| blob | f6a24a42dc85f8f0f5207ff9ac794ab2fc62f55d |
1 ;;; mule-cmds.el --- commands for multilingual environment -*-coding: iso-2022-7bit -*-
3 ;; Copyright (C) 1997-2012 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 ,(purecopy "Default") describe-specified-language-support))
65 map))
67 (defvar setup-language-environment-map
68 (let ((map (make-sparse-keymap "Set Language Environment")))
69 (define-key map
71 map))
84 revert-buffer-with-coding-system
85 :enable buffer-file-name
114 map))
142 ,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"
362 ;; The file-name coding system on Darwin systems is always utf-8.
369 ;; Prevent default-terminal-coding-system from converting ^M to ^J.
372 ;; Preserve eol-type from existing default-process-coding-systems.
373 ;; On non-unix-like systems in particular, these may have been set
374 ;; carefully by the user, or by the startup code, to deal with the
375 ;; users shell appropriately, so should not be altered by changing
376 ;; language environment.
387 "Add CODING-SYSTEM at the front of the priority list for automatic detection.
388 This also sets the following coding systems:
389 o coding system of a newly created buffer
390 o default coding system for subprocess I/O
391 This also sets the following values:
392 o default value used as `file-name-coding-system' for converting file names
393 o default value for the command `set-terminal-coding-system'
394 o default value for the command `set-keyboard-coding-system'
396 If CODING-SYSTEM specifies a certain type of EOL conversion, the coding
397 systems set by this function will use that type of EOL conversion.
399 A coding system that requires automatic detection of text+encoding
400 \(e.g. undecided, unix) can't be preferred.
402 To prefer, for instance, utf-8, say the following:
404 \(prefer-coding-system 'utf-8)"
416 base coding-system)))
417 ;; If they asked for specific EOL conversion, honor that.
424 (message "The default value of `file-name-coding-system' was not changed because the specified coding system is not suitable for file names.")))))
427 "If non-nil, a predicate function to sort coding systems.
429 It is called with two coding systems, and should return t if the first
435 "Sort coding system list CODINGS by a priority of each coding system.
436 Return the sorted list. CODINGS is modified by side effects.
438 If a coding system is most preferred, it has the highest priority.
439 Otherwise, coding systems that correspond to MIME charsets have
440 higher priorities. Among them, a coding system included in the
441 `coding-system' key of the current language environment has higher
442 priority. See also the documentation of `language-info-alist'.
444 If the variable `sort-coding-systems-predicate' (which see) is
445 non-nil, it is used to sort CODINGS instead."
455 ;; We calculate the priority number 0..255 by
456 ;; using the 8 bits PMMLCEII as this:
457 ;; P: 1 if most preferred.
458 ;; MM: greater than 0 if mime-charset.
459 ;; L: 1 if one of the current lang. env.'s codings.
460 ;; C: 1 if one of codings listed in the category list.
461 ;; E: 1 if not XXX-with-esc
462 ;; II: if iso-2022 based, 0..3, else 1.
467 ;; Prefer coding systems corresponding to a
468 ;; MIME charset.
470 ;; Lower utf-16 priority so that we
471 ;; normally prefer utf-8 to it, and put
472 ;; x-ctext below that.
488 ;; For ISO based coding systems, prefer
489 ;; one that doesn't use designation nor
490 ;; locking/single shifting.
501 ))))))
506 "Return a list of proper coding systems to encode a text between FROM and TO.
508 If FROM is a string, find coding systems in that instead of the buffer.
509 All coding systems in the list can safely encode any multibyte characters
510 in the text.
512 If the text contains no multibyte characters, return a list of a single
513 element `undecided'."
516 ;; The text contains only ASCII characters. Any coding
517 ;; systems are safe.
519 ;; We need copy-sequence because sorting will alter the argument.
523 "Return a list of proper coding systems to encode STRING.
524 All coding systems in the list can safely encode any multibyte characters
525 in STRING.
527 If STRING contains no multibyte characters, return a list of a single
528 element `undecided'."
532 "Return a list of proper coding systems to encode characters of CHARSETS.
533 CHARSETS is a list of character sets.
535 This only finds coding systems of type `charset', whose
536 `:charset-list' property includes all of CHARSETS (plus `ascii' for
537 ASCII-compatible coding systems). It was used in older versions of
538 Emacs, but is unlikely to be what you really want now."
539 ;; Deal with aliases.
542 charsets))
563 t))
568 "Find multibyte characters in the region specified by FROM and TO.
569 If FROM is a string, find multibyte characters in the string.
570 The return value is an alist of the following format:
571 ((CHARSET COUNT CHAR ...) ...)
572 where
573 CHARSET is a character set,
574 COUNT is a number of characters,
575 CHARs are the characters found from the character set.
576 Optional 3rd arg MAXCOUNT limits how many CHARs are put in the above list.
577 Optional 4th arg EXCLUDES is a list of character sets to be ignored."
579 charset char)
614 "Search forward from point for a character that is not encodable.
615 It asks which coding system to check.
616 If such a character is found, set point after that character.
617 Otherwise, don't move point.
619 When called from a program, the value is the position of the unencodable
620 character found, or nil if all characters are encodable."
625 default))))
630 pos))
633 "Most recent coding system explicitly specified by the user when asked.
634 This variable is set whenever Emacs asks the user which coding system
635 to use in order to write a file. If you set it to nil explicitly,
636 then call `write-region', then afterward this variable will be non-nil
640 "If non-nil, a function to control the behavior of coding system selection.
641 The meaning is the same as the argument ACCEPT-DEFAULT-P of the
642 function `select-safe-coding-system' (which see). This variable
647 "Select interactively a coding system for the region FROM ... TO.
648 FROM can be a string, as in `write-region'.
649 CODINGS is the list of base coding systems known to be safe for this region,
650 typically obtained with `find-coding-systems-region'.
651 UNSAFE is a list of coding systems known to be unsafe for this region.
652 REJECTED is a list of coding systems which were safe but for some reason
653 were not recommended in the particular context.
654 DEFAULT is the coding system to use by default in the query."
655 ;; At first, if some defaults are unsafe, record at most 11
656 ;; problematic characters and their positions for them by turning
657 ;; (CODING ...)
658 ;; into
659 ;; ((CODING (POS . CHAR) (POS . CHAR) ...) ...)
674 unsafe)))
676 ;; Change each safe coding system to the corresponding
677 ;; mime-charset name if it is also a coding system. Such a name
678 ;; is more friendly to users.
680 mime-charset)
688 ;; Don't offer variations with locking shift, which you
689 ;; basically never want.
698 ;; Remove raw-text, emacs-mule and no-conversion unless nothing
699 ;; else is available.
708 coding-system)
710 ;; If some defaults are unsafe, make sure the offending
711 ;; buffer is displayed.
715 unsafe))))
716 ;; Then ask users to select one from CODINGS while showing
717 ;; the reason why none of the defaults are not used.
726 "These default coding systems were tried to encode"
741 but are not recommended for encoding text in this context,
776 'help-echo
777 "mouse-2, RET: jump to this character"
778 'help-function func1
781 "..."
784 'help-echo
785 "mouse-2, RET: next unencodable character"
786 'help-function func2
809 ;; Read a coding system.
813 default))
818 coding-system))
821 accept-default-p file)
822 "Ask a user to select a safe coding system from candidates.
823 The candidates of coding systems which can safely encode a text
824 between FROM and TO are shown in a popup window. Among them, the most
825 proper one is suggested as the default.
827 The list of `buffer-file-coding-system' of the current buffer, the
828 default `buffer-file-coding-system', and the most preferred coding
829 system (if it corresponds to a MIME charset) is treated as the
830 default coding system list. Among them, the first one that safely
831 encodes the text is normally selected silently and returned without
832 any user interaction. See also the command `prefer-coding-system'.
834 However, the user is queried if the chosen coding system is
835 inconsistent with what would be selected by `find-auto-coding' from
836 coding cookies &c. if the contents of the region were read from a
837 file. (That could lead to data corruption in a file subsequently
838 re-visited and edited.)
840 Optional 3rd arg DEFAULT-CODING-SYSTEM specifies a coding system or a
841 list of coding systems to be prepended to the default coding system
842 list. However, if DEFAULT-CODING-SYSTEM is a list and the first
843 element is t, the cdr part is used as the default coding system list,
844 i.e. current `buffer-file-coding-system', default `buffer-file-coding-system',
845 and the most preferred coding system are not used.
847 Optional 4th arg ACCEPT-DEFAULT-P, if non-nil, is a function to
848 determine the acceptability of the silently selected coding system.
849 It is called with that coding system, and should return nil if it
850 should not be silently selected and thus user interaction is required.
852 Optional 5th arg FILE is the file name to use for this purpose.
853 That is different from `buffer-file-name' when handling `write-region'
854 \(for example).
856 The variable `select-safe-coding-system-accept-default-p', if non-nil,
857 overrides ACCEPT-DEFAULT-P.
859 Kludgy feature: if FROM is a string, the string is the target text,
860 and TO is ignored."
865 auto-cs)
867 ;; Find an auto-coding that is specified for the current
868 ;; buffer and file from the region FROM and TO.
879 'mule
881 Invalid coding system `%s' is specified
882 for the current buffer/file by the %s.
883 It is highly recommended to fix it before writing to a file."
896 ;; Change elements of the list to (coding . base-coding).
899 default-coding-system))
902 ;; If the file has a coding cookie, use it regardless of any
903 ;; other setting.
910 ;; If buffer-file-coding-system is not nil nor undecided, append it
911 ;; to the defaults.
922 ;; If default buffer-file-coding-system is not nil nor undecided,
923 ;; append it to the defaults.
933 base)))))))
935 ;; If the most preferred coding system has the property mime-charset,
936 ;; append it to the defaults.
938 base)
950 ;; Decide the eol-type from the top of the default codings,
951 ;; current buffer-file-coding-system, or default buffer-file-coding-system.
957 buffer-file-coding-system)))
971 safe rejected unsafe)
973 ;; Any coding system is ok.
975 ;; Reverse the list so that elements are accumulated in safe,
976 ;; rejected, and unsafe in the correct order.
979 ;; Classify the defaults into safe, rejected, and unsafe.
991 ;; If all the defaults failed, ask a user.
996 ;; Check we're not inconsistent with what `coding:' spec &c would
997 ;; give when file is re-read.
998 ;; But don't do this if we explicitly ignored the cookie
999 ;; by using `find-file-literally'.
1002 coding-system
1004 ;; Merge coding-system and auto-cs as far as possible.
1013 coding-system auto-cs))
1016 auto-cs coding-system))))
1020 coding-system eol-type-2)))
1023 auto-cs eol-type-1)))))))
1026 ;; Don't barf if writing a compressed file, say.
1027 ;; This check perhaps isn't ideal, but is probably
1028 ;; the best thing to do.
1033 %s specified by file contents. Really save (else edit coding cookies \
1038 coding-system)))
1043 "Return a coding system to encode the outgoing message of the current buffer.
1044 It at first tries the first coding system found in these variables
1045 in this order:
1046 (1) local value of `buffer-file-coding-system'
1047 (2) value of `sendmail-coding-system'
1048 (3) value of `default-sendmail-coding-system'
1049 (4) default value of `buffer-file-coding-system'
1050 If the found coding system can't encode the current buffer,
1051 or none of them are bound to a coding system,
1052 it asks the user to select a proper coding system."
1054 buffer-file-coding-system)
1055 sendmail-coding-system
1056 default-sendmail-coding-system
1059 ;; We should never use no-conversion for outgoing mail.
1065 coding)))
1066 \f
1067 ;;; Language support stuff.
1070 "Alist of language environment definitions.
1071 Each element looks like:
1072 (LANGUAGE-NAME . ((KEY . INFO) ...))
1073 where LANGUAGE-NAME is a string, the name of the language environment,
1074 KEY is a symbol denoting the kind of information, and
1075 INFO is the data associated with KEY.
1076 Meaningful values for KEY include
1078 documentation value is documentation of what this language environment
1079 is meant for, and how to use it.
1080 charset value is a list of the character sets mainly used
1081 by this language environment.
1082 sample-text value is an expression which is evalled to generate
1083 a line of text written using characters appropriate
1084 for this language environment.
1085 setup-function value is a function to call to switch to this
1086 language environment.
1087 exit-function value is a function to call to leave this
1088 language environment.
1089 coding-system value is a list of coding systems that are good for
1090 saving text written in this language environment.
1091 This list serves as suggestions to the user;
1092 in effect, as a kind of documentation.
1093 coding-priority value is a list of coding systems for this language
1094 environment, in order of decreasing priority.
1095 This is used to set up the coding system priority
1096 list when you switch to this language environment.
1097 nonascii-translation
1098 value is a charset of dimension one to use for
1099 converting a unibyte character to multibyte
1100 and vice versa.
1101 input-method value is a default input method for this language
1102 environment.
1103 features value is a list of features requested in this
1104 language environment.
1105 ctext-non-standard-encodings
1106 value is a list of non-standard encoding names used
1107 in extended segments of CTEXT. See the variable
1108 `ctext-non-standard-encodings' for more detail.
1110 The following key takes effect only when multibyte characters are
1111 globally disabled, i.e. the default value of `enable-multibyte-characters'
1112 is nil (which is an obsolete and deprecated use):
1114 unibyte-display value is a coding system to encode characters for
1115 the terminal. Characters in the range of 160 to
1116 255 display not as octal escapes, but as non-ASCII
1120 "Return information listed under KEY for language environment LANG-ENV.
1121 KEY is a symbol denoting the kind of information.
1122 For a list of useful values for KEY and their meanings,
1123 see `language-info-alist'."
1131 "Modify part of the definition of language environment LANG-ENV.
1132 Specifically, this stores the information INFO under KEY
1133 in the definition of this language environment.
1134 KEY is a symbol denoting the kind of information.
1135 INFO is the value for that information.
1137 For a list of useful values for KEY and their meanings,
1138 see `language-info-alist'."
1157 "Internal use only.
1158 Arguments are the same as `set-language-info'."
1170 ;; Update the custom-type of `current-language-environment'.
1178 "Store ALIST as the definition of language environment LANG-ENV.
1179 ALIST is an alist of KEY and INFO values. See the documentation of
1180 `language-info-alist' for the meanings of KEY and INFO.
1182 Optional arg PARENTS is a list of parent menu names; it specifies
1183 where to put this language environment in the
1184 Describe Language Environment and Set Language Environment menus.
1186 in the European submenu in each of those two menus."
1195 map parent-symbol parent prompt)
1201 ;; This prompt string is for define-prefix-command, so
1202 ;; that the map it creates will be suitable for a menu.
1223 ;; Set up menu items for this language env.
1238 "Read a language environment name which has information for KEY.
1239 If KEY is nil, read any language environment.
1240 Prompt with PROMPT. DEFAULT is the default choice of language environment.
1241 This returns a language environment name as a string."
1244 language-info-alist
1247 t nil nil default)))
1251 name)))
1252 \f
1253 ;;; Multilingual input methods.
1255 "LEIM: Libraries of Emacs Input Methods."
1259 "Name of LEIM list file.
1260 This file contains a list of libraries of Emacs input methods (LEIM)
1261 in the format of Lisp expression for registering each input method.
1265 ";;; %s -- list of LEIM (Library of Emacs Input Method) -*-coding: utf-8;-*-
1266 ;;
1267 ;; This file is automatically generated.
1268 ;;
1269 ;; This file contains a list of LEIM (Library of Emacs Input Method)
1270 ;; methods in the same directory as this file. Loading this file
1271 ;; registers all the input methods in Emacs.
1272 ;;
1273 ;; Each entry has the form:
1274 ;; (register-input-method
1275 ;; INPUT-METHOD LANGUAGE-NAME ACTIVATE-FUNC
1276 ;; TITLE DESCRIPTION
1277 ;; ARG ...)
1278 ;; See the function `register-input-method' for the meanings of the arguments.
1279 ;;
1280 ;; If this directory is included in `load-path', Emacs automatically
1281 ;; loads this file at startup time.
1283 "
1284 leim-list-file-name)
1288 "Regexp matching head of each entry in LEIM list file.
1293 "List of functions to call to update LEIM list file.
1297 "Update LEIM list file in directories DIRS."
1302 "The current input method for multilingual text.
1313 "String widget with completion for input method."
1314 :completions
1321 "Default input method for multilingual text (a string).
1322 This is the input method activated automatically by the command
1327 mule-input-method-string)
1333 "History list of input methods read from the minibuffer.
1335 Maximum length of the history list is determined by the value
1341 "Function to call for inactivating the current input method.
1342 Every input method should set this to an appropriate value when activated.
1343 This function is called with no argument.
1345 This function should never change the value of `current-input-method'.
1351 "Function to call for describing the current input method.
1357 "Alist of input method names vs how to use them.
1358 Each element has the form:
1359 (INPUT-METHOD LANGUAGE-ENV ACTIVATE-FUNC TITLE DESCRIPTION ARGS...)
1361 ;;;###autoload
1365 "Register INPUT-METHOD as an input method for language environment LANG-ENV.
1367 INPUT-METHOD and LANG-ENV are symbols or strings.
1368 ACTIVATE-FUNC is a function to call to activate this method.
1369 TITLE is a string to show in the mode line when this method is active.
1370 DESCRIPTION is a string describing this method and what it is good for.
1371 The ARGS, if any, are passed as arguments to ACTIVATE-FUNC.
1372 All told, the arguments to ACTIVATE-FUNC are INPUT-METHOD and the ARGS.
1375 created at Emacs build time, registering all Quail input methods
1376 contained in the Emacs distribution.
1378 In case you want to register a new Quail input method by yourself, be
1379 careful to use the same input method title as given in the third
1380 parameter of `quail-define-package'. (If the values are different, the
1381 string specified in this function takes precedence.)
1383 The commands `describe-input-method' and `list-input-methods' need
1384 these duplicated values to show some information about input methods
1385 without loading the relevant Quail packages.
1402 "Read a name of input method from a minibuffer prompting with PROMPT.
1403 If DEFAULT is non-nil, use that as the default,
1404 and substitute it into PROMPT at the first `%s'.
1405 If INHIBIT-NULL is non-nil, null input signals an error.
1407 The return value is a string."
1411 ;; As it is quite normal to change input method in the
1412 ;; minibuffer, we must enable it even if
1413 ;; enable-recursive-minibuffers is currently nil.
1415 ;; This binding is necessary because input-method-history is
1416 ;; buffer local.
1418 nil t nil 'input-method-history
1419 default)))
1423 input-method
1428 "Switch to input method INPUT-METHOD for the current buffer.
1429 If some other input method is already active, turn it off first.
1430 If INPUT-METHOD is nil, deactivate any current input method."
1457 "Turn off the current input method."
1468 current-input-method-title nil)
1476 "Select and activate input method INPUT-METHOD for the current buffer.
1477 This also sets the default input method to the one you specify.
1478 If INPUT-METHOD is nil, this function turns off the input method, and
1479 also causes you to be prompted for a name of an input method the next
1481 When called interactively, the optional arg INTERACTIVE is non-nil,
1482 which marks the variable `default-input-method' as set for Custom buffers.
1485 To deactivate it programmatically, use `inactivate-input-method'."
1490 default t)
1491 t)))
1496 default-input-method)
1502 "Enable or disable multilingual text input method for the current buffer.
1503 Only one input method can be enabled at any time in a given buffer.
1505 The normal action is to enable an input method if none was enabled,
1506 and disable the current one otherwise. Which input method to enable
1507 can be determined in various ways--either the one most recently used,
1508 or the one specified by `default-input-method', or as a last resort
1509 by reading the name of an input method in the minibuffer.
1511 With a prefix argument ARG, read an input method name with the minibuffer
1512 and enable that one. The default is the most recent input method specified
1513 \(not including the currently active input method, if any).
1515 When called interactively, the optional argument INTERACTIVE is non-nil,
1516 which marks the variable `default-input-method' as set for Custom buffers."
1533 default t))
1534 default))
1544 "Describe input method INPUT-METHOD."
1574 "Describe the input method currently in use.
1575 This is a subroutine for `describe-input-method'."
1581 current-input-method)
1586 "Read a multilingual string from minibuffer, prompting with string PROMPT.
1587 The input method selected last time is activated in minibuffer.
1588 If optional second argument INITIAL-INPUT is non-nil, insert it in the
1589 minibuffer initially.
1590 Optional 3rd argument INPUT-METHOD specifies the input method to be activated
1591 instead of the one selected last time. It is a symbol or a string."
1594 current-input-method
1595 default-input-method
1606 ;; Variables to control behavior of input methods. All input methods
1607 ;; should react to these variables.
1610 "A flag to control extra guidance given by input methods.
1611 The value should be nil, t, `complex-only', or `default'.
1613 The extra guidance is done by showing list of available keys in echo
1614 area. When you use the input method in the minibuffer, the guidance
1615 is shown at the bottom short window (split from the existing window).
1617 If the value is t, extra guidance is always given, if the value is
1618 nil, extra guidance is always suppressed.
1620 If the value is `complex-only', only complex input methods such as
1621 `chinese-py' and `japanese' give extra guidance.
1623 If the value is `default', complex input methods always give extra
1624 guidance, but simple input methods give it only when you are not in
1625 the minibuffer.
1627 See also the variable `input-method-highlight-flag'."
1633 "If this flag is non-nil, input methods highlight partially-entered text.
1634 For instance, while you are in the middle of a Quail input method sequence,
1635 the text inserted so far is temporarily underlined.
1636 The underlining goes away when you finish or abort the input method sequence.
1637 See also the variable `input-method-verbose-flag'."
1642 "Normal hook run just after an input method is activated.
1644 The variable `current-input-method' keeps the input method name
1645 just activated."
1650 "Normal hook run just after an input method is inactivated.
1652 The variable `current-input-method' still keeps the input method name
1653 just inactivated."
1658 "Normal hook run just after an input method insert some chunk of text."
1663 "This flag controls when an input method returns.
1664 Usually, the input method does not return while there's a possibility
1665 that it may find a different translation if a user types another key.
1666 But, if this flag is non-nil, the input method returns as soon as the
1670 "This flag controls how an input method shows an intermediate key sequence.
1671 Usually, the input method inserts the intermediate key sequence,
1672 or candidate translations corresponding to the sequence,
1673 at point in the current buffer.
1674 But, if this flag is non-nil, it displays them in echo area instead."
1679 "This flag controls the behavior of an input method on invalid key input.
1680 Usually, when a user types a key which doesn't start any character
1681 handled by the input method, the key is handled by turning off the
1682 input method temporarily. After that key, the input method is re-enabled.
1685 \f
1687 "Normal hook run after some language environment is set.
1689 When you set some hook function here, that effect usually should not
1690 be inherited to another language environment. So, you had better set
1691 another function in `exit-language-environment-hook' (which see) to
1692 cancel the effect."
1697 "Normal hook run after exiting from some language environment.
1698 When this hook is run, the variable `current-language-environment'
1699 is still bound to the language environment being exited.
1701 This hook is mainly used for canceling the effect of
1702 `set-language-environment-hook' (which see)."
1709 "Switch to a specified language environment."
1722 "The last language environment specified with `set-language-environment'.
1724 to using the function `set-language-environment'."
1731 current-language-environment)
1732 language-info-alist t))
1734 ;; custom type will be updated with `set-language-info'.
1745 "Reset multilingual environment of Emacs to the default status.
1747 The default status is as follows:
1749 The default value of `buffer-file-coding-system' is nil.
1750 The default coding system for process I/O is nil.
1751 The default value for the command `set-terminal-coding-system' is nil.
1752 The default value for the command `set-keyboard-coding-system' is nil.
1754 The order of priorities of coding systems are as follows:
1755 utf-8
1756 iso-2022-7bit
1757 iso-latin-1
1758 iso-2022-7bit-lock
1759 iso-2022-8bit-ss2
1760 emacs-mule
1761 raw-text"
1763 ;; This function formerly set default-enable-multibyte-characters to t,
1764 ;; but that is incorrect. It should not alter the unibyte/multibyte choice.
1767 'utf-8
1768 'iso-2022-7bit
1769 'iso-latin-1
1770 'iso-2022-7bit-lock
1771 'iso-2022-8bit-ss2
1772 'emacs-mule
1777 ;; On Darwin systems, this should be utf-8, but when this file is loaded
1778 ;; utf-8 is not yet defined, so we set it in set-locale-environment instead.
1780 ;; Preserve eol-type from existing default-process-coding-systems.
1781 ;; On non-unix-like systems in particular, these may have been set
1782 ;; carefully by the user, or by the startup code, to deal with the
1783 ;; users shell appropriately, so should not be altered by changing
1784 ;; language environment.
1786 ;; When bootstrapping, coding-systems are not defined yet, so
1787 ;; we need to catch the error from check-coding-system.
1800 ;; Put the highest priority to the charset iso-8859-1 to prefer the
1801 ;; registry iso8859-1 over iso8859-2 in font selection. It also
1802 ;; makes unibyte-display-via-language-environment to use iso-8859-1
1803 ;; as the unibyte charset.
1806 ;; Don't alter the terminal and keyboard coding systems here.
1807 ;; The terminal still supports the same coding system
1808 ;; that it supported a minute ago.
1809 ;; (set-terminal-coding-system-internal nil)
1810 ;; (set-keyboard-coding-system-internal nil)
1812 ;; Back in Emacs-20, it was necessary to provide some fallback implicit
1813 ;; conversion, because almost no packages handled coding-system issues.
1814 ;; Nowadays it'd just paper over bugs.
1815 ;; (set-unibyte-charset 'iso-8859-1)
1816 )
1820 (defun set-display-table-and-terminal-coding-system (language-name &optional coding-system display)
1821 "Set up the display table and terminal coding system for LANGUAGE-NAME."
1827 ;; The following 2 lines undo the 8-bit display that we set up
1828 ;; in standard-display-european-internal, which see. This is in
1829 ;; case the user has used standard-display-european earlier in
1830 ;; this session.
1837 "Set up multilingual environment for using LANGUAGE-NAME.
1838 This sets the coding system priority and the default input method
1839 and sometimes other things. LANGUAGE-NAME should be a string
1841 specifies the character set for the major languages of Western Europe.
1843 If there is a prior value for `current-language-environment', this
1844 runs the hook `exit-language-environment-hook'. After setting up
1845 the new language environment, it runs `set-language-environment-hook'."
1847 nil
1864 ;; The features might set up coding systems.
1876 ;; Unibyte setups if necessary.
1886 current-iso639-language))
1892 "An Emacs charset."
1894 :completions
1904 widget))
1908 "Customizations of language environment parameters.
1909 Value is an alist with elements like those of `language-info-alist'.
1910 These are used to set values in `language-info-alist' which replace
1911 the defaults. A typical use is replacing the default input method for
1914 This option is intended for use at startup. Removing items doesn't
1915 remove them from the language info until you next restart Emacs.
1917 Setting this variable directly does not take effect.
1918 See `set-language-info-alist' for use in programs."
1923 ;; Can't do this before language environments are set up.
1925 ;; modify language-info-alist
1928 ;; re-set the environment in case its parameters changed
1932 :completions
1936 action language-info-alist string pred))))
1937 :value-type
1955 ;; Actually set up direct output of non-ASCII characters.
1957 ;; Unibyte Emacs on MS-DOS wants to display all 8-bit characters with
1958 ;; the native font, and codes 160 and 146 stand for something very
1959 ;; different there.
1963 ;; Most X fonts used to do the wrong thing for latin-1 code 160.
1965 ;; XFree86 4 has fixed the fonts.
1969 ;; Make non-line-break space display as a plain space.
1972 ;; don't contain a character at that position. Map it to the ASCII
1973 ;; apostrophe. [This is actually RIGHT SINGLE QUOTATION MARK,
1974 ;; U+2019, normally from the windows-1252 character set. XFree 4
1975 ;; fonts probably have the appropriate glyph at this position,
1976 ;; so they could use standard-display-8bit. It's better to use a
1977 ;; proper windows-1252 coding system. --fx]
1978 (aset standard-display-table (unibyte-char-to-multibyte 146) [39]))))
1980 (defun set-language-environment-coding-systems (language-name)
1981 "Do various coding system setups for language environment LANGUAGE-NAME."
1982 (let* ((priority (get-language-info language-name 'coding-priority))
1983 (default-coding (car priority))
1984 ;; If the default buffer-file-coding-system is nil, don't use
1985 ;; coding-system-eol-type, because it treats nil as
1986 ;; `no-conversion'. The default buffer-file-coding-system is set
1987 ;; to nil by reset-language-environment, and in that case we
1988 ;; want to have here the native EOL type for each platform.
1989 ;; FIXME: there should be a common code that runs both on
1990 ;; startup and here to set the default EOL type correctly.
1991 ;; Right now, DOS/Windows platforms set this on dos-w32.el,
1992 ;; which works only as long as the order of loading files at
1993 ;; dump time and calling functions at startup is not modified
1994 ;; significantly, i.e. as long as this function is called
1995 ;; _after_ the default buffer-file-coding-system was set by
1996 ;; dos-w32.el.
1997 (eol-type
1998 (coding-system-eol-type
1999 (or (default-value 'buffer-file-coding-system)
2000 (if (memq system-type '(windows-nt ms-dos)) 'dos 'unix)))))
2001 (when priority
2002 (set-default-coding-systems
2003 (if (memq eol-type '(0 1 2 unix dos mac))
2004 (coding-system-change-eol-conversion default-coding eol-type)
2005 default-coding))
2006 (setq default-sendmail-coding-system default-coding)
2007 (apply 'set-coding-system-priority priority))))
2009 (defun set-language-environment-input-method (language-name)
2010 "Do various input method setups for language environment LANGUAGE-NAME."
2011 (let ((input-method (get-language-info language-name 'input-method)))
2012 (when input-method
2013 (setq default-input-method input-method)
2014 (if input-method-history
2015 (setq input-method-history
2016 (cons input-method
2017 (delete input-method input-method-history)))))))
2019 (defun set-language-environment-nonascii-translation (language-name)
2020 "Do unibyte/multibyte translation setup for language environment LANGUAGE-NAME."
2021 ;; Note: For DOS, we assumed that the charset cpXXX is already
2022 ;; defined.
2023 (let ((nonascii (get-language-info language-name 'nonascii-translation)))
2024 (if (eq window-system 'pc)
2025 (setq nonascii (intern (format "cp%d" dos-codepage))))
2026 (or (and (charsetp nonascii)
2027 (get-charset-property nonascii :ascii-compatible-p))
2028 (setq nonascii 'iso-8859-1))
2029 ;; Back in Emacs-20, it was necessary to provide some fallback implicit
2030 ;; conversion, because almost no packages handled coding-system issues.
2031 ;; Nowadays it'd just paper over bugs.
2032 ;; (set-unibyte-charset nonascii)
2033 ))
2035 (defun set-language-environment-charset (language-name)
2036 "Do various charset setups for language environment LANGUAGE-NAME."
2037 ;; Put higher priorities to such charsets that are supported by the
2038 ;; coding systems of higher priorities in this environment.
2039 (let ((charsets (get-language-info language-name 'charset)))
2040 (dolist (coding (get-language-info language-name 'coding-priority))
2041 (let ((list (coding-system-charset-list coding)))
2042 (if (consp list)
2043 (setq charsets (append charsets list)))))
2044 (if charsets
2045 (apply 'set-charset-priority charsets))))
2047 (defun set-language-environment-unibyte (language-name)
2048 "Do various unibyte-mode setups for language environment LANGUAGE-NAME."
2049 (set-display-table-and-terminal-coding-system language-name))
2051 (defun princ-list (&rest args)
2053 (mapc #'princ args)
2055 (make-obsolete 'princ-list "use mapc and princ instead" "23.3")
2057 (put 'describe-specified-language-support 'apropos-inhibit t)
2059 ;; Print language-specific information such as input methods,
2060 ;; charsets, and coding systems. This function is intended to be
2061 ;; called from the menu:
2062 ;; [menu-bar mule describe-language-environment LANGUAGE]
2063 ;; and should not run it by `M-x describe-current-input-method-function'.
2064 (defun describe-specified-language-support ()
2065 "Describe how Emacs supports the specified language environment."
2066 (interactive)
2067 (let (language-name)
2068 (if (not (and (symbolp last-command-event)
2069 (or (not (eq last-command-event 'Default))
2070 (setq last-command-event 'English))
2071 (setq language-name (symbol-name last-command-event))))
2072 (error "This command should only be called from the menu bar"))
2073 (describe-language-environment language-name)))
2075 (defun describe-language-environment (language-name)
2076 "Describe how Emacs supports language environment LANGUAGE-NAME."
2077 (interactive
2078 (list (read-language-name
2079 'documentation
2080 "Describe language environment (default current choice): ")))
2081 (if (null language-name)
2082 (setq language-name current-language-environment))
2083 (if (or (null language-name)
2084 (null (get-language-info language-name 'documentation)))
2085 (error "No documentation for the specified language"))
2086 (if (symbolp language-name)
2087 (setq language-name (symbol-name language-name)))
2088 (dolist (feature (get-language-info language-name 'features))
2089 (require feature))
2090 (let ((doc (get-language-info language-name 'documentation)))
2091 (help-setup-xref (list #'describe-language-environment language-name)
2092 (called-interactively-p 'interactive))
2093 (with-output-to-temp-buffer (help-buffer)
2094 (with-current-buffer standard-output
2096 (if (stringp doc)
2098 (condition-case nil
2099 (let ((str (eval (get-language-info language-name 'sample-text))))
2100 (if (stringp str)
2104 (error nil))
2105 (let ((input-method (get-language-info language-name 'input-method))
2106 (l (copy-sequence input-method-alist))
2107 (first t))
2108 (when (and input-method
2109 (setq input-method (assoc input-method l)))
2111 (setq l (cons input-method (delete input-method l))
2112 first nil))
2113 (dolist (elt l)
2114 (when (or (eq input-method elt)
2115 (eq t (compare-strings language-name nil nil
2116 (nth 1 elt) nil nil t)))
2117 (when first
2119 (setq first nil))
2120 (insert " " (car elt))
2121 (search-backward (car elt))
2122 (help-xref-button 0 'help-input-method (car elt))
2123 (goto-char (point-max))
2125 (if (stringp (nth 3 elt)) (nth 3 elt) (car (nth 3 elt)))
2127 (or first
2130 (let ((l (get-language-info language-name 'charset)))
2131 (if (null l)
2133 (while l
2134 (insert " " (symbol-name (car l)))
2135 (search-backward (symbol-name (car l)))
2136 (help-xref-button 0 'help-character-set (car l))
2137 (goto-char (point-max))
2139 (setq l (cdr l)))))
2142 (let ((l (get-language-info language-name 'coding-system)))
2143 (if (null l)
2145 (while l
2146 (insert " " (symbol-name (car l)))
2147 (search-backward (symbol-name (car l)))
2148 (help-xref-button 0 'help-coding-system (car l))
2149 (goto-char (point-max))
2150 (insert " (`"
2151 (coding-system-mnemonic (car l))
2153 (coding-system-doc-string (car l))
2155 (let ((aliases (coding-system-aliases (car l))))
2156 (when aliases
2158 (while aliases
2159 (insert " " (symbol-name (car aliases)))
2160 (setq aliases (cdr aliases)))
2162 (setq l (cdr l)))))))))
2163 \f
2164 ;;; Locales.
2166 (defvar locale-translation-file-name nil
2167 "File name for the system's file of locale-name aliases, or nil if none.")
2169 ;; The following definitions might as well be marked as constants and
2170 ;; purecopied, since they're normally used on startup, and probably
2171 ;; should reflect the facilities of the base Emacs.
2172 (defconst locale-language-names
2173 (purecopy
2174 '(
2175 ;; Locale names of the form LANGUAGE[_TERRITORY][.CODESET][@MODIFIER]
2176 ;; as specified in the Single Unix Spec, Version 2.
2177 ;; LANGUAGE is a language code taken from ISO 639:1988 (E/F)
2178 ;; with additions from ISO 639/RA Newsletter No.1/1989;
2179 ;; see Internet RFC 2165 (1997-06) and
2180 ;; http://www.evertype.com/standards/iso639/iso639-en.html
2181 ;; TERRITORY is a country code taken from ISO 3166
2182 ;; http://www.din.de/gremien/nas/nabd/iso3166ma/codlstp1/en_listp1.html.
2183 ;; CODESET and MODIFIER are implementation-dependent.
2185 ;; jasonr comments: MS Windows uses three letter codes for
2186 ;; languages instead of the two letter ISO codes that POSIX
2187 ;; uses. In most cases the first two letters are the same, so
2188 ;; most of the regexps in locale-language-names work. Japanese
2189 ;; and Chinese are exceptions, which are listed in the
2190 ;; non-standard section at the bottom of locale-language-names.
2192 ("aa_DJ" . "Latin-1") ; Afar
2193 ("aa" . "UTF-8")
2194 ;; ab Abkhazian
2195 ("af" . "Latin-1") ; Afrikaans
2196 ("am" "Ethiopic" utf-8) ; Amharic
2197 ("an" . "Latin-9") ; Aragonese
2198 ("ar" . "Arabic")
2199 ; as Assamese
2200 ; ay Aymara
2201 ("az" . "UTF-8") ; Azerbaijani
2202 ; ba Bashkir
2203 ("be" "Belarusian" cp1251) ; Belarusian [Byelorussian until early 1990s]
2204 ("bg" "Bulgarian" cp1251) ; Bulgarian
2205 ; bh Bihari
2206 ; bi Bislama
2207 ("bn" . "UTF-8") ; Bengali, Bangla
2208 ("bo" . "Tibetan")
2209 ("br" . "Latin-1") ; Breton
2210 ("bs" . "Latin-2") ; Bosnian
2211 ("byn" . "UTF-8") ; Bilin; Blin
2212 ("ca" . "Latin-1") ; Catalan
2213 ; co Corsican
2214 ("cs" "Czech" iso-8859-2)
2215 ("cy" "Welsh" iso-8859-14)
2216 ("da" . "Latin-1") ; Danish
2217 ("de" "German" iso-8859-1)
2218 ; dv Divehi
2219 ; dz Bhutani
2220 ("el" "Greek" iso-8859-7)
2221 ;; Users who specify "en" explicitly typically want Latin-1, not ASCII.
2222 ;; That's actually what the GNU locales define, modulo things like
2223 ;; en_IN -- fx.
2224 ("en_IN" "English" utf-8) ; glibc uses utf-8 for English in India
2225 ("en" "English" iso-8859-1) ; English
2226 ("eo" . "Esperanto") ; Esperanto
2227 ("es" "Spanish" iso-8859-1)
2228 ("et" . "Latin-1") ; Estonian
2229 ("eu" . "Latin-1") ; Basque
2230 ("fa" . "UTF-8") ; Persian
2231 ("fi" . "Latin-1") ; Finnish
2232 ("fj" . "Latin-1") ; Fiji
2233 ("fo" . "Latin-1") ; Faroese
2234 ("fr" "French" iso-8859-1) ; French
2235 ("fy" . "Latin-1") ; Frisian
2236 ("ga" . "Latin-1") ; Irish Gaelic (new orthography)
2237 ("gd" . "Latin-9") ; Scots Gaelic
2238 ("gez" "Ethiopic" utf-8) ; Geez
2239 ("gl" . "Latin-1") ; Gallegan; Galician
2240 ; gn Guarani
2241 ("gu" . "UTF-8") ; Gujarati
2242 ("gv" . "Latin-1") ; Manx Gaelic
2243 ; ha Hausa
2244 ("he" "Hebrew" iso-8859-8)
2245 ("hi" "Devanagari" utf-8) ; Hindi
2246 ("hr" "Croatian" iso-8859-2) ; Croatian
2247 ("hu" . "Latin-2") ; Hungarian
2248 ; hy Armenian
2249 ; ia Interlingua
2250 ("id" . "Latin-1") ; Indonesian
2251 ; ie Interlingue
2252 ; ik Inupiak
2253 ("is" . "Latin-1") ; Icelandic
2254 ("it" "Italian" iso-8859-1) ; Italian
2255 ; iu Inuktitut
2256 ("iw" "Hebrew" iso-8859-8)
2257 ("ja" "Japanese" euc-jp)
2258 ; jw Javanese
2259 ("ka" "Georgian" georgian-ps) ; Georgian
2260 ; kk Kazakh
2261 ("kl" . "Latin-1") ; Greenlandic
2262 ; km Cambodian
2263 ("kn" "Kannada" utf-8)
2264 ("ko" "Korean" euc-kr)
2265 ; ks Kashmiri
2266 ; ku Kurdish
2267 ("kw" . "Latin-1") ; Cornish
2268 ; ky Kirghiz
2269 ("la" . "Latin-1") ; Latin
2270 ("lb" . "Latin-1") ; Luxemburgish
2271 ("lg" . "Laint-6") ; Ganda
2272 ; ln Lingala
2273 ("lo" "Lao" utf-8) ; Laothian
2274 ("lt" "Lithuanian" iso-8859-13)
2275 ("lv" . "Latvian") ; Latvian, Lettish
2276 ; mg Malagasy
2277 ("mi" . "Latin-7") ; Maori
2278 ("mk" "Cyrillic-ISO" iso-8859-5) ; Macedonian
2279 ("ml" "Malayalam" utf-8)
2280 ("mn" . "UTF-8") ; Mongolian
2281 ; mo Moldavian
2282 ("mr" "Devanagari" utf-8) ; Marathi
2283 ("ms" . "Latin-1") ; Malay
2284 ("mt" . "Latin-3") ; Maltese
2285 ; my Burmese
2286 ; na Nauru
2287 ("nb" . "Latin-1") ; Norwegian
2288 ("ne" "Devanagari" utf-8) ; Nepali
2289 ("nl" "Dutch" iso-8859-1)
2290 ("no" . "Latin-1") ; Norwegian
2291 ("oc" . "Latin-1") ; Occitan
2292 ("om_ET" . "UTF-8") ; (Afan) Oromo
2293 ("om" . "Latin-1") ; (Afan) Oromo
2294 ; or Oriya
2295 ("pa" . "UTF-8") ; Punjabi
2296 ("pl" . "Latin-2") ; Polish
2297 ; ps Pashto, Pushto
2298 ("pt" . "Latin-1") ; Portuguese
2299 ; qu Quechua
2300 ("rm" . "Latin-1") ; Rhaeto-Romanic
2301 ; rn Kirundi
2302 ("ro" "Romanian" iso-8859-2)
2303 ("ru_RU" "Russian" iso-8859-5)
2304 ("ru_UA" "Russian" koi8-u)
2305 ; rw Kinyarwanda
2306 ("sa" . "Devanagari") ; Sanskrit
2307 ; sd Sindhi
2308 ("se" . "UTF-8") ; Northern Sami
2309 ; sg Sangho
2310 ("sh" . "Latin-2") ; Serbo-Croatian
2311 ; si Sinhalese
2312 ("sid" . "UTF-8") ; Sidamo
2313 ("sk" "Slovak" iso-8859-2)
2314 ("sl" "Slovenian" iso-8859-2)
2315 ; sm Samoan
2316 ; sn Shona
2317 ("so_ET" "UTF-8") ; Somali
2318 ("so" "Latin-1") ; Somali
2319 ("sq" . "Latin-1") ; Albanian
2320 ("sr" . "Latin-2") ; Serbian (Latin alphabet)
2321 ; ss Siswati
2322 ("st" . "Latin-1") ; Sesotho
2323 ; su Sundanese
2324 ("sv" "Swedish" iso-8859-1) ; Swedish
2325 ("sw" . "Latin-1") ; Swahili
2326 ("ta" "Tamil" utf-8)
2327 ("te" . "UTF-8") ; Telugu
2328 ("tg" "Tajik" koi8-t)
2329 ("th" "Thai" tis-620)
2330 ("ti" "Ethiopic" utf-8) ; Tigrinya
2331 ("tig_ER" . "UTF-8") ; Tigre
2332 ; tk Turkmen
2333 ("tl" . "Latin-1") ; Tagalog
2334 ; tn Setswana
2335 ; to Tonga
2336 ("tr" "Turkish" iso-8859-9)
2337 ; ts Tsonga
2338 ("tt" . "UTF-8") ; Tatar
2339 ; tw Twi
2340 ; ug Uighur
2341 ("uk" "Ukrainian" koi8-u)
2342 ("ur" . "UTF-8") ; Urdu
2343 ("uz_UZ@cyrillic" . "UTF-8"); Uzbek
2344 ("uz" . "Latin-1") ; Uzbek
2345 ("vi" "Vietnamese" utf-8)
2346 ; vo Volapuk
2347 ("wa" . "Latin-1") ; Walloon
2348 ; wo Wolof
2349 ("xh" . "Latin-1") ; Xhosa
2350 ("yi" . "Windows-1255") ; Yiddish
2351 ; yo Yoruba
2352 ; za Zhuang
2353 ("zh_HK" . "Chinese-Big5")
2354 ; zh_HK/BIG5-HKSCS \
2355 ("zh_TW" . "Chinese-Big5")
2356 ("zh_CN.GB2312" "Chinese-GB")
2357 ("zh_CN.GBK" "Chinese-GBK")
2358 ("zh_CN.GB18030" "Chinese-GB18030")
2359 ("zh_CN.UTF-8" . "Chinese-GBK")
2360 ("zh_CN" . "Chinese-GB")
2361 ("zh" . "Chinese-GB")
2362 ("zu" . "Latin-1") ; Zulu
2364 ;; ISO standard locales
2365 ("c$" . "ASCII")
2366 ("posix$" . "ASCII")
2368 ;; The "IPA" Emacs language environment does not correspond
2369 ;; to any ISO 639 code, so let it stand for itself.
2370 ("ipa$" . "IPA")
2372 ;; Nonstandard or obsolete language codes
2373 ("cz" . "Czech") ; e.g. Solaris 2.6
2374 ("ee" . "Latin-4") ; Estonian, e.g. X11R6.4
2375 ("iw" . "Hebrew") ; e.g. X11R6.4
2376 ("sp" . "Cyrillic-ISO") ; Serbian (Cyrillic alphabet), e.g. X11R6.4
2377 ("su" . "Latin-1") ; Finnish, e.g. Solaris 2.6
2378 ("jp" . "Japanese") ; e.g. MS Windows
2379 ("chs" . "Chinese-GBK") ; MS Windows Chinese Simplified
2380 ("cht" . "Chinese-BIG5") ; MS Windows Chinese Traditional
2381 ("gbz" . "UTF-8") ; MS Windows Dari Persian
2382 ("div" . "UTF-8") ; MS Windows Divehi (Maldives)
2383 ("wee" . "Latin-2") ; MS Windows Lower Sorbian
2384 ("wen" . "Latin-2") ; MS Windows Upper Sorbian
2385 ))
2386 "Alist of locale regexps vs the corresponding languages and coding systems.
2387 Each element has this form:
2388 \(LOCALE-REGEXP LANG-ENV CODING-SYSTEM)
2389 The first element whose LOCALE-REGEXP matches the start of a
2390 downcased locale specifies the LANG-ENV \(language environment)
2391 and CODING-SYSTEM corresponding to that locale. If there is no
2392 appropriate language environment, the element may have this form:
2393 \(LOCALE-REGEXP . LANG-ENV)
2394 In this case, LANG-ENV is one of generic language environments for an
2397 (defconst locale-charset-language-names
2398 (purecopy
2407 ;; utf-8@euro exists, so put this last. (@euro really specifies
2408 ;; the currency, rather than the charset.)
2410 "List of pairs of locale regexps and charset language names.
2411 The first element whose locale regexp matches the start of a downcased locale
2412 specifies the language name whose charset corresponds to that locale.
2413 This language name is used if the locale is not listed in
2414 `locale-language-names'.")
2416 (defconst locale-preferred-coding-systems
2417 (purecopy
2426 ;; utf-8@euro exists, so put this after utf-8. (@euro really
2427 ;; specifies the currency, rather than the charset.)
2428 (".*@euro" . iso-8859-15)
2429 ("koi8-?r" . koi8-r)
2430 ("koi8-?u" . koi8-u)
2431 ("tcvn" . tcvn)
2432 ("big5[-_]?hkscs" . big5-hkscs)
2433 ("big5" . big5)
2434 ("euc-?tw" . euc-tw)
2435 ("euc-?cn" . euc-cn)
2436 ("gb2312" . gb2312)
2437 ("gbk" . gbk)
2438 ("gb18030" . gb18030)
2439 ("ja.*[._]euc" . japanese-iso-8bit)
2440 ("ja.*[._]jis7" . iso-2022-jp)
2441 ("ja.*[._]pck" . japanese-shift-jis)
2442 ("ja.*[._]sjis" . japanese-shift-jis)
2443 ("jpn" . japanese-shift-jis) ; MS-Windows uses this.
2444 ))
2445 "List of pairs of locale regexps and preferred coding systems.
2446 The first element whose locale regexp matches the start of a downcased locale
2447 specifies the coding system to prefer when using that locale.
2448 This coding system is used if the locale specifies a specific charset.")
2450 (defun locale-name-match (key alist)
2451 "Search for KEY in ALIST, which should be a list of regexp-value pairs.
2452 Return the value corresponding to the first regexp that matches the
2453 start of KEY, or nil if there is no match."
2454 (let (element)
2455 (while (and alist (not element))
2457 (setq element (car alist)))
2458 (setq alist (cdr alist)))
2459 (cdr element)))
2461 (defun locale-charset-match-p (charset1 charset2)
2462 "Whether charset names (strings) CHARSET1 and CHARSET2 are equivalent.
2463 Matching is done ignoring case and any hyphens and underscores in the
2464 names. E.g. `ISO_8859-1' and `iso88591' both match `iso-8859-1'."
2465 (setq charset1 (replace-regexp-in-string "[-_]" "" charset1))
2466 (setq charset2 (replace-regexp-in-string "[-_]" "" charset2))
2467 (eq t (compare-strings charset1 nil nil charset2 nil nil t)))
2469 (defvar locale-charset-alist nil
2470 "Coding system alist keyed on locale-style charset name.
2471 Used by `locale-charset-to-coding-system'.")
2473 (defun locale-charset-to-coding-system (charset)
2474 "Find coding system corresponding to CHARSET.
2475 CHARSET is any sort of non-Emacs charset name, such as might be used
2476 in a locale codeset, or elsewhere. It is matched to a coding system
2477 first by case-insensitive lookup in `locale-charset-alist'. Then
2478 matches are looked for in the coding system list, treating case and
2479 the characters `-' and `_' as insignificant. The coding system base
2481 `iso-latin-2' is returned."
2482 (or (car (assoc-string charset locale-charset-alist t))
2483 (let ((cs coding-system-alist)
2484 c)
2485 (while (and (not c) cs)
2486 (if (locale-charset-match-p charset (caar cs))
2487 (setq c (intern (caar cs)))
2488 (pop cs)))
2489 (if c (coding-system-base c)))))
2491 ;; Fixme: This ought to deal with the territory part of the locale
2492 ;; too, for setting things such as calendar holidays, ps-print paper
2493 ;; size, spelling dictionary.
2495 (defun locale-translate (locale)
2496 "Expand LOCALE according to `locale-translation-file-name', if possible.
2498 (if locale-translation-file-name
2499 (with-temp-buffer
2500 (set-buffer-multibyte nil)
2501 (insert-file-contents locale-translation-file-name)
2502 (if (re-search-forward
2504 (buffer-substring (point) (line-end-position))
2505 locale))
2506 locale))
2508 (defun set-locale-environment (&optional locale-name frame)
2509 "Set up multilingual environment for using LOCALE-NAME.
2510 This sets the language environment, the coding system priority,
2511 the default input method and sometimes other things.
2513 LOCALE-NAME should be a string which is the name of a locale supported
2514 by the system. Often it is of the form xx_XX.CODE, where xx is a
2515 language, XX is a country, and CODE specifies a character set and
2517 a locale for Japanese in Japan using the `japanese-iso-8bit'
2518 coding-system. The name may also have a modifier suffix, e.g. `@euro'
2519 or `@cyrillic'.
2521 If LOCALE-NAME is nil, its value is taken from the environment
2522 variables LC_ALL, LC_CTYPE and LANG (the first one that is set).
2524 The locale names supported by your system can typically be found in a
2525 directory named `/usr/share/locale' or `/usr/lib/locale'. LOCALE-NAME
2526 will be translated according to the table specified by
2527 `locale-translation-file-name'.
2529 If FRAME is non-nil, only set the keyboard coding system and the
2530 terminal coding system for the terminal of that frame, and don't
2531 touch session-global parameters like the language environment.
2533 See also `locale-charset-language-names', `locale-language-names',
2534 `locale-preferred-coding-systems' and `locale-coding-system'."
2535 (interactive "sSet environment for locale: ")
2537 ;; Do this at runtime for the sake of binaries possibly transported
2538 ;; to a system without X.
2539 (setq locale-translation-file-name
2540 (let ((files
2541 '("/usr/share/X11/locale/locale.alias" ; e.g. X11R7
2542 "/usr/lib/X11/locale/locale.alias" ; e.g. X11R6.4
2543 "/usr/X11R6/lib/X11/locale/locale.alias" ; XFree86, e.g. RedHat 4.2
2544 "/usr/openwin/lib/locale/locale.alias" ; e.g. Solaris 2.6
2545 ;;
2546 ;; The following name appears after the X-related names above,
2547 ;; since the X-related names are what X actually uses.
2548 "/usr/share/locale/locale.alias" ; GNU/Linux sans X
2549 )))
2550 (while (and files (not (file-exists-p (car files))))
2551 (setq files (cdr files)))
2552 (car files)))
2554 (let ((locale locale-name))
2556 (unless locale
2557 ;; Use the first of these three environment variables
2558 ;; that has a nonempty value.
2559 (let ((vars '("LC_ALL" "LC_CTYPE" "LANG")))
2560 (while (and vars
2561 (= 0 (length locale))) ; nil or empty string
2562 (setq locale (getenv (pop vars) frame)))))
2564 (when locale
2565 (setq locale (locale-translate locale))
2567 ;; Leave the system locales alone if the caller did not specify
2568 ;; an explicit locale name, as their defaults are set from
2569 ;; LC_MESSAGES and LC_TIME, not LC_CTYPE, and the user might not
2570 ;; want to set them to the same value as LC_CTYPE.
2571 (when locale-name
2572 (setq system-messages-locale locale)
2573 (setq system-time-locale locale))
2575 (if (string-match "^[a-z][a-z]" locale)
2576 (setq current-iso639-language (intern (match-string 0 locale)))))
2578 (setq woman-locale
2579 (or system-messages-locale
2580 (let ((msglocale (getenv "LC_MESSAGES" frame)))
2581 (if (zerop (length msglocale))
2582 locale
2583 (locale-translate msglocale)))))
2585 (when locale
2586 (setq locale (downcase locale))
2588 (let ((language-name
2589 (locale-name-match locale locale-language-names))
2590 (charset-language-name
2591 (locale-name-match locale locale-charset-language-names))
2592 (default-eol-type (coding-system-eol-type
2593 (default-value 'buffer-file-coding-system)))
2594 (coding-system
2595 (or (locale-name-match locale locale-preferred-coding-systems)
2596 (when locale
2598 (locale-charset-to-coding-system
2599 (match-string 1 locale)))))))
2601 (if (consp language-name)
2602 ;; locale-language-names specify both lang-env and coding.
2603 ;; But, what specified in locale-preferred-coding-systems
2604 ;; has higher priority.
2605 (setq coding-system (or coding-system
2606 (nth 1 language-name))
2607 language-name (car language-name))
2608 ;; Otherwise, if locale is not listed in locale-language-names,
2609 ;; use what listed in locale-charset-language-names.
2610 (if (not language-name)
2611 (setq language-name charset-language-name)))
2613 ;; If a specific EOL conversion was specified in the default
2614 ;; buffer-file-coding-system, preserve it in the coding system
2615 ;; we will be using from now on.
2616 (if (and (memq default-eol-type '(0 1 2 unix dos mac))
2617 coding-system
2618 (coding-system-p coding-system))
2619 (setq coding-system (coding-system-change-eol-conversion
2620 coding-system default-eol-type)))
2622 (when language-name
2624 ;; Set up for this character set. This is now the right way
2625 ;; to do it for both unibyte and multibyte modes.
2626 (unless frame
2627 (set-language-environment language-name))
2629 ;; If the default enable-multibyte-characters is nil,
2630 ;; we are using single-byte characters,
2631 ;; so the display table and terminal coding system are irrelevant.
2632 (when (default-value 'enable-multibyte-characters)
2633 (set-display-table-and-terminal-coding-system
2634 language-name coding-system frame))
2636 ;; Set the `keyboard-coding-system' if appropriate (tty
2637 ;; only). At least X and MS Windows can generate
2638 ;; multilingual input.
2639 ;; XXX This was disabled unless `window-system', but that
2640 ;; leads to buggy behavior when a tty frame is opened
2641 ;; later. Setting the keyboard coding system has no adverse
2642 ;; effect on X, so let's do it anyway. -- Lorentey
2643 (let ((kcs (or coding-system
2644 (car (get-language-info language-name
2645 'coding-system)))))
2646 (if kcs (set-keyboard-coding-system kcs frame)))
2648 (unless frame
2649 (setq locale-coding-system
2650 (car (get-language-info language-name 'coding-priority)))))
2652 (when (and (not frame)
2653 coding-system
2654 (not (coding-system-equal coding-system
2655 locale-coding-system)))
2656 (prefer-coding-system coding-system)
2657 ;; Fixme: perhaps prefer-coding-system should set this too.
2658 ;; But it's not the time to do such a fundamental change.
2659 (setq default-sendmail-coding-system coding-system)
2660 (setq locale-coding-system coding-system))))
2662 ;; On Windows, override locale-coding-system,
2663 ;; default-file-name-coding-system, keyboard-coding-system,
2664 ;; terminal-coding-system with system codepage.
2665 (when (boundp 'w32-ansi-code-page)
2666 (let ((code-page-coding (intern (format "cp%d" w32-ansi-code-page))))
2667 (when (coding-system-p code-page-coding)
2668 (unless frame (setq locale-coding-system code-page-coding))
2669 (set-keyboard-coding-system code-page-coding frame)
2670 (set-terminal-coding-system code-page-coding frame)
2671 ;; Set default-file-name-coding-system last, so that Emacs
2672 ;; doesn't try to use cpNNNN when it defines keyboard and
2673 ;; terminal encoding. That's because the above two lines
2674 ;; will want to load code-pages.el, where cpNNNN are
2675 ;; defined; if default-file-name-coding-system were set to
2676 ;; cpNNNN while these two lines run, Emacs will want to use
2677 ;; it for encoding the file name it wants to load. And that
2678 ;; will fail, since cpNNNN is not yet usable until
2679 ;; code-pages.el finishes loading.
2680 (setq default-file-name-coding-system code-page-coding))))
2682 (when (eq system-type 'darwin)
2683 ;; On Darwin, file names are always encoded in utf-8, no matter
2684 ;; the locale.
2685 (setq default-file-name-coding-system 'utf-8)
2686 ;; Mac OS X's Terminal.app by default uses utf-8 regardless of
2687 ;; the locale.
2688 (when (and (null window-system)
2689 (equal (getenv "TERM_PROGRAM" frame) "Apple_Terminal"))
2690 (set-terminal-coding-system 'utf-8)
2691 (set-keyboard-coding-system 'utf-8)))
2693 ;; Default to A4 paper if we're not in a C, POSIX or US locale.
2694 ;; (See comments in Flocale_info.)
2695 (unless frame
2696 (let ((locale locale)
2697 (paper (locale-info 'paper)))
2698 (if paper
2699 ;; This will always be null at the time of writing.
2700 (cond
2701 ((equal paper '(216 279))
2702 (setq ps-paper-type 'letter))
2703 ((equal paper '(210 297))
2704 (setq ps-paper-type 'a4)))
2705 (let ((vars '("LC_ALL" "LC_PAPER" "LANG")))
2706 (while (and vars (= 0 (length locale)))
2707 (setq locale (getenv (pop vars) frame))))
2708 (when locale
2709 ;; As of glibc 2.2.5, these are the only US Letter locales,
2710 ;; and the rest are A4.
2711 (setq ps-paper-type
2712 (or (locale-name-match locale '(("c$" . letter)
2713 ("posix$" . letter)
2714 (".._us" . letter)
2715 (".._pr" . letter)
2716 (".._ca" . letter)
2717 ("enu$" . letter) ; Windows
2718 ("esu$" . letter)
2719 ("enc$" . letter)
2720 ("frc$" . letter)))
2721 'a4)))))))
2722 nil)
2723 \f
2724 ;;; Character property
2726 (put 'char-code-property-table 'char-table-extra-slots 5)
2728 (defun define-char-code-property (name table &optional docstring)
2729 "Define NAME as a character code property given by TABLE.
2730 TABLE is a char-table of purpose `char-code-property-table' with
2731 these extra slots:
2732 1st: NAME.
2733 2nd: Function to call to get a property value of a character.
2734 It is called with three arguments CHAR, VAL, and TABLE, where
2735 CHAR is a character, VAL is the value of (aref TABLE CHAR).
2736 3rd: Function to call to put a property value of a character.
2737 It is called with the same arguments as above.
2738 4th: Function to call to get a description string of a property value.
2739 It is called with one argument VALUE, a property value.
2740 5th: Data used by the above functions.
2742 TABLE may be a name of file to load to build a char-table. The
2743 file should contain a call of `define-char-code-property' with a
2744 char-table of the above format as the argument TABLE.
2746 TABLE may also be nil, in which case no property value is pre-assigned.
2748 Optional 3rd argument DOCSTRING is a documentation string of the property.
2750 See also the documentation of `get-char-code-property' and
2751 `put-char-code-property'."
2752 (or (symbolp name)
2753 (error "Not a symbol: %s" name))
2754 (if (char-table-p table)
2755 (or (and (eq (char-table-subtype table) 'char-code-property-table)
2756 (eq (char-table-extra-slot table 0) name))
2757 (error "Invalid char-table: %s" table))
2758 (or (stringp table)
2759 (error "Not a char-table nor a file name: %s" table)))
2760 (if (stringp table) (setq table (purecopy table)))
2761 (let ((slot (assq name char-code-property-alist)))
2762 (if slot
2763 (setcdr slot table)
2764 (setq char-code-property-alist
2765 (cons (cons name table) char-code-property-alist))))
2766 (put name 'char-code-property-documentation (purecopy docstring)))
2768 (defvar char-code-property-table
2769 (make-char-table 'char-code-property-table)
2770 "Char-table containing a property list of each character code.
2771 This table is used for properties not listed in `char-code-property-alist'.
2772 See also the documentation of `get-char-code-property' and
2773 `put-char-code-property'.")
2775 (defun get-char-code-property (char propname)
2776 "Return the value of CHAR's PROPNAME property."
2777 (let ((table (unicode-property-table-internal propname)))
2778 (if table
2779 (let ((func (char-table-extra-slot table 1)))
2780 (if (functionp func)
2781 (funcall func char (aref table char) table)
2782 (get-unicode-property-internal table char)))
2783 (plist-get (aref char-code-property-table char) propname))))
2785 (defun put-char-code-property (char propname value)
2786 "Store CHAR's PROPNAME property with VALUE.
2787 It can be retrieved with `(get-char-code-property CHAR PROPNAME)'."
2788 (let ((table (unicode-property-table-internal propname)))
2789 (if table
2790 (let ((func (char-table-extra-slot table 2)))
2791 (if (functionp func)
2792 (funcall func char value table)
2793 (put-unicode-property-internal table char value)))
2794 (let* ((plist (aref char-code-property-table char))
2795 (x (plist-put plist propname value)))
2796 (or (eq x plist)
2797 (aset char-code-property-table char x))))
2798 value))
2800 (defun char-code-property-description (prop value)
2801 "Return a description string of character property PROP's value VALUE.
2802 If there's no description string for VALUE, return nil."
2803 (let ((table (unicode-property-table-internal prop)))
2804 (if table
2805 (let ((func (char-table-extra-slot table 3)))
2806 (if (functionp func)
2807 (funcall func value))))))
2809 \f
2810 ;; Pretty description of encoded string
2812 ;; Alist of ISO 2022 control code vs the corresponding mnemonic string.
2813 (defconst iso-2022-control-alist
2821 (defun encoded-string-description (str coding-system)
2822 "Return a pretty description of STR that is encoded by CODING-SYSTEM."
2823 (setq str (string-as-unibyte str))
2824 (mapconcat
2825 (if (and coding-system (eq (coding-system-type coding-system) 'iso-2022))
2826 ;; Try to get a pretty description for ISO 2022 escape sequences.
2827 (function (lambda (x) (or (cdr (assq x iso-2022-control-alist))
2828 (format "#x%02X" x))))
2829 (function (lambda (x) (format "#x%02X" x))))
2830 str " "))
2832 (defun encode-coding-char (char coding-system &optional charset)
2833 "Encode CHAR by CODING-SYSTEM and return the resulting string.
2834 If CODING-SYSTEM can't safely encode CHAR, return nil.
2835 The 3rd optional argument CHARSET, if non-nil, is a charset preferred
2836 on encoding."
2837 (let* ((str1 (string-as-multibyte (string char)))
2838 (str2 (string-as-multibyte (string char char)))
2839 (found (find-coding-systems-string str1))
2840 enc1 enc2 i1 i2)
2841 (if (and (consp found)
2842 (eq (car found) 'undecided))
2843 str1
2844 (when (memq (coding-system-base coding-system) found)
2845 ;; We must find the encoded string of CHAR. But, just encoding
2846 ;; CHAR will put extra control sequences (usually to designate
2847 ;; ASCII charset) at the tail if type of CODING is ISO 2022.
2848 ;; To exclude such tailing bytes, we at first encode one-char
2849 ;; string and two-char string, then check how many bytes at the
2850 ;; tail of both encoded strings are the same.
2852 (when charset
2853 (put-text-property 0 1 'charset charset str1)
2854 (put-text-property 0 2 'charset charset str2))
2855 (setq enc1 (encode-coding-string str1 coding-system)
2856 i1 (length enc1)
2857 enc2 (encode-coding-string str2 coding-system)
2858 i2 (length enc2))
2859 (while (and (> i1 0) (= (aref enc1 (1- i1)) (aref enc2 (1- i2))))
2860 (setq i1 (1- i1) i2 (1- i2)))
2862 ;; Now (substring enc1 i1) and (substring enc2 i2) are the same,
2863 ;; and they are the extra control sequences at the tail to
2864 ;; exclude.
2865 (substring enc2 0 i2)))))
2867 ;; Backwards compatibility. These might be better with :init-value t,
2868 ;; but that breaks loadup.
2869 (define-minor-mode unify-8859-on-encoding-mode
2870 "Exists only for backwards compatibility."
2871 :group 'mule
2872 :global t)
2873 ;; Doc said "obsolete" in 23.1, this statement only added in 24.1.
2874 (make-obsolete 'unify-8859-on-encoding-mode "don't use it." "23.1")
2876 (define-minor-mode unify-8859-on-decoding-mode
2877 "Exists only for backwards compatibility."
2878 :group 'mule
2879 :global t)
2880 ;; Doc said "obsolete" in 23.1, this statement only added in 24.1.
2881 (make-obsolete 'unify-8859-on-decoding-mode "don't use it." "23.1")
2883 (defvar nonascii-insert-offset 0)
2884 (make-obsolete-variable 'nonascii-insert-offset "do not use it." "23.1")
2885 (defvar nonascii-translation-table nil)
2886 (make-obsolete-variable 'nonascii-translation-table "do not use it." "23.1")
2888 (defvar ucs-names nil
2889 "Alist of cached (CHAR-NAME . CHAR-CODE) pairs.")
2891 (defun ucs-names ()
2892 "Return alist of (CHAR-NAME . CHAR-CODE) pairs cached in `ucs-names'."
2893 (or ucs-names
2894 (let ((bmp-ranges
2895 '((#x0000 . #x33FF)
2896 ;; (#x3400 . #x4DBF) CJK Ideographs Extension A
2897 (#x4DC0 . #x4DFF)
2898 ;; (#x4E00 . #x9FFF) CJK Unified Ideographs
2899 (#xA000 . #xD7FF)
2900 ;; (#xD800 . #xFAFF) Surrogate/Private
2901 (#xFB00 . #xFFFD)))
2902 (upper-ranges
2903 '((#x10000 . #x134FF)
2904 ;; (#x13500 . #x167FF) unused
2905 (#x16800 . #x16A3F)
2906 ;; (#x16A40 . #x1AFFF) unused
2907 (#x1B000 . #x1B0FF)
2908 ;; (#x1B100 . #x1CFFF) unused
2909 (#x1D000 . #x1FFFF)
2910 ;; (#x20000 . #xDFFFF) CJK Ideograph Extension A, B, etc, unused
2911 (#xE0000 . #xE01FF)))
2912 (gc-cons-threshold 10000000)
2913 c end name names)
2914 (dolist (range bmp-ranges)
2915 (setq c (car range)
2916 end (cdr range))
2917 (while (<= c end)
2918 (if (setq name (get-char-code-property c 'name))
2919 (push (cons name c) names))
2920 (if (setq name (get-char-code-property c 'old-name))
2921 (push (cons name c) names))
2922 (setq c (1+ c))))
2923 (dolist (range upper-ranges)
2924 (setq c (car range)
2925 end (cdr range))
2926 (while (<= c end)
2927 (if (setq name (get-char-code-property c 'name))
2928 (push (cons name c) names))
2929 (setq c (1+ c))))
2930 (setq ucs-names names))))
2932 (defun read-char-by-name (prompt)
2933 "Read a character by its Unicode name or hex number string.
2934 Display PROMPT and read a string that represents a character by its
2935 Unicode property `name' or `old-name'.
2937 This function returns the character as a number.
2939 You can type a few of the first letters of the Unicode name and
2940 use completion. If you type a substring of the Unicode name
2941 preceded by an asterisk `*' and use completion, it will show all
2942 the characters whose names include that substring, not necessarily
2943 at the beginning of the name.
2945 This function also accepts a hexadecimal number of Unicode code
2946 point or a number in hash notation, e.g. #o21430 for octal,
2947 #x2318 for hex, or #10r8984 for decimal."
2948 (let* ((completion-ignore-case t)
2949 (input (completing-read
2950 prompt
2951 (lambda (string pred action)
2952 (if (eq action 'metadata)
2953 '(metadata (category . unicode-name))
2954 (complete-with-action action (ucs-names) string pred))))))
2955 (cond
2957 (string-to-number input 16))
2959 (read input))
2960 (t
2961 (cdr (assoc-string input (ucs-names) t))))))
2963 (defun ucs-insert (character &optional count inherit)
2964 "Insert COUNT copies of CHARACTER of the given Unicode code point.
2965 Interactively, prompts for a Unicode character name or a hex number
2966 using `read-char-by-name'.
2968 You can type a few of the first letters of the Unicode name and
2969 use completion. If you type a substring of the Unicode name
2970 preceded by an asterisk `*' and use completion, it will show all
2971 the characters whose names include that substring, not necessarily
2972 at the beginning of the name.
2974 This function also accepts a hexadecimal number of Unicode code
2975 point or a number in hash notation, e.g. #o21430 for octal,
2976 #x2318 for hex, or #10r8984 for decimal.
2978 The optional third arg INHERIT (non-nil when called interactively),
2979 says to inherit text properties from adjoining text, if those
2980 properties are sticky."
2981 (interactive
2982 (list (read-char-by-name "Unicode (name or hex): ")
2983 (prefix-numeric-value current-prefix-arg)
2984 t))
2985 (unless count (setq count 1))
2986 (if (and (stringp character)
2988 (setq character (string-to-number character 16)))
2989 (cond
2990 ((null character)
2991 (error "Not a Unicode character"))
2992 ((not (integerp character))
2993 (error "Not a Unicode character code: %S" character))
2994 ((or (< character 0) (> character #x10FFFF))
2995 (error "Not a Unicode character code: 0x%X" character)))
2996 (if inherit
2997 (dotimes (i count) (insert-and-inherit character))
2998 (dotimes (i count) (insert character))))
3002 ;;; mule-cmds.el ends here