| blob | f29e11b8e2639c52b47a1d0fc45215bd85d2f587 |
1 ;;; mule-cmds.el --- commands for mulitilingual environment -*-coding: iso-2022-7bit -*-
2 ;; Copyright (C) 1995, 2003 Electrotechnical Laboratory, JAPAN.
3 ;; Licensed to the Free Software Foundation.
4 ;; Copyright (C) 2000, 2001, 2002, 2003 Free Software Foundation, Inc.
6 ;; Keywords: mule, multilingual
8 ;; This file is part of GNU Emacs.
10 ;; GNU Emacs is free software; you can redistribute it and/or modify
11 ;; it under the terms of the GNU General Public License as published by
12 ;; the Free Software Foundation; either version 2, or (at your option)
13 ;; any later version.
15 ;; GNU Emacs is distributed in the hope that it will be useful,
16 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
17 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18 ;; GNU General Public License for more details.
20 ;; You should have received a copy of the GNU General Public License
21 ;; along with GNU Emacs; see the file COPYING. If not, write to the
22 ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330,
23 ;; Boston, MA 02111-1307, USA.
25 ;;; Commentary:
27 ;;; Code:
31 ;;; MULE related key bindings and menus.
36 ;; Keep "C-x C-m ..." for mule specific commands.
69 t)
72 t)
75 t)
80 t)
89 t)
92 t)
95 describe-language-environment-map
97 t)
101 t)
104 t)
111 t)
116 t)
119 t)
123 t)
126 :enable buffer-file-name
128 t)
132 t)
135 t)
140 t)
145 t)
148 t)
153 t)
158 t)
164 t)
168 [Default] '(menu-item "Default" setup-specified-language-environment))
170 (define-key describe-language-environment-map
173 ;; This should be a single character key binding because users use it
174 ;; very frequently while editing multilingual text. Now we can use
175 ;; only two such keys: "\C-\\" and "\C-^", but the latter is not
176 ;; convenient because it requires shifting on most keyboards. An
177 ;; alternative is "\C-\]" which is now bound to `abort-recursive-edit'
178 ;; but it won't be used that frequently.
181 ;; This is no good because people often type Shift-SPC
182 ;; meaning to type SPC. -- rms.
183 ;; ;; Here's an alternative key binding for X users (Shift-SPACE).
184 ;; (define-key global-map [?\S- ] 'toggle-input-method)
186 ;;; Mule related hyperlinks.
194 ;; Note starting with word-syntax character:
198 "Return a coding system which differs from CODING-SYSTEM in eol conversion.
199 The returned coding system converts end-of-line by EOL-TYPE
200 but text as the same way as CODING-SYSTEM.
201 EOL-TYPE should be `unix', `dos', `mac', or nil.
202 If EOL-TYPE is nil, the returned coding system detects
203 how end-of-line is formatted automatically while decoding.
205 EOL-TYPE can be specified by an integer 0, 1, or 2.
206 They means `unix', `dos', and `mac' respectively."
215 coding-system
219 base
221 coding-system
227 "Return a coding system which differs from CODING-SYSTEM in text conversion.
228 The returned coding system converts text by CODING
229 but end-of-line as the same way as CODING-SYSTEM.
230 If CODING is nil, the returned coding system detects
231 how text is formatted automatically while decoding."
238 "Change whether this buffer uses multibyte characters.
239 With arg, use multibyte characters if the arg is positive.
241 Note that this command does not convert the byte contents of
242 the buffer; it only changes the way those bytes are interpreted.
243 In general, therefore, this command *changes* the sequence of
244 characters that the current buffer contains.
246 We suggest you avoid using use this command unless you know what you
247 are doing. If you use it by mistake, and the buffer is now displayed
248 wrong, use this command again to toggle back to the right mode."
257 "Display the HELLO file which list up many languages and characters."
259 ;; We have to decode the file in any environment.
265 "Execute an I/O command using the specified coding system."
269 t))
270 buffer-file-coding-system)))
275 default))))
279 prefix)
284 ;; Process keys bound in `universal-argument-map'.
290 ;; Have to bind `last-command-char' here so that
291 ;; `digit-argument', for instance, can compute the
292 ;; prefix arg.
296 ;; This is the final call to `universal-argument-other-key', which
297 ;; set's the final `prefix-arg.
301 ;; Read the command to execute with the given prefix arg.
314 "Set default value of various coding systems to CODING-SYSTEM.
315 This sets the following coding systems:
316 o coding system of a newly created buffer
317 o default coding system for subprocess I/O
318 This also sets the following values:
319 o default value used as `file-name-coding-system' for converting file names.
320 o default value for the command `set-terminal-coding-system' (not on MSDOS)
321 o default value for the command `set-keyboard-coding-system'."
326 ;; If coding-system is nil, honor that on MS-DOS as well, so
327 ;; that they could reset the terminal coding system.
331 ;; Preserve eol-type from existing default-process-coding-systems.
332 ;; On non-unix-like systems in particular, these may have been set
333 ;; carefully by the user, or by the startup code, to deal with the
334 ;; users shell appropriately, so should not be altered by changing
335 ;; language environment.
349 "Add CODING-SYSTEM at the front of the priority list for automatic detection.
350 This also sets the following coding systems:
351 o coding system of a newly created buffer
352 o default coding system for subprocess I/O
353 This also sets the following values:
354 o default value used as `file-name-coding-system' for converting file names.
355 o default value for the command `set-terminal-coding-system' (not on MSDOS)
356 o default value for the command `set-keyboard-coding-system'
358 If CODING-SYSTEM specifies a certain type of EOL conversion, the coding
359 systems set by this function will use that type of EOL conversion.
361 This command does not change the default value of terminal coding system
362 for MS-DOS terminal, because DOS terminals only support a single coding
363 system, and Emacs automatically sets the default to that coding system at
364 startup.
366 A coding system that requires automatic detection of text
367 encoding (e.g. undecided, unix) can't be preferred.
369 See also `coding-category-list' and `coding-system-category'."
377 ;; CODING-SYSTEM is no-conversion or undecided.
382 ;; We must change the order.
386 base coding-system))
387 ;; If they asked for specific EOL conversion, honor that.
395 "If non-nil, a predicate function to sort coding systems.
397 It is called with two coding systems, and should return t if the first
403 "Sort coding system list CODINGS by a priority of each coding system.
405 If a coding system is most preferred, it has the highest priority.
406 Otherwise, a coding system corresponds to some MIME charset has higher
407 priorities. Among them, a coding system included in `coding-system'
408 key of the current language environment has higher priorities. See
409 also the documentation of `language-info-alist'.
411 If the variable `sort-coding-systems-predicate' (which see) is
412 non-nil, it is used to sort CODINGS in the different way than above."
416 coding-category-list))
423 ;; We calculate the priority number 0..255 by
424 ;; using the 8 bits PMMLCEII as this:
425 ;; P: 1 iff most preferred.
426 ;; MM: greater than 0 iff mime-charset.
427 ;; L: 1 iff one of the current lang. env.'s codings.
428 ;; C: 1 iff one of codings listed in the category list.
429 ;; E: 1 iff not XXX-with-esc
430 ;; II: if iso-2022 based, 0..3, else 1.
435 ;; Prefer coding systems corresponding to a
436 ;; MIME charset.
438 ;; Lower utf-16 priority so that we
439 ;; normally prefer utf-8 to it, and put
440 ;; x-ctext below that.
455 ;; For ISO based coding systems, prefer
456 ;; one that doesn't use escape sequences.
463 0
471 "Return a list of proper coding systems to encode a text between FROM and TO.
472 If FROM is a string, find coding systems in that instead of the buffer.
473 All coding systems in the list can safely encode any multibyte characters
474 in the text.
476 If the text contains no multibyte characters, return a list of a single
477 element `undecided'."
480 ;; The text contains only ASCII characters. Any coding
481 ;; systems are safe.
483 ;; We need copy-sequence because sorting will alter the argument.
487 "Return a list of proper coding systems to encode STRING.
488 All coding systems in the list can safely encode any multibyte characters
489 in STRING.
491 If STRING contains no multibyte characters, return a list of a single
492 element `undecided'."
496 "Return a list of proper coding systems to encode characters of CHARSETS.
497 CHARSETS is a list of character sets.
498 It actually checks at most the first 96 characters of each charset.
499 So, if a charset of dimension two is included in CHARSETS, the value may
500 contain a coding system that can't encode all characters of the charset."
510 charset l str)
534 codings))))
537 "Find multibyte characters in the region specified by FROM and TO.
538 If FROM is a string, find multibyte characters in the string.
539 The return value is an alist of the following format:
540 ((CHARSET COUNT CHAR ...) ...)
541 where
542 CHARSET is a character set,
543 COUNT is a number of characters,
544 CHARs are the characters found from the character set.
545 Optional 3rd arg MAXCOUNT limits how many CHARs are put in the above list.
546 Optional 4th arg EXCLUDE is a list of character sets to be ignored.
548 For invalid characters, CHARs are actually strings."
550 charset char)
559 eight-bit-control eight-bit-graphic))
591 "Search forward from point for a character that is not encodable.
592 It asks which coding system to check.
593 If such a character is found, set point after that character.
594 Otherwise, don't move point.
596 When called from a program, the value is a position of the found character,
597 or nil if all characters are encodable."
602 default))))
607 pos))
611 "Most recent coding system explicitly specified by the user when asked.
612 This variable is set whenever Emacs asks the user which coding system
613 to use in order to write a file. If you set it to nil explicitly,
614 then call `write-region', then afterward this variable will be non-nil
618 "If non-nil, a function to control the behaviour of coding system selection.
619 The meaning is the same as the argument ACCEPT-DEFAULT-P of the
620 function `select-safe-coding-system' (which see). This variable
624 accept-default-p file)
625 "Ask a user to select a safe coding system from candidates.
626 The candidates of coding systems which can safely encode a text
627 between FROM and TO are shown in a popup window. Among them, the most
628 proper one is suggested as the default.
630 The list of `buffer-file-coding-system' of the current buffer,
631 the `default-buffer-file-coding-system', and the
632 most preferred coding system (if it corresponds to a MIME charset) is
633 treated as the default coding system list. Among them, the first one
634 that safely encodes the text is normally selected silently and
635 returned without any user interaction. See also the command
636 `prefer-coding-system'.
638 However, the user is queried if the chosen coding system is
639 inconsistent with what would be selected by `set-auto-coding' from
640 coding cookies &c. if the contents of the region were read from a
641 file. (That could lead to data corruption in a file subsequently
642 re-visited and edited.)
644 Optional 3rd arg DEFAULT-CODING-SYSTEM specifies a coding system or a
645 list of coding systems to be prepended to the default coding system
646 list. However, if DEFAULT-CODING-SYSTEM is a list and the first
647 element is t, the cdr part is used as the defualt coding system list,
648 i.e. `buffer-file-coding-system', `default-buffer-file-coding-system',
649 and the most preferred coding system are not used.
651 Optional 4th arg ACCEPT-DEFAULT-P, if non-nil, is a function to
652 determine the acceptability of the silently selected coding system.
653 It is called with that coding system, and should return nil if it
654 should not be silently selected and thus user interaction is required.
656 Optional 5th arg FILE is the file name to use for this purpose.
657 That is different from `buffer-file-name' when handling `write-region'
658 \(for example).
660 The variable `select-safe-coding-system-accept-default-p', if
661 non-nil, overrides ACCEPT-DEFAULT-P.
663 Kludgy feature: if FROM is a string, the string is the target text,
664 and TO is ignored."
674 ;; Change elements of the list to (coding . base-coding).
677 default-coding-system))
679 ;; From now on, the list of defaults is reversed.
683 ;; If buffer-file-coding-system is not nil nor undecided, append it
684 ;; to the defaults.
690 default-coding-system))))
692 ;; If default-buffer-file-coding-system is not nil nor undecided,
693 ;; append it to the defaults.
699 default-coding-system))))
701 ;; If the most preferred coding system has the property mime-charset,
702 ;; append it to the defaults.
704 preferred base)
712 default-coding-system)))))
720 safe rejected unsafe)
722 ;; Any coding system is ok.
724 ;; Classify the defaults into safe, rejected, and unsafe.
735 ;; If all the defaults failed, ask a user.
737 ;; At first, if some defaults are unsafe, record at most 11
738 ;; problematic characters and their positions for them by turning
739 ;; (CODING ...)
740 ;; into
741 ;; ((CODING (POS . CHAR) (POS . CHAR) ...) ...)
752 unsafe))
760 unsafe))))
762 ;; Change each safe coding system to the corresponding
763 ;; mime-charset name if it is also a coding system. Such a name
764 ;; is more friendly to users.
766 mime-charset)
773 ;; Don't offer variations with locking shift, which you
774 ;; basically never want.
783 ;; Remove raw-text, emacs-mule and no-conversion unless nothing
784 ;; else is available.
793 ;; If some defaults are unsafe, make sure the offending
794 ;; buffer is displayed.
798 unsafe))))
799 ;; Then ask users to select one from CODINGS while showing
800 ;; the reason why none of the defaults are not used.
810 "These default coding systems were tried to encode"
820 default-coding-system)
825 but it is not recommended for encoding text in this context,
859 'help-echo
860 "mouse-2, RET: jump to this character"
861 'help-function func1
864 "..."
866 'help-echo
867 "mouse-2, RET: next unencodable character"
868 'help-function func2
873 unsafe)
881 "\
886 codings)
892 ;; Read a coding system.
897 default-coding-system)
898 default-coding-system))
912 ;; Check we're not inconsistent with what `coding:' spec &c would
913 ;; give when file is re-read.
914 ;; But don't do this if we explicitly ignored the cookie
915 ;; by using `find-file-literally'.
925 ;; Don't barf if writing a compressed file, say.
926 ;; This check perhaps isn't ideal, but is probably
927 ;; the best thing to do.
933 %s specified by file contents. Really save (else edit coding cookies \
936 coding-system))
941 "Return a coding system to encode the outgoing message of the current buffer.
942 It at first tries the first coding system found in these variables
943 in this order:
944 (1) local value of `buffer-file-coding-system'
945 (2) value of `sendmail-coding-system'
946 (3) value of `default-sendmail-coding-system'
947 (4) value of `default-buffer-file-coding-system'
948 If the found coding system can't encode the current buffer,
949 or none of them are bound to a coding system,
950 it asks the user to select a proper coding system."
952 buffer-file-coding-system)
953 sendmail-coding-system
954 default-sendmail-coding-system
955 default-buffer-file-coding-system)))
957 ;; We should never use no-conversion for outgoing mail.
963 coding)))
964 \f
965 ;;; Language support stuff.
968 "Alist of language environment definitions.
969 Each element looks like:
970 (LANGUAGE-NAME . ((KEY . INFO) ...))
971 where LANGUAGE-NAME is a string, the name of the language environment,
972 KEY is a symbol denoting the kind of information, and
973 INFO is the data associated with KEY.
974 Meaningful values for KEY include
976 documentation value is documentation of what this language environment
977 is meant for, and how to use it.
978 charset value is a list of the character sets used by this
979 language environment.
980 sample-text value is an expression which is evalled to generate
981 a line of text written using characters appropriate
982 for this language environment.
983 setup-function value is a function to call to switch to this
984 language environment.
985 exit-function value is a function to call to leave this
986 language environment.
987 coding-system value is a list of coding systems that are good
988 for saving text written in this language environment.
989 This list serves as suggestions to the user;
990 in effect, as a kind of documentation.
991 coding-priority value is a list of coding systems for this language
992 environment, in order of decreasing priority.
993 This is used to set up the coding system priority
994 list when you switch to this language environment.
995 nonascii-translation
996 value is a translation table to be set in the
997 variable `nonascii-translation-table' in this
998 language environment, or a character set from
999 which `nonascii-insert-offset' is calculated.
1000 input-method value is a default input method for this language
1001 environment.
1002 features value is a list of features requested in this
1003 language environment.
1005 The following keys take effect only when multibyte characters are
1006 globally disabled, i.e. the value of `default-enable-multibyte-characters'
1007 is nil.
1009 unibyte-syntax value is a library name to load to set
1010 unibyte 8-bit character syntaxes for this
1011 language environment.
1013 unibyte-display value is a coding system to encode characters
1014 for the terminal. Characters in the range
1015 of 160 to 255 display not as octal escapes,
1016 but as non-ASCII characters in this language
1020 "Return information listed under KEY for language environment LANG-ENV.
1021 KEY is a symbol denoting the kind of information.
1022 For a list of useful values for KEY and their meanings,
1023 see `language-info-alist'."
1031 "Modify part of the definition of language environment LANG-ENV.
1032 Specifically, this stores the information INFO under KEY
1033 in the definition of this language environment.
1034 KEY is a symbol denoting the kind of information.
1035 INFO is the value for that information.
1037 For a list of useful values for KEY and their meanings,
1038 see `language-info-alist'."
1052 ;; Update the custom-type of `current-language-environment'.
1060 "Store ALIST as the definition of language environment LANG-ENV.
1061 ALIST is an alist of KEY and INFO values. See the documentation of
1062 `language-info-alist' for the meanings of KEY and INFO.
1064 Optional arg PARENTS is a list of parent menu names; it specifies
1065 where to put this language environment in the
1066 Describe Language Environment and Set Language Environment menus.
1068 in the European submenu in each of those two menus."
1075 map parent-symbol parent prompt)
1081 ;; This prompt string is for define-prefix-command, so
1082 ;; that the map it creates will be suitable for a menu.
1103 ;; Set up menu items for this language env.
1116 "Read a language environment name which has information for KEY.
1117 If KEY is nil, read any language environment.
1118 Prompt with PROMPT. DEFAULT is the default choice of language environment.
1119 This returns a language environment name as a string."
1122 language-info-alist
1125 t nil nil default)))
1129 name)))
1130 \f
1131 ;;; Multilingual input methods.
1133 "LEIM: Libraries of Emacs Input Methods."
1137 "Name of LEIM list file.
1138 This file contains a list of libraries of Emacs input methods (LEIM)
1139 in the format of Lisp expression for registering each input method.
1143 ";;; %s -- list of LEIM (Library of Emacs Input Method) -*-coding: iso-2022-7bit;-*-
1144 ;;
1145 ;; This file contains a list of LEIM (Library of Emacs Input Method)
1146 ;; methods in the same directory as this file. Loading this file
1147 ;; registers all the input methods in Emacs.
1148 ;;
1149 ;; Each entry has the form:
1150 ;; (register-input-method
1151 ;; INPUT-METHOD LANGUAGE-NAME ACTIVATE-FUNC
1152 ;; TITLE DESCRIPTION
1153 ;; ARG ...)
1154 ;; See the function `register-input-method' for the meanings of the arguments.
1155 ;;
1156 ;; If this directory is included in load-path, Emacs automatically
1157 ;; loads this file at startup time.
1159 "
1160 leim-list-file-name)
1164 "Regexp matching head of each entry in LEIM list file.
1169 "List of functions to call to update LEIM list file.
1173 "Update LEIM list file in directories DIRS."
1180 "The current input method for multilingual text.
1191 "*Default input method for multilingual text (a string).
1192 This is the input method activated automatically by the command
1197 :completion-ignore-case t
1198 :complete-function widget-string-complete
1199 :completion-alist input-method-alist
1211 "Function to call for inactivating the current input method.
1212 Every input method should set this to an appropriate value when activated.
1213 This function is called with no argument.
1215 This function should never change the value of `current-input-method'.
1221 "Function to call for describing the current input method.
1227 "Alist of input method names vs how to use them.
1228 Each element has the form:
1229 (INPUT-METHOD LANGUAGE-ENV ACTIVATE-FUNC TITLE DESCRIPTION ARGS...)
1233 "Register INPUT-METHOD as an input method for language environment LANG-ENV.
1234 INPUT-METHOD and LANG-ENV are symbols or strings.
1236 The remaining arguments are:
1237 ACTIVATE-FUNC, TITLE, DESCRIPTION, and ARGS...
1238 ACTIVATE-FUNC is a function to call to activate this method.
1239 TITLE is a string to show in the mode line when this method is active.
1240 DESCRIPTION is a string describing this method and what it is good for.
1241 The ARGS, if any, are passed as arguments to ACTIVATE-FUNC.
1242 All told, the arguments to ACTIVATE-FUNC are INPUT-METHOD and the ARGS.
1245 created at Emacs build time, registering all Quail input methods
1246 contained in the Emacs distribution.
1248 In case you want to register a new Quail input method by yourself, be
1249 careful to use the same input method title as given in the third
1250 parameter of `quail-define-package'. (If the values are different, the
1251 string specified in this function takes precedence.)
1253 The commands `describe-input-method' and `list-input-methods' need
1254 these duplicated values to show some information about input methods
1255 without loading the relevant Quail packages."
1268 "Read a name of input method from a minibuffer prompting with PROMPT.
1269 If DEFAULT is non-nil, use that as the default,
1270 and substitute it into PROMPT at the first `%s'.
1271 If INHIBIT-NULL is non-nil, null input signals an error.
1273 The return value is a string."
1277 ;; As it is quite normal to change input method in the
1278 ;; minibuffer, we must enable it even if
1279 ;; enable-recursive-minibuffers is currently nil.
1281 ;; This binding is necessary because input-method-history is
1282 ;; buffer local.
1284 nil t nil 'input-method-history
1285 default)))
1289 input-method
1294 "Switch to input method INPUT-METHOD for the current buffer.
1295 If some other input method is already active, turn it off first.
1296 If INPUT-METHOD is nil, deactivate any current input method."
1323 "Turn off the current input method."
1336 input-method-function nil
1337 current-input-method-title nil)
1341 "Select and activate input method INPUT-METHOD for the current buffer.
1342 This also sets the default input method to the one you specify.
1343 If INPUT-METHOD is nil, this function turns off the input method, and
1344 also causes you to be prompted for a name of an input method the next
1353 default t))))
1358 default-input-method)
1361 "Enable or disable multilingual text input method for the current buffer.
1362 Only one input method can be enabled at any time in a given buffer.
1364 The normal action is to enable an input method if none was
1365 enabled, and disable the current one otherwise. Which input method
1366 to enable can be determined in various ways--either the one most
1367 recently used, or the one specified by `default-input-method', or
1368 as a last resort by reading the name of an input method in the
1369 minibuffer.
1371 With a prefix argument, read an input method name with the minibuffer
1372 and enable that one. The default is the most recent input method specified
1373 \(not including the currently active input method, if any)."
1387 default t))
1388 default))
1396 "Describe input method INPUT-METHOD."
1426 "Describe the input method currently in use.
1427 This is a subroutine for `describe-input-method'."
1433 current-input-method)
1438 "Read a multilingual string from minibuffer, prompting with string PROMPT.
1439 The input method selected last time is activated in minibuffer.
1440 If optional second arg INITIAL-INPUT is non-nil, insert it in the minibuffer
1441 initially.
1442 Optional 3rd argument INPUT-METHOD specifies the input method
1443 to be activated instead of the one selected last time. It is a symbol
1444 or a string."
1447 current-input-method
1448 default-input-method
1459 ;; Variables to control behavior of input methods. All input methods
1460 ;; should react to these variables.
1463 "*A flag to control extra guidance given by input methods.
1464 The value should be nil, t, `complex-only', or `default'.
1466 The extra guidance is done by showing list of available keys in echo
1467 area. When you use the input method in the minibuffer, the guidance
1468 is shown at the bottom short window (split from the existing window).
1470 If the value is t, extra guidance is always given, if the value is
1471 nil, extra guidance is always suppressed.
1473 If the value is `complex-only', only complex input methods such as
1474 `chinese-py' and `japanese' give extra guidance.
1476 If the value is `default', complex input methods always give extra
1477 guidance, but simple input methods give it only when you are not in
1478 the minibuffer.
1480 See also the variable `input-method-highlight-flag'."
1486 "*If this flag is non-nil, input methods highlight partially-entered text.
1487 For instance, while you are in the middle of a Quail input method sequence,
1488 the text inserted so far is temporarily underlined.
1489 The underlining goes away when you finish or abort the input method sequence.
1490 See also the variable `input-method-verbose-flag'."
1495 "Normal hook run just after an input method is activated.
1497 The variable `current-input-method' keeps the input method name
1501 "Normal hook run just after an input method is inactivated.
1503 The variable `current-input-method' still keeps the input method name
1510 "This flag controls when an input method returns.
1511 Usually, the input method does not return while there's a possibility
1512 that it may find a different translation if a user types another key.
1513 But, it this flag is non-nil, the input method returns as soon as
1517 "This flag controls how an input method shows an intermediate key sequence.
1518 Usually, the input method inserts the intermediate key sequence,
1519 or candidate translations corresponding to the sequence,
1520 at point in the current buffer.
1524 "This flag controls the behaviour of an input method on invalid key input.
1525 Usually, when a user types a key which doesn't start any character
1526 handled by the input method, the key is handled by turning off the
1527 input method temporarily. After that key, the input method is re-enabled.
1530 \f
1532 "Normal hook run after some language environment is set.
1534 When you set some hook function here, that effect usually should not
1535 be inherited to another language environment. So, you had better set
1536 another function in `exit-language-environment-hook' (which see) to
1540 "Normal hook run after exiting from some language environment.
1541 When this hook is run, the variable `current-language-environment'
1542 is still bound to the language environment being exited.
1544 This hook is mainly used for canceling the effect of
1550 "Switch to a specified language environment."
1563 "The last language environment specified with `set-language-environment'.
1565 to using the function `set-language-environment'."
1572 current-language-environment)
1573 language-info-alist))
1575 ;; custom type will be updated with `set-language-info'.
1586 "Reset multilingual environment of Emacs to the default status.
1588 The default status is as follows:
1590 The default value of `buffer-file-coding-system' is nil.
1591 The default coding system for process I/O is nil.
1592 The default value for the command `set-terminal-coding-system' is nil.
1593 The default value for the command `set-keyboard-coding-system' is nil.
1595 The order of priorities of coding categories and the coding system
1596 bound to each category are as follows
1597 coding category coding system
1598 --------------------------------------------------
1599 coding-category-iso-8-1 iso-latin-1
1600 coding-category-iso-8-2 iso-latin-1
1601 coding-category-utf-8 mule-utf-8
1602 coding-category-utf-16-be mule-utf-16be-with-signature
1603 coding-category-utf-16-le mule-utf-16le-with-signature
1604 coding-category-iso-7-tight iso-2022-jp
1605 coding-category-iso-7 iso-2022-7bit
1606 coding-category-iso-7-else iso-2022-7bit-lock
1607 coding-category-iso-8-else iso-2022-8bit-ss2
1608 coding-category-emacs-mule emacs-mule
1609 coding-category-raw-text raw-text
1610 coding-category-sjis japanese-shift-jis
1611 coding-category-big5 chinese-big5
1612 coding-category-ccl nil
1613 coding-category-binary no-conversion"
1615 ;; This function formerly set default-enable-multibyte-characters to t,
1616 ;; but that is incorrect. It should not alter the unibyte/multibyte choice.
1619 coding-category-iso-7 'iso-2022-7bit
1620 coding-category-iso-8-1 'iso-latin-1
1621 coding-category-iso-8-2 'iso-latin-1
1622 coding-category-iso-7-else 'iso-2022-7bit-lock
1623 coding-category-iso-8-else 'iso-2022-8bit-ss2
1624 coding-category-emacs-mule 'emacs-mule
1625 coding-category-raw-text 'raw-text
1626 coding-category-sjis 'japanese-shift-jis
1627 coding-category-big5 'chinese-big5
1628 coding-category-utf-16-be 'mule-utf-16be-with-signature
1629 coding-category-utf-16-le 'mule-utf-16le-with-signature
1630 coding-category-utf-8 'mule-utf-8
1631 coding-category-ccl nil
1636 coding-category-iso-8-2
1637 coding-category-utf-8
1638 coding-category-utf-16-be
1639 coding-category-utf-16-le
1640 coding-category-iso-7-tight
1641 coding-category-iso-7
1642 coding-category-iso-7-else
1643 coding-category-iso-8-else
1644 coding-category-emacs-mule
1645 coding-category-raw-text
1646 coding-category-sjis
1647 coding-category-big5
1648 coding-category-ccl
1649 coding-category-binary))
1656 ;; Preserve eol-type from existing default-process-coding-systems.
1657 ;; On non-unix-like systems in particular, these may have been set
1658 ;; carefully by the user, or by the startup code, to deal with the
1659 ;; users shell appropriately, so should not be altered by changing
1660 ;; language environment.
1662 ;; When bootstrapping, coding-systems are not defined yet, so
1663 ;; we need to catch the error from check-coding-system.
1676 ;; Don't alter the terminal and keyboard coding systems here.
1677 ;; The terminal still supports the same coding system
1678 ;; that it supported a minute ago.
1679 ;; (set-terminal-coding-system-internal nil)
1680 ;; (set-keyboard-coding-system-internal nil)
1688 "Set up the display table and terminal coding system for LANGUAGE-NAME."
1692 ;; The following 2 lines undo the 8-bit display that we set up
1693 ;; in standard-display-european-internal, which see. This is in
1694 ;; case the user has used standard-display-european earlier in
1695 ;; this session. (The MS-DOS port doesn't use that setup, so it
1696 ;; doesn't need to undo it.)
1704 "Set up multi-lingual environment for using LANGUAGE-NAME.
1705 This sets the coding system priority and the default input method
1706 and sometimes other things. LANGUAGE-NAME should be a string
1708 specifies the character set for the major languages of Western Europe."
1710 nil
1726 default-buffer-file-coding-system)))
1729 ;; The features might set up coding systems.
1753 ;; DOS terminals' default is to use a special non-ASCII translation
1754 ;; table as appropriate for the installed codepage.
1759 ;; Unibyte setups if necessary.
1761 ;; Syntax and case table.
1766 ;; No information for syntax and case. Reset to the defaults.
1796 ;; Actually set up direct output of non-ASCII characters.
1798 ;; Unibyte Emacs on MS-DOS wants to display all 8-bit characters with
1799 ;; the native font, and codes 160 and 146 stand for something very
1800 ;; different there.
1803 ;; Make non-line-break space display as a plain space.
1804 ;; Most X fonts do the wrong thing for code 160.
1806 ;; With luck, non-Latin-1 fonts are more recent and so don't
1807 ;; have this bug.
1809 ;; Most Windows programs send out apostrophes as \222. Most X fonts
1810 ;; don't contain a character at that position. Map it to the ASCII
1811 ;; apostrophe. [This is actually RIGHT SINGLE QUOTATION MARK,
1812 ;; U+2019, normally from the windows-1252 character set. XFree 4
1813 ;; fonts probably have the appropriate glyph at this position,
1814 ;; so they could use standard-display-8bit. It's better to use a
1815 ;; proper windows-1252 coding system. --fx]
1817 ;; XFree86 4 has changed most of the fonts from their designed
1818 ;; versions such that `' no longer appears as balanced quotes.
1819 ;; Assume it has iso10646 fonts installed, so we can display
1820 ;; balanced quotes.
1821 (when (and (eq window-system 'x)
1822 (string= "The XFree86 Project, Inc" (x-server-vendor))
1823 (> (aref (number-to-string (nth 2 (x-server-version))) 0)
1824 ?3))
1825 (aset standard-display-table ?' [?\e$,1ry\e(B])
1826 (aset standard-display-table ?` [?\e$,1rx\e(B])
1827 ;; The fonts don't have the relevant bug.
1828 (aset standard-display-table 160 nil)
1829 (aset standard-display-table (make-char 'latin-iso8859-1 160)
1830 nil)))))
1832 (defun set-language-environment-coding-systems (language-name
1833 &optional eol-type)
1834 "Do various coding system setups for language environment LANGUAGE-NAME.
1836 The optional arg EOL-TYPE specifies the eol-type of the default value
1837 of `buffer-file-coding-system' set by this function."
1838 (let* ((priority (get-language-info language-name 'coding-priority))
1839 (default-coding (car priority)))
1840 (if priority
1841 (let ((categories (mapcar 'coding-system-category priority)))
1842 (set-default-coding-systems
1843 (if (memq eol-type '(0 1 2 unix dos mac))
1844 (coding-system-change-eol-conversion default-coding eol-type)
1845 default-coding))
1846 (setq default-sendmail-coding-system default-coding)
1847 (set-coding-priority categories)
1848 (while priority
1849 (set (car categories) (car priority))
1850 (setq priority (cdr priority) categories (cdr categories)))
1851 (update-coding-systems-internal)))))
1853 (defsubst princ-list (&rest args)
1855 (while args (princ (car args)) (setq args (cdr args)))
1858 (put 'describe-specified-language-support 'apropos-inhibit t)
1860 ;; Print language-specific information such as input methods,
1861 ;; charsets, and coding systems. This function is intended to be
1862 ;; called from the menu:
1863 ;; [menu-bar mule describe-language-environment LANGUAGE]
1864 ;; and should not run it by `M-x describe-current-input-method-function'.
1865 (defun describe-specified-language-support ()
1866 "Describe how Emacs supports the specified language environment."
1867 (interactive)
1868 (let (language-name)
1869 (if (not (and (symbolp last-command-event)
1870 (or (not (eq last-command-event 'Default))
1871 (setq last-command-event 'English))
1872 (setq language-name (symbol-name last-command-event))))
1873 (error "Bogus calling sequence"))
1874 (describe-language-environment language-name)))
1876 (defun describe-language-environment (language-name)
1877 "Describe how Emacs supports language environment LANGUAGE-NAME."
1878 (interactive
1879 (list (read-language-name
1880 'documentation
1881 "Describe language environment (default, current choice): ")))
1882 (if (null language-name)
1883 (setq language-name current-language-environment))
1884 (if (or (null language-name)
1885 (null (get-language-info language-name 'documentation)))
1886 (error "No documentation for the specified language"))
1887 (if (symbolp language-name)
1888 (setq language-name (symbol-name language-name)))
1889 (dolist (feature (get-language-info language-name 'features))
1890 (require feature))
1891 (let ((doc (get-language-info language-name 'documentation))
1892 pos)
1893 (help-setup-xref (list #'describe-language-environment language-name)
1894 (interactive-p))
1895 (with-output-to-temp-buffer (help-buffer)
1896 (save-excursion
1897 (set-buffer standard-output)
1899 (if (stringp doc)
1901 (condition-case nil
1902 (let ((str (eval (get-language-info language-name 'sample-text))))
1903 (if (stringp str)
1905 (error nil))
1906 (let ((input-method (get-language-info language-name 'input-method))
1907 (l (copy-sequence input-method-alist)))
1908 (insert "Input methods")
1909 (when input-method
1910 (insert " (default, " input-method ")")
1911 (setq input-method (assoc input-method input-method-alist))
1912 (setq l (cons input-method (delete input-method l))))
1914 (while l
1915 (when (string= language-name (nth 1 (car l)))
1916 (insert " " (car (car l)))
1917 (search-backward (car (car l)))
1918 (help-xref-button 0 'help-input-method (car (car l)))
1919 (goto-char (point-max))
1921 (if (stringp (nth 3 (car l)))
1922 (nth 3 (car l))
1923 (car (nth 3 (car l))))
1925 (setq l (cdr l)))
1928 (let ((l (get-language-info language-name 'charset)))
1929 (if (null l)
1931 (while l
1932 (insert " " (symbol-name (car l)))
1933 (search-backward (symbol-name (car l)))
1934 (help-xref-button 0 'help-character-set (car l))
1935 (goto-char (point-max))
1937 (setq l (cdr l)))))
1940 (let ((l (get-language-info language-name 'coding-system)))
1941 (if (null l)
1943 (while l
1944 (insert " " (symbol-name (car l)))
1945 (search-backward (symbol-name (car l)))
1946 (help-xref-button 0 'help-coding-system (car l))
1947 (goto-char (point-max))
1948 (insert " (`"
1949 (coding-system-mnemonic (car l))
1951 (coding-system-doc-string (car l))
1953 (let ((aliases (coding-system-get (car l)
1954 'alias-coding-systems)))
1955 (when aliases
1957 (while aliases
1958 (insert " " (symbol-name (car aliases)))
1959 (setq aliases (cdr aliases)))
1961 (setq l (cdr l)))))))))
1962 \f
1963 ;;; Locales.
1965 (defvar locale-translation-file-name nil
1966 "File name for the system's file of locale-name aliases, or nil if none.")
1968 ;; The following definitions might as well be marked as constants and
1969 ;; purecopied, since they're normally used on startup, and probably
1970 ;; should reflect the facilities of the base Emacs.
1971 (defconst locale-language-names
1972 (purecopy
1973 '(
1974 ;; Locale names of the form LANGUAGE[_TERRITORY][.CODESET][@MODIFIER]
1975 ;; as specified in the Single Unix Spec, Version 2.
1976 ;; LANGUAGE is a language code taken from ISO 639:1988 (E/F)
1977 ;; with additions from ISO 639/RA Newsletter No.1/1989;
1978 ;; see Internet RFC 2165 (1997-06) and
1979 ;; http://www.evertype.com/standards/iso639/iso639-en.html
1980 ;; TERRITORY is a country code taken from ISO 3166
1981 ;; http://www.din.de/gremien/nas/nabd/iso3166ma/codlstp1/en_listp1.html.
1982 ;; CODESET and MODIFIER are implementation-dependent.
1984 ; aa Afar
1985 ; ab Abkhazian
1986 ("af" . "Latin-1") ; Afrikaans
1987 ("am" . "Ethiopic") ; Amharic
1988 ; ar Arabic glibc uses 8859-6
1989 ; as Assamese
1990 ; ay Aymara
1991 ; az Azerbaijani
1992 ; ba Bashkir
1993 ("be" . "Belarusian") ; Belarusian [Byelorussian until early 1990s]
1994 ("bg" . "Bulgarian") ; Bulgarian
1995 ; bh Bihari
1996 ; bi Bislama
1997 ; bn Bengali, Bangla
1998 ("bo" . "Tibetan")
1999 ("br" . "Latin-1") ; Breton
2000 ("bs" . "Latin-2") ; Bosnian
2001 ("ca" . "Latin-1") ; Catalan
2002 ; co Corsican
2003 ("cs" . "Czech")
2004 ("cy" . "Welsh") ; Welsh [glibc uses Latin-8. Did this change?]
2005 ("da" . "Latin-1") ; Danish
2006 ("de" . "German")
2007 ; dz Bhutani
2008 ("el" . "Greek")
2009 ;; Users who specify "en" explicitly typically want Latin-1, not ASCII.
2010 ;; That's actually what the GNU locales define, modulo things like
2011 ;; en_IN -- fx.
2012 ("en" . "Latin-1") ; English
2013 ("eo" . "Latin-3") ; Esperanto
2014 ("es" . "Spanish")
2015 ("et" . "Latin-4") ; Estonian
2016 ("eu" . "Latin-1") ; Basque
2017 ; fa Persian glibc uses utf-8
2018 ("fi" . "Latin-1") ; Finnish
2019 ; fj Fiji
2020 ("fo" . "Latin-1") ; Faroese
2021 ("fr" . "French") ; French
2022 ("fy" . "Latin-1") ; Frisian
2023 ("ga" . "Latin-1") ; Irish Gaelic (new orthography)
2024 ("gd" . "Latin-1") ; Scots Gaelic
2025 ("gl" . "Latin-1") ; Galician
2026 ; gn Guarani
2027 ; gu Gujarati
2028 ("gv" . "Latin-8") ; Manx Gaelic glibc uses 8859-1
2029 ; ha Hausa
2030 ("he" . "Hebrew")
2031 ("hi" . "Devanagari") ; Hindi glibc uses utf-8
2032 ("hr" . "Croatian") ; Croatian
2033 ("hu" . "Latin-2") ; Hungarian
2034 ; hy Armenian
2035 ; ia Interlingua
2036 ("id" . "Latin-1") ; Indonesian
2037 ; ie Interlingue
2038 ; ik Inupiak
2039 ("is" . "Latin-1") ; Icelandic
2040 ("it" . "Italian") ; Italian
2041 ; iu Inuktitut
2042 ("ja" . "Japanese")
2043 ; jw Javanese
2044 ("ka" . "Georgian") ; Georgian
2045 ; kk Kazakh
2046 ("kl" . "Latin-1") ; Greenlandic
2047 ; km Cambodian
2048 ; kn Kannada
2049 ("ko" . "Korean")
2050 ; ks Kashmiri
2051 ; ku Kurdish
2052 ("kw" . "Latin-1") ; Cornish
2053 ; ky Kirghiz
2054 ("la" . "Latin-1") ; Latin
2055 ("lb" . "Latin-1") ; Luxemburgish
2056 ; ln Lingala
2057 ("lo" . "Lao") ; Laothian
2058 ("lt" . "Lithuanian")
2059 ("lv" . "Latvian") ; Latvian, Lettish
2060 ; mg Malagasy
2061 ("mi" . "Latin-7") ; Maori
2062 ("mk" . "Cyrillic-ISO") ; Macedonian
2063 ; ml Malayalam
2064 ; mn Mongolian
2065 ; mo Moldavian
2066 ("mr" . "Devanagari") ; Marathi glibc uses utf-8
2067 ("ms" . "Latin-1") ; Malay
2068 ("mt" . "Latin-3") ; Maltese
2069 ; my Burmese
2070 ; na Nauru
2071 ("ne" . "Devanagari") ; Nepali
2072 ("nl" . "Dutch")
2073 ("no" . "Latin-1") ; Norwegian
2074 ("oc" . "Latin-1") ; Occitan
2075 ; om (Afan) Oromo
2076 ; or Oriya
2077 ; pa Punjabi
2078 ("pl" . "Latin-2") ; Polish
2079 ; ps Pashto, Pushto
2080 ("pt" . "Latin-1") ; Portuguese
2081 ; qu Quechua
2082 ("rm" . "Latin-1") ; Rhaeto-Romanic
2083 ; rn Kirundi
2084 ("ro" . "Romanian")
2085 ("ru.*[_.]koi8" . "Russian")
2086 ("ru" . "Cyrillic-ISO") ; Russian
2087 ; rw Kinyarwanda
2088 ("sa" . "Devanagari") ; Sanskrit
2089 ; sd Sindhi
2090 ; se Northern Sami
2091 ; sg Sangho
2092 ("sh" . "Latin-2") ; Serbo-Croatian
2093 ; si Sinhalese
2094 ("sk" . "Slovak")
2095 ("sl" . "Slovenian")
2096 ; sm Samoan
2097 ; sn Shona
2098 ; so Somali
2099 ("sq" . "Latin-1") ; Albanian
2100 ("sr" . "Latin-2") ; Serbian (Latin alphabet)
2101 ("sr_YU@cyrillic" . "Cyrillic-ISO") ; per glibc
2102 ; ss Siswati
2103 ; st Sesotho
2104 ; su Sundanese
2105 ("sv" . "Swedish") ; Swedish
2106 ("sw" . "Latin-1") ; Swahili
2107 ; ta Tamil glibc uses utf-8
2108 ; te Telugu glibc uses utf-8
2109 ("tg" . "Tajik")
2110 ("th" . "Thai")
2111 ; ti Tigrinya
2112 ; tk Turkmen
2113 ("tl" . "Latin-1") ; Tagalog
2114 ; tn Setswana
2115 ; to Tonga
2116 ("tr" . "Turkish")
2117 ; ts Tsonga
2118 ; tt Tatar
2119 ; tw Twi
2120 ; ug Uighur
2121 ("uk" . "Ukrainian") ; Ukrainian
2122 ; ur Urdu glibc uses utf-8
2123 ("uz" . "Latin-1") ; Uzbek
2124 ("vi" . "Vietnamese") ; glibc uses utf-8
2125 ; vo Volapuk
2126 ("wa" . "Latin-1") ; Walloon
2127 ; wo Wolof
2128 ; xh Xhosa
2129 ("yi" . "Windows-1255") ; Yiddish
2130 ; yo Yoruba
2131 ; za Zhuang
2133 ; glibc:
2134 ; zh_CN.GB18030/GB18030 \
2135 ; zh_CN.GBK/GBK \
2136 ; zh_HK/BIG5-HKSCS \
2138 ("zh.*[._]big5" . "Chinese-BIG5")
2139 ("zh.*[._]gbk" . nil) ; Solaris 2.7; has gbk-0 as well as GB 2312.1980-0
2140 ("zh_tw" . "Chinese-CNS") ; glibc uses big5
2141 ("zh_tw[._]euc-tw" . "Chinese-EUC-TW")
2142 ("zh" . "Chinese-GB")
2143 ; zu Zulu
2145 ;; ISO standard locales
2146 ("c$" . "ASCII")
2147 ("posix$" . "ASCII")
2149 ;; The "IPA" Emacs language environment does not correspond
2150 ;; to any ISO 639 code, so let it stand for itself.
2151 ("ipa$" . "IPA")
2153 ;; Nonstandard or obsolete language codes
2154 ("cz" . "Czech") ; e.g. Solaris 2.6
2155 ("ee" . "Latin-4") ; Estonian, e.g. X11R6.4
2156 ("iw" . "Hebrew") ; e.g. X11R6.4
2157 ("sp" . "Cyrillic-ISO") ; Serbian (Cyrillic alphabet), e.g. X11R6.4
2158 ("su" . "Latin-1") ; Finnish, e.g. Solaris 2.6
2159 ("jp" . "Japanese") ; e.g. MS Windows
2160 ("chs" . "Chinese-GB") ; MS Windows Chinese Simplified
2161 ("cht" . "Chinese-BIG5") ; MS Windows Chinese Traditional
2162 ))
2163 "List of pairs of locale regexps and language names.
2164 The first element whose locale regexp matches the start of a downcased locale
2165 specifies the language name corresponding to that locale.
2166 If the language name is nil, there is no corresponding language environment.")
2168 (defconst locale-charset-language-names
2169 (purecopy
2178 ;; utf-8@euro exists, so put this last. (@euro really specifies
2179 ;; the currency, rather than the charset.)
2181 "List of pairs of locale regexps and charset language names.
2182 The first element whose locale regexp matches the start of a downcased locale
2183 specifies the language name whose charset corresponds to that locale.
2184 This language name is used if its charsets disagree with the charsets of
2185 the language name that would otherwise be used for this locale.")
2187 (defconst locale-preferred-coding-systems
2188 (purecopy
2189 '(("ja.*[._]euc" . japanese-iso-8bit)
2190 ("ja.*[._]jis7" . iso-2022-jp)
2191 ("ja.*[._]pck" . japanese-shift-jis)
2192 ("ja.*[._]sjis" . japanese-shift-jis)
2193 ("jpn" . japanese-shift-jis) ; MS-Windows uses this.
2194 (".*[._]utf" . utf-8)))
2195 "List of pairs of locale regexps and preferred coding systems.
2196 The first element whose locale regexp matches the start of a downcased locale
2197 specifies the coding system to prefer when using that locale.")
2199 (defun locale-name-match (key alist)
2200 "Search for KEY in ALIST, which should be a list of regexp-value pairs.
2201 Return the value corresponding to the first regexp that matches the
2202 start of KEY, or nil if there is no match."
2203 (let (element)
2204 (while (and alist (not element))
2206 (setq element (car alist)))
2207 (setq alist (cdr alist)))
2208 (cdr element)))
2210 (defun locale-charset-match-p (charset1 charset2)
2211 "Whether charset names (strings) CHARSET1 and CHARSET2 are equivalent.
2212 Matching is done ignoring case and any hyphens and underscores in the
2213 names. E.g. `ISO_8859-1' and `iso88591' both match `iso-8859-1'."
2214 (setq charset1 (replace-regexp-in-string "[-_]" "" charset1))
2215 (setq charset2 (replace-regexp-in-string "[-_]" "" charset2))
2216 (eq t (compare-strings charset1 nil nil charset2 nil nil t)))
2218 (defvar locale-charset-alist nil
2219 "Coding system alist keyed on locale-style charset name.
2220 Used by `locale-charset-to-coding-system'.")
2222 (defun locale-charset-to-coding-system (charset)
2223 "Find coding system corresponding to CHARSET.
2224 CHARSET is any sort of non-Emacs charset name, such as might be used
2225 in a locale codeset, or elsewhere. It is matched to a coding system
2226 first by case-insensitive lookup in `locale-charset-alist'. Then
2227 matches are looked for in the coding system list, treating case and
2228 the characters `-' and `_' as insignificant. The coding system base
2230 `iso-latin-2' is returned."
2231 (or (car (assoc-ignore-case charset locale-charset-alist))
2232 (let ((cs coding-system-alist)
2233 c)
2234 (while (and (not c) cs)
2235 (if (locale-charset-match-p charset (caar cs))
2236 (setq c (intern (caar cs)))
2237 (pop cs)))
2238 (coding-system-base c))))
2240 ;; Fixme: This ought to deal with the territory part of the locale
2241 ;; too, for setting things such as calendar holidays, ps-print paper
2242 ;; size, spelling dictionary.
2244 (defun set-locale-environment (&optional locale-name)
2245 "Set up multi-lingual environment for using LOCALE-NAME.
2246 This sets the language environment, the coding system priority,
2247 the default input method and sometimes other things.
2249 LOCALE-NAME should be a string which is the name of a locale supported
2250 by the system. Often it is of the form xx_XX.CODE, where xx is a
2251 language, XX is a country, and CODE specifies a character set and
2253 a locale for Japanese in Japan using the `japanese-iso-8bit'
2254 coding-system. The name may also have a modifier suffix, e.g. `@euro'
2255 or `@cyrillic'.
2257 If LOCALE-NAME is nil, its value is taken from the environment
2258 variables LC_ALL, LC_CTYPE and LANG (the first one that is set).
2260 The locale names supported by your system can typically be found in a
2261 directory named `/usr/share/locale' or `/usr/lib/locale'. LOCALE-NAME
2262 will be translated according to the table specified by
2263 `locale-translation-file-name'.
2265 See also `locale-charset-language-names', `locale-language-names',
2266 `locale-preferred-coding-systems' and `locale-coding-system'."
2267 (interactive "sSet environment for locale: ")
2269 ;; Do this at runtime for the sake of binaries possibly transported
2270 ;; to a system without X.
2271 (setq locale-translation-file-name
2272 (let ((files
2273 '("/usr/lib/X11/locale/locale.alias" ; e.g. X11R6.4
2274 "/usr/X11R6/lib/X11/locale/locale.alias" ; XFree86, e.g. RedHat 4.2
2275 "/usr/openwin/lib/locale/locale.alias" ; e.g. Solaris 2.6
2276 ;;
2277 ;; The following name appears after the X-related names above,
2278 ;; since the X-related names are what X actually uses.
2279 "/usr/share/locale/locale.alias" ; GNU/Linux sans X
2280 )))
2281 (while (and files (not (file-exists-p (car files))))
2282 (setq files (cdr files)))
2283 (car files)))
2285 (let ((locale locale-name))
2287 (unless locale
2288 ;; Use the first of these three environment variables
2289 ;; that has a nonempty value.
2290 (let ((vars '("LC_ALL" "LC_CTYPE" "LANG")))
2291 (while (and vars
2292 (= 0 (length locale))) ; nil or empty string
2293 (setq locale (getenv (pop vars))))))
2295 (when locale
2297 ;; Translate "swedish" into "sv_SE.ISO8859-1", and so on,
2298 ;; using the translation file that many systems have.
2299 (when locale-translation-file-name
2300 (with-temp-buffer
2301 (insert-file-contents locale-translation-file-name)
2302 (when (re-search-forward
2304 (setq locale (buffer-substring (point) (line-end-position))))))
2306 ;; Leave the system locales alone if the caller did not specify
2307 ;; an explicit locale name, as their defaults are set from
2308 ;; LC_MESSAGES and LC_TIME, not LC_CTYPE, and the user might not
2309 ;; want to set them to the same value as LC_CTYPE.
2310 (when locale-name
2311 (setq system-messages-locale locale)
2312 (setq system-time-locale locale))
2314 (setq locale (downcase locale))
2316 (let ((language-name
2317 (locale-name-match locale locale-language-names))
2318 (charset-language-name
2319 (locale-name-match locale locale-charset-language-names))
2320 (coding-system
2321 (or (locale-name-match locale locale-preferred-coding-systems)
2322 (when locale
2324 (locale-charset-to-coding-system
2325 (match-string 1 locale)))))))
2327 ;; Give preference to charset-language-name over language-name.
2328 (if (and charset-language-name
2329 (not
2330 (equal (get-language-info language-name 'charset)
2331 (get-language-info charset-language-name 'charset))))
2332 (setq language-name charset-language-name))
2334 (when language-name
2336 ;; Set up for this character set. This is now the right way
2337 ;; to do it for both unibyte and multibyte modes.
2338 (set-language-environment language-name)
2340 ;; If default-enable-multibyte-characters is nil,
2341 ;; we are using single-byte characters,
2342 ;; so the display table and terminal coding system are irrelevant.
2343 (when default-enable-multibyte-characters
2344 (set-display-table-and-terminal-coding-system language-name))
2346 ;; Set the `keyboard-coding-system' if appropriate (tty
2347 ;; only). At least X and MS Windows can generate
2348 ;; multilingual input.
2349 (unless window-system
2350 (let ((kcs (or coding-system
2351 (car (get-language-info language-name
2352 'coding-system)))))
2353 (if kcs (set-keyboard-coding-system kcs))))
2355 (setq locale-coding-system
2356 (car (get-language-info language-name 'coding-priority))))
2358 (when coding-system
2359 (prefer-coding-system coding-system)
2360 (setq locale-coding-system coding-system))))
2362 ;; Default to A4 paper if we're not in a C, POSIX or US locale.
2363 ;; (See comments in Flocale_info.)
2364 (let ((locale locale)
2365 (paper (locale-info 'paper)))
2366 (if paper
2367 ;; This will always be null at the time of writing.
2368 (cond
2369 ((equal paper '(216 279))
2370 (setq ps-paper-type 'letter))
2371 ((equal paper '(210 297))
2372 (setq ps-paper-type 'a4)))
2373 (let ((vars '("LC_ALL" "LC_PAPER" "LANG")))
2374 (while (and vars (= 0 (length locale)))
2375 (setq locale (getenv (pop vars)))))
2376 (when locale
2377 ;; As of glibc 2.2.5, these are the only US Letter locales,
2378 ;; and the rest are A4.
2379 (setq ps-paper-type
2380 (or (locale-name-match locale '(("c$" . letter)
2381 ("posix$" . letter)
2382 (".._us" . letter)
2383 (".._pr" . letter)
2384 (".._ca" . letter)))
2385 'a4))))))
2386 nil)
2387 \f
2388 ;;; Charset property
2390 (defun get-charset-property (charset propname)
2391 "Return the value of CHARSET's PROPNAME property.
2392 This is the last value stored with
2393 (put-charset-property CHARSET PROPNAME VALUE)."
2394 (and (not (eq charset 'composition))
2395 (plist-get (charset-plist charset) propname)))
2397 (defun put-charset-property (charset propname value)
2398 "Store CHARSETS's PROPNAME property with value VALUE.
2399 It can be retrieved with `(get-charset-property CHARSET PROPNAME)'."
2400 (or (eq charset 'composition)
2401 (set-charset-plist charset
2402 (plist-put (charset-plist charset) propname value))))
2404 ;;; Character code property
2405 (put 'char-code-property-table 'char-table-extra-slots 0)
2407 (defvar char-code-property-table
2408 (make-char-table 'char-code-property-table)
2409 "Char-table containing a property list of each character code.
2411 See also the documentation of `get-char-code-property' and
2412 `put-char-code-property'.")
2414 (defun get-char-code-property (char propname)
2415 "Return the value of CHAR's PROPNAME property in `char-code-property-table'."
2416 (let ((plist (aref char-code-property-table char)))
2417 (if (listp plist)
2418 (car (cdr (memq propname plist))))))
2420 (defun put-char-code-property (char propname value)
2421 "Store CHAR's PROPNAME property with VALUE in `char-code-property-table'.
2422 It can be retrieved with `(get-char-code-property CHAR PROPNAME)'."
2423 (let ((plist (aref char-code-property-table char)))
2424 (if plist
2425 (let ((slot (memq propname plist)))
2426 (if slot
2427 (setcar (cdr slot) value)
2428 (nconc plist (list propname value))))
2429 (aset char-code-property-table char (list propname value)))))
2431 \f
2432 ;; Pretty description of encoded string
2434 ;; Alist of ISO 2022 control code vs the corresponding mnemonic string.
2435 (defvar iso-2022-control-alist
2443 (defun encoded-string-description (str coding-system)
2444 "Return a pretty description of STR that is encoded by CODING-SYSTEM."
2445 (setq str (string-as-unibyte str))
2446 (mapconcat
2447 (if (and coding-system (eq (coding-system-type coding-system) 2))
2448 ;; Try to get a pretty description for ISO 2022 escape sequences.
2449 (function (lambda (x) (or (cdr (assq x iso-2022-control-alist))
2450 (format "0x%02X" x))))
2451 (function (lambda (x) (format "0x%02X" x))))
2452 str " "))
2454 (defun encode-coding-char (char coding-system)
2455 "Encode CHAR by CODING-SYSTEM and return the resulting string.
2456 If CODING-SYSTEM can't safely encode CHAR, return nil."
2457 (let ((str1 (string-as-multibyte (char-to-string char)))
2458 (str2 (string-as-multibyte (make-string 2 char)))
2459 (safe-chars (and coding-system
2460 (coding-system-get coding-system 'safe-chars)))
2461 (charset (char-charset char))
2462 enc1 enc2 i1 i2)
2463 (when (or (eq safe-chars t)
2464 (eq charset 'ascii)
2465 (and safe-chars (aref safe-chars char)))
2466 ;; We must find the encoded string of CHAR. But, just encoding
2467 ;; CHAR will put extra control sequences (usually to designate
2468 ;; ASCII charset) at the tail if type of CODING is ISO 2022.
2469 ;; To exclude such tailing bytes, we at first encode one-char
2470 ;; string and two-char string, then check how many bytes at the
2471 ;; tail of both encoded strings are the same.
2473 (setq enc1 (encode-coding-string str1 coding-system)
2474 i1 (length enc1)
2475 enc2 (encode-coding-string str2 coding-system)
2476 i2 (length enc2))
2477 (while (and (> i1 0) (= (aref enc1 (1- i1)) (aref enc2 (1- i2))))
2478 (setq i1 (1- i1) i2 (1- i2)))
2480 ;; Now (substring enc1 i1) and (substring enc2 i2) are the same,
2481 ;; and they are the extra control sequences at the tail to
2482 ;; exclude.
2483 (substring enc2 0 i2))))
2486 ;;; arch-tag: b382c432-4b36-460e-bf4c-05efd0bb18dc
2487 ;;; mule-cmds.el ends here