1 ;;; mule.el --- basic commands for mulitilingual environment
3 ;; Copyright (C) 1995 Electrotechnical Laboratory, JAPAN.
4 ;; Licensed to the Free Software Foundation.
5 ;; Copyright (C) 2001, 2002 Free Software Foundation, Inc.
7 ;; Keywords: mule, multilingual, character set, coding system
9 ;; This file is part of GNU Emacs.
11 ;; GNU Emacs is free software; you can redistribute it and/or modify
12 ;; it under the terms of the GNU General Public License as published by
13 ;; the Free Software Foundation; either version 2, or (at your option)
16 ;; GNU Emacs is distributed in the hope that it will be useful,
17 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
18 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
19 ;; GNU General Public License for more details.
21 ;; You should have received a copy of the GNU General Public License
22 ;; along with GNU Emacs; see the file COPYING. If not, write to the
23 ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330,
24 ;; Boston, MA 02111-1307, USA.
30 (defconst mule-version "5.0 (SAKAKI)" "\
31 Version number and name of this version of MULE (multilingual environment).")
33 (defconst mule-version-date "1999.12.7" "\
34 Distribution date of this version of MULE (multilingual environment).")
36 (defun load-with-code-conversion (fullname file &optional noerror nomessage)
37 "Execute a file of Lisp code named FILE whose absolute name is FULLNAME.
38 The file contents are decoded before evaluation if necessary.
39 If optional second arg NOERROR is non-nil,
40 report no error if FILE doesn't exist.
41 Print messages at start and end of loading unless
42 optional third arg NOMESSAGE is non-nil.
43 Return t if file exists."
44 (if (null (file-readable-p fullname))
46 (signal 'file-error (list "Cannot open load file" file)))
47 ;; Read file with code conversion, and then eval.
49 ;; To avoid any autoloading, set default-major-mode to
51 ;; So that we don't get completely screwed if the
52 ;; file is encoded in some complicated character set,
53 ;; read it with real decoding, as a multibyte buffer,
54 ;; even if this is a --unibyte Emacs session.
55 (let ((default-major-mode 'fundamental-mode)
56 (default-enable-multibyte-characters t))
57 ;; We can't use `generate-new-buffer' because files.el
59 (get-buffer-create (generate-new-buffer-name " *load*"))))
61 (source (save-match-data (string-match "\\.el\\'" fullname))))
64 (message "Loading %s (source)..." file)
65 (message "Loading %s..." file)))
67 (push file preloaded-file-list))
69 (let ((load-file-name fullname)
70 (set-auto-coding-for-load t)
71 (inhibit-file-name-operation nil))
74 (insert-file-contents fullname)
75 ;; If the loaded file was inserted with no-conversion or
76 ;; raw-text coding system, make the buffer unibyte.
77 ;; Otherwise, eval-buffer might try to interpret random
78 ;; binary junk as multibyte characters.
79 (if (and enable-multibyte-characters
80 (or (eq (coding-system-type last-coding-system-used) 5)
81 (eq last-coding-system-used 'no-conversion)))
82 (set-buffer-multibyte nil))
83 ;; Make `kill-buffer' quiet.
84 (set-buffer-modified-p nil))
85 ;; Have the original buffer current while we eval.
86 (eval-buffer buffer nil file
87 ;; If this Emacs is running with --unibyte,
88 ;; convert multibyte strings to unibyte
89 ;; after reading them.
90 ;; (not default-enable-multibyte-characters)
93 (let (kill-buffer-hook kill-buffer-query-functions)
94 (kill-buffer buffer)))
95 (let ((hook (assoc file after-load-alist)))
97 (mapcar (function eval) (cdr hook))))
98 (unless (or nomessage noninteractive)
100 (message "Loading %s (source)...done" file)
101 (message "Loading %s...done" file)))
104 ;; API (Application Program Interface) for charsets.
106 (defsubst charset-quoted-standard-p (obj)
107 "Return t if OBJ is a quoted symbol, and is the name of a standard charset."
108 (and (listp obj) (eq (car obj) 'quote)
109 (symbolp (car-safe (cdr obj)))
110 (let ((vector (get (car-safe (cdr obj)) 'charset)))
111 (and (vectorp vector)
112 (< (aref vector 0) 160)))))
114 (defsubst charsetp (object)
115 "T if OBJECT is a charset."
116 (and (symbolp object) (vectorp (get object 'charset))))
118 (defsubst charset-info (charset)
119 "Return a vector of information of CHARSET.
120 The elements of the vector are:
121 CHARSET-ID, BYTES, DIMENSION, CHARS, WIDTH, DIRECTION,
122 LEADING-CODE-BASE, LEADING-CODE-EXT,
123 ISO-FINAL-CHAR, ISO-GRAPHIC-PLANE,
124 REVERSE-CHARSET, SHORT-NAME, LONG-NAME, DESCRIPTION,
127 CHARSET-ID (integer) is the identification number of the charset.
128 BYTES (integer) is the length of multi-byte form of a character in
129 the charset: one of 1, 2, 3, and 4.
130 DIMENSION (integer) is the number of bytes to represent a character of
132 CHARS (integer) is the number of characters in a dimension: 94 or 96.
133 WIDTH (integer) is the number of columns a character in the charset
134 occupies on the screen: one of 0, 1, and 2.
135 DIRECTION (integer) is the rendering direction of characters in the
136 charset when rendering. If 0, render from left to right, else
137 render from right to left.
138 LEADING-CODE-BASE (integer) is the base leading-code for the
140 LEADING-CODE-EXT (integer) is the extended leading-code for the
141 charset. All charsets of less than 0xA0 has the value 0.
142 ISO-FINAL-CHAR (character) is the final character of the
143 corresponding ISO 2022 charset. If the charset is not assigned
144 any final character, the value is -1.
145 ISO-GRAPHIC-PLANE (integer) is the graphic plane to be invoked
146 while encoding to variants of ISO 2022 coding system, one of the
147 following: 0/graphic-plane-left(GL), 1/graphic-plane-right(GR).
148 If the charset is not assigned any final character, the value is -1.
149 REVERSE-CHARSET (integer) is the charset which differs only in
150 LEFT-TO-RIGHT value from the charset. If there's no such a
151 charset, the value is -1.
152 SHORT-NAME (string) is the short name to refer to the charset.
153 LONG-NAME (string) is the long name to refer to the charset
154 DESCRIPTION (string) is the description string of the charset.
155 PLIST (property list) may contain any type of information a user
156 want to put and get by functions `put-charset-property' and
157 `get-charset-property' respectively."
158 (get charset 'charset))
160 ;; It is better not to use backquote in this file,
161 ;; because that makes a bootstrapping problem
162 ;; if you need to recompile all the Lisp files using interpreted code.
164 (defmacro charset-id (charset)
165 "Return charset identification number of CHARSET."
166 (if (charset-quoted-standard-p charset)
167 (aref (charset-info (nth 1 charset)) 0)
168 (list 'aref (list 'charset-info charset) 0)))
170 (defmacro charset-bytes (charset)
171 "Return bytes of CHARSET.
172 See the function `charset-info' for more detail."
173 (if (charset-quoted-standard-p charset)
174 (aref (charset-info (nth 1 charset)) 1)
175 (list 'aref (list 'charset-info charset) 1)))
177 (defmacro charset-dimension (charset)
178 "Return dimension of CHARSET.
179 See the function `charset-info' for more detail."
180 (if (charset-quoted-standard-p charset)
181 (aref (charset-info (nth 1 charset)) 2)
182 (list 'aref (list 'charset-info charset) 2)))
184 (defmacro charset-chars (charset)
185 "Return character numbers contained in a dimension of CHARSET.
186 See the function `charset-info' for more detail."
187 (if (charset-quoted-standard-p charset)
188 (aref (charset-info (nth 1 charset)) 3)
189 (list 'aref (list 'charset-info charset) 3)))
191 (defmacro charset-width (charset)
192 "Return width (how many column occupied on a screen) of CHARSET.
193 See the function `charset-info' for more detail."
194 (if (charset-quoted-standard-p charset)
195 (aref (charset-info (nth 1 charset)) 4)
196 (list 'aref (list 'charset-info charset) 4)))
198 (defmacro charset-direction (charset)
199 "Return direction of CHARSET.
200 See the function `charset-info' for more detail."
201 (if (charset-quoted-standard-p charset)
202 (aref (charset-info (nth 1 charset)) 5)
203 (list 'aref (list 'charset-info charset) 5)))
205 (defmacro charset-iso-final-char (charset)
206 "Return final char of CHARSET.
207 See the function `charset-info' for more detail."
208 (if (charset-quoted-standard-p charset)
209 (aref (charset-info (nth 1 charset)) 8)
210 (list 'aref (list 'charset-info charset) 8)))
212 (defmacro charset-iso-graphic-plane (charset)
213 "Return graphic plane of CHARSET.
214 See the function `charset-info' for more detail."
215 (if (charset-quoted-standard-p charset)
216 (aref (charset-info (nth 1 charset)) 9)
217 (list 'aref (list 'charset-info charset) 9)))
219 (defmacro charset-reverse-charset (charset)
220 "Return reverse charset of CHARSET.
221 See the function `charset-info' for more detail."
222 (if (charset-quoted-standard-p charset)
223 (aref (charset-info (nth 1 charset)) 10)
224 (list 'aref (list 'charset-info charset) 10)))
226 (defmacro charset-short-name (charset)
227 "Return short name of CHARSET.
228 See the function `charset-info' for more detail."
229 (if (charset-quoted-standard-p charset)
230 (aref (charset-info (nth 1 charset)) 11)
231 (list 'aref (list 'charset-info charset) 11)))
233 (defmacro charset-long-name (charset)
234 "Return long name of CHARSET.
235 See the function `charset-info' for more detail."
236 (if (charset-quoted-standard-p charset)
237 (aref (charset-info (nth 1 charset)) 12)
238 (list 'aref (list 'charset-info charset) 12)))
240 (defmacro charset-description (charset)
241 "Return description of CHARSET.
242 See the function `charset-info' for more detail."
243 (if (charset-quoted-standard-p charset)
244 (aref (charset-info (nth 1 charset)) 13)
245 (list 'aref (list 'charset-info charset) 13)))
247 (defmacro charset-plist (charset)
248 "Return list charset property of CHARSET.
249 See the function `charset-info' for more detail."
251 (if (charset-quoted-standard-p charset)
252 (charset-info (nth 1 charset))
253 (list 'charset-info charset))
256 (defun set-charset-plist (charset plist)
257 "Set CHARSET's property list to PLIST, and return PLIST."
258 (aset (charset-info charset) 14 plist))
260 (defun make-char (charset &optional code1 code2)
261 "Return a character of CHARSET whose position codes are CODE1 and CODE2.
262 CODE1 and CODE2 are optional, but if you don't supply
263 sufficient position codes, return a generic character which stands for
264 all characters or group of characters in the character set.
265 A generic character can be used to index a char table (e.g. syntax-table).
267 Such character sets as ascii, eight-bit-control, and eight-bit-graphic
268 don't have corresponding generic characters. If CHARSET is one of
269 them and you don't supply CODE1, return the character of the smallest
272 If CODE1 or CODE2 are invalid (out of range), this function signals an
273 error. However, the eighth bit of both CODE1 and CODE2 is zeroed
274 before they are used to index CHARSET. Thus you may use, say, the
275 actual ISO 8859 character code rather than subtracting 128, as you
276 would need to index the corresponding Emacs charset."
277 (make-char-internal (charset-id charset) code1 code2))
279 (put 'make-char 'byte-compile
282 (let ((charset (nth 1 form)))
283 (if (charset-quoted-standard-p charset)
284 (byte-compile-normal-call
285 (cons 'make-char-internal
286 (cons (charset-id (nth 1 charset)) (nthcdr 2 form))))
287 (byte-compile-normal-call
288 (cons 'make-char-internal
289 (cons (list 'charset-id charset) (nthcdr 2 form)))))))))
291 (defun charset-list ()
292 "Return list of charsets ever defined.
294 This function is provided for backward compatibility.
295 Now we have the variable `charset-list'."
298 (defsubst generic-char-p (char)
299 "Return t if and only if CHAR is a generic character.
300 See also the documentation of `make-char'."
302 (let ((l (split-char char)))
303 (and (or (= (nth 1 l) 0) (eq (nth 2 l) 0))
304 (not (eq (car l) 'composition))))))
306 (defun decode-char (ccs code-point &optional restriction)
307 "Return character specified by coded character set CCS and CODE-POINT in it.
308 Return nil if such a character is not supported.
309 Currently the only supported coded character set is `ucs' (ISO/IEC
310 10646: Universal Multi-Octet Coded Character Set), and the result is
311 translated through the translation-table named
312 `utf-translation-table-for-decode' or the translation-hash-table named
313 `utf-subst-table-for-decode'.
315 Optional argument RESTRICTION specifies a way to map the pair of CCS
316 and CODE-POINT to a character. Currently not supported and just ignored."
319 (or (utf-lookup-subst-table-for-decode code-point)
324 (make-char 'latin-iso8859-1 code-point))
325 ((< code-point #x2500)
326 (setq code-point (- code-point #x0100))
327 (make-char 'mule-unicode-0100-24ff
328 (+ (/ code-point 96) 32) (+ (% code-point 96) 32)))
329 ((< code-point #x3400)
330 (setq code-point (- code-point #x2500))
331 (make-char 'mule-unicode-2500-33ff
332 (+ (/ code-point 96) 32) (+ (% code-point 96) 32)))
333 ((and (>= code-point #xe000) (< code-point #x10000))
334 (setq code-point (- code-point #xe000))
335 (make-char 'mule-unicode-e000-ffff
336 (+ (/ code-point 96) 32)
337 (+ (% code-point 96) 32))))))
339 (or (aref (get 'utf-translation-table-for-decode
340 'translation-table) c)
343 (defun encode-char (char ccs &optional restriction)
344 "Return code-point in coded character set CCS that corresponds to CHAR.
345 Return nil if CHAR is not included in CCS.
346 Currently the only supported coded character set is `ucs' (ISO/IEC
347 10646: Universal Multi-Octet Coded Character Set), and CHAR is first
348 translated through the translation-table named
349 `utf-translation-table-for-encode' or the translation-hash-table named
350 `utf-subst-table-for-encode'.
352 CHAR should be in one of these charsets:
353 ascii, latin-iso8859-1, mule-unicode-0100-24ff, mule-unicode-2500-33ff,
354 mule-unicode-e000-ffff, eight-bit-control
355 Otherwise, return nil.
357 Optional argument RESTRICTION specifies a way to map CHAR to a
358 code-point in CCS. Currently not supported and just ignored."
359 (let* ((split (split-char char))
360 (charset (car split))
363 (or (utf-lookup-subst-table-for-encode char)
364 (let ((table (get 'utf-translation-table-for-encode
365 'translation-table)))
366 (setq trans (aref table char))
368 (setq split (split-char trans)
369 charset (car split)))
370 (cond ((eq charset 'ascii)
372 ((eq charset 'latin-iso8859-1)
373 (+ (nth 1 split) 128))
374 ((eq charset 'mule-unicode-0100-24ff)
375 (+ #x0100 (+ (* (- (nth 1 split) 32) 96)
376 (- (nth 2 split) 32))))
377 ((eq charset 'mule-unicode-2500-33ff)
378 (+ #x2500 (+ (* (- (nth 1 split) 32) 96)
379 (- (nth 2 split) 32))))
380 ((eq charset 'mule-unicode-e000-ffff)
381 (+ #xe000 (+ (* (- (nth 1 split) 32) 96)
382 (- (nth 2 split) 32))))
383 ((eq charset 'eight-bit-control)
387 ;; Coding system stuff
389 ;; Coding system is a symbol that has the property `coding-system'.
391 ;; The value of the property `coding-system' is a vector of the
393 ;; [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
394 ;; We call this vector as coding-spec. See comments in src/coding.c
397 (defconst coding-spec-type-idx 0)
398 (defconst coding-spec-mnemonic-idx 1)
399 (defconst coding-spec-doc-string-idx 2)
400 (defconst coding-spec-plist-idx 3)
401 (defconst coding-spec-flags-idx 4)
403 ;; PLIST is a property list of a coding system. To share PLIST among
404 ;; alias coding systems, a coding system has PLIST in coding-spec
405 ;; instead of having it in normal property list of Lisp symbol.
406 ;; Here's a list of coding system properties currently being used.
410 ;; The value is a coding category the coding system belongs to. The
411 ;; function `make-coding-system' sets this value automatically
412 ;; unless its argument PROPERTIES specifies this property.
414 ;; o alias-coding-systems
416 ;; The value is a list of coding systems of the same alias group. The
417 ;; first element is the coding system made at first, which we call as
418 ;; `base coding system'. The function `make-coding-system' sets this
419 ;; value automatically and `define-coding-system-alias' updates it.
421 ;; See the documentation of make-coding-system for the meanings of the
422 ;; following properties.
424 ;; o post-read-conversion
425 ;; o pre-write-conversion
426 ;; o translation-table-for-decode
427 ;; o translation-table-for-encode
431 ;; o valid-codes (meaningful only for a coding system based on CCL)
434 (defsubst coding-system-spec (coding-system)
435 "Return coding-spec of CODING-SYSTEM."
436 (get (check-coding-system coding-system) 'coding-system))
438 (defun coding-system-type (coding-system)
439 "Return the coding type of CODING-SYSTEM.
440 A coding type is an integer value indicating the encoding method
441 of CODING-SYSTEM. See the function `make-coding-system' for more detail."
442 (aref (coding-system-spec coding-system) coding-spec-type-idx))
444 (defun coding-system-mnemonic (coding-system)
445 "Return the mnemonic character of CODING-SYSTEM.
446 The mnemonic character of a coding system is used in mode line
447 to indicate the coding system. If the arg is nil, return ?-."
448 (let ((spec (coding-system-spec coding-system)))
449 (if spec (aref spec coding-spec-mnemonic-idx) ?-)))
451 (defun coding-system-doc-string (coding-system)
452 "Return the documentation string for CODING-SYSTEM."
453 (aref (coding-system-spec coding-system) coding-spec-doc-string-idx))
455 (defun coding-system-plist (coding-system)
456 "Return the property list of CODING-SYSTEM."
457 (aref (coding-system-spec coding-system) coding-spec-plist-idx))
459 (defun coding-system-flags (coding-system)
460 "Return `flags' of CODING-SYSTEM.
461 A `flags' of a coding system is a vector of length 32 indicating detailed
462 information of a coding system. See the function `make-coding-system'
464 (aref (coding-system-spec coding-system) coding-spec-flags-idx))
466 (defun coding-system-get (coding-system prop)
467 "Extract a value from CODING-SYSTEM's property list for property PROP."
468 (plist-get (coding-system-plist coding-system) prop))
470 (defun coding-system-put (coding-system prop val)
471 "Change value in CODING-SYSTEM's property list PROP to VAL."
472 (let ((plist (coding-system-plist coding-system)))
474 (plist-put plist prop val)
475 (aset (coding-system-spec coding-system) coding-spec-plist-idx
478 (defun coding-system-category (coding-system)
479 "Return the coding category of CODING-SYSTEM.
480 See also `coding-category-list'."
481 (coding-system-get coding-system 'coding-category))
483 (defun coding-system-base (coding-system)
484 "Return the base coding system of CODING-SYSTEM.
485 A base coding system is what made by `make-coding-system'.
486 Any alias nor subsidiary coding systems are not base coding system."
487 (car (coding-system-get coding-system 'alias-coding-systems)))
489 (defalias 'coding-system-parent 'coding-system-base)
490 (make-obsolete 'coding-system-parent 'coding-system-base "20.3")
492 ;; Coding system also has a property `eol-type'.
494 ;; This property indicates how the coding system handles end-of-line
495 ;; format. The value is integer 0, 1, 2, or a vector of three coding
496 ;; systems. Each integer value 0, 1, and 2 indicates the format of
497 ;; end-of-line LF, CRLF, and CR respectively. A vector value
498 ;; indicates that the format of end-of-line should be detected
499 ;; automatically. Nth element of the vector is the subsidiary coding
500 ;; system whose `eol-type' property is N.
502 (defun coding-system-eol-type (coding-system)
503 "Return eol-type of CODING-SYSTEM.
504 An eol-type is integer 0, 1, 2, or a vector of coding systems.
506 Integer values 0, 1, and 2 indicate a format of end-of-line; LF,
507 CRLF, and CR respectively.
509 A vector value indicates that a format of end-of-line should be
510 detected automatically. Nth element of the vector is the subsidiary
511 coding system whose eol-type is N."
512 (get coding-system 'eol-type))
514 (defun coding-system-eol-type-mnemonic (coding-system)
515 "Return the string indicating end-of-line format of CODING-SYSTEM."
516 (let* ((eol-type (coding-system-eol-type coding-system))
517 (val (cond ((eq eol-type 0) eol-mnemonic-unix)
518 ((eq eol-type 1) eol-mnemonic-dos)
519 ((eq eol-type 2) eol-mnemonic-mac)
520 (t eol-mnemonic-undecided))))
523 (char-to-string val))))
525 (defun coding-system-lessp (x y)
526 (cond ((eq x 'no-conversion) t)
527 ((eq y 'no-conversion) nil)
528 ((eq x 'emacs-mule) t)
529 ((eq y 'emacs-mule) nil)
530 ((eq x 'undecided) t)
531 ((eq y 'undecided) nil)
532 (t (let ((c1 (coding-system-mnemonic x))
533 (c2 (coding-system-mnemonic y)))
534 (or (< (downcase c1) (downcase c2))
535 (and (not (> (downcase c1) (downcase c2)))
538 (defun coding-system-equal (coding-system-1 coding-system-2)
539 "Return t if and only if CODING-SYSTEM-1 and CODING-SYSTEM-2 are identical.
540 Two coding systems are identical if two symbols are equal
541 or one is an alias of the other."
542 (or (eq coding-system-1 coding-system-2)
543 (and (equal (coding-system-spec coding-system-1)
544 (coding-system-spec coding-system-2))
545 (let ((eol-type-1 (coding-system-eol-type coding-system-1))
546 (eol-type-2 (coding-system-eol-type coding-system-2)))
547 (or (eq eol-type-1 eol-type-2)
548 (and (vectorp eol-type-1) (vectorp eol-type-2)))))))
550 (defun add-to-coding-system-list (coding-system)
551 "Add CODING-SYSTEM to `coding-system-list' while keeping it sorted."
552 (if (or (null coding-system-list)
553 (coding-system-lessp coding-system (car coding-system-list)))
554 (setq coding-system-list (cons coding-system coding-system-list))
555 (let ((len (length coding-system-list))
556 mid (tem coding-system-list))
558 (setq mid (nthcdr (/ len 2) tem))
559 (if (coding-system-lessp (car mid) coding-system)
561 len (- len (/ len 2)))
562 (setq len (/ len 2))))
563 (setcdr tem (cons coding-system (cdr tem))))))
565 (defun coding-system-list (&optional base-only)
566 "Return a list of all existing non-subsidiary coding systems.
567 If optional arg BASE-ONLY is non-nil, only base coding systems are listed.
568 The value doesn't include subsidiary coding systems which are what
569 made from bases and aliases automatically for various end-of-line
570 formats (e.g. iso-latin-1-unix, koi8-r-dos)."
571 (let* ((codings (copy-sequence coding-system-list))
572 (tail (cons nil codings)))
573 ;; Remove subsidiary coding systems (eol variants) and alias
574 ;; coding systems (if necessary).
576 (let* ((coding (car (cdr tail)))
577 (aliases (coding-system-get coding 'alias-coding-systems)))
579 ;; CODING is an eol variant if not in ALIASES.
580 (not (memq coding aliases))
581 ;; CODING is an alias if it is not car of ALIASES.
582 (and base-only (not (eq coding (car aliases)))))
583 (setcdr tail (cdr (cdr tail)))
584 (setq tail (cdr tail)))))
587 (defun map-charset-chars (func charset)
588 "Use FUNC to map over all characters in CHARSET for side effects.
589 FUNC is a function of two args, the start and end (inclusive) of a
590 character code range. Thus FUNC should iterate over [START, END]."
591 (let* ((dim (charset-dimension charset))
592 (chars (charset-chars charset))
593 (start (if (= chars 94)
598 (make-char charset start)
599 (make-char charset (+ start chars -1)))
602 (make-char charset (+ i start) start)
603 (make-char charset (+ i start) (+ start chars -1)))))))
605 (defalias 'register-char-codings 'ignore "")
606 (make-obsolete 'register-char-codings
607 "it exists just for backward compatibility, and does nothing."
610 (defconst char-coding-system-table nil
611 "This is an obsolete variable.
612 It exists just for backward compatibility, and the value is always nil.")
614 (defun make-subsidiary-coding-system (coding-system)
615 "Make subsidiary coding systems (eol-type variants) of CODING-SYSTEM."
616 (let ((coding-spec (coding-system-spec coding-system))
617 (subsidiaries (vector (intern (format "%s-unix" coding-system))
618 (intern (format "%s-dos" coding-system))
619 (intern (format "%s-mac" coding-system))))
623 (put (aref subsidiaries i) 'coding-system coding-spec)
624 (put (aref subsidiaries i) 'eol-type i)
625 (add-to-coding-system-list (aref subsidiaries i))
626 (setq coding-system-alist
627 (cons (list (symbol-name (aref subsidiaries i)))
628 coding-system-alist))
632 (defun transform-make-coding-system-args (name type &optional doc-string props)
633 "For internal use only.
634 Transform XEmacs style args for `make-coding-system' to Emacs style.
635 Value is a list of transformed arguments."
636 (let ((mnemonic (string-to-char (or (plist-get props 'mnemonic) "?")))
637 (eol-type (plist-get props 'eol-type))
640 ((eq eol-type 'lf) (setq eol-type 'unix))
641 ((eq eol-type 'crlf) (setq eol-type 'dos))
642 ((eq eol-type 'cr) (setq eol-type 'mac)))
643 (if (setq tmp (plist-get props 'post-read-conversion))
644 (setq properties (plist-put properties 'post-read-conversion tmp)))
645 (if (setq tmp (plist-get props 'pre-write-conversion))
646 (setq properties (plist-put properties 'pre-write-conversion tmp)))
648 ((eq type 'shift-jis)
649 `(,name 1 ,mnemonic ,doc-string () ,properties ,eol-type))
650 ((eq type 'iso2022) ; This is not perfect.
651 (if (plist-get props 'escape-quoted)
652 (error "escape-quoted is not supported: %S"
653 `(,name ,type ,doc-string ,props)))
654 (let ((g0 (plist-get props 'charset-g0))
655 (g1 (plist-get props 'charset-g1))
656 (g2 (plist-get props 'charset-g2))
657 (g3 (plist-get props 'charset-g3))
660 (eq (cadr (assoc 'latin-jisx0201
661 (plist-get props 'input-charset-conversion)))
663 (eq (cadr (assoc 'ascii
664 (plist-get props 'output-charset-conversion)))
668 (eq (cadr (assoc 'japanese-jisx0208-1978
669 (plist-get props 'input-charset-conversion)))
671 (eq (cadr (assoc 'japanese-jisx0208
672 (plist-get props 'output-charset-conversion)))
673 'japanese-jisx0208-1978))))
675 (if (plist-get props 'force-g0-on-output)
679 (if (plist-get props 'force-g1-on-output)
683 (if (plist-get props 'force-g2-on-output)
687 (if (plist-get props 'force-g3-on-output)
690 `(,name 2 ,mnemonic ,doc-string
692 ,(plist-get props 'short)
693 ,(not (plist-get props 'no-ascii-eol))
694 ,(not (plist-get props 'no-ascii-cntl))
695 ,(plist-get props 'seven)
697 ,(not (plist-get props 'lock-shift))
700 ,(plist-get props 'no-iso6429)
702 ,properties ,eol-type)))
704 `(,name 3 ,mnemonic ,doc-string () ,properties ,eol-type))
706 `(,name 4 ,mnemonic ,doc-string
707 (,(plist-get props 'decode) . ,(plist-get props 'encode))
708 ,properties ,eol-type))
710 (error "unsupported XEmacs style make-coding-style arguments: %S"
711 `(,name ,type ,doc-string ,props))))))
713 (defun make-coding-system (coding-system type mnemonic doc-string
718 "Define a new coding system CODING-SYSTEM (symbol).
719 Remaining arguments are TYPE, MNEMONIC, DOC-STRING, FLAGS (optional),
720 and PROPERTIES (optional) which construct a coding-spec of CODING-SYSTEM
721 in the following format:
722 [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
724 TYPE is an integer value indicating the type of the coding system as follows:
725 0: Emacs internal format,
726 1: Shift-JIS (or MS-Kanji) used mainly on Japanese PCs,
727 2: ISO-2022 including many variants,
728 3: Big5 used mainly on Chinese PCs,
729 4: private, CCL programs provide encoding/decoding algorithm,
730 5: Raw-text, which means that text contains random 8-bit codes.
732 MNEMONIC is a character to be displayed on mode line for the coding system.
734 DOC-STRING is a documentation string for the coding system.
736 FLAGS specifies more detailed information of the coding system as follows:
738 If TYPE is 2 (ISO-2022), FLAGS is a list of these elements:
739 CHARSET0, CHARSET1, CHARSET2, CHARSET3, SHORT-FORM,
740 ASCII-EOL, ASCII-CNTL, SEVEN, LOCKING-SHIFT, SINGLE-SHIFT,
741 USE-ROMAN, USE-OLDJIS, NO-ISO6429, INIT-BOL, DESIGNATION-BOL,
742 SAFE, ACCEPT-LATIN-EXTRA-CODE.
743 CHARSETn are character sets initially designated to Gn graphic registers.
744 If CHARSETn is nil, Gn is never used.
745 If CHARSETn is t, Gn can be used but nothing designated initially.
746 If CHARSETn is a list of character sets, those character sets are
747 designated to Gn on output, but nothing designated to Gn initially.
748 But, character set `ascii' can be designated only to G0.
749 SHORT-FORM non-nil means use short designation sequence on output.
750 ASCII-EOL non-nil means designate ASCII to g0 at end of line on output.
751 ASCII-CNTL non-nil means designate ASCII to g0 before control codes and
753 SEVEN non-nil means use 7-bit code only on output.
754 LOCKING-SHIFT non-nil means use locking-shift.
755 SINGLE-SHIFT non-nil means use single-shift.
756 USE-ROMAN non-nil means designate JIS0201-1976-Roman instead of ASCII.
757 USE-OLDJIS non-nil means designate JIS0208-1976 instead of JIS0208-1983.
758 NO-ISO6429 non-nil means not use ISO6429's direction specification.
759 INIT-BOL non-nil means any designation state is assumed to be reset
760 to initial at each beginning of line on output.
761 DESIGNATION-BOL non-nil means designation sequences should be placed
762 at beginning of line on output.
763 SAFE non-nil means convert unsafe characters to `?' on output.
764 Characters not specified in the property `safe-charsets' nor
765 `safe-chars' are unsafe.
766 ACCEPT-LATIN-EXTRA-CODE non-nil means code-detection routine accepts
767 a code specified in `latin-extra-code-table' (which see) as a valid
768 code of the coding system.
770 If TYPE is 4 (private), FLAGS should be a cons of CCL programs, for
771 decoding and encoding. CCL programs should be specified by their
774 PROPERTIES is an alist of properties vs the corresponding values. The
775 following properties are recognized:
777 o post-read-conversion
779 The value is a function to call after some text is inserted and
780 decoded by the coding system itself and before any functions in
781 `after-insert-functions' are called. The argument of this
782 function is the same as for a function in
783 `after-insert-file-functions', i.e. LENGTH of the text inserted,
784 with point at the head of the text to be decoded.
786 o pre-write-conversion
788 The value is a function to call after all functions in
789 `write-region-annotate-functions' and `buffer-file-format' are
790 called, and before the text is encoded by the coding system itself.
791 The arguments to this function are the same as those of a function
792 in `write-region-annotate-functions', i.e. FROM and TO, specifying
795 o translation-table-for-decode
797 The value is a translation table to be applied on decoding. See
798 the function `make-translation-table' for the format of translation
799 table. This is not applicable to type 4 (CCL-based) coding systems.
801 o translation-table-for-encode
803 The value is a translation table to be applied on encoding. This is
804 not applicable to type 4 (CCL-based) coding systems.
808 The value is a char table. If a character has non-nil value in it,
809 the character is safely supported by the coding system. This
810 overrides the specification of safe-charsets.
814 The value is a list of charsets safely supported by the coding
815 system. The value t means that all charsets Emacs handles are
816 supported. Even if some charset is not in this list, it doesn't
817 mean that the charset can't be encoded in the coding system;
818 it just means that some other receiver of text encoded
819 in the coding system won't be able to handle that charset.
823 The value is a symbol whose name is the `MIME-charset' parameter of
826 o mime-text-unsuitable
828 A non-nil value means the `mime-charset' property names a charset
829 which is unsuitable for the top-level media type \"text\".
831 o valid-codes (meaningful only for a coding system based on CCL)
833 The value is a list to indicate valid byte ranges of the encoded
834 file. Each element of the list is an integer or a cons of integer.
835 In the former case, the integer value is a valid byte code. In the
836 latter case, the integers specify the range of valid byte codes.
838 o composition (meaningful only when TYPE is 0 or 2)
840 If the value is non-nil, the coding system preserves composition
843 These properties are set in PLIST, a property list. This function
844 also sets properties `coding-category' and `alias-coding-systems'
847 EOL-TYPE specifies the EOL type of the coding-system in one of the
850 o symbol (unix, dos, or mac)
852 The symbol `unix' means Unix-like EOL (LF), `dos' means
853 DOS-like EOL (CRLF), and `mac' means MAC-like EOL (CR).
855 o number (0, 1, or 2)
857 The number 0, 1, and 2 mean UNIX, DOS, and MAC-like EOL
860 o vector of coding-systems of length 3
862 The EOL type is detected automatically for the coding system.
863 And, according to the detected EOL type, one of the coding
864 systems in the vector is selected. Elements of the vector
865 corresponds to Unix-like EOL, DOS-like EOL, and Mac-like EOL
868 Kludgy features for backward compatibility:
870 1. If TYPE is 4 and car or cdr of FLAGS is a vector, the vector is
871 treated as a compiled CCL code.
873 2. If PROPERTIES is just a list of character sets, the list is set as
874 a value of `safe-charsets' in PLIST."
876 ;; For compatiblity with XEmacs, we check the type of TYPE. If it
877 ;; is a symbol, perhaps, this function is called with XEmacs-style
878 ;; arguments. Here, try to transform that kind of arguments to
881 (let ((args (transform-make-coding-system-args coding-system type
882 mnemonic doc-string)))
883 (setq coding-system (car args)
885 mnemonic (nth 2 args)
886 doc-string (nth 3 args)
888 properties (nth 5 args)
889 eol-type (nth 6 args))))
891 ;; Set a value of `coding-system' property.
892 (let ((coding-spec (make-vector 5 nil))
893 (no-initial-designation t)
894 (no-alternative-designation t)
895 (accept-latin-extra-code nil)
897 (if (or (not (integerp type)) (< type 0) (> type 5))
898 (error "TYPE argument must be 0..5"))
899 (if (or (not (integerp mnemonic)) (<= mnemonic ? ) (> mnemonic 127))
900 (error "MNEMONIC argument must be an ASCII printable character"))
901 (aset coding-spec coding-spec-type-idx type)
902 (aset coding-spec coding-spec-mnemonic-idx mnemonic)
903 (aset coding-spec coding-spec-doc-string-idx
904 (purecopy (if (stringp doc-string) doc-string "")))
906 (setq coding-category 'coding-category-emacs-mule))
908 (setq coding-category 'coding-category-sjis))
909 ((= type 2) ; ISO2022
911 (vec (make-vector 32 nil))
915 (let ((charset (car fl)))
916 (if (and no-initial-designation
918 (or (charsetp charset)
920 (charsetp (car charset)))))
921 (setq no-initial-designation nil))
922 (if (charsetp charset)
923 (if (= i 1) (setq g1-designation charset))
928 (setq elt (car tail))
930 (setq no-alternative-designation nil)
931 (if (and elt (not (charsetp elt)))
932 (error "Invalid charset: %s" elt)))
933 (setq tail (cdr tail)))
934 (setq g1-designation (car charset)))
937 (setq no-alternative-designation nil)
938 (error "Invalid charset: %s" charset)))))
939 (aset vec i charset))
940 (setq fl (cdr fl) i (1+ i)))
941 (while (and (< i 32) fl)
942 (aset vec i (car fl))
943 (if (and (= i 16) ; ACCEPT-LATIN-EXTRA-CODE
945 (setq accept-latin-extra-code t))
946 (setq fl (cdr fl) i (1+ i)))
947 (aset coding-spec 4 vec)
948 (setq coding-category
949 (if (aref vec 8) ; Use locking-shift.
950 (or (and (aref vec 7) 'coding-category-iso-7-else)
951 'coding-category-iso-8-else)
952 (if (aref vec 7) ; 7-bit only.
953 (if (aref vec 9) ; Use single-shift.
954 'coding-category-iso-7-else
955 (if no-alternative-designation
956 'coding-category-iso-7-tight
957 'coding-category-iso-7))
958 (if (or no-initial-designation
959 (not no-alternative-designation))
960 'coding-category-iso-8-else
961 (if (and (charsetp g1-designation)
962 (= (charset-dimension g1-designation) 2))
963 'coding-category-iso-8-2
964 'coding-category-iso-8-1)))))))
966 (setq coding-category 'coding-category-big5))
967 ((= type 4) ; private
968 (setq coding-category 'coding-category-ccl)
969 (if (not (consp flags))
970 (error "Invalid FLAGS argument for TYPE 4 (CCL)")
971 (let ((decoder (check-ccl-program
973 (intern (format "%s-decoder" coding-system))))
974 (encoder (check-ccl-program
976 (intern (format "%s-encoder" coding-system)))))
977 (if (and decoder encoder)
978 (aset coding-spec 4 (cons decoder encoder))
979 (error "Invalid FLAGS argument for TYPE 4 (CCL)")))))
981 (setq coding-category 'coding-category-raw-text)))
983 (let ((plist (list 'coding-category coding-category
984 'alias-coding-systems (list coding-system))))
985 (if no-initial-designation
986 (plist-put plist 'no-initial-designation t))
988 (or (eq properties t)
989 (not (consp (car properties)))))
990 ;; In the old version, the arg PROPERTIES is a list to be
991 ;; set in PLIST as a value of property `safe-charsets'.
992 (setq properties (list (cons 'safe-charsets properties))))
993 ;; In the current version PROPERTIES is a property list.
994 ;; Reflect it into PLIST one by one while handling safe-chars
996 (let ((safe-charsets (cdr (assq 'safe-charsets properties)))
997 (safe-chars (cdr (assq 'safe-chars properties)))
1000 ;; If only safe-charsets is specified, make a char-table from
1001 ;; it, and store that char-table as the value of `safe-chars'.
1002 (if (and (not safe-chars) safe-charsets)
1004 (if (eq safe-charsets t)
1006 (setq safe-chars (make-char-table 'safe-chars))
1007 (while safe-charsets
1008 (setq charset (car safe-charsets)
1009 safe-charsets (cdr safe-charsets))
1010 (cond ((eq charset 'ascii)) ; just ignore
1011 ((eq charset 'eight-bit-control)
1014 (aset safe-chars i t)
1016 ((eq charset 'eight-bit-graphic)
1019 (aset safe-chars i t)
1022 (aset safe-chars (make-char charset) t))))
1023 (if accept-latin-extra-code
1026 (if (aref latin-extra-code-table i)
1027 (aset safe-chars i t))
1029 (setq l (cons (cons 'safe-chars safe-chars) l))))
1031 (setq prop (car (car l)) val (cdr (car l)) l (cdr l))
1032 (if (eq prop 'safe-chars)
1034 (if (and (symbolp val)
1035 (get val 'translation-table))
1036 (setq safe-chars (get val 'translation-table)))
1037 (setq val safe-chars)))
1038 (plist-put plist prop val)))
1039 ;; The property `coding-category' may have been set differently
1040 ;; through PROPERTIES.
1041 (setq coding-category (plist-get plist 'coding-category))
1042 (aset coding-spec coding-spec-plist-idx plist))
1043 (put coding-system 'coding-system coding-spec)
1044 (put coding-system 'coding-system-define-form nil)
1045 (put coding-category 'coding-systems
1046 (cons coding-system (get coding-category 'coding-systems))))
1048 ;; Next, set a value of `eol-type' property.
1050 ;; If EOL-TYPE is nil, set a vector of subsidiary coding
1051 ;; systems, each corresponds to a coding system for the detected
1053 (setq eol-type (make-subsidiary-coding-system coding-system)))
1055 (cond ((or (eq eol-type 'unix) (null eol-type))
1061 ((or (and (vectorp eol-type)
1062 (= (length eol-type) 3))
1063 (and (numberp eol-type)
1064 (and (>= eol-type 0)
1068 (error "Invalid EOL-TYPE spec:%S" eol-type))))
1069 (put coding-system 'eol-type eol-type)
1071 (define-coding-system-internal coding-system)
1073 ;; At last, register CODING-SYSTEM in `coding-system-list' and
1074 ;; `coding-system-alist'.
1075 (add-to-coding-system-list coding-system)
1076 (setq coding-system-alist (cons (list (symbol-name coding-system))
1077 coding-system-alist))
1079 ;; For a coding system of cateogory iso-8-1 and iso-8-2, create
1080 ;; XXX-with-esc variants.
1081 (let ((coding-category (coding-system-category coding-system)))
1082 (if (or (eq coding-category 'coding-category-iso-8-1)
1083 (eq coding-category 'coding-category-iso-8-2))
1084 (let ((esc (intern (concat (symbol-name coding-system) "-with-esc")))
1085 (doc (format "Same as %s but can handle any charsets by ISO's escape sequences." coding-system))
1086 (safe-charsets (assq 'safe-charsets properties))
1087 (mime-charset (assq 'mime-charset properties)))
1089 (setcdr safe-charsets t)
1090 (setq properties (cons (cons 'safe-charsets t) properties)))
1092 (setcdr mime-charset nil))
1093 (make-coding-system esc type mnemonic doc
1094 (if (listp (car flags))
1095 (cons (append (car flags) '(t)) (cdr flags))
1096 (cons (list (car flags) t) (cdr flags)))
1101 (put 'safe-chars 'char-table-extra-slots 0)
1103 (defun define-coding-system-alias (alias coding-system)
1104 "Define ALIAS as an alias for coding system CODING-SYSTEM."
1105 (put alias 'coding-system (coding-system-spec coding-system))
1106 (put alias 'coding-system-define-form nil)
1107 (add-to-coding-system-list alias)
1108 (setq coding-system-alist (cons (list (symbol-name alias))
1109 coding-system-alist))
1110 (let ((eol-type (coding-system-eol-type coding-system)))
1111 (if (vectorp eol-type)
1113 (nconc (coding-system-get alias 'alias-coding-systems) (list alias))
1114 (put alias 'eol-type (make-subsidiary-coding-system alias)))
1115 (put alias 'eol-type eol-type))))
1117 (defun merge-coding-systems (first second)
1118 "Fill in any unspecified aspects of coding system FIRST from SECOND.
1119 Return the resulting coding system."
1120 (let ((base (coding-system-base second))
1121 (eol (coding-system-eol-type second)))
1122 ;; If FIRST doesn't specify text conversion, merge with that of SECOND.
1123 (if (eq (coding-system-base first) 'undecided)
1124 (setq first (coding-system-change-text-conversion first base)))
1125 ;; If FIRST doesn't specify eol conversion, merge with that of SECOND.
1126 (if (and (vectorp (coding-system-eol-type first))
1127 (numberp eol) (>= eol 0) (<= eol 2))
1128 (setq first (coding-system-change-eol-conversion
1132 (defun autoload-coding-system (symbol form)
1133 "Define SYMBOL as a coding-system that is defined on demand.
1135 FROM is a form to evaluate to define the coding-system."
1136 (put symbol 'coding-system-define-form form)
1137 (setq coding-system-alist (cons (list (symbol-name symbol))
1138 coding-system-alist)))
1140 (defun set-buffer-file-coding-system (coding-system &optional force nomodify)
1141 "Set the file coding-system of the current buffer to CODING-SYSTEM.
1142 This means that when you save the buffer, it will be converted
1143 according to CODING-SYSTEM. For a list of possible values of CODING-SYSTEM,
1144 use \\[list-coding-systems].
1146 If CODING-SYSTEM leaves the text conversion unspecified, or if it
1147 leaves the end-of-line conversion unspecified, FORCE controls what to
1148 do. If FORCE is nil, get the unspecified aspect (or aspects) from the
1149 buffer's previous `buffer-file-coding-system' value (if it is
1150 specified there). Otherwise, leave it unspecified.
1152 This marks the buffer modified so that the succeeding \\[save-buffer]
1153 surely saves the buffer with CODING-SYSTEM. From a program, if you
1154 don't want to mark the buffer modified, specify t for NOMODIFY.
1155 If you know exactly what coding system you want to use,
1156 just set the variable `buffer-file-coding-system' directly."
1157 (interactive "zCoding system for saving file (default, nil): \nP")
1158 (check-coding-system coding-system)
1159 (if (and coding-system buffer-file-coding-system (null force))
1161 (merge-coding-systems coding-system buffer-file-coding-system)))
1162 (setq buffer-file-coding-system coding-system)
1163 ;; This is in case of an explicit call. Normally, `normal-mode' and
1164 ;; `set-buffer-major-mode-hook' take care of setting the table.
1165 (if (fboundp 'ucs-set-table-for-input) ; don't lose when building
1166 (ucs-set-table-for-input))
1168 (set-buffer-modified-p t))
1169 (force-mode-line-update))
1171 (defun revert-buffer-with-coding-system (coding-system &optional force)
1172 "Visit the current buffer's file again using coding system CODING-SYSTEM.
1173 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1175 If CODING-SYSTEM leaves the text conversion unspecified, or if it
1176 leaves the end-of-line conversion unspecified, FORCE controls what to
1177 do. If FORCE is nil, get the unspecified aspect (or aspects) from the
1178 buffer's previous `buffer-file-coding-system' value (if it is
1179 specified there). Otherwise, determine it from the file contents as
1180 usual for visiting a file."
1181 (interactive "zCoding system for visited file (default, nil): \nP")
1182 (check-coding-system coding-system)
1183 (if (and coding-system buffer-file-coding-system (null force))
1185 (merge-coding-systems coding-system buffer-file-coding-system)))
1186 (let ((coding-system-for-read coding-system))
1189 (defun set-file-name-coding-system (coding-system)
1190 "Set coding system for decoding and encoding file names to CODING-SYSTEM.
1191 It actually just set the variable `file-name-coding-system' (which
1192 see) to CODING-SYSTEM."
1193 (interactive "zCoding system for file names (default, nil): ")
1194 (check-coding-system coding-system)
1195 (setq file-name-coding-system coding-system))
1197 (defvar default-terminal-coding-system nil
1198 "Default value for the terminal coding system.
1199 This is normally set according to the selected language environment.
1200 See also the command `set-terminal-coding-system'.")
1202 (defun set-terminal-coding-system (coding-system)
1203 "Set coding system of your terminal to CODING-SYSTEM.
1204 All text output to the terminal will be encoded
1205 with the specified coding system.
1206 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1207 The default is determined by the selected language environment
1208 or by the previous use of this command."
1210 (list (let ((default (if (and (not (terminal-coding-system))
1211 default-terminal-coding-system)
1212 default-terminal-coding-system)))
1214 (format "Coding system for terminal display (default, %s): "
1217 (if (and (not coding-system)
1218 (not (terminal-coding-system)))
1219 (setq coding-system default-terminal-coding-system))
1221 (setq default-terminal-coding-system coding-system))
1222 (set-terminal-coding-system-internal coding-system)
1223 (redraw-frame (selected-frame)))
1225 (defvar default-keyboard-coding-system nil
1226 "Default value of the keyboard coding system.
1227 This is normally set according to the selected language environment.
1228 See also the command `set-keyboard-coding-system'.")
1230 (defun set-keyboard-coding-system (coding-system)
1231 "Set coding system for keyboard input to CODING-SYSTEM.
1232 In addition, this command enables Encoded-kbd minor mode.
1233 \(If CODING-SYSTEM is nil, Encoded-kbd mode is turned off -- see
1234 `encoded-kbd-mode'.)
1235 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1236 The default is determined by the selected language environment
1237 or by the previous use of this command."
1239 (list (let ((default (if (and (not (keyboard-coding-system))
1240 default-keyboard-coding-system)
1241 default-keyboard-coding-system)))
1243 (format "Coding system for keyboard input (default, %s): "
1246 (if (and (not coding-system)
1247 (not (keyboard-coding-system)))
1248 (setq coding-system default-keyboard-coding-system))
1250 (setq default-keyboard-coding-system coding-system))
1251 (set-keyboard-coding-system-internal coding-system)
1252 (setq keyboard-coding-system coding-system)
1253 (encoded-kbd-mode (if coding-system 1 0)))
1255 (defcustom keyboard-coding-system nil
1256 "Specify coding system for keyboard input.
1257 If you set this on a terminal which can't distinguish Meta keys from
1258 8-bit characters, you will have to use ESC to type Meta characters.
1259 See Info node `Specify Coding' and Info node `Single-Byte Character Support'.
1261 On non-windowing terminals, this is set from the locale by default.
1263 Setting this variable directly does not take effect;
1264 use either \\[customize] or \\[set-keyboard-coding-system]."
1265 :type '(coding-system :tag "Coding system")
1266 :link '(info-link "(emacs)Specify Coding")
1267 :link '(info-link "(emacs)Single-Byte Character Support")
1268 :set (lambda (symbol value)
1269 ;; Don't load encoded-kbd-mode unnecessarily.
1270 (if (or value (boundp 'encoded-kbd-mode))
1271 (set-keyboard-coding-system value)
1272 (set-default 'keyboard-coding-system nil))) ; must initialize
1277 (defun set-buffer-process-coding-system (decoding encoding)
1278 "Set coding systems for the process associated with the current buffer.
1279 DECODING is the coding system to be used to decode input from the process,
1280 ENCODING is the coding system to be used to encode output to the process.
1282 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems]."
1284 "zCoding-system for output from the process: \nzCoding-system for input to the process: ")
1285 (let ((proc (get-buffer-process (current-buffer))))
1287 (error "No process")
1288 (check-coding-system decoding)
1289 (check-coding-system encoding)
1290 (set-process-coding-system proc decoding encoding)))
1291 (force-mode-line-update))
1293 (defalias 'set-clipboard-coding-system 'set-selection-coding-system)
1295 (defun set-selection-coding-system (coding-system)
1296 "Make CODING-SYSTEM used for communicating with other X clients.
1297 When sending or receiving text via cut_buffer, selection, and clipboard,
1298 the text is encoded or decoded by CODING-SYSTEM."
1299 (interactive "zCoding system for X selection: ")
1300 (check-coding-system coding-system)
1301 (setq selection-coding-system coding-system))
1303 ;; Coding system lastly specified by the command
1304 ;; set-next-selection-coding-system.
1305 (defvar last-next-selection-coding-system nil)
1307 (defun set-next-selection-coding-system (coding-system)
1308 "Make CODING-SYSTEM used for the next communication with other X clients.
1309 This setting is effective for the next communication only."
1311 (list (read-coding-system
1312 (if last-next-selection-coding-system
1313 (format "Coding system for the next X selection (default, %S): "
1314 last-next-selection-coding-system)
1315 "Coding system for the next X selection: ")
1316 last-next-selection-coding-system)))
1318 (setq last-next-selection-coding-system coding-system)
1319 (setq coding-system last-next-selection-coding-system))
1320 (check-coding-system coding-system)
1322 (setq next-selection-coding-system coding-system))
1324 (defun set-coding-priority (arg)
1325 "Set priority of coding categories according to ARG.
1326 ARG is a list of coding categories ordered by priority."
1328 (current-list (copy-sequence coding-category-list)))
1329 ;; Check the validity of ARG while deleting coding categories in
1330 ;; ARG from CURRENT-LIST. We assume that CODING-CATEGORY-LIST
1331 ;; contains all coding categories.
1333 (if (or (null (get (car l) 'coding-category-index))
1334 (null (memq (car l) current-list)))
1335 (error "Invalid or duplicated element in argument: %s" arg))
1336 (setq current-list (delq (car l) current-list))
1338 ;; Update `coding-category-list' and return it.
1339 (setq coding-category-list (append arg current-list))
1340 (set-coding-priority-internal)))
1344 (defvar ctext-non-standard-encodings-alist
1345 '(("big5-0" big5 2 (chinese-big5-1 chinese-big5-2))
1346 ("ISO8859-14" iso-8859-14 1 latin-iso8859-14)
1347 ("ISO8859-15" iso-8859-15 1 latin-iso8859-15))
1348 "Alist of non-standard encoding names vs the corresponding usages in CTEXT.
1350 It controls how extended segments of a compound text are handled
1351 by the coding system `compound-text-with-extensions'.
1353 Each element has the form (ENCODING-NAME CODING-SYSTEM N-OCTET CHARSET).
1355 ENCODING-NAME is an encoding name of an \"extended segments\".
1357 CODING-SYSTEM is the coding-system to encode (or decode) the
1358 characters into (or from) the extended segment.
1360 N-OCTET is the number of octets (bytes) that encodes a character
1361 in the segment. It can be 0 (meaning the number of octets per
1362 character is variable), 1, 2, 3, or 4.
1364 CHARSET is a charater set containing characters that are encoded
1365 in the segment. It can be a list of character sets. It can also
1366 be a char-table, in which case characters that have non-nil value
1367 in the char-table are the target.
1369 On decoding CTEXT, all encoding names listed here are recognized.
1371 On encoding CTEXT, encoding names in the variable
1372 `ctext-non-standard-encodings' (which see) and in the information
1373 listed for the current language environment under the key
1374 `ctext-non-standard-encodings' are used.")
1376 (defvar ctext-non-standard-encodings
1378 "List of non-standard encoding names used in extended segments of CTEXT.
1379 Each element must be one of the names listed in the variable
1380 `ctext-non-standard-encodings-alist' (which see).")
1382 (defvar ctext-non-standard-encodings-regexp
1383 (string-to-multibyte
1385 ;; For non-standard encodings.
1386 "\\(\e%/[0-4][\200-\377][\200-\377]\\([^\002]+\\)\002\\)"
1388 ;; For UTF-8 encoding.
1389 "\\(\e%G[^\e]*\e%@\\)")))
1391 ;; Functions to support "Non-Standard Character Set Encodings" defined
1392 ;; by the COMPOUND-TEXT spec. They also support "The UTF-8 encoding"
1393 ;; described in the section 7 of the documentation of COMPOUND-TEXT
1394 ;; distributed with XFree86.
1396 (defun ctext-post-read-conversion (len)
1397 "Decode LEN characters encoded as Compound Text with Extended Segments."
1400 (let ((case-fold-search nil)
1401 (in-workbuf (string= (buffer-name) " *code-converting-work*"))
1402 last-coding-system-used
1405 (narrow-to-region (point) (+ (point) len)))
1407 (set-buffer-multibyte t))
1408 (while (re-search-forward ctext-non-standard-encodings-regexp
1410 (setq pos (match-beginning 0))
1411 (if (match-beginning 1)
1412 ;; ESC % / [0-4] M L --ENCODING-NAME-- \002 --BYTES--
1413 (let* ((M (char-after (+ pos 4)))
1414 (L (char-after (+ pos 5)))
1415 (encoding (match-string 2))
1416 (encoding-info (assoc-string
1418 ctext-non-standard-encodings-alist t))
1419 (coding (if encoding-info
1420 (nth 1 encoding-info)
1421 (setq encoding (intern (downcase encoding)))
1422 (and (coding-system-p encoding)
1424 (setq bytes (- (+ (* (- M 128) 128) (- L 128))
1425 (- (point) (+ pos 6))))
1427 (delete-region pos (point))
1428 (forward-char bytes)
1429 (decode-coding-region (- (point) bytes) (point) coding)))
1430 ;; ESC % G --UTF-8-BYTES-- ESC % @
1432 (delete-region pos (+ pos 3))
1433 (decode-coding-region pos (point) 'utf-8))))
1434 (goto-char (point-min))
1435 (- (point-max) (point)))))
1437 ;; Return a char table of extended segment usage for each character.
1438 ;; Each value of the char table is nil, one of the elements of
1439 ;; `ctext-non-standard-encodings-alist', or the symbol `utf-8'.
1441 (defun ctext-non-standard-encodings-table ()
1442 (let ((table (make-char-table 'translation-table)))
1443 (aset table (make-char 'mule-unicode-0100-24ff) 'utf-8)
1444 (aset table (make-char 'mule-unicode-2500-33ff) 'utf-8)
1445 (aset table (make-char 'mule-unicode-e000-ffff) 'utf-8)
1446 (dolist (encoding (reverse
1448 (get-language-info current-language-environment
1449 'ctext-non-standard-encodings)
1450 ctext-non-standard-encodings)))
1451 (let* ((slot (assoc encoding ctext-non-standard-encodings-alist))
1452 (charset (nth 3 slot)))
1454 (cond ((charsetp charset)
1455 (aset table (make-char charset) slot))
1457 (dolist (elt charset)
1458 (aset table (make-char elt) slot)))
1459 ((char-table-p charset)
1460 (map-char-table #'(lambda (k v)
1461 (if (and v (> k 128)) (aset table k slot)))
1465 (defun ctext-pre-write-conversion (from to)
1466 "Encode characters between FROM and TO as Compound Text w/Extended Segments.
1468 If FROM is a string, or if the current buffer is not the one set up for us
1469 by encode-coding-string, generate a new temp buffer, insert the
1470 text, and convert it in the temporary buffer. Otherwise, convert in-place."
1472 (let ((workbuf (get-buffer-create " *code-conversion-work*")))
1473 ;; Setup a working buffer if necessary.
1474 (cond ((stringp from)
1475 (set-buffer workbuf)
1477 (set-buffer-multibyte (multibyte-string-p from))
1479 ((not (eq (current-buffer) workbuf))
1480 (let ((buf (current-buffer))
1481 (multibyte enable-multibyte-characters))
1482 (set-buffer workbuf)
1484 (set-buffer-multibyte multibyte)
1485 (insert-buffer-substring buf from to)))))
1487 ;; Now we can encode the whole buffer.
1488 (let ((encoding-table (ctext-non-standard-encodings-table))
1489 last-coding-system-used
1490 last-pos last-encoding-info
1491 encoding-info end-pos)
1492 (goto-char (setq last-pos (point-min)))
1493 (setq end-pos (point-marker))
1494 (while (re-search-forward "[^\000-\177]+" nil t)
1495 ;; Found a sequence of non-ASCII characters.
1496 (setq last-pos (match-beginning 0)
1497 last-encoding-info (aref encoding-table (char-after last-pos)))
1498 (set-marker end-pos (match-end 0))
1499 (goto-char (1+ last-pos))
1503 (if (< (point) end-pos)
1504 (aref encoding-table (following-char))))
1505 (unless (eq last-encoding-info encoding-info)
1506 (cond ((consp last-encoding-info)
1507 ;; Encode the previous range using an extended
1509 (let ((encoding-name (car last-encoding-info))
1510 (coding-system (nth 1 last-encoding-info))
1511 (noctets (nth 2 last-encoding-info))
1513 (encode-coding-region last-pos (point) coding-system)
1514 (setq len (+ (length encoding-name) 1
1515 (- (point) last-pos)))
1517 (goto-char last-pos)
1518 (insert (string-to-multibyte
1519 (format "\e%%/%d%c%c%s
\ 2"
1524 ((eq last-encoding-info 'utf-8)
1525 ;; Encode the previous range using UTF-8 encoding
1527 (encode-coding-region last-pos (point) 'mule-utf-8)
1529 (goto-char last-pos)
1532 (setq last-pos (point)
1533 last-encoding-info encoding-info))
1534 (if (< (point) end-pos)
1536 (throw 'tag nil)))))
1537 (set-marker end-pos nil)
1538 (goto-char (point-min))))
1539 ;; Must return nil, as build_annotations_2 expects that.
1544 (defcustom auto-coding-alist
1545 '(("\\.\\(arc\\|zip\\|lzh\\|zoo\\|[jew]ar\\|xpi\\)\\'" . no-conversion)
1546 ("\\.\\(ARC\\|ZIP\\|LZH\\|ZOO\\|[JEW]AR\\|XPI\\)\\'" . no-conversion)
1547 ("\\.\\(sx[dmicw]\\|tar\\|tgz\\)\\'" . no-conversion)
1548 ("\\.\\(gz\\|Z\\|bz\\|bz2\\|gpg\\)\\'" . no-conversion)
1549 ("\\.\\(jpe?g\\|png\\|gif\\|tiff?\\|p[bpgn]m\\)\\'" . no-conversion)
1550 ("/#[^/]+#\\'" . emacs-mule))
1551 "Alist of filename patterns vs corresponding coding systems.
1552 Each element looks like (REGEXP . CODING-SYSTEM).
1553 A file whose name matches REGEXP is decoded by CODING-SYSTEM on reading.
1555 The settings in this alist take priority over `coding:' tags
1556 in the file (see the function `set-auto-coding')
1557 and the contents of `file-coding-system-alist'."
1560 :type '(repeat (cons (regexp :tag "File name regexp")
1561 (symbol :tag "Coding system"))))
1563 (defcustom auto-coding-regexp-alist
1564 '(("^BABYL OPTIONS:[ \t]*-\\*-[ \t]*rmail[ \t]*-\\*-" . no-conversion))
1565 "Alist of patterns vs corresponding coding systems.
1566 Each element looks like (REGEXP . CODING-SYSTEM).
1567 A file whose first bytes match REGEXP is decoded by CODING-SYSTEM on reading.
1569 The settings in this alist take priority over `coding:' tags
1570 in the file (see the function `set-auto-coding')
1571 and the contents of `file-coding-system-alist'."
1574 :type '(repeat (cons (regexp :tag "Regexp")
1575 (symbol :tag "Coding system"))))
1577 ;; See the bottom of this file for built-in auto coding functions.
1578 (defcustom auto-coding-functions '(sgml-xml-auto-coding-function
1579 sgml-html-meta-auto-coding-function)
1580 "A list of functions which attempt to determine a coding system.
1582 Each function in this list should be written to operate on the
1583 current buffer, but should not modify it in any way. The buffer
1584 will contain undecoded text of parts of the file. Each function
1585 should take one argument, SIZE, which says how many
1586 characters (starting from point) it should look at.
1588 If one of these functions succeeds in determining a coding
1589 system, it should return that coding system. Otherwise, it
1592 If a file has a `coding:' tag, that takes precedence over these
1593 functions, so they won't be called at all."
1596 :type '(repeat function))
1598 (defvar set-auto-coding-for-load nil
1599 "Non-nil means look for `load-coding' property instead of `coding'.
1600 This is used for loading and byte-compiling Emacs Lisp files.")
1602 (defun auto-coding-alist-lookup (filename)
1603 "Return the coding system specified by `auto-coding-alist' for FILENAME."
1604 (let ((alist auto-coding-alist)
1605 (case-fold-search (memq system-type '(vax-vms windows-nt ms-dos cygwin)))
1607 (while (and alist (not coding-system))
1608 (if (string-match (car (car alist)) filename)
1609 (setq coding-system (cdr (car alist)))
1610 (setq alist (cdr alist))))
1613 (defun set-auto-coding (filename size)
1614 "Return coding system for a file FILENAME of which SIZE bytes follow point.
1615 These bytes should include at least the first 1k of the file
1616 and the last 3k of the file, but the middle may be omitted.
1618 The function checks FILENAME against the variable `auto-coding-alist'.
1619 If FILENAME doesn't match any entries in the variable, it checks the
1620 contents of the current buffer following point against
1621 `auto-coding-regexp-alist'. If no match is found, it checks for a
1622 `coding:' tag in the first one or two lines following point. If no
1623 `coding:' tag is found, it checks any local variables list in the last
1624 3K bytes out of the SIZE bytes. Finally, if none of these methods
1625 succeed, it checks to see if any function in `auto-coding-functions'
1628 The return value is the specified coding system, or nil if nothing is
1631 The variable `set-auto-coding-function' (which see) is set to this
1632 function by default."
1633 (or (auto-coding-alist-lookup filename)
1634 ;; Try using `auto-coding-regexp-alist'.
1636 (let ((alist auto-coding-regexp-alist)
1638 (while (and alist (not coding-system))
1639 (let ((regexp (car (car alist))))
1640 (when (re-search-forward regexp (+ (point) size) t)
1641 (setq coding-system (cdr (car alist)))))
1642 (setq alist (cdr alist)))
1644 (let* ((case-fold-search t)
1645 (head-start (point))
1646 (head-end (+ head-start (min size 1024)))
1647 (tail-start (+ head-start (max (- size 3072) 0)))
1648 (tail-end (+ head-start size))
1649 coding-system head-found tail-found pos)
1650 ;; Try a short cut by searching for the string "coding:"
1651 ;; and for "unibyte:" at the head and tail of SIZE bytes.
1652 (setq head-found (or (search-forward "coding:" head-end t)
1653 (search-forward "unibyte:" head-end t)))
1654 (if (and head-found (> head-found tail-start))
1655 ;; Head and tail are overlapped.
1656 (setq tail-found head-found)
1657 (goto-char tail-start)
1658 (setq tail-found (or (search-forward "coding:" tail-end t)
1659 (search-forward "unibyte:" tail-end t))))
1661 ;; At first check the head.
1663 (goto-char head-start)
1664 (setq head-end (set-auto-mode-1))
1665 (setq head-start (point))
1666 (when (and head-end (< head-found head-end))
1667 (goto-char head-start)
1668 (when (and set-auto-coding-for-load
1670 "\\(.*;\\)?[ \t]*unibyte:[ \t]*\\([^ ;]+\\)"
1672 (setq coding-system 'raw-text))
1673 (when (and (not coding-system)
1675 "\\(.*;\\)?[ \t]*coding:[ \t]*\\([^ ;]+\\)"
1677 (setq coding-system (intern (match-string 2)))
1678 (or (coding-system-p coding-system)
1679 (setq coding-system nil)))))
1681 ;; If no coding: tag in the head, check the tail.
1682 ;; Here we must pay attention to the case that the end-of-line
1683 ;; is just "\r" and we can't use "^" nor "$" in regexp.
1684 (when (and tail-found (not coding-system))
1685 (goto-char tail-start)
1686 (re-search-forward "[\r\n]\^L" nil t)
1687 (if (re-search-forward
1688 "[\r\n]\\([^[\r\n]*\\)[ \t]*Local Variables:[ \t]*\\([^\r\n]*\\)[\r\n]"
1690 ;; The prefix is what comes before "local variables:" in its
1691 ;; line. The suffix is what comes after "local variables:"
1693 (let* ((prefix (regexp-quote (match-string 1)))
1694 (suffix (regexp-quote (match-string 2)))
1698 ;; N.B. without the \n below, the regexp can
1700 "[ \t]*coding[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1705 "[ \t]*unibyte[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1708 (concat "[\r\n]" prefix "[ \t]*End *:[ \t]*" suffix
1711 (forward-char -1) ; skip back \r or \n.
1712 (re-search-forward re-end tail-end 'move)
1713 (setq tail-end (point))
1715 (when (and set-auto-coding-for-load
1716 (re-search-forward re-unibyte tail-end t))
1717 (setq coding-system 'raw-text))
1718 (when (and (not coding-system)
1719 (re-search-forward re-coding tail-end t))
1720 (setq coding-system (intern (match-string 1)))
1721 (or (coding-system-p coding-system)
1722 (setq coding-system nil))))))
1724 ;; Finally, try all the `auto-coding-functions'.
1725 (let ((funcs auto-coding-functions)
1726 (coding-system nil))
1727 (while (and funcs (not coding-system))
1728 (setq coding-system (condition-case e
1730 (goto-char (point-min))
1731 (funcall (pop funcs) size))
1735 (setq set-auto-coding-function 'set-auto-coding)
1737 ;; This variable is set in these two cases:
1738 ;; (1) A file is read by a coding system specified explicitly.
1739 ;; after-insert-file-set-coding sets this value to
1740 ;; coding-system-for-read.
1741 ;; (2) A buffer is saved.
1742 ;; After writing, basic-save-buffer-1 sets this value to
1743 ;; last-coding-system-used.
1744 ;; This variable is used for decoding in revert-buffer.
1745 (defvar buffer-file-coding-system-explicit nil
1746 "The file coding system explicitly specified for the current buffer.
1747 Internal use only.")
1748 (make-variable-buffer-local 'buffer-file-coding-system-explicit)
1749 (put 'buffer-file-coding-system-explicit 'permanent-local t)
1751 (defun after-insert-file-set-coding (inserted &optional visit)
1752 "Set `buffer-file-coding-system' of current buffer after text is inserted.
1753 INSERTED is the number of characters that were inserted, as figured
1754 in the situation before this function. Return the number of characters
1755 inserted, as figured in the situation after. The two numbers can be
1756 different if the buffer has become unibyte.
1757 The optional second arg VISIT non-nil means that we are visiting a file."
1759 coding-system-for-read
1760 (not (eq coding-system-for-read 'auto-save-coding)))
1761 (setq buffer-file-coding-system-explicit coding-system-for-read))
1762 (if last-coding-system-used
1763 (let ((coding-system
1764 (find-new-buffer-file-coding-system last-coding-system-used))
1765 (modified-p (buffer-modified-p)))
1767 ;; Tell set-buffer-file-coding-system not to mark the file
1768 ;; as modified; we just read it, and it's supposed to be unmodified.
1769 ;; Marking it modified would try to lock it, which would
1770 ;; check the modtime, and we don't want to do that again now.
1771 (set-buffer-file-coding-system coding-system t t)
1772 (if (and enable-multibyte-characters
1773 (or (eq coding-system 'no-conversion)
1774 (eq (coding-system-type coding-system) 5))
1775 ;; If buffer was unmodified and the size is the
1776 ;; same as INSERTED, we must be visiting it.
1778 (= (buffer-size) inserted))
1779 ;; For coding systems no-conversion and raw-text...,
1780 ;; edit the buffer as unibyte.
1781 (let ((pos-marker (copy-marker (+ (point) inserted)))
1783 (buffer-file-name nil))
1784 (set-buffer-multibyte nil)
1785 (setq inserted (- pos-marker (point)))))
1786 (set-buffer-modified-p modified-p))))
1789 ;; The coding-spec and eol-type of coding-system returned is decided
1790 ;; independently in the following order.
1791 ;; 1. That of buffer-file-coding-system locally bound.
1792 ;; 2. That of CODING.
1794 (defun find-new-buffer-file-coding-system (coding)
1795 "Return a coding system for a buffer when a file of CODING is inserted.
1796 The local variable `buffer-file-coding-system' of the current buffer
1797 is set to the returned value.
1798 Return nil if there's no need to set `buffer-file-coding-system'."
1799 (let (local-coding local-eol
1800 found-coding found-eol
1803 ;; Nothing found about coding.
1806 ;; Get information of `buffer-file-coding-system' in LOCAL-EOL
1807 ;; and LOCAL-CODING.
1808 (setq local-eol (coding-system-eol-type buffer-file-coding-system))
1809 (if (null (numberp local-eol))
1810 ;; But eol-type is not yet set.
1811 (setq local-eol nil))
1812 (if (and buffer-file-coding-system
1813 (not (eq (coding-system-type buffer-file-coding-system) t)))
1814 ;; This is not `undecided'.
1815 (setq local-coding (coding-system-base buffer-file-coding-system)))
1817 (if (and (local-variable-p 'buffer-file-coding-system)
1818 local-eol local-coding)
1819 ;; The current buffer has already set full coding-system, we
1820 ;; had better not change it.
1823 (setq found-eol (coding-system-eol-type coding))
1824 (if (null (numberp found-eol))
1825 ;; But eol-type is not found.
1826 ;; If EOL conversions are inhibited, force unix eol-type.
1827 (setq found-eol (if inhibit-eol-conversion 0)))
1828 (if (eq (coding-system-type coding) t)
1829 (setq found-coding 'undecided)
1830 (setq found-coding (coding-system-base coding)))
1832 (if (and (not found-eol) (eq found-coding 'undecided))
1833 ;; No valid coding information found.
1836 ;; Some coding information (eol or text) found.
1838 ;; The local setting takes precedence over the found one.
1839 (setq new-coding (if (local-variable-p 'buffer-file-coding-system)
1840 (or local-coding found-coding)
1841 (or found-coding local-coding)))
1842 (setq new-eol (if (local-variable-p 'buffer-file-coding-system)
1843 (or local-eol found-eol)
1844 (or found-eol local-eol)))
1846 (let ((eol-type (coding-system-eol-type new-coding)))
1847 (if (and (numberp new-eol) (vectorp eol-type))
1848 (aref eol-type new-eol)
1851 (defun modify-coding-system-alist (target-type regexp coding-system)
1852 "Modify one of look up tables for finding a coding system on I/O operation.
1853 There are three of such tables, `file-coding-system-alist',
1854 `process-coding-system-alist', and `network-coding-system-alist'.
1856 TARGET-TYPE specifies which of them to modify.
1857 If it is `file', it affects `file-coding-system-alist' (which see).
1858 If it is `process', it affects `process-coding-system-alist' (which see).
1859 If it is `network', it affects `network-coding-system-alist' (which see).
1861 REGEXP is a regular expression matching a target of I/O operation.
1862 The target is a file name if TARGET-TYPE is `file', a program name if
1863 TARGET-TYPE is `process', or a network service name or a port number
1864 to connect to if TARGET-TYPE is `network'.
1866 CODING-SYSTEM is a coding system to perform code conversion on the I/O
1867 operation, or a cons cell (DECODING . ENCODING) specifying the coding systems
1868 for decoding and encoding respectively,
1869 or a function symbol which, when called, returns such a cons cell."
1870 (or (memq target-type '(file process network))
1871 (error "Invalid target type: %s" target-type))
1872 (or (stringp regexp)
1873 (and (eq target-type 'network) (integerp regexp))
1874 (error "Invalid regular expression: %s" regexp))
1875 (if (symbolp coding-system)
1876 (if (not (fboundp coding-system))
1878 (check-coding-system coding-system)
1879 (setq coding-system (cons coding-system coding-system))))
1880 (check-coding-system (car coding-system))
1881 (check-coding-system (cdr coding-system)))
1882 (cond ((eq target-type 'file)
1883 (let ((slot (assoc regexp file-coding-system-alist)))
1885 (setcdr slot coding-system)
1886 (setq file-coding-system-alist
1887 (cons (cons regexp coding-system)
1888 file-coding-system-alist)))))
1889 ((eq target-type 'process)
1890 (let ((slot (assoc regexp process-coding-system-alist)))
1892 (setcdr slot coding-system)
1893 (setq process-coding-system-alist
1894 (cons (cons regexp coding-system)
1895 process-coding-system-alist)))))
1897 (let ((slot (assoc regexp network-coding-system-alist)))
1899 (setcdr slot coding-system)
1900 (setq network-coding-system-alist
1901 (cons (cons regexp coding-system)
1902 network-coding-system-alist)))))))
1904 (defun decode-coding-inserted-region (from to filename
1905 &optional visit beg end replace)
1906 "Decode the region between FROM and TO as if it is read from file FILENAME.
1907 The idea is that the text between FROM and TO was just inserted somehow.
1908 Optional arguments VISIT, BEG, END, and REPLACE are the same as those
1909 of the function `insert-file-contents'.
1910 Part of the job of this function is setting `buffer-undo-list' appropriately."
1913 (let ((coding coding-system-for-read)
1916 ;; Temporarily turn off undo recording, if we're decoding the
1917 ;; text of a visited file.
1918 (setq buffer-undo-list t)
1919 ;; Otherwise, if we can recognize the undo elt for the insertion,
1920 ;; remove it and get ready to replace it later.
1921 ;; In the mean time, turn off undo recording.
1922 (let ((last (car-safe buffer-undo-list)))
1923 (if (and (consp last) (eql (car last) from) (eql (cdr last) to))
1924 (setq undo-list-saved (cdr buffer-undo-list)
1925 buffer-undo-list t))))
1926 (narrow-to-region from to)
1927 (goto-char (point-min))
1929 (setq coding (funcall set-auto-coding-function
1930 filename (- (point-max) (point-min)))))
1932 (setq coding (car (find-operation-coding-system
1933 'insert-file-contents
1934 filename visit beg end replace))))
1935 (if (coding-system-p coding)
1936 (or enable-multibyte-characters
1938 (coding-system-change-text-conversion coding 'raw-text)))
1941 (decode-coding-region (point-min) (point-max) coding)
1942 (setq last-coding-system-used coding))
1943 ;; If we're decoding the text of a visited file,
1944 ;; the undo list should start out empty.
1946 (setq buffer-undo-list nil)
1947 ;; If we decided to replace the undo entry for the insertion,
1950 (setq buffer-undo-list
1951 (cons (cons from (point-max)) undo-list-saved))))))))
1953 (defun recode-region (start end new-coding coding)
1954 "Re-decode the region (previously decoded by CODING) by NEW-CODING."
1956 (list (region-beginning) (region-end)
1957 (read-coding-system "Text was really in: ")
1958 (let ((coding (or buffer-file-coding-system last-coding-system-used)))
1960 (concat "But was interpreted as"
1961 (if coding (format " (default %S): " coding) ": "))
1963 (or (and new-coding coding)
1964 (error "Coding system not specified"))
1965 ;; Check it before we encode the region.
1966 (check-coding-system new-coding)
1968 (narrow-to-region start end)
1969 (encode-coding-region (point-min) (point-max) coding)
1970 (decode-coding-region (point-min) (point-max) new-coding)))
1972 (defun make-translation-table (&rest args)
1973 "Make a translation table from arguments.
1974 A translation table is a char table intended for character
1975 translation in CCL programs.
1977 Each argument is a list of elements of the form (FROM . TO), where FROM
1978 is a character to be translated to TO.
1980 FROM can be a generic character (see `make-char'). In this case, TO is
1981 a generic character containing the same number of characters, or an
1982 ordinary character. If FROM and TO are both generic characters, all
1983 characters belonging to FROM are translated to characters belonging to TO
1984 without changing their position code(s).
1986 The arguments and forms in each argument are processed in the given
1987 order, and if a previous form already translates TO to some other
1988 character, say TO-ALT, FROM is also translated to TO-ALT."
1989 (let ((table (make-char-table 'translation-table))
1992 (let ((elts (car args)))
1994 (let* ((from (car (car elts)))
1995 (from-i 0) ; degree of freedom of FROM
1996 (from-rev (nreverse (split-char from)))
1997 (to (cdr (car elts)))
1998 (to-i 0) ; degree of freedom of TO
1999 (to-rev (nreverse (split-char to))))
2000 ;; Check numbers of heading 0s in FROM-REV and TO-REV.
2001 (while (eq (car from-rev) 0)
2002 (setq from-i (1+ from-i) from-rev (cdr from-rev)))
2003 (while (eq (car to-rev) 0)
2004 (setq to-i (1+ to-i) to-rev (cdr to-rev)))
2005 (if (and (/= from-i to-i) (/= to-i 0))
2006 (error "Invalid character pair (%d . %d)" from to))
2007 ;; If we have already translated TO to TO-ALT, FROM should
2008 ;; also be translated to TO-ALT. But, this is only if TO
2009 ;; is a generic character or TO-ALT is not a generic
2011 (let ((to-alt (aref table to)))
2013 (or (> to-i 0) (not (generic-char-p to-alt))))
2016 (set-char-table-default table from to)
2017 (aset table from to))
2018 ;; If we have already translated some chars to FROM, they
2019 ;; should also be translated to TO.
2020 (let ((l (assq from revlist)))
2027 (setq l (cdr l)) ))))
2028 ;; Now update REVLIST.
2029 (let ((l (assq to revlist)))
2031 (setcdr l (cons from (cdr l)))
2032 (setq revlist (cons (list to from) revlist)))))
2033 (setq elts (cdr elts))))
2034 (setq args (cdr args)))
2035 ;; Return TABLE just created.
2038 (defun make-translation-table-from-vector (vec)
2039 "Make translation table from decoding vector VEC.
2040 VEC is an array of 256 elements to map unibyte codes to multibyte
2041 characters. Elements may be nil for undefined code points.
2042 See also the variable `nonascii-translation-table'."
2043 (let ((table (make-char-table 'translation-table))
2044 (rev-table (make-char-table 'translation-table))
2047 (setq ch (aref vec i))
2051 (aset rev-table ch i))))
2052 (set-char-table-extra-slot table 0 rev-table)
2055 (defun define-translation-table (symbol &rest args)
2056 "Define SYMBOL as the name of translation table made by ARGS.
2057 This sets up information so that the table can be used for
2058 translations in a CCL program.
2060 If the first element of ARGS is a char-table whose purpose is
2061 `translation-table', just define SYMBOL to name it. (Note that this
2062 function does not bind SYMBOL.)
2064 Any other ARGS should be suitable as arguments of the function
2065 `make-translation-table' (which see).
2067 This function sets properties `translation-table' and
2068 `translation-table-id' of SYMBOL to the created table itself and the
2069 identification number of the table respectively. It also registers
2070 the table in `translation-table-vector'."
2071 (let ((table (if (and (char-table-p (car args))
2072 (eq (char-table-subtype (car args))
2073 'translation-table))
2075 (apply 'make-translation-table args)))
2076 (len (length translation-table-vector))
2079 (put symbol 'translation-table table)
2082 (setq translation-table-vector
2083 (vconcat translation-table-vector (make-vector len nil))))
2084 (let ((slot (aref translation-table-vector id)))
2086 (eq (car slot) symbol))
2088 (aset translation-table-vector id (cons symbol table))
2090 (setq id (1+ id)))))
2091 (put symbol 'translation-table-id id)
2094 (defun translate-region (start end table)
2095 "From START to END, translate characters according to TABLE.
2096 TABLE is a string or a char-table.
2097 If TABLE is a string, the Nth character in it is the mapping
2098 for the character with code N.
2099 If TABLE is a char-table, the element for character N is the mapping
2100 for the character with code N.
2101 It returns the number of characters changed."
2103 (list (region-beginning)
2106 (dotimes (i (length translation-table-vector))
2107 (if (consp (aref translation-table-vector i))
2108 (push (list (symbol-name
2109 (car (aref translation-table-vector i)))) l)))
2111 (error "No translation table defined"))
2113 (setq table (completing-read "Translation table: " l nil t)))
2116 (let ((val (get table 'translation-table)))
2117 (or (char-table-p val)
2118 (error "Invalid translation table name: %s" table))
2120 (translate-region-internal start end table))
2122 (put 'with-category-table 'lisp-indent-function 1)
2124 (defmacro with-category-table (table &rest body)
2125 "Evaluate BODY with category table of current buffer set to TABLE.
2126 The category table of the current buffer is saved, BODY is evaluated,
2127 then the saved table is restored, even in case of an abnormal exit.
2128 Value is what BODY returns."
2129 (let ((old-table (make-symbol "old-table"))
2130 (old-buffer (make-symbol "old-buffer")))
2131 `(let ((,old-table (category-table))
2132 (,old-buffer (current-buffer)))
2135 (set-category-table ,table)
2137 (save-current-buffer
2138 (set-buffer ,old-buffer)
2139 (set-category-table ,old-table))))))
2141 (defun define-translation-hash-table (symbol table)
2142 "Define SYMBOL as the name of the hash translation TABLE for use in CCL.
2144 Analogous to `define-translation-table', but updates
2145 `translation-hash-table-vector' and the table is for use in the CCL
2146 `lookup-integer' and `lookup-character' functions."
2147 (unless (and (symbolp symbol)
2148 (hash-table-p table))
2149 (error "Bad args to define-translation-hash-table"))
2150 (let ((len (length translation-hash-table-vector))
2153 (put symbol 'translation-hash-table table)
2156 (setq translation-hash-table-vector
2157 (vconcat translation-hash-table-vector [nil])))
2158 (let ((slot (aref translation-hash-table-vector id)))
2160 (eq (car slot) symbol))
2162 (aset translation-hash-table-vector id (cons symbol table))
2164 (setq id (1+ id)))))
2165 (put symbol 'translation-hash-table-id id)
2168 ;;; Initialize some variables.
2170 (put 'use-default-ascent 'char-table-extra-slots 0)
2171 (setq use-default-ascent (make-char-table 'use-default-ascent))
2172 (put 'ignore-relative-composition 'char-table-extra-slots 0)
2173 (setq ignore-relative-composition
2174 (make-char-table 'ignore-relative-composition))
2177 ;;; Built-in auto-coding-functions:
2179 (defun sgml-xml-auto-coding-function (size)
2180 "Determine whether the buffer is XML, and if so, its encoding.
2181 This function is intended to be added to `auto-coding-functions'."
2182 (setq size (+ (point) size))
2183 (when (re-search-forward "\\`[[:space:]\n]*<\\?xml" size t)
2184 (let ((end (save-excursion
2186 (re-search-forward "\"\\s-*\\?>" size t))))
2188 (if (re-search-forward "encoding=\"\\(.+?\\)\"" end t)
2189 (let* ((match (match-string 1))
2190 (sym (intern (downcase match))))
2191 (if (coding-system-p sym)
2193 (message "Warning: unknown coding system \"%s\"" match)
2197 (defun sgml-html-meta-auto-coding-function (size)
2198 "If the buffer has an HTML meta tag, use it to determine encoding.
2199 This function is intended to be added to `auto-coding-functions'."
2200 (setq size (min (+ (point) size)
2201 ;; Only search forward 10 lines
2205 (when (and (search-forward "<html" size t)
2206 (re-search-forward "<meta\\s-+http-equiv=\"content-type\"\\s-+content=\"text/\\sw+;\\s-*charset=\\(.+?\\)\"" size t))
2207 (let* ((match (match-string 1))
2208 (sym (intern (downcase match))))
2209 (if (coding-system-p sym)
2211 (message "Warning: unknown coding system \"%s\"" match)
2217 ;;; arch-tag: 9aebaa6e-0e8a-40a9-b857-cb5d04a39e7c
2218 ;;; mule.el ends here