| blob | e9b769d9c126d7ab2b32a6df458a559182d2d4fd |
1 ;;; mule-cmds.el --- commands for mulitilingual environment
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.
73 t)
76 t)
79 t)
84 t)
93 t)
96 t)
99 describe-language-environment-map
101 t)
105 t)
108 t)
115 t)
120 t)
123 t)
127 t)
130 :enable buffer-file-name
132 t)
136 t)
139 t)
144 t)
149 t)
152 t)
157 t)
162 t)
168 t)
172 [Default] '(menu-item "Default" setup-specified-language-environment))
174 (define-key describe-language-environment-map
177 ;; This should be a single character key binding because users use it
178 ;; very frequently while editing multilingual text. Now we can use
179 ;; only two such keys: "\C-\\" and "\C-^", but the latter is not
180 ;; convenient because it requires shifting on most keyboards. An
181 ;; alternative is "\C-\]" which is now bound to `abort-recursive-edit'
182 ;; but it won't be used that frequently.
185 ;; This is no good because people often type Shift-SPC
186 ;; meaning to type SPC. -- rms.
187 ;; ;; Here's an alternative key binding for X users (Shift-SPACE).
188 ;; (define-key global-map [?\S- ] 'toggle-input-method)
190 ;;; Mule related hyperlinks.
198 ;; Note starting with word-syntax character:
202 "Return a coding system which differs from CODING-SYSTEM in eol conversion.
203 The returned coding system converts end-of-line by EOL-TYPE
204 but text as the same way as CODING-SYSTEM.
205 EOL-TYPE should be `unix', `dos', `mac', or nil.
206 If EOL-TYPE is nil, the returned coding system detects
207 how end-of-line is formatted automatically while decoding.
209 EOL-TYPE can be specified by an integer 0, 1, or 2.
210 They means `unix', `dos', and `mac' respectively."
219 coding-system
223 base
225 coding-system
231 "Return a coding system which differs from CODING-SYSTEM in text conversion.
232 The returned coding system converts text by CODING
233 but end-of-line as the same way as CODING-SYSTEM.
234 If CODING is nil, the returned coding system detects
235 how text is formatted automatically while decoding."
242 "Change whether this buffer uses multibyte characters.
243 With arg, use multibyte characters if the arg is positive.
245 Note that this command does not convert the byte contents of
246 the buffer; it only changes the way those bytes are interpreted.
247 In general, therefore, this command *changes* the sequence of
248 characters that the current buffer contains.
250 We suggest you avoid using use this command unless you know what you
251 are doing. If you use it by mistake, and the buffer is now displayed
252 wrong, use this command again to toggle back to the right mode."
261 "Display the HELLO file which list up many languages and characters."
263 ;; We have to decode the file in any environment.
269 "Execute an I/O command using the specified coding system."
273 t))
274 buffer-file-coding-system)))
279 default))))
283 prefix)
288 ;; Process keys bound in `universal-argument-map'.
294 ;; Have to bind `last-command-char' here so that
295 ;; `digit-argument', for isntance, can compute the
296 ;; prefix arg.
300 ;; This is the final call to `univeral-argument-other-key', which
301 ;; set's the final `prefix-arg.
305 ;; Read the command to execute with the given prefix arg.
318 "Set default value of various coding systems to CODING-SYSTEM.
319 This sets the following coding systems:
320 o coding system of a newly created buffer
321 o default coding system for subprocess I/O
322 This also sets the following values:
323 o default value used as `file-name-coding-system' for converting file names.
324 o default value for the command `set-terminal-coding-system' (not on MSDOS)
325 o default value for the command `set-keyboard-coding-system'."
330 ;; If coding-system is nil, honor that on MS-DOS as well, so
331 ;; that they could reset the terminal coding system.
335 ;; Preserve eol-type from existing default-process-coding-systems.
336 ;; On non-unix-like systems in particular, these may have been set
337 ;; carefully by the user, or by the startup code, to deal with the
338 ;; users shell appropriately, so should not be altered by changing
339 ;; language environment.
353 "Add CODING-SYSTEM at the front of the priority list for automatic detection.
354 This also sets the following coding systems:
355 o coding system of a newly created buffer
356 o default coding system for subprocess I/O
357 This also sets the following values:
358 o default value used as `file-name-coding-system' for converting file names.
359 o default value for the command `set-terminal-coding-system' (not on MSDOS)
360 o default value for the command `set-keyboard-coding-system'
362 If CODING-SYSTEM specifies a certain type of EOL conversion, the coding
363 systems set by this function will use that type of EOL conversion.
365 This command does not change the default value of terminal coding system
366 for MS-DOS terminal, because DOS terminals only support a single coding
367 system, and Emacs automatically sets the default to that coding system at
368 startup.
370 A coding system that requires automatic detection of text
371 encoding (e.g. undecided, unix) can't be preferred.
373 See also `coding-category-list' and `coding-system-category'."
381 ;; CODING-SYSTEM is no-conversion or undecided.
386 ;; We must change the order.
390 base coding-system))
391 ;; If they asked for specific EOL conversion, honor that.
399 "If non-nil, a predicate function to sort coding systems.
401 It is called with two coding systems, and should return t if the first
407 "Sort coding system list CODINGS by a priority of each coding system.
409 If a coding system is most preferred, it has the highest priority.
410 Otherwise, a coding system corresponds to some MIME charset has higher
411 priorities. Among them, a coding system included in `coding-system'
412 key of the current language environment has higher priorities. See
413 also the documentation of `language-info-alist'.
415 If the variable `sort-coding-systems-predicate' (which see) is
416 non-nil, it is used to sort CODINGS in the different way than above."
420 coding-category-list))
427 ;; We calculate the priority number 0..255 by
428 ;; using the 8 bits PMMLCEII as this:
429 ;; P: 1 iff most preferred.
430 ;; MM: greater than 0 iff mime-charset.
431 ;; L: 1 iff one of the current lang. env.'s codings.
432 ;; C: 1 iff one of codings listed in the category list.
433 ;; E: 1 iff not XXX-with-esc
434 ;; II: if iso-2022 based, 0..3, else 1.
439 ;; Prefer coding systems corresponding to a
440 ;; MIME charset.
442 ;; Lower utf-16 priority so that we
443 ;; normally prefer utf-8 to it, and put
444 ;; x-ctext below that.
459 ;; For ISO based coding systems, prefer
460 ;; one that doesn't use escape sequences.
467 0
475 "Return a list of proper coding systems to encode a text between FROM and TO.
476 If FROM is a string, find coding systems in that instead of the buffer.
477 All coding systems in the list can safely encode any multibyte characters
478 in the text.
480 If the text contains no multibyte characters, return a list of a single
481 element `undecided'."
484 ;; The text contains only ASCII characters. Any coding
485 ;; systems are safe.
487 ;; We need copy-sequence because sorting will alter the argument.
491 "Return a list of proper coding systems to encode STRING.
492 All coding systems in the list can safely encode any multibyte characters
493 in STRING.
495 If STRING contains no multibyte characters, return a list of a single
496 element `undecided'."
500 "Return a list of proper coding systems to encode characters of CHARSETS.
501 CHARSETS is a list of character sets.
502 It actually checks at most the first 96 characters of each charset.
503 So, if a charset of dimension two is included in CHARSETS, the value may
504 contain a coding system that can't encode all characters of the charset."
514 charset l str)
538 codings))))
541 "Find multibyte characters in the region specified by FROM and TO.
542 If FROM is a string, find multibyte characters in the string.
543 The return value is an alist of the following format:
544 ((CHARSET COUNT CHAR ...) ...)
545 where
546 CHARSET is a character set,
547 COUNT is a number of characters,
548 CHARs are found characters of the character set.
549 Optional 3rd arg MAXCOUNT limits how many CHARs are put in the above list.
550 Optional 4th arg EXCLUDE is a list of character sets to be ignored.
552 For invalid characters, CHARs are actually strings."
554 charset char)
563 eight-bit-control eight-bit-graphic))
595 "Search forward from point for a character that is not encodable.
596 It asks which coding system to check.
597 If such a character is found, set point after that character.
598 Otherwise, don't move point.
600 When called from a program, the value is a position of the found character,
601 or nil if all characters are encodable."
606 default))))
611 pos))
615 "Most recent coding system explicitly specified by the user when asked.
616 This variable is set whenever Emacs asks the user which coding system
617 to use in order to write a file. If you set it to nil explicitly,
618 then call `write-region', then afterward this variable will be non-nil
622 "If non-nil, a function to control the behaviour of coding system selection.
623 The meaning is the same as the argument ACCEPT-DEFAULT-P of the
624 function `select-safe-coding-system' (which see). This variable
628 accept-default-p file)
629 "Ask a user to select a safe coding system from candidates.
630 The candidates of coding systems which can safely encode a text
631 between FROM and TO are shown in a popup window. Among them, the most
632 proper one is suggested as the default.
634 The list of `buffer-file-coding-system' of the current buffer,
635 the `default-buffer-file-coding-system', and the
636 most preferred coding system (if it corresponds to a MIME charset) is
637 treated as the default coding system list. Among them, the first one
638 that safely encodes the text is normally selected silently and
639 returned without any user interaction. See also the command
640 `prefer-coding-system'.
642 However, the user is queried if the chosen coding system is
643 inconsistent with what would be selected by `set-auto-coding' from
644 coding cookies &c. if the contents of the region were read from a
645 file. (That could lead to data corruption in a file subsequently
646 re-visited and edited.)
648 Optional 3rd arg DEFAULT-CODING-SYSTEM specifies a coding system or a
649 list of coding systems to be prepended to the default coding system
650 list. However, if DEFAULT-CODING-SYSTEM is a list and the first
651 element is t, the cdr part is used as the defualt coding system list,
652 i.e. `buffer-file-coding-system', `default-buffer-file-coding-system',
653 and the most preferred coding system are not used.
655 Optional 4th arg ACCEPT-DEFAULT-P, if non-nil, is a function to
656 determine the acceptability of the silently selected coding system.
657 It is called with that coding system, and should return nil if it
658 should not be silently selected and thus user interaction is required.
660 Optional 5th arg FILE is the file name to use for this purpose.
661 That is different from `buffer-file-name' when handling `write-region'
662 \(for example).
664 The variable `select-safe-coding-system-accept-default-p', if
665 non-nil, overrides ACCEPT-DEFAULT-P.
667 Kludgy feature: if FROM is a string, the string is the target text,
668 and TO is ignored."
678 ;; Change elements of the list to (coding . base-coding).
681 default-coding-system))
683 ;; From now on, the list of defaults is reversed.
687 ;; If buffer-file-coding-system is not nil nor undecided, append it
688 ;; to the defaults.
694 default-coding-system))))
696 ;; If default-buffer-file-coding-system is not nil nor undecided,
697 ;; append it to the defaults.
703 default-coding-system))))
705 ;; If the most preferred coding system has the property mime-charset,
706 ;; append it to the defaults.
708 preferred base)
716 default-coding-system)))))
724 safe rejected unsafe)
726 ;; Any coding system is ok.
728 ;; Classify the defaults into safe, rejected, and unsafe.
739 ;; If all the defaults failed, ask a user.
741 ;; At first, if some defaults are unsafe, record at most 11
742 ;; problematic characters and their positions for them by turning
743 ;; (CODING ...)
744 ;; into
745 ;; ((CODING (POS . CHAR) (POS . CHAR) ...) ...)
756 unsafe))
764 unsafe))))
766 ;; Change each safe coding system to the corresponding
767 ;; mime-charset name if it is also a coding system. Such a name
768 ;; is more friendly to users.
770 mime-charset)
777 ;; Don't offer variations with locking shift, which you
778 ;; basically never want.
787 ;; Remove raw-text, emacs-mule and no-conversion unless nothing
788 ;; else is available.
797 ;; If some defaults are unsafe, make sure the offending
798 ;; buffer is displayed.
802 unsafe))))
803 ;; Then ask users to select one from CODINGS while showing
804 ;; the reason why none of the defaults are not used.
814 "These default coding systems were tried to encode"
824 default-coding-system)
829 but it is not recommended for encoding text in this context,
863 'help-echo
864 "mouse-2, RET: jump to this character"
865 'help-function func1
868 "..."
870 'help-echo
871 "mouse-2, RET: next unencodable character"
872 'help-function func2
877 unsafe)
885 "\
890 codings)
896 ;; Read a coding system.
901 default-coding-system)
902 default-coding-system))
916 ;; Check we're not inconsistent with what `coding:' spec &c would
917 ;; give when file is re-read.
918 ;; But don't do this if we explicitly ignored the cookie
919 ;; by using `find-file-literally'.
929 ;; Don't barf if writing a compressed file, say.
930 ;; This check perhaps isn't ideal, but is probably
931 ;; the best thing to do.
937 %s specified by file contents. Really save (else edit coding cookies \
940 coding-system))
945 "Return a coding system to encode the outgoing message of the current buffer.
946 It at first tries the first coding system found in these variables
947 in this order:
948 (1) local value of `buffer-file-coding-system'
949 (2) value of `sendmail-coding-system'
950 (3) value of `default-sendmail-coding-system'
951 (4) value of `default-buffer-file-coding-system'
952 If the found coding system can't encode the current buffer,
953 or none of them are bound to a coding system,
954 it asks the user to select a proper coding system."
956 buffer-file-coding-system)
957 sendmail-coding-system
958 default-sendmail-coding-system
959 default-buffer-file-coding-system)))
961 ;; We should never use no-conversion for outgoing mail.
967 coding)))
968 \f
969 ;;; Language support stuff.
972 "Alist of language environment definitions.
973 Each element looks like:
974 (LANGUAGE-NAME . ((KEY . INFO) ...))
975 where LANGUAGE-NAME is a string, the name of the language environment,
976 KEY is a symbol denoting the kind of information, and
977 INFO is the data associated with KEY.
978 Meaningful values for KEY include
980 documentation value is documentation of what this language environment
981 is meant for, and how to use it.
982 charset value is a list of the character sets used by this
983 language environment.
984 sample-text value is an expression which is evalled to generate
985 a line of text written using characters appropriate
986 for this language environment.
987 setup-function value is a function to call to switch to this
988 language environment.
989 exit-function value is a function to call to leave this
990 language environment.
991 coding-system value is a list of coding systems that are good
992 for saving text written in this language environment.
993 This list serves as suggestions to the user;
994 in effect, as a kind of documentation.
995 coding-priority value is a list of coding systems for this language
996 environment, in order of decreasing priority.
997 This is used to set up the coding system priority
998 list when you switch to this language environment.
999 nonascii-translation
1000 value is a translation table to be set in the
1001 variable `nonascii-translation-table' in this
1002 language environment, or a character set from
1003 which `nonascii-insert-offset' is calculated.
1004 input-method value is a default input method for this language
1005 environment.
1006 features value is a list of features requested in this
1007 language environment.
1009 The following keys take effect only when multibyte characters are
1010 globally disabled, i.e. the value of `default-enable-multibyte-characters'
1011 is nil.
1013 unibyte-syntax value is a library name to load to set
1014 unibyte 8-bit character syntaxes for this
1015 language environment.
1017 unibyte-display value is a coding system to encode characters
1018 for the terminal. Characters in the range
1019 of 160 to 255 display not as octal escapes,
1020 but as non-ASCII characters in this language
1024 "Return information listed under KEY for language environment LANG-ENV.
1025 KEY is a symbol denoting the kind of information.
1026 For a list of useful values for KEY and their meanings,
1027 see `language-info-alist'."
1035 "Modify part of the definition of language environment LANG-ENV.
1036 Specifically, this stores the information INFO under KEY
1037 in the definition of this language environment.
1038 KEY is a symbol denoting the kind of information.
1039 INFO is the value for that information.
1041 For a list of useful values for KEY and their meanings,
1042 see `language-info-alist'."
1056 ;; Update the custom-type of `current-language-environment'.
1064 "Store ALIST as the definition of language environment LANG-ENV.
1065 ALIST is an alist of KEY and INFO values. See the documentation of
1066 `language-info-alist' for the meanings of KEY and INFO.
1068 Optional arg PARENTS is a list of parent menu names; it specifies
1069 where to put this language environment in the
1070 Describe Language Environment and Set Language Environment menus.
1072 in the European submenu in each of those two menus."
1079 map parent-symbol parent prompt)
1085 ;; This prompt string is for define-prefix-command, so
1086 ;; that the map it creates will be suitable for a menu.
1107 ;; Set up menu items for this language env.
1120 "Read a language environment name which has information for KEY.
1121 If KEY is nil, read any language environment.
1122 Prompt with PROMPT. DEFAULT is the default choice of language environment.
1123 This returns a language environment name as a string."
1126 language-info-alist
1129 t nil nil default)))
1133 name)))
1134 \f
1135 ;;; Multilingual input methods.
1137 "LEIM: Libraries of Emacs Input Methods."
1141 "Name of LEIM list file.
1142 This file contains a list of libraries of Emacs input methods (LEIM)
1143 in the format of Lisp expression for registering each input method.
1147 ";;; %s -- list of LEIM (Library of Emacs Input Method) -*-coding: iso-2022-7bit;-*-
1148 ;;
1149 ;; This file contains a list of LEIM (Library of Emacs Input Method)
1150 ;; methods in the same directory as this file. Loading this file
1151 ;; registers all the input methods in Emacs.
1152 ;;
1153 ;; Each entry has the form:
1154 ;; (register-input-method
1155 ;; INPUT-METHOD LANGUAGE-NAME ACTIVATE-FUNC
1156 ;; TITLE DESCRIPTION
1157 ;; ARG ...)
1158 ;; See the function `register-input-method' for the meanings of the arguments.
1159 ;;
1160 ;; If this directory is included in load-path, Emacs automatically
1161 ;; loads this file at startup time.
1163 "
1164 leim-list-file-name)
1168 "Regexp matching head of each entry in LEIM list file.
1173 "List of functions to call to update LEIM list file.
1177 "Update LEIM list file in directories DIRS."
1184 "The current input method for multilingual text.
1195 "*Default input method for multilingual text (a string).
1196 This is the input method activated automatically by the command
1210 "Function to call for inactivating the current input method.
1211 Every input method should set this to an appropriate value when activated.
1212 This function is called with no argument.
1214 This function should never change the value of `current-input-method'.
1220 "Function to call for describing the current input method.
1226 "Alist of input method names vs how to use them.
1227 Each element has the form:
1228 (INPUT-METHOD LANGUAGE-ENV ACTIVATE-FUNC TITLE DESCRIPTION ARGS...)
1232 "Register INPUT-METHOD as an input method for language environment LANG-ENV.
1233 INPUT-METHOD and LANG-ENV are symbols or strings.
1235 The remaining arguments are:
1236 ACTIVATE-FUNC, TITLE, DESCRIPTION, and ARGS...
1237 ACTIVATE-FUNC is a function to call to activate this method.
1238 TITLE is a string to show in the mode line when this method is active.
1239 DESCRIPTION is a string describing this method and what it is good for.
1240 The ARGS, if any, are passed as arguments to ACTIVATE-FUNC.
1241 All told, the arguments to ACTIVATE-FUNC are INPUT-METHOD and the ARGS.
1244 created at Emacs build time, registering all Quail input methods
1245 contained in the Emacs distribution.
1247 In case you want to register a new Quail input method by yourself, be
1248 careful to use the same input method title as given in the third
1249 parameter of `quail-define-package'. (If the values are different, the
1250 string specified in this function takes precedence.)
1252 The commands `describe-input-method' and `list-input-methods' need
1253 these duplicated values to show some information about input methods
1254 without loading the relevant Quail packages."
1267 "Read a name of input method from a minibuffer prompting with PROMPT.
1268 If DEFAULT is non-nil, use that as the default,
1269 and substitute it into PROMPT at the first `%s'.
1270 If INHIBIT-NULL is non-nil, null input signals an error.
1272 The return value is a string."
1276 ;; As it is quite normal to change input method in the
1277 ;; minibuffer, we must enable it even if
1278 ;; enable-recursive-minibuffers is currently nil.
1280 ;; This binding is necessary because input-method-history is
1281 ;; buffer local.
1283 nil t nil 'input-method-history
1284 default)))
1288 input-method
1293 "Switch to input method INPUT-METHOD for the current buffer.
1294 If some other input method is already active, turn it off first.
1295 If INPUT-METHOD is nil, deactivate any current input method."
1322 "Turn off the current input method."
1335 input-method-function nil
1336 current-input-method-title nil)
1340 "Select and activate input method INPUT-METHOD for the current buffer.
1341 This also sets the default input method to the one you specify.
1342 If INPUT-METHOD is nil, this function turns off the input method, and
1343 also causes you to be prompted for a name of an input method the next
1352 default t))))
1357 default-input-method)
1360 "Enable or disable multilingual text input method for the current buffer.
1361 Only one input method can be enabled at any time in a given buffer.
1363 The normal action is to enable an input method if none was
1364 enabled, and disable the current one otherwise. Which input method
1365 to enable can be determined in various ways--either the one most
1366 recently used, or the one specified by `default-input-method', or
1367 as a last resort by reading the name of an input method in the
1368 minibuffer.
1370 With a prefix argument, read an input method name with the minibuffer
1371 and enable that one. The default is the most recent input method specified
1372 \(not including the currently active input method, if any)."
1386 default t))
1387 default))
1395 "Describe input method INPUT-METHOD."
1425 "Describe the input method currently in use.
1426 This is a subroutine for `describe-input-method'."
1432 current-input-method)
1437 "Read a multilingual string from minibuffer, prompting with string PROMPT.
1438 The input method selected last time is activated in minibuffer.
1439 If optional second arg INITIAL-INPUT is non-nil, insert it in the minibuffer
1440 initially.
1441 Optional 3rd argument INPUT-METHOD specifies the input method
1442 to be activated instead of the one selected last time. It is a symbol
1443 or a string."
1446 current-input-method
1447 default-input-method
1458 ;; Variables to control behavior of input methods. All input methods
1459 ;; should react to these variables.
1462 "*A flag to control extra guidance given by input methods.
1463 The value should be nil, t, `complex-only', or `default'.
1465 The extra guidance is done by showing list of available keys in echo
1466 area. When you use the input method in the minibuffer, the guidance
1467 is shown at the bottom short window (split from the existing window).
1469 If the value is t, extra guidance is always given, if the value is
1470 nil, extra guidance is always suppressed.
1472 If the value is `complex-only', only complex input methods such as
1473 `chinese-py' and `japanese' give extra guidance.
1475 If the value is `default', complex input methods always give extra
1476 guidance, but simple input methods give it only when you are not in
1477 the minibuffer.
1479 See also the variable `input-method-highlight-flag'."
1485 "*If this flag is non-nil, input methods highlight partially-entered text.
1486 For instance, while you are in the middle of a Quail input method sequence,
1487 the text inserted so far is temporarily underlined.
1488 The underlining goes away when you finish or abort the input method sequence.
1489 See also the variable `input-method-verbose-flag'."
1494 "Normal hook run just after an input method is activated.
1496 The variable `current-input-method' keeps the input method name
1500 "Normal hook run just after an input method is inactivated.
1502 The variable `current-input-method' still keeps the input method name
1509 "This flag controls when an input method returns.
1510 Usually, the input method does not return while there's a possibility
1511 that it may find a different translation if a user types another key.
1512 But, it this flag is non-nil, the input method returns as soon as
1516 "This flag controls how an input method shows an intermediate key sequence.
1517 Usually, the input method inserts the intermediate key sequence,
1518 or candidate translations corresponding to the sequence,
1519 at point in the current buffer.
1523 "This flag controls the behaviour of an input method on invalid key input.
1524 Usually, when a user types a key which doesn't start any character
1525 handled by the input method, the key is handled by turning off the
1526 input method temporarily. After that key, the input method is re-enabled.
1529 \f
1531 "Normal hook run after some language environment is set.
1533 When you set some hook function here, that effect usually should not
1534 be inherited to another language environment. So, you had better set
1535 another function in `exit-language-environment-hook' (which see) to
1539 "Normal hook run after exiting from some language environment.
1540 When this hook is run, the variable `current-language-environment'
1541 is still bound to the language environment being exited.
1543 This hook is mainly used for canceling the effect of
1549 "Switch to a specified language environment."
1562 "The last language environment specified with `set-language-environment'.
1564 to using the function `set-language-environment'."
1571 current-language-environment)
1572 language-info-alist))
1574 ;; custom type will be updated with `set-language-info'.
1585 "Reset multilingual environment of Emacs to the default status.
1587 The default status is as follows:
1589 The default value of `buffer-file-coding-system' is nil.
1590 The default coding system for process I/O is nil.
1591 The default value for the command `set-terminal-coding-system' is nil.
1592 The default value for the command `set-keyboard-coding-system' is nil.
1594 The order of priorities of coding categories and the coding system
1595 bound to each category are as follows
1596 coding category coding system
1597 --------------------------------------------------
1598 coding-category-iso-8-1 iso-latin-1
1599 coding-category-iso-8-2 iso-latin-1
1600 coding-category-utf-8 mule-utf-8
1601 coding-category-utf-16-be mule-utf-16-be-with-signature
1602 coding-category-utf-16-le mule-utf-16-le-with-signature
1603 coding-category-iso-7-tight iso-2022-jp
1604 coding-category-iso-7 iso-2022-7bit
1605 coding-category-iso-7-else iso-2022-7bit-lock
1606 coding-category-iso-8-else iso-2022-8bit-ss2
1607 coding-category-emacs-mule emacs-mule
1608 coding-category-raw-text raw-text
1609 coding-category-sjis japanese-shift-jis
1610 coding-category-big5 chinese-big5
1611 coding-category-ccl nil
1612 coding-category-binary no-conversion"
1614 ;; This function formerly set default-enable-multibyte-characters to t,
1615 ;; but that is incorrect. It should not alter the unibyte/multibyte choice.
1618 coding-category-iso-7 'iso-2022-7bit
1619 coding-category-iso-8-1 'iso-latin-1
1620 coding-category-iso-8-2 'iso-latin-1
1621 coding-category-iso-7-else 'iso-2022-7bit-lock
1622 coding-category-iso-8-else 'iso-2022-8bit-ss2
1623 coding-category-emacs-mule 'emacs-mule
1624 coding-category-raw-text 'raw-text
1625 coding-category-sjis 'japanese-shift-jis
1626 coding-category-big5 'chinese-big5
1627 coding-category-utf-16-be 'mule-utf-16-be-with-signature
1628 coding-category-utf-16-le 'mule-utf-16-le-with-signature
1629 coding-category-utf-8 'mule-utf-8
1630 coding-category-ccl nil
1635 coding-category-iso-8-2
1636 coding-category-utf-8
1637 coding-category-utf-16-be
1638 coding-category-utf-16-le
1639 coding-category-iso-7-tight
1640 coding-category-iso-7
1641 coding-category-iso-7-else
1642 coding-category-iso-8-else
1643 coding-category-emacs-mule
1644 coding-category-raw-text
1645 coding-category-sjis
1646 coding-category-big5
1647 coding-category-ccl
1648 coding-category-binary))
1655 ;; Preserve eol-type from existing default-process-coding-systems.
1656 ;; On non-unix-like systems in particular, these may have been set
1657 ;; carefully by the user, or by the startup code, to deal with the
1658 ;; users shell appropriately, so should not be altered by changing
1659 ;; language environment.
1661 ;; When bootstrapping, coding-systems are not defined yet, so
1662 ;; we need to catch the error from check-coding-system.
1675 ;; Don't alter the terminal and keyboard coding systems here.
1676 ;; The terminal still supports the same coding system
1677 ;; that it supported a minute ago.
1678 ;; (set-terminal-coding-system-internal nil)
1679 ;; (set-keyboard-coding-system-internal nil)
1687 "Set up the display table and terminal coding system for LANGUAGE-NAME."
1697 "Set up multi-lingual environment for using LANGUAGE-NAME.
1698 This sets the coding system priority and the default input method
1699 and sometimes other things. LANGUAGE-NAME should be a string
1701 specifies the character set for the major languages of Western Europe."
1703 nil
1717 default-buffer-file-coding-system)))
1720 ;; The features might set up coding systems.
1744 ;; DOS terminals' default is to use a special non-ASCII translation
1745 ;; table as appropriate for the installed codepage.
1750 ;; Unibyte setups if necessary.
1752 ;; Syntax and case table.
1757 ;; No information for syntax and case. Reset to the defaults.
1787 ;; Actually set up direct output of non-ASCII characters.
1789 ;; Unibyte Emacs on MS-DOS wants to display all 8-bit characters with
1790 ;; the native font, and codes 160 and 146 stand for something very
1791 ;; different there.
1794 ;; Make non-line-break space display as a plain space.
1795 ;; Most X fonts do the wrong thing for code 160.
1797 ;; With luck, non-Latin-1 fonts are more recent and so don't
1798 ;; have this bug.
1800 ;; Most Windows programs send out apostrophes as \222. Most X fonts
1801 ;; don't contain a character at that position. Map it to the ASCII
1802 ;; apostrophe. [This is actually RIGHT SINGLE QUOTATION MARK,
1803 ;; U+2019, normally from the windows-1252 character set. XFree 4
1804 ;; fonts probably have the appropriate glyph at this position,
1805 ;; so they could use standard-display-8bit. It's better to use a
1806 ;; proper windows-1252 coding system. --fx]
1809 (defun set-language-environment-coding-systems (language-name
1810 &optional eol-type)
1811 "Do various coding system setups for language environment LANGUAGE-NAME.
1813 The optional arg EOL-TYPE specifies the eol-type of the default value
1814 of `buffer-file-coding-system' set by this function."
1815 (let* ((priority (get-language-info language-name 'coding-priority))
1816 (default-coding (car priority)))
1817 (if priority
1818 (let ((categories (mapcar 'coding-system-category priority)))
1819 (set-default-coding-systems
1820 (if (memq eol-type '(0 1 2 unix dos mac))
1821 (coding-system-change-eol-conversion default-coding eol-type)
1822 default-coding))
1823 (setq default-sendmail-coding-system default-coding)
1824 (set-coding-priority categories)
1825 (while priority
1826 (set (car categories) (car priority))
1827 (setq priority (cdr priority) categories (cdr categories)))
1828 (update-coding-systems-internal)))))
1830 (defsubst princ-list (&rest args)
1832 (while args (princ (car args)) (setq args (cdr args)))
1835 (put 'describe-specified-language-support 'apropos-inhibit t)
1837 ;; Print a language specific information such as input methods,
1838 ;; charsets, and coding systems. This function is intended to be
1839 ;; called from the menu:
1840 ;; [menu-bar mule describe-language-environment LANGUAGE]
1841 ;; and should not run it by `M-x describe-current-input-method-function'.
1842 (defun describe-specified-language-support ()
1843 "Describe how Emacs supports the specified language environment."
1844 (interactive)
1845 (let (language-name)
1846 (if (not (and (symbolp last-command-event)
1847 (or (not (eq last-command-event 'Default))
1848 (setq last-command-event 'English))
1849 (setq language-name (symbol-name last-command-event))))
1850 (error "Bogus calling sequence"))
1851 (describe-language-environment language-name)))
1853 (defun describe-language-environment (language-name)
1854 "Describe how Emacs supports language environment LANGUAGE-NAME."
1855 (interactive
1856 (list (read-language-name
1857 'documentation
1858 "Describe language environment (default, current choice): ")))
1859 (if (null language-name)
1860 (setq language-name current-language-environment))
1861 (if (or (null language-name)
1862 (null (get-language-info language-name 'documentation)))
1863 (error "No documentation for the specified language"))
1864 (if (symbolp language-name)
1865 (setq language-name (symbol-name language-name)))
1866 (dolist (feature (get-language-info language-name 'features))
1867 (require feature))
1868 (let ((doc (get-language-info language-name 'documentation))
1869 pos)
1870 (help-setup-xref (list #'describe-language-environment language-name)
1871 (interactive-p))
1872 (with-output-to-temp-buffer (help-buffer)
1873 (save-excursion
1874 (set-buffer standard-output)
1876 (if (stringp doc)
1878 (condition-case nil
1879 (let ((str (eval (get-language-info language-name 'sample-text))))
1880 (if (stringp str)
1882 (error nil))
1883 (let ((input-method (get-language-info language-name 'input-method))
1884 (l (copy-sequence input-method-alist)))
1885 (insert "Input methods")
1886 (when input-method
1887 (insert " (default, " input-method ")")
1888 (setq input-method (assoc input-method input-method-alist))
1889 (setq l (cons input-method (delete input-method l))))
1891 (while l
1892 (when (string= language-name (nth 1 (car l)))
1893 (insert " " (car (car l)))
1894 (search-backward (car (car l)))
1895 (help-xref-button 0 'help-input-method (car (car l)))
1896 (goto-char (point-max))
1898 (if (stringp (nth 3 (car l)))
1899 (nth 3 (car l))
1900 (car (nth 3 (car l))))
1902 (setq l (cdr l)))
1905 (let ((l (get-language-info language-name 'charset)))
1906 (if (null l)
1908 (while l
1909 (insert " " (symbol-name (car l)))
1910 (search-backward (symbol-name (car l)))
1911 (help-xref-button 0 'help-character-set (car l))
1912 (goto-char (point-max))
1914 (setq l (cdr l)))))
1917 (let ((l (get-language-info language-name 'coding-system)))
1918 (if (null l)
1920 (while l
1921 (insert " " (symbol-name (car l)))
1922 (search-backward (symbol-name (car l)))
1923 (help-xref-button 0 'help-coding-system (car l))
1924 (goto-char (point-max))
1925 (insert " (`"
1926 (coding-system-mnemonic (car l))
1928 (coding-system-doc-string (car l))
1930 (let ((aliases (coding-system-get (car l)
1931 'alias-coding-systems)))
1932 (when aliases
1934 (while aliases
1935 (insert " " (symbol-name (car aliases)))
1936 (setq aliases (cdr aliases)))
1938 (setq l (cdr l)))))))))
1939 \f
1940 ;;; Locales.
1942 (defvar locale-translation-file-name nil
1943 "File name for the system's file of locale-name aliases, or nil if none.")
1945 ;; The following definitions might as well be marked as constants and
1946 ;; purecopied, since they're normally used on startup, and probably
1947 ;; should reflect the facilities of the base Emacs.
1948 (defconst locale-language-names
1949 (purecopy
1950 '(
1951 ;; Locale names of the form LANGUAGE[_TERRITORY][.CODESET][@MODIFIER]
1952 ;; as specified in the Single Unix Spec, Version 2.
1953 ;; LANGUAGE is a language code taken from ISO 639:1988 (E/F)
1954 ;; with additions from ISO 639/RA Newsletter No.1/1989;
1955 ;; see Internet RFC 2165 (1997-06) and
1956 ;; http://www.evertype.com/standards/iso639/iso639-en.html
1957 ;; TERRITORY is a country code taken from ISO 3166
1958 ;; http://www.din.de/gremien/nas/nabd/iso3166ma/codlstp1/en_listp1.html.
1959 ;; CODESET and MODIFIER are implementation-dependent.
1961 ; aa Afar
1962 ; ab Abkhazian
1963 ("af" . "Latin-1") ; Afrikaans
1964 ("am" . "Ethiopic") ; Amharic
1965 ; ar Arabic glibc uses 8859-6
1966 ; as Assamese
1967 ; ay Aymara
1968 ; az Azerbaijani
1969 ; ba Bashkir
1970 ("be" . "Belarusian") ; Belarusian [Byelorussian until early 1990s]
1971 ("bg" . "Bulgarian") ; Bulgarian
1972 ; bh Bihari
1973 ; bi Bislama
1974 ; bn Bengali, Bangla
1975 ("bo" . "Tibetan")
1976 ("br" . "Latin-1") ; Breton
1977 ("bs" . "Latin-2") ; Bosnian
1978 ("ca" . "Latin-1") ; Catalan
1979 ; co Corsican
1980 ("cs" . "Czech")
1981 ("cy" . "Welsh") ; Welsh [glibc uses Latin-8. Did this change?]
1982 ("da" . "Latin-1") ; Danish
1983 ("de" . "German")
1984 ; dz Bhutani
1985 ("el" . "Greek")
1986 ;; Users who specify "en" explicitly typically want Latin-1, not ASCII.
1987 ;; That's actually what the GNU locales define, modulo things like
1988 ;; en_IN -- fx.
1989 ("en" . "Latin-1") ; English
1990 ("eo" . "Latin-3") ; Esperanto
1991 ("es" . "Spanish")
1992 ("et" . "Latin-4") ; Estonian
1993 ("eu" . "Latin-1") ; Basque
1994 ; fa Persian glibc uses utf-8
1995 ("fi" . "Latin-1") ; Finnish
1996 ; fj Fiji
1997 ("fo" . "Latin-1") ; Faroese
1998 ("fr" . "French") ; French
1999 ("fy" . "Latin-1") ; Frisian
2000 ("ga" . "Latin-1") ; Irish Gaelic (new orthography)
2001 ("gd" . "Latin-1") ; Scots Gaelic
2002 ("gl" . "Latin-1") ; Galician
2003 ; gn Guarani
2004 ; gu Gujarati
2005 ("gv" . "Latin-8") ; Manx Gaelic glibc uses 8859-1
2006 ; ha Hausa
2007 ("he" . "Hebrew")
2008 ("hi" . "Devanagari") ; Hindi glibc uses utf-8
2009 ("hr" . "Croatian") ; Croatian
2010 ("hu" . "Latin-2") ; Hungarian
2011 ; hy Armenian
2012 ; ia Interlingua
2013 ("id" . "Latin-1") ; Indonesian
2014 ; ie Interlingue
2015 ; ik Inupiak
2016 ("is" . "Latin-1") ; Icelandic
2017 ("it" . "Italian") ; Italian
2018 ; iu Inuktitut
2019 ("ja" . "Japanese")
2020 ; jw Javanese
2021 ("ka" . "Georgian") ; Georgian
2022 ; kk Kazakh
2023 ("kl" . "Latin-1") ; Greenlandic
2024 ; km Cambodian
2025 ; kn Kannada
2026 ("ko" . "Korean")
2027 ; ks Kashmiri
2028 ; ku Kurdish
2029 ("kw" . "Latin-1") ; Cornish
2030 ; ky Kirghiz
2031 ("la" . "Latin-1") ; Latin
2032 ("lb" . "Latin-1") ; Luxemburgish
2033 ; ln Lingala
2034 ("lo" . "Lao") ; Laothian
2035 ("lt" . "Lithuanian")
2036 ("lv" . "Latvian") ; Latvian, Lettish
2037 ; mg Malagasy
2038 ("mi" . "Latin-7") ; Maori
2039 ("mk" . "Cyrillic-ISO") ; Macedonian
2040 ; ml Malayalam
2041 ; mn Mongolian
2042 ; mo Moldavian
2043 ("mr" . "Devanagari") ; Marathi glibc uses utf-8
2044 ("ms" . "Latin-1") ; Malay
2045 ("mt" . "Latin-3") ; Maltese
2046 ; my Burmese
2047 ; na Nauru
2048 ("ne" . "Devanagari") ; Nepali
2049 ("nl" . "Dutch")
2050 ("no" . "Latin-1") ; Norwegian
2051 ("oc" . "Latin-1") ; Occitan
2052 ; om (Afan) Oromo
2053 ; or Oriya
2054 ; pa Punjabi
2055 ("pl" . "Latin-2") ; Polish
2056 ; ps Pashto, Pushto
2057 ("pt" . "Latin-1") ; Portuguese
2058 ; qu Quechua
2059 ("rm" . "Latin-1") ; Rhaeto-Romanic
2060 ; rn Kirundi
2061 ("ro" . "Romanian")
2062 ("ru.*[_.]koi8" . "Russian")
2063 ("ru" . "Cyrillic-ISO") ; Russian
2064 ; rw Kinyarwanda
2065 ("sa" . "Devanagari") ; Sanskrit
2066 ; sd Sindhi
2067 ; se Northern Sami
2068 ; sg Sangho
2069 ("sh" . "Latin-2") ; Serbo-Croatian
2070 ; si Sinhalese
2071 ("sk" . "Slovak")
2072 ("sl" . "Slovenian")
2073 ; sm Samoan
2074 ; sn Shona
2075 ; so Somali
2076 ("sq" . "Latin-1") ; Albanian
2077 ("sr" . "Latin-2") ; Serbian (Latin alphabet)
2078 ("sr_YU@cyrillic" . "Cyrillic-ISO") ; per glibc
2079 ; ss Siswati
2080 ; st Sesotho
2081 ; su Sundanese
2082 ("sv" . "Swedish") ; Swedish
2083 ("sw" . "Latin-1") ; Swahili
2084 ; ta Tamil glibc uses utf-8
2085 ; te Telugu glibc uses utf-8
2086 ("tg" . "Tajik")
2087 ("th" . "Thai")
2088 ; ti Tigrinya
2089 ; tk Turkmen
2090 ("tl" . "Latin-1") ; Tagalog
2091 ; tn Setswana
2092 ; to Tonga
2093 ("tr" . "Turkish")
2094 ; ts Tsonga
2095 ; tt Tatar
2096 ; tw Twi
2097 ; ug Uighur
2098 ("uk" . "Ukrainian") ; Ukrainian
2099 ; ur Urdu glibc uses utf-8
2100 ("uz" . "Latin-1") ; Uzbek
2101 ("vi" . "Vietnamese") ; glibc uses utf-8
2102 ; vo Volapuk
2103 ("wa" . "Latin-1") ; Walloon
2104 ; wo Wolof
2105 ; xh Xhosa
2106 ("yi" . "Windows-1255") ; Yiddish
2107 ; yo Yoruba
2108 ; za Zhuang
2110 ; glibc:
2111 ; zh_CN.GB18030/GB18030 \
2112 ; zh_CN.GBK/GBK \
2113 ; zh_HK/BIG5-HKSCS \
2115 ("zh.*[._]big5" . "Chinese-BIG5")
2116 ("zh.*[._]gbk" . nil) ; Solaris 2.7; has gbk-0 as well as GB 2312.1980-0
2117 ("zh_tw" . "Chinese-CNS") ; glibc uses big5
2118 ("zh_tw[._]euc-tw" . "Chinese-EUC-TW")
2119 ("zh" . "Chinese-GB")
2120 ; zu Zulu
2122 ;; ISO standard locales
2123 ("c$" . "ASCII")
2124 ("posix$" . "ASCII")
2126 ;; The "IPA" Emacs language environment does not correspond
2127 ;; to any ISO 639 code, so let it stand for itself.
2128 ("ipa$" . "IPA")
2130 ;; Nonstandard or obsolete language codes
2131 ("cz" . "Czech") ; e.g. Solaris 2.6
2132 ("ee" . "Latin-4") ; Estonian, e.g. X11R6.4
2133 ("iw" . "Hebrew") ; e.g. X11R6.4
2134 ("sp" . "Cyrillic-ISO") ; Serbian (Cyrillic alphabet), e.g. X11R6.4
2135 ("su" . "Latin-1") ; Finnish, e.g. Solaris 2.6
2136 ("jp" . "Japanese") ; e.g. MS Windows
2137 ("chs" . "Chinese-GB") ; MS Windows Chinese Simplified
2138 ("cht" . "Chinese-BIG5") ; MS Windows Chinese Traditional
2139 ))
2140 "List of pairs of locale regexps and language names.
2141 The first element whose locale regexp matches the start of a downcased locale
2142 specifies the language name corresponding to that locale.
2143 If the language name is nil, there is no corresponding language environment.")
2145 (defconst locale-charset-language-names
2146 (purecopy
2155 ;; utf-8@euro exists, so put this last. (@euro really specifies
2156 ;; the currency, rather than the charset.)
2158 "List of pairs of locale regexps and charset language names.
2159 The first element whose locale regexp matches the start of a downcased locale
2160 specifies the language name whose charsets corresponds to that locale.
2161 This language name is used if its charsets disagree with the charsets of
2162 the language name that would otherwise be used for this locale.")
2164 (defconst locale-preferred-coding-systems
2165 (purecopy
2166 '(("ja.*[._]euc" . japanese-iso-8bit)
2167 ("ja.*[._]jis7" . iso-2022-jp)
2168 ("ja.*[._]pck" . japanese-shift-jis)
2169 ("ja.*[._]sjis" . japanese-shift-jis)
2170 ("jpn" . japanese-shift-jis) ; MS-Windows uses this.
2171 (".*[._]utf" . utf-8)))
2172 "List of pairs of locale regexps and preferred coding systems.
2173 The first element whose locale regexp matches the start of a downcased locale
2174 specifies the coding system to prefer when using that locale.")
2176 (defun locale-name-match (key alist)
2177 "Search for KEY in ALIST, which should be a list of regexp-value pairs.
2178 Return the value corresponding to the first regexp that matches the
2179 start of KEY, or nil if there is no match."
2180 (let (element)
2181 (while (and alist (not element))
2183 (setq element (car alist)))
2184 (setq alist (cdr alist)))
2185 (cdr element)))
2187 (defun set-locale-environment (&optional locale-name)
2188 "Set up multi-lingual environment for using LOCALE-NAME.
2189 This sets the language environment, the coding system priority,
2190 the default input method and sometimes other things.
2192 LOCALE-NAME should be a string
2193 which is the name of a locale supported by the system;
2194 often it is of the form xx_XX.CODE, where xx is a language,
2195 XX is a country, and CODE specifies a character set and coding system.
2197 for Japanese in Japan using the `japanese-iso-8bit' coding-system.
2199 If LOCALE-NAME is nil, its value is taken from the environment
2200 variables LC_ALL, LC_CTYPE and LANG (the first one that is set).
2202 The locale names supported by your system can typically be found in a
2203 directory named `/usr/share/locale' or `/usr/lib/locale'. LOCALE-NAME
2204 will be translated according to the table specified by
2205 `locale-translation-file-name'.
2207 See also `locale-charset-language-names', `locale-language-names',
2208 `locale-preferred-coding-systems' and `locale-coding-system'."
2209 (interactive "sSet environment for locale: ")
2211 ;; Do this at runtime for the sake of binaries possibly transported
2212 ;; to a system without X.
2213 (setq locale-translation-file-name
2214 (let ((files
2215 '("/usr/lib/X11/locale/locale.alias" ; e.g. X11R6.4
2216 "/usr/X11R6/lib/X11/locale/locale.alias" ; e.g. RedHat 4.2
2217 "/usr/openwin/lib/locale/locale.alias" ; e.g. Solaris 2.6
2218 ;;
2219 ;; The following name appears after the X-related names above,
2220 ;; since the X-related names are what X actually uses.
2221 "/usr/share/locale/locale.alias" ; GNU/Linux sans X
2222 )))
2223 (while (and files (not (file-exists-p (car files))))
2224 (setq files (cdr files)))
2225 (car files)))
2227 (let ((locale locale-name))
2229 (unless locale
2230 ;; Use the first of these three environment variables
2231 ;; that has a nonempty value.
2232 (let ((vars '("LC_ALL" "LC_CTYPE" "LANG")))
2233 (while (and vars
2234 (= 0 (length locale))) ; nil or empty string
2235 (setq locale (getenv (pop vars))))))
2237 (when locale
2239 ;; Translate "swedish" into "sv_SE.ISO8859-1", and so on,
2240 ;; using the translation file that many systems have.
2241 (when locale-translation-file-name
2242 (with-temp-buffer
2243 (insert-file-contents locale-translation-file-name)
2244 (when (re-search-forward
2246 (setq locale (buffer-substring (point) (line-end-position))))))
2248 ;; Leave the system locales alone if the caller did not specify
2249 ;; an explicit locale name, as their defaults are set from
2250 ;; LC_MESSAGES and LC_TIME, not LC_CTYPE, and the user might not
2251 ;; want to set them to the same value as LC_CTYPE.
2252 (when locale-name
2253 (setq system-messages-locale locale)
2254 (setq system-time-locale locale))
2256 (setq locale (downcase locale))
2258 (let ((language-name
2259 (locale-name-match locale locale-language-names))
2260 (charset-language-name
2261 (locale-name-match locale locale-charset-language-names))
2262 (coding-system
2263 (locale-name-match locale locale-preferred-coding-systems)))
2265 ;; Give preference to charset-language-name over language-name.
2266 (if (and charset-language-name
2267 (not
2268 (equal (get-language-info language-name 'charset)
2269 (get-language-info charset-language-name 'charset))))
2270 (setq language-name charset-language-name))
2272 (when language-name
2274 ;; Set up for this character set. This is now the right way
2275 ;; to do it for both unibyte and multibyte modes.
2276 (set-language-environment language-name)
2278 ;; If default-enable-multibyte-characters is nil,
2279 ;; we are using single-byte characters,
2280 ;; so the display table and terminal coding system are irrelevant.
2281 (when default-enable-multibyte-characters
2282 (set-display-table-and-terminal-coding-system language-name))
2284 ;; Set the `keyboard-coding-system' if appropriate (tty
2285 ;; only). At least X and MS Windows can generate
2286 ;; multilingual input.
2287 (unless window-system
2288 (let ((kcs (or coding-system
2289 (car (get-language-info language-name
2290 'coding-system)))))
2291 (if kcs (set-keyboard-coding-system kcs))))
2293 (setq locale-coding-system
2294 (car (get-language-info language-name 'coding-priority))))
2296 (when coding-system
2297 (prefer-coding-system coding-system)
2298 (setq locale-coding-system coding-system))))
2300 ;; Default to A4 paper if we're not in a C, POSIX or US locale.
2301 ;; (See comments in Flanginfo.)
2302 (let ((locale locale)
2303 (paper (langinfo 'paper)))
2304 (if paper
2305 ;; This will always be null at the time of writing.
2306 (cond
2307 ((equal paper '(216 279))
2308 (setq ps-paper-type 'letter))
2309 ((equal paper '(210 297))
2310 (setq ps-paper-type 'a4)))
2311 (let ((vars '("LC_ALL" "LC_PAPER" "LANG")))
2312 (while (and vars (= 0 (length locale)))
2313 (setq locale (getenv (pop vars)))))
2314 (when locale
2315 ;; As of glibc 2.2.5, these are the only US Letter locales,
2316 ;; and the rest are A4.
2317 (setq ps-paper-type
2318 (or (locale-name-match locale '(("c$" . letter)
2319 ("posix$" . letter)
2320 (".._us" . letter)
2321 (".._pr" . letter)
2322 (".._ca" . letter)))
2323 'a4))))))
2324 nil)
2325 \f
2326 ;;; Charset property
2328 (defun get-charset-property (charset propname)
2329 "Return the value of CHARSET's PROPNAME property.
2330 This is the last value stored with
2331 (put-charset-property CHARSET PROPNAME VALUE)."
2332 (and (not (eq charset 'composition))
2333 (plist-get (charset-plist charset) propname)))
2335 (defun put-charset-property (charset propname value)
2336 "Store CHARSETS's PROPNAME property with value VALUE.
2337 It can be retrieved with `(get-charset-property CHARSET PROPNAME)'."
2338 (or (eq charset 'composition)
2339 (set-charset-plist charset
2340 (plist-put (charset-plist charset) propname value))))
2342 ;;; Character code property
2343 (put 'char-code-property-table 'char-table-extra-slots 0)
2345 (defvar char-code-property-table
2346 (make-char-table 'char-code-property-table)
2347 "Char-table containing a property list of each character code.
2349 See also the documentation of `get-char-code-property' and
2350 `put-char-code-property'.")
2352 (defun get-char-code-property (char propname)
2353 "Return the value of CHAR's PROPNAME property in `char-code-property-table'."
2354 (let ((plist (aref char-code-property-table char)))
2355 (if (listp plist)
2356 (car (cdr (memq propname plist))))))
2358 (defun put-char-code-property (char propname value)
2359 "Store CHAR's PROPNAME property with VALUE in `char-code-property-table'.
2360 It can be retrieved with `(get-char-code-property CHAR PROPNAME)'."
2361 (let ((plist (aref char-code-property-table char)))
2362 (if plist
2363 (let ((slot (memq propname plist)))
2364 (if slot
2365 (setcar (cdr slot) value)
2366 (nconc plist (list propname value))))
2367 (aset char-code-property-table char (list propname value)))))
2369 \f
2370 ;; Pretty description of encoded string
2372 ;; Alist of ISO 2022 control code vs the corresponding mnemonic string.
2373 (defvar iso-2022-control-alist
2381 (defun encoded-string-description (str coding-system)
2382 "Return a pretty description of STR that is encoded by CODING-SYSTEM."
2383 (setq str (string-as-unibyte str))
2384 (mapconcat
2385 (if (and coding-system (eq (coding-system-type coding-system) 2))
2386 ;; Try to get a pretty description for ISO 2022 escape sequences.
2387 (function (lambda (x) (or (cdr (assq x iso-2022-control-alist))
2388 (format "%02X" x))))
2389 (function (lambda (x) (format "0x%02X" x))))
2390 str " "))
2392 (defun encode-coding-char (char coding-system)
2393 "Encode CHAR by CODING-SYSTEM and return the resulting string.
2394 If CODING-SYSTEM can't safely encode CHAR, return nil."
2395 (let ((str1 (string-as-multibyte (char-to-string char)))
2396 (str2 (string-as-multibyte (make-string 2 char)))
2397 (safe-chars (and coding-system
2398 (coding-system-get coding-system 'safe-chars)))
2399 (charset (char-charset char))
2400 enc1 enc2 i1 i2)
2401 (when (or (eq safe-chars t)
2402 (eq charset 'ascii)
2403 (and safe-chars (aref safe-chars char)))
2404 ;; We must find the encoded string of CHAR. But, just encoding
2405 ;; CHAR will put extra control sequences (usually to designate
2406 ;; ASCII charaset) at the tail if type of CODING is ISO 2022.
2407 ;; To exclude such tailing bytes, we at first encode one-char
2408 ;; string and two-char string, then check how many bytes at the
2409 ;; tail of both encoded strings are the same.
2411 (setq enc1 (encode-coding-string str1 coding-system)
2412 i1 (length enc1)
2413 enc2 (encode-coding-string str2 coding-system)
2414 i2 (length enc2))
2415 (while (and (> i1 0) (= (aref enc1 (1- i1)) (aref enc2 (1- i2))))
2416 (setq i1 (1- i1) i2 (1- i2)))
2418 ;; Now (substring enc1 i1) and (substring enc2 i2) are the same,
2419 ;; and they are the extra control sequences at the tail to
2420 ;; exclude.
2421 (substring enc2 0 i2))))
2424 ;;; mule-cmds.el ends here