| blob | f83e64e3921028faa6938243b60c6f9462072325 |
1 ;;; mule-cmds.el --- commands for multilingual environment -*-coding: utf-8 -*-
3 ;; Copyright (C) 1997-2014 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:
40 ;;; MULE related key bindings and menus.
55 map)
58 ;; Keep "C-x C-m ..." for mule specific commands.
64 [Default] '(menu-item "Default" describe-specified-language-support))
65 map))
67 (defvar setup-language-environment-map
68 (let ((map (make-sparse-keymap "Set Language Environment")))
69 (bindings--define-key map
71 map))
104 revert-buffer-with-coding-system
105 :enable buffer-file-name
114 map))
131 ,describe-language-environment-map
155 map)
158 ;; This should be a single character key binding because users use it
159 ;; very frequently while editing multilingual text. Now we can use
160 ;; only two such keys: "\C-\\" and "\C-^", but the latter is not
161 ;; convenient because it requires shifting on most keyboards. An
162 ;; alternative is "\C-\]" which is now bound to `abort-recursive-edit'
163 ;; but it won't be used that frequently.
166 ;; This is no good because people often type Shift-SPC
167 ;; meaning to type SPC. -- rms.
168 ;; ;; Here's an alternative key binding for X users (Shift-SPACE).
169 ;; (define-key global-map [?\S- ] 'toggle-input-method)
171 ;;; Mule related hyperlinks.
179 ;; Note starting with word-syntax character:
183 "Return a coding system which differs from CODING-SYSTEM in EOL conversion.
184 The returned coding system converts end-of-line by EOL-TYPE
185 but text as the same way as CODING-SYSTEM.
186 EOL-TYPE should be `unix', `dos', `mac', or nil.
187 If EOL-TYPE is nil, the returned coding system detects
188 how end-of-line is formatted automatically while decoding.
190 EOL-TYPE can be specified by an integer 0, 1, or 2.
191 They means `unix', `dos', and `mac' respectively."
197 ;; We call `coding-system-base' before `coding-system-eol-type',
198 ;; because the coding-system may not be initialized until then.
203 coding-system
206 base)
208 coding-system)
214 "Return a coding system which differs from CODING-SYSTEM in text conversion.
215 The returned coding system converts text by CODING
216 but end-of-line as the same way as CODING-SYSTEM.
217 If CODING is nil, the returned coding system detects
218 how text is formatted automatically while decoding."
224 ;; Canonicalize the coding system name NAME by removing some prefixes
225 ;; and delimiter characters. Support function of
226 ;; coding-system-from-name.
229 ;; "ms950", "ibm950", "windows-950" -> "cp950"
232 ;; "iso-8859-1" -> "8859-1", "iso-2022-jp" ->"2022-jp"
235 ;; Delete "-", "_", " ", "/" but do distinguish "16-be" and "16be".
243 name)))
246 "Return a coding system whose name matches with NAME (string or symbol)."
251 sym
260 name)
262 elt eol-type)
263 elt)))))))))
266 "Change whether this buffer uses multibyte characters.
267 With ARG, use multibyte characters if the ARG is positive.
269 Note that this command does not convert the byte contents of
270 the buffer; it only changes the way those bytes are interpreted.
271 In general, therefore, this command *changes* the sequence of
272 characters that the current buffer contains.
274 We suggest you avoid using this command unless you know what you are
275 doing. If you use it by mistake, and the buffer is now displayed
276 wrong, use this command again to toggle back to the right mode."
285 "Display the HELLO file, which lists many languages and characters."
287 ;; We have to decode the file in any environment.
292 "Execute an I/O command using the specified coding system."
297 buffer-file-coding-system)))
302 default))))
306 prefix)
307 ;; read-key-sequence ignores quit, so make an explicit check.
308 ;; Like many places, this assumes quit == C-g, but it need not be.
314 ;; Process keys bound in `universal-argument-map'.
320 ;; Have to bind `last-command-event' here so that
321 ;; `digit-argument', for instance, can compute the
322 ;; prefix arg.
326 ;; This is the final call to `universal-argument-other-key', which
327 ;; set's the final `prefix-arg.
331 ;; Read the command to execute with the given prefix arg.
344 "Set default value of various coding systems to CODING-SYSTEM.
345 This sets the following coding systems:
346 o coding system of a newly created buffer
347 o default coding system for subprocess I/O
348 This also sets the following values:
349 o default value used as `file-name-coding-system' for converting file names
350 if CODING-SYSTEM is ASCII-compatible
351 o default value for the command `set-terminal-coding-system'
352 o default value for the command `set-keyboard-coding-system'
353 if CODING-SYSTEM is ASCII-compatible"
358 ;; The file-name coding system on Darwin systems is always utf-8.
365 ;; Prevent default-terminal-coding-system from converting ^M to ^J.
368 ;; Preserve eol-type from existing default-process-coding-systems.
369 ;; On non-unix-like systems in particular, these may have been set
370 ;; carefully by the user, or by the startup code, to deal with the
371 ;; users shell appropriately, so should not be altered by changing
372 ;; language environment.
383 "Add CODING-SYSTEM at the front of the priority list for automatic detection.
384 This also sets the following coding systems:
385 o coding system of a newly created buffer
386 o default coding system for subprocess I/O
387 This also sets the following values:
388 o default value used as `file-name-coding-system' for converting file names
389 o default value for the command `set-terminal-coding-system'
390 o default value for the command `set-keyboard-coding-system'
392 If CODING-SYSTEM specifies a certain type of EOL conversion, the coding
393 systems set by this function will use that type of EOL conversion.
395 A coding system that requires automatic detection of text+encoding
396 \(e.g. undecided, unix) can't be preferred.
398 To prefer, for instance, utf-8, say the following:
400 \(prefer-coding-system 'utf-8)"
412 base coding-system)))
413 ;; If they asked for specific EOL conversion, honor that.
420 (message "The default value of `file-name-coding-system' was not changed because the specified coding system is not suitable for file names.")))))
423 "If non-nil, a predicate function to sort coding systems.
425 It is called with two coding systems, and should return t if the first
431 "Sort coding system list CODINGS by a priority of each coding system.
432 Return the sorted list. CODINGS is modified by side effects.
434 If a coding system is most preferred, it has the highest priority.
435 Otherwise, coding systems that correspond to MIME charsets have
436 higher priorities. Among them, a coding system included in the
437 `coding-system' key of the current language environment has higher
438 priority. See also the documentation of `language-info-alist'.
440 If the variable `sort-coding-systems-predicate' (which see) is
441 non-nil, it is used to sort CODINGS instead."
451 ;; We calculate the priority number 0..255 by
452 ;; using the 8 bits PMMLCEII as this:
453 ;; P: 1 if most preferred.
454 ;; MM: greater than 0 if mime-charset.
455 ;; L: 1 if one of the current lang. env.'s codings.
456 ;; C: 1 if one of codings listed in the category list.
457 ;; E: 1 if not XXX-with-esc
458 ;; II: if iso-2022 based, 0..3, else 1.
463 ;; Prefer coding systems corresponding to a
464 ;; MIME charset.
466 ;; Lower utf-16 priority so that we
467 ;; normally prefer utf-8 to it, and put
468 ;; x-ctext below that.
484 ;; For ISO based coding systems, prefer
485 ;; one that doesn't use designation nor
486 ;; locking/single shifting.
497 ))))))
502 "Return a list of proper coding systems to encode a text between FROM and TO.
504 If FROM is a string, find coding systems in that instead of the buffer.
505 All coding systems in the list can safely encode any multibyte characters
506 in the text.
508 If the text contains no multibyte characters, return a list of a single
509 element `undecided'."
512 ;; The text contains only ASCII characters. Any coding
513 ;; systems are safe.
515 ;; We need copy-sequence because sorting will alter the argument.
519 "Return a list of proper coding systems to encode STRING.
520 All coding systems in the list can safely encode any multibyte characters
521 in STRING.
523 If STRING contains no multibyte characters, return a list of a single
524 element `undecided'."
528 "Return a list of proper coding systems to encode characters of CHARSETS.
529 CHARSETS is a list of character sets.
531 This only finds coding systems of type `charset', whose
532 `:charset-list' property includes all of CHARSETS (plus `ascii' for
533 ASCII-compatible coding systems). It was used in older versions of
534 Emacs, but is unlikely to be what you really want now."
535 ;; Deal with aliases.
538 charsets))
559 t))
564 "Find multibyte characters in the region specified by FROM and TO.
565 If FROM is a string, find multibyte characters in the string.
566 The return value is an alist of the following format:
567 ((CHARSET COUNT CHAR ...) ...)
568 where
569 CHARSET is a character set,
570 COUNT is a number of characters,
571 CHARs are the characters found from the character set.
572 Optional 3rd arg MAXCOUNT limits how many CHARs are put in the above list.
573 Optional 4th arg EXCLUDES is a list of character sets to be ignored."
575 charset char)
610 "Search forward from point for a character that is not encodable.
611 It asks which coding system to check.
612 If such a character is found, set point after that character.
613 Otherwise, don't move point.
615 When called from a program, the value is the position of the unencodable
616 character found, or nil if all characters are encodable."
621 default))))
626 pos))
629 "Most recent coding system explicitly specified by the user when asked.
630 This variable is set whenever Emacs asks the user which coding system
631 to use in order to write a file. If you set it to nil explicitly,
632 then call `write-region', then afterward this variable will be non-nil
636 "If non-nil, a function to control the behavior of coding system selection.
637 The meaning is the same as the argument ACCEPT-DEFAULT-P of the
638 function `select-safe-coding-system' (which see). This variable
642 "Return a list of coding systems presumably more user-friendly than CODINGS."
643 ;; Change each safe coding system to the corresponding
644 ;; mime-charset name if it is also a coding system. Such a name
645 ;; is more friendly to users.
651 mime-charset cs)))
652 codings))
654 ;; Don't offer variations with locking shift, which you
655 ;; basically never want.
664 ;; Remove raw-text, emacs-mule and no-conversion unless nothing
665 ;; else is available.
669 codings))
673 "Select interactively a coding system for the region FROM ... TO.
674 FROM can be a string, as in `write-region'.
675 CODINGS is the list of base coding systems known to be safe for this region,
676 typically obtained with `find-coding-systems-region'.
677 UNSAFE is a list of coding systems known to be unsafe for this region.
678 REJECTED is a list of coding systems which were safe but for some reason
679 were not recommended in the particular context.
680 DEFAULT is the coding system to use by default in the query."
681 ;; At first, if some defaults are unsafe, record at most 11
682 ;; problematic characters and their positions for them by turning
683 ;; (CODING ...)
684 ;; into
685 ;; ((CODING (POS . CHAR) (POS . CHAR) ...) ...)
700 unsafe)))
706 coding-system)
708 ;; If some defaults are unsafe, make sure the offending
709 ;; buffer is displayed.
713 unsafe))))
714 ;; Then ask users to select one from CODINGS while showing
715 ;; the reason why none of the defaults are not used.
724 "These default coding systems were tried to encode"
739 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)
1493 'inactivate-input-method
1497 "Select and activate input method INPUT-METHOD for the current buffer.
1498 This also sets the default input method to the one you specify.
1499 If INPUT-METHOD is nil, this function turns off the input method, and
1500 also causes you to be prompted for a name of an input method the next
1502 When called interactively, the optional arg INTERACTIVE is non-nil,
1503 which marks the variable `default-input-method' as set for Custom buffers.
1506 To deactivate it programmatically, use `deactivate-input-method'."
1511 default t)
1512 t)))
1517 default-input-method)
1523 "Enable or disable multilingual text input method for the current buffer.
1524 Only one input method can be enabled at any time in a given buffer.
1526 The normal action is to enable an input method if none was enabled,
1527 and disable the current one otherwise. Which input method to enable
1528 can be determined in various ways--either the one most recently used,
1529 or the one specified by `default-input-method', or as a last resort
1530 by reading the name of an input method in the minibuffer.
1532 With a prefix argument ARG, read an input method name with the minibuffer
1533 and enable that one. The default is the most recent input method specified
1534 \(not including the currently active input method, if any).
1536 When called interactively, the optional argument INTERACTIVE is non-nil,
1537 which marks the variable `default-input-method' as set for Custom buffers."
1554 default t))
1555 default))
1565 "Describe input method INPUT-METHOD."
1595 "Describe the input method currently in use.
1596 This is a subroutine for `describe-input-method'."
1602 current-input-method)
1607 "Read a multilingual string from minibuffer, prompting with string PROMPT.
1608 The input method selected last time is activated in minibuffer.
1609 If optional second argument INITIAL-INPUT is non-nil, insert it in the
1610 minibuffer initially.
1611 Optional 3rd argument INPUT-METHOD specifies the input method to be activated
1612 instead of the one selected last time. It is a symbol or a string."
1615 current-input-method
1616 default-input-method
1627 ;; Variables to control behavior of input methods. All input methods
1628 ;; should react to these variables.
1631 "A flag to control extra guidance given by input methods.
1632 The value should be nil, t, `complex-only', or `default'.
1634 The extra guidance is done by showing list of available keys in echo
1635 area. When you use the input method in the minibuffer, the guidance
1636 is shown at the bottom short window (split from the existing window).
1638 If the value is t, extra guidance is always given, if the value is
1639 nil, extra guidance is always suppressed.
1641 If the value is `complex-only', only complex input methods such as
1642 `chinese-py' and `japanese' give extra guidance.
1644 If the value is `default', complex input methods always give extra
1645 guidance, but simple input methods give it only when you are not in
1646 the minibuffer.
1648 See also the variable `input-method-highlight-flag'."
1654 "If this flag is non-nil, input methods highlight partially-entered text.
1655 For instance, while you are in the middle of a Quail input method sequence,
1656 the text inserted so far is temporarily underlined.
1657 The underlining goes away when you finish or abort the input method sequence.
1658 See also the variable `input-method-verbose-flag'."
1663 "Normal hook run just after an input method is activated.
1665 The variable `current-input-method' keeps the input method name
1666 just activated."
1671 'input-method-inactivate-hook
1675 "Normal hook run just after an input method is deactivated.
1677 The variable `current-input-method' still keeps the input method name
1678 just deactivated."
1684 "Normal hook run just after an input method insert some chunk of text."
1689 "This flag controls when an input method returns.
1690 Usually, the input method does not return while there's a possibility
1691 that it may find a different translation if a user types another key.
1692 But, if this flag is non-nil, the input method returns as soon as the
1696 "This flag controls how an input method shows an intermediate key sequence.
1697 Usually, the input method inserts the intermediate key sequence,
1698 or candidate translations corresponding to the sequence,
1699 at point in the current buffer.
1700 But, if this flag is non-nil, it displays them in echo area instead."
1705 "This flag controls the behavior of an input method on invalid key input.
1706 Usually, when a user types a key which doesn't start any character
1707 handled by the input method, the key is handled by turning off the
1708 input method temporarily. After that key, the input method is re-enabled.
1711 \f
1713 "Normal hook run after some language environment is set.
1715 When you set some hook function here, that effect usually should not
1716 be inherited to another language environment. So, you had better set
1717 another function in `exit-language-environment-hook' (which see) to
1718 cancel the effect."
1723 "Normal hook run after exiting from some language environment.
1724 When this hook is run, the variable `current-language-environment'
1725 is still bound to the language environment being exited.
1727 This hook is mainly used for canceling the effect of
1728 `set-language-environment-hook' (which see)."
1735 "Switch to a specified language environment."
1748 "The last language environment specified with `set-language-environment'.
1750 to using the function `set-language-environment'."
1757 current-language-environment)
1758 language-info-alist t))
1760 ;; custom type will be updated with `set-language-info'.
1771 "Reset multilingual environment of Emacs to the default status.
1773 The default status is as follows:
1775 The default value of `buffer-file-coding-system' is nil.
1776 The default coding system for process I/O is nil.
1777 The default value for the command `set-terminal-coding-system' is nil.
1778 The default value for the command `set-keyboard-coding-system' is nil.
1780 The order of priorities of coding systems are as follows:
1781 utf-8
1782 iso-2022-7bit
1783 iso-latin-1
1784 iso-2022-7bit-lock
1785 iso-2022-8bit-ss2
1786 emacs-mule
1787 raw-text"
1789 ;; This function formerly set default-enable-multibyte-characters to t,
1790 ;; but that is incorrect. It should not alter the unibyte/multibyte choice.
1793 'utf-8
1794 'iso-2022-7bit
1795 'iso-latin-1
1796 'iso-2022-7bit-lock
1797 'iso-2022-8bit-ss2
1798 'emacs-mule
1803 ;; On Darwin systems, this should be utf-8, but when this file is loaded
1804 ;; utf-8 is not yet defined, so we set it in set-locale-environment instead.
1806 ;; Preserve eol-type from existing default-process-coding-systems.
1807 ;; On non-unix-like systems in particular, these may have been set
1808 ;; carefully by the user, or by the startup code, to deal with the
1809 ;; users shell appropriately, so should not be altered by changing
1810 ;; language environment.
1812 ;; When bootstrapping, coding-systems are not defined yet, so
1813 ;; we need to catch the error from check-coding-system.
1826 ;; Put the highest priority to the charset iso-8859-1 to prefer the
1827 ;; registry iso8859-1 over iso8859-2 in font selection. It also
1828 ;; makes unibyte-display-via-language-environment to use iso-8859-1
1829 ;; as the unibyte charset.
1832 ;; Don't alter the terminal and keyboard coding systems here.
1833 ;; The terminal still supports the same coding system
1834 ;; that it supported a minute ago.
1835 ;; (set-terminal-coding-system-internal nil)
1836 ;; (set-keyboard-coding-system-internal nil)
1838 ;; Back in Emacs-20, it was necessary to provide some fallback implicit
1839 ;; conversion, because almost no packages handled coding-system issues.
1840 ;; Nowadays it'd just paper over bugs.
1841 ;; (set-unibyte-charset 'iso-8859-1)
1842 )
1846 (defun set-display-table-and-terminal-coding-system (language-name &optional coding-system display)
1847 "Set up the display table and terminal coding system for LANGUAGE-NAME."
1853 ;; The following 2 lines undo the 8-bit display that we set up
1854 ;; in standard-display-european-internal, which see. This is in
1855 ;; case the user has used standard-display-european earlier in
1856 ;; this session.
1863 "Set up multilingual environment for using LANGUAGE-NAME.
1864 This sets the coding system priority and the default input method
1865 and sometimes other things. LANGUAGE-NAME should be a string
1867 specifies the character set for the major languages of Western Europe.
1869 If there is a prior value for `current-language-environment', this
1870 runs the hook `exit-language-environment-hook'. After setting up
1871 the new language environment, it runs `set-language-environment-hook'."
1873 nil
1890 ;; The features might set up coding systems.
1902 ;; Unibyte setups if necessary.
1912 current-iso639-language))
1918 "An Emacs charset."
1920 :completions
1930 widget))
1934 "Customizations of language environment parameters.
1935 Value is an alist with elements like those of `language-info-alist'.
1936 These are used to set values in `language-info-alist' which replace
1937 the defaults. A typical use is replacing the default input method for
1940 This option is intended for use at startup. Removing items doesn't
1941 remove them from the language info until you next restart Emacs.
1943 Setting this variable directly does not take effect.
1944 See `set-language-info-alist' for use in programs."
1949 ;; Can't do this before language environments are set up.
1951 ;; modify language-info-alist
1954 ;; re-set the environment in case its parameters changed
1958 :completions
1962 action language-info-alist string pred))))
1963 :value-type
1981 ;; Actually set up direct output of non-ASCII characters.
1983 ;; Unibyte Emacs on MS-DOS wants to display all 8-bit characters with
1984 ;; the native font, and codes 160 and 146 stand for something very
1985 ;; different there.
1989 ;; Most X fonts used to do the wrong thing for latin-1 code 160.
1991 ;; XFree86 4 has fixed the fonts.
1995 ;; Make non-line-break space display as a plain space.
1998 ;; don't contain a character at that position. Map it to the ASCII
1999 ;; apostrophe. [This is actually RIGHT SINGLE QUOTATION MARK,
2000 ;; U+2019, normally from the windows-1252 character set. XFree 4
2001 ;; fonts probably have the appropriate glyph at this position,
2002 ;; so they could use standard-display-8bit. It's better to use a
2003 ;; proper windows-1252 coding system. --fx]
2004 (aset standard-display-table (unibyte-char-to-multibyte 146) [39]))))
2006 (defun set-language-environment-coding-systems (language-name)
2007 "Do various coding system setups for language environment LANGUAGE-NAME."
2008 (let* ((priority (get-language-info language-name 'coding-priority))
2009 (default-coding (car priority))
2010 ;; If the default buffer-file-coding-system is nil, don't use
2011 ;; coding-system-eol-type, because it treats nil as
2012 ;; `no-conversion'. The default buffer-file-coding-system is set
2013 ;; to nil by reset-language-environment, and in that case we
2014 ;; want to have here the native EOL type for each platform.
2015 ;; FIXME: there should be a common code that runs both on
2016 ;; startup and here to set the default EOL type correctly.
2017 ;; Right now, DOS/Windows platforms set this on dos-w32.el,
2018 ;; which works only as long as the order of loading files at
2019 ;; dump time and calling functions at startup is not modified
2020 ;; significantly, i.e. as long as this function is called
2021 ;; _after_ the default buffer-file-coding-system was set by
2022 ;; dos-w32.el.
2023 (eol-type
2024 (coding-system-eol-type
2025 (or (default-value 'buffer-file-coding-system)
2026 (if (memq system-type '(windows-nt ms-dos)) 'dos 'unix)))))
2027 (when priority
2028 (set-default-coding-systems
2029 (if (memq eol-type '(0 1 2 unix dos mac))
2030 (coding-system-change-eol-conversion default-coding eol-type)
2031 default-coding))
2032 (setq default-sendmail-coding-system default-coding)
2033 (apply 'set-coding-system-priority priority))))
2035 (defun set-language-environment-input-method (language-name)
2036 "Do various input method setups for language environment LANGUAGE-NAME."
2037 (let ((input-method (get-language-info language-name 'input-method)))
2038 (when input-method
2039 (setq default-input-method input-method)
2040 (if input-method-history
2041 (setq input-method-history
2042 (cons input-method
2043 (delete input-method input-method-history)))))))
2045 (defun set-language-environment-nonascii-translation (language-name)
2046 "Do unibyte/multibyte translation setup for language environment LANGUAGE-NAME."
2047 ;; Note: For DOS, we assumed that the charset cpXXX is already
2048 ;; defined.
2049 (let ((nonascii (get-language-info language-name 'nonascii-translation)))
2050 (if (eq window-system 'pc)
2051 (setq nonascii (intern (format "cp%d" dos-codepage))))
2052 (or (and (charsetp nonascii)
2053 (get-charset-property nonascii :ascii-compatible-p))
2054 (setq nonascii 'iso-8859-1))
2055 ;; Back in Emacs-20, it was necessary to provide some fallback implicit
2056 ;; conversion, because almost no packages handled coding-system issues.
2057 ;; Nowadays it'd just paper over bugs.
2058 ;; (set-unibyte-charset nonascii)
2059 ))
2061 (defun set-language-environment-charset (language-name)
2062 "Do various charset setups for language environment LANGUAGE-NAME."
2063 ;; Put higher priorities to such charsets that are supported by the
2064 ;; coding systems of higher priorities in this environment.
2065 (let ((charsets (get-language-info language-name 'charset)))
2066 (dolist (coding (get-language-info language-name 'coding-priority))
2067 (let ((list (coding-system-charset-list coding)))
2068 (if (consp list)
2069 (setq charsets (append charsets list)))))
2070 (if charsets
2071 (apply 'set-charset-priority charsets))))
2073 (defun set-language-environment-unibyte (language-name)
2074 "Do various unibyte-mode setups for language environment LANGUAGE-NAME."
2075 (set-display-table-and-terminal-coding-system language-name))
2077 (defun princ-list (&rest args)
2079 (declare (obsolete "use mapc and princ instead." "23.3"))
2080 (mapc #'princ args)
2083 (put 'describe-specified-language-support 'apropos-inhibit t)
2085 ;; Print language-specific information such as input methods,
2086 ;; charsets, and coding systems. This function is intended to be
2087 ;; called from the menu:
2088 ;; [menu-bar mule describe-language-environment LANGUAGE]
2089 ;; and should not run it by `M-x describe-current-input-method-function'.
2090 (defun describe-specified-language-support ()
2091 "Describe how Emacs supports the specified language environment."
2092 (interactive)
2093 (let (language-name)
2094 (if (not (and (symbolp last-command-event)
2095 (or (not (eq last-command-event 'Default))
2096 (setq last-command-event 'English))
2097 (setq language-name (symbol-name last-command-event))))
2098 (error "This command should only be called from the menu bar"))
2099 (describe-language-environment language-name)))
2101 (defun describe-language-environment (language-name)
2102 "Describe how Emacs supports language environment LANGUAGE-NAME."
2103 (interactive
2104 (list (read-language-name
2105 'documentation
2106 "Describe language environment (default current choice): ")))
2107 (if (null language-name)
2108 (setq language-name current-language-environment))
2109 (if (or (null language-name)
2110 (null (get-language-info language-name 'documentation)))
2111 (error "No documentation for the specified language"))
2112 (if (symbolp language-name)
2113 (setq language-name (symbol-name language-name)))
2114 (dolist (feature (get-language-info language-name 'features))
2115 (require feature))
2116 (let ((doc (get-language-info language-name 'documentation)))
2117 (help-setup-xref (list #'describe-language-environment language-name)
2118 (called-interactively-p 'interactive))
2119 (with-output-to-temp-buffer (help-buffer)
2120 (with-current-buffer standard-output
2122 (if (stringp doc)
2124 (condition-case nil
2125 (let ((str (eval (get-language-info language-name 'sample-text))))
2126 (if (stringp str)
2130 (error nil))
2131 (let ((input-method (get-language-info language-name 'input-method))
2132 (l (copy-sequence input-method-alist))
2133 (first t))
2134 (when (and input-method
2135 (setq input-method (assoc input-method l)))
2137 (setq l (cons input-method (delete input-method l))
2138 first nil))
2139 (dolist (elt l)
2140 (when (or (eq input-method elt)
2141 (eq t (compare-strings language-name nil nil
2142 (nth 1 elt) nil nil t)))
2143 (when first
2145 (setq first nil))
2146 (insert " " (car elt))
2147 (search-backward (car elt))
2148 (help-xref-button 0 'help-input-method (car elt))
2149 (goto-char (point-max))
2151 (if (stringp (nth 3 elt)) (nth 3 elt) (car (nth 3 elt)))
2153 (or first
2156 (let ((l (get-language-info language-name 'charset)))
2157 (if (null l)
2159 (while l
2160 (insert " " (symbol-name (car l)))
2161 (search-backward (symbol-name (car l)))
2162 (help-xref-button 0 'help-character-set (car l))
2163 (goto-char (point-max))
2165 (setq l (cdr l)))))
2168 (let ((l (get-language-info language-name 'coding-system)))
2169 (if (null l)
2171 (while l
2172 (insert " " (symbol-name (car l)))
2173 (search-backward (symbol-name (car l)))
2174 (help-xref-button 0 'help-coding-system (car l))
2175 (goto-char (point-max))
2176 (insert " (`"
2177 (coding-system-mnemonic (car l))
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" . "Latin-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 (defun locale-translate (locale)
2522 "Expand LOCALE according to `locale-translation-file-name', if possible.
2524 (if locale-translation-file-name
2525 (with-temp-buffer
2526 (set-buffer-multibyte nil)
2527 (insert-file-contents locale-translation-file-name)
2528 (if (re-search-forward
2530 (buffer-substring (point) (line-end-position))
2531 locale))
2532 locale))
2534 (defun set-locale-environment (&optional locale-name frame)
2535 "Set up multilingual environment for using LOCALE-NAME.
2536 This sets the language environment, the coding system priority,
2537 the default input method and sometimes other things.
2539 LOCALE-NAME should be a string which is the name of a locale supported
2540 by the system. Often it is of the form xx_XX.CODE, where xx is a
2541 language, XX is a country, and CODE specifies a character set and
2543 a locale for Japanese in Japan using the `japanese-iso-8bit'
2544 coding-system. The name may also have a modifier suffix, e.g. `@euro'
2545 or `@cyrillic'.
2547 If LOCALE-NAME is nil, its value is taken from the environment
2548 variables LC_ALL, LC_CTYPE and LANG (the first one that is set).
2550 The locale names supported by your system can typically be found in a
2551 directory named `/usr/share/locale' or `/usr/lib/locale'. LOCALE-NAME
2552 will be translated according to the table specified by
2553 `locale-translation-file-name'.
2555 If FRAME is non-nil, only set the keyboard coding system and the
2556 terminal coding system for the terminal of that frame, and don't
2557 touch session-global parameters like the language environment.
2559 See also `locale-charset-language-names', `locale-language-names',
2560 `locale-preferred-coding-systems' and `locale-coding-system'."
2561 (interactive "sSet environment for locale: ")
2563 ;; Do this at runtime for the sake of binaries possibly transported
2564 ;; to a system without X.
2565 (setq locale-translation-file-name
2566 (let ((files
2567 '("/usr/share/X11/locale/locale.alias" ; e.g. X11R7
2568 "/usr/lib/X11/locale/locale.alias" ; e.g. X11R6.4
2569 "/usr/X11R6/lib/X11/locale/locale.alias" ; XFree86, e.g. RedHat 4.2
2570 "/usr/openwin/lib/locale/locale.alias" ; e.g. Solaris 2.6
2571 ;;
2572 ;; The following name appears after the X-related names above,
2573 ;; since the X-related names are what X actually uses.
2574 "/usr/share/locale/locale.alias" ; GNU/Linux sans X
2575 )))
2576 (while (and files (not (file-exists-p (car files))))
2577 (setq files (cdr files)))
2578 (car files)))
2580 (let ((locale locale-name))
2582 (unless locale
2583 ;; Use the first of these three environment variables
2584 ;; that has a nonempty value.
2585 (let ((vars '("LC_ALL" "LC_CTYPE" "LANG")))
2586 (while (and vars
2587 (= 0 (length locale))) ; nil or empty string
2588 (setq locale (getenv (pop vars) frame)))))
2590 (when locale
2591 (setq locale (locale-translate locale))
2593 ;; Leave the system locales alone if the caller did not specify
2594 ;; an explicit locale name, as their defaults are set from
2595 ;; LC_MESSAGES and LC_TIME, not LC_CTYPE, and the user might not
2596 ;; want to set them to the same value as LC_CTYPE.
2597 (when locale-name
2598 (setq system-messages-locale locale)
2599 (setq system-time-locale locale))
2601 (if (string-match "^[a-z][a-z]" locale)
2602 (setq current-iso639-language (intern (match-string 0 locale)))))
2604 (setq woman-locale
2605 (or system-messages-locale
2606 (let ((msglocale (getenv "LC_MESSAGES" frame)))
2607 (if (zerop (length msglocale))
2608 locale
2609 (locale-translate msglocale)))))
2611 (when locale
2612 (setq locale (downcase locale))
2614 (let ((language-name
2615 (locale-name-match locale locale-language-names))
2616 (charset-language-name
2617 (locale-name-match locale locale-charset-language-names))
2618 (default-eol-type (coding-system-eol-type
2619 (default-value 'buffer-file-coding-system)))
2620 (coding-system
2621 (or (locale-name-match locale locale-preferred-coding-systems)
2622 (when locale
2624 (locale-charset-to-coding-system
2625 (match-string 1 locale)))))))
2627 (if (consp language-name)
2628 ;; locale-language-names specify both lang-env and coding.
2629 ;; But, what specified in locale-preferred-coding-systems
2630 ;; has higher priority.
2631 (setq coding-system (or coding-system
2632 (nth 1 language-name))
2633 language-name (car language-name))
2634 ;; Otherwise, if locale is not listed in locale-language-names,
2635 ;; use what listed in locale-charset-language-names.
2636 (if (not language-name)
2637 (setq language-name charset-language-name)))
2639 ;; If a specific EOL conversion was specified in the default
2640 ;; buffer-file-coding-system, preserve it in the coding system
2641 ;; we will be using from now on.
2642 (if (and (memq default-eol-type '(0 1 2 unix dos mac))
2643 coding-system
2644 (coding-system-p coding-system))
2645 (setq coding-system (coding-system-change-eol-conversion
2646 coding-system default-eol-type)))
2648 (when language-name
2650 ;; Set up for this character set. This is now the right way
2651 ;; to do it for both unibyte and multibyte modes.
2652 (unless frame
2653 (set-language-environment language-name))
2655 ;; If the default enable-multibyte-characters is nil,
2656 ;; we are using single-byte characters,
2657 ;; so the display table and terminal coding system are irrelevant.
2658 (when (default-value 'enable-multibyte-characters)
2659 (set-display-table-and-terminal-coding-system
2660 language-name coding-system frame))
2662 ;; Set the `keyboard-coding-system' if appropriate (tty
2663 ;; only). At least X and MS Windows can generate
2664 ;; multilingual input.
2665 ;; XXX This was disabled unless `window-system', but that
2666 ;; leads to buggy behavior when a tty frame is opened
2667 ;; later. Setting the keyboard coding system has no adverse
2668 ;; effect on X, so let's do it anyway. -- Lorentey
2669 (let ((kcs (or coding-system
2670 (car (get-language-info language-name
2671 'coding-system)))))
2672 (if kcs (set-keyboard-coding-system kcs frame)))
2674 (unless frame
2675 (setq locale-coding-system
2676 (car (get-language-info language-name 'coding-priority)))))
2678 (when (and (not frame)
2679 coding-system
2680 (not (coding-system-equal coding-system
2681 locale-coding-system)))
2682 (prefer-coding-system coding-system)
2683 ;; Fixme: perhaps prefer-coding-system should set this too.
2684 ;; But it's not the time to do such a fundamental change.
2685 (setq default-sendmail-coding-system coding-system)
2686 (setq locale-coding-system coding-system))))
2688 ;; On Windows, override locale-coding-system,
2689 ;; default-file-name-coding-system, keyboard-coding-system,
2690 ;; terminal-coding-system with system codepage.
2691 (when (and (eq system-type 'windows-nt)
2692 (boundp 'w32-ansi-code-page))
2693 (let ((code-page-coding (intern (format "cp%d" w32-ansi-code-page))))
2694 (when (coding-system-p code-page-coding)
2695 (unless frame (setq locale-coding-system code-page-coding))
2696 (set-keyboard-coding-system code-page-coding frame)
2697 (set-terminal-coding-system code-page-coding frame)
2698 (setq default-file-name-coding-system code-page-coding))))
2700 (when (eq system-type 'darwin)
2701 ;; On Darwin, file names are always encoded in utf-8, no matter
2702 ;; the locale.
2703 (setq default-file-name-coding-system 'utf-8)
2704 ;; Mac OS X's Terminal.app by default uses utf-8 regardless of
2705 ;; the locale.
2706 (when (and (null window-system)
2707 (equal (getenv "TERM_PROGRAM" frame) "Apple_Terminal"))
2708 (set-terminal-coding-system 'utf-8)
2709 (set-keyboard-coding-system 'utf-8)))
2711 ;; Default to A4 paper if we're not in a C, POSIX or US locale.
2712 ;; (See comments in Flocale_info.)
2713 (unless frame
2714 (let ((locale locale)
2715 (paper (locale-info 'paper)))
2716 (if paper
2717 ;; This will always be null at the time of writing.
2718 (cond
2719 ((equal paper '(216 279))
2720 (setq ps-paper-type 'letter))
2721 ((equal paper '(210 297))
2722 (setq ps-paper-type 'a4)))
2723 (let ((vars '("LC_ALL" "LC_PAPER" "LANG")))
2724 (while (and vars (= 0 (length locale)))
2725 (setq locale (getenv (pop vars) frame))))
2726 (when locale
2727 ;; As of glibc 2.2.5, these are the only US Letter locales,
2728 ;; and the rest are A4.
2729 (setq ps-paper-type
2730 (or (locale-name-match locale '(("c$" . letter)
2731 ("posix$" . letter)
2732 (".._us" . letter)
2733 (".._pr" . letter)
2734 (".._ca" . letter)
2735 ("enu$" . letter) ; Windows
2736 ("esu$" . letter)
2737 ("enc$" . letter)
2738 ("frc$" . letter)))
2739 'a4)))))))
2740 nil)
2741 \f
2742 ;;; Character property
2744 (put 'char-code-property-table 'char-table-extra-slots 5)
2746 (defun define-char-code-property (name table &optional docstring)
2747 "Define NAME as a character code property given by TABLE.
2748 TABLE is a char-table of purpose `char-code-property-table' with
2749 these extra slots:
2750 1st: NAME.
2751 2nd: Function to call to get a property value of a character.
2752 It is called with three arguments CHAR, VAL, and TABLE, where
2753 CHAR is a character, VAL is the value of (aref TABLE CHAR).
2754 3rd: Function to call to put a property value of a character.
2755 It is called with the same arguments as above.
2756 4th: Function to call to get a description string of a property value.
2757 It is called with one argument VALUE, a property value.
2758 5th: Data used by the above functions.
2760 TABLE may be a name of file to load to build a char-table. The
2761 file should contain a call of `define-char-code-property' with a
2762 char-table of the above format as the argument TABLE.
2764 TABLE may also be nil, in which case no property value is pre-assigned.
2766 Optional 3rd argument DOCSTRING is a documentation string of the property.
2768 See also the documentation of `get-char-code-property' and
2769 `put-char-code-property'."
2770 (or (symbolp name)
2771 (error "Not a symbol: %s" name))
2772 (if (char-table-p table)
2773 (or (and (eq (char-table-subtype table) 'char-code-property-table)
2774 (eq (char-table-extra-slot table 0) name))
2775 (error "Invalid char-table: %s" table))
2776 (or (stringp table)
2777 (error "Not a char-table nor a file name: %s" table)))
2778 (if (stringp table) (setq table (purecopy table)))
2779 (let ((slot (assq name char-code-property-alist)))
2780 (if slot
2781 (setcdr slot table)
2782 (setq char-code-property-alist
2783 (cons (cons name table) char-code-property-alist))))
2784 (put name 'char-code-property-documentation (purecopy docstring)))
2786 (defvar char-code-property-table
2787 (make-char-table 'char-code-property-table)
2788 "Char-table containing a property list of each character code.
2789 This table is used for properties not listed in `char-code-property-alist'.
2790 See also the documentation of `get-char-code-property' and
2791 `put-char-code-property'.")
2793 (defun get-char-code-property (char propname)
2794 "Return the value of CHAR's PROPNAME property."
2795 (let ((table (unicode-property-table-internal propname)))
2796 (if table
2797 (let ((func (char-table-extra-slot table 1)))
2798 (if (functionp func)
2799 (funcall func char (aref table char) table)
2800 (get-unicode-property-internal table char)))
2801 (plist-get (aref char-code-property-table char) propname))))
2803 (defun put-char-code-property (char propname value)
2804 "Store CHAR's PROPNAME property with VALUE.
2805 It can be retrieved with `(get-char-code-property CHAR PROPNAME)'."
2806 (let ((table (unicode-property-table-internal propname)))
2807 (if table
2808 (let ((func (char-table-extra-slot table 2)))
2809 (if (functionp func)
2810 (funcall func char value table)
2811 (put-unicode-property-internal table char value)))
2812 (let* ((plist (aref char-code-property-table char))
2813 (x (plist-put plist propname value)))
2814 (or (eq x plist)
2815 (aset char-code-property-table char x))))
2816 value))
2818 (defun char-code-property-description (prop value)
2819 "Return a description string of character property PROP's value VALUE.
2820 If there's no description string for VALUE, return nil."
2821 (let ((table (unicode-property-table-internal prop)))
2822 (if table
2823 (let ((func (char-table-extra-slot table 3)))
2824 (if (functionp func)
2825 (funcall func value))))))
2827 \f
2828 ;; Pretty description of encoded string
2830 ;; Alist of ISO 2022 control code vs the corresponding mnemonic string.
2831 (defconst iso-2022-control-alist
2839 (defun encoded-string-description (str coding-system)
2840 "Return a pretty description of STR that is encoded by CODING-SYSTEM."
2841 (setq str (string-as-unibyte str))
2842 (mapconcat
2843 (if (and coding-system (eq (coding-system-type coding-system) 'iso-2022))
2844 ;; Try to get a pretty description for ISO 2022 escape sequences.
2845 (function (lambda (x) (or (cdr (assq x iso-2022-control-alist))
2846 (format "#x%02X" x))))
2847 (function (lambda (x) (format "#x%02X" x))))
2848 str " "))
2850 (defun encode-coding-char (char coding-system &optional charset)
2851 "Encode CHAR by CODING-SYSTEM and return the resulting string.
2852 If CODING-SYSTEM can't safely encode CHAR, return nil.
2853 The 3rd optional argument CHARSET, if non-nil, is a charset preferred
2854 on encoding."
2855 (let* ((str1 (string-as-multibyte (string char)))
2856 (str2 (string-as-multibyte (string char char)))
2857 (found (find-coding-systems-string str1))
2858 enc1 enc2 i1 i2)
2859 (if (and (consp found)
2860 (eq (car found) 'undecided))
2861 str1
2862 (when (memq (coding-system-base coding-system) found)
2863 ;; We must find the encoded string of CHAR. But, just encoding
2864 ;; CHAR will put extra control sequences (usually to designate
2865 ;; ASCII charset) at the tail if type of CODING is ISO 2022.
2866 ;; To exclude such tailing bytes, we at first encode one-char
2867 ;; string and two-char string, then check how many bytes at the
2868 ;; tail of both encoded strings are the same.
2870 (when charset
2871 (put-text-property 0 1 'charset charset str1)
2872 (put-text-property 0 2 'charset charset str2))
2873 (setq enc1 (encode-coding-string str1 coding-system)
2874 i1 (length enc1)
2875 enc2 (encode-coding-string str2 coding-system)
2876 i2 (length enc2))
2877 (while (and (> i1 0) (= (aref enc1 (1- i1)) (aref enc2 (1- i2))))
2878 (setq i1 (1- i1) i2 (1- i2)))
2880 ;; Now (substring enc1 i1) and (substring enc2 i2) are the same,
2881 ;; and they are the extra control sequences at the tail to
2882 ;; exclude.
2883 (substring enc2 0 i2)))))
2885 ;; Backwards compatibility. These might be better with :init-value t,
2886 ;; but that breaks loadup.
2887 (define-minor-mode unify-8859-on-encoding-mode
2888 "Exists only for backwards compatibility."
2889 :group 'mule
2890 :global t)
2891 ;; Doc said "obsolete" in 23.1, this statement only added in 24.1.
2892 (make-obsolete 'unify-8859-on-encoding-mode "don't use it." "23.1")
2894 (define-minor-mode unify-8859-on-decoding-mode
2895 "Exists only for backwards compatibility."
2896 :group 'mule
2897 :global t)
2898 ;; Doc said "obsolete" in 23.1, this statement only added in 24.1.
2899 (make-obsolete 'unify-8859-on-decoding-mode "don't use it." "23.1")
2901 (defvar nonascii-insert-offset 0)
2902 (make-obsolete-variable 'nonascii-insert-offset "do not use it." "23.1")
2903 (defvar nonascii-translation-table nil)
2904 (make-obsolete-variable 'nonascii-translation-table "do not use it." "23.1")
2906 (defvar ucs-names nil
2907 "Alist of cached (CHAR-NAME . CHAR-CODE) pairs.")
2909 (defun ucs-names ()
2910 "Return alist of (CHAR-NAME . CHAR-CODE) pairs cached in `ucs-names'."
2911 (or ucs-names
2912 (let ((bmp-ranges
2913 '((#x0000 . #x33FF)
2914 ;; (#x3400 . #x4DBF) CJK Ideographs Extension A
2915 (#x4DC0 . #x4DFF)
2916 ;; (#x4E00 . #x9FFF) CJK Unified Ideographs
2917 (#xA000 . #xD7FF)
2918 ;; (#xD800 . #xFAFF) Surrogate/Private
2919 (#xFB00 . #xFFFD)))
2920 (upper-ranges
2921 '((#x10000 . #x134FF)
2922 ;; (#x13500 . #x167FF) unused
2923 (#x16800 . #x16A3F)
2924 ;; (#x16A40 . #x1AFFF) unused
2925 (#x1B000 . #x1B0FF)
2926 ;; (#x1B100 . #x1CFFF) unused
2927 (#x1D000 . #x1FFFF)
2928 ;; (#x20000 . #xDFFFF) CJK Ideograph Extension A, B, etc, unused
2929 (#xE0000 . #xE01FF)))
2930 (gc-cons-threshold 10000000)
2931 c end name names)
2932 (dolist (range bmp-ranges)
2933 (setq c (car range)
2934 end (cdr range))
2935 (while (<= c end)
2936 (if (setq name (get-char-code-property c 'name))
2937 (push (cons name c) names))
2938 (if (setq name (get-char-code-property c 'old-name))
2939 (push (cons name c) names))
2940 (setq c (1+ c))))
2941 (dolist (range upper-ranges)
2942 (setq c (car range)
2943 end (cdr range))
2944 (while (<= c end)
2945 (if (setq name (get-char-code-property c 'name))
2946 (push (cons name c) names))
2947 (setq c (1+ c))))
2948 (setq ucs-names names))))
2950 (defun read-char-by-name (prompt)
2951 "Read a character by its Unicode name or hex number string.
2952 Display PROMPT and read a string that represents a character by its
2953 Unicode property `name' or `old-name'.
2955 This function returns the character as a number.
2957 You can type a few of the first letters of the Unicode name and
2958 use completion. If you type a substring of the Unicode name
2959 preceded by an asterisk `*' and use completion, it will show all
2960 the characters whose names include that substring, not necessarily
2961 at the beginning of the name.
2963 This function also accepts a hexadecimal number of Unicode code
2964 point or a number in hash notation, e.g. #o21430 for octal,
2965 #x2318 for hex, or #10r8984 for decimal."
2966 (let* ((enable-recursive-minibuffers t)
2967 (completion-ignore-case t)
2968 (input
2969 (completing-read
2970 prompt
2971 (lambda (string pred action)
2972 (if (eq action 'metadata)
2973 '(metadata (category . unicode-name))
2974 (complete-with-action action (ucs-names) string pred)))))
2975 (char
2976 (cond
2978 (string-to-number input 16))
2980 (read input))
2981 (t
2982 (cdr (assoc-string input (ucs-names) t))))))
2983 (unless (characterp char)
2984 (error "Invalid character"))
2985 char))
2987 (define-obsolete-function-alias 'ucs-insert 'insert-char "24.3")
2990 ;;; mule-cmds.el ends here