| blob | bdba8eeb11282aae1eb7746c1452364c15292f57 |
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.
363 ;; Prevent default-terminal-coding-system from converting ^M to ^J.
366 ;; Preserve eol-type from existing default-process-coding-systems.
367 ;; On non-unix-like systems in particular, these may have been set
368 ;; carefully by the user, or by the startup code, to deal with the
369 ;; users shell appropriately, so should not be altered by changing
370 ;; language environment.
381 "Add CODING-SYSTEM at the front of the priority list for automatic detection.
382 This also sets the following coding systems:
383 o coding system of a newly created buffer
384 o default coding system for subprocess I/O
385 This also sets the following values:
386 o default value used as `file-name-coding-system' for converting file names
387 o default value for the command `set-terminal-coding-system'
388 o default value for the command `set-keyboard-coding-system'
390 If CODING-SYSTEM specifies a certain type of EOL conversion, the coding
391 systems set by this function will use that type of EOL conversion.
393 A coding system that requires automatic detection of text+encoding
394 \(e.g. undecided, unix) can't be preferred.
396 To prefer, for instance, utf-8, say the following:
410 base coding-system)))
411 ;; If they asked for specific EOL conversion, honor that.
418 (message "The default value of `file-name-coding-system' was not changed because the specified coding system is not suitable for file names.")))))
421 "If non-nil, a predicate function to sort coding systems.
423 It is called with two coding systems, and should return t if the first
429 "Sort coding system list CODINGS by a priority of each coding system.
430 Return the sorted list. CODINGS is modified by side effects.
432 If a coding system is most preferred, it has the highest priority.
433 Otherwise, coding systems that correspond to MIME charsets have
434 higher priorities. Among them, a coding system included in the
435 `coding-system' key of the current language environment has higher
436 priority. See also the documentation of `language-info-alist'.
438 If the variable `sort-coding-systems-predicate' (which see) is
439 non-nil, it is used to sort CODINGS instead."
449 ;; We calculate the priority number 0..255 by
450 ;; using the 8 bits PMMLCEII as this:
451 ;; P: 1 if most preferred.
452 ;; MM: greater than 0 if mime-charset.
453 ;; L: 1 if one of the current lang. env.'s codings.
454 ;; C: 1 if one of codings listed in the category list.
455 ;; E: 1 if not XXX-with-esc
456 ;; II: if iso-2022 based, 0..3, else 1.
461 ;; Prefer coding systems corresponding to a
462 ;; MIME charset.
464 ;; Lower utf-16 priority so that we
465 ;; normally prefer utf-8 to it, and put
466 ;; x-ctext below that.
482 ;; For ISO based coding systems, prefer
483 ;; one that doesn't use designation nor
484 ;; locking/single shifting.
495 ))))))
500 "Return a list of proper coding systems to encode a text between FROM and TO.
502 If FROM is a string, find coding systems in that instead of the buffer.
503 All coding systems in the list can safely encode any multibyte characters
504 in the text.
506 If the text contains no multibyte characters, return a list of a single
507 element `undecided'."
510 ;; The text contains only ASCII characters. Any coding
511 ;; systems are safe.
513 ;; We need copy-sequence because sorting will alter the argument.
517 "Return a list of proper coding systems to encode STRING.
518 All coding systems in the list can safely encode any multibyte characters
519 in STRING.
521 If STRING contains no multibyte characters, return a list of a single
522 element `undecided'."
526 "Return a list of proper coding systems to encode characters of CHARSETS.
527 CHARSETS is a list of character sets.
529 This only finds coding systems of type `charset', whose
530 `:charset-list' property includes all of CHARSETS (plus `ascii' for
531 ASCII-compatible coding systems). It was used in older versions of
532 Emacs, but is unlikely to be what you really want now."
533 ;; Deal with aliases.
536 charsets))
557 t))
562 "Find multibyte characters in the region specified by FROM and TO.
563 If FROM is a string, find multibyte characters in the string.
564 The return value is an alist of the following format:
565 ((CHARSET COUNT CHAR ...) ...)
566 where
567 CHARSET is a character set,
568 COUNT is a number of characters,
569 CHARs are the characters found from the character set.
570 Optional 3rd arg MAXCOUNT limits how many CHARs are put in the above list.
571 Optional 4th arg EXCLUDES is a list of character sets to be ignored."
573 charset char)
608 "Search forward from point for a character that is not encodable.
609 It asks which coding system to check.
610 If such a character is found, set point after that character.
611 Otherwise, don't move point.
613 When called from a program, the value is the position of the unencodable
614 character found, or nil if all characters are encodable."
619 default))))
624 pos))
627 "Most recent coding system explicitly specified by the user when asked.
628 This variable is set whenever Emacs asks the user which coding system
629 to use in order to write a file. If you set it to nil explicitly,
630 then call `write-region', then afterward this variable will be non-nil
634 "If non-nil, a function to control the behavior of coding system selection.
635 The meaning is the same as the argument ACCEPT-DEFAULT-P of the
636 function `select-safe-coding-system' (which see). This variable
640 "Return a list of coding systems presumably more user-friendly than CODINGS."
641 ;; Change each safe coding system to the corresponding
642 ;; mime-charset name if it is also a coding system. Such a name
643 ;; is more friendly to users.
649 mime-charset cs)))
650 codings))
652 ;; Don't offer variations with locking shift, which you
653 ;; basically never want.
662 ;; Remove raw-text, emacs-mule and no-conversion unless nothing
663 ;; else is available.
667 codings))
671 "Select interactively a coding system for the region FROM ... TO.
672 FROM can be a string, as in `write-region'.
673 CODINGS is the list of base coding systems known to be safe for this region,
674 typically obtained with `find-coding-systems-region'.
675 UNSAFE is a list of coding systems known to be unsafe for this region.
676 REJECTED is a list of coding systems which were safe but for some reason
677 were not recommended in the particular context.
678 DEFAULT is the coding system to use by default in the query."
679 ;; At first, if some defaults are unsafe, record at most 11
680 ;; problematic characters and their positions for them by turning
681 ;; (CODING ...)
682 ;; into
683 ;; ((CODING (POS . CHAR) (POS . CHAR) ...) ...)
698 unsafe)))
704 coding-system)
706 ;; If some defaults are unsafe, make sure the offending
707 ;; buffer is displayed.
711 unsafe))))
712 ;; Then ask users to select one from CODINGS while showing
713 ;; the reason why none of the defaults are not used.
722 "These default coding systems were tried to encode"
737 but are not recommended for encoding text in this context,
773 'help-echo
774 "mouse-2, RET: jump to this character"
775 'help-function func1
778 "..."
781 'help-echo
782 "mouse-2, RET: next unencodable character"
783 'help-function func2
806 ;; Read a coding system.
810 default))
815 coding-system))
818 accept-default-p file)
819 "Ask a user to select a safe coding system from candidates.
820 The candidates of coding systems which can safely encode a text
821 between FROM and TO are shown in a popup window. Among them, the most
822 proper one is suggested as the default.
824 The list of `buffer-file-coding-system' of the current buffer, the
825 default `buffer-file-coding-system', and the most preferred coding
826 system (if it corresponds to a MIME charset) is treated as the
827 default coding system list. Among them, the first one that safely
828 encodes the text is normally selected silently and returned without
829 any user interaction. See also the command `prefer-coding-system'.
831 However, the user is queried if the chosen coding system is
832 inconsistent with what would be selected by `find-auto-coding' from
833 coding cookies &c. if the contents of the region were read from a
834 file. (That could lead to data corruption in a file subsequently
835 re-visited and edited.)
837 Optional 3rd arg DEFAULT-CODING-SYSTEM specifies a coding system or a
838 list of coding systems to be prepended to the default coding system
839 list. However, if DEFAULT-CODING-SYSTEM is a list and the first
840 element is t, the cdr part is used as the default coding system list,
841 i.e. current `buffer-file-coding-system', default `buffer-file-coding-system',
842 and the most preferred coding system are not used.
844 Optional 4th arg ACCEPT-DEFAULT-P, if non-nil, is a function to
845 determine the acceptability of the silently selected coding system.
846 It is called with that coding system, and should return nil if it
847 should not be silently selected and thus user interaction is required.
849 Optional 5th arg FILE is the file name to use for this purpose.
850 That is different from `buffer-file-name' when handling `write-region'
851 \(for example).
853 The variable `select-safe-coding-system-accept-default-p', if non-nil,
854 overrides ACCEPT-DEFAULT-P.
856 Kludgy feature: if FROM is a string, the string is the target text,
857 and TO is ignored."
862 auto-cs)
864 ;; Find an auto-coding that is specified for the current
865 ;; buffer and file from the region FROM and TO.
876 'mule
878 Invalid coding system `%s' is specified
879 for the current buffer/file by the %s.
880 It is highly recommended to fix it before writing to a file."
893 ;; Change elements of the list to (coding . base-coding).
896 default-coding-system))
899 ;; If the file has a coding cookie, use it regardless of any
900 ;; other setting.
907 ;; If buffer-file-coding-system is not nil nor undecided, append it
908 ;; to the defaults.
919 ;; If default buffer-file-coding-system is not nil nor undecided,
920 ;; append it to the defaults.
930 base)))))))
932 ;; If the most preferred coding system has the property mime-charset,
933 ;; append it to the defaults.
935 base)
947 ;; Decide the eol-type from the top of the default codings,
948 ;; current buffer-file-coding-system, or default buffer-file-coding-system.
954 buffer-file-coding-system)))
968 safe rejected unsafe)
970 ;; Any coding system is ok.
972 ;; Reverse the list so that elements are accumulated in safe,
973 ;; rejected, and unsafe in the correct order.
976 ;; Classify the defaults into safe, rejected, and unsafe.
979 ;; This is safe. Is it acceptable?
982 ;; No, not acceptable.
984 ;; Yes, acceptable.
986 ;; This is not safe.
988 ;; If there are safe ones, the first one is what we want.
992 ;; If all the defaults failed, ask a user.
997 ;; Check we're not inconsistent with what `coding:' spec &c would
998 ;; give when file is re-read.
999 ;; But don't do this if we explicitly ignored the cookie
1000 ;; by using `find-file-literally'.
1003 coding-system
1005 ;; Merge coding-system and auto-cs as far as possible.
1014 coding-system auto-cs))
1017 auto-cs coding-system))))
1021 coding-system eol-type-2)))
1024 auto-cs eol-type-1)))))))
1027 ;; Don't barf if writing a compressed file, say.
1028 ;; This check perhaps isn't ideal, but is probably
1029 ;; the best thing to do.
1034 %s specified by file contents. Really save (else edit coding cookies \
1047 coding-system)))
1052 "Return a coding system to encode the outgoing message of the current buffer.
1053 It at first tries the first coding system found in these variables
1054 in this order:
1055 (1) local value of `buffer-file-coding-system'
1056 (2) value of `sendmail-coding-system'
1057 (3) value of `default-sendmail-coding-system'
1058 (4) default value of `buffer-file-coding-system'
1059 If the found coding system can't encode the current buffer,
1060 or none of them are bound to a coding system,
1061 it asks the user to select a proper coding system."
1063 buffer-file-coding-system)
1064 sendmail-coding-system
1065 default-sendmail-coding-system
1068 ;; We should never use no-conversion for outgoing mail.
1074 coding)))
1075 \f
1076 ;;; Language support stuff.
1079 "Alist of language environment definitions.
1080 Each element looks like:
1081 (LANGUAGE-NAME . ((KEY . INFO) ...))
1082 where LANGUAGE-NAME is a string, the name of the language environment,
1083 KEY is a symbol denoting the kind of information, and
1084 INFO is the data associated with KEY.
1085 Meaningful values for KEY include
1087 documentation value is documentation of what this language environment
1088 is meant for, and how to use it.
1089 charset value is a list of the character sets mainly used
1090 by this language environment.
1091 sample-text value is an expression which is evalled to generate
1092 a line of text written using characters appropriate
1093 for this language environment.
1094 setup-function value is a function to call to switch to this
1095 language environment.
1096 exit-function value is a function to call to leave this
1097 language environment.
1098 coding-system value is a list of coding systems that are good for
1099 saving text written in this language environment.
1100 This list serves as suggestions to the user;
1101 in effect, as a kind of documentation.
1102 coding-priority value is a list of coding systems for this language
1103 environment, in order of decreasing priority.
1104 This is used to set up the coding system priority
1105 list when you switch to this language environment.
1106 nonascii-translation
1107 value is a charset of dimension one to use for
1108 converting a unibyte character to multibyte
1109 and vice versa.
1110 input-method value is a default input method for this language
1111 environment.
1112 features value is a list of features requested in this
1113 language environment.
1114 ctext-non-standard-encodings
1115 value is a list of non-standard encoding names used
1116 in extended segments of CTEXT. See the variable
1117 `ctext-non-standard-encodings' for more detail.
1119 The following key takes effect only when multibyte characters are
1120 globally disabled, i.e. the default value of `enable-multibyte-characters'
1121 is nil (which is an obsolete and deprecated use):
1123 unibyte-display value is a coding system to encode characters for
1124 the terminal. Characters in the range of 160 to
1125 255 display not as octal escapes, but as non-ASCII
1129 "Return information listed under KEY for language environment LANG-ENV.
1130 KEY is a symbol denoting the kind of information.
1131 For a list of useful values for KEY and their meanings,
1132 see `language-info-alist'."
1140 "Modify part of the definition of language environment LANG-ENV.
1141 Specifically, this stores the information INFO under KEY
1142 in the definition of this language environment.
1143 KEY is a symbol denoting the kind of information.
1144 INFO is the value for that information.
1146 For a list of useful values for KEY and their meanings,
1147 see `language-info-alist'."
1166 "Internal use only.
1167 Arguments are the same as `set-language-info'."
1179 ;; Update the custom-type of `current-language-environment'.
1187 "Store ALIST as the definition of language environment LANG-ENV.
1188 ALIST is an alist of KEY and INFO values. See the documentation of
1189 `language-info-alist' for the meanings of KEY and INFO.
1191 Optional arg PARENTS is a list of parent menu names; it specifies
1192 where to put this language environment in the
1193 Describe Language Environment and Set Language Environment menus.
1195 in the European submenu in each of those two menus."
1204 map parent-symbol parent prompt)
1210 ;; This prompt string is for define-prefix-command, so
1211 ;; that the map it creates will be suitable for a menu.
1232 ;; Set up menu items for this language env.
1247 "Read a language environment name which has information for KEY.
1248 If KEY is nil, read any language environment.
1249 Prompt with PROMPT. DEFAULT is the default choice of language environment.
1250 This returns a language environment name as a string."
1253 language-info-alist
1256 t nil nil default)))
1260 name)))
1261 \f
1262 ;;; Multilingual input methods.
1264 "LEIM: Libraries of Emacs Input Methods."
1268 "Name of LEIM list file.
1269 This file contains a list of libraries of Emacs input methods (LEIM)
1270 in the format of Lisp expression for registering each input method.
1274 ";;; %s -- list of LEIM (Library of Emacs Input Method) -*-coding: utf-8;-*-
1275 ;;
1276 ;; This file is automatically generated.
1277 ;;
1278 ;; This file contains a list of LEIM (Library of Emacs Input Method)
1279 ;; methods in the same directory as this file. Loading this file
1280 ;; registers all the input methods in Emacs.
1281 ;;
1282 ;; Each entry has the form:
1283 ;; (register-input-method
1284 ;; INPUT-METHOD LANGUAGE-NAME ACTIVATE-FUNC
1285 ;; TITLE DESCRIPTION
1286 ;; ARG ...)
1287 ;; See the function `register-input-method' for the meanings of the arguments.
1288 ;;
1289 ;; If this directory is included in `load-path', Emacs automatically
1290 ;; loads this file at startup time.
1292 "
1293 leim-list-file-name)
1297 "Regexp matching head of each entry in LEIM list file.
1302 "List of functions to call to update LEIM list file.
1306 "Update LEIM list file in directories DIRS."
1311 "The current input method for multilingual text.
1322 "String widget with completion for input method."
1323 :completions
1330 "Default input method for multilingual text (a string).
1331 This is the input method activated automatically by the command
1336 mule-input-method-string)
1342 "History list of input methods read from the minibuffer.
1344 Maximum length of the history list is determined by the value
1350 'inactivate-current-input-method-function
1353 "Function to call for deactivating the current input method.
1354 Every input method should set this to an appropriate value when activated.
1355 This function is called with no argument.
1357 This function should never change the value of `current-input-method'.
1363 "Function to call for describing the current input method.
1369 "Alist of input method names vs how to use them.
1370 Each element has the form:
1371 (INPUT-METHOD LANGUAGE-ENV ACTIVATE-FUNC TITLE DESCRIPTION ARGS...)
1373 ;;;###autoload
1377 "Register INPUT-METHOD as an input method for language environment LANG-ENV.
1379 INPUT-METHOD and LANG-ENV are symbols or strings.
1380 ACTIVATE-FUNC is a function to call to activate this method.
1381 TITLE is a string to show in the mode line when this method is active.
1382 DESCRIPTION is a string describing this method and what it is good for.
1383 The ARGS, if any, are passed as arguments to ACTIVATE-FUNC.
1384 All told, the arguments to ACTIVATE-FUNC are INPUT-METHOD and the ARGS.
1387 created at Emacs build time, registering all Quail input methods
1388 contained in the Emacs distribution.
1390 In case you want to register a new Quail input method by yourself, be
1391 careful to use the same input method title as given in the third
1392 parameter of `quail-define-package'. (If the values are different, the
1393 string specified in this function takes precedence.)
1395 The commands `describe-input-method' and `list-input-methods' need
1396 these duplicated values to show some information about input methods
1397 without loading the relevant Quail packages.
1414 "Read a name of input method from a minibuffer prompting with PROMPT.
1415 If DEFAULT is non-nil, use that as the default,
1416 and substitute it into PROMPT at the first `%s'.
1417 If INHIBIT-NULL is non-nil, null input signals an error.
1419 The return value is a string."
1423 ;; As it is quite normal to change input method in the
1424 ;; minibuffer, we must enable it even if
1425 ;; enable-recursive-minibuffers is currently nil.
1427 ;; This binding is necessary because input-method-history is
1428 ;; buffer local.
1430 nil t nil 'input-method-history
1433 default))))
1437 input-method
1442 "Switch to input method INPUT-METHOD for the current buffer.
1443 If some other input method is already active, turn it off first.
1444 If INPUT-METHOD is nil, deactivate any current input method."
1471 "Turn off the current input method."
1482 current-input-method-title nil)
1492 'inactivate-input-method
1496 "Select and activate input method INPUT-METHOD for the current buffer.
1497 This also sets the default input method to the one you specify.
1498 If INPUT-METHOD is nil, this function turns off the input method, and
1499 also causes you to be prompted for a name of an input method the next
1501 When called interactively, the optional arg INTERACTIVE is non-nil,
1502 which marks the variable `default-input-method' as set for Custom buffers.
1505 To deactivate it programmatically, use `deactivate-input-method'."
1510 default t)
1511 t)))
1516 default-input-method)
1522 "Enable or disable multilingual text input method for the current buffer.
1523 Only one input method can be enabled at any time in a given buffer.
1525 The normal action is to enable an input method if none was enabled,
1526 and disable the current one otherwise. Which input method to enable
1527 can be determined in various ways--either the one most recently used,
1528 or the one specified by `default-input-method', or as a last resort
1529 by reading the name of an input method in the minibuffer.
1531 With a prefix argument ARG, read an input method name with the minibuffer
1532 and enable that one. The default is the most recent input method specified
1533 \(not including the currently active input method, if any).
1535 When called interactively, the optional argument INTERACTIVE is non-nil,
1536 which marks the variable `default-input-method' as set for Custom buffers."
1553 default t))
1554 default))
1564 "Describe input method INPUT-METHOD."
1594 "Describe the input method currently in use.
1595 This is a subroutine for `describe-input-method'."
1601 current-input-method)
1606 "Read a multilingual string from minibuffer, prompting with string PROMPT.
1607 The input method selected last time is activated in minibuffer.
1608 If optional second argument INITIAL-INPUT is non-nil, insert it in the
1609 minibuffer initially.
1610 Optional 3rd argument INPUT-METHOD specifies the input method to be activated
1611 instead of the one selected last time. It is a symbol or a string."
1614 current-input-method
1615 default-input-method
1626 ;; Variables to control behavior of input methods. All input methods
1627 ;; should react to these variables.
1630 "A flag to control extra guidance given by input methods.
1631 The value should be nil, t, `complex-only', or `default'.
1633 The extra guidance is done by showing list of available keys in echo
1634 area. When you use the input method in the minibuffer, the guidance
1635 is shown at the bottom short window (split from the existing window).
1637 If the value is t, extra guidance is always given, if the value is
1638 nil, extra guidance is always suppressed.
1640 If the value is `complex-only', only complex input methods such as
1641 `chinese-py' and `japanese' give extra guidance.
1643 If the value is `default', complex input methods always give extra
1644 guidance, but simple input methods give it only when you are not in
1645 the minibuffer.
1647 See also the variable `input-method-highlight-flag'."
1653 "If this flag is non-nil, input methods highlight partially-entered text.
1654 For instance, while you are in the middle of a Quail input method sequence,
1655 the text inserted so far is temporarily underlined.
1656 The underlining goes away when you finish or abort the input method sequence.
1657 See also the variable `input-method-verbose-flag'."
1662 "Normal hook run just after an input method is activated.
1664 The variable `current-input-method' keeps the input method name
1665 just activated."
1670 'input-method-inactivate-hook
1674 "Normal hook run just after an input method is deactivated.
1676 The variable `current-input-method' still keeps the input method name
1677 just deactivated."
1683 "Normal hook run just after an input method insert some chunk of text."
1688 "This flag controls when an input method returns.
1689 Usually, the input method does not return while there's a possibility
1690 that it may find a different translation if a user types another key.
1691 But, if this flag is non-nil, the input method returns as soon as the
1695 "This flag controls how an input method shows an intermediate key sequence.
1696 Usually, the input method inserts the intermediate key sequence,
1697 or candidate translations corresponding to the sequence,
1698 at point in the current buffer.
1699 But, if this flag is non-nil, it displays them in echo area instead."
1704 "This flag controls the behavior of an input method on invalid key input.
1705 Usually, when a user types a key which doesn't start any character
1706 handled by the input method, the key is handled by turning off the
1707 input method temporarily. After that key, the input method is re-enabled.
1710 \f
1712 "Normal hook run after some language environment is set.
1714 When you set some hook function here, that effect usually should not
1715 be inherited to another language environment. So, you had better set
1716 another function in `exit-language-environment-hook' (which see) to
1717 cancel the effect."
1722 "Normal hook run after exiting from some language environment.
1723 When this hook is run, the variable `current-language-environment'
1724 is still bound to the language environment being exited.
1726 This hook is mainly used for canceling the effect of
1727 `set-language-environment-hook' (which see)."
1734 "Switch to a specified language environment."
1747 "The last language environment specified with `set-language-environment'.
1749 to using the function `set-language-environment'."
1756 current-language-environment)
1757 language-info-alist t))
1759 ;; custom type will be updated with `set-language-info'.
1770 "Reset multilingual environment of Emacs to the default status.
1772 The default status is as follows:
1774 The default value of `buffer-file-coding-system' is nil.
1775 The default coding system for process I/O is nil.
1776 The default value for the command `set-terminal-coding-system' is nil.
1777 The default value for the command `set-keyboard-coding-system' is nil.
1779 The order of priorities of coding systems are as follows:
1780 utf-8
1781 iso-2022-7bit
1782 iso-latin-1
1783 iso-2022-7bit-lock
1784 iso-2022-8bit-ss2
1785 emacs-mule
1786 raw-text"
1788 ;; This function formerly set default-enable-multibyte-characters to t,
1789 ;; but that is incorrect. It should not alter the unibyte/multibyte choice.
1792 'utf-8
1793 'iso-2022-7bit
1794 'iso-latin-1
1795 'iso-2022-7bit-lock
1796 'iso-2022-8bit-ss2
1797 'emacs-mule
1802 ;; On Darwin systems, this should be utf-8, but when this file is loaded
1803 ;; utf-8 is not yet defined, so we set it in set-locale-environment instead.
1805 ;; Preserve eol-type from existing default-process-coding-systems.
1806 ;; On non-unix-like systems in particular, these may have been set
1807 ;; carefully by the user, or by the startup code, to deal with the
1808 ;; users shell appropriately, so should not be altered by changing
1809 ;; language environment.
1811 ;; When bootstrapping, coding-systems are not defined yet, so
1812 ;; we need to catch the error from check-coding-system.
1825 ;; Put the highest priority to the charset iso-8859-1 to prefer the
1826 ;; registry iso8859-1 over iso8859-2 in font selection. It also
1827 ;; makes unibyte-display-via-language-environment to use iso-8859-1
1828 ;; as the unibyte charset.
1831 ;; Don't alter the terminal and keyboard coding systems here.
1832 ;; The terminal still supports the same coding system
1833 ;; that it supported a minute ago.
1834 ;; (set-terminal-coding-system-internal nil)
1835 ;; (set-keyboard-coding-system-internal nil)
1837 ;; Back in Emacs-20, it was necessary to provide some fallback implicit
1838 ;; conversion, because almost no packages handled coding-system issues.
1839 ;; Nowadays it'd just paper over bugs.
1840 ;; (set-unibyte-charset 'iso-8859-1)
1841 )
1845 (defun set-display-table-and-terminal-coding-system (language-name &optional coding-system display)
1846 "Set up the display table and terminal coding system for LANGUAGE-NAME."
1852 ;; The following 2 lines undo the 8-bit display that we set up
1853 ;; in standard-display-european-internal, which see. This is in
1854 ;; case the user has used standard-display-european earlier in
1855 ;; this session.
1862 "Set up multilingual environment for using LANGUAGE-NAME.
1863 This sets the coding system priority and the default input method
1864 and sometimes other things. LANGUAGE-NAME should be a string
1866 specifies the character set for the major languages of Western Europe.
1868 If there is a prior value for `current-language-environment', this
1869 runs the hook `exit-language-environment-hook'. After setting up
1870 the new language environment, it runs `set-language-environment-hook'."
1872 nil
1889 ;; The features might set up coding systems.
1901 ;; Unibyte setups if necessary.
1911 current-iso639-language))
1917 "An Emacs charset."
1919 :completions
1929 widget))
1933 "Customizations of language environment parameters.
1934 Value is an alist with elements like those of `language-info-alist'.
1935 These are used to set values in `language-info-alist' which replace
1936 the defaults. A typical use is replacing the default input method for
1939 This option is intended for use at startup. Removing items doesn't
1940 remove them from the language info until you next restart Emacs.
1942 Setting this variable directly does not take effect.
1943 See `set-language-info-alist' for use in programs."
1948 ;; Can't do this before language environments are set up.
1950 ;; modify language-info-alist
1953 ;; re-set the environment in case its parameters changed
1957 :completions
1961 action language-info-alist string pred))))
1962 :value-type
1980 ;; Actually set up direct output of non-ASCII characters.
1982 ;; Unibyte Emacs on MS-DOS wants to display all 8-bit characters with
1983 ;; the native font, and codes 160 and 146 stand for something very
1984 ;; different there.
1988 ;; Most X fonts used to do the wrong thing for latin-1 code 160.
1990 ;; XFree86 4 has fixed the fonts.
1994 ;; Make non-line-break space display as a plain space.
1997 ;; don't contain a character at that position. Map it to the ASCII
1998 ;; apostrophe. [This is actually RIGHT SINGLE QUOTATION MARK,
1999 ;; U+2019, normally from the windows-1252 character set. XFree 4
2000 ;; fonts probably have the appropriate glyph at this position,
2001 ;; so they could use standard-display-8bit. It's better to use a
2002 ;; proper windows-1252 coding system. --fx]
2003 (aset standard-display-table (unibyte-char-to-multibyte 146) [39]))))
2005 (defun set-language-environment-coding-systems (language-name)
2006 "Do various coding system setups for language environment LANGUAGE-NAME."
2007 (let* ((priority (get-language-info language-name 'coding-priority))
2008 (default-coding (car priority))
2009 ;; If the default buffer-file-coding-system is nil, don't use
2010 ;; coding-system-eol-type, because it treats nil as
2011 ;; `no-conversion'. The default buffer-file-coding-system is set
2012 ;; to nil by reset-language-environment, and in that case we
2013 ;; want to have here the native EOL type for each platform.
2014 ;; FIXME: there should be a common code that runs both on
2015 ;; startup and here to set the default EOL type correctly.
2016 ;; Right now, DOS/Windows platforms set this on dos-w32.el,
2017 ;; which works only as long as the order of loading files at
2018 ;; dump time and calling functions at startup is not modified
2019 ;; significantly, i.e. as long as this function is called
2020 ;; _after_ the default buffer-file-coding-system was set by
2021 ;; dos-w32.el.
2022 (eol-type
2023 (coding-system-eol-type
2024 (or (default-value 'buffer-file-coding-system)
2025 (if (memq system-type '(windows-nt ms-dos)) 'dos 'unix)))))
2026 (when priority
2027 (set-default-coding-systems
2028 (if (memq eol-type '(0 1 2 unix dos mac))
2029 (coding-system-change-eol-conversion default-coding eol-type)
2030 default-coding))
2031 (setq default-sendmail-coding-system default-coding)
2032 (apply 'set-coding-system-priority priority))))
2034 (defun set-language-environment-input-method (language-name)
2035 "Do various input method setups for language environment LANGUAGE-NAME."
2036 (let ((input-method (get-language-info language-name 'input-method)))
2037 (when input-method
2038 (setq default-input-method input-method)
2039 (if input-method-history
2040 (setq input-method-history
2041 (cons input-method
2042 (delete input-method input-method-history)))))))
2044 (defun set-language-environment-nonascii-translation (language-name)
2045 "Do unibyte/multibyte translation setup for language environment LANGUAGE-NAME."
2046 ;; Note: For DOS, we assumed that the charset cpXXX is already
2047 ;; defined.
2048 (let ((nonascii (get-language-info language-name 'nonascii-translation)))
2049 (if (eq window-system 'pc)
2050 (setq nonascii (intern (format "cp%d" dos-codepage))))
2051 (or (and (charsetp nonascii)
2052 (get-charset-property nonascii :ascii-compatible-p))
2053 (setq nonascii 'iso-8859-1))
2054 ;; Back in Emacs-20, it was necessary to provide some fallback implicit
2055 ;; conversion, because almost no packages handled coding-system issues.
2056 ;; Nowadays it'd just paper over bugs.
2057 ;; (set-unibyte-charset nonascii)
2058 ))
2060 (defun set-language-environment-charset (language-name)
2061 "Do various charset setups for language environment LANGUAGE-NAME."
2062 ;; Put higher priorities to such charsets that are supported by the
2063 ;; coding systems of higher priorities in this environment.
2064 (let ((charsets (get-language-info language-name 'charset)))
2065 (dolist (coding (get-language-info language-name 'coding-priority))
2066 (let ((list (coding-system-charset-list coding)))
2067 (if (consp list)
2068 (setq charsets (append charsets list)))))
2069 (if charsets
2070 (apply 'set-charset-priority charsets))))
2072 (defun set-language-environment-unibyte (language-name)
2073 "Do various unibyte-mode setups for language environment LANGUAGE-NAME."
2074 (set-display-table-and-terminal-coding-system language-name))
2076 (defun princ-list (&rest args)
2078 (declare (obsolete "use mapc and princ instead." "23.3"))
2079 (mapc #'princ args)
2082 (put 'describe-specified-language-support 'apropos-inhibit t)
2084 ;; Print language-specific information such as input methods,
2085 ;; charsets, and coding systems. This function is intended to be
2086 ;; called from the menu:
2087 ;; [menu-bar mule describe-language-environment LANGUAGE]
2088 ;; and should not run it by `M-x describe-current-input-method-function'.
2089 (defun describe-specified-language-support ()
2090 "Describe how Emacs supports the specified language environment."
2091 (interactive)
2092 (let (language-name)
2093 (if (not (and (symbolp last-command-event)
2094 (or (not (eq last-command-event 'Default))
2095 (setq last-command-event 'English))
2096 (setq language-name (symbol-name last-command-event))))
2097 (error "This command should only be called from the menu bar"))
2098 (describe-language-environment language-name)))
2100 (defun describe-language-environment (language-name)
2101 "Describe how Emacs supports language environment LANGUAGE-NAME."
2102 (interactive
2103 (list (read-language-name
2104 'documentation
2105 "Describe language environment (default current choice): ")))
2106 (if (null language-name)
2107 (setq language-name current-language-environment))
2108 (if (or (null language-name)
2109 (null (get-language-info language-name 'documentation)))
2110 (error "No documentation for the specified language"))
2111 (if (symbolp language-name)
2112 (setq language-name (symbol-name language-name)))
2113 (dolist (feature (get-language-info language-name 'features))
2114 (require feature))
2115 (let ((doc (get-language-info language-name 'documentation)))
2116 (help-setup-xref (list #'describe-language-environment language-name)
2117 (called-interactively-p 'interactive))
2118 (with-output-to-temp-buffer (help-buffer)
2119 (with-current-buffer standard-output
2121 (if (stringp doc)
2123 (condition-case nil
2124 (let ((str (eval (get-language-info language-name 'sample-text))))
2125 (if (stringp str)
2129 (error nil))
2130 (let ((input-method (get-language-info language-name 'input-method))
2131 (l (copy-sequence input-method-alist))
2132 (first t))
2133 (when (and input-method
2134 (setq input-method (assoc input-method l)))
2136 (setq l (cons input-method (delete input-method l))
2137 first nil))
2138 (dolist (elt l)
2139 (when (or (eq input-method elt)
2140 (eq t (compare-strings language-name nil nil
2141 (nth 1 elt) nil nil t)))
2142 (when first
2144 (setq first nil))
2145 (insert " " (car elt))
2146 (search-backward (car elt))
2147 (help-xref-button 0 'help-input-method (car elt))
2148 (goto-char (point-max))
2150 (if (stringp (nth 3 elt)) (nth 3 elt) (car (nth 3 elt)))
2152 (or first
2155 (let ((l (get-language-info language-name 'charset)))
2156 (if (null l)
2158 (while l
2159 (insert " " (symbol-name (car l)))
2160 (search-backward (symbol-name (car l)))
2161 (help-xref-button 0 'help-character-set (car l))
2162 (goto-char (point-max))
2164 (setq l (cdr l)))))
2167 (let ((l (get-language-info language-name 'coding-system)))
2168 (if (null l)
2170 (while l
2171 (insert " " (symbol-name (car l)))
2172 (search-backward (symbol-name (car l)))
2173 (help-xref-button 0 'help-coding-system (car l))
2174 (goto-char (point-max))
2175 (insert (substitute-command-keys " (`")
2176 (coding-system-mnemonic (car l))
2178 (substitute-command-keys
2179 (coding-system-doc-string (car l)))
2181 (let ((aliases (coding-system-aliases (car l))))
2182 (when aliases
2184 (while aliases
2185 (insert " " (symbol-name (car aliases)))
2186 (setq aliases (cdr aliases)))
2188 (setq l (cdr l)))))))))
2189 \f
2190 ;;; Locales.
2192 (defvar locale-translation-file-name nil
2193 "File name for the system's file of locale-name aliases, or nil if none.")
2195 ;; The following definitions might as well be marked as constants and
2196 ;; purecopied, since they're normally used on startup, and probably
2197 ;; should reflect the facilities of the base Emacs.
2198 (defconst locale-language-names
2199 (purecopy
2200 '(
2201 ;; Locale names of the form LANGUAGE[_TERRITORY][.CODESET][@MODIFIER]
2202 ;; as specified in the Single Unix Spec, Version 2.
2203 ;; LANGUAGE is a language code taken from ISO 639:1988 (E/F)
2204 ;; with additions from ISO 639/RA Newsletter No.1/1989;
2205 ;; see Internet RFC 2165 (1997-06) and
2206 ;; http://www.evertype.com/standards/iso639/iso639-en.html
2207 ;; TERRITORY is a country code taken from ISO 3166
2208 ;; http://www.din.de/gremien/nas/nabd/iso3166ma/codlstp1/en_listp1.html.
2209 ;; CODESET and MODIFIER are implementation-dependent.
2211 ;; jasonr comments: MS Windows uses three letter codes for
2212 ;; languages instead of the two letter ISO codes that POSIX
2213 ;; uses. In most cases the first two letters are the same, so
2214 ;; most of the regexps in locale-language-names work. Japanese
2215 ;; and Chinese are exceptions, which are listed in the
2216 ;; non-standard section at the bottom of locale-language-names.
2218 ("aa_DJ" . "Latin-1") ; Afar
2219 ("aa" . "UTF-8")
2220 ;; ab Abkhazian
2221 ("af" . "Latin-1") ; Afrikaans
2222 ("am" "Ethiopic" utf-8) ; Amharic
2223 ("an" . "Latin-9") ; Aragonese
2224 ("ar" . "Arabic")
2225 ; as Assamese
2226 ; ay Aymara
2227 ("az" . "UTF-8") ; Azerbaijani
2228 ; ba Bashkir
2229 ("be" "Belarusian" cp1251) ; Belarusian [Byelorussian until early 1990s]
2230 ("bg" "Bulgarian" cp1251) ; Bulgarian
2231 ; bh Bihari
2232 ; bi Bislama
2233 ("bn" . "UTF-8") ; Bengali, Bangla
2234 ("bo" . "Tibetan")
2235 ("br" . "Latin-1") ; Breton
2236 ("bs" . "Latin-2") ; Bosnian
2237 ("byn" . "UTF-8") ; Bilin; Blin
2238 ("ca" "Catalan" iso-8859-1) ; Catalan
2239 ; co Corsican
2240 ("cs" "Czech" iso-8859-2)
2241 ("cy" "Welsh" iso-8859-14)
2242 ("da" . "Latin-1") ; Danish
2243 ("de" "German" iso-8859-1)
2244 ; dv Divehi
2245 ; dz Bhutani
2246 ("el" "Greek" iso-8859-7)
2247 ;; Users who specify "en" explicitly typically want Latin-1, not ASCII.
2248 ;; That's actually what the GNU locales define, modulo things like
2249 ;; en_IN -- fx.
2250 ("en_IN" "English" utf-8) ; glibc uses utf-8 for English in India
2251 ("en" "English" iso-8859-1) ; English
2252 ("eo" . "Esperanto") ; Esperanto
2253 ("es" "Spanish" iso-8859-1)
2254 ("et" . "Latin-1") ; Estonian
2255 ("eu" . "Latin-1") ; Basque
2256 ("fa" . "UTF-8") ; Persian
2257 ("fi" . "Latin-1") ; Finnish
2258 ("fj" . "Latin-1") ; Fiji
2259 ("fo" . "Latin-1") ; Faroese
2260 ("fr" "French" iso-8859-1) ; French
2261 ("fy" . "Latin-1") ; Frisian
2262 ("ga" . "Latin-1") ; Irish Gaelic (new orthography)
2263 ("gd" . "Latin-9") ; Scots Gaelic
2264 ("gez" "Ethiopic" utf-8) ; Geez
2265 ("gl" . "Latin-1") ; Gallegan; Galician
2266 ; gn Guarani
2267 ("gu" . "UTF-8") ; Gujarati
2268 ("gv" . "Latin-1") ; Manx Gaelic
2269 ; ha Hausa
2270 ("he" "Hebrew" iso-8859-8)
2271 ("hi" "Devanagari" utf-8) ; Hindi
2272 ("hr" "Croatian" iso-8859-2) ; Croatian
2273 ("hu" . "Latin-2") ; Hungarian
2274 ; hy Armenian
2275 ; ia Interlingua
2276 ("id" . "Latin-1") ; Indonesian
2277 ; ie Interlingue
2278 ; ik Inupiak
2279 ("is" . "Latin-1") ; Icelandic
2280 ("it" "Italian" iso-8859-1) ; Italian
2281 ; iu Inuktitut
2282 ("iw" "Hebrew" iso-8859-8)
2283 ("ja" "Japanese" euc-jp)
2284 ; jw Javanese
2285 ("ka" "Georgian" georgian-ps) ; Georgian
2286 ; kk Kazakh
2287 ("kl" . "Latin-1") ; Greenlandic
2288 ; km Cambodian
2289 ("kn" "Kannada" utf-8)
2290 ("ko" "Korean" euc-kr)
2291 ; ks Kashmiri
2292 ; ku Kurdish
2293 ("kw" . "Latin-1") ; Cornish
2294 ; ky Kirghiz
2295 ("la" . "Latin-1") ; Latin
2296 ("lb" . "Latin-1") ; Luxemburgish
2297 ("lg" . "Laint-6") ; Ganda
2298 ; ln Lingala
2299 ("lo" "Lao" utf-8) ; Laothian
2300 ("lt" "Lithuanian" iso-8859-13)
2301 ("lv" . "Latvian") ; Latvian, Lettish
2302 ; mg Malagasy
2303 ("mi" . "Latin-7") ; Maori
2304 ("mk" "Cyrillic-ISO" iso-8859-5) ; Macedonian
2305 ("ml" "Malayalam" utf-8)
2306 ("mn" . "UTF-8") ; Mongolian
2307 ; mo Moldavian
2308 ("mr" "Devanagari" utf-8) ; Marathi
2309 ("ms" . "Latin-1") ; Malay
2310 ("mt" . "Latin-3") ; Maltese
2311 ; my Burmese
2312 ; na Nauru
2313 ("nb" . "Latin-1") ; Norwegian
2314 ("ne" "Devanagari" utf-8) ; Nepali
2315 ("nl" "Dutch" iso-8859-1)
2316 ("no" . "Latin-1") ; Norwegian
2317 ("oc" . "Latin-1") ; Occitan
2318 ("om_ET" . "UTF-8") ; (Afan) Oromo
2319 ("om" . "Latin-1") ; (Afan) Oromo
2320 ; or Oriya
2321 ("pa" . "UTF-8") ; Punjabi
2322 ("pl" . "Latin-2") ; Polish
2323 ; ps Pashto, Pushto
2324 ("pt" . "Latin-1") ; Portuguese
2325 ; qu Quechua
2326 ("rm" . "Latin-1") ; Rhaeto-Romanic
2327 ; rn Kirundi
2328 ("ro" "Romanian" iso-8859-2)
2329 ("ru_RU" "Russian" iso-8859-5)
2330 ("ru_UA" "Russian" koi8-u)
2331 ; rw Kinyarwanda
2332 ("sa" . "Devanagari") ; Sanskrit
2333 ; sd Sindhi
2334 ("se" . "UTF-8") ; Northern Sami
2335 ; sg Sangho
2336 ("sh" . "Latin-2") ; Serbo-Croatian
2337 ; si Sinhalese
2338 ("sid" . "UTF-8") ; Sidamo
2339 ("sk" "Slovak" iso-8859-2)
2340 ("sl" "Slovenian" iso-8859-2)
2341 ; sm Samoan
2342 ; sn Shona
2343 ("so_ET" "UTF-8") ; Somali
2344 ("so" "Latin-1") ; Somali
2345 ("sq" . "Latin-1") ; Albanian
2346 ("sr" . "Latin-2") ; Serbian (Latin alphabet)
2347 ; ss Siswati
2348 ("st" . "Latin-1") ; Sesotho
2349 ; su Sundanese
2350 ("sv" "Swedish" iso-8859-1) ; Swedish
2351 ("sw" . "Latin-1") ; Swahili
2352 ("ta" "Tamil" utf-8)
2353 ("te" . "UTF-8") ; Telugu
2354 ("tg" "Tajik" koi8-t)
2355 ("th" "Thai" tis-620)
2356 ("ti" "Ethiopic" utf-8) ; Tigrinya
2357 ("tig_ER" . "UTF-8") ; Tigre
2358 ; tk Turkmen
2359 ("tl" . "Latin-1") ; Tagalog
2360 ; tn Setswana
2361 ; to Tonga
2362 ("tr" "Turkish" iso-8859-9)
2363 ; ts Tsonga
2364 ("tt" . "UTF-8") ; Tatar
2365 ; tw Twi
2366 ; ug Uighur
2367 ("uk" "Ukrainian" koi8-u)
2368 ("ur" . "UTF-8") ; Urdu
2369 ("uz_UZ@cyrillic" . "UTF-8"); Uzbek
2370 ("uz" . "Latin-1") ; Uzbek
2371 ("vi" "Vietnamese" utf-8)
2372 ; vo Volapuk
2373 ("wa" . "Latin-1") ; Walloon
2374 ; wo Wolof
2375 ("xh" . "Latin-1") ; Xhosa
2376 ("yi" . "Windows-1255") ; Yiddish
2377 ; yo Yoruba
2378 ; za Zhuang
2379 ("zh_HK" . "Chinese-Big5")
2380 ; zh_HK/BIG5-HKSCS \
2381 ("zh_TW" . "Chinese-Big5")
2382 ("zh_CN.GB2312" "Chinese-GB")
2383 ("zh_CN.GBK" "Chinese-GBK")
2384 ("zh_CN.GB18030" "Chinese-GB18030")
2385 ("zh_CN.UTF-8" . "Chinese-GBK")
2386 ("zh_CN" . "Chinese-GB")
2387 ("zh" . "Chinese-GB")
2388 ("zu" . "Latin-1") ; Zulu
2390 ;; ISO standard locales
2391 ("c$" . "ASCII")
2392 ("posix$" . "ASCII")
2394 ;; The "IPA" Emacs language environment does not correspond
2395 ;; to any ISO 639 code, so let it stand for itself.
2396 ("ipa$" . "IPA")
2398 ;; Nonstandard or obsolete language codes
2399 ("cz" . "Czech") ; e.g. Solaris 2.6
2400 ("ee" . "Latin-4") ; Estonian, e.g. X11R6.4
2401 ("iw" . "Hebrew") ; e.g. X11R6.4
2402 ("sp" . "Cyrillic-ISO") ; Serbian (Cyrillic alphabet), e.g. X11R6.4
2403 ("su" . "Latin-1") ; Finnish, e.g. Solaris 2.6
2404 ("jp" . "Japanese") ; e.g. MS Windows
2405 ("chs" . "Chinese-GBK") ; MS Windows Chinese Simplified
2406 ("cht" . "Chinese-BIG5") ; MS Windows Chinese Traditional
2407 ("gbz" . "UTF-8") ; MS Windows Dari Persian
2408 ("div" . "UTF-8") ; MS Windows Divehi (Maldives)
2409 ("wee" . "Latin-2") ; MS Windows Lower Sorbian
2410 ("wen" . "Latin-2") ; MS Windows Upper Sorbian
2411 ))
2412 "Alist of locale regexps vs the corresponding languages and coding systems.
2413 Each element has this form:
2414 (LOCALE-REGEXP LANG-ENV CODING-SYSTEM)
2415 The first element whose LOCALE-REGEXP matches the start of a
2416 downcased locale specifies the LANG-ENV \(language environment)
2417 and CODING-SYSTEM corresponding to that locale. If there is no
2418 appropriate language environment, the element may have this form:
2419 (LOCALE-REGEXP . LANG-ENV)
2420 In this case, LANG-ENV is one of generic language environments for an
2423 (defconst locale-charset-language-names
2424 (purecopy
2433 ;; utf-8@euro exists, so put this last. (@euro really specifies
2434 ;; the currency, rather than the charset.)
2436 "List of pairs of locale regexps and charset language names.
2437 The first element whose locale regexp matches the start of a downcased locale
2438 specifies the language name whose charset corresponds to that locale.
2439 This language name is used if the locale is not listed in
2440 `locale-language-names'.")
2442 (defconst locale-preferred-coding-systems
2443 (purecopy
2452 ;; utf-8@euro exists, so put this after utf-8. (@euro really
2453 ;; specifies the currency, rather than the charset.)
2454 (".*@euro" . iso-8859-15)
2455 ("koi8-?r" . koi8-r)
2456 ("koi8-?u" . koi8-u)
2457 ("tcvn" . tcvn)
2458 ("big5[-_]?hkscs" . big5-hkscs)
2459 ("big5" . big5)
2460 ("euc-?tw" . euc-tw)
2461 ("euc-?cn" . euc-cn)
2462 ("gb2312" . gb2312)
2463 ("gbk" . gbk)
2464 ("gb18030" . gb18030)
2465 ("ja.*[._]euc" . japanese-iso-8bit)
2466 ("ja.*[._]jis7" . iso-2022-jp)
2467 ("ja.*[._]pck" . japanese-shift-jis)
2468 ("ja.*[._]sjis" . japanese-shift-jis)
2469 ("jpn" . japanese-shift-jis) ; MS-Windows uses this.
2470 ))
2471 "List of pairs of locale regexps and preferred coding systems.
2472 The first element whose locale regexp matches the start of a downcased locale
2473 specifies the coding system to prefer when using that locale.
2474 This coding system is used if the locale specifies a specific charset.")
2476 (defun locale-name-match (key alist)
2477 "Search for KEY in ALIST, which should be a list of regexp-value pairs.
2478 Return the value corresponding to the first regexp that matches the
2479 start of KEY, or nil if there is no match."
2480 (let (element)
2481 (while (and alist (not element))
2483 (setq element (car alist)))
2484 (setq alist (cdr alist)))
2485 (cdr element)))
2487 (defun locale-charset-match-p (charset1 charset2)
2488 "Whether charset names (strings) CHARSET1 and CHARSET2 are equivalent.
2489 Matching is done ignoring case and any hyphens and underscores in the
2490 names. E.g. `ISO_8859-1' and `iso88591' both match `iso-8859-1'."
2491 (setq charset1 (replace-regexp-in-string "[-_]" "" charset1))
2492 (setq charset2 (replace-regexp-in-string "[-_]" "" charset2))
2493 (eq t (compare-strings charset1 nil nil charset2 nil nil t)))
2495 (defvar locale-charset-alist nil
2496 "Coding system alist keyed on locale-style charset name.
2497 Used by `locale-charset-to-coding-system'.")
2499 (defun locale-charset-to-coding-system (charset)
2500 "Find coding system corresponding to CHARSET.
2501 CHARSET is any sort of non-Emacs charset name, such as might be used
2502 in a locale codeset, or elsewhere. It is matched to a coding system
2503 first by case-insensitive lookup in `locale-charset-alist'. Then
2504 matches are looked for in the coding system list, treating case and
2505 the characters `-' and `_' as insignificant. The coding system base
2507 `iso-latin-2' is returned."
2508 (or (car (assoc-string charset locale-charset-alist t))
2509 (let ((cs coding-system-alist)
2510 c)
2511 (while (and (not c) cs)
2512 (if (locale-charset-match-p charset (caar cs))
2513 (setq c (intern (caar cs)))
2514 (pop cs)))
2515 (if c (coding-system-base c)))))
2517 ;; Fixme: This ought to deal with the territory part of the locale
2518 ;; too, for setting things such as calendar holidays, ps-print paper
2519 ;; size, spelling dictionary.
2521 (declare-function w32-get-console-codepage "w32proc.c" ())
2522 (declare-function w32-get-console-output-codepage "w32proc.c" ())
2524 (defun locale-translate (locale)
2525 "Expand LOCALE according to `locale-translation-file-name', if possible.
2527 (if locale-translation-file-name
2528 (with-temp-buffer
2529 (set-buffer-multibyte nil)
2530 (insert-file-contents locale-translation-file-name)
2531 (if (re-search-forward
2533 (buffer-substring (point) (line-end-position))
2534 locale))
2535 locale))
2537 (defun set-locale-environment (&optional locale-name frame)
2538 "Set up multilingual environment for using LOCALE-NAME.
2539 This sets the language environment, the coding system priority,
2540 the default input method and sometimes other things.
2542 LOCALE-NAME should be a string which is the name of a locale supported
2543 by the system. Often it is of the form xx_XX.CODE, where xx is a
2544 language, XX is a country, and CODE specifies a character set and
2546 a locale for Japanese in Japan using the `japanese-iso-8bit'
2547 coding-system. The name may also have a modifier suffix, e.g. `@euro'
2548 or `@cyrillic'.
2550 If LOCALE-NAME is nil, its value is taken from the environment
2551 variables LC_ALL, LC_CTYPE and LANG (the first one that is set).
2553 The locale names supported by your system can typically be found in a
2554 directory named `/usr/share/locale' or `/usr/lib/locale'. LOCALE-NAME
2555 will be translated according to the table specified by
2556 `locale-translation-file-name'.
2558 If FRAME is non-nil, only set the keyboard coding system and the
2559 terminal coding system for the terminal of that frame, and don't
2560 touch session-global parameters like the language environment.
2562 See also `locale-charset-language-names', `locale-language-names',
2563 `locale-preferred-coding-systems' and `locale-coding-system'."
2564 (interactive "sSet environment for locale: ")
2566 ;; Do this at runtime for the sake of binaries possibly transported
2567 ;; to a system without X.
2568 (setq locale-translation-file-name
2569 (let ((files
2570 '("/usr/share/X11/locale/locale.alias" ; e.g. X11R7
2571 "/usr/lib/X11/locale/locale.alias" ; e.g. X11R6.4
2572 "/usr/X11R6/lib/X11/locale/locale.alias" ; XFree86, e.g. RedHat 4.2
2573 "/usr/openwin/lib/locale/locale.alias" ; e.g. Solaris 2.6
2574 ;;
2575 ;; The following name appears after the X-related names above,
2576 ;; since the X-related names are what X actually uses.
2577 "/usr/share/locale/locale.alias" ; GNU/Linux sans X
2578 )))
2579 (while (and files (not (file-exists-p (car files))))
2580 (setq files (cdr files)))
2581 (car files)))
2583 (let ((locale locale-name))
2585 (unless locale
2586 ;; Use the first of these three environment variables
2587 ;; that has a nonempty value.
2588 (let ((vars '("LC_ALL" "LC_CTYPE" "LANG")))
2589 (while (and vars
2590 (= 0 (length locale))) ; nil or empty string
2591 (setq locale (getenv (pop vars) frame)))))
2593 (when locale
2594 (setq locale (locale-translate locale))
2596 ;; Leave the system locales alone if the caller did not specify
2597 ;; an explicit locale name, as their defaults are set from
2598 ;; LC_MESSAGES and LC_TIME, not LC_CTYPE, and the user might not
2599 ;; want to set them to the same value as LC_CTYPE.
2600 (when locale-name
2601 (setq system-messages-locale locale)
2602 (setq system-time-locale locale))
2604 (if (string-match "^[a-z][a-z]" locale)
2605 ;; The value of 'current-iso639-language' is matched against
2606 ;; the ':lang' property of font-spec objects when selecting
2607 ;; and prioritizing available fonts for displaying
2608 ;; characters; see fontset.c.
2609 (setq current-iso639-language
2610 ;; The call to 'downcase' is for w32, where the
2611 ;; MS-Windows locale names are in caps, as in "ENU",
2612 ;; the equivalent of the Posix "en_US". Since the
2613 ;; match mentioned above uses memq, and ':lang'
2614 ;; properties have lower-case values, the letter-case
2615 ;; must match exactly.
2616 (intern (downcase (match-string 0 locale))))))
2618 (setq woman-locale
2619 (or system-messages-locale
2620 (let ((msglocale (getenv "LC_MESSAGES" frame)))
2621 (if (zerop (length msglocale))
2622 locale
2623 (locale-translate msglocale)))))
2625 (when locale
2626 (setq locale (downcase locale))
2628 (let ((language-name
2629 (locale-name-match locale locale-language-names))
2630 (charset-language-name
2631 (locale-name-match locale locale-charset-language-names))
2632 (default-eol-type (coding-system-eol-type
2633 (default-value 'buffer-file-coding-system)))
2634 (coding-system
2635 (or (locale-name-match locale locale-preferred-coding-systems)
2636 (when locale
2638 (locale-charset-to-coding-system
2639 (match-string 1 locale)))))))
2641 (if (consp language-name)
2642 ;; locale-language-names specify both lang-env and coding.
2643 ;; But, what specified in locale-preferred-coding-systems
2644 ;; has higher priority.
2645 (setq coding-system (or coding-system
2646 (nth 1 language-name))
2647 language-name (car language-name))
2648 ;; Otherwise, if locale is not listed in locale-language-names,
2649 ;; use what listed in locale-charset-language-names.
2650 (if (not language-name)
2651 (setq language-name charset-language-name)))
2653 ;; If a specific EOL conversion was specified in the default
2654 ;; buffer-file-coding-system, preserve it in the coding system
2655 ;; we will be using from now on.
2656 (if (and (memq default-eol-type '(0 1 2 unix dos mac))
2657 coding-system
2658 (coding-system-p coding-system))
2659 (setq coding-system (coding-system-change-eol-conversion
2660 coding-system default-eol-type)))
2662 (when language-name
2664 ;; Set up for this character set. This is now the right way
2665 ;; to do it for both unibyte and multibyte modes.
2666 (unless frame
2667 (set-language-environment language-name))
2669 ;; If the default enable-multibyte-characters is nil,
2670 ;; we are using single-byte characters,
2671 ;; so the display table and terminal coding system are irrelevant.
2672 (when (default-value 'enable-multibyte-characters)
2673 (set-display-table-and-terminal-coding-system
2674 language-name coding-system frame))
2676 ;; Set the `keyboard-coding-system' if appropriate (tty
2677 ;; only). At least X and MS Windows can generate
2678 ;; multilingual input.
2679 ;; XXX This was disabled unless `window-system', but that
2680 ;; leads to buggy behavior when a tty frame is opened
2681 ;; later. Setting the keyboard coding system has no adverse
2682 ;; effect on X, so let's do it anyway. -- Lorentey
2683 (let ((kcs (or coding-system
2684 (car (get-language-info language-name
2685 'coding-system)))))
2686 (if kcs (set-keyboard-coding-system kcs frame)))
2688 (unless frame
2689 (setq locale-coding-system
2690 (car (get-language-info language-name 'coding-priority)))))
2692 (when (and (not frame)
2693 coding-system
2694 (not (coding-system-equal coding-system
2695 locale-coding-system)))
2696 (prefer-coding-system coding-system)
2697 ;; Fixme: perhaps prefer-coding-system should set this too.
2698 ;; But it's not the time to do such a fundamental change.
2699 (setq default-sendmail-coding-system coding-system)
2700 (setq locale-coding-system coding-system))))
2702 ;; On Windows, override locale-coding-system,
2703 ;; default-file-name-coding-system, keyboard-coding-system,
2704 ;; terminal-coding-system with the ANSI or console codepage.
2705 (when (and (eq system-type 'windows-nt)
2706 (boundp 'w32-ansi-code-page))
2707 (let* ((ansi-code-page-coding
2708 (intern (format "cp%d" w32-ansi-code-page)))
2709 (code-page-coding
2710 (if noninteractive
2711 (intern (format "cp%d" (w32-get-console-codepage)))
2712 ansi-code-page-coding))
2713 (output-coding
2714 (if noninteractive
2715 (intern (format "cp%d" (w32-get-console-output-codepage)))
2716 code-page-coding)))
2717 (when (coding-system-p code-page-coding)
2718 (or output-coding (setq output-coding code-page-coding))
2719 (unless frame (setq locale-coding-system code-page-coding))
2720 (set-keyboard-coding-system code-page-coding frame)
2721 (set-terminal-coding-system output-coding frame)
2722 (setq default-file-name-coding-system ansi-code-page-coding))))
2724 (when (eq system-type 'darwin)
2725 ;; On Darwin, file names are always encoded in utf-8, no matter
2726 ;; the locale.
2727 (setq default-file-name-coding-system 'utf-8)
2728 ;; macOS's Terminal.app by default uses utf-8 regardless of
2729 ;; the locale.
2730 (when (and (null window-system)
2731 (equal (getenv "TERM_PROGRAM" frame) "Apple_Terminal"))
2732 (set-terminal-coding-system 'utf-8)
2733 (set-keyboard-coding-system 'utf-8)))
2735 ;; Default to A4 paper if we're not in a C, POSIX or US locale.
2736 ;; (See comments in Flocale_info.)
2737 (unless frame
2738 (let ((paper (locale-info 'paper))
2739 locale)
2740 (if paper
2741 ;; This will always be null at the time of writing.
2742 (cond
2743 ((equal paper '(216 279))
2744 (setq ps-paper-type 'letter))
2745 ((equal paper '(210 297))
2746 (setq ps-paper-type 'a4)))
2747 (let ((vars '("LC_ALL" "LC_PAPER" "LANG")))
2748 (while (and vars (= 0 (length locale)))
2749 (setq locale (getenv (pop vars) frame))))
2750 (when locale
2751 ;; As of glibc 2.2.5, these are the only US Letter locales,
2752 ;; and the rest are A4.
2753 (setq ps-paper-type
2754 (or (locale-name-match locale '(("c$" . letter)
2755 ("posix$" . letter)
2756 (".._us" . letter)
2757 (".._pr" . letter)
2758 (".._ca" . letter)
2759 ("enu$" . letter) ; Windows
2760 ("esu$" . letter)
2761 ("enc$" . letter)
2762 ("frc$" . letter)))
2763 'a4)))))))
2764 nil)
2765 \f
2766 ;;; Character property
2768 (put 'char-code-property-table 'char-table-extra-slots 5)
2770 (defun define-char-code-property (name table &optional docstring)
2771 "Define NAME as a character code property given by TABLE.
2772 TABLE is a char-table of purpose `char-code-property-table' with
2773 these extra slots:
2774 1st: NAME.
2775 2nd: Function to call to get a property value of a character.
2776 It is called with three arguments CHAR, VAL, and TABLE, where
2777 CHAR is a character, VAL is the value of (aref TABLE CHAR).
2778 3rd: Function to call to put a property value of a character.
2779 It is called with the same arguments as above.
2780 4th: Function to call to get a description string of a property value.
2781 It is called with one argument VALUE, a property value.
2782 5th: Data used by the above functions.
2784 TABLE may be a name of file to load to build a char-table. The
2785 file should contain a call of `define-char-code-property' with a
2786 char-table of the above format as the argument TABLE.
2788 TABLE may also be nil, in which case no property value is pre-assigned.
2790 Optional 3rd argument DOCSTRING is a documentation string of the property.
2792 See also the documentation of `get-char-code-property' and
2793 `put-char-code-property'."
2794 (or (symbolp name)
2795 (error "Not a symbol: %s" name))
2796 (if (char-table-p table)
2797 (or (and (eq (char-table-subtype table) 'char-code-property-table)
2798 (eq (char-table-extra-slot table 0) name))
2799 (error "Invalid char-table: %s" table))
2800 (or (stringp table)
2801 (error "Not a char-table nor a file name: %s" table)))
2802 (if (stringp table) (setq table (purecopy table)))
2803 (setf (alist-get name char-code-property-alist) table)
2804 (put name 'char-code-property-documentation (purecopy docstring)))
2806 (defvar char-code-property-table
2807 (make-char-table 'char-code-property-table)
2808 "Char-table containing a property list of each character code.
2809 This table is used for properties not listed in `char-code-property-alist'.
2810 See also the documentation of `get-char-code-property' and
2811 `put-char-code-property'.")
2813 (defun get-char-code-property (char propname)
2814 "Return the value of CHAR's PROPNAME property."
2815 (let ((table (unicode-property-table-internal propname)))
2816 (if table
2817 (let ((func (char-table-extra-slot table 1)))
2818 (if (functionp func)
2819 (funcall func char (aref table char) table)
2820 (get-unicode-property-internal table char)))
2821 (plist-get (aref char-code-property-table char) propname))))
2823 (defun put-char-code-property (char propname value)
2824 "Store CHAR's PROPNAME property with VALUE.
2825 It can be retrieved with `(get-char-code-property CHAR PROPNAME)'."
2826 (let ((table (unicode-property-table-internal propname)))
2827 (if table
2828 (let ((func (char-table-extra-slot table 2)))
2829 (if (functionp func)
2830 (funcall func char value table)
2831 (put-unicode-property-internal table char value)))
2832 (let* ((plist (aref char-code-property-table char))
2833 (x (plist-put plist propname value)))
2834 (or (eq x plist)
2835 (aset char-code-property-table char x))))
2836 value))
2838 (defun char-code-property-description (prop value)
2839 "Return a description string of character property PROP's value VALUE.
2840 If there's no description string for VALUE, return nil."
2841 (let ((table (unicode-property-table-internal prop)))
2842 (if table
2843 (let ((func (char-table-extra-slot table 3)))
2844 (if (functionp func)
2845 (funcall func value))))))
2847 \f
2848 ;; Pretty description of encoded string
2850 ;; Alist of ISO 2022 control code vs the corresponding mnemonic string.
2851 (defconst iso-2022-control-alist
2859 (defun encoded-string-description (str coding-system)
2860 "Return a pretty description of STR that is encoded by CODING-SYSTEM."
2861 (setq str (string-as-unibyte str))
2862 (mapconcat
2863 (if (and coding-system (eq (coding-system-type coding-system) 'iso-2022))
2864 ;; Try to get a pretty description for ISO 2022 escape sequences.
2865 (function (lambda (x) (or (cdr (assq x iso-2022-control-alist))
2866 (format "#x%02X" x))))
2867 (function (lambda (x) (format "#x%02X" x))))
2868 str " "))
2870 (defun encode-coding-char (char coding-system &optional charset)
2871 "Encode CHAR by CODING-SYSTEM and return the resulting string.
2872 If CODING-SYSTEM can't safely encode CHAR, return nil.
2873 The 3rd optional argument CHARSET, if non-nil, is a charset preferred
2874 on encoding."
2875 (let* ((str1 (string-as-multibyte (string char)))
2876 (str2 (string-as-multibyte (string char char)))
2877 (found (find-coding-systems-string str1))
2878 enc1 enc2 i1 i2)
2879 (if (and (consp found)
2880 (eq (car found) 'undecided))
2881 str1
2882 (when (memq (coding-system-base coding-system) found)
2883 ;; We must find the encoded string of CHAR. But, just encoding
2884 ;; CHAR will put extra control sequences (usually to designate
2885 ;; ASCII charset) at the tail if type of CODING is ISO 2022.
2886 ;; To exclude such tailing bytes, we at first encode one-char
2887 ;; string and two-char string, then check how many bytes at the
2888 ;; tail of both encoded strings are the same.
2890 (when charset
2891 (put-text-property 0 1 'charset charset str1)
2892 (put-text-property 0 2 'charset charset str2))
2893 (setq enc1 (encode-coding-string str1 coding-system)
2894 i1 (length enc1)
2895 enc2 (encode-coding-string str2 coding-system)
2896 i2 (length enc2))
2897 (while (and (> i1 0) (= (aref enc1 (1- i1)) (aref enc2 (1- i2))))
2898 (setq i1 (1- i1) i2 (1- i2)))
2900 ;; Now (substring enc1 i1) and (substring enc2 i2) are the same,
2901 ;; and they are the extra control sequences at the tail to
2902 ;; exclude.
2903 (substring enc2 0 i2)))))
2905 ;; Backwards compatibility. These might be better with :init-value t,
2906 ;; but that breaks loadup.
2907 (define-minor-mode unify-8859-on-encoding-mode
2908 "Exists only for backwards compatibility."
2909 :group 'mule
2910 :global t)
2911 ;; Doc said "obsolete" in 23.1, this statement only added in 24.1.
2912 (make-obsolete 'unify-8859-on-encoding-mode "don't use it." "23.1")
2914 (define-minor-mode unify-8859-on-decoding-mode
2915 "Exists only for backwards compatibility."
2916 :group 'mule
2917 :global t)
2918 ;; Doc said "obsolete" in 23.1, this statement only added in 24.1.
2919 (make-obsolete 'unify-8859-on-decoding-mode "don't use it." "23.1")
2921 (defvar nonascii-insert-offset 0)
2922 (make-obsolete-variable 'nonascii-insert-offset "do not use it." "23.1")
2923 (defvar nonascii-translation-table nil)
2924 (make-obsolete-variable 'nonascii-translation-table "do not use it." "23.1")
2926 (defvar ucs-names nil
2927 "Alist of cached (CHAR-NAME . CHAR-CODE) pairs.")
2929 (defun ucs-names ()
2930 "Return alist of (CHAR-NAME . CHAR-CODE) pairs cached in `ucs-names'."
2931 (or ucs-names
2932 (let ((ranges
2933 '((#x0000 . #x33FF)
2934 ;; (#x3400 . #x4DBF) CJK Ideographs Extension A
2935 (#x4DC0 . #x4DFF)
2936 ;; (#x4E00 . #x9FFF) CJK Unified Ideographs
2937 (#xA000 . #xD7FF)
2938 ;; (#xD800 . #xFAFF) Surrogate/Private
2939 (#xFB00 . #x134FF)
2940 ;; (#x13500 . #x143FF) unused
2941 (#x14400 . #x14646)
2942 ;; (#x14647 . #x167FF) unused
2943 (#x16800 . #x16F9F)
2944 (#x16FE0 . #x16FE0)
2945 ;; (#x17000 . #x187FF) Tangut Ideographs
2946 ;; (#x18800 . #x18AFF) Tangut Components
2947 ;; (#x18B00 . #x1AFFF) unused
2948 (#x1B000 . #x1B0FF)
2949 ;; (#x1B100 . #x1BBFF) unused
2950 (#x1BC00 . #x1BCAF)
2951 ;; (#x1BCB0 . #x1CFFF) unused
2952 (#x1D000 . #x1FFFF)
2953 ;; (#x20000 . #xDFFFF) CJK Ideograph Extension A, B, etc, unused
2954 (#xE0000 . #xE01FF)))
2955 (gc-cons-threshold 10000000)
2956 names)
2957 (dolist (range ranges)
2958 (let ((c (car range))
2959 (end (cdr range)))
2960 (while (<= c end)
2961 (let ((new-name (get-char-code-property c 'name))
2962 (old-name (get-char-code-property c 'old-name)))
2963 ;; In theory this code could end up pushing an "old-name" that
2964 ;; shadows a "new-name" but in practice every time an
2965 ;; `old-name' conflicts with a `new-name', the newer one has a
2966 ;; higher code, so it gets pushed later!
2967 (if new-name (push (cons new-name c) names))
2968 (if old-name (push (cons old-name c) names))
2969 (setq c (1+ c))))))
2970 ;; Special case for "BELL" which is apparently the only char which
2971 ;; doesn't have a new name and whose old-name is shadowed by a newer
2972 ;; char with that name.
2973 (setq ucs-names `(("BELL (BEL)" . 7) ,@names)))))
2975 (defun mule--ucs-names-annotation (name)
2976 ;; FIXME: It would be much better to add this annotation before rather than
2977 ;; after the char name, so the annotations are aligned.
2978 ;; FIXME: The default behavior of displaying annotations in italics
2979 ;; doesn't work well here.
2980 (let ((char (assoc name ucs-names)))
2981 (when char (format " (%c)" (cdr char)))))
2983 (defun char-from-name (string &optional ignore-case)
2984 "Return a character as a number from its Unicode name STRING.
2985 If optional IGNORE-CASE is non-nil, ignore case in STRING.
2986 Return nil if STRING does not name a character."
2987 (or (cdr (assoc-string string (ucs-names) ignore-case))
2989 (when minus
2990 ;; Parse names like "VARIATION SELECTOR-17" and "CJK
2991 ;; COMPATIBILITY IDEOGRAPH-F900" that are not in ucs-names.
2992 (ignore-errors
2993 (let* ((case-fold-search ignore-case)
2995 (minus-num (string-to-number (substring string minus)
2996 (if vs 10 16)))
2997 (vs-offset (if vs (if (< minus-num -16) #xE00EF #xFDFF) 0))
2998 (code (- vs-offset minus-num))
2999 (name (get-char-code-property code 'name)))
3000 (when (eq t (compare-strings string nil nil name nil nil
3001 ignore-case))
3002 code)))))))
3004 (defun read-char-by-name (prompt)
3005 "Read a character by its Unicode name or hex number string.
3006 Display PROMPT and read a string that represents a character by its
3007 Unicode property `name' or `old-name'.
3009 This function returns the character as a number.
3011 You can type a few of the first letters of the Unicode name and
3012 use completion. If you type a substring of the Unicode name
3013 preceded by an asterisk `*' and use completion, it will show all
3014 the characters whose names include that substring, not necessarily
3015 at the beginning of the name.
3021 as names, not numbers."
3022 (let* ((enable-recursive-minibuffers t)
3023 (completion-ignore-case t)
3024 (input
3025 (completing-read
3026 prompt
3027 (lambda (string pred action)
3028 (if (eq action 'metadata)
3029 '(metadata
3030 (annotation-function . mule--ucs-names-annotation)
3031 (category . unicode-name))
3032 (complete-with-action action (ucs-names) string pred)))))
3033 (char
3034 (cond
3035 ((char-from-name input t))
3037 (ignore-errors (string-to-number input 16)))
3039 input)
3040 (ignore-errors (read input))))))
3041 (unless (characterp char)
3042 (error "Invalid character"))
3043 char))
3045 (define-obsolete-function-alias 'ucs-insert 'insert-char "24.3")
3048 ;;; mule-cmds.el ends here