1 ;;; mule.el --- basic commands for mulitilingual environment
3 ;; Copyright (C) 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005
4 ;; Free Software Foundation, Inc.
5 ;; Copyright (C) 1995, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005
6 ;; National Institute of Advanced Industrial Science and Technology (AIST)
7 ;; Registration Number H14PRO021
9 ;; Keywords: mule, multilingual, character set, coding system
11 ;; This file is part of GNU Emacs.
13 ;; GNU Emacs is free software; you can redistribute it and/or modify
14 ;; it under the terms of the GNU General Public License as published by
15 ;; the Free Software Foundation; either version 2, or (at your option)
18 ;; GNU Emacs is distributed in the hope that it will be useful,
19 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
20 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
21 ;; GNU General Public License for more details.
23 ;; You should have received a copy of the GNU General Public License
24 ;; along with GNU Emacs; see the file COPYING. If not, write to the
25 ;; Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
26 ;; Boston, MA 02110-1301, USA.
32 (defconst mule-version "5.0 (SAKAKI)" "\
33 Version number and name of this version of MULE (multilingual environment).")
35 (defconst mule-version-date "1999.12.7" "\
36 Distribution date of this version of MULE (multilingual environment).")
38 (defun load-with-code-conversion (fullname file &optional noerror nomessage)
39 "Execute a file of Lisp code named FILE whose absolute name is FULLNAME.
40 The file contents are decoded before evaluation if necessary.
41 If optional second arg NOERROR is non-nil,
42 report no error if FILE doesn't exist.
43 Print messages at start and end of loading unless
44 optional third arg NOMESSAGE is non-nil.
45 Return t if file exists."
46 (if (null (file-readable-p fullname))
48 (signal 'file-error (list "Cannot open load file" file)))
49 ;; Read file with code conversion, and then eval.
51 ;; To avoid any autoloading, set default-major-mode to
53 ;; So that we don't get completely screwed if the
54 ;; file is encoded in some complicated character set,
55 ;; read it with real decoding, as a multibyte buffer,
56 ;; even if this is a --unibyte Emacs session.
57 (let ((default-major-mode 'fundamental-mode)
58 (default-enable-multibyte-characters t))
59 ;; We can't use `generate-new-buffer' because files.el
61 (get-buffer-create (generate-new-buffer-name " *load*"))))
63 (source (save-match-data (string-match "\\.el\\'" fullname))))
66 (message "Loading %s (source)..." file)
67 (message "Loading %s..." file)))
69 (push file preloaded-file-list))
71 (let ((load-file-name fullname)
72 (set-auto-coding-for-load t)
73 (inhibit-file-name-operation nil))
76 ;; Don't let deactivate-mark remain set.
77 (let (deactivate-mark)
78 (insert-file-contents fullname))
79 ;; If the loaded file was inserted with no-conversion or
80 ;; raw-text coding system, make the buffer unibyte.
81 ;; Otherwise, eval-buffer might try to interpret random
82 ;; binary junk as multibyte characters.
83 (if (and enable-multibyte-characters
84 (or (eq (coding-system-type last-coding-system-used) 5)
85 (eq last-coding-system-used 'no-conversion)))
86 (set-buffer-multibyte nil))
87 ;; Make `kill-buffer' quiet.
88 (set-buffer-modified-p nil))
89 ;; Have the original buffer current while we eval.
90 (eval-buffer buffer nil
91 ;; This is compatible with what `load' does.
92 (if purify-flag file fullname)
93 ;; If this Emacs is running with --unibyte,
94 ;; convert multibyte strings to unibyte
95 ;; after reading them.
96 ;; (not default-enable-multibyte-characters)
99 (let (kill-buffer-hook kill-buffer-query-functions)
100 (kill-buffer buffer)))
101 (let ((hook (assoc file after-load-alist)))
103 (mapcar (function eval) (cdr hook))))
104 (unless (or nomessage noninteractive)
106 (message "Loading %s (source)...done" file)
107 (message "Loading %s...done" file)))
110 ;; API (Application Program Interface) for charsets.
112 (defsubst charset-quoted-standard-p (obj)
113 "Return t if OBJ is a quoted symbol, and is the name of a standard charset."
114 (and (listp obj) (eq (car obj) 'quote)
115 (symbolp (car-safe (cdr obj)))
116 (let ((vector (get (car-safe (cdr obj)) 'charset)))
117 (and (vectorp vector)
118 (< (aref vector 0) 160)))))
120 (defsubst charsetp (object)
121 "T if OBJECT is a charset."
122 (and (symbolp object) (vectorp (get object 'charset))))
124 (defsubst charset-info (charset)
125 "Return a vector of information of CHARSET.
126 The elements of the vector are:
127 CHARSET-ID, BYTES, DIMENSION, CHARS, WIDTH, DIRECTION,
128 LEADING-CODE-BASE, LEADING-CODE-EXT,
129 ISO-FINAL-CHAR, ISO-GRAPHIC-PLANE,
130 REVERSE-CHARSET, SHORT-NAME, LONG-NAME, DESCRIPTION,
133 CHARSET-ID (integer) is the identification number of the charset.
134 BYTES (integer) is the length of multi-byte form of a character in
135 the charset: one of 1, 2, 3, and 4.
136 DIMENSION (integer) is the number of bytes to represent a character of
138 CHARS (integer) is the number of characters in a dimension: 94 or 96.
139 WIDTH (integer) is the number of columns a character in the charset
140 occupies on the screen: one of 0, 1, and 2.
141 DIRECTION (integer) is the rendering direction of characters in the
142 charset when rendering. If 0, render from left to right, else
143 render from right to left.
144 LEADING-CODE-BASE (integer) is the base leading-code for the
146 LEADING-CODE-EXT (integer) is the extended leading-code for the
147 charset. All charsets of less than 0xA0 has the value 0.
148 ISO-FINAL-CHAR (character) is the final character of the
149 corresponding ISO 2022 charset. If the charset is not assigned
150 any final character, the value is -1.
151 ISO-GRAPHIC-PLANE (integer) is the graphic plane to be invoked
152 while encoding to variants of ISO 2022 coding system, one of the
153 following: 0/graphic-plane-left(GL), 1/graphic-plane-right(GR).
154 If the charset is not assigned any final character, the value is -1.
155 REVERSE-CHARSET (integer) is the charset which differs only in
156 LEFT-TO-RIGHT value from the charset. If there's no such a
157 charset, the value is -1.
158 SHORT-NAME (string) is the short name to refer to the charset.
159 LONG-NAME (string) is the long name to refer to the charset
160 DESCRIPTION (string) is the description string of the charset.
161 PLIST (property list) may contain any type of information a user
162 want to put and get by functions `put-charset-property' and
163 `get-charset-property' respectively."
164 (get charset 'charset))
166 ;; It is better not to use backquote in this file,
167 ;; because that makes a bootstrapping problem
168 ;; if you need to recompile all the Lisp files using interpreted code.
170 (defmacro charset-id (charset)
171 "Return charset identification number of CHARSET."
172 (if (charset-quoted-standard-p charset)
173 (aref (charset-info (nth 1 charset)) 0)
174 (list 'aref (list 'charset-info charset) 0)))
176 (defmacro charset-bytes (charset)
177 "Return bytes of CHARSET.
178 See the function `charset-info' for more detail."
179 (if (charset-quoted-standard-p charset)
180 (aref (charset-info (nth 1 charset)) 1)
181 (list 'aref (list 'charset-info charset) 1)))
183 (defmacro charset-dimension (charset)
184 "Return dimension of CHARSET.
185 See the function `charset-info' for more detail."
186 (if (charset-quoted-standard-p charset)
187 (aref (charset-info (nth 1 charset)) 2)
188 (list 'aref (list 'charset-info charset) 2)))
190 (defmacro charset-chars (charset)
191 "Return character numbers contained in a dimension of CHARSET.
192 See the function `charset-info' for more detail."
193 (if (charset-quoted-standard-p charset)
194 (aref (charset-info (nth 1 charset)) 3)
195 (list 'aref (list 'charset-info charset) 3)))
197 (defmacro charset-width (charset)
198 "Return width (how many column occupied on a screen) of CHARSET.
199 See the function `charset-info' for more detail."
200 (if (charset-quoted-standard-p charset)
201 (aref (charset-info (nth 1 charset)) 4)
202 (list 'aref (list 'charset-info charset) 4)))
204 (defmacro charset-direction (charset)
205 "Return direction of CHARSET.
206 See the function `charset-info' for more detail."
207 (if (charset-quoted-standard-p charset)
208 (aref (charset-info (nth 1 charset)) 5)
209 (list 'aref (list 'charset-info charset) 5)))
211 (defmacro charset-iso-final-char (charset)
212 "Return final char of CHARSET.
213 See the function `charset-info' for more detail."
214 (if (charset-quoted-standard-p charset)
215 (aref (charset-info (nth 1 charset)) 8)
216 (list 'aref (list 'charset-info charset) 8)))
218 (defmacro charset-iso-graphic-plane (charset)
219 "Return graphic plane of CHARSET.
220 See the function `charset-info' for more detail."
221 (if (charset-quoted-standard-p charset)
222 (aref (charset-info (nth 1 charset)) 9)
223 (list 'aref (list 'charset-info charset) 9)))
225 (defmacro charset-reverse-charset (charset)
226 "Return reverse charset of CHARSET.
227 See the function `charset-info' for more detail."
228 (if (charset-quoted-standard-p charset)
229 (aref (charset-info (nth 1 charset)) 10)
230 (list 'aref (list 'charset-info charset) 10)))
232 (defmacro charset-short-name (charset)
233 "Return short name of CHARSET.
234 See the function `charset-info' for more detail."
235 (if (charset-quoted-standard-p charset)
236 (aref (charset-info (nth 1 charset)) 11)
237 (list 'aref (list 'charset-info charset) 11)))
239 (defmacro charset-long-name (charset)
240 "Return long name of CHARSET.
241 See the function `charset-info' for more detail."
242 (if (charset-quoted-standard-p charset)
243 (aref (charset-info (nth 1 charset)) 12)
244 (list 'aref (list 'charset-info charset) 12)))
246 (defmacro charset-description (charset)
247 "Return description of CHARSET.
248 See the function `charset-info' for more detail."
249 (if (charset-quoted-standard-p charset)
250 (aref (charset-info (nth 1 charset)) 13)
251 (list 'aref (list 'charset-info charset) 13)))
253 (defmacro charset-plist (charset)
254 "Return list charset property of CHARSET.
255 See the function `charset-info' for more detail."
257 (if (charset-quoted-standard-p charset)
258 (charset-info (nth 1 charset))
259 (list 'charset-info charset))
262 (defun set-charset-plist (charset plist)
263 "Set CHARSET's property list to PLIST, and return PLIST."
264 (aset (charset-info charset) 14 plist))
266 (defun make-char (charset &optional code1 code2)
267 "Return a character of CHARSET whose position codes are CODE1 and CODE2.
268 CODE1 and CODE2 are optional, but if you don't supply
269 sufficient position codes, return a generic character which stands for
270 all characters or group of characters in the character set.
271 A generic character can be used to index a char table (e.g. syntax-table).
273 Such character sets as ascii, eight-bit-control, and eight-bit-graphic
274 don't have corresponding generic characters. If CHARSET is one of
275 them and you don't supply CODE1, return the character of the smallest
278 If CODE1 or CODE2 are invalid (out of range), this function signals an
279 error. However, the eighth bit of both CODE1 and CODE2 is zeroed
280 before they are used to index CHARSET. Thus you may use, say, the
281 actual ISO 8859 character code rather than subtracting 128, as you
282 would need to index the corresponding Emacs charset."
283 (make-char-internal (charset-id charset) code1 code2))
285 (put 'make-char 'byte-compile
288 (let ((charset (nth 1 form)))
289 (if (charset-quoted-standard-p charset)
290 (byte-compile-normal-call
291 (cons 'make-char-internal
292 (cons (charset-id (nth 1 charset)) (nthcdr 2 form))))
293 (byte-compile-normal-call
294 (cons 'make-char-internal
295 (cons (list 'charset-id charset) (nthcdr 2 form)))))))))
297 (defun charset-list ()
298 "Return list of charsets ever defined.
300 This function is provided for backward compatibility.
301 Now we have the variable `charset-list'."
304 (defsubst generic-char-p (char)
305 "Return t if and only if CHAR is a generic character.
306 See also the documentation of `make-char'."
308 (let ((l (split-char char)))
309 (and (or (= (nth 1 l) 0) (eq (nth 2 l) 0))
310 (not (eq (car l) 'composition))))))
312 (defun decode-char (ccs code-point &optional restriction)
313 "Return character specified by coded character set CCS and CODE-POINT in it.
314 Return nil if such a character is not supported.
315 Currently the only supported coded character set is `ucs' (ISO/IEC
316 10646: Universal Multi-Octet Coded Character Set), and the result is
317 translated through the translation-table named
318 `utf-translation-table-for-decode', or through the
319 translation-hash-table named `utf-subst-table-for-decode'
320 \(if `utf-translate-cjk-mode' is non-nil).
322 Optional argument RESTRICTION specifies a way to map the pair of CCS
323 and CODE-POINT to a character. Currently not supported and just ignored."
326 (or (and utf-translate-cjk-mode
327 (utf-lookup-subst-table-for-decode code-point))
332 (make-char 'latin-iso8859-1 code-point))
333 ((< code-point #x2500)
334 (setq code-point (- code-point #x0100))
335 (make-char 'mule-unicode-0100-24ff
336 (+ (/ code-point 96) 32) (+ (% code-point 96) 32)))
337 ((< code-point #x3400)
338 (setq code-point (- code-point #x2500))
339 (make-char 'mule-unicode-2500-33ff
340 (+ (/ code-point 96) 32) (+ (% code-point 96) 32)))
341 ((and (>= code-point #xe000) (< code-point #x10000))
342 (setq code-point (- code-point #xe000))
343 (make-char 'mule-unicode-e000-ffff
344 (+ (/ code-point 96) 32)
345 (+ (% code-point 96) 32))))))
347 (or (aref (get 'utf-translation-table-for-decode
348 'translation-table) c)
351 (defun encode-char (char ccs &optional restriction)
352 "Return code-point in coded character set CCS that corresponds to CHAR.
353 Return nil if CHAR is not included in CCS.
354 Currently the only supported coded character set is `ucs' (ISO/IEC
355 10646: Universal Multi-Octet Coded Character Set), and CHAR is first
356 translated through the translation-table named
357 `utf-translation-table-for-encode', or through the
358 translation-hash-table named `utf-subst-table-for-encode' \(if
359 `utf-translate-cjk-mode' is non-nil).
361 CHAR should be in one of these charsets:
362 ascii, latin-iso8859-1, mule-unicode-0100-24ff, mule-unicode-2500-33ff,
363 mule-unicode-e000-ffff, eight-bit-control
364 Otherwise, return nil.
366 Optional argument RESTRICTION specifies a way to map CHAR to a
367 code-point in CCS. Currently not supported and just ignored."
368 (let* ((split (split-char char))
369 (charset (car split))
372 (or (and utf-translate-cjk-mode
373 (utf-lookup-subst-table-for-encode char))
374 (let ((table (get 'utf-translation-table-for-encode
375 'translation-table)))
376 (setq trans (aref table char))
378 (setq split (split-char trans)
379 charset (car split)))
380 (cond ((eq charset 'ascii)
382 ((eq charset 'latin-iso8859-1)
383 (+ (nth 1 split) 128))
384 ((eq charset 'mule-unicode-0100-24ff)
385 (+ #x0100 (+ (* (- (nth 1 split) 32) 96)
386 (- (nth 2 split) 32))))
387 ((eq charset 'mule-unicode-2500-33ff)
388 (+ #x2500 (+ (* (- (nth 1 split) 32) 96)
389 (- (nth 2 split) 32))))
390 ((eq charset 'mule-unicode-e000-ffff)
391 (+ #xe000 (+ (* (- (nth 1 split) 32) 96)
392 (- (nth 2 split) 32))))
393 ((eq charset 'eight-bit-control)
397 ;; Coding system stuff
399 ;; Coding system is a symbol that has the property `coding-system'.
401 ;; The value of the property `coding-system' is a vector of the
403 ;; [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
404 ;; We call this vector as coding-spec. See comments in src/coding.c
407 (defconst coding-spec-type-idx 0)
408 (defconst coding-spec-mnemonic-idx 1)
409 (defconst coding-spec-doc-string-idx 2)
410 (defconst coding-spec-plist-idx 3)
411 (defconst coding-spec-flags-idx 4)
413 ;; PLIST is a property list of a coding system. To share PLIST among
414 ;; alias coding systems, a coding system has PLIST in coding-spec
415 ;; instead of having it in normal property list of Lisp symbol.
416 ;; Here's a list of coding system properties currently being used.
420 ;; The value is a coding category the coding system belongs to. The
421 ;; function `make-coding-system' sets this value automatically
422 ;; unless its argument PROPERTIES specifies this property.
424 ;; o alias-coding-systems
426 ;; The value is a list of coding systems of the same alias group. The
427 ;; first element is the coding system made at first, which we call as
428 ;; `base coding system'. The function `make-coding-system' sets this
429 ;; value automatically and `define-coding-system-alias' updates it.
431 ;; See the documentation of make-coding-system for the meanings of the
432 ;; following properties.
434 ;; o post-read-conversion
435 ;; o pre-write-conversion
436 ;; o translation-table-for-decode
437 ;; o translation-table-for-encode
441 ;; o valid-codes (meaningful only for a coding system based on CCL)
444 (defsubst coding-system-spec (coding-system)
445 "Return coding-spec of CODING-SYSTEM."
446 (get (check-coding-system coding-system) 'coding-system))
448 (defun coding-system-type (coding-system)
449 "Return the coding type of CODING-SYSTEM.
450 A coding type is an integer value indicating the encoding method
451 of CODING-SYSTEM. See the function `make-coding-system' for more detail."
452 (aref (coding-system-spec coding-system) coding-spec-type-idx))
454 (defun coding-system-mnemonic (coding-system)
455 "Return the mnemonic character of CODING-SYSTEM.
456 The mnemonic character of a coding system is used in mode line
457 to indicate the coding system. If the arg is nil, return ?-."
458 (let ((spec (coding-system-spec coding-system)))
459 (if spec (aref spec coding-spec-mnemonic-idx) ?-)))
461 (defun coding-system-doc-string (coding-system)
462 "Return the documentation string for CODING-SYSTEM."
463 (aref (coding-system-spec coding-system) coding-spec-doc-string-idx))
465 (defun coding-system-plist (coding-system)
466 "Return the property list of CODING-SYSTEM."
467 (aref (coding-system-spec coding-system) coding-spec-plist-idx))
469 (defun coding-system-flags (coding-system)
470 "Return `flags' of CODING-SYSTEM.
471 A `flags' of a coding system is a vector of length 32 indicating detailed
472 information of a coding system. See the function `make-coding-system'
474 (aref (coding-system-spec coding-system) coding-spec-flags-idx))
476 (defun coding-system-get (coding-system prop)
477 "Extract a value from CODING-SYSTEM's property list for property PROP."
478 (plist-get (coding-system-plist coding-system) prop))
480 (defun coding-system-put (coding-system prop val)
481 "Change value in CODING-SYSTEM's property list PROP to VAL."
482 (let ((plist (coding-system-plist coding-system)))
484 (plist-put plist prop val)
485 (aset (coding-system-spec coding-system) coding-spec-plist-idx
488 (defun coding-system-category (coding-system)
489 "Return the coding category of CODING-SYSTEM.
490 See also `coding-category-list'."
491 (coding-system-get coding-system 'coding-category))
493 (defun coding-system-base (coding-system)
494 "Return the base coding system of CODING-SYSTEM.
495 A base coding system is what made by `make-coding-system'.
496 Any alias nor subsidiary coding systems are not base coding system."
497 (car (coding-system-get coding-system 'alias-coding-systems)))
499 ;; Coding system also has a property `eol-type'.
501 ;; This property indicates how the coding system handles end-of-line
502 ;; format. The value is integer 0, 1, 2, or a vector of three coding
503 ;; systems. Each integer value 0, 1, and 2 indicates the format of
504 ;; end-of-line LF, CRLF, and CR respectively. A vector value
505 ;; indicates that the format of end-of-line should be detected
506 ;; automatically. Nth element of the vector is the subsidiary coding
507 ;; system whose `eol-type' property is N.
509 (defun coding-system-eol-type (coding-system)
510 "Return eol-type of CODING-SYSTEM.
511 An eol-type is integer 0, 1, 2, or a vector of coding systems.
513 Integer values 0, 1, and 2 indicate a format of end-of-line; LF,
514 CRLF, and CR respectively.
516 A vector value indicates that a format of end-of-line should be
517 detected automatically. Nth element of the vector is the subsidiary
518 coding system whose eol-type is N."
519 (get coding-system 'eol-type))
521 (defun coding-system-eol-type-mnemonic (coding-system)
522 "Return the string indicating end-of-line format of CODING-SYSTEM."
523 (let* ((eol-type (coding-system-eol-type coding-system))
524 (val (cond ((eq eol-type 0) eol-mnemonic-unix)
525 ((eq eol-type 1) eol-mnemonic-dos)
526 ((eq eol-type 2) eol-mnemonic-mac)
527 (t eol-mnemonic-undecided))))
530 (char-to-string val))))
532 (defun coding-system-lessp (x y)
533 (cond ((eq x 'no-conversion) t)
534 ((eq y 'no-conversion) nil)
535 ((eq x 'emacs-mule) t)
536 ((eq y 'emacs-mule) nil)
537 ((eq x 'undecided) t)
538 ((eq y 'undecided) nil)
539 (t (let ((c1 (coding-system-mnemonic x))
540 (c2 (coding-system-mnemonic y)))
541 (or (< (downcase c1) (downcase c2))
542 (and (not (> (downcase c1) (downcase c2)))
545 (defun coding-system-equal (coding-system-1 coding-system-2)
546 "Return t if and only if CODING-SYSTEM-1 and CODING-SYSTEM-2 are identical.
547 Two coding systems are identical if two symbols are equal
548 or one is an alias of the other."
549 (or (eq coding-system-1 coding-system-2)
550 (and (equal (coding-system-spec coding-system-1)
551 (coding-system-spec coding-system-2))
552 (let ((eol-type-1 (coding-system-eol-type coding-system-1))
553 (eol-type-2 (coding-system-eol-type coding-system-2)))
554 (or (eq eol-type-1 eol-type-2)
555 (and (vectorp eol-type-1) (vectorp eol-type-2)))))))
557 (defun add-to-coding-system-list (coding-system)
558 "Add CODING-SYSTEM to `coding-system-list' while keeping it sorted."
559 (if (or (null coding-system-list)
560 (coding-system-lessp coding-system (car coding-system-list)))
561 (setq coding-system-list (cons coding-system coding-system-list))
562 (let ((len (length coding-system-list))
563 mid (tem coding-system-list))
565 (setq mid (nthcdr (/ len 2) tem))
566 (if (coding-system-lessp (car mid) coding-system)
568 len (- len (/ len 2)))
569 (setq len (/ len 2))))
570 (setcdr tem (cons coding-system (cdr tem))))))
572 (defun coding-system-list (&optional base-only)
573 "Return a list of all existing non-subsidiary coding systems.
574 If optional arg BASE-ONLY is non-nil, only base coding systems are listed.
575 The value doesn't include subsidiary coding systems which are what
576 made from bases and aliases automatically for various end-of-line
577 formats (e.g. iso-latin-1-unix, koi8-r-dos)."
578 (let* ((codings (copy-sequence coding-system-list))
579 (tail (cons nil codings)))
580 ;; Remove subsidiary coding systems (eol variants) and alias
581 ;; coding systems (if necessary).
583 (let* ((coding (car (cdr tail)))
584 (aliases (coding-system-get coding 'alias-coding-systems)))
586 ;; CODING is an eol variant if not in ALIASES.
587 (not (memq coding aliases))
588 ;; CODING is an alias if it is not car of ALIASES.
589 (and base-only (not (eq coding (car aliases)))))
590 (setcdr tail (cdr (cdr tail)))
591 (setq tail (cdr tail)))))
594 (defun map-charset-chars (func charset)
595 "Use FUNC to map over all characters in CHARSET for side effects.
596 FUNC is a function of two args, the start and end (inclusive) of a
597 character code range. Thus FUNC should iterate over [START, END]."
598 (let* ((dim (charset-dimension charset))
599 (chars (charset-chars charset))
600 (start (if (= chars 94)
605 (make-char charset start)
606 (make-char charset (+ start chars -1)))
609 (make-char charset (+ i start) start)
610 (make-char charset (+ i start) (+ start chars -1)))))))
612 (defalias 'register-char-codings 'ignore "")
613 (make-obsolete 'register-char-codings
614 "it exists just for backward compatibility, and does nothing."
617 (defconst char-coding-system-table nil
618 "This is an obsolete variable.
619 It exists just for backward compatibility, and the value is always nil.")
621 (defun make-subsidiary-coding-system (coding-system)
622 "Make subsidiary coding systems (eol-type variants) of CODING-SYSTEM."
623 (let ((coding-spec (coding-system-spec coding-system))
624 (subsidiaries (vector (intern (format "%s-unix" coding-system))
625 (intern (format "%s-dos" coding-system))
626 (intern (format "%s-mac" coding-system))))
629 (setq elt (aref subsidiaries i))
630 (put elt 'coding-system coding-spec)
631 (put elt 'eol-type i)
632 (put elt 'coding-system-define-form nil)
633 (add-to-coding-system-list elt)
634 (or (assoc (symbol-name elt) coding-system-alist)
635 (setq coding-system-alist
636 (cons (list (symbol-name elt)) coding-system-alist))))
639 (defun transform-make-coding-system-args (name type &optional doc-string props)
640 "For internal use only.
641 Transform XEmacs style args for `make-coding-system' to Emacs style.
642 Value is a list of transformed arguments."
643 (let ((mnemonic (string-to-char (or (plist-get props 'mnemonic) "?")))
644 (eol-type (plist-get props 'eol-type))
647 ((eq eol-type 'lf) (setq eol-type 'unix))
648 ((eq eol-type 'crlf) (setq eol-type 'dos))
649 ((eq eol-type 'cr) (setq eol-type 'mac)))
650 (if (setq tmp (plist-get props 'post-read-conversion))
651 (setq properties (plist-put properties 'post-read-conversion tmp)))
652 (if (setq tmp (plist-get props 'pre-write-conversion))
653 (setq properties (plist-put properties 'pre-write-conversion tmp)))
655 ((eq type 'shift-jis)
656 `(,name 1 ,mnemonic ,doc-string () ,properties ,eol-type))
657 ((eq type 'iso2022) ; This is not perfect.
658 (if (plist-get props 'escape-quoted)
659 (error "escape-quoted is not supported: %S"
660 `(,name ,type ,doc-string ,props)))
661 (let ((g0 (plist-get props 'charset-g0))
662 (g1 (plist-get props 'charset-g1))
663 (g2 (plist-get props 'charset-g2))
664 (g3 (plist-get props 'charset-g3))
667 (eq (cadr (assoc 'latin-jisx0201
668 (plist-get props 'input-charset-conversion)))
670 (eq (cadr (assoc 'ascii
671 (plist-get props 'output-charset-conversion)))
675 (eq (cadr (assoc 'japanese-jisx0208-1978
676 (plist-get props 'input-charset-conversion)))
678 (eq (cadr (assoc 'japanese-jisx0208
679 (plist-get props 'output-charset-conversion)))
680 'japanese-jisx0208-1978))))
682 (if (plist-get props 'force-g0-on-output)
686 (if (plist-get props 'force-g1-on-output)
690 (if (plist-get props 'force-g2-on-output)
694 (if (plist-get props 'force-g3-on-output)
697 `(,name 2 ,mnemonic ,doc-string
699 ,(plist-get props 'short)
700 ,(not (plist-get props 'no-ascii-eol))
701 ,(not (plist-get props 'no-ascii-cntl))
702 ,(plist-get props 'seven)
704 ,(not (plist-get props 'lock-shift))
707 ,(plist-get props 'no-iso6429)
709 ,properties ,eol-type)))
711 `(,name 3 ,mnemonic ,doc-string () ,properties ,eol-type))
713 `(,name 4 ,mnemonic ,doc-string
714 (,(plist-get props 'decode) . ,(plist-get props 'encode))
715 ,properties ,eol-type))
717 (error "unsupported XEmacs style make-coding-style arguments: %S"
718 `(,name ,type ,doc-string ,props))))))
720 (defun make-coding-system (coding-system type mnemonic doc-string
725 "Define a new coding system CODING-SYSTEM (symbol).
726 Remaining arguments are TYPE, MNEMONIC, DOC-STRING, FLAGS (optional),
727 and PROPERTIES (optional) which construct a coding-spec of CODING-SYSTEM
728 in the following format:
729 [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
731 TYPE is an integer value indicating the type of the coding system as follows:
732 0: Emacs internal format,
733 1: Shift-JIS (or MS-Kanji) used mainly on Japanese PCs,
734 2: ISO-2022 including many variants,
735 3: Big5 used mainly on Chinese PCs,
736 4: private, CCL programs provide encoding/decoding algorithm,
737 5: Raw-text, which means that text contains random 8-bit codes.
739 MNEMONIC is a character to be displayed on mode line for the coding system.
741 DOC-STRING is a documentation string for the coding system.
743 FLAGS specifies more detailed information of the coding system as follows:
745 If TYPE is 2 (ISO-2022), FLAGS is a list of these elements:
746 CHARSET0, CHARSET1, CHARSET2, CHARSET3, SHORT-FORM,
747 ASCII-EOL, ASCII-CNTL, SEVEN, LOCKING-SHIFT, SINGLE-SHIFT,
748 USE-ROMAN, USE-OLDJIS, NO-ISO6429, INIT-BOL, DESIGNATION-BOL,
749 SAFE, ACCEPT-LATIN-EXTRA-CODE.
750 CHARSETn are character sets initially designated to Gn graphic registers.
751 If CHARSETn is nil, Gn is never used.
752 If CHARSETn is t, Gn can be used but nothing designated initially.
753 If CHARSETn is a list of character sets, those character sets are
754 designated to Gn on output, but nothing designated to Gn initially.
755 But, character set `ascii' can be designated only to G0.
756 SHORT-FORM non-nil means use short designation sequence on output.
757 ASCII-EOL non-nil means designate ASCII to g0 at end of line on output.
758 ASCII-CNTL non-nil means designate ASCII to g0 before control codes and
760 SEVEN non-nil means use 7-bit code only on output.
761 LOCKING-SHIFT non-nil means use locking-shift.
762 SINGLE-SHIFT non-nil means use single-shift.
763 USE-ROMAN non-nil means designate JIS0201-1976-Roman instead of ASCII.
764 USE-OLDJIS non-nil means designate JIS0208-1976 instead of JIS0208-1983.
765 NO-ISO6429 non-nil means not use ISO6429's direction specification.
766 INIT-BOL non-nil means any designation state is assumed to be reset
767 to initial at each beginning of line on output.
768 DESIGNATION-BOL non-nil means designation sequences should be placed
769 at beginning of line on output.
770 SAFE non-nil means convert unsafe characters to `?' on output.
771 Characters not specified in the property `safe-charsets' nor
772 `safe-chars' are unsafe.
773 ACCEPT-LATIN-EXTRA-CODE non-nil means code-detection routine accepts
774 a code specified in `latin-extra-code-table' (which see) as a valid
775 code of the coding system.
777 If TYPE is 4 (private), FLAGS should be a cons of CCL programs, for
778 decoding and encoding. CCL programs should be specified by their
781 PROPERTIES is an alist of properties vs the corresponding values. The
782 following properties are recognized:
784 o post-read-conversion
786 The value is a function to call after some text is inserted and
787 decoded by the coding system itself and before any functions in
788 `after-insert-functions' are called. The argument of this
789 function is the same as for a function in
790 `after-insert-file-functions', i.e. LENGTH of the text inserted,
791 with point at the head of the text to be decoded.
793 o pre-write-conversion
795 The value is a function to call after all functions in
796 `write-region-annotate-functions' and `buffer-file-format' are
797 called, and before the text is encoded by the coding system itself.
798 The arguments to this function are the same as those of a function
799 in `write-region-annotate-functions', i.e. FROM and TO, specifying
802 o translation-table-for-decode
804 The value is a translation table to be applied on decoding. See
805 the function `make-translation-table' for the format of translation
806 table. This is not applicable to type 4 (CCL-based) coding systems.
808 o translation-table-for-encode
810 The value is a translation table to be applied on encoding. This is
811 not applicable to type 4 (CCL-based) coding systems.
815 The value is a char table. If a character has non-nil value in it,
816 the character is safely supported by the coding system. This
817 overrides the specification of safe-charsets.
821 The value is a list of charsets safely supported by the coding
822 system. The value t means that all charsets Emacs handles are
823 supported. Even if some charset is not in this list, it doesn't
824 mean that the charset can't be encoded in the coding system;
825 it just means that some other receiver of text encoded
826 in the coding system won't be able to handle that charset.
830 The value is a symbol whose name is the `MIME-charset' parameter of
833 o mime-text-unsuitable
835 A non-nil value means the `mime-charset' property names a charset
836 which is unsuitable for the top-level media type \"text\".
838 o valid-codes (meaningful only for a coding system based on CCL)
840 The value is a list to indicate valid byte ranges of the encoded
841 file. Each element of the list is an integer or a cons of integer.
842 In the former case, the integer value is a valid byte code. In the
843 latter case, the integers specify the range of valid byte codes.
845 o composition (meaningful only when TYPE is 0 or 2)
847 If the value is non-nil, the coding system preserves composition
852 If the value is non-nil, the coding system is not compatible
853 with ASCII, which means it encodes or decodes ASCII character
854 string to the different byte sequence.
856 These properties are set in PLIST, a property list. This function
857 also sets properties `coding-category' and `alias-coding-systems'
860 EOL-TYPE specifies the EOL type of the coding-system in one of the
863 o symbol (unix, dos, or mac)
865 The symbol `unix' means Unix-like EOL (LF), `dos' means
866 DOS-like EOL (CRLF), and `mac' means MAC-like EOL (CR).
868 o number (0, 1, or 2)
870 The number 0, 1, and 2 mean UNIX, DOS, and MAC-like EOL
873 o vector of coding-systems of length 3
875 The EOL type is detected automatically for the coding system.
876 And, according to the detected EOL type, one of the coding
877 systems in the vector is selected. Elements of the vector
878 corresponds to Unix-like EOL, DOS-like EOL, and Mac-like EOL
881 Kludgy features for backward compatibility:
883 1. If TYPE is 4 and car or cdr of FLAGS is a vector, the vector is
884 treated as a compiled CCL code.
886 2. If PROPERTIES is just a list of character sets, the list is set as
887 a value of `safe-charsets' in PLIST."
889 ;; For compatiblity with XEmacs, we check the type of TYPE. If it
890 ;; is a symbol, perhaps, this function is called with XEmacs-style
891 ;; arguments. Here, try to transform that kind of arguments to
894 (let ((args (transform-make-coding-system-args coding-system type
895 mnemonic doc-string)))
896 (setq coding-system (car args)
898 mnemonic (nth 2 args)
899 doc-string (nth 3 args)
901 properties (nth 5 args)
902 eol-type (nth 6 args))))
904 ;; Set a value of `coding-system' property.
905 (let ((coding-spec (make-vector 5 nil))
906 (no-initial-designation t)
907 (no-alternative-designation t)
908 (accept-latin-extra-code nil)
910 (if (or (not (integerp type)) (< type 0) (> type 5))
911 (error "TYPE argument must be 0..5"))
912 (if (or (not (integerp mnemonic)) (<= mnemonic ? ) (> mnemonic 127))
913 (error "MNEMONIC argument must be an ASCII printable character"))
914 (aset coding-spec coding-spec-type-idx type)
915 (aset coding-spec coding-spec-mnemonic-idx mnemonic)
916 (aset coding-spec coding-spec-doc-string-idx
917 (purecopy (if (stringp doc-string) doc-string "")))
919 (setq coding-category 'coding-category-emacs-mule))
921 (setq coding-category 'coding-category-sjis))
922 ((= type 2) ; ISO2022
924 (vec (make-vector 32 nil))
928 (let ((charset (car fl)))
929 (if (and no-initial-designation
931 (or (charsetp charset)
933 (charsetp (car charset)))))
934 (setq no-initial-designation nil))
935 (if (charsetp charset)
936 (if (= i 1) (setq g1-designation charset))
941 (setq elt (car tail))
943 (setq no-alternative-designation nil)
944 (if (and elt (not (charsetp elt)))
945 (error "Invalid charset: %s" elt)))
946 (setq tail (cdr tail)))
947 (setq g1-designation (car charset)))
950 (setq no-alternative-designation nil)
951 (error "Invalid charset: %s" charset)))))
952 (aset vec i charset))
953 (setq fl (cdr fl) i (1+ i)))
954 (while (and (< i 32) fl)
955 (aset vec i (car fl))
956 (if (and (= i 16) ; ACCEPT-LATIN-EXTRA-CODE
958 (setq accept-latin-extra-code t))
959 (setq fl (cdr fl) i (1+ i)))
960 (aset coding-spec 4 vec)
961 (setq coding-category
962 (if (aref vec 8) ; Use locking-shift.
963 (or (and (aref vec 7) 'coding-category-iso-7-else)
964 'coding-category-iso-8-else)
965 (if (aref vec 7) ; 7-bit only.
966 (if (aref vec 9) ; Use single-shift.
967 'coding-category-iso-7-else
968 (if no-alternative-designation
969 'coding-category-iso-7-tight
970 'coding-category-iso-7))
971 (if (or no-initial-designation
972 (not no-alternative-designation))
973 'coding-category-iso-8-else
974 (if (and (charsetp g1-designation)
975 (= (charset-dimension g1-designation) 2))
976 'coding-category-iso-8-2
977 'coding-category-iso-8-1)))))))
979 (setq coding-category 'coding-category-big5))
980 ((= type 4) ; private
981 (setq coding-category 'coding-category-ccl)
982 (if (not (consp flags))
983 (error "Invalid FLAGS argument for TYPE 4 (CCL)")
984 (let ((decoder (check-ccl-program
986 (intern (format "%s-decoder" coding-system))))
987 (encoder (check-ccl-program
989 (intern (format "%s-encoder" coding-system)))))
990 (if (and decoder encoder)
991 (aset coding-spec 4 (cons decoder encoder))
992 (error "Invalid FLAGS argument for TYPE 4 (CCL)")))))
994 (setq coding-category 'coding-category-raw-text)))
996 (let ((plist (list 'coding-category coding-category
997 'alias-coding-systems (list coding-system))))
998 (if no-initial-designation
999 (plist-put plist 'no-initial-designation t))
1001 (or (eq properties t)
1002 (not (consp (car properties)))))
1003 ;; In the old version, the arg PROPERTIES is a list to be
1004 ;; set in PLIST as a value of property `safe-charsets'.
1005 (setq properties (list (cons 'safe-charsets properties))))
1006 ;; In the current version PROPERTIES is a property list.
1007 ;; Reflect it into PLIST one by one while handling safe-chars
1009 (let ((safe-charsets (cdr (assq 'safe-charsets properties)))
1010 (safe-chars (cdr (assq 'safe-chars properties)))
1013 ;; If only safe-charsets is specified, make a char-table from
1014 ;; it, and store that char-table as the value of `safe-chars'.
1015 (if (and (not safe-chars) safe-charsets)
1017 (if (eq safe-charsets t)
1019 (setq safe-chars (make-char-table 'safe-chars))
1020 (while safe-charsets
1021 (setq charset (car safe-charsets)
1022 safe-charsets (cdr safe-charsets))
1023 (cond ((eq charset 'ascii)) ; just ignore
1024 ((eq charset 'eight-bit-control)
1027 (aset safe-chars i t)
1029 ((eq charset 'eight-bit-graphic)
1032 (aset safe-chars i t)
1035 (aset safe-chars (make-char charset) t))))
1036 (if accept-latin-extra-code
1039 (if (aref latin-extra-code-table i)
1040 (aset safe-chars i t))
1042 (setq l (cons (cons 'safe-chars safe-chars) l))))
1044 (setq prop (car (car l)) val (cdr (car l)) l (cdr l))
1045 (if (eq prop 'safe-chars)
1047 (if (and (symbolp val)
1048 (get val 'translation-table))
1049 (setq safe-chars (get val 'translation-table)))
1050 (setq val safe-chars)))
1051 (plist-put plist prop val)))
1052 ;; The property `coding-category' may have been set differently
1053 ;; through PROPERTIES.
1054 (setq coding-category (plist-get plist 'coding-category))
1055 (aset coding-spec coding-spec-plist-idx plist))
1056 (put coding-system 'coding-system coding-spec)
1057 (put coding-system 'coding-system-define-form nil)
1058 (put coding-category 'coding-systems
1059 (cons coding-system (get coding-category 'coding-systems))))
1061 ;; Next, set a value of `eol-type' property.
1063 ;; If EOL-TYPE is nil, set a vector of subsidiary coding
1064 ;; systems, each corresponds to a coding system for the detected
1066 (setq eol-type (make-subsidiary-coding-system coding-system)))
1068 (cond ((or (eq eol-type 'unix) (null eol-type))
1074 ((or (and (vectorp eol-type)
1075 (= (length eol-type) 3))
1076 (and (numberp eol-type)
1077 (and (>= eol-type 0)
1081 (error "Invalid EOL-TYPE spec:%S" eol-type))))
1082 (put coding-system 'eol-type eol-type)
1084 (define-coding-system-internal coding-system)
1086 ;; At last, register CODING-SYSTEM in `coding-system-list' and
1087 ;; `coding-system-alist'.
1088 (add-to-coding-system-list coding-system)
1089 (or (assoc (symbol-name coding-system) coding-system-alist)
1090 (setq coding-system-alist (cons (list (symbol-name coding-system))
1091 coding-system-alist)))
1093 ;; For a coding system of cateogory iso-8-1 and iso-8-2, create
1094 ;; XXX-with-esc variants.
1095 (let ((coding-category (coding-system-category coding-system)))
1096 (if (or (eq coding-category 'coding-category-iso-8-1)
1097 (eq coding-category 'coding-category-iso-8-2))
1098 (let ((esc (intern (concat (symbol-name coding-system) "-with-esc")))
1099 (doc (format "Same as %s but can handle any charsets by ISO's escape sequences." coding-system))
1100 (safe-charsets (assq 'safe-charsets properties))
1101 (mime-charset (assq 'mime-charset properties)))
1103 (setcdr safe-charsets t)
1104 (setq properties (cons (cons 'safe-charsets t) properties)))
1106 (setcdr mime-charset nil))
1107 (make-coding-system esc type mnemonic doc
1108 (if (listp (car flags))
1109 (cons (append (car flags) '(t)) (cdr flags))
1110 (cons (list (car flags) t) (cdr flags)))
1115 (put 'safe-chars 'char-table-extra-slots 0)
1117 (defun define-coding-system-alias (alias coding-system)
1118 "Define ALIAS as an alias for coding system CODING-SYSTEM."
1119 (put alias 'coding-system (coding-system-spec coding-system))
1120 (put alias 'coding-system-define-form nil)
1121 (add-to-coding-system-list alias)
1122 (or (assoc (symbol-name alias) coding-system-alist)
1123 (setq coding-system-alist (cons (list (symbol-name alias))
1124 coding-system-alist)))
1125 (let ((eol-type (coding-system-eol-type coding-system)))
1126 (if (vectorp eol-type)
1128 (nconc (coding-system-get alias 'alias-coding-systems) (list alias))
1129 (put alias 'eol-type (make-subsidiary-coding-system alias)))
1130 (put alias 'eol-type eol-type))))
1132 (defun merge-coding-systems (first second)
1133 "Fill in any unspecified aspects of coding system FIRST from SECOND.
1134 Return the resulting coding system."
1135 (let ((base (coding-system-base second))
1136 (eol (coding-system-eol-type second)))
1137 ;; If FIRST doesn't specify text conversion, merge with that of SECOND.
1138 (if (eq (coding-system-base first) 'undecided)
1139 (setq first (coding-system-change-text-conversion first base)))
1140 ;; If FIRST doesn't specify eol conversion, merge with that of SECOND.
1141 (if (and (vectorp (coding-system-eol-type first))
1142 (numberp eol) (>= eol 0) (<= eol 2))
1143 (setq first (coding-system-change-eol-conversion
1147 (defun autoload-coding-system (symbol form)
1148 "Define SYMBOL as a coding-system that is defined on demand.
1150 FROM is a form to evaluate to define the coding-system."
1151 (put symbol 'coding-system-define-form form)
1152 (setq coding-system-alist (cons (list (symbol-name symbol))
1153 coding-system-alist))
1154 (dolist (elt '("-unix" "-dos" "-mac"))
1155 (let ((name (concat (symbol-name symbol) elt)))
1156 (put (intern name) 'coding-system-define-form form)
1157 (setq coding-system-alist (cons (list name) coding-system-alist)))))
1159 (defun set-buffer-file-coding-system (coding-system &optional force nomodify)
1160 "Set the file coding-system of the current buffer to CODING-SYSTEM.
1161 This means that when you save the buffer, it will be converted
1162 according to CODING-SYSTEM. For a list of possible values of CODING-SYSTEM,
1163 use \\[list-coding-systems].
1165 If CODING-SYSTEM leaves the text conversion unspecified, or if it
1166 leaves the end-of-line conversion unspecified, FORCE controls what to
1167 do. If FORCE is nil, get the unspecified aspect (or aspects) from the
1168 buffer's previous `buffer-file-coding-system' value (if it is
1169 specified there). Otherwise, leave it unspecified.
1171 This marks the buffer modified so that the succeeding \\[save-buffer]
1172 surely saves the buffer with CODING-SYSTEM. From a program, if you
1173 don't want to mark the buffer modified, specify t for NOMODIFY.
1174 If you know exactly what coding system you want to use,
1175 just set the variable `buffer-file-coding-system' directly."
1176 (interactive "zCoding system for saving file (default nil): \nP")
1177 (check-coding-system coding-system)
1178 (if (and coding-system buffer-file-coding-system (null force))
1180 (merge-coding-systems coding-system buffer-file-coding-system)))
1181 (setq buffer-file-coding-system coding-system)
1182 ;; This is in case of an explicit call. Normally, `normal-mode' and
1183 ;; `set-buffer-major-mode-hook' take care of setting the table.
1184 (if (fboundp 'ucs-set-table-for-input) ; don't lose when building
1185 (ucs-set-table-for-input))
1187 (set-buffer-modified-p t))
1188 (force-mode-line-update))
1190 (defun revert-buffer-with-coding-system (coding-system &optional force)
1191 "Visit the current buffer's file again using coding system CODING-SYSTEM.
1192 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1194 If CODING-SYSTEM leaves the text conversion unspecified, or if it
1195 leaves the end-of-line conversion unspecified, FORCE controls what to
1196 do. If FORCE is nil, get the unspecified aspect (or aspects) from the
1197 buffer's previous `buffer-file-coding-system' value (if it is
1198 specified there). Otherwise, determine it from the file contents as
1199 usual for visiting a file."
1200 (interactive "zCoding system for visited file (default nil): \nP")
1201 (check-coding-system coding-system)
1202 (if (and coding-system buffer-file-coding-system (null force))
1204 (merge-coding-systems coding-system buffer-file-coding-system)))
1205 (let ((coding-system-for-read coding-system))
1208 (defun set-file-name-coding-system (coding-system)
1209 "Set coding system for decoding and encoding file names to CODING-SYSTEM.
1210 It actually just set the variable `file-name-coding-system' (which
1211 see) to CODING-SYSTEM."
1212 (interactive "zCoding system for file names (default nil): ")
1213 (check-coding-system coding-system)
1214 (if (and coding-system
1215 (coding-system-get coding-system 'ascii-incompatible))
1216 (error "%s is not ASCII-compatible" coding-system))
1217 (setq file-name-coding-system coding-system))
1219 (defvar default-terminal-coding-system nil
1220 "Default value for the terminal coding system.
1221 This is normally set according to the selected language environment.
1222 See also the command `set-terminal-coding-system'.")
1224 (defun set-terminal-coding-system (coding-system &optional display)
1225 "Set coding system of terminal output to CODING-SYSTEM.
1226 All text output to DISPLAY will be encoded
1227 with the specified coding system.
1229 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1230 The default is determined by the selected language environment
1231 or by the previous use of this command.
1233 DISPLAY may be a display id, a frame, or nil for the selected frame's display.
1234 The setting has no effect on graphical displays."
1236 (list (let ((default (if (and (not (terminal-coding-system))
1237 default-terminal-coding-system)
1238 default-terminal-coding-system)))
1240 (format "Coding system for terminal display (default %s): "
1243 (if (and (not coding-system)
1244 (not (terminal-coding-system)))
1245 (setq coding-system default-terminal-coding-system))
1247 (setq default-terminal-coding-system coding-system))
1248 (set-terminal-coding-system-internal coding-system display)
1249 (redraw-frame (selected-frame)))
1251 (defvar default-keyboard-coding-system nil
1252 "Default value of the keyboard coding system.
1253 This is normally set according to the selected language environment.
1254 See also the command `set-keyboard-coding-system'.")
1256 (defun set-keyboard-coding-system (coding-system &optional display)
1257 "Set coding system for keyboard input on DISPLAY to CODING-SYSTEM.
1258 In addition, this command calls `encoded-kbd-setup-display' to set up the
1259 translation of keyboard input events to the specified coding system.
1261 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1262 The default is determined by the selected language environment
1263 or by the previous use of this command.
1265 DISPLAY may be a display id, a frame, or nil for the selected frame's display.
1266 The setting has no effect on graphical displays."
1268 (list (let ((default (if (and (not (keyboard-coding-system))
1269 default-keyboard-coding-system)
1270 default-keyboard-coding-system)))
1272 (format "Coding system for keyboard input (default %s): "
1275 (if (and (not coding-system)
1276 (not (keyboard-coding-system)))
1277 (setq coding-system default-keyboard-coding-system))
1279 (setq default-keyboard-coding-system coding-system))
1280 (if (and coding-system
1281 (coding-system-get coding-system 'ascii-incompatible))
1282 (error "%s is not ASCII-compatible" coding-system))
1283 (set-keyboard-coding-system-internal coding-system display)
1284 (setq keyboard-coding-system coding-system)
1285 (encoded-kbd-setup-display display))
1287 (defcustom keyboard-coding-system nil
1288 "Specify coding system for keyboard input.
1289 If you set this on a terminal which can't distinguish Meta keys from
1290 8-bit characters, you will have to use ESC to type Meta characters.
1291 See Info node `Specify Coding' and Info node `Single-Byte Character Support'.
1293 On non-windowing terminals, this is set from the locale by default.
1295 Setting this variable directly does not take effect;
1296 use either \\[customize] or \\[set-keyboard-coding-system]."
1297 :type '(coding-system :tag "Coding system")
1298 :link '(info-link "(emacs)Specify Coding")
1299 :link '(info-link "(emacs)Single-Byte Character Support")
1300 :set (lambda (symbol value)
1301 ;; Don't load encoded-kb unnecessarily.
1302 (if (or value (boundp 'encoded-kbd-setup-display))
1303 (set-keyboard-coding-system value)
1304 (set-default 'keyboard-coding-system nil))) ; must initialize
1309 (defun set-buffer-process-coding-system (decoding encoding)
1310 "Set coding systems for the process associated with the current buffer.
1311 DECODING is the coding system to be used to decode input from the process,
1312 ENCODING is the coding system to be used to encode output to the process.
1314 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems]."
1316 "zCoding-system for output from the process: \nzCoding-system for input to the process: ")
1317 (let ((proc (get-buffer-process (current-buffer))))
1319 (error "No process")
1320 (check-coding-system decoding)
1321 (check-coding-system encoding)
1322 (set-process-coding-system proc decoding encoding)))
1323 (force-mode-line-update))
1325 (defalias 'set-clipboard-coding-system 'set-selection-coding-system)
1327 (defun set-selection-coding-system (coding-system)
1328 "Make CODING-SYSTEM used for communicating with other X clients.
1329 When sending or receiving text via cut_buffer, selection, and clipboard,
1330 the text is encoded or decoded by CODING-SYSTEM."
1331 (interactive "zCoding system for X selection: ")
1332 (check-coding-system coding-system)
1333 (setq selection-coding-system coding-system))
1335 ;; Coding system lastly specified by the command
1336 ;; set-next-selection-coding-system.
1337 (defvar last-next-selection-coding-system nil)
1339 (defun set-next-selection-coding-system (coding-system)
1340 "Make CODING-SYSTEM used for the next communication with other X clients.
1341 This setting is effective for the next communication only."
1343 (list (read-coding-system
1344 (if last-next-selection-coding-system
1345 (format "Coding system for the next X selection (default %S): "
1346 last-next-selection-coding-system)
1347 "Coding system for the next X selection: ")
1348 last-next-selection-coding-system)))
1350 (setq last-next-selection-coding-system coding-system)
1351 (setq coding-system last-next-selection-coding-system))
1352 (check-coding-system coding-system)
1354 (setq next-selection-coding-system coding-system))
1356 (defun set-coding-priority (arg)
1357 "Set priority of coding categories according to ARG.
1358 ARG is a list of coding categories ordered by priority."
1360 (current-list (copy-sequence coding-category-list)))
1361 ;; Check the validity of ARG while deleting coding categories in
1362 ;; ARG from CURRENT-LIST. We assume that CODING-CATEGORY-LIST
1363 ;; contains all coding categories.
1365 (if (or (null (get (car l) 'coding-category-index))
1366 (null (memq (car l) current-list)))
1367 (error "Invalid or duplicated element in argument: %s" arg))
1368 (setq current-list (delq (car l) current-list))
1370 ;; Update `coding-category-list' and return it.
1371 (setq coding-category-list (append arg current-list))
1372 (set-coding-priority-internal)))
1376 (defvar ctext-non-standard-encodings-alist
1377 '(("big5-0" big5 2 (chinese-big5-1 chinese-big5-2))
1378 ("ISO8859-14" iso-8859-14 1 latin-iso8859-14)
1379 ("ISO8859-15" iso-8859-15 1 latin-iso8859-15))
1380 "Alist of non-standard encoding names vs the corresponding usages in CTEXT.
1382 It controls how extended segments of a compound text are handled
1383 by the coding system `compound-text-with-extensions'.
1385 Each element has the form (ENCODING-NAME CODING-SYSTEM N-OCTET CHARSET).
1387 ENCODING-NAME is an encoding name of an \"extended segments\".
1389 CODING-SYSTEM is the coding-system to encode (or decode) the
1390 characters into (or from) the extended segment.
1392 N-OCTET is the number of octets (bytes) that encodes a character
1393 in the segment. It can be 0 (meaning the number of octets per
1394 character is variable), 1, 2, 3, or 4.
1396 CHARSET is a charater set containing characters that are encoded
1397 in the segment. It can be a list of character sets. It can also
1398 be a char-table, in which case characters that have non-nil value
1399 in the char-table are the target.
1401 On decoding CTEXT, all encoding names listed here are recognized.
1403 On encoding CTEXT, encoding names in the variable
1404 `ctext-non-standard-encodings' (which see) and in the information
1405 listed for the current language environment under the key
1406 `ctext-non-standard-encodings' are used.")
1408 (defvar ctext-non-standard-encodings
1410 "List of non-standard encoding names used in extended segments of CTEXT.
1411 Each element must be one of the names listed in the variable
1412 `ctext-non-standard-encodings-alist' (which see).")
1414 (defvar ctext-non-standard-encodings-regexp
1415 (string-to-multibyte
1417 ;; For non-standard encodings.
1418 "\\(\e%/[0-4][\200-\377][\200-\377]\\([^\002]+\\)\002\\)"
1420 ;; For UTF-8 encoding.
1421 "\\(\e%G[^\e]*\e%@\\)")))
1423 ;; Functions to support "Non-Standard Character Set Encodings" defined
1424 ;; by the COMPOUND-TEXT spec. They also support "The UTF-8 encoding"
1425 ;; described in the section 7 of the documentation of COMPOUND-TEXT
1426 ;; distributed with XFree86.
1428 (defun ctext-post-read-conversion (len)
1429 "Decode LEN characters encoded as Compound Text with Extended Segments."
1432 (let ((case-fold-search nil)
1433 (in-workbuf (string= (buffer-name) " *code-converting-work*"))
1434 last-coding-system-used
1437 (narrow-to-region (point) (+ (point) len)))
1439 (set-buffer-multibyte t))
1440 (while (re-search-forward ctext-non-standard-encodings-regexp
1442 (setq pos (match-beginning 0))
1443 (if (match-beginning 1)
1444 ;; ESC % / [0-4] M L --ENCODING-NAME-- \002 --BYTES--
1445 (let* ((M (char-after (+ pos 4)))
1446 (L (char-after (+ pos 5)))
1447 (encoding (match-string 2))
1448 (encoding-info (assoc-string
1450 ctext-non-standard-encodings-alist t))
1451 (coding (if encoding-info
1452 (nth 1 encoding-info)
1453 (setq encoding (intern (downcase encoding)))
1454 (and (coding-system-p encoding)
1456 (setq bytes (- (+ (* (- M 128) 128) (- L 128))
1457 (- (point) (+ pos 6))))
1459 (delete-region pos (point))
1460 (forward-char bytes)
1461 (decode-coding-region (- (point) bytes) (point) coding)))
1462 ;; ESC % G --UTF-8-BYTES-- ESC % @
1464 (delete-region pos (+ pos 3))
1465 (decode-coding-region pos (point) 'utf-8))))
1466 (goto-char (point-min))
1467 (- (point-max) (point)))))
1469 ;; Return a char table of extended segment usage for each character.
1470 ;; Each value of the char table is nil, one of the elements of
1471 ;; `ctext-non-standard-encodings-alist', or the symbol `utf-8'.
1473 (defun ctext-non-standard-encodings-table ()
1474 (let ((table (make-char-table 'translation-table)))
1475 (aset table (make-char 'mule-unicode-0100-24ff) 'utf-8)
1476 (aset table (make-char 'mule-unicode-2500-33ff) 'utf-8)
1477 (aset table (make-char 'mule-unicode-e000-ffff) 'utf-8)
1478 (dolist (encoding (reverse
1480 (get-language-info current-language-environment
1481 'ctext-non-standard-encodings)
1482 ctext-non-standard-encodings)))
1483 (let* ((slot (assoc encoding ctext-non-standard-encodings-alist))
1484 (charset (nth 3 slot)))
1486 (cond ((charsetp charset)
1487 (aset table (make-char charset) slot))
1489 (dolist (elt charset)
1490 (aset table (make-char elt) slot)))
1491 ((char-table-p charset)
1492 (map-char-table #'(lambda (k v)
1493 (if (and v (> k 128)) (aset table k slot)))
1497 (defun ctext-pre-write-conversion (from to)
1498 "Encode characters between FROM and TO as Compound Text w/Extended Segments.
1500 If FROM is a string, or if the current buffer is not the one set up for us
1501 by encode-coding-string, generate a new temp buffer, insert the
1502 text, and convert it in the temporary buffer. Otherwise, convert in-place."
1504 (let ((workbuf (get-buffer-create " *code-conversion-work*")))
1505 ;; Setup a working buffer if necessary.
1506 (cond ((stringp from)
1507 (set-buffer workbuf)
1509 (set-buffer-multibyte (multibyte-string-p from))
1511 ((not (eq (current-buffer) workbuf))
1512 (let ((buf (current-buffer))
1513 (multibyte enable-multibyte-characters))
1514 (set-buffer workbuf)
1516 (set-buffer-multibyte multibyte)
1517 (insert-buffer-substring buf from to)))))
1519 ;; Now we can encode the whole buffer.
1520 (let ((encoding-table (ctext-non-standard-encodings-table))
1521 last-coding-system-used
1522 last-pos last-encoding-info
1523 encoding-info end-pos)
1524 (goto-char (setq last-pos (point-min)))
1525 (setq end-pos (point-marker))
1526 (while (re-search-forward "[^\000-\177]+" nil t)
1527 ;; Found a sequence of non-ASCII characters.
1528 (setq last-pos (match-beginning 0)
1529 last-encoding-info (aref encoding-table (char-after last-pos)))
1530 (set-marker end-pos (match-end 0))
1531 (goto-char (1+ last-pos))
1535 (if (< (point) end-pos)
1536 (aref encoding-table (following-char))))
1537 (unless (eq last-encoding-info encoding-info)
1538 (cond ((consp last-encoding-info)
1539 ;; Encode the previous range using an extended
1541 (let ((encoding-name (car last-encoding-info))
1542 (coding-system (nth 1 last-encoding-info))
1543 (noctets (nth 2 last-encoding-info))
1545 (encode-coding-region last-pos (point) coding-system)
1546 (setq len (+ (length encoding-name) 1
1547 (- (point) last-pos)))
1549 (goto-char last-pos)
1550 (insert (string-to-multibyte
1551 (format "\e%%/%d%c%c%s
\ 2"
1556 ((eq last-encoding-info 'utf-8)
1557 ;; Encode the previous range using UTF-8 encoding
1559 (encode-coding-region last-pos (point) 'mule-utf-8)
1561 (goto-char last-pos)
1564 (setq last-pos (point)
1565 last-encoding-info encoding-info))
1566 (if (< (point) end-pos)
1568 (throw 'tag nil)))))
1569 (set-marker end-pos nil)
1570 (goto-char (point-min))))
1571 ;; Must return nil, as build_annotations_2 expects that.
1576 (defcustom auto-coding-alist
1577 '(("\\.\\(arc\\|zip\\|lzh\\|zoo\\|[jew]ar\\|xpi\\)\\'" . no-conversion)
1578 ("\\.\\(ARC\\|ZIP\\|LZH\\|ZOO\\|[JEW]AR\\|XPI\\)\\'" . no-conversion)
1579 ("\\.\\(sx[dmicw]\\|tar\\|tgz\\)\\'" . no-conversion)
1580 ("\\.\\(gz\\|Z\\|bz\\|bz2\\|gpg\\)\\'" . no-conversion)
1581 ("\\.\\(jpe?g\\|png\\|gif\\|tiff?\\|p[bpgn]m\\)\\'" . no-conversion)
1582 ("/#[^/]+#\\'" . emacs-mule))
1583 "Alist of filename patterns vs corresponding coding systems.
1584 Each element looks like (REGEXP . CODING-SYSTEM).
1585 A file whose name matches REGEXP is decoded by CODING-SYSTEM on reading.
1587 The settings in this alist take priority over `coding:' tags
1588 in the file (see the function `set-auto-coding')
1589 and the contents of `file-coding-system-alist'."
1592 :type '(repeat (cons (regexp :tag "File name regexp")
1593 (symbol :tag "Coding system"))))
1595 (defcustom auto-coding-regexp-alist
1596 '(("^BABYL OPTIONS:[ \t]*-\\*-[ \t]*rmail[ \t]*-\\*-" . no-conversion))
1597 "Alist of patterns vs corresponding coding systems.
1598 Each element looks like (REGEXP . CODING-SYSTEM).
1599 A file whose first bytes match REGEXP is decoded by CODING-SYSTEM on reading.
1601 The settings in this alist take priority over `coding:' tags
1602 in the file (see the function `set-auto-coding')
1603 and the contents of `file-coding-system-alist'."
1606 :type '(repeat (cons (regexp :tag "Regexp")
1607 (symbol :tag "Coding system"))))
1609 (defun auto-coding-regexp-alist-lookup (from to)
1610 "Lookup `auto-coding-regexp-alist' for the contents of the current buffer.
1611 The value is a coding system is specified for the region FROM and TO,
1615 (let ((alist auto-coding-regexp-alist)
1617 (while (and alist (not coding-system))
1618 (let ((regexp (car (car alist))))
1619 (if enable-multibyte-characters
1620 (setq regexp (string-to-multibyte regexp)))
1621 (if (re-search-forward regexp to t)
1622 (setq coding-system (cdr (car alist)))
1623 (setq alist (cdr alist)))))
1626 ;; See the bottom of this file for built-in auto coding functions.
1627 (defcustom auto-coding-functions '(sgml-xml-auto-coding-function
1628 sgml-html-meta-auto-coding-function)
1629 "A list of functions which attempt to determine a coding system.
1631 Each function in this list should be written to operate on the
1632 current buffer, but should not modify it in any way. The buffer
1633 will contain undecoded text of parts of the file. Each function
1634 should take one argument, SIZE, which says how many
1635 characters (starting from point) it should look at.
1637 If one of these functions succeeds in determining a coding
1638 system, it should return that coding system. Otherwise, it
1641 If a file has a `coding:' tag, that takes precedence over these
1642 functions, so they won't be called at all."
1645 :type '(repeat function))
1647 (defvar set-auto-coding-for-load nil
1648 "Non-nil means look for `load-coding' property instead of `coding'.
1649 This is used for loading and byte-compiling Emacs Lisp files.")
1651 (defun auto-coding-alist-lookup (filename)
1652 "Return the coding system specified by `auto-coding-alist' for FILENAME."
1653 (let ((alist auto-coding-alist)
1654 (case-fold-search (memq system-type '(vax-vms windows-nt ms-dos cygwin)))
1656 (while (and alist (not coding-system))
1657 (if (string-match (car (car alist)) filename)
1658 (setq coding-system (cdr (car alist)))
1659 (setq alist (cdr alist))))
1662 (defun find-auto-coding (filename size)
1663 "Find a coding system for a file FILENAME of which SIZE bytes follow point.
1664 These bytes should include at least the first 1k of the file
1665 and the last 3k of the file, but the middle may be omitted.
1667 The function checks FILENAME against the variable `auto-coding-alist'.
1668 If FILENAME doesn't match any entries in the variable, it checks the
1669 contents of the current buffer following point against
1670 `auto-coding-regexp-alist'. If no match is found, it checks for a
1671 `coding:' tag in the first one or two lines following point. If no
1672 `coding:' tag is found, it checks any local variables list in the last
1673 3K bytes out of the SIZE bytes. Finally, if none of these methods
1674 succeed, it checks to see if any function in `auto-coding-functions'
1677 If a coding system is specifed, the return value is a
1678 cons (CODING . SOURCE), where CODING is the specified coding
1679 system and SOURCE is a symbol `auto-coding-alist',
1680 `auto-coding-regexp-alist', `coding:', or `auto-coding-functions'
1681 indicating by what CODING is specified. Note that the validity
1682 of CODING is not checked; it's callers responsibility to check
1685 If nothing is specified, the return value is nil."
1686 (or (let ((coding-system (auto-coding-alist-lookup filename)))
1688 (cons coding-system 'auto-coding-alist)))
1689 ;; Try using `auto-coding-regexp-alist'.
1690 (let ((coding-system (auto-coding-regexp-alist-lookup (point)
1693 (cons coding-system 'auto-coding-regexp-alist)))
1694 (let* ((case-fold-search t)
1695 (head-start (point))
1696 (head-end (+ head-start (min size 1024)))
1697 (tail-start (+ head-start (max (- size 3072) 0)))
1698 (tail-end (+ head-start size))
1699 coding-system head-found tail-found pos)
1700 ;; Try a short cut by searching for the string "coding:"
1701 ;; and for "unibyte:" at the head and tail of SIZE bytes.
1702 (setq head-found (or (search-forward "coding:" head-end t)
1703 (search-forward "unibyte:" head-end t)))
1704 (if (and head-found (> head-found tail-start))
1705 ;; Head and tail are overlapped.
1706 (setq tail-found head-found)
1707 (goto-char tail-start)
1708 (setq tail-found (or (search-forward "coding:" tail-end t)
1709 (search-forward "unibyte:" tail-end t))))
1711 ;; At first check the head.
1713 (goto-char head-start)
1714 (setq head-end (set-auto-mode-1))
1715 (setq head-start (point))
1716 (when (and head-end (< head-found head-end))
1717 (goto-char head-start)
1718 (when (and set-auto-coding-for-load
1720 "\\(.*;\\)?[ \t]*unibyte:[ \t]*\\([^ ;]+\\)"
1722 (setq coding-system 'raw-text))
1723 (when (and (not coding-system)
1725 "\\(.*;\\)?[ \t]*coding:[ \t]*\\([^ ;]+\\)"
1727 (setq coding-system (intern (match-string 2))))))
1729 ;; If no coding: tag in the head, check the tail.
1730 ;; Here we must pay attention to the case that the end-of-line
1731 ;; is just "\r" and we can't use "^" nor "$" in regexp.
1732 (when (and tail-found (not coding-system))
1733 (goto-char tail-start)
1734 (re-search-forward "[\r\n]\^L" nil t)
1735 (if (re-search-forward
1736 "[\r\n]\\([^[\r\n]*\\)[ \t]*Local Variables:[ \t]*\\([^\r\n]*\\)[\r\n]"
1738 ;; The prefix is what comes before "local variables:" in its
1739 ;; line. The suffix is what comes after "local variables:"
1741 (let* ((prefix (regexp-quote (match-string 1)))
1742 (suffix (regexp-quote (match-string 2)))
1746 ;; N.B. without the \n below, the regexp can
1748 "[ \t]*coding[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1753 "[ \t]*unibyte[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1756 (concat "[\r\n]" prefix "[ \t]*End *:[ \t]*" suffix
1759 (forward-char -1) ; skip back \r or \n.
1760 (re-search-forward re-end tail-end 'move)
1761 (setq tail-end (point))
1763 (when (and set-auto-coding-for-load
1764 (re-search-forward re-unibyte tail-end t))
1765 (setq coding-system 'raw-text))
1766 (when (and (not coding-system)
1767 (re-search-forward re-coding tail-end t))
1768 (setq coding-system (intern (match-string 1)))))))
1770 (cons coding-system :coding)))
1771 ;; Finally, try all the `auto-coding-functions'.
1772 (let ((funcs auto-coding-functions)
1773 (coding-system nil))
1774 (while (and funcs (not coding-system))
1775 (setq coding-system (condition-case e
1777 (goto-char (point-min))
1778 (funcall (pop funcs) size))
1781 (cons coding-system 'auto-coding-functions)))))
1783 (defun set-auto-coding (filename size)
1784 "Return coding system for a file FILENAME of which SIZE bytes follow point.
1785 See `find-auto-coding' for how the coding system is found.
1786 Return nil if an invalid coding system is found.
1788 The variable `set-auto-coding-function' (which see) is set to this
1789 function by default."
1790 (let ((found (find-auto-coding filename size)))
1791 (if (and found (coding-system-p (car found)))
1794 (setq set-auto-coding-function 'set-auto-coding)
1796 ;; This variable is set in these two cases:
1797 ;; (1) A file is read by a coding system specified explicitly.
1798 ;; after-insert-file-set-coding sets this value to
1799 ;; coding-system-for-read.
1800 ;; (2) A buffer is saved.
1801 ;; After writing, basic-save-buffer-1 sets this value to
1802 ;; last-coding-system-used.
1803 ;; This variable is used for decoding in revert-buffer.
1804 (defvar buffer-file-coding-system-explicit nil
1805 "The file coding system explicitly specified for the current buffer.
1806 Internal use only.")
1807 (make-variable-buffer-local 'buffer-file-coding-system-explicit)
1808 (put 'buffer-file-coding-system-explicit 'permanent-local t)
1810 (defun after-insert-file-set-coding (inserted &optional visit)
1811 "Set `buffer-file-coding-system' of current buffer after text is inserted.
1812 INSERTED is the number of characters that were inserted, as figured
1813 in the situation before this function. Return the number of characters
1814 inserted, as figured in the situation after. The two numbers can be
1815 different if the buffer has become unibyte.
1816 The optional second arg VISIT non-nil means that we are visiting a file."
1818 coding-system-for-read
1819 (not (eq coding-system-for-read 'auto-save-coding)))
1820 (setq buffer-file-coding-system-explicit coding-system-for-read))
1821 (if last-coding-system-used
1822 (let ((coding-system
1823 (find-new-buffer-file-coding-system last-coding-system-used))
1824 (modified-p (buffer-modified-p)))
1826 ;; Tell set-buffer-file-coding-system not to mark the file
1827 ;; as modified; we just read it, and it's supposed to be unmodified.
1828 ;; Marking it modified would try to lock it, which would
1829 ;; check the modtime, and we don't want to do that again now.
1830 (set-buffer-file-coding-system coding-system t t)
1831 (if (and enable-multibyte-characters
1832 (or (eq coding-system 'no-conversion)
1833 (eq (coding-system-type coding-system) 5))
1834 ;; If buffer was unmodified and the size is the
1835 ;; same as INSERTED, we must be visiting it.
1837 (= (buffer-size) inserted))
1838 ;; For coding systems no-conversion and raw-text...,
1839 ;; edit the buffer as unibyte.
1840 (let ((pos-marker (copy-marker (+ (point) inserted)))
1842 (buffer-file-name nil))
1843 (set-buffer-multibyte nil)
1844 (setq inserted (- pos-marker (point)))))
1845 (set-buffer-modified-p modified-p))))
1848 ;; The coding-spec and eol-type of coding-system returned is decided
1849 ;; independently in the following order.
1850 ;; 1. That of buffer-file-coding-system locally bound.
1851 ;; 2. That of CODING.
1853 (defun find-new-buffer-file-coding-system (coding)
1854 "Return a coding system for a buffer when a file of CODING is inserted.
1855 The local variable `buffer-file-coding-system' of the current buffer
1856 is set to the returned value.
1857 Return nil if there's no need to set `buffer-file-coding-system'."
1858 (let (local-coding local-eol
1859 found-coding found-eol
1862 ;; Nothing found about coding.
1865 ;; Get information of `buffer-file-coding-system' in LOCAL-EOL
1866 ;; and LOCAL-CODING.
1867 (setq local-eol (coding-system-eol-type buffer-file-coding-system))
1868 (if (null (numberp local-eol))
1869 ;; But eol-type is not yet set.
1870 (setq local-eol nil))
1871 (if (and buffer-file-coding-system
1872 (not (eq (coding-system-type buffer-file-coding-system) t)))
1873 ;; This is not `undecided'.
1874 (setq local-coding (coding-system-base buffer-file-coding-system)))
1876 (if (and (local-variable-p 'buffer-file-coding-system)
1877 local-eol local-coding)
1878 ;; The current buffer has already set full coding-system, we
1879 ;; had better not change it.
1882 (setq found-eol (coding-system-eol-type coding))
1883 (if (null (numberp found-eol))
1884 ;; But eol-type is not found.
1885 ;; If EOL conversions are inhibited, force unix eol-type.
1886 (setq found-eol (if inhibit-eol-conversion 0)))
1887 (if (eq (coding-system-type coding) t)
1888 (setq found-coding 'undecided)
1889 (setq found-coding (coding-system-base coding)))
1891 (if (and (not found-eol) (eq found-coding 'undecided))
1892 ;; No valid coding information found.
1895 ;; Some coding information (eol or text) found.
1897 ;; The local setting takes precedence over the found one.
1898 (setq new-coding (if (local-variable-p 'buffer-file-coding-system)
1899 (or local-coding found-coding)
1900 (or found-coding local-coding)))
1901 (setq new-eol (if (local-variable-p 'buffer-file-coding-system)
1902 (or local-eol found-eol)
1903 (or found-eol local-eol)))
1905 (let ((eol-type (coding-system-eol-type new-coding)))
1906 (if (and (numberp new-eol) (vectorp eol-type))
1907 (aref eol-type new-eol)
1910 (defun modify-coding-system-alist (target-type regexp coding-system)
1911 "Modify one of look up tables for finding a coding system on I/O operation.
1912 There are three of such tables, `file-coding-system-alist',
1913 `process-coding-system-alist', and `network-coding-system-alist'.
1915 TARGET-TYPE specifies which of them to modify.
1916 If it is `file', it affects `file-coding-system-alist' (which see).
1917 If it is `process', it affects `process-coding-system-alist' (which see).
1918 If it is `network', it affects `network-coding-system-alist' (which see).
1920 REGEXP is a regular expression matching a target of I/O operation.
1921 The target is a file name if TARGET-TYPE is `file', a program name if
1922 TARGET-TYPE is `process', or a network service name or a port number
1923 to connect to if TARGET-TYPE is `network'.
1925 CODING-SYSTEM is a coding system to perform code conversion on the I/O
1926 operation, or a cons cell (DECODING . ENCODING) specifying the coding systems
1927 for decoding and encoding respectively,
1928 or a function symbol which, when called, returns such a cons cell."
1929 (or (memq target-type '(file process network))
1930 (error "Invalid target type: %s" target-type))
1931 (or (stringp regexp)
1932 (and (eq target-type 'network) (integerp regexp))
1933 (error "Invalid regular expression: %s" regexp))
1934 (if (symbolp coding-system)
1935 (if (not (fboundp coding-system))
1937 (check-coding-system coding-system)
1938 (setq coding-system (cons coding-system coding-system))))
1939 (check-coding-system (car coding-system))
1940 (check-coding-system (cdr coding-system)))
1941 (cond ((eq target-type 'file)
1942 (let ((slot (assoc regexp file-coding-system-alist)))
1944 (setcdr slot coding-system)
1945 (setq file-coding-system-alist
1946 (cons (cons regexp coding-system)
1947 file-coding-system-alist)))))
1948 ((eq target-type 'process)
1949 (let ((slot (assoc regexp process-coding-system-alist)))
1951 (setcdr slot coding-system)
1952 (setq process-coding-system-alist
1953 (cons (cons regexp coding-system)
1954 process-coding-system-alist)))))
1956 (let ((slot (assoc regexp network-coding-system-alist)))
1958 (setcdr slot coding-system)
1959 (setq network-coding-system-alist
1960 (cons (cons regexp coding-system)
1961 network-coding-system-alist)))))))
1963 (defun decode-coding-inserted-region (from to filename
1964 &optional visit beg end replace)
1965 "Decode the region between FROM and TO as if it is read from file FILENAME.
1966 The idea is that the text between FROM and TO was just inserted somehow.
1967 Optional arguments VISIT, BEG, END, and REPLACE are the same as those
1968 of the function `insert-file-contents'.
1969 Part of the job of this function is setting `buffer-undo-list' appropriately."
1972 (let ((coding coding-system-for-read)
1975 ;; Temporarily turn off undo recording, if we're decoding the
1976 ;; text of a visited file.
1977 (setq buffer-undo-list t)
1978 ;; Otherwise, if we can recognize the undo elt for the insertion,
1979 ;; remove it and get ready to replace it later.
1980 ;; In the mean time, turn off undo recording.
1981 (let ((last (car-safe buffer-undo-list)))
1982 (if (and (consp last) (eql (car last) from) (eql (cdr last) to))
1983 (setq undo-list-saved (cdr buffer-undo-list)
1984 buffer-undo-list t))))
1985 (narrow-to-region from to)
1986 (goto-char (point-min))
1988 (setq coding (funcall set-auto-coding-function
1989 filename (- (point-max) (point-min)))))
1991 (setq coding (car (find-operation-coding-system
1992 'insert-file-contents
1993 filename visit beg end replace))))
1994 (if (coding-system-p coding)
1995 (or enable-multibyte-characters
1997 (coding-system-change-text-conversion coding 'raw-text)))
2000 (decode-coding-region (point-min) (point-max) coding)
2001 (setq last-coding-system-used coding))
2002 ;; If we're decoding the text of a visited file,
2003 ;; the undo list should start out empty.
2005 (setq buffer-undo-list nil)
2006 ;; If we decided to replace the undo entry for the insertion,
2009 (setq buffer-undo-list
2010 (cons (cons from (point-max)) undo-list-saved))))))))
2012 (defun recode-region (start end new-coding coding)
2013 "Re-decode the region (previously decoded by CODING) by NEW-CODING."
2015 (list (region-beginning) (region-end)
2016 (read-coding-system "Text was really in: ")
2017 (let ((coding (or buffer-file-coding-system last-coding-system-used)))
2019 (concat "But was interpreted as"
2020 (if coding (format " (default %S): " coding) ": "))
2022 (or (and new-coding coding)
2023 (error "Coding system not specified"))
2024 ;; Check it before we encode the region.
2025 (check-coding-system new-coding)
2027 (narrow-to-region start end)
2028 (encode-coding-region (point-min) (point-max) coding)
2029 (decode-coding-region (point-min) (point-max) new-coding)))
2031 (defun make-translation-table (&rest args)
2032 "Make a translation table from arguments.
2033 A translation table is a char table intended for character
2034 translation in CCL programs.
2036 Each argument is a list of elements of the form (FROM . TO), where FROM
2037 is a character to be translated to TO.
2039 FROM can be a generic character (see `make-char'). In this case, TO is
2040 a generic character containing the same number of characters, or an
2041 ordinary character. If FROM and TO are both generic characters, all
2042 characters belonging to FROM are translated to characters belonging to TO
2043 without changing their position code(s).
2045 The arguments and forms in each argument are processed in the given
2046 order, and if a previous form already translates TO to some other
2047 character, say TO-ALT, FROM is also translated to TO-ALT."
2048 (let ((table (make-char-table 'translation-table))
2051 (let ((elts (car args)))
2053 (let* ((from (car (car elts)))
2054 (from-i 0) ; degree of freedom of FROM
2055 (from-rev (nreverse (split-char from)))
2056 (to (cdr (car elts)))
2057 (to-i 0) ; degree of freedom of TO
2058 (to-rev (nreverse (split-char to))))
2059 ;; Check numbers of heading 0s in FROM-REV and TO-REV.
2060 (while (eq (car from-rev) 0)
2061 (setq from-i (1+ from-i) from-rev (cdr from-rev)))
2062 (while (eq (car to-rev) 0)
2063 (setq to-i (1+ to-i) to-rev (cdr to-rev)))
2064 (if (and (/= from-i to-i) (/= to-i 0))
2065 (error "Invalid character pair (%d . %d)" from to))
2066 ;; If we have already translated TO to TO-ALT, FROM should
2067 ;; also be translated to TO-ALT. But, this is only if TO
2068 ;; is a generic character or TO-ALT is not a generic
2070 (let ((to-alt (aref table to)))
2072 (or (> to-i 0) (not (generic-char-p to-alt))))
2075 (set-char-table-default table from to)
2076 (aset table from to))
2077 ;; If we have already translated some chars to FROM, they
2078 ;; should also be translated to TO.
2079 (let ((l (assq from revlist)))
2086 (setq l (cdr l)) ))))
2087 ;; Now update REVLIST.
2088 (let ((l (assq to revlist)))
2090 (setcdr l (cons from (cdr l)))
2091 (setq revlist (cons (list to from) revlist)))))
2092 (setq elts (cdr elts))))
2093 (setq args (cdr args)))
2094 ;; Return TABLE just created.
2097 (defun make-translation-table-from-vector (vec)
2098 "Make translation table from decoding vector VEC.
2099 VEC is an array of 256 elements to map unibyte codes to multibyte
2100 characters. Elements may be nil for undefined code points.
2101 See also the variable `nonascii-translation-table'."
2102 (let ((table (make-char-table 'translation-table))
2103 (rev-table (make-char-table 'translation-table))
2106 (setq ch (aref vec i))
2110 (aset rev-table ch i))))
2111 (set-char-table-extra-slot table 0 rev-table)
2114 (defun define-translation-table (symbol &rest args)
2115 "Define SYMBOL as the name of translation table made by ARGS.
2116 This sets up information so that the table can be used for
2117 translations in a CCL program.
2119 If the first element of ARGS is a char-table whose purpose is
2120 `translation-table', just define SYMBOL to name it. (Note that this
2121 function does not bind SYMBOL.)
2123 Any other ARGS should be suitable as arguments of the function
2124 `make-translation-table' (which see).
2126 This function sets properties `translation-table' and
2127 `translation-table-id' of SYMBOL to the created table itself and the
2128 identification number of the table respectively. It also registers
2129 the table in `translation-table-vector'."
2130 (let ((table (if (and (char-table-p (car args))
2131 (eq (char-table-subtype (car args))
2132 'translation-table))
2134 (apply 'make-translation-table args)))
2135 (len (length translation-table-vector))
2138 (put symbol 'translation-table table)
2141 (setq translation-table-vector
2142 (vconcat translation-table-vector (make-vector len nil))))
2143 (let ((slot (aref translation-table-vector id)))
2145 (eq (car slot) symbol))
2147 (aset translation-table-vector id (cons symbol table))
2149 (setq id (1+ id)))))
2150 (put symbol 'translation-table-id id)
2153 (defun translate-region (start end table)
2154 "From START to END, translate characters according to TABLE.
2155 TABLE is a string or a char-table.
2156 If TABLE is a string, the Nth character in it is the mapping
2157 for the character with code N.
2158 If TABLE is a char-table, the element for character N is the mapping
2159 for the character with code N.
2160 It returns the number of characters changed."
2162 (list (region-beginning)
2165 (dotimes (i (length translation-table-vector))
2166 (if (consp (aref translation-table-vector i))
2167 (push (list (symbol-name
2168 (car (aref translation-table-vector i)))) l)))
2170 (error "No translation table defined"))
2172 (setq table (completing-read "Translation table: " l nil t)))
2175 (let ((val (get table 'translation-table)))
2176 (or (char-table-p val)
2177 (error "Invalid translation table name: %s" table))
2179 (translate-region-internal start end table))
2181 (put 'with-category-table 'lisp-indent-function 1)
2183 (defmacro with-category-table (table &rest body)
2184 "Evaluate BODY with category table of current buffer set to TABLE.
2185 The category table of the current buffer is saved, BODY is evaluated,
2186 then the saved table is restored, even in case of an abnormal exit.
2187 Value is what BODY returns."
2188 (let ((old-table (make-symbol "old-table"))
2189 (old-buffer (make-symbol "old-buffer")))
2190 `(let ((,old-table (category-table))
2191 (,old-buffer (current-buffer)))
2194 (set-category-table ,table)
2196 (save-current-buffer
2197 (set-buffer ,old-buffer)
2198 (set-category-table ,old-table))))))
2200 (defun define-translation-hash-table (symbol table)
2201 "Define SYMBOL as the name of the hash translation TABLE for use in CCL.
2203 Analogous to `define-translation-table', but updates
2204 `translation-hash-table-vector' and the table is for use in the CCL
2205 `lookup-integer' and `lookup-character' functions."
2206 (unless (and (symbolp symbol)
2207 (hash-table-p table))
2208 (error "Bad args to define-translation-hash-table"))
2209 (let ((len (length translation-hash-table-vector))
2212 (put symbol 'translation-hash-table table)
2215 (setq translation-hash-table-vector
2216 (vconcat translation-hash-table-vector [nil])))
2217 (let ((slot (aref translation-hash-table-vector id)))
2219 (eq (car slot) symbol))
2221 (aset translation-hash-table-vector id (cons symbol table))
2223 (setq id (1+ id)))))
2224 (put symbol 'translation-hash-table-id id)
2227 ;;; Initialize some variables.
2229 (put 'use-default-ascent 'char-table-extra-slots 0)
2230 (setq use-default-ascent (make-char-table 'use-default-ascent))
2231 (put 'ignore-relative-composition 'char-table-extra-slots 0)
2232 (setq ignore-relative-composition
2233 (make-char-table 'ignore-relative-composition))
2236 ;;; Built-in auto-coding-functions:
2238 (defun sgml-xml-auto-coding-function (size)
2239 "Determine whether the buffer is XML, and if so, its encoding.
2240 This function is intended to be added to `auto-coding-functions'."
2241 (setq size (+ (point) size))
2242 (when (re-search-forward "\\`[[:space:]\n]*<\\?xml" size t)
2243 (let ((end (save-excursion
2245 (re-search-forward "[\"']\\s-*\\?>" size t))))
2247 (if (re-search-forward "encoding=[\"']\\(.+?\\)[\"']" end t)
2248 (let* ((match (match-string 1))
2249 (sym (intern (downcase match))))
2250 (if (coding-system-p sym)
2252 (message "Warning: unknown coding system \"%s\"" match)
2256 (defun sgml-html-meta-auto-coding-function (size)
2257 "If the buffer has an HTML meta tag, use it to determine encoding.
2258 This function is intended to be added to `auto-coding-functions'."
2259 (setq size (min (+ (point) size)
2260 ;; Only search forward 10 lines
2264 (when (and (search-forward "<html" size t)
2265 (re-search-forward "<meta\\s-+http-equiv=\"content-type\"\\s-+content=\"text/\\sw+;\\s-*charset=\\(.+?\\)\"" size t))
2266 (let* ((match (match-string 1))
2267 (sym (intern (downcase match))))
2268 (if (coding-system-p sym)
2270 (message "Warning: unknown coding system \"%s\"" match)
2276 ;;; arch-tag: 9aebaa6e-0e8a-40a9-b857-cb5d04a39e7c
2277 ;;; mule.el ends here