up pushing an "old-name" that
2965 ;; shadows a "new-name" but in practice every time an
2966 ;; `old-name' conflicts with a `new-name', the newer one has a
2967 ;; higher code, so it gets pushed later!
2968 (if new-name (puthash new-name c names))
2969 (if old-name (puthash old-name c names))
2970 (setq c (1+ c))))))
2971 ;; Special case for "BELL" which is apparently the only char which
2972 ;; doesn't have a new name and whose old-name is shadowed by a newer
2973 ;; char with that name.
2975 (setq ucs-names names))))
2977 (defun mule--ucs-names-annotation (name)
2978 ;; FIXME: It would be much better to add this annotation before rather than
2979 ;; after the char name, so the annotations are aligned.
2980 ;; FIXME: The default behavior of displaying annotations in italics
2981 ;; doesn't work well here.
2982 (let ((char (gethash name ucs-names)))
2983 (when char (format " (%c)" char))))
2985 (defun char-from-name (string &optional ignore-case)
2986 "Return a character as a number from its Unicode name STRING.
2987 If optional IGNORE-CASE is non-nil, ignore case in STRING.
2988 Return nil if STRING does not name a character."
2989 (or (gethash (if ignore-case (upcase string) string) (ucs-names))
2991 (when minus
2992 ;; Parse names like "VARIATION SELECTOR-17" and "CJK
2993 ;; COMPATIBILITY IDEOGRAPH-F900" that are not in ucs-names.
2994 (ignore-errors
2995 (let* ((case-fold-search ignore-case)
2997 (minus-num (string-to-number (substring string minus)
2998 (if vs 10 16)))
2999 (vs-offset (if vs (if (< minus-num -16) #xE00EF #xFDFF) 0))
3000 (code (- vs-offset minus-num))
3001 (name (get-char-code-property code 'name)))
3002 (when (eq t (compare-strings string nil nil name nil nil
3003 ignore-case))
3004 code)))))))
3006 (defun read-char-by-name (prompt)
3007 "Read a character by its Unicode name or hex number string.
3008 Display PROMPT and read a string that represents a character by its
3009 Unicode property `name' or `old-name'.
3011 This function returns the character as a number.
3013 You can type a few of the first letters of the Unicode name and
3014 use completion. If you type a substring of the Unicode name
3015 preceded by an asterisk `*' and use completion, it will show all
3016 the characters whose names include that substring, not necessarily
3017 at the beginning of the name.
3023 as names, not numbers."
3024 (let* ((enable-recursive-minibuffers t)
3025 (completion-ignore-case t)
3026 (input
3027 (completing-read
3028 prompt
3029 (lambda (string pred action)
3030 (if (eq action 'metadata)
3031 '(metadata
3032 (annotation-function . mule--ucs-names-annotation)
3033 (category . unicode-name))
3034 (complete-with-action action (ucs-names) string pred)))))
3035 (char
3036 (cond
3037 ((char-from-name input t))
3039 (ignore-errors (string-to-number input 16)))
3041 input)
3042 (ignore-errors (read input))))))
3043 (unless (characterp char)
3044 (error "Invalid character"))
3045 char))
3047 (define-obsolete-function-alias 'ucs-insert 'insert-char "24.3")
3050 ;;; mule-cmds.el ends here