| blob | f393b0408aa12c01187d6b42f6929e99d6753107 |
1 ;;; mule-cmds.el --- commands for mulitilingual environment
2 ;; Copyright (C) 1995 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."
427 ;; Prefer coding systems corresponding to a
428 ;; MIME charset.
430 ;; Lower utf-16 priority so that we
431 ;; normally prefer utf-8 to it, and put
432 ;; x-ctext below that.
444 ;; For ISO based coding systems, prefer
445 ;; one that doesn't use escape sequences.
452 0
460 "Return a list of proper coding systems to encode a text between FROM and TO.
461 If FROM is a string, find coding systems in that instead of the buffer.
462 All coding systems in the list can safely encode any multibyte characters
463 in the text.
465 If the text contains no multibyte characters, return a list of a single
466 element `undecided'."
469 ;; The text contains only ASCII characters. Any coding
470 ;; systems are safe.
472 ;; We need copy-sequence because sorting will alter the argument.
476 "Return a list of proper coding systems to encode STRING.
477 All coding systems in the list can safely encode any multibyte characters
478 in STRING.
480 If STRING contains no multibyte characters, return a list of a single
481 element `undecided'."
485 "Return a list of proper coding systems to encode characters of CHARSETS.
486 CHARSETS is a list of character sets."
496 charset l ll)
513 "Find multibyte characters in the region specified by FROM and TO.
514 If FROM is a string, find multibyte characters in the string.
515 The return value is an alist of the following format:
516 ((CHARSET COUNT CHAR ...) ...)
517 where
518 CHARSET is a character set,
519 COUNT is a number of characters,
520 CHARs are found characters of the character set.
521 Optional 3rd arg MAXCOUNT limits how many CHARs are put in the above list.
522 Optional 4th arg EXCLUDE is a list of character sets to be ignored.
524 For invalid characters, CHARs are actually strings."
526 charset char)
535 eight-bit-control eight-bit-graphic))
567 "Search forward from point for a character that is not encodable.
568 It asks which coding system to check.
569 If such a character is found, set point after that character.
570 Otherwise, don't move point.
572 When called from a program, the value is a position of the found character,
573 or nil if all characters are encodable."
578 default))))
583 pos))
587 "Most recent coding system explicitly specified by the user when asked.
588 This variable is set whenever Emacs asks the user which coding system
589 to use in order to write a file. If you set it to nil explicitly,
590 then call `write-region', then afterward this variable will be non-nil
594 "If non-nil, a function to control the behaviour of coding system selection.
595 The meaning is the same as the argument ACCEPT-DEFAULT-P of the
596 function `select-safe-coding-system' (which see). This variable
600 accept-default-p file)
601 "Ask a user to select a safe coding system from candidates.
602 The candidates of coding systems which can safely encode a text
603 between FROM and TO are shown in a popup window. Among them, the most
604 proper one is suggested as the default.
606 The list of `buffer-file-coding-system' of the current buffer and the
607 most preferred coding system (if it corresponds to a MIME charset) is
608 treated as the default coding system list. Among them, the first one
609 that safely encodes the text is normally selected silently and
610 returned without any user interaction. See also the command
611 `prefer-coding-system'.
613 However, the user is queried if the chosen coding system is
614 inconsistent with what would be selected by `set-auto-coding' from
615 coding cookies &c. if the contents of the region were read from a
616 file. (That could lead to data corruption in a file subsequently
617 re-visited and edited.)
619 Optional 3rd arg DEFAULT-CODING-SYSTEM specifies a coding system or a
620 list of coding systems to be prepended to the default coding system
621 list. However, if DEFAULT-CODING-SYSTEM is a list and the first
622 element is t, the cdr part is used as the defualt coding system list,
623 i.e. `buffer-file-coding-system' and the most prepended coding system
624 is not used.
626 Optional 4th arg ACCEPT-DEFAULT-P, if non-nil, is a function to
627 determine the acceptability of the silently selected coding system.
628 It is called with that coding system, and should return nil if it
629 should not be silently selected and thus user interaction is required.
631 Optional 5th arg FILE is the file name to use for this purpose.
632 That is different from `buffer-file-name' when handling `write-region'
633 \(for example).
635 The variable `select-safe-coding-system-accept-default-p', if
636 non-nil, overrides ACCEPT-DEFAULT-P.
638 Kludgy feature: if FROM is a string, the string is the target text,
639 and TO is ignored."
649 ;; Change elements of the list to (coding . base-coding).
652 default-coding-system))
655 ;; If buffer-file-coding-system is not nil nor undecided, append it
656 ;; to the defaults.
665 ;; If the most preferred coding system has the property mime-charset,
666 ;; append it to the defaults.
668 preferred base)
686 safe rejected unsafe)
688 ;; Any coding system is ok.
690 ;; Classify the defaults into safe, rejected, and unsafe.
701 ;; If all the defaults failed, ask a user.
703 ;; At first, if some defaults are unsafe, record at most 11
704 ;; problematic characters and their positions for them by turning
705 ;; (CODING ...)
706 ;; into
707 ;; ((CODING (POS . CHAR) (POS . CHAR) ...) ...)
718 unsafe))
726 unsafe))))
728 ;; Change each safe coding system to the corresponding
729 ;; mime-charset name if it is also a coding system. Such a name
730 ;; is more friendly to users.
732 mime-charset)
739 ;; Don't offer variations with locking shift, which you
740 ;; basically never want.
749 ;; Remove raw-text, emacs-mule and no-conversion unless nothing
750 ;; else is available.
759 ;; If some defaults are unsafe, make sure the offending
760 ;; buffer is displayed.
764 unsafe))))
765 ;; Then ask users to select one from CODINGS while showing
766 ;; the reason why none of the defaults are not used.
776 "These default coding systems were tried to encode"
786 default-coding-system)
791 but it is not recommended for encoding text in this context,
825 'help-echo
826 "mouse-2, RET: jump to this character"
827 'help-function func1
830 "..."
832 'help-echo
833 "mouse-2, RET: next unencodable character"
834 'help-function func2
839 unsafe)
847 "\
852 codings)
858 ;; Read a coding system.
863 default-coding-system)
864 default-coding-system))
878 ;; Check we're not inconsistent with what `coding:' spec &c would
879 ;; give when file is re-read.
880 ;; But don't do this if we explicitly ignored the cookie
881 ;; by using `find-file-literally'.
891 ;; Don't barf if writing a compressed file, say.
892 ;; This check perhaps isn't ideal, but is probably
893 ;; the best thing to do.
899 %s specified by file contents. Really save (else edit coding cookies \
902 coding-system))
907 "Return a coding system to encode the outgoing message of the current buffer.
908 It at first tries the first coding system found in these variables
909 in this order:
910 (1) local value of `buffer-file-coding-system'
911 (2) value of `sendmail-coding-system'
912 (3) value of `default-sendmail-coding-system'
913 (4) value of `default-buffer-file-coding-system'
914 If the found coding system can't encode the current buffer,
915 or none of them are bound to a coding system,
916 it asks the user to select a proper coding system."
918 buffer-file-coding-system)
919 sendmail-coding-system
920 default-sendmail-coding-system
921 default-buffer-file-coding-system)))
923 ;; We should never use no-conversion for outgoing mail.
929 coding)))
930 \f
931 ;;; Language support stuff.
934 "Alist of language environment definitions.
935 Each element looks like:
936 (LANGUAGE-NAME . ((KEY . INFO) ...))
937 where LANGUAGE-NAME is a string, the name of the language environment,
938 KEY is a symbol denoting the kind of information, and
939 INFO is the data associated with KEY.
940 Meaningful values for KEY include
942 documentation value is documentation of what this language environment
943 is meant for, and how to use it.
944 charset value is a list of the character sets used by this
945 language environment.
946 sample-text value is an expression which is evalled to generate
947 a line of text written using characters appropriate
948 for this language environment.
949 setup-function value is a function to call to switch to this
950 language environment.
951 exit-function value is a function to call to leave this
952 language environment.
953 coding-system value is a list of coding systems that are good
954 for saving text written in this language environment.
955 This list serves as suggestions to the user;
956 in effect, as a kind of documentation.
957 coding-priority value is a list of coding systems for this language
958 environment, in order of decreasing priority.
959 This is used to set up the coding system priority
960 list when you switch to this language environment.
961 nonascii-translation
962 value is a translation table to be set in the
963 variable `nonascii-translation-table' in this
964 language environment, or a character set from
965 which `nonascii-insert-offset' is calculated.
966 input-method value is a default input method for this language
967 environment.
968 features value is a list of features requested in this
969 language environment.
971 The following keys take effect only when multibyte characters are
972 globally disabled, i.e. the value of `default-enable-multibyte-characters'
973 is nil.
975 unibyte-syntax value is a library name to load to set
976 unibyte 8-bit character syntaxes for this
977 language environment.
979 unibyte-display value is a coding system to encode characters
980 for the terminal. Characters in the range
981 of 160 to 255 display not as octal escapes,
982 but as non-ASCII characters in this language
986 "Return information listed under KEY for language environment LANG-ENV.
987 KEY is a symbol denoting the kind of information.
988 For a list of useful values for KEY and their meanings,
989 see `language-info-alist'."
997 "Modify part of the definition of language environment LANG-ENV.
998 Specifically, this stores the information INFO under KEY
999 in the definition of this language environment.
1000 KEY is a symbol denoting the kind of information.
1001 INFO is the value for that information.
1003 For a list of useful values for KEY and their meanings,
1004 see `language-info-alist'."
1018 ;; Update the custom-type of `current-language-environment'.
1026 "Store ALIST as the definition of language environment LANG-ENV.
1027 ALIST is an alist of KEY and INFO values. See the documentation of
1028 `language-info-alist' for the meanings of KEY and INFO.
1030 Optional arg PARENTS is a list of parent menu names; it specifies
1031 where to put this language environment in the
1032 Describe Language Environment and Set Language Environment menus.
1034 in the European submenu in each of those two menus."
1041 map parent-symbol parent prompt)
1047 ;; This prompt string is for define-prefix-command, so
1048 ;; that the map it creates will be suitable for a menu.
1069 ;; Set up menu items for this language env.
1082 "Read a language environment name which has information for KEY.
1083 If KEY is nil, read any language environment.
1084 Prompt with PROMPT. DEFAULT is the default choice of language environment.
1085 This returns a language environment name as a string."
1088 language-info-alist
1091 t nil nil default)))
1095 name)))
1096 \f
1097 ;;; Multilingual input methods.
1099 "LEIM: Libraries of Emacs Input Methods."
1103 "Name of LEIM list file.
1104 This file contains a list of libraries of Emacs input methods (LEIM)
1105 in the format of Lisp expression for registering each input method.
1109 ";;; %s -- list of LEIM (Library of Emacs Input Method) -*-coding: iso-2022-7bit;-*-
1110 ;;
1111 ;; This file contains a list of LEIM (Library of Emacs Input Method)
1112 ;; methods in the same directory as this file. Loading this file
1113 ;; registers all the input methods in Emacs.
1114 ;;
1115 ;; Each entry has the form:
1116 ;; (register-input-method
1117 ;; INPUT-METHOD LANGUAGE-NAME ACTIVATE-FUNC
1118 ;; TITLE DESCRIPTION
1119 ;; ARG ...)
1120 ;; See the function `register-input-method' for the meanings of the arguments.
1121 ;;
1122 ;; If this directory is included in load-path, Emacs automatically
1123 ;; loads this file at startup time.
1125 "
1126 leim-list-file-name)
1130 "Regexp matching head of each entry in LEIM list file.
1135 "List of functions to call to update LEIM list file.
1139 "Update LEIM list file in directories DIRS."
1146 "The current input method for multilingual text.
1157 "*Default input method for multilingual text (a string).
1158 This is the input method activated automatically by the command
1172 "Function to call for inactivating the current input method.
1173 Every input method should set this to an appropriate value when activated.
1174 This function is called with no argument.
1176 This function should never change the value of `current-input-method'.
1182 "Function to call for describing the current input method.
1188 "Alist of input method names vs how to use them.
1189 Each element has the form:
1190 (INPUT-METHOD LANGUAGE-ENV ACTIVATE-FUNC TITLE DESCRIPTION ARGS...)
1194 "Register INPUT-METHOD as an input method for language environment LANG-ENV.
1195 INPUT-METHOD and LANG-ENV are symbols or strings.
1197 The remaining arguments are:
1198 ACTIVATE-FUNC, TITLE, DESCRIPTION, and ARGS...
1199 ACTIVATE-FUNC is a function to call to activate this method.
1200 TITLE is a string to show in the mode line when this method is active.
1201 DESCRIPTION is a string describing this method and what it is good for.
1202 The ARGS, if any, are passed as arguments to ACTIVATE-FUNC.
1203 All told, the arguments to ACTIVATE-FUNC are INPUT-METHOD and the ARGS.
1206 created at Emacs build time, registering all Quail input methods
1207 contained in the Emacs distribution.
1209 In case you want to register a new Quail input method by yourself, be
1210 careful to use the same input method title as given in the third
1211 parameter of `quail-define-package'. (If the values are different, the
1212 string specified in this function takes precedence.)
1214 The commands `describe-input-method' and `list-input-methods' need
1215 these duplicated values to show some information about input methods
1216 without loading the relevant Quail packages."
1229 "Read a name of input method from a minibuffer prompting with PROMPT.
1230 If DEFAULT is non-nil, use that as the default,
1231 and substitute it into PROMPT at the first `%s'.
1232 If INHIBIT-NULL is non-nil, null input signals an error.
1234 The return value is a string."
1238 ;; As it is quite normal to change input method in the
1239 ;; minibuffer, we must enable it even if
1240 ;; enable-recursive-minibuffers is currently nil.
1242 ;; This binding is necessary because input-method-history is
1243 ;; buffer local.
1245 nil t nil 'input-method-history
1246 default)))
1250 input-method
1255 "Switch to input method INPUT-METHOD for the current buffer.
1256 If some other input method is already active, turn it off first.
1257 If INPUT-METHOD is nil, deactivate any current input method."
1284 "Turn off the current input method."
1297 input-method-function nil
1298 current-input-method-title nil)
1302 "Select and activate input method INPUT-METHOD for the current buffer.
1303 This also sets the default input method to the one you specify.
1304 If INPUT-METHOD is nil, this function turns off the input method, and
1305 also causes you to be prompted for a name of an input method the next
1314 default t))))
1319 default-input-method)
1322 "Enable or disable multilingual text input method for the current buffer.
1323 Only one input method can be enabled at any time in a given buffer.
1325 The normal action is to enable an input method if none was
1326 enabled, and disable the current one otherwise. Which input method
1327 to enable can be determined in various ways--either the one most
1328 recently used, or the one specified by `default-input-method', or
1329 as a last resort by reading the name of an input method in the
1330 minibuffer.
1332 With a prefix argument, read an input method name with the minibuffer
1333 and enable that one. The default is the most recent input method specified
1334 \(not including the currently active input method, if any)."
1348 default t))
1349 default))
1357 "Describe input method INPUT-METHOD."
1387 "Describe the input method currently in use.
1388 This is a subroutine for `describe-input-method'."
1394 current-input-method)
1399 "Read a multilingual string from minibuffer, prompting with string PROMPT.
1400 The input method selected last time is activated in minibuffer.
1401 If optional second arg INITIAL-INPUT is non-nil, insert it in the minibuffer
1402 initially.
1403 Optional 3rd argument INPUT-METHOD specifies the input method
1404 to be activated instead of the one selected last time. It is a symbol
1405 or a string."
1408 current-input-method
1409 default-input-method
1420 ;; Variables to control behavior of input methods. All input methods
1421 ;; should react to these variables.
1424 "*A flag to control extra guidance given by input methods.
1425 The value should be nil, t, `complex-only', or `default'.
1427 The extra guidance is done by showing list of available keys in echo
1428 area. When you use the input method in the minibuffer, the guidance
1429 is shown at the bottom short window (split from the existing window).
1431 If the value is t, extra guidance is always given, if the value is
1432 nil, extra guidance is always suppressed.
1434 If the value is `complex-only', only complex input methods such as
1435 `chinese-py' and `japanese' give extra guidance.
1437 If the value is `default', complex input methods always give extra
1438 guidance, but simple input methods give it only when you are not in
1439 the minibuffer.
1441 See also the variable `input-method-highlight-flag'."
1447 "*If this flag is non-nil, input methods highlight partially-entered text.
1448 For instance, while you are in the middle of a Quail input method sequence,
1449 the text inserted so far is temporarily underlined.
1450 The underlining goes away when you finish or abort the input method sequence.
1451 See also the variable `input-method-verbose-flag'."
1456 "Normal hook run just after an input method is activated.
1458 The variable `current-input-method' keeps the input method name
1462 "Normal hook run just after an input method is inactivated.
1464 The variable `current-input-method' still keeps the input method name
1471 "This flag controls when an input method returns.
1472 Usually, the input method does not return while there's a possibility
1473 that it may find a different translation if a user types another key.
1474 But, it this flag is non-nil, the input method returns as soon as
1478 "This flag controls how an input method shows an intermediate key sequence.
1479 Usually, the input method inserts the intermediate key sequence,
1480 or candidate translations corresponding to the sequence,
1481 at point in the current buffer.
1485 "This flag controls the behaviour of an input method on invalid key input.
1486 Usually, when a user types a key which doesn't start any character
1487 handled by the input method, the key is handled by turning off the
1488 input method temporarily. After that key, the input method is re-enabled.
1491 \f
1493 "Normal hook run after some language environment is set.
1495 When you set some hook function here, that effect usually should not
1496 be inherited to another language environment. So, you had better set
1497 another function in `exit-language-environment-hook' (which see) to
1501 "Normal hook run after exiting from some language environment.
1502 When this hook is run, the variable `current-language-environment'
1503 is still bound to the language environment being exited.
1505 This hook is mainly used for canceling the effect of
1511 "Switch to a specified language environment."
1524 "The last language environment specified with `set-language-environment'.
1526 to using the function `set-language-environment'."
1533 current-language-environment)
1534 language-info-alist))
1536 ;; custom type will be updated with `set-language-info'.
1547 "Reset multilingual environment of Emacs to the default status.
1549 The default status is as follows:
1551 The default value of `buffer-file-coding-system' is nil.
1552 The default coding system for process I/O is nil.
1553 The default value for the command `set-terminal-coding-system' is nil.
1554 The default value for the command `set-keyboard-coding-system' is nil.
1556 The order of priorities of coding categories and the coding system
1557 bound to each category are as follows
1558 coding category coding system
1559 --------------------------------------------------
1560 coding-category-iso-8-2 iso-latin-1
1561 coding-category-iso-8-1 iso-latin-1
1562 coding-category-iso-7-tight iso-2022-jp
1563 coding-category-iso-7 iso-2022-7bit
1564 coding-category-iso-7-else iso-2022-7bit-lock
1565 coding-category-iso-8-else iso-2022-8bit-ss2
1566 coding-category-emacs-mule emacs-mule
1567 coding-category-raw-text raw-text
1568 coding-category-sjis japanese-shift-jis
1569 coding-category-big5 chinese-big5
1570 coding-category-ccl nil
1571 coding-category-binary no-conversion
1572 coding-category-utf-16-be nil
1573 coding-category-utf-16-le nil
1574 coding-category-utf-8 mule-utf-8"
1576 ;; This function formerly set default-enable-multibyte-characters to t,
1577 ;; but that is incorrect. It should not alter the unibyte/multibyte choice.
1580 coding-category-iso-7 'iso-2022-7bit
1581 coding-category-iso-8-1 'iso-latin-1
1582 coding-category-iso-8-2 'iso-latin-1
1583 coding-category-iso-7-else 'iso-2022-7bit-lock
1584 coding-category-iso-8-else 'iso-2022-8bit-ss2
1585 coding-category-emacs-mule 'emacs-mule
1586 coding-category-raw-text 'raw-text
1587 coding-category-sjis 'japanese-shift-jis
1588 coding-category-big5 'chinese-big5
1589 coding-category-utf-16-be nil
1590 coding-category-utf-16-le nil
1591 coding-category-utf-8 'mule-utf-8
1592 coding-category-ccl nil
1597 coding-category-iso-8-2
1598 coding-category-iso-7-tight
1599 coding-category-iso-7
1600 coding-category-iso-7-else
1601 coding-category-iso-8-else
1602 coding-category-emacs-mule
1603 coding-category-raw-text
1604 coding-category-sjis
1605 coding-category-big5
1606 coding-category-ccl
1607 coding-category-binary
1608 coding-category-utf-16-be
1609 coding-category-utf-16-le
1610 coding-category-utf-8))
1616 ;; Preserve eol-type from existing default-process-coding-systems.
1617 ;; On non-unix-like systems in particular, these may have been set
1618 ;; carefully by the user, or by the startup code, to deal with the
1619 ;; users shell appropriately, so should not be altered by changing
1620 ;; language environment.
1622 ;; When bootstrapping, coding-systems are not defined yet, so
1623 ;; we need to catch the error from check-coding-system.
1636 ;; Don't alter the terminal and keyboard coding systems here.
1637 ;; The terminal still supports the same coding system
1638 ;; that it supported a minute ago.
1639 ;; (set-terminal-coding-system-internal nil)
1640 ;; (set-keyboard-coding-system-internal nil)
1648 "Set up the display table and terminal coding system for LANGUAGE-NAME."
1658 "Set up multi-lingual environment for using LANGUAGE-NAME.
1659 This sets the coding system priority and the default input method
1660 and sometimes other things. LANGUAGE-NAME should be a string
1662 specifies the character set for the major languages of Western Europe."
1664 nil
1678 default-buffer-file-coding-system)))
1681 ;; The features might set up coding systems.
1705 ;; DOS terminals' default is to use a special non-ASCII translation
1706 ;; table as appropriate for the installed codepage.
1711 ;; Unibyte setups if necessary.
1713 ;; Syntax and case table.
1718 ;; No information for syntax and case. Reset to the defaults.
1748 ;; Actually set up direct output of non-ASCII characters.
1750 ;; Unibyte Emacs on MS-DOS wants to display all 8-bit characters with
1751 ;; the native font, and codes 160 and 146 stand for something very
1752 ;; different there.
1755 ;; Make non-line-break space display as a plain space.
1756 ;; Most X fonts do the wrong thing for code 160.
1758 ;; With luck, non-Latin-1 fonts are more recent and so don't
1759 ;; have this bug.
1761 ;; Most Windows programs send out apostrophes as \222. Most X fonts
1762 ;; don't contain a character at that position. Map it to the ASCII
1763 ;; apostrophe. [This is actually RIGHT SINGLE QUOTATION MARK,
1764 ;; U+2019, normally from the windows-1252 character set. XFree 4
1765 ;; fonts probably have the appropriate glyph at this position,
1766 ;; so they could use standard-display-8bit. It's better to use a
1767 ;; proper windows-1252 coding system. --fx]
1770 (defun set-language-environment-coding-systems (language-name
1771 &optional eol-type)
1772 "Do various coding system setups for language environment LANGUAGE-NAME.
1774 The optional arg EOL-TYPE specifies the eol-type of the default value
1775 of `buffer-file-coding-system' set by this function."
1776 (let* ((priority (get-language-info language-name 'coding-priority))
1777 (default-coding (car priority)))
1778 (if priority
1779 (let ((categories (mapcar 'coding-system-category priority)))
1780 (set-default-coding-systems
1781 (if (memq eol-type '(0 1 2 unix dos mac))
1782 (coding-system-change-eol-conversion default-coding eol-type)
1783 default-coding))
1784 (setq default-sendmail-coding-system default-coding)
1785 (set-coding-priority categories)
1786 (while priority
1787 (set (car categories) (car priority))
1788 (setq priority (cdr priority) categories (cdr categories)))
1789 (update-coding-systems-internal)))))
1791 (defsubst princ-list (&rest args)
1793 (while args (princ (car args)) (setq args (cdr args)))
1796 (put 'describe-specified-language-support 'apropos-inhibit t)
1798 ;; Print a language specific information such as input methods,
1799 ;; charsets, and coding systems. This function is intended to be
1800 ;; called from the menu:
1801 ;; [menu-bar mule describe-language-environment LANGUAGE]
1802 ;; and should not run it by `M-x describe-current-input-method-function'.
1803 (defun describe-specified-language-support ()
1804 "Describe how Emacs supports the specified language environment."
1805 (interactive)
1806 (let (language-name)
1807 (if (not (and (symbolp last-command-event)
1808 (or (not (eq last-command-event 'Default))
1809 (setq last-command-event 'English))
1810 (setq language-name (symbol-name last-command-event))))
1811 (error "Bogus calling sequence"))
1812 (describe-language-environment language-name)))
1814 (defun describe-language-environment (language-name)
1815 "Describe how Emacs supports language environment LANGUAGE-NAME."
1816 (interactive
1817 (list (read-language-name
1818 'documentation
1819 "Describe language environment (default, current choice): ")))
1820 (if (null language-name)
1821 (setq language-name current-language-environment))
1822 (if (or (null language-name)
1823 (null (get-language-info language-name 'documentation)))
1824 (error "No documentation for the specified language"))
1825 (if (symbolp language-name)
1826 (setq language-name (symbol-name language-name)))
1827 (let ((doc (get-language-info language-name 'documentation))
1828 pos)
1829 (help-setup-xref (list #'describe-language-environment language-name)
1830 (interactive-p))
1831 (with-output-to-temp-buffer (help-buffer)
1832 (save-excursion
1833 (set-buffer standard-output)
1835 (if (stringp doc)
1837 (condition-case nil
1838 (let ((str (eval (get-language-info language-name 'sample-text))))
1839 (if (stringp str)
1841 (error nil))
1842 (let ((input-method (get-language-info language-name 'input-method))
1843 (l (copy-sequence input-method-alist)))
1844 (insert "Input methods")
1845 (when input-method
1846 (insert " (default, " input-method ")")
1847 (setq input-method (assoc input-method input-method-alist))
1848 (setq l (cons input-method (delete input-method l))))
1850 (while l
1851 (when (string= language-name (nth 1 (car l)))
1852 (insert " " (car (car l)))
1853 (search-backward (car (car l)))
1854 (help-xref-button 0 'help-input-method (car (car l)))
1855 (goto-char (point-max))
1857 (if (stringp (nth 3 (car l)))
1858 (nth 3 (car l))
1859 (car (nth 3 (car l))))
1861 (setq l (cdr l)))
1864 (let ((l (get-language-info language-name 'charset)))
1865 (if (null l)
1867 (while l
1868 (insert " " (symbol-name (car l)))
1869 (search-backward (symbol-name (car l)))
1870 (help-xref-button 0 'help-character-set (car l))
1871 (goto-char (point-max))
1873 (setq l (cdr l)))))
1876 (let ((l (get-language-info language-name 'coding-system)))
1877 (if (null l)
1879 (while l
1880 (insert " " (symbol-name (car l)))
1881 (search-backward (symbol-name (car l)))
1882 (help-xref-button 0 'help-coding-system (car l))
1883 (goto-char (point-max))
1884 (insert " (`"
1885 (coding-system-mnemonic (car l))
1887 (coding-system-doc-string (car l))
1889 (let ((aliases (coding-system-get (car l)
1890 'alias-coding-systems)))
1891 (when aliases
1893 (while aliases
1894 (insert " " (symbol-name (car aliases)))
1895 (setq aliases (cdr aliases)))
1897 (setq l (cdr l)))))))))
1898 \f
1899 ;;; Locales.
1901 (defvar locale-translation-file-name nil
1902 "File name for the system's file of locale-name aliases, or nil if none.")
1904 ;; The following definitions might as well be marked as constants and
1905 ;; purecopied, since they're normally used on startup, and probably
1906 ;; should reflect the facilities of the base Emacs.
1907 (defconst locale-language-names
1908 (purecopy
1909 '(
1910 ;; Locale names of the form LANGUAGE[_TERRITORY][.CODESET][@MODIFIER]
1911 ;; as specified in the Single Unix Spec, Version 2.
1912 ;; LANGUAGE is a language code taken from ISO 639:1988 (E/F)
1913 ;; with additions from ISO 639/RA Newsletter No.1/1989;
1914 ;; see Internet RFC 2165 (1997-06) and
1915 ;; http://www.evertype.com/standards/iso639/iso639-en.html
1916 ;; TERRITORY is a country code taken from ISO 3166
1917 ;; http://www.din.de/gremien/nas/nabd/iso3166ma/codlstp1/en_listp1.html.
1918 ;; CODESET and MODIFIER are implementation-dependent.
1920 ; aa Afar
1921 ; ab Abkhazian
1922 ("af" . "Latin-1") ; Afrikaans
1923 ("am" . "Ethiopic") ; Amharic
1924 ; ar Arabic glibc uses 8859-6
1925 ; as Assamese
1926 ; ay Aymara
1927 ; az Azerbaijani
1928 ; ba Bashkir
1929 ("be" . "Belarusian") ; Belarusian [Byelorussian until early 1990s]
1930 ("bg" . "Bulgarian") ; Bulgarian
1931 ; bh Bihari
1932 ; bi Bislama
1933 ; bn Bengali, Bangla
1934 ("bo" . "Tibetan")
1935 ("br" . "Latin-1") ; Breton
1936 ("bs" . "Latin-2") ; Bosnian
1937 ("ca" . "Latin-1") ; Catalan
1938 ; co Corsican
1939 ("cs" . "Czech")
1940 ("cy" . "Welsh") ; Welsh [glibc uses Latin-8. Did this change?]
1941 ("da" . "Latin-1") ; Danish
1942 ("de" . "German")
1943 ; dz Bhutani
1944 ("el" . "Greek")
1945 ;; Users who specify "en" explicitly typically want Latin-1, not ASCII.
1946 ;; That's actually what the GNU locales define, modulo things like
1947 ;; en_IN -- fx.
1948 ("en" . "Latin-1") ; English
1949 ("eo" . "Latin-3") ; Esperanto
1950 ("es" . "Spanish")
1951 ("et" . "Latin-4") ; Estonian
1952 ("eu" . "Latin-1") ; Basque
1953 ; fa Persian glibc uses utf-8
1954 ("fi" . "Latin-1") ; Finnish
1955 ; fj Fiji
1956 ("fo" . "Latin-1") ; Faroese
1957 ("fr" . "French") ; French
1958 ("fy" . "Latin-1") ; Frisian
1959 ("ga" . "Latin-1") ; Irish Gaelic (new orthography)
1960 ("gd" . "Latin-1") ; Scots Gaelic
1961 ("gl" . "Latin-1") ; Galician
1962 ; gn Guarani
1963 ; gu Gujarati
1964 ("gv" . "Latin-8") ; Manx Gaelic glibc uses 8859-1
1965 ; ha Hausa
1966 ("he" . "Hebrew")
1967 ("hi" . "Devanagari") ; Hindi glibc uses utf-8
1968 ("hr" . "Croatian") ; Croatian
1969 ("hu" . "Latin-2") ; Hungarian
1970 ; hy Armenian
1971 ; ia Interlingua
1972 ("id" . "Latin-1") ; Indonesian
1973 ; ie Interlingue
1974 ; ik Inupiak
1975 ("is" . "Latin-1") ; Icelandic
1976 ("it" . "Italian") ; Italian
1977 ; iu Inuktitut
1978 ("ja" . "Japanese")
1979 ; jw Javanese
1980 ("ka" . "Georgian") ; Georgian
1981 ; kk Kazakh
1982 ("kl" . "Latin-1") ; Greenlandic
1983 ; km Cambodian
1984 ; kn Kannada
1985 ("ko" . "Korean")
1986 ; ks Kashmiri
1987 ; ku Kurdish
1988 ("kw" . "Latin-1") ; Cornish
1989 ; ky Kirghiz
1990 ("la" . "Latin-1") ; Latin
1991 ("lb" . "Latin-1") ; Luxemburgish
1992 ; ln Lingala
1993 ("lo" . "Lao") ; Laothian
1994 ("lt" . "Lithuanian")
1995 ("lv" . "Latvian") ; Latvian, Lettish
1996 ; mg Malagasy
1997 ("mi" . "Latin-7") ; Maori
1998 ("mk" . "Cyrillic-ISO") ; Macedonian
1999 ; ml Malayalam
2000 ; mn Mongolian
2001 ; mo Moldavian
2002 ("mr" . "Devanagari") ; Marathi glibc uses utf-8
2003 ("ms" . "Latin-1") ; Malay
2004 ("mt" . "Latin-3") ; Maltese
2005 ; my Burmese
2006 ; na Nauru
2007 ("ne" . "Devanagari") ; Nepali
2008 ("nl" . "Dutch")
2009 ("no" . "Latin-1") ; Norwegian
2010 ("oc" . "Latin-1") ; Occitan
2011 ; om (Afan) Oromo
2012 ; or Oriya
2013 ; pa Punjabi
2014 ("pl" . "Latin-2") ; Polish
2015 ; ps Pashto, Pushto
2016 ("pt" . "Latin-1") ; Portuguese
2017 ; qu Quechua
2018 ("rm" . "Latin-1") ; Rhaeto-Romanic
2019 ; rn Kirundi
2020 ("ro" . "Romanian")
2021 ("ru.*[_.]koi8" . "Russian")
2022 ("ru" . "Cyrillic-ISO") ; Russian
2023 ; rw Kinyarwanda
2024 ("sa" . "Devanagari") ; Sanskrit
2025 ; sd Sindhi
2026 ; se Northern Sami
2027 ; sg Sangho
2028 ("sh" . "Latin-2") ; Serbo-Croatian
2029 ; si Sinhalese
2030 ("sk" . "Slovak")
2031 ("sl" . "Slovenian")
2032 ; sm Samoan
2033 ; sn Shona
2034 ; so Somali
2035 ("sq" . "Latin-1") ; Albanian
2036 ("sr" . "Latin-2") ; Serbian (Latin alphabet)
2037 ("sr_YU@cyrillic" . "Cyrillic-ISO") ; per glibc
2038 ; ss Siswati
2039 ; st Sesotho
2040 ; su Sundanese
2041 ("sv" . "Swedish") ; Swedish
2042 ("sw" . "Latin-1") ; Swahili
2043 ; ta Tamil glibc uses utf-8
2044 ; te Telugu glibc uses utf-8
2045 ("tg" . "Tajik")
2046 ("th" . "Thai")
2047 ; ti Tigrinya
2048 ; tk Turkmen
2049 ("tl" . "Latin-1") ; Tagalog
2050 ; tn Setswana
2051 ; to Tonga
2052 ("tr" . "Turkish")
2053 ; ts Tsonga
2054 ; tt Tatar
2055 ; tw Twi
2056 ; ug Uighur
2057 ("uk" . "Ukrainian") ; Ukrainian
2058 ; ur Urdu glibc uses utf-8
2059 ("uz" . "Latin-1") ; Uzbek
2060 ("vi" . "Vietnamese") ; glibc uses utf-8
2061 ; vo Volapuk
2062 ("wa" . "Latin-1") ; Walloon
2063 ; wo Wolof
2064 ; xh Xhosa
2065 ("yi" . "Windows-1255") ; Yiddish
2066 ; yo Yoruba
2067 ; za Zhuang
2069 ; glibc:
2070 ; zh_CN.GB18030/GB18030 \
2071 ; zh_CN.GBK/GBK \
2072 ; zh_HK/BIG5-HKSCS \
2074 ("zh.*[._]big5" . "Chinese-BIG5")
2075 ("zh.*[._]gbk" . nil) ; Solaris 2.7; has gbk-0 as well as GB 2312.1980-0
2076 ("zh_tw" . "Chinese-CNS") ; glibc uses big5
2077 ("zh_tw[._]euc-tw" . "Chinese-EUC-TW")
2078 ("zh" . "Chinese-GB")
2079 ; zu Zulu
2081 ;; ISO standard locales
2082 ("c$" . "ASCII")
2083 ("posix$" . "ASCII")
2085 ;; The "IPA" Emacs language environment does not correspond
2086 ;; to any ISO 639 code, so let it stand for itself.
2087 ("ipa$" . "IPA")
2089 ;; Nonstandard or obsolete language codes
2090 ("cz" . "Czech") ; e.g. Solaris 2.6
2091 ("ee" . "Latin-4") ; Estonian, e.g. X11R6.4
2092 ("iw" . "Hebrew") ; e.g. X11R6.4
2093 ("sp" . "Cyrillic-ISO") ; Serbian (Cyrillic alphabet), e.g. X11R6.4
2094 ("su" . "Latin-1") ; Finnish, e.g. Solaris 2.6
2095 ("jp" . "Japanese") ; e.g. MS Windows
2096 ("chs" . "Chinese-GB") ; MS Windows Chinese Simplified
2097 ("cht" . "Chinese-BIG5") ; MS Windows Chinese Traditional
2098 ))
2099 "List of pairs of locale regexps and language names.
2100 The first element whose locale regexp matches the start of a downcased locale
2101 specifies the language name corresponding to that locale.
2102 If the language name is nil, there is no corresponding language environment.")
2104 (defconst locale-charset-language-names
2105 (purecopy
2114 ;; utf-8@euro exists, so put this last. (@euro really specifies
2115 ;; the currency, rather than the charset.)
2117 "List of pairs of locale regexps and charset language names.
2118 The first element whose locale regexp matches the start of a downcased locale
2119 specifies the language name whose charsets corresponds to that locale.
2120 This language name is used if its charsets disagree with the charsets of
2121 the language name that would otherwise be used for this locale.")
2123 (defconst locale-preferred-coding-systems
2124 (purecopy
2125 '(("ja.*[._]euc" . japanese-iso-8bit)
2126 ("ja.*[._]jis7" . iso-2022-jp)
2127 ("ja.*[._]pck" . japanese-shift-jis)
2128 ("ja.*[._]sjis" . japanese-shift-jis)
2129 ("jpn" . japanese-shift-jis) ; MS-Windows uses this.
2130 (".*[._]utf" . utf-8)))
2131 "List of pairs of locale regexps and preferred coding systems.
2132 The first element whose locale regexp matches the start of a downcased locale
2133 specifies the coding system to prefer when using that locale.")
2135 (defun locale-name-match (key alist)
2136 "Search for KEY in ALIST, which should be a list of regexp-value pairs.
2137 Return the value corresponding to the first regexp that matches the
2138 start of KEY, or nil if there is no match."
2139 (let (element)
2140 (while (and alist (not element))
2142 (setq element (car alist)))
2143 (setq alist (cdr alist)))
2144 (cdr element)))
2146 (defun set-locale-environment (&optional locale-name)
2147 "Set up multi-lingual environment for using LOCALE-NAME.
2148 This sets the language environment, the coding system priority,
2149 the default input method and sometimes other things.
2151 LOCALE-NAME should be a string
2152 which is the name of a locale supported by the system;
2153 often it is of the form xx_XX.CODE, where xx is a language,
2154 XX is a country, and CODE specifies a character set and coding system.
2156 for Japanese in Japan using the `japanese-iso-8bit' coding-system.
2158 If LOCALE-NAME is nil, its value is taken from the environment
2159 variables LC_ALL, LC_CTYPE and LANG (the first one that is set).
2161 The locale names supported by your system can typically be found in a
2162 directory named `/usr/share/locale' or `/usr/lib/locale'. LOCALE-NAME
2163 will be translated according to the table specified by
2164 `locale-translation-file-name'.
2166 See also `locale-charset-language-names', `locale-language-names',
2167 `locale-preferred-coding-systems' and `locale-coding-system'."
2168 (interactive "sSet environment for locale: ")
2170 ;; Do this at runtime for the sake of binaries possibly transported
2171 ;; to a system without X.
2172 (setq locale-translation-file-name
2173 (let ((files
2174 '("/usr/lib/X11/locale/locale.alias" ; e.g. X11R6.4
2175 "/usr/X11R6/lib/X11/locale/locale.alias" ; e.g. RedHat 4.2
2176 "/usr/openwin/lib/locale/locale.alias" ; e.g. Solaris 2.6
2177 ;;
2178 ;; The following name appears after the X-related names above,
2179 ;; since the X-related names are what X actually uses.
2180 "/usr/share/locale/locale.alias" ; GNU/Linux sans X
2181 )))
2182 (while (and files (not (file-exists-p (car files))))
2183 (setq files (cdr files)))
2184 (car files)))
2186 (let ((locale locale-name))
2188 (unless locale
2189 ;; Use the first of these three environment variables
2190 ;; that has a nonempty value.
2191 (let ((vars '("LC_ALL" "LC_CTYPE" "LANG")))
2192 (while (and vars
2193 (= 0 (length locale))) ; nil or empty string
2194 (setq locale (getenv (pop vars))))))
2196 (when locale
2198 ;; Translate "swedish" into "sv_SE.ISO8859-1", and so on,
2199 ;; using the translation file that many systems have.
2200 (when locale-translation-file-name
2201 (with-temp-buffer
2202 (insert-file-contents locale-translation-file-name)
2203 (when (re-search-forward
2205 (setq locale (buffer-substring (point) (line-end-position))))))
2207 ;; Leave the system locales alone if the caller did not specify
2208 ;; an explicit locale name, as their defaults are set from
2209 ;; LC_MESSAGES and LC_TIME, not LC_CTYPE, and the user might not
2210 ;; want to set them to the same value as LC_CTYPE.
2211 (when locale-name
2212 (setq system-messages-locale locale)
2213 (setq system-time-locale locale))
2215 (setq locale (downcase locale))
2217 (let ((language-name
2218 (locale-name-match locale locale-language-names))
2219 (charset-language-name
2220 (locale-name-match locale locale-charset-language-names))
2221 (coding-system
2222 (locale-name-match locale locale-preferred-coding-systems)))
2224 ;; Give preference to charset-language-name over language-name.
2225 (if (and charset-language-name
2226 (not
2227 (equal (get-language-info language-name 'charset)
2228 (get-language-info charset-language-name 'charset))))
2229 (setq language-name charset-language-name))
2231 (when language-name
2233 ;; Set up for this character set. This is now the right way
2234 ;; to do it for both unibyte and multibyte modes.
2235 (set-language-environment language-name)
2237 ;; If default-enable-multibyte-characters is nil,
2238 ;; we are using single-byte characters,
2239 ;; so the display table and terminal coding system are irrelevant.
2240 (when default-enable-multibyte-characters
2241 (set-display-table-and-terminal-coding-system language-name))
2243 ;; Set the `keyboard-coding-system' if appropriate (tty
2244 ;; only). At least X and MS Windows can generate
2245 ;; multilingual input.
2246 (unless window-system
2247 (let ((kcs (or coding-system
2248 (car (get-language-info language-name
2249 'coding-system)))))
2250 (if kcs (set-keyboard-coding-system kcs))))
2252 (setq locale-coding-system
2253 (car (get-language-info language-name 'coding-priority))))
2255 (when coding-system
2256 (prefer-coding-system coding-system)
2257 (setq locale-coding-system coding-system))))
2259 ;; Default to A4 paper if we're not in a C, POSIX or US locale.
2260 ;; (See comments in Flanginfo.)
2261 (let ((locale locale)
2262 (paper (langinfo 'paper)))
2263 (if paper
2264 ;; This will always be null at the time of writing.
2265 (cond
2266 ((equal paper '(216 279))
2267 (setq ps-paper-type 'letter))
2268 ((equal paper '(210 297))
2269 (setq ps-paper-type 'a4)))
2270 (let ((vars '("LC_ALL" "LC_PAPER" "LANG")))
2271 (while (and vars (= 0 (length locale)))
2272 (setq locale (getenv (pop vars)))))
2273 (when locale
2274 ;; As of glibc 2.2.5, these are the only US Letter locales,
2275 ;; and the rest are A4.
2276 (setq ps-paper-type
2277 (or (locale-name-match locale '(("c$" . letter)
2278 ("posix$" . letter)
2279 (".._us" . letter)
2280 (".._pr" . letter)
2281 (".._ca" . letter)))
2282 'a4))))))
2283 nil)
2284 \f
2285 ;;; Charset property
2287 (defun get-charset-property (charset propname)
2288 "Return the value of CHARSET's PROPNAME property.
2289 This is the last value stored with
2290 (put-charset-property CHARSET PROPNAME VALUE)."
2291 (and (not (eq charset 'composition))
2292 (plist-get (charset-plist charset) propname)))
2294 (defun put-charset-property (charset propname value)
2295 "Store CHARSETS's PROPNAME property with value VALUE.
2296 It can be retrieved with `(get-charset-property CHARSET PROPNAME)'."
2297 (or (eq charset 'composition)
2298 (set-charset-plist charset
2299 (plist-put (charset-plist charset) propname value))))
2301 ;;; Character code property
2302 (put 'char-code-property-table 'char-table-extra-slots 0)
2304 (defvar char-code-property-table
2305 (make-char-table 'char-code-property-table)
2306 "Char-table containing a property list of each character code.
2308 See also the documentation of `get-char-code-property' and
2309 `put-char-code-property'.")
2311 (defun get-char-code-property (char propname)
2312 "Return the value of CHAR's PROPNAME property in `char-code-property-table'."
2313 (let ((plist (aref char-code-property-table char)))
2314 (if (listp plist)
2315 (car (cdr (memq propname plist))))))
2317 (defun put-char-code-property (char propname value)
2318 "Store CHAR's PROPNAME property with VALUE in `char-code-property-table'.
2319 It can be retrieved with `(get-char-code-property CHAR PROPNAME)'."
2320 (let ((plist (aref char-code-property-table char)))
2321 (if plist
2322 (let ((slot (memq propname plist)))
2323 (if slot
2324 (setcar (cdr slot) value)
2325 (nconc plist (list propname value))))
2326 (aset char-code-property-table char (list propname value)))))
2328 \f
2329 ;; Pretty description of encoded string
2331 ;; Alist of ISO 2022 control code vs the corresponding mnemonic string.
2332 (defvar iso-2022-control-alist
2340 (defun encoded-string-description (str coding-system)
2341 "Return a pretty description of STR that is encoded by CODING-SYSTEM."
2342 (setq str (string-as-unibyte str))
2343 (mapconcat
2344 (if (and coding-system (eq (coding-system-type coding-system) 2))
2345 ;; Try to get a pretty description for ISO 2022 escape sequences.
2346 (function (lambda (x) (or (cdr (assq x iso-2022-control-alist))
2347 (format "%02X" x))))
2348 (function (lambda (x) (format "0x%02X" x))))
2349 str " "))
2351 (defun encode-coding-char (char coding-system)
2352 "Encode CHAR by CODING-SYSTEM and return the resulting string.
2353 If CODING-SYSTEM can't safely encode CHAR, return nil."
2354 (let ((str1 (string-as-multibyte (char-to-string char)))
2355 (str2 (string-as-multibyte (make-string 2 char)))
2356 (safe-chars (and coding-system
2357 (coding-system-get coding-system 'safe-chars)))
2358 (charset (char-charset char))
2359 enc1 enc2 i1 i2)
2360 (when (or (eq safe-chars t)
2361 (eq charset 'ascii)
2362 (and safe-chars (aref safe-chars char)))
2363 ;; We must find the encoded string of CHAR. But, just encoding
2364 ;; CHAR will put extra control sequences (usually to designate
2365 ;; ASCII charaset) at the tail if type of CODING is ISO 2022.
2366 ;; To exclude such tailing bytes, we at first encode one-char
2367 ;; string and two-char string, then check how many bytes at the
2368 ;; tail of both encoded strings are the same.
2370 (setq enc1 (encode-coding-string str1 coding-system)
2371 i1 (length enc1)
2372 enc2 (encode-coding-string str2 coding-system)
2373 i2 (length enc2))
2374 (while (and (> i1 0) (= (aref enc1 (1- i1)) (aref enc2 (1- i2))))
2375 (setq i1 (1- i1) i2 (1- i2)))
2377 ;; Now (substring enc1 i1) and (substring enc2 i2) are the same,
2378 ;; and they are the extra control sequences at the tail to
2379 ;; exclude.
2380 (substring enc2 0 i2))))
2383 ;;; mule-cmds.el ends here