| blob | 2ee7c484bea7d4265de5df8ff97c90777d43a67e |
1 ;;; mule-cmds.el --- commands for mulitilingual environment
3 ;; Copyright (C) 1995 Electrotechnical Laboratory, JAPAN.
4 ;; Licensed to the Free Software Foundation.
5 ;; Copyright (C) 2000, 2001, 2002 Free Software Foundation, Inc.
7 ;; Keywords: mule, multilingual
9 ;; This file is part of GNU Emacs.
11 ;; GNU Emacs is free software; you can redistribute it and/or modify
12 ;; it under the terms of the GNU General Public License as published by
13 ;; the Free Software Foundation; either version 2, or (at your option)
14 ;; any later version.
16 ;; GNU Emacs is distributed in the hope that it will be useful,
17 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
18 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
19 ;; GNU General Public License for more details.
21 ;; You should have received a copy of the GNU General Public License
22 ;; along with GNU Emacs; see the file COPYING. If not, write to the
23 ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330,
24 ;; Boston, MA 02111-1307, USA.
26 ;;; Commentary:
28 ;;; Code:
32 ;;; MULE related key bindings and menus.
37 ;; 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)
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 isntance, can compute the
292 ;; prefix arg.
296 ;; This is the final call to `univeral-argument-other-key', which
297 ;; set's the final `prefix-arg.
301 ;; Read the command to execute with the given prefix arg.
313 "Set default value of various coding systems to CODING-SYSTEM.
314 This sets the following coding systems:
315 o coding system of a newly created buffer
316 o default coding system for subprocess I/O
317 This also sets the following values:
318 o default value used as `file-name-coding-system' for converting file names.
319 o default value for the command `set-terminal-coding-system' (not on MSDOS)
320 o default value for the command `set-keyboard-coding-system'."
325 ;; If coding-system is nil, honor that on MS-DOS as well, so
326 ;; that they could reset the terminal coding system.
330 ;; Preserve eol-type from existing default-process-coding-systems.
331 ;; On non-unix-like systems in particular, these may have been set
332 ;; carefully by the user, or by the startup code, to deal with the
333 ;; users shell appropriately, so should not be altered by changing
334 ;; language environment.
348 "Add CODING-SYSTEM at the front of the priority list for automatic detection.
349 This also sets the following coding systems:
350 o coding system of a newly created buffer
351 o default coding system for subprocess I/O
352 This also sets the following values:
353 o default value used as `file-name-coding-system' for converting file names.
354 o default value for the command `set-terminal-coding-system' (not on MSDOS)
355 o default value for the command `set-keyboard-coding-system'
357 If CODING-SYSTEM specifies a certain type of EOL conversion, the coding
358 systems set by this function will use that type of EOL conversion.
360 This command does not change the default value of terminal coding system
361 for MS-DOS terminal, because DOS terminals only support a single coding
362 system, and Emacs automatically sets the default to that coding system at
363 startup.
365 A coding system that requires automatic detection of text
366 encoding (e.g. undecided, unix) can't be preferred.
368 See also `coding-category-list' and `coding-system-category'."
376 ;; CODING-SYSTEM is no-conversion or undecided.
381 ;; We must change the order.
385 base coding-system))
386 ;; If they asked for specific EOL conversion, honor that.
394 "If non-nil, a predicate function to sort coding systems.
396 It is called with two coding systems, and should return t if the first
402 "Sort coding system list CODINGS by a priority of each coding system.
404 If a coding system is most preferred, it has the highest priority.
405 Otherwise, a coding system corresponds to some MIME charset has higher
406 priorities. Among them, a coding system included in `coding-system'
407 key of the current language environment has higher priorities. See
408 also the documentation of `language-info-alist'.
410 If the variable `sort-coding-systems-predicate' (which see) is
411 non-nil, it is used to sort CODINGS in the different way than above."
430 ;; For ISO based coding systems, prefer
431 ;; one that doesn't use escape sequences.
438 0
446 "Return a list of proper coding systems to encode a text between FROM and TO.
447 If FROM is a string, find coding systems in that instead of the buffer.
448 All coding systems in the list can safely encode any multibyte characters
449 in the text.
451 If the text contains no multibyte characters, return a list of a single
452 element `undecided'."
455 ;; The text contains only ASCII characters. Any coding
456 ;; systems are safe.
458 ;; We need copy-sequence because sorting will alter the argument.
462 "Return a list of proper coding systems to encode STRING.
463 All coding systems in the list can safely encode any multibyte characters
464 in STRING.
466 If STRING contains no multibyte characters, return a list of a single
467 element `undecided'."
471 "Return a list of proper coding systems to encode characters of CHARSETS.
472 CHARSETS is a list of character sets."
482 charset l ll)
499 "Find multibyte characters in the region specified by FROM and TO.
500 If FROM is a string, find multibyte characters in the string.
501 The return value is an alist of the following format:
502 ((CHARSET COUNT CHAR ...) ...)
503 where
504 CHARSET is a character set,
505 COUNT is a number of characters,
506 CHARs are found characters of the character set.
507 Optional 3rd arg MAXCOUNT limits how many CHARs are put in the above list.
508 Optional 4th arg EXCLUDE is a list of character sets to be ignored.
510 For invalid characters, CHARs are actually strings."
512 charset char)
521 eight-bit-control eight-bit-graphic))
553 "Search forward from point for a character that is not encodable.
554 It asks which coding system to check.
555 If such a character is found, set point after that character.
556 Otherwise, don't move point.
558 When called from a program, the value is a position of the found character,
559 or nil if all characters are encodable."
564 default))))
569 pos))
573 "Most recent coding system explicitly specified by the user when asked.
574 This variable is set whenever Emacs asks the user which coding system
575 to use in order to write a file. If you set it to nil explicitly,
576 then call `write-region', then afterward this variable will be non-nil
580 "If non-nil, a function to control the behaviour of coding system selection.
581 The meaning is the same as the argument ACCEPT-DEFAULT-P of the
582 function `select-safe-coding-system' (which see). This variable
586 accept-default-p file)
587 "Ask a user to select a safe coding system from candidates.
588 The candidates of coding systems which can safely encode a text
589 between FROM and TO are shown in a popup window. Among them, the most
590 proper one is suggested as the default.
592 The list of `buffer-file-coding-system' of the current buffer and the
593 most preferred coding system (if it corresponds to a MIME charset) is
594 treated as the default coding system list. Among them, the first one
595 that safely encodes the text is normally selected silently and
596 returned without any user interaction. See also the command
597 `prefer-coding-system'.
599 However, the user is queried if the chosen coding system is
600 inconsistent with what would be selected by `set-auto-coding' from
601 coding cookies &c. if the contents of the region were read from a
602 file. (That could lead to data corruption in a file subsequently
603 re-visited and edited.)
605 Optional 3rd arg DEFAULT-CODING-SYSTEM specifies a coding system or a
606 list of coding systems to be prepended to the default coding system
607 list.
609 Optional 4th arg ACCEPT-DEFAULT-P, if non-nil, is a function to
610 determine the acceptability of the silently selected coding system.
611 It is called with that coding system, and should return nil if it
612 should not be silently selected and thus user interaction is required.
614 Optional 5th arg FILE is the file name to use for this purpose.
615 That is different from `buffer-file-name' when handling `write-region'
616 \(for example).
618 The variable `select-safe-coding-system-accept-default-p', if
619 non-nil, overrides ACCEPT-DEFAULT-P.
621 Kludgy feature: if FROM is a string, the string is the target text,
622 and TO is ignored."
627 ;; Change elements of the list to (coding . base-coding).
630 default-coding-system))
632 ;; If buffer-file-coding-system is not nil nor undecided, append it
633 ;; to the defaults.
643 ;; If the most preferred coding system has the property mime-charset,
644 ;; append it to the defaults.
646 preferred base)
664 safe rejected unsafe)
666 ;; Any coding system is ok.
668 ;; Classify the defaults into safe, rejected, and unsafe.
679 ;; If all the defaults failed, ask a user.
681 ;; At first, if some defaults are unsafe, record at most 11
682 ;; problematic characters and their positions for them by turning
683 ;; (CODING ...)
684 ;; into
685 ;; ((CODING (POS . CHAR) (POS . CHAR) ...) ...)
696 unsafe))
704 unsafe))))
706 ;; Change each safe coding system to the corresponding
707 ;; mime-charset name if it is also a coding system. Such a name
708 ;; is more friendly to users.
710 mime-charset)
717 ;; Don't offer variations with locking shift, which you
718 ;; basically never want.
729 ;; If some defaults are unsafe, make sure the offending
730 ;; buffer is displayed.
734 unsafe))))
735 ;; Then ask users to select one from CODINGS while showing
736 ;; the reason why none of the defaults are not used.
746 "These default coding systems were tried to encode"
756 default-coding-system)
761 but it is not recommended for encoding text in this context,
795 'help-echo
796 "mouse-2, RET: jump to this character"
797 'help-function func1
800 "..."
802 'help-echo
803 "mouse-2, RET: next unencodable character"
804 'help-function func2
809 unsafe)
817 "\
822 codings)
826 ;; Read a coding system.
830 codings))
834 safe-names nil t nil nil
837 coding-system last-coding-system-specified)))
849 ;; Check we're not inconsistent with what `coding:' spec &c would
850 ;; give when file is re-read.
851 ;; But don't do this if we explicitly ignored the cookie
852 ;; by using `find-file-literally'.
862 ;; Don't barf if writing a compressed file, say.
863 ;; This check perhaps isn't ideal, but is probably
864 ;; the best thing to do.
870 %s specified by file contents. Really save (else edit coding cookies \
873 coding-system))
878 "Return a coding system to encode the outgoing message of the current buffer.
879 It at first tries the first coding system found in these variables
880 in this order:
881 (1) local value of `buffer-file-coding-system'
882 (2) value of `sendmail-coding-system'
883 (3) value of `default-sendmail-coding-system'
884 (4) value of `default-buffer-file-coding-system'
885 If the found coding system can't encode the current buffer,
886 or none of them are bound to a coding system,
887 it asks the user to select a proper coding system."
889 buffer-file-coding-system)
890 sendmail-coding-system
891 default-sendmail-coding-system
892 default-buffer-file-coding-system)))
894 ;; We should never use no-conversion for outgoing mail.
900 coding)))
901 \f
902 ;;; Language support stuff.
905 "Alist of language environment definitions.
906 Each element looks like:
907 (LANGUAGE-NAME . ((KEY . INFO) ...))
908 where LANGUAGE-NAME is a string, the name of the language environment,
909 KEY is a symbol denoting the kind of information, and
910 INFO is the data associated with KEY.
911 Meaningful values for KEY include
913 documentation value is documentation of what this language environment
914 is meant for, and how to use it.
915 charset value is a list of the character sets used by this
916 language environment.
917 sample-text value is an expression which is evalled to generate
918 a line of text written using characters appropriate
919 for this language environment.
920 setup-function value is a function to call to switch to this
921 language environment.
922 exit-function value is a function to call to leave this
923 language environment.
924 coding-system value is a list of coding systems that are good
925 for saving text written in this language environment.
926 This list serves as suggestions to the user;
927 in effect, as a kind of documentation.
928 coding-priority value is a list of coding systems for this language
929 environment, in order of decreasing priority.
930 This is used to set up the coding system priority
931 list when you switch to this language environment.
932 nonascii-translation
933 value is a translation table to be set in the
934 variable `nonascii-translation-table' in this
935 language environment, or a character set from
936 which `nonascii-insert-offset' is calculated.
937 input-method value is a default input method for this language
938 environment.
939 features value is a list of features requested in this
940 language environment.
942 The following keys take effect only when multibyte characters are
943 globally disabled, i.e. the value of `default-enable-multibyte-characters'
944 is nil.
946 unibyte-syntax value is a library name to load to set
947 unibyte 8-bit character syntaxes for this
948 language environment.
950 unibyte-display value is a coding system to encode characters
951 for the terminal. Characters in the range
952 of 160 to 255 display not as octal escapes,
953 but as non-ASCII characters in this language
957 "Return information listed under KEY for language environment LANG-ENV.
958 KEY is a symbol denoting the kind of information.
959 For a list of useful values for KEY and their meanings,
960 see `language-info-alist'."
968 "Modify part of the definition of language environment LANG-ENV.
969 Specifically, this stores the information INFO under KEY
970 in the definition of this language environment.
971 KEY is a symbol denoting the kind of information.
972 INFO is the value for that information.
974 For a list of useful values for KEY and their meanings,
975 see `language-info-alist'."
989 ;; Update the custom-type of `current-language-environment'.
998 "Store ALIST as the definition of language environment LANG-ENV.
999 ALIST is an alist of KEY and INFO values. See the documentation of
1000 `language-info-alist' for the meanings of KEY and INFO.
1002 Optional arg PARENTS is a list of parent menu names; it specifies
1003 where to put this language environment in the
1004 Describe Language Environment and Set Language Environment menus.
1006 in the European submenu in each of those two menus."
1013 map parent-symbol parent prompt)
1019 ;; This prompt string is for define-prefix-command, so
1020 ;; that the map it creates will be suitable for a menu.
1041 ;; Set up menu items for this language env.
1054 "Read a language environment name which has information for KEY.
1055 If KEY is nil, read any language environment.
1056 Prompt with PROMPT. DEFAULT is the default choice of language environment.
1057 This returns a language environment name as a string."
1060 language-info-alist
1063 t nil nil default)))
1067 name)))
1068 \f
1069 ;;; Multilingual input methods.
1071 "LEIM: Libraries of Emacs Input Methods."
1075 "Name of LEIM list file.
1076 This file contains a list of libraries of Emacs input methods (LEIM)
1077 in the format of Lisp expression for registering each input method.
1081 ";;; %s -- list of LEIM (Library of Emacs Input Method) -*-coding: iso-2022-7bit;-*-
1082 ;;
1083 ;; This file contains a list of LEIM (Library of Emacs Input Method)
1084 ;; methods in the same directory as this file. Loading this file
1085 ;; registers all the input methods in Emacs.
1086 ;;
1087 ;; Each entry has the form:
1088 ;; (register-input-method
1089 ;; INPUT-METHOD LANGUAGE-NAME ACTIVATE-FUNC
1090 ;; TITLE DESCRIPTION
1091 ;; ARG ...)
1092 ;; See the function `register-input-method' for the meanings of the arguments.
1093 ;;
1094 ;; If this directory is included in load-path, Emacs automatically
1095 ;; loads this file at startup time.
1097 "
1098 leim-list-file-name)
1102 "Regexp matching head of each entry in LEIM list file.
1107 "List of functions to call to update LEIM list file.
1111 "Update LEIM list file in directories DIRS."
1118 "The current input method for multilingual text.
1129 "*Default input method for multilingual text (a string).
1130 This is the input method activated automatically by the command
1144 "Function to call for inactivating the current input method.
1145 Every input method should set this to an appropriate value when activated.
1146 This function is called with no argument.
1148 This function should never change the value of `current-input-method'.
1154 "Function to call for describing the current input method.
1160 "Alist of input method names vs how to use them.
1161 Each element has the form:
1162 (INPUT-METHOD LANGUAGE-ENV ACTIVATE-FUNC TITLE DESCRIPTION ARGS...)
1166 "Register INPUT-METHOD as an input method for language environment LANG-ENV.
1167 INPUT-METHOD and LANG-ENV are symbols or strings.
1169 The remaining arguments are:
1170 ACTIVATE-FUNC, TITLE, DESCRIPTION, and ARGS...
1171 ACTIVATE-FUNC is a function to call to activate this method.
1172 TITLE is a string to show in the mode line when this method is active.
1173 DESCRIPTION is a string describing this method and what it is good for.
1174 The ARGS, if any, are passed as arguments to ACTIVATE-FUNC.
1175 All told, the arguments to ACTIVATE-FUNC are INPUT-METHOD and the ARGS.
1178 created at Emacs build time, registering all Quail input methods
1179 contained in the Emacs distribution.
1181 In case you want to register a new Quail input method by yourself, be
1182 careful to use the same input method title as given in the third
1183 parameter of `quail-define-package'. (If the values are different, the
1184 string specified in this function takes precedence.)
1186 The commands `describe-input-method' and `list-input-methods' need
1187 these duplicated values to show some information about input methods
1188 without loading the relevant Quail packages."
1201 "Read a name of input method from a minibuffer prompting with PROMPT.
1202 If DEFAULT is non-nil, use that as the default,
1203 and substitute it into PROMPT at the first `%s'.
1204 If INHIBIT-NULL is non-nil, null input signals an error.
1206 The return value is a string."
1210 ;; As it is quite normal to change input method in the
1211 ;; minibuffer, we must enable it even if
1212 ;; enable-recursive-minibuffers is currently nil.
1214 ;; This binding is necessary because input-method-history is
1215 ;; buffer local.
1217 nil t nil 'input-method-history
1218 default)))
1222 input-method
1227 "Switch to input method INPUT-METHOD for the current buffer.
1228 If some other input method is already active, turn it off first.
1229 If INPUT-METHOD is nil, deactivate any current input method."
1256 "Turn off the current input method."
1269 input-method-function nil
1270 current-input-method-title nil)
1274 "Select and activate input method INPUT-METHOD for the current buffer.
1275 This also sets the default input method to the one you specify.
1276 If INPUT-METHOD is nil, this function turns off the input method, and
1277 also causes you to be prompted for a name of an input method the next
1286 default t))))
1291 default-input-method)
1294 "Enable or disable multilingual text input method for the current buffer.
1295 Only one input method can be enabled at any time in a given buffer.
1297 The normal action is to enable an input method if none was
1298 enabled, and disable the current one otherwise. Which input method
1299 to enable can be determined in various ways--either the one most
1300 recently used, or the one specified by `default-input-method', or
1301 as a last resort by reading the name of an input method in the
1302 minibuffer.
1304 With a prefix argument, read an input method name with the minibuffer
1305 and enable that one. The default is the most recent input method specified
1306 \(not including the currently active input method, if any)."
1320 default t))
1321 default))
1329 "Describe input method INPUT-METHOD."
1359 "Describe the input method currently in use.
1360 This is a subroutine for `describe-input-method'."
1366 current-input-method)
1371 "Read a multilingual string from minibuffer, prompting with string PROMPT.
1372 The input method selected last time is activated in minibuffer.
1373 If optional second arg INITIAL-INPUT is non-nil, insert it in the minibuffer
1374 initially.
1375 Optional 3rd argument INPUT-METHOD specifies the input method
1376 to be activated instead of the one selected last time. It is a symbol
1377 or a string."
1380 current-input-method
1381 default-input-method
1392 ;; Variables to control behavior of input methods. All input methods
1393 ;; should react to these variables.
1396 "*A flag to control extra guidance given by input methods.
1397 The value should be nil, t, `complex-only', or `default'.
1399 The extra guidance is done by showing list of available keys in echo
1400 area. When you use the input method in the minibuffer, the guidance
1401 is shown at the bottom short window (split from the existing window).
1403 If the value is t, extra guidance is always given, if the value is
1404 nil, extra guidance is always suppressed.
1406 If the value is `complex-only', only complex input methods such as
1407 `chinese-py' and `japanese' give extra guidance.
1409 If the value is `default', complex input methods always give extra
1410 guidance, but simple input methods give it only when you are not in
1411 the minibuffer.
1413 See also the variable `input-method-highlight-flag'."
1419 "*If this flag is non-nil, input methods highlight partially-entered text.
1420 For instance, while you are in the middle of a Quail input method sequence,
1421 the text inserted so far is temporarily underlined.
1422 The underlining goes away when you finish or abort the input method sequence.
1423 See also the variable `input-method-verbose-flag'."
1428 "Normal hook run just after an input method is activated.
1430 The variable `current-input-method' keeps the input method name
1434 "Normal hook run just after an input method is inactivated.
1436 The variable `current-input-method' still keeps the input method name
1443 "This flag controls when an input method returns.
1444 Usually, the input method does not return while there's a possibility
1445 that it may find a different translation if a user types another key.
1446 But, it this flag is non-nil, the input method returns as soon as
1450 "This flag controls how an input method shows an intermediate key sequence.
1451 Usually, the input method inserts the intermediate key sequence,
1452 or candidate translations corresponding to the sequence,
1453 at point in the current buffer.
1457 "This flag controls the behaviour of an input method on invalid key input.
1458 Usually, when a user types a key which doesn't start any character
1459 handled by the input method, the key is handled by turning off the
1460 input method temporarily. After that key, the input method is re-enabled.
1463 \f
1465 "Normal hook run after some language environment is set.
1467 When you set some hook function here, that effect usually should not
1468 be inherited to another language environment. So, you had better set
1469 another function in `exit-language-environment-hook' (which see) to
1473 "Normal hook run after exiting from some language environment.
1474 When this hook is run, the variable `current-language-environment'
1475 is still bound to the language environment being exited.
1477 This hook is mainly used for canceling the effect of
1483 "Switch to a specified language environment."
1496 "The last language environment specified with `set-language-environment'.
1498 to using the function `set-language-environment'."
1505 current-language-environment)
1506 language-info-alist))
1508 ;; custom type will be updated with `set-language-info'.
1520 "Reset multilingual environment of Emacs to the default status.
1522 The default status is as follows:
1524 The default value of `buffer-file-coding-system' is nil.
1525 The default coding system for process I/O is nil.
1526 The default value for the command `set-terminal-coding-system' is nil.
1527 The default value for the command `set-keyboard-coding-system' is nil.
1529 The order of priorities of coding categories and the coding system
1530 bound to each category are as follows
1531 coding category coding system
1532 --------------------------------------------------
1533 coding-category-iso-8-2 iso-latin-1
1534 coding-category-iso-8-1 iso-latin-1
1535 coding-category-iso-7-tight iso-2022-jp
1536 coding-category-iso-7 iso-2022-7bit
1537 coding-category-iso-7-else iso-2022-7bit-lock
1538 coding-category-iso-8-else iso-2022-8bit-ss2
1539 coding-category-emacs-mule emacs-mule
1540 coding-category-raw-text raw-text
1541 coding-category-sjis japanese-shift-jis
1542 coding-category-big5 chinese-big5
1543 coding-category-ccl nil
1544 coding-category-binary no-conversion
1545 coding-category-utf-16-be nil
1546 coding-category-utf-16-le nil
1547 coding-category-utf-8 mule-utf-8"
1549 ;; This function formerly set default-enable-multibyte-characters to t,
1550 ;; but that is incorrect. It should not alter the unibyte/multibyte choice.
1553 coding-category-iso-7 'iso-2022-7bit
1554 coding-category-iso-8-1 'iso-latin-1
1555 coding-category-iso-8-2 'iso-latin-1
1556 coding-category-iso-7-else 'iso-2022-7bit-lock
1557 coding-category-iso-8-else 'iso-2022-8bit-ss2
1558 coding-category-emacs-mule 'emacs-mule
1559 coding-category-raw-text 'raw-text
1560 coding-category-sjis 'japanese-shift-jis
1561 coding-category-big5 'chinese-big5
1562 coding-category-utf-16-be nil
1563 coding-category-utf-16-le nil
1564 coding-category-utf-8 'mule-utf-8
1565 coding-category-ccl nil
1570 coding-category-iso-8-2
1571 coding-category-iso-7-tight
1572 coding-category-iso-7
1573 coding-category-iso-7-else
1574 coding-category-iso-8-else
1575 coding-category-emacs-mule
1576 coding-category-raw-text
1577 coding-category-sjis
1578 coding-category-big5
1579 coding-category-ccl
1580 coding-category-binary
1581 coding-category-utf-16-be
1582 coding-category-utf-16-le
1583 coding-category-utf-8))
1589 ;; Preserve eol-type from existing default-process-coding-systems.
1590 ;; On non-unix-like systems in particular, these may have been set
1591 ;; carefully by the user, or by the startup code, to deal with the
1592 ;; users shell appropriately, so should not be altered by changing
1593 ;; language environment.
1595 ;; When bootstrapping, coding-systems are not defined yet, so
1596 ;; we need to catch the error from check-coding-system.
1609 ;; Don't alter the terminal and keyboard coding systems here.
1610 ;; The terminal still supports the same coding system
1611 ;; that it supported a minute ago.
1612 ;; (set-terminal-coding-system-internal nil)
1613 ;; (set-keyboard-coding-system-internal nil)
1621 "Set up the display table and terminal coding system for LANGUAGE-NAME."
1631 "Set up multi-lingual environment for using LANGUAGE-NAME.
1632 This sets the coding system priority and the default input method
1633 and sometimes other things. LANGUAGE-NAME should be a string
1635 specifies the character set for the major languages of Western Europe."
1637 nil
1651 default-buffer-file-coding-system)))
1654 ;; The features might set up coding systems.
1678 ;; DOS terminals' default is to use a special non-ASCII translation
1679 ;; table as appropriate for the installed codepage.
1684 ;; Unibyte setups if necessary.
1686 ;; Syntax and case table.
1691 ;; No information for syntax and case. Reset to the defaults.
1721 ;; Actually set up direct output of non-ASCII characters.
1723 ;; Unibyte Emacs on MS-DOS wants to display all 8-bit characters with
1724 ;; the native font, and codes 160 and 146 stand for something very
1725 ;; different there.
1728 ;; Make non-line-break space display as a plain space.
1729 ;; Most X fonts do the wrong thing for code 160.
1731 ;; With luck, non-Latin-1 fonts are more recent and so don't
1732 ;; have this bug.
1734 ;; Most Windows programs send out apostrophes as \222. Most X fonts
1735 ;; don't contain a character at that position. Map it to the ASCII
1736 ;; apostrophe. [This is actually RIGHT SINGLE QUOTATION MARK,
1737 ;; U+2019, normally from the windows-1252 character set. XFree 4
1738 ;; fonts probably have the appropriate glyph at this position,
1739 ;; so they could use standard-display-8bit. It's better to use a
1740 ;; proper windows-1252 coding system. --fx]
1743 (defun set-language-environment-coding-systems (language-name
1744 &optional eol-type)
1745 "Do various coding system setups for language environment LANGUAGE-NAME.
1747 The optional arg EOL-TYPE specifies the eol-type of the default value
1748 of `buffer-file-coding-system' set by this function."
1749 (let* ((priority (get-language-info language-name 'coding-priority))
1750 (default-coding (car priority)))
1751 (if priority
1752 (let ((categories (mapcar 'coding-system-category priority)))
1753 (set-default-coding-systems
1754 (if (memq eol-type '(0 1 2 unix dos mac))
1755 (coding-system-change-eol-conversion default-coding eol-type)
1756 default-coding))
1757 (setq default-sendmail-coding-system default-coding)
1758 (set-coding-priority categories)
1759 (while priority
1760 (set (car categories) (car priority))
1761 (setq priority (cdr priority) categories (cdr categories)))
1762 (update-coding-systems-internal)))))
1764 (defsubst princ-list (&rest args)
1766 (while args (princ (car args)) (setq args (cdr args)))
1769 (put 'describe-specified-language-support 'apropos-inhibit t)
1771 ;; Print a language specific information such as input methods,
1772 ;; charsets, and coding systems. This function is intended to be
1773 ;; called from the menu:
1774 ;; [menu-bar mule describe-language-environment LANGUAGE]
1775 ;; and should not run it by `M-x describe-current-input-method-function'.
1776 (defun describe-specified-language-support ()
1777 "Describe how Emacs supports the specified language environment."
1778 (interactive)
1779 (let (language-name)
1780 (if (not (and (symbolp last-command-event)
1781 (or (not (eq last-command-event 'Default))
1782 (setq last-command-event 'English))
1783 (setq language-name (symbol-name last-command-event))))
1784 (error "Bogus calling sequence"))
1785 (describe-language-environment language-name)))
1787 (defun describe-language-environment (language-name)
1788 "Describe how Emacs supports language environment LANGUAGE-NAME."
1789 (interactive
1790 (list (read-language-name
1791 'documentation
1792 "Describe language environment (default, current choice): ")))
1793 (if (null language-name)
1794 (setq language-name current-language-environment))
1795 (if (or (null language-name)
1796 (null (get-language-info language-name 'documentation)))
1797 (error "No documentation for the specified language"))
1798 (if (symbolp language-name)
1799 (setq language-name (symbol-name language-name)))
1800 (let ((doc (get-language-info language-name 'documentation))
1801 pos)
1802 (help-setup-xref (list #'describe-language-environment language-name)
1803 (interactive-p))
1804 (with-output-to-temp-buffer (help-buffer)
1805 (save-excursion
1806 (set-buffer standard-output)
1808 (if (stringp doc)
1810 (condition-case nil
1811 (let ((str (eval (get-language-info language-name 'sample-text))))
1812 (if (stringp str)
1814 (error nil))
1815 (let ((input-method (get-language-info language-name 'input-method))
1816 (l (copy-sequence input-method-alist)))
1817 (insert "Input methods")
1818 (when input-method
1819 (insert " (default, " input-method ")")
1820 (setq input-method (assoc input-method input-method-alist))
1821 (setq l (cons input-method (delete input-method l))))
1823 (while l
1824 (when (string= language-name (nth 1 (car l)))
1825 (insert " " (car (car l)))
1826 (search-backward (car (car l)))
1827 (help-xref-button 0 'help-input-method (car (car l)))
1828 (goto-char (point-max))
1830 (if (stringp (nth 3 (car l)))
1831 (nth 3 (car l))
1832 (car (nth 3 (car l))))
1834 (setq l (cdr l)))
1837 (let ((l (get-language-info language-name 'charset)))
1838 (if (null l)
1840 (while l
1841 (insert " " (symbol-name (car l)))
1842 (search-backward (symbol-name (car l)))
1843 (help-xref-button 0 'help-character-set (car l))
1844 (goto-char (point-max))
1846 (setq l (cdr l)))))
1849 (let ((l (get-language-info language-name 'coding-system)))
1850 (if (null l)
1852 (while l
1853 (insert " " (symbol-name (car l)))
1854 (search-backward (symbol-name (car l)))
1855 (help-xref-button 0 'help-coding-system (car l))
1856 (goto-char (point-max))
1857 (insert " (`"
1858 (coding-system-mnemonic (car l))
1860 (coding-system-doc-string (car l))
1862 (let ((aliases (coding-system-get (car l)
1863 'alias-coding-systems)))
1864 (when aliases
1866 (while aliases
1867 (insert " " (symbol-name (car aliases)))
1868 (setq aliases (cdr aliases)))
1870 (setq l (cdr l)))))))))
1871 \f
1872 ;;; Locales.
1874 (defvar locale-translation-file-name nil
1875 "File name for the system's file of locale-name aliases, or nil if none.")
1877 ;; The following definitions might as well be marked as constants and
1878 ;; purecopied, since they're normally used on startup, and probably
1879 ;; should reflect the facilities of the base Emacs.
1880 (defconst locale-language-names
1881 (purecopy
1882 '(
1883 ;; Locale names of the form LANGUAGE[_TERRITORY][.CODESET][@MODIFIER]
1884 ;; as specified in the Single Unix Spec, Version 2.
1885 ;; LANGUAGE is a language code taken from ISO 639:1988 (E/F)
1886 ;; with additions from ISO 639/RA Newsletter No.1/1989;
1887 ;; see Internet RFC 2165 (1997-06) and
1888 ;; http://www.evertype.com/standards/iso639/iso639-en.html
1889 ;; TERRITORY is a country code taken from ISO 3166
1890 ;; http://www.din.de/gremien/nas/nabd/iso3166ma/codlstp1/en_listp1.html.
1891 ;; CODESET and MODIFIER are implementation-dependent.
1893 ; aa Afar
1894 ; ab Abkhazian
1895 ("af" . "Latin-1") ; Afrikaans
1896 ("am" . "Ethiopic") ; Amharic
1897 ; ar Arabic glibc uses 8859-6
1898 ; as Assamese
1899 ; ay Aymara
1900 ; az Azerbaijani
1901 ; ba Bashkir
1902 ("be" . "Belarusian") ; Belarusian [Byelorussian until early 1990s]
1903 ("bg" . "Bulgarian") ; Bulgarian
1904 ; bh Bihari
1905 ; bi Bislama
1906 ; bn Bengali, Bangla
1907 ("bo" . "Tibetan")
1908 ("br" . "Latin-1") ; Breton
1909 ("bs" . "Latin-2") ; Bosnian
1910 ("ca" . "Latin-1") ; Catalan
1911 ; co Corsican
1912 ("cs" . "Czech")
1913 ("cy" . "Welsh") ; Welsh
1914 ("da" . "Latin-1") ; Danish
1915 ("de" . "German")
1916 ; dz Bhutani
1917 ("el" . "Greek")
1918 ;; Users who specify "en" explicitly typically want Latin-1, not ASCII.
1919 ;; That's actually what the GNU locales define, modulo things like
1920 ;; en_IN -- fx.
1921 ("en" . "Latin-1") ; English
1922 ("eo" . "Latin-3") ; Esperanto
1923 ("es" . "Spanish")
1924 ("et" . "Latin-4") ; Estonian
1925 ("eu" . "Latin-1") ; Basque
1926 ; fa Persian glibc uses utf-8
1927 ("fi" . "Latin-1") ; Finnish
1928 ; fj Fiji
1929 ("fo" . "Latin-1") ; Faroese
1930 ("fr" . "French") ; French
1931 ("fy" . "Latin-1") ; Frisian
1932 ("ga" . "Latin-1") ; Irish Gaelic (new orthography)
1933 ("gd" . "Latin-1") ; Scots Gaelic
1934 ("gl" . "Latin-1") ; Galician
1935 ; gn Guarani
1936 ; gu Gujarati
1937 ("gv" . "Latin-8") ; Manx Gaelic glibc uses 8859-1
1938 ; ha Hausa
1939 ("he" . "Hebrew")
1940 ("hi" . "Devanagari") ; Hindi glibc uses utf-8
1941 ("hr" . "Latin-2") ; Croatian
1942 ("hu" . "Latin-2") ; Hungarian
1943 ; hy Armenian
1944 ; ia Interlingua
1945 ("id" . "Latin-1") ; Indonesian
1946 ; ie Interlingue
1947 ; ik Inupiak
1948 ("is" . "Latin-1") ; Icelandic
1949 ("it" . "Latin-1") ; Italian
1950 ; iu Inuktitut
1951 ("ja" . "Japanese")
1952 ; jw Javanese
1953 ("ka" . "Georgian") ; Georgian
1954 ; kk Kazakh
1955 ("kl" . "Latin-1") ; Greenlandic
1956 ; km Cambodian
1957 ; kn Kannada
1958 ("ko" . "Korean")
1959 ; ks Kashmiri
1960 ; ku Kurdish
1961 ("kw" . "Latin-1") ; Cornish
1962 ; ky Kirghiz
1963 ("la" . "Latin-1") ; Latin
1964 ("lb" . "Latin-1") ; Luxemburgish
1965 ; ln Lingala
1966 ("lo" . "Lao") ; Laothian
1967 ("lt" . "Lithuanian")
1968 ("lv" . "Latvian") ; Latvian, Lettish
1969 ; mg Malagasy
1970 ("mi" . "Latin-7") ; Maori
1971 ("mk" . "Cyrillic-ISO") ; Macedonian
1972 ; ml Malayalam
1973 ; mn Mongolian
1974 ; mo Moldavian
1975 ("mr" . "Devanagari") ; Marathi glibc uses utf-8
1976 ("ms" . "Latin-1") ; Malay
1977 ("mt" . "Latin-3") ; Maltese
1978 ; my Burmese
1979 ; na Nauru
1980 ("ne" . "Devanagari") ; Nepali
1981 ("nl" . "Dutch")
1982 ("no" . "Latin-1") ; Norwegian
1983 ("oc" . "Latin-1") ; Occitan
1984 ; om (Afan) Oromo
1985 ; or Oriya
1986 ; pa Punjabi
1987 ("pl" . "Latin-2") ; Polish
1988 ; ps Pashto, Pushto
1989 ("pt" . "Latin-1") ; Portuguese
1990 ; qu Quechua
1991 ("rm" . "Latin-1") ; Rhaeto-Romanic
1992 ; rn Kirundi
1993 ("ro" . "Romanian")
1994 ("ru.*[_.]koi8" . "Cyrillic-KOI8") ; Russian
1995 ("ru" . "Cyrillic-ISO") ; Russian
1996 ; rw Kinyarwanda
1997 ("sa" . "Devanagari") ; Sanskrit
1998 ; sd Sindhi
1999 ; se Northern Sami
2000 ; sg Sangho
2001 ("sh" . "Latin-2") ; Serbo-Croatian
2002 ; si Sinhalese
2003 ("sk" . "Slovak")
2004 ("sl" . "Slovenian")
2005 ; sm Samoan
2006 ; sn Shona
2007 ; so Somali
2008 ("sq" . "Latin-1") ; Albanian
2009 ("sr" . "Latin-2") ; Serbian (Latin alphabet)
2010 ("sr_YU@cyrillic" . "Cyrillic-ISO") ; per glibc
2011 ; ss Siswati
2012 ; st Sesotho
2013 ; su Sundanese
2014 ("sv" . "Latin-1") ; Swedish
2015 ("sw" . "Latin-1") ; Swahili
2016 ; ta Tamil glibc uses utf-8
2017 ; te Telugu glibc uses utf-8
2018 ("tg" . "Tajik")
2019 ("th" . "Thai")
2020 ; ti Tigrinya
2021 ; tk Turkmen
2022 ("tl" . "Latin-1") ; Tagalog
2023 ; tn Setswana
2024 ; to Tonga
2025 ("tr" . "Latin-5") ; Turkish
2026 ; ts Tsonga
2027 ; tt Tatar
2028 ; tw Twi
2029 ; ug Uighur
2030 ("uk" . "Ukrainian") ; Ukrainian
2031 ; ur Urdu glibc uses utf-8
2032 ("uz" . "Latin-1") ; Uzbek
2033 ("vi" . "Vietnamese") ; glibc uses utf-8
2034 ; vo Volapuk
2035 ("wa" . "Latin-1") ; Walloon
2036 ; wo Wolof
2037 ; xh Xhosa
2038 ("yi" . "Windows-1255") ; Yiddish
2039 ; yo Yoruba
2040 ; za Zhuang
2042 ; glibc:
2043 ; zh_CN.GB18030/GB18030 \
2044 ; zh_CN.GBK/GBK \
2045 ; zh_HK/BIG5-HKSCS \
2046 ; zh_TW/BIG5 \
2047 ; zh_TW.EUC-TW/EUC-TW \
2049 ("zh.*[._]big5" . "Chinese-BIG5")
2050 ("zh.*[._]gbk" . nil) ; Solaris 2.7; has gbk-0 as well as GB 2312.1980-0
2051 ("zh_tw" . "Chinese-CNS")
2052 ("zh" . "Chinese-GB")
2053 ; zu Zulu
2055 ;; ISO standard locales
2056 ("c$" . "ASCII")
2057 ("posix$" . "ASCII")
2059 ;; The "IPA" Emacs language environment does not correspond
2060 ;; to any ISO 639 code, so let it stand for itself.
2061 ("ipa$" . "IPA")
2063 ;; Nonstandard or obsolete language codes
2064 ("cz" . "Czech") ; e.g. Solaris 2.6
2065 ("ee" . "Latin-4") ; Estonian, e.g. X11R6.4
2066 ("iw" . "Hebrew") ; e.g. X11R6.4
2067 ("sp" . "Cyrillic-ISO") ; Serbian (Cyrillic alphabet), e.g. X11R6.4
2068 ("su" . "Latin-1") ; Finnish, e.g. Solaris 2.6
2069 ("jp" . "Japanese") ; e.g. MS Windows
2070 ("chs" . "Chinese-GB") ; MS Windows Chinese Simplified
2071 ("cht" . "Chinese-BIG5") ; MS Windows Chinese Traditional
2072 ))
2073 "List of pairs of locale regexps and language names.
2074 The first element whose locale regexp matches the start of a downcased locale
2075 specifies the language name corresponding to that locale.
2076 If the language name is nil, there is no corresponding language environment.")
2078 (defconst locale-charset-language-names
2079 (purecopy
2088 ;; utf-8@euro exists, so put this last. (@euro really specifies
2089 ;; the currency, rather than the charset.)
2091 "List of pairs of locale regexps and charset language names.
2092 The first element whose locale regexp matches the start of a downcased locale
2093 specifies the language name whose charsets corresponds to that locale.
2094 This language name is used if its charsets disagree with the charsets of
2095 the language name that would otherwise be used for this locale.")
2097 (defconst locale-preferred-coding-systems
2098 (purecopy
2099 '(("ja.*[._]euc" . japanese-iso-8bit)
2100 ("ja.*[._]jis7" . iso-2022-jp)
2101 ("ja.*[._]pck" . japanese-shift-jis)
2102 ("ja.*[._]sjis" . japanese-shift-jis)
2103 ("jpn" . japanese-shift-jis) ; MS-Windows uses this.
2104 (".*[._]utf" . utf-8)))
2105 "List of pairs of locale regexps and preferred coding systems.
2106 The first element whose locale regexp matches the start of a downcased locale
2107 specifies the coding system to prefer when using that locale.")
2109 (defconst standard-keyboard-coding-systems
2110 (purecopy
2111 '(iso-latin-1 iso-latin-2 iso-latin-3 iso-latin-4 iso-latin-5
2112 iso-latin-6 iso-latin-7 iso-latin-8 iso-latin-9 koi8-u koi8-r))
2113 "Coding systems that are commonly used for keyboards.
2114 `set-locale-environment' will set the `keyboard-coding-system' if the
2115 coding-system specified by the locale setting is a member of this list.")
2117 (defun locale-name-match (key alist)
2118 "Search for KEY in ALIST, which should be a list of regexp-value pairs.
2119 Return the value corresponding to the first regexp that matches the
2120 start of KEY, or nil if there is no match."
2121 (let (element)
2122 (while (and alist (not element))
2124 (setq element (car alist)))
2125 (setq alist (cdr alist)))
2126 (cdr element)))
2128 (defun set-locale-environment (&optional locale-name)
2129 "Set up multi-lingual environment for using LOCALE-NAME.
2130 This sets the language environment, the coding system priority,
2131 the default input method and sometimes other things.
2133 LOCALE-NAME should be a string
2134 which is the name of a locale supported by the system;
2135 often it is of the form xx_XX.CODE, where xx is a language,
2136 XX is a country, and CODE specifies a character set and coding system.
2138 for Japanese in Japan using the `japanese-iso-8bit' coding-system.
2140 If LOCALE-NAME is nil, its value is taken from the environment
2141 variables LC_ALL, LC_CTYPE and LANG (the first one that is set).
2143 The locale names supported by your system can typically be found in a
2144 directory named `/usr/share/locale' or `/usr/lib/locale'. LOCALE-NAME
2145 will be translated according to the table specified by
2146 `locale-translation-file-name'.
2148 See also `locale-charset-language-names', `locale-language-names',
2149 `locale-preferred-coding-systems' and `locale-coding-system'."
2150 (interactive "sSet environment for locale: ")
2152 ;; Do this at runtime for the sake of binaries possibly transported
2153 ;; to a system without X.
2154 (setq locale-translation-file-name
2155 (let ((files
2156 '("/usr/lib/X11/locale/locale.alias" ; e.g. X11R6.4
2157 "/usr/X11R6/lib/X11/locale/locale.alias" ; e.g. RedHat 4.2
2158 "/usr/openwin/lib/locale/locale.alias" ; e.g. Solaris 2.6
2159 ;;
2160 ;; The following name appears after the X-related names above,
2161 ;; since the X-related names are what X actually uses.
2162 "/usr/share/locale/locale.alias" ; GNU/Linux sans X
2163 )))
2164 (while (and files (not (file-exists-p (car files))))
2165 (setq files (cdr files)))
2166 (car files)))
2168 (let ((locale locale-name))
2170 (unless locale
2171 ;; Use the first of these three environment variables
2172 ;; that has a nonempty value.
2173 (let ((vars '("LC_ALL" "LC_CTYPE" "LANG")))
2174 (while (and vars (not (setq locale (getenv (car vars)))))
2175 (setq vars (cdr vars)))))
2177 (when locale
2179 ;; Translate "swedish" into "sv_SE.ISO8859-1", and so on,
2180 ;; using the translation file that many systems have.
2181 (when locale-translation-file-name
2182 (with-temp-buffer
2183 (insert-file-contents locale-translation-file-name)
2184 (when (re-search-forward
2186 (setq locale (buffer-substring (point) (line-end-position))))))
2188 ;; Leave the system locales alone if the caller did not specify
2189 ;; an explicit locale name, as their defaults are set from
2190 ;; LC_MESSAGES and LC_TIME, not LC_CTYPE, and the user might not
2191 ;; want to set them to the same value as LC_CTYPE.
2192 (when locale-name
2193 (setq system-messages-locale locale)
2194 (setq system-time-locale locale))
2196 (setq locale (downcase locale))
2198 (let ((language-name
2199 (locale-name-match locale locale-language-names))
2200 (charset-language-name
2201 (locale-name-match locale locale-charset-language-names))
2202 (coding-system
2203 (locale-name-match locale locale-preferred-coding-systems)))
2205 ;; Give preference to charset-language-name over language-name.
2206 (if (and charset-language-name
2207 (not
2208 (equal (get-language-info language-name 'charset)
2209 (get-language-info charset-language-name 'charset))))
2210 (setq language-name charset-language-name))
2212 (when language-name
2214 ;; Set up for this character set. This is now the right way
2215 ;; to do it for both unibyte and multibyte modes.
2216 (set-language-environment language-name)
2218 ;; If default-enable-multibyte-characters is nil,
2219 ;; we are using single-byte characters,
2220 ;; so the display table and terminal coding system are irrelevant.
2221 (when default-enable-multibyte-characters
2222 (set-display-table-and-terminal-coding-system language-name))
2224 ;; Set the `keyboard-coding-system' if appropriate.
2225 (let ((kcs (or coding-system
2226 (car (get-language-info language-name
2227 'coding-system)))))
2228 (if (memq kcs standard-keyboard-coding-systems)
2229 (set-keyboard-coding-system kcs)))
2231 (setq locale-coding-system
2232 (car (get-language-info language-name 'coding-priority))))
2234 (when coding-system
2235 (prefer-coding-system coding-system)
2236 (setq locale-coding-system coding-system))))))
2237 \f
2238 ;;; Charset property
2240 (defun get-charset-property (charset propname)
2241 "Return the value of CHARSET's PROPNAME property.
2242 This is the last value stored with
2243 (put-charset-property CHARSET PROPNAME VALUE)."
2244 (and (not (eq charset 'composition))
2245 (plist-get (charset-plist charset) propname)))
2247 (defun put-charset-property (charset propname value)
2248 "Store CHARSETS's PROPNAME property with value VALUE.
2249 It can be retrieved with `(get-charset-property CHARSET PROPNAME)'."
2250 (or (eq charset 'composition)
2251 (set-charset-plist charset
2252 (plist-put (charset-plist charset) propname value))))
2254 ;;; Character code property
2255 (put 'char-code-property-table 'char-table-extra-slots 0)
2257 (defvar char-code-property-table
2258 (make-char-table 'char-code-property-table)
2259 "Char-table containing a property list of each character code.
2261 See also the documentation of `get-char-code-property' and
2262 `put-char-code-property'.")
2264 (defun get-char-code-property (char propname)
2265 "Return the value of CHAR's PROPNAME property in `char-code-property-table'."
2266 (let ((plist (aref char-code-property-table char)))
2267 (if (listp plist)
2268 (car (cdr (memq propname plist))))))
2270 (defun put-char-code-property (char propname value)
2271 "Store CHAR's PROPNAME property with VALUE in `char-code-property-table'.
2272 It can be retrieved with `(get-char-code-property CHAR PROPNAME)'."
2273 (let ((plist (aref char-code-property-table char)))
2274 (if plist
2275 (let ((slot (memq propname plist)))
2276 (if slot
2277 (setcar (cdr slot) value)
2278 (nconc plist (list propname value))))
2279 (aset char-code-property-table char (list propname value)))))
2281 \f
2282 ;; Pretty description of encoded string
2284 ;; Alist of ISO 2022 control code vs the corresponding mnemonic string.
2285 (defvar iso-2022-control-alist
2293 (defun encoded-string-description (str coding-system)
2294 "Return a pretty description of STR that is encoded by CODING-SYSTEM."
2295 (setq str (string-as-unibyte str))
2296 (mapconcat
2297 (if (and coding-system (eq (coding-system-type coding-system) 2))
2298 ;; Try to get a pretty description for ISO 2022 escape sequences.
2299 (function (lambda (x) (or (cdr (assq x iso-2022-control-alist))
2300 (format "%02X" x))))
2301 (function (lambda (x) (format "0x%02X" x))))
2302 str " "))
2304 (defun encode-coding-char (char coding-system)
2305 "Encode CHAR by CODING-SYSTEM and return the resulting string.
2306 If CODING-SYSTEM can't safely encode CHAR, return nil."
2307 (let ((str1 (string-as-multibyte (char-to-string char)))
2308 (str2 (string-as-multibyte (make-string 2 char)))
2309 (safe-chars (and coding-system
2310 (coding-system-get coding-system 'safe-chars)))
2311 (charset (char-charset char))
2312 enc1 enc2 i1 i2)
2313 (when (or (eq safe-chars t)
2314 (eq charset 'ascii)
2315 (and safe-chars (aref safe-chars char)))
2316 ;; We must find the encoded string of CHAR. But, just encoding
2317 ;; CHAR will put extra control sequences (usually to designate
2318 ;; ASCII charaset) at the tail if type of CODING is ISO 2022.
2319 ;; To exclude such tailing bytes, we at first encode one-char
2320 ;; string and two-char string, then check how many bytes at the
2321 ;; tail of both encoded strings are the same.
2323 (setq enc1 (encode-coding-string str1 coding-system)
2324 i1 (length enc1)
2325 enc2 (encode-coding-string str2 coding-system)
2326 i2 (length enc2))
2327 (while (and (> i1 0) (= (aref enc1 (1- i1)) (aref enc2 (1- i2))))
2328 (setq i1 (1- i1) i2 (1- i2)))
2330 ;; Now (substring enc1 i1) and (substring enc2 i2) are the same,
2331 ;; and they are the extra control sequences at the tail to
2332 ;; exclude.
2333 (substring enc2 0 i2))))
2336 ;;; mule-cmds.el ends here