1 ;;; mule.el --- basic commands for mulitilingual environment
3 ;; Copyright (C) 1995 Electrotechnical Laboratory, JAPAN.
4 ;; Licensed to the Free Software Foundation.
6 ;; Keywords: mule, multilingual, character set, coding system
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)
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.
27 (defconst mule-version "5.0 (AOI)" "\
28 Version number and name of this version of MULE (multilingual environment).")
30 (defconst mule-version-date "1999.12.7" "\
31 Distribution date of this version of MULE (multilingual environment).")
33 (defun load-with-code-conversion (fullname file &optional noerror nomessage)
34 "Execute a file of Lisp code named FILE whose absolute name is FULLNAME.
35 The file contents are decoded before evaluation if necessary.
36 If optional second arg NOERROR is non-nil,
37 report no error if FILE doesn't exist.
38 Print messages at start and end of loading unless
39 optional third arg NOMESSAGE is non-nil.
40 Return t if file exists."
41 (if (null (file-readable-p fullname))
43 (signal 'file-error (list "Cannot open load file" file)))
44 ;; Read file with code conversion, and then eval.
46 ;; To avoid any autoloading, set default-major-mode to
48 ;; So that we don't get completely screwed if the
49 ;; file is encoded in some complicated character set,
50 ;; read it with real decoding, as a multibyte buffer,
51 ;; even if this is a --unibyte Emacs session.
52 (let ((default-major-mode 'fundamental-mode)
53 (default-enable-multibyte-characters t))
54 ;; We can't use `generate-new-buffer' because files.el
56 (get-buffer-create (generate-new-buffer-name " *load*"))))
58 (source (save-match-data (string-match "\\.el\\'" fullname))))
61 (message "Loading %s (source)..." file)
62 (message "Loading %s..." file)))
64 (setq preloaded-file-list (cons file preloaded-file-list)))
66 (let ((load-file-name fullname)
67 (set-auto-coding-for-load t)
68 (inhibit-file-name-operation nil))
71 (insert-file-contents fullname)
72 ;; If the loaded file was inserted with no-conversion or
73 ;; raw-text coding system, make the buffer unibyte.
74 ;; Otherwise, eval-buffer might try to interpret random
75 ;; binary junk as multibyte characters.
76 (if (and enable-multibyte-characters
77 (or (eq (coding-system-type last-coding-system-used) 5)
78 (eq last-coding-system-used 'no-conversion)))
79 (set-buffer-multibyte nil))
80 ;; Make `kill-buffer' quiet.
81 (set-buffer-modified-p nil))
82 ;; Have the original buffer current while we eval.
83 (eval-buffer buffer nil file
84 ;; If this Emacs is running with --unibyte,
85 ;; convert multibyte strings to unibyte
86 ;; after reading them.
87 ;; (not default-enable-multibyte-characters)
90 (let (kill-buffer-hook kill-buffer-query-functions)
91 (kill-buffer buffer)))
92 (let ((hook (assoc file after-load-alist)))
94 (mapcar (function eval) (cdr hook))))
95 (unless (or nomessage noninteractive)
97 (message "Loading %s (source)...done" file)
98 (message "Loading %s...done" file)))
101 ;; API (Application Program Interface) for charsets.
103 ;; Return t if OBJ is a quoted symbol
104 ;; and the symbol is the name of a standard charset.
105 (defsubst charset-quoted-standard-p (obj)
106 (and (listp obj) (eq (car obj) 'quote)
107 (symbolp (car-safe (cdr obj)))
108 (let ((vector (get (car-safe (cdr obj)) 'charset)))
109 (and (vectorp vector)
110 (< (aref vector 0) 160)))))
112 (defsubst charsetp (object)
113 "T if OBJECT is a charset."
114 (and (symbolp object) (vectorp (get object 'charset))))
116 (defsubst charset-info (charset)
117 "Return a vector of information of CHARSET.
118 The elements of the vector are:
119 CHARSET-ID, BYTES, DIMENSION, CHARS, WIDTH, DIRECTION,
120 LEADING-CODE-BASE, LEADING-CODE-EXT,
121 ISO-FINAL-CHAR, ISO-GRAPHIC-PLANE,
122 REVERSE-CHARSET, SHORT-NAME, LONG-NAME, DESCRIPTION,
125 CHARSET-ID (integer) is the identification number of the charset.
126 BYTES (integer) is the length of multi-byte form of a character in
127 the charset: one of 1, 2, 3, and 4.
128 DIMENSION (integer) is the number of bytes to represent a character of
130 CHARS (integer) is the number of characters in a dimension: 94 or 96.
131 WIDTH (integer) is the number of columns a character in the charset
132 occupies on the screen: one of 0, 1, and 2.
133 DIRECTION (integer) is the rendering direction of characters in the
134 charset when rendering. If 0, render from left to right, else
135 render from right to left.
136 LEADING-CODE-BASE (integer) is the base leading-code for the
138 LEADING-CODE-EXT (integer) is the extended leading-code for the
139 charset. All charsets of less than 0xA0 has the value 0.
140 ISO-FINAL-CHAR (character) is the final character of the
141 corresponding ISO 2022 charset.
142 ISO-GRAPHIC-PLANE (integer) is the graphic plane to be invoked
143 while encoding to variants of ISO 2022 coding system, one of the
144 following: 0/graphic-plane-left(GL), 1/graphic-plane-right(GR).
145 REVERSE-CHARSET (integer) is the charset which differs only in
146 LEFT-TO-RIGHT value from the charset. If there's no such a
147 charset, the value is -1.
148 SHORT-NAME (string) is the short name to refer to the charset.
149 LONG-NAME (string) is the long name to refer to the charset
150 DESCRIPTION (string) is the description string of the charset.
151 PLIST (property list) may contain any type of information a user
152 want to put and get by functions `put-charset-property' and
153 `get-charset-property' respectively."
154 (get charset 'charset))
156 ;; It is better not to use backquote in this file,
157 ;; because that makes a bootstrapping problem
158 ;; if you need to recompile all the Lisp files using interpreted code.
160 (defmacro charset-id (charset)
161 "Return charset identification number of CHARSET."
162 (if (charset-quoted-standard-p charset)
163 (aref (charset-info (nth 1 charset)) 0)
164 (list 'aref (list 'charset-info charset) 0)))
166 (defmacro charset-bytes (charset)
167 "Return bytes of CHARSET.
168 See the function `charset-info' for more detail."
169 (if (charset-quoted-standard-p charset)
170 (aref (charset-info (nth 1 charset)) 1)
171 (list 'aref (list 'charset-info charset) 1)))
173 (defmacro charset-dimension (charset)
174 "Return dimension of CHARSET.
175 See the function `charset-info' for more detail."
176 (if (charset-quoted-standard-p charset)
177 (aref (charset-info (nth 1 charset)) 2)
178 (list 'aref (list 'charset-info charset) 2)))
180 (defmacro charset-chars (charset)
181 "Return character numbers contained in a dimension of CHARSET.
182 See the function `charset-info' for more detail."
183 (if (charset-quoted-standard-p charset)
184 (aref (charset-info (nth 1 charset)) 3)
185 (list 'aref (list 'charset-info charset) 3)))
187 (defmacro charset-width (charset)
188 "Return width (how many column occupied on a screen) of CHARSET.
189 See the function `charset-info' for more detail."
190 (if (charset-quoted-standard-p charset)
191 (aref (charset-info (nth 1 charset)) 4)
192 (list 'aref (list 'charset-info charset) 4)))
194 (defmacro charset-direction (charset)
195 "Return direction of CHARSET.
196 See the function `charset-info' for more detail."
197 (if (charset-quoted-standard-p charset)
198 (aref (charset-info (nth 1 charset)) 5)
199 (list 'aref (list 'charset-info charset) 5)))
201 (defmacro charset-iso-final-char (charset)
202 "Return final char of CHARSET.
203 See the function `charset-info' for more detail."
204 (if (charset-quoted-standard-p charset)
205 (aref (charset-info (nth 1 charset)) 8)
206 (list 'aref (list 'charset-info charset) 8)))
208 (defmacro charset-iso-graphic-plane (charset)
209 "Return graphic plane of CHARSET.
210 See the function `charset-info' for more detail."
211 (if (charset-quoted-standard-p charset)
212 (aref (charset-info (nth 1 charset)) 9)
213 (list 'aref (list 'charset-info charset) 9)))
215 (defmacro charset-reverse-charset (charset)
216 "Return reverse charset of CHARSET.
217 See the function `charset-info' for more detail."
218 (if (charset-quoted-standard-p charset)
219 (aref (charset-info (nth 1 charset)) 10)
220 (list 'aref (list 'charset-info charset) 10)))
222 (defmacro charset-short-name (charset)
223 "Return short name of CHARSET.
224 See the function `charset-info' for more detail."
225 (if (charset-quoted-standard-p charset)
226 (aref (charset-info (nth 1 charset)) 11)
227 (list 'aref (list 'charset-info charset) 11)))
229 (defmacro charset-long-name (charset)
230 "Return long name of CHARSET.
231 See the function `charset-info' for more detail."
232 (if (charset-quoted-standard-p charset)
233 (aref (charset-info (nth 1 charset)) 12)
234 (list 'aref (list 'charset-info charset) 12)))
236 (defmacro charset-description (charset)
237 "Return description of CHARSET.
238 See the function `charset-info' for more detail."
239 (if (charset-quoted-standard-p charset)
240 (aref (charset-info (nth 1 charset)) 13)
241 (list 'aref (list 'charset-info charset) 13)))
243 (defmacro charset-plist (charset)
244 "Return list charset property of CHARSET.
245 See the function `charset-info' for more detail."
247 (if (charset-quoted-standard-p charset)
248 (charset-info (nth 1 charset))
249 (list 'charset-info charset))
252 (defun set-charset-plist (charset plist)
253 "Set CHARSET's property list to PLIST, and return PLIST."
254 (aset (charset-info charset) 14 plist))
256 (defun make-char (charset &optional c1 c2)
257 "Return a character of CHARSET and position codes CODE1 and CODE2.
258 CODE1 and CODE2 are optional, but if you don't supply
259 sufficient position codes, return a generic character which stands for
260 all characters or group of characters in the character set.
261 A generic character can be used to index a char table (e.g. syntax-table)."
262 (make-char-internal (charset-id charset) c1 c2))
264 (put 'make-char 'byte-compile
267 (let ((charset (nth 1 form)))
268 (if (charset-quoted-standard-p charset)
269 (byte-compile-normal-call
270 (cons 'make-char-internal
271 (cons (charset-id (nth 1 charset)) (nthcdr 2 form))))
272 (byte-compile-normal-call
273 (cons 'make-char-internal
274 (cons (list 'charset-id charset) (nthcdr 2 form)))))))))
276 (defun charset-list ()
277 "Return list of charsets ever defined.
279 This function is provided for backward compatibility.
280 Now we have the variable `charset-list'."
283 (defsubst generic-char-p (char)
284 "Return t if and only if CHAR is a generic character.
285 See also the documentation of make-char."
287 (let ((l (split-char char)))
288 (and (or (= (nth 1 l) 0) (eq (nth 2 l) 0))
289 (not (eq (car l) 'composition))))))
292 ;; Coding system staffs
294 ;; Coding system is a symbol that has the property `coding-system'.
296 ;; The value of the property `coding-system' is a vector of the
298 ;; [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
299 ;; We call this vector as coding-spec. See comments in src/coding.c
302 (defconst coding-spec-type-idx 0)
303 (defconst coding-spec-mnemonic-idx 1)
304 (defconst coding-spec-doc-string-idx 2)
305 (defconst coding-spec-plist-idx 3)
306 (defconst coding-spec-flags-idx 4)
308 ;; PLIST is a property list of a coding system. To share PLIST among
309 ;; alias coding systems, a coding system has PLIST in coding-spec
310 ;; instead of having it in normal property list of Lisp symbol.
311 ;; Here's a list of coding system properties currently being used.
315 ;; The value is a coding category the coding system belongs to. The
316 ;; function `make-coding-system' sets this value automatically
317 ;; unless its argument PROPERTIES specifies this property.
319 ;; o alias-coding-systems
321 ;; The value is a list of coding systems of the same alias group. The
322 ;; first element is the coding system made at first, which we call as
323 ;; `base coding system'. The function `make-coding-system' sets this
324 ;; value automatically and `define-coding-system-alias' updates it.
326 ;; o post-read-conversion
328 ;; The value is a function to call after some text is inserted and
329 ;; decoded by the coding system itself and before any functions in
330 ;; `after-insert-functions' are called. The arguments to this
331 ;; function is the same as those of a function in
332 ;; `after-insert-functions', i.e. LENGTH of a text while putting point
333 ;; at the head of the text to be decoded
335 ;; o pre-write-conversion
337 ;; The value is a function to call after all functions in
338 ;; `write-region-annotate-functions' and `buffer-file-format' are
339 ;; called, and before the text is encoded by the coding system itself.
340 ;; The arguments to this function is the same as those of a function
341 ;; in `write-region-annotate-functions', i.e. FROM and TO specifying
344 ;; o translation-table-for-decode
346 ;; The value is a translation table to be applied on decoding. See
347 ;; the function `make-translation-table' for the format of translation
350 ;; o translation-table-for-encode
352 ;; The value is a translation table to be applied on encoding.
356 ;; The value is a list of charsets safely supported by the coding
357 ;; system. The value t means that all charsets Emacs handles are
358 ;; supported. Even if some charset is not in this list, it doesn't
359 ;; mean that the charset can't be encoded in the coding system,
360 ;; instead, it just means that some other receiver of a text encoded
361 ;; in the coding system won't be able to handle that charset.
365 ;; The value is a symbol of which name is `MIME-charset' parameter of
366 ;; the coding system.
368 ;; o charset-origin-alist
370 ;; The value is a list of this form:
371 ;; (CHARSET EXTERNAL-CHARSET-NAME ENCODING-FUNCTION).
372 ;; ENCODING-FUNCTION is a function to encode a character in CHARSET
373 ;; to the code in EXTERNAL-CHARSET-NAME. The command what-cursor-position
374 ;; uses this information of the buffer-file-coding-system.
375 ;; ENCODING-FUNCTION may be a translation table or a symbol whose
376 ;; property `translation-table' is a translation table. In these case,
377 ;; the translation table is used to encode the character.
379 ;; o valid-codes (meaningful only for a coding system based on CCL)
381 ;; The value is a list to indicate valid byte ranges of the encoded
382 ;; file. Each element of the list is an integer or a cons of integer.
383 ;; In the former case, the integer value is a valid byte code. In the
384 ;; latter case, the integers specifies the range of valid byte codes.
387 ;; Return coding-spec of CODING-SYSTEM
388 (defsubst coding-system-spec (coding-system)
389 (get (check-coding-system coding-system) 'coding-system))
391 (defun coding-system-type (coding-system)
392 "Return the coding type of CODING-SYSTEM.
393 A coding type is an integer value indicating the encoding method
394 of CODING-SYSTEM. See the function `make-coding-system' for more detail."
395 (aref (coding-system-spec coding-system) coding-spec-type-idx))
397 (defun coding-system-mnemonic (coding-system)
398 "Return the mnemonic character of CODING-SYSTEM.
399 The mnemonic character of a coding system is used in mode line
400 to indicate the coding system. If the arg is nil, return ?-."
401 (let ((spec (coding-system-spec coding-system)))
402 (if spec (aref spec coding-spec-mnemonic-idx) ?-)))
404 (defun coding-system-doc-string (coding-system)
405 "Return the documentation string for CODING-SYSTEM."
406 (aref (coding-system-spec coding-system) coding-spec-doc-string-idx))
408 (defun coding-system-plist (coding-system)
409 "Return the property list of CODING-SYSTEM."
410 (aref (coding-system-spec coding-system) coding-spec-plist-idx))
412 (defun coding-system-flags (coding-system)
413 "Return `flags' of CODING-SYSTEM.
414 A `flags' of a coding system is a vector of length 32 indicating detailed
415 information of a coding system. See the function `make-coding-system'
417 (aref (coding-system-spec coding-system) coding-spec-flags-idx))
419 (defun coding-system-get (coding-system prop)
420 "Extract a value from CODING-SYSTEM's property list for property PROP."
421 (plist-get (coding-system-plist coding-system) prop))
423 (defun coding-system-put (coding-system prop val)
424 "Change value in CODING-SYSTEM's property list PROP to VAL."
425 (let ((plist (coding-system-plist coding-system)))
427 (plist-put plist prop val)
428 (aset (coding-system-spec coding-system) coding-spec-plist-idx
431 (defun coding-system-category (coding-system)
432 "Return the coding category of CODING-SYSTEM."
433 (coding-system-get coding-system 'coding-category))
435 (defun coding-system-base (coding-system)
436 "Return the base coding system of CODING-SYSTEM.
437 A base coding system is what made by `make-coding-system'.
438 Any alias nor subsidiary coding systems are not base coding system."
439 (car (coding-system-get coding-system 'alias-coding-systems)))
441 (defalias 'coding-system-parent 'coding-system-base)
442 (make-obsolete 'coding-system-parent 'coding-system-base "20.3")
444 ;; Coding system also has a property `eol-type'.
446 ;; This property indicates how the coding system handles end-of-line
447 ;; format. The value is integer 0, 1, 2, or a vector of three coding
448 ;; systems. Each integer value 0, 1, and 2 indicates the format of
449 ;; end-of-line LF, CRLF, and CR respectively. A vector value
450 ;; indicates that the format of end-of-line should be detected
451 ;; automatically. Nth element of the vector is the subsidiary coding
452 ;; system whose `eol-type' property is N.
454 (defun coding-system-eol-type (coding-system)
455 "Return eol-type of CODING-SYSTEM.
456 An eol-type is integer 0, 1, 2, or a vector of coding systems.
458 Integer values 0, 1, and 2 indicate a format of end-of-line; LF,
459 CRLF, and CR respectively.
461 A vector value indicates that a format of end-of-line should be
462 detected automatically. Nth element of the vector is the subsidiary
463 coding system whose eol-type is N."
464 (get coding-system 'eol-type))
466 (defun coding-system-lessp (x y)
467 (cond ((eq x 'no-conversion) t)
468 ((eq y 'no-conversion) nil)
469 ((eq x 'emacs-mule) t)
470 ((eq y 'emacs-mule) nil)
471 ((eq x 'undecided) t)
472 ((eq y 'undecided) nil)
473 (t (let ((c1 (coding-system-mnemonic x))
474 (c2 (coding-system-mnemonic y)))
475 (or (< (downcase c1) (downcase c2))
476 (and (not (> (downcase c1) (downcase c2)))
479 ;; Add CODING-SYSTEM to coding-system-list while keeping it sorted.
480 (defun add-to-coding-system-list (coding-system)
481 (if (or (null coding-system-list)
482 (coding-system-lessp coding-system (car coding-system-list)))
483 (setq coding-system-list (cons coding-system coding-system-list))
484 (let ((len (length coding-system-list))
485 mid (tem coding-system-list))
487 (setq mid (nthcdr (/ len 2) tem))
488 (if (coding-system-lessp (car mid) coding-system)
490 len (- len (/ len 2)))
491 (setq len (/ len 2))))
492 (setcdr tem (cons coding-system (cdr tem))))))
494 (defun coding-system-list (&optional base-only)
495 "Return a list of all existing coding systems.
496 If optional arg BASE-ONLY is non-nil, only base coding systems are listed."
497 (let* ((codings (copy-sequence coding-system-list))
498 (tail (cons nil codings)))
499 ;; Remove subsidiary coding systems (eol variants) and alias
500 ;; coding systems (if necessary).
502 (let* ((coding (car (cdr tail)))
503 (aliases (coding-system-get coding 'alias-coding-systems)))
505 ;; CODING is an eol variant if not in ALIASES.
506 (not (memq coding aliases))
507 ;; CODING is an alias if it is not car of ALIASES.
508 (and base-only (not (eq coding (car aliases)))))
509 (setcdr tail (cdr (cdr tail)))
510 (setq tail (cdr tail)))))
513 ;; Make subsidiary coding systems (eol-type variants) of CODING-SYSTEM.
514 (defun make-subsidiary-coding-system (coding-system)
515 (let ((coding-spec (coding-system-spec coding-system))
516 (subsidiaries (vector (intern (format "%s-unix" coding-system))
517 (intern (format "%s-dos" coding-system))
518 (intern (format "%s-mac" coding-system))))
522 (put (aref subsidiaries i) 'coding-system coding-spec)
523 (put (aref subsidiaries i) 'eol-type i)
524 (add-to-coding-system-list (aref subsidiaries i))
525 (setq coding-system-alist
526 (cons (list (symbol-name (aref subsidiaries i)))
527 coding-system-alist))
531 (defun make-coding-system (coding-system type mnemonic doc-string
536 "Define a new coding system CODING-SYSTEM (symbol).
537 Remaining arguments are TYPE, MNEMONIC, DOC-STRING, FLAGS (optional),
538 and PROPERTIES (optional) which construct a coding-spec of CODING-SYSTEM
539 in the following format:
540 [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
542 TYPE is an integer value indicating the type of the coding system as follows:
543 0: Emacs internal format,
544 1: Shift-JIS (or MS-Kanji) used mainly on Japanese PC,
545 2: ISO-2022 including many variants,
546 3: Big5 used mainly on Chinese PC,
547 4: private, CCL programs provide encoding/decoding algorithm,
548 5: Raw-text, which means that text contains random 8-bit codes.
550 MNEMONIC is a character to be displayed on mode line for the coding system.
552 DOC-STRING is a documentation string for the coding system.
554 FLAGS specifies more detailed information of the coding system as follows:
556 If TYPE is 2 (ISO-2022), FLAGS is a list of these elements:
557 CHARSET0, CHARSET1, CHARSET2, CHARSET3, SHORT-FORM,
558 ASCII-EOL, ASCII-CNTL, SEVEN, LOCKING-SHIFT, SINGLE-SHIFT,
559 USE-ROMAN, USE-OLDJIS, NO-ISO6429, INIT-BOL, DESIGNATION-BOL,
560 SAFE, ACCEPT-LATIN-EXTRA-CODE.
561 CHARSETn are character sets initially designated to Gn graphic registers.
562 If CHARSETn is nil, Gn is never used.
563 If CHARSETn is t, Gn can be used but nothing designated initially.
564 If CHARSETn is a list of character sets, those character sets are
565 designated to Gn on output, but nothing designated to Gn initially.
566 But, character set `ascii' can be designated only to G0.
567 SHORT-FORM non-nil means use short designation sequence on output.
568 ASCII-EOL non-nil means designate ASCII to g0 at end of line on output.
569 ASCII-CNTL non-nil means designate ASCII to g0 before control codes and
571 SEVEN non-nil means use 7-bit code only on output.
572 LOCKING-SHIFT non-nil means use locking-shift.
573 SINGLE-SHIFT non-nil means use single-shift.
574 USE-ROMAN non-nil means designate JIS0201-1976-Roman instead of ASCII.
575 USE-OLDJIS non-nil means designate JIS0208-1976 instead of JIS0208-1983.
576 NO-ISO6429 non-nil means not use ISO6429's direction specification.
577 INIT-BOL non-nil means any designation state is assumed to be reset
578 to initial at each beginning of line on output.
579 DESIGNATION-BOL non-nil means designation sequences should be placed
580 at beginning of line on output.
581 SAFE non-nil means convert unsafe characters to `?' on output.
582 Unsafe characters are what not specified in SAFE-CHARSET.
583 ACCEPT-LATIN-EXTRA-CODE non-nil means code-detection routine accepts
584 a code specified in `latin-extra-code-table' (which see) as a valid
585 code of the coding system.
587 If TYPE is 4 (private), FLAGS should be a cons of CCL programs, for
588 decoding and encoding. CCL programs should be specified by their
591 PROPERTIES is an alist of properties vs the corresponding values.
592 These properties are set in PLIST, a property list. This function
593 also sets properties `coding-category' and `alias-coding-systems'
596 EOL-TYPE specifies the EOL type of the coding-system in one of the
599 o symbol (unix, dos, or mac)
601 The symbol `unix' means Unix-like EOL (LF), `dos' means
602 DOS-like EOL (CRLF), and `mac' means MAC-like EOL (CR).
604 o number (0, 1, or 2)
606 The number 0, 1, and 2 mean UNIX, DOS, and MAC-like EOL
609 o vector of coding-systems of length 3
611 The EOL type is detected automatically for the coding system.
612 And, according to the detected EOL type, one of the coding
613 systems in the vector is selected. Elements of the vector
614 corresponds to Unix-liek EOL, DOS-like EOL, and Mac-like EOL
617 Kludgy features for backward compatibility:
619 1. If TYPE is 4 and car or cdr of FLAGS is a vector, the vector is
620 treated as a compiled CCL code.
622 2. If PROPERTIES is just a list of character sets, the list is set as
623 a value of `safe-charsets' in PLIST."
624 ;; Set a value of `coding-system' property.
625 (let ((coding-spec (make-vector 5 nil))
626 (no-initial-designation t)
627 (no-alternative-designation t)
629 (if (or (not (integerp type)) (< type 0) (> type 5))
630 (error "TYPE argument must be 0..5"))
631 (if (or (not (integerp mnemonic)) (<= mnemonic ? ) (> mnemonic 127))
632 (error "MNEMONIC argument must be an ASCII printable character."))
633 (aset coding-spec coding-spec-type-idx type)
634 (aset coding-spec coding-spec-mnemonic-idx mnemonic)
635 (aset coding-spec coding-spec-doc-string-idx
636 (purecopy (if (stringp doc-string) doc-string "")))
638 (setq coding-category 'coding-category-emacs-mule))
640 (setq coding-category 'coding-category-sjis))
641 ((= type 2) ; ISO2022
643 (vec (make-vector 32 nil))
647 (let ((charset (car fl)))
648 (if (and no-initial-designation
650 (or (charsetp charset)
652 (charsetp (car charset)))))
653 (setq no-initial-designation nil))
654 (if (charsetp charset)
655 (if (= i 1) (setq g1-designation charset))
660 (setq elt (car tail))
662 (setq no-alternative-designation nil)
663 (if (and elt (not (charsetp elt)))
664 (error "Invalid charset: %s" elt)))
665 (setq tail (cdr tail)))
666 (setq g1-designation (car charset)))
669 (setq no-alternative-designation nil)
670 (error "Invalid charset: %s" charset)))))
671 (aset vec i charset))
672 (setq fl (cdr fl) i (1+ i)))
673 (while (and (< i 32) fl)
674 (aset vec i (car fl))
675 (setq fl (cdr fl) i (1+ i)))
676 (aset coding-spec 4 vec)
677 (setq coding-category
678 (if (aref vec 8) ; Use locking-shift.
679 (or (and (aref vec 7) 'coding-category-iso-7-else)
680 'coding-category-iso-8-else)
681 (if (aref vec 7) ; 7-bit only.
682 (if (aref vec 9) ; Use single-shift.
683 'coding-category-iso-7-else
684 (if no-alternative-designation
685 'coding-category-iso-7-tight
686 'coding-category-iso-7))
687 (if (or no-initial-designation
688 (not no-alternative-designation))
689 'coding-category-iso-8-else
690 (if (and (charsetp g1-designation)
691 (= (charset-dimension g1-designation) 2))
692 'coding-category-iso-8-2
693 'coding-category-iso-8-1)))))))
695 (setq coding-category 'coding-category-big5))
696 ((= type 4) ; private
697 (setq coding-category 'coding-category-ccl)
698 (if (not (consp flags))
699 (error "Invalid FLAGS argument for TYPE 4 (CCL)")
700 (let ((decoder (check-ccl-program
702 (intern (format "%s-decoder" coding-system))))
703 (encoder (check-ccl-program
705 (intern (format "%s-encoder" coding-system)))))
706 (if (and decoder encoder)
707 (aset coding-spec 4 (cons decoder encoder))
708 (error "Invalid FLAGS argument for TYPE 4 (CCL)")))))
710 (setq coding-category 'coding-category-raw-text)))
712 (let ((plist (list 'coding-category coding-category
713 'alias-coding-systems (list coding-system))))
714 (if no-initial-designation
715 (plist-put plist 'no-initial-designation t))
717 (or (eq properties t)
718 (not (consp (car properties)))))
719 ;; In the old version, the arg PROPERTIES is a list to be
720 ;; set in PLIST as a value of property `safe-charsets'.
721 (plist-put plist 'safe-charsets properties)
722 ;; In the current version PROPERTIES is a property list.
723 ;; Reflect it into PLIST one by one.
724 (let ((l properties))
726 (plist-put plist (car (car l)) (cdr (car l)))
728 ;; The property `coding-category' may have been set differently
729 ;; through PROPERTIES.
730 (setq coding-category (plist-get plist 'coding-category))
731 (aset coding-spec coding-spec-plist-idx plist))
732 (put coding-system 'coding-system coding-spec)
733 (put coding-category 'coding-systems
734 (cons coding-system (get coding-category 'coding-systems))))
736 ;; Next, set a value of `eol-type' property.
737 (if (and (not eol-type)
738 (or (<= type 3) (= type 5)))
739 ;; If EOL-TYPE is nil, set a vector of subsidiary coding
740 ;; systems, each corresponds to a coding system for the detected
742 (setq eol-type (make-subsidiary-coding-system coding-system)))
744 (cond ((or (eq eol-type 'unix) (null eol-type))
750 ((or (and (vectorp eol-type)
751 (= (length eol-type) 3))
752 (and (numberp eol-type)
757 (error "Invalid EOL-TYPE spec:%S" eol-type))))
758 (put coding-system 'eol-type eol-type)
760 ;; At last, register CODING-SYSTEM in `coding-system-list' and
761 ;; `coding-system-alist'.
762 (add-to-coding-system-list coding-system)
763 (setq coding-system-alist (cons (list (symbol-name coding-system))
764 coding-system-alist))
766 ;; For a coding system of cateogory iso-8-1 and iso-8-2, create
767 ;; XXX-with-esc variants.
768 (let ((coding-category (coding-system-category coding-system)))
769 (if (or (eq coding-category 'coding-category-iso-8-1)
770 (eq coding-category 'coding-category-iso-8-2))
771 (let ((esc (intern (concat (symbol-name coding-system) "-with-esc")))
772 (doc (format "Same as %s but can handle any charsets by ISO's escape sequences." coding-system)))
773 (make-coding-system esc type mnemonic doc
774 (if (listp (car flags))
775 (cons (append (car flags) '(t)) (cdr flags))
776 (cons (list (car flags) t) (cdr flags)))
778 (coding-system-put esc 'mime-charset nil)
779 (coding-system-put esc 'safe-charsets t))))
783 (defun define-coding-system-alias (alias coding-system)
784 "Define ALIAS as an alias for coding system CODING-SYSTEM."
785 (put alias 'coding-system (coding-system-spec coding-system))
786 (nconc (coding-system-get alias 'alias-coding-systems) (list alias))
787 (add-to-coding-system-list alias)
788 (setq coding-system-alist (cons (list (symbol-name alias))
789 coding-system-alist))
790 (let ((eol-type (coding-system-eol-type coding-system)))
791 (if (vectorp eol-type)
792 (put alias 'eol-type (make-subsidiary-coding-system alias))
793 (put alias 'eol-type eol-type))))
795 (defun set-buffer-file-coding-system (coding-system &optional force)
796 "Set the file coding-system of the current buffer to CODING-SYSTEM.
797 This means that when you save the buffer, it will be converted
798 according to CODING-SYSTEM. For a list of possible values of CODING-SYSTEM,
799 use \\[list-coding-systems].
801 If the buffer's previous file coding-system value specifies end-of-line
802 conversion, and CODING-SYSTEM does not specify one, CODING-SYSTEM is
803 merged with the already-specified end-of-line conversion.
804 However, if the optional prefix argument FORCE is non-nil,
805 then CODING-SYSTEM is used exactly as specified.
807 This marks the buffer modified so that the succeeding \\[save-buffer]
808 surely saves the buffer with CODING-SYSTEM. From a program, if you
809 don't want to mark the buffer modified, just set the variable
810 `buffer-file-coding-system' directly."
811 (interactive "zCoding system for visited file (default, nil): \nP")
812 (check-coding-system coding-system)
814 (let ((x (coding-system-eol-type buffer-file-coding-system))
815 (y (coding-system-eol-type coding-system)))
816 (if (and (numberp x) (>= x 0) (<= x 2) (vectorp y))
817 (setq coding-system (aref y x)))))
818 (setq buffer-file-coding-system coding-system)
819 (set-buffer-modified-p t)
820 (force-mode-line-update))
822 (defvar default-terminal-coding-system nil
823 "Default value for the terminal coding system.
824 This is normally set according to the selected language environment.
825 See also the command `set-terminal-coding-system'.")
827 (defun set-terminal-coding-system (coding-system)
828 "Set coding system of your terminal to CODING-SYSTEM.
829 All text output to the terminal will be encoded
830 with the specified coding system.
831 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
832 The default is determined by the selected language environment
833 or by the previous use of this command."
835 (list (let ((default (if (and (not (terminal-coding-system))
836 default-terminal-coding-system)
837 default-terminal-coding-system)))
839 (format "Coding system for terminal display (default, %s): "
842 (if (and (not coding-system)
843 (not (terminal-coding-system)))
844 (setq coding-system default-terminal-coding-system))
846 (setq default-terminal-coding-system coding-system))
847 (set-terminal-coding-system-internal coding-system)
848 (redraw-frame (selected-frame)))
850 (defvar default-keyboard-coding-system nil
851 "Default value of the keyboard coding system.
852 This is normally set according to the selected language environment.
853 See also the command `set-keyboard-coding-system'.")
855 (defun set-keyboard-coding-system (coding-system)
856 "Set coding system for keyboard input to CODING-SYSTEM.
857 In addition, this command enables Encoded-kbd minor mode.
858 \(If CODING-SYSTEM is nil, Encoded-kbd mode is turned off.)
859 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
860 The default is determined by the selected language environment
861 or by the previous use of this command."
863 (list (let ((default (if (and (not (keyboard-coding-system))
864 default-keyboard-coding-system)
865 default-keyboard-coding-system)))
867 (format "Coding system for keyboard input (default, %s): "
870 (if (and (not coding-system)
871 (not (keyboard-coding-system)))
872 (setq coding-system default-keyboard-coding-system))
874 (setq default-keyboard-coding-system coding-system))
875 (set-keyboard-coding-system-internal coding-system)
876 (encoded-kbd-mode (if coding-system 1 0)))
878 (defun set-buffer-process-coding-system (decoding encoding)
879 "Set coding systems for the process associated with the current buffer.
880 DECODING is the coding system to be used to decode input from the process,
881 ENCODING is the coding system to be used to encode output to the process.
883 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems]."
885 "zCoding-system for process input: \nzCoding-system for process output: ")
886 (let ((proc (get-buffer-process (current-buffer))))
889 (check-coding-system decoding)
890 (check-coding-system encoding)
891 (set-process-coding-system proc decoding encoding)))
892 (force-mode-line-update))
894 (defalias 'set-clipboard-coding-system 'set-selection-coding-system)
896 (defun set-selection-coding-system (coding-system)
897 "Make CODING-SYSTEM used for communicating with other X clients .
898 When sending or receiving text via cut_buffer, selection, and clipboard,
899 the text is encoded or decoded by CODING-SYSTEM."
900 (interactive "zCoding system for X selection: ")
901 (check-coding-system coding-system)
902 (setq selection-coding-system coding-system))
904 ;; Coding system lastly specified by the command
905 ;; set-next-selection-coding-system.
906 (defvar last-next-selection-coding-system nil)
908 (defun set-next-selection-coding-system (coding-system)
909 "Make CODING-SYSTEM used for the next communication with other X clients.
910 This setting is effective for the next communication only."
912 (list (read-coding-system
913 (if last-next-selection-coding-system
914 (format "Coding system for the next X selection (default, %S): "
915 last-next-selection-coding-system)
916 "Coding system for the next X selection: ")
917 last-next-selection-coding-system)))
919 (setq last-next-selection-coding-system coding-system)
920 (setq coding-system last-next-selection-coding-system))
921 (check-coding-system coding-system)
923 (setq next-selection-coding-system coding-system))
925 (defun set-coding-priority (arg)
926 "Set priority of coding categories according to LIST.
927 LIST is a list of coding categories ordered by priority."
929 (current-list (copy-sequence coding-category-list)))
930 ;; Check the validity of ARG while deleting coding categories in
931 ;; ARG from CURRENT-LIST. We assume that CODING-CATEGORY-LIST
932 ;; contains all coding categories.
934 (if (or (null (get (car l) 'coding-category-index))
935 (null (memq (car l) current-list)))
936 (error "Invalid or duplicated element in argument: %s" arg))
937 (setq current-list (delq (car l) current-list))
939 ;; Update `coding-category-list' and return it.
940 (setq coding-category-list (append arg current-list))
941 (set-coding-priority-internal)))
945 (defvar auto-coding-alist
946 '(("\\.\\(arc\\|zip\\|lzh\\|zoo\\|jar\\|tar\\|tgz\\)\\'" . no-conversion)
947 ("\\.\\(ARC\\|ZIP\\|LZH\\|ZOO\\|JAR\\|TAR\\|TGZ\\)\\'" . no-conversion))
948 "Alist of filename patterns vs corresponding coding systems.
949 Each element looks like (REGEXP . CODING-SYSTEM).
950 A file whose name matches REGEXP is decoded by CODING-SYSTEM on reading.
952 The settings in this alist take priority over `coding:' tags
953 in the file (see the function `set-auto-coding')
954 and the contents of `file-coding-system-alist'.")
956 (defvar set-auto-coding-for-load nil
957 "Non-nil means look for `load-coding' property instead of `coding'.
958 This is used for loading and byte-compiling Emacs Lisp files.")
960 (defun auto-coding-alist-lookup (filename)
961 "Return the coding system specified by `auto-coding-alist' for FILENAME."
962 (let ((alist auto-coding-alist)
963 (case-fold-search (memq system-type '(vax-vms windows-nt ms-dos)))
965 (while (and alist (not coding-system))
966 (if (string-match (car (car alist)) filename)
967 (setq coding-system (cdr (car alist)))
968 (setq alist (cdr alist))))
971 (defun set-auto-coding (filename size)
972 "Return coding system for a file FILENAME of which SIZE bytes follow point.
973 These bytes should include at least the first 1k of the file
974 and the last 3k of the file, but the middle may be omitted.
976 It checks FILENAME against the variable `auto-coding-alist'.
977 If FILENAME doesn't match any entries in the variable,
978 it checks for a `coding:' tag in the first one or two lines following
979 point. If no `coding:' tag is found, it checks for local variables
980 list in the last 3K bytes out of the SIZE bytes.
982 The return value is the specified coding system,
983 or nil if nothing specified.
985 The variable `set-auto-coding-function' (which see) is set to this
986 function by default."
987 (let ((coding-system (auto-coding-alist-lookup filename)))
990 (let* ((case-fold-search t)
992 (head-end (+ head-start (min size 1024)))
993 (tail-start (+ head-start (max (- size 3072) 0)))
994 (tail-end (+ head-start size))
995 coding-system head-found tail-found pos)
996 ;; Try a short cut by searching for the string "coding:"
997 ;; and for "unibyte:" at the head and tail of SIZE bytes.
998 (setq head-found (or (search-forward "coding:" head-end t)
999 (search-forward "unibyte:" head-end t)))
1000 (if (and head-found (> head-found tail-start))
1001 ;; Head and tail are overlapped.
1002 (setq tail-found head-found)
1003 (goto-char tail-start)
1004 (setq tail-found (or (search-forward "coding:" tail-end t)
1005 (search-forward "unibyte:" tail-end t))))
1007 ;; At first check the head.
1009 (goto-char head-start)
1010 (setq pos (re-search-forward "[\n\r]" head-end t))
1012 (= (char-after head-start) ?#)
1013 (= (char-after (1+ head-start)) ?!))
1014 ;; If the file begins with "#!" (exec interpreter magic),
1015 ;; look for coding frobs in the first two lines. You cannot
1016 ;; necessarily put them in the first line of such a file
1017 ;; without screwing up the interpreter invocation.
1018 (setq pos (search-forward "\n" head-end t)))
1019 (if pos (setq head-end pos))
1020 (when (< head-found head-end)
1021 (goto-char head-start)
1022 (when (and set-auto-coding-for-load
1024 "-\\*-\\(.*;\\)?[ \t]*unibyte:[ \t]*\\([^ ;]+\\)"
1026 (setq coding-system 'raw-text))
1027 (when (and (not coding-system)
1029 "-\\*-\\(.*;\\)?[ \t]*coding:[ \t]*\\([^ ;]+\\)"
1031 (setq coding-system (intern (match-string 2)))
1032 (or (coding-system-p coding-system)
1033 (setq coding-system nil)))))
1035 ;; If no coding: tag in the head, check the tail.
1036 (when (and tail-found (not coding-system))
1037 (goto-char tail-start)
1038 (search-forward "\n\^L" nil t)
1039 (if (re-search-forward
1040 "^\\(.*\\)[ \t]*Local Variables:[ \t]*\\(.*\\)$" tail-end t)
1041 ;; The prefix is what comes before "local variables:" in its
1042 ;; line. The suffix is what comes after "local variables:"
1044 (let* ((prefix (regexp-quote (match-string 1)))
1045 (suffix (regexp-quote (match-string 2)))
1049 "[ \t]*coding[ \t]*:[ \t]*\\([^ \t]+\\)[ \t]*"
1054 "[ \t]*unibyte[ \t]*:[ \t]*\\([^ \t]+\\)[ \t]*"
1057 (concat "^" prefix "[ \t]*end *:[ \t]*" suffix "$"))
1059 (re-search-forward re-end tail-end 'move)
1060 (setq tail-end (point))
1062 (when (and set-auto-coding-for-load
1063 (re-search-forward re-unibyte tail-end t))
1064 (setq coding-system 'raw-text))
1065 (when (and (not coding-system)
1066 (re-search-forward re-coding tail-end t))
1067 (setq coding-system (intern (match-string 1)))
1068 (or (coding-system-p coding-system)
1069 (setq coding-system nil))))))
1072 (setq set-auto-coding-function 'set-auto-coding)
1074 ;; Set buffer-file-coding-system of the current buffer after some text
1076 (defun after-insert-file-set-buffer-file-coding-system (inserted)
1077 (if last-coding-system-used
1078 (let ((coding-system
1079 (find-new-buffer-file-coding-system last-coding-system-used))
1080 (modified-p (buffer-modified-p)))
1082 (set-buffer-file-coding-system coding-system)
1083 (if (and enable-multibyte-characters
1084 (or (eq coding-system 'no-conversion)
1085 (eq (coding-system-type coding-system) 5))
1086 ;; If buffer was unmodified and the size is the
1087 ;; same as INSERTED, we must be visiting it.
1089 (= (buffer-size) inserted))
1090 ;; For coding systems no-conversion and raw-text...,
1091 ;; edit the buffer as unibyte.
1092 (let ((pos-byte (position-bytes (+ (point) inserted))))
1093 (set-buffer-multibyte nil)
1094 (setq inserted (- pos-byte (position-bytes (point))))))
1095 (set-buffer-modified-p modified-p))))
1098 (add-hook 'after-insert-file-functions
1099 'after-insert-file-set-buffer-file-coding-system)
1101 ;; The coding-spec and eol-type of coding-system returned is decided
1102 ;; independently in the following order.
1103 ;; 1. That of buffer-file-coding-system locally bound.
1104 ;; 2. That of CODING.
1106 (defun find-new-buffer-file-coding-system (coding)
1107 "Return a coding system for a buffer when a file of CODING is inserted.
1108 The local variable `buffer-file-coding-system' of the current buffer
1109 is set to the returned value.
1110 Return nil if there's no need to set `buffer-file-coding-system'."
1111 (let (local-coding local-eol
1112 found-coding found-eol
1115 ;; Nothing found about coding.
1118 ;; Get information of `buffer-file-coding-system' in LOCAL-EOL
1119 ;; and LOCAL-CODING.
1120 (setq local-eol (coding-system-eol-type buffer-file-coding-system))
1121 (if (null (numberp local-eol))
1122 ;; But eol-type is not yet set.
1123 (setq local-eol nil))
1124 (if (and buffer-file-coding-system
1125 (not (eq (coding-system-type buffer-file-coding-system) t)))
1126 ;; This is not `undecided'.
1127 (setq local-coding (coding-system-base buffer-file-coding-system)))
1129 (if (and (local-variable-p 'buffer-file-coding-system)
1130 local-eol local-coding)
1131 ;; The current buffer has already set full coding-system, we
1132 ;; had better not change it.
1135 (setq found-eol (coding-system-eol-type coding))
1136 (if (null (numberp found-eol))
1137 ;; But eol-type is not found.
1138 ;; If EOL conversions are inhibited, force unix eol-type.
1139 (setq found-eol (if inhibit-eol-conversion 0)))
1140 (if (eq (coding-system-type coding) t)
1141 (setq found-coding 'undecided)
1142 (setq found-coding (coding-system-base coding)))
1144 (if (and (not found-eol) (eq found-coding 'undecided))
1145 ;; No valid coding information found.
1148 ;; Some coding information (eol or text) found.
1150 ;; The local setting takes precedence over the found one.
1151 (setq new-coding (if (local-variable-p 'buffer-file-coding-system)
1152 (or local-coding found-coding)
1153 (or found-coding local-coding)))
1154 (setq new-eol (if (local-variable-p 'buffer-file-coding-system)
1155 (or local-eol found-eol)
1156 (or found-eol local-eol)))
1158 (let ((eol-type (coding-system-eol-type new-coding)))
1159 (if (and (numberp new-eol) (vectorp eol-type))
1160 (aref eol-type new-eol)
1163 (defun modify-coding-system-alist (target-type regexp coding-system)
1164 "Modify one of look up tables for finding a coding system on I/O operation.
1165 There are three of such tables, `file-coding-system-alist',
1166 `process-coding-system-alist', and `network-coding-system-alist'.
1168 TARGET-TYPE specifies which of them to modify.
1169 If it is `file', it affects `file-coding-system-alist' (which see).
1170 If it is `process', it affects `process-coding-system-alist' (which see).
1171 If it is `network', it affects `network-coding-system-alist' (which see).
1173 REGEXP is a regular expression matching a target of I/O operation.
1174 The target is a file name if TARGET-TYPE is `file', a program name if
1175 TARGET-TYPE is `process', or a network service name or a port number
1176 to connect to if TARGET-TYPE is `network'.
1178 CODING-SYSTEM is a coding system to perform code conversion on the I/O
1179 operation, or a cons cell (DECODING . ENCODING) specifying the coding systems
1180 for decoding and encoding respectively,
1181 or a function symbol which, when called, returns such a cons cell."
1182 (or (memq target-type '(file process network))
1183 (error "Invalid target type: %s" target-type))
1184 (or (stringp regexp)
1185 (and (eq target-type 'network) (integerp regexp))
1186 (error "Invalid regular expression: %s" regexp))
1187 (if (symbolp coding-system)
1188 (if (not (fboundp coding-system))
1190 (check-coding-system coding-system)
1191 (setq coding-system (cons coding-system coding-system))))
1192 (check-coding-system (car coding-system))
1193 (check-coding-system (cdr coding-system)))
1194 (cond ((eq target-type 'file)
1195 (let ((slot (assoc regexp file-coding-system-alist)))
1197 (setcdr slot coding-system)
1198 (setq file-coding-system-alist
1199 (cons (cons regexp coding-system)
1200 file-coding-system-alist)))))
1201 ((eq target-type 'process)
1202 (let ((slot (assoc regexp process-coding-system-alist)))
1204 (setcdr slot coding-system)
1205 (setq process-coding-system-alist
1206 (cons (cons regexp coding-system)
1207 process-coding-system-alist)))))
1209 (let ((slot (assoc regexp network-coding-system-alist)))
1211 (setcdr slot coding-system)
1212 (setq network-coding-system-alist
1213 (cons (cons regexp coding-system)
1214 network-coding-system-alist)))))))
1216 (defun make-translation-table (&rest args)
1217 "Make a translation table (char table) from arguments.
1218 Each argument is a list of the form (FROM . TO),
1219 where FROM is a character to be translated to TO.
1221 FROM can be a generic character (see `make-char'). In this case, TO is
1222 a generic character containing the same number of characters, or a
1223 ordinary character. If FROM and TO are both generic characters, all
1224 characters belonging to FROM are translated to characters belonging to TO
1225 without changing their position code(s)."
1226 (let ((table (make-char-table 'translation-table))
1229 (let ((elts (car args)))
1231 (let* ((from (car (car elts)))
1232 (from-i 0) ; degree of freedom of FROM
1233 (from-rev (nreverse (split-char from)))
1234 (to (cdr (car elts)))
1235 (to-i 0) ; degree of freedom of TO
1236 (to-rev (nreverse (split-char to))))
1237 ;; Check numbers of heading 0s in FROM-REV and TO-REV.
1238 (while (eq (car from-rev) 0)
1239 (setq from-i (1+ from-i) from-rev (cdr from-rev)))
1240 (while (eq (car to-rev) 0)
1241 (setq to-i (1+ to-i) to-rev (cdr to-rev)))
1242 (if (and (/= from-i to-i) (/= to-i 0))
1243 (error "Invalid character pair (%d . %d)" from to))
1244 ;; If we have already translated TO to TO-ALT, FROM should
1245 ;; also be translated to TO-ALT. But, this is only if TO
1246 ;; is a generic character or TO-ALT is not a generic
1248 (let ((to-alt (aref table to)))
1250 (or (> to-i 0) (not (generic-char-p to-alt))))
1253 (set-char-table-default table from to)
1254 (aset table from to))
1255 ;; If we have already translated some chars to FROM, they
1256 ;; should also be translated to TO.
1257 (let ((l (assq from revlist)))
1264 (setq l (cdr l)) ))))
1265 ;; Now update REVLIST.
1266 (let ((l (assq to revlist)))
1268 (setcdr l (cons from (cdr l)))
1269 (setq revlist (cons (list to from) revlist)))))
1270 (setq elts (cdr elts))))
1271 (setq args (cdr args)))
1272 ;; Return TABLE just created.
1275 (defun make-translation-table-from-vector (vec)
1276 "Make translation table from decoding vector VEC.
1277 VEC is an array of 256 elements to map unibyte codes to multibyte characters.
1278 See also the variable `nonascii-translation-table'."
1279 (let ((table (make-char-table 'translation-table))
1280 (rev-table (make-char-table 'translation-table))
1284 (setq ch (aref vec i))
1287 (aset rev-table ch i))
1289 (set-char-table-extra-slot table 0 rev-table)
1292 (defun define-translation-table (symbol &rest args)
1293 "Define SYMBOL as a name of translation table made by ARGS.
1295 If the first element of ARGS is a char-table of which purpose is
1296 translation-table, just define SYMBOL as the name of it.
1298 In the other case, ARGS are the same as arguments to the function
1299 `make-translation-table' (which see).
1301 This function sets properties `translation-table' and
1302 `translation-table-id' of SYMBOL to the created table itself and
1303 identification number of the table respectively."
1304 (let ((table (if (and (char-table-p (car args))
1305 (eq (char-table-subtype (car args))
1306 'translation-table))
1308 (apply 'make-translation-table args)))
1309 (len (length translation-table-vector))
1312 (put symbol 'translation-table table)
1315 (setq translation-table-vector
1316 (vconcat translation-table-vector (make-vector len nil))))
1317 (let ((slot (aref translation-table-vector id)))
1319 (eq (car slot) symbol))
1321 (aset translation-table-vector id (cons symbol table))
1323 (setq id (1+ id)))))
1324 (put symbol 'translation-table-id id)
1327 (put 'with-category-table 'lisp-indent-function 1)
1329 (defmacro with-category-table (category-table &rest body)
1330 `(let ((current-category-table (category-table)))
1331 (set-category-table ,category-table)
1334 (set-category-table current-category-table))))
1336 ;;; Initialize some variables.
1338 (put 'use-default-ascent 'char-table-extra-slots 0)
1339 (setq use-default-ascent (make-char-table 'use-default-ascent))
1340 (put 'ignore-relative-composition 'char-table-extra-slots 0)
1341 (setq ignore-relative-composition
1342 (make-char-table 'ignore-relative-composition))
1347 ;;; mule.el ends here