1 ;;; mule.el --- basic commands for mulitilingual environment
3 ;; Copyright (C) 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007
4 ;; Free Software Foundation, Inc.
5 ;; Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
7 ;; National Institute of Advanced Industrial Science and Technology (AIST)
8 ;; Registration Number H14PRO021
10 ;; Keywords: mule, multilingual, character set, coding system
12 ;; This file is part of GNU Emacs.
14 ;; GNU Emacs is free software; you can redistribute it and/or modify
15 ;; it under the terms of the GNU General Public License as published by
16 ;; the Free Software Foundation; either version 2, or (at your option)
19 ;; GNU Emacs is distributed in the hope that it will be useful,
20 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
21 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
22 ;; GNU General Public License for more details.
24 ;; You should have received a copy of the GNU General Public License
25 ;; along with GNU Emacs; see the file COPYING. If not, write to the
26 ;; Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
27 ;; Boston, MA 02110-1301, USA.
33 (defconst mule-version
"5.0 (SAKAKI)" "\
34 Version number and name of this version of MULE (multilingual environment).")
36 (defconst mule-version-date
"1999.12.7" "\
37 Distribution date of this version of MULE (multilingual environment).")
39 (defun load-with-code-conversion (fullname file
&optional noerror nomessage
)
40 "Execute a file of Lisp code named FILE whose absolute name is FULLNAME.
41 The file contents are decoded before evaluation if necessary.
42 If optional third arg NOERROR is non-nil,
43 report no error if FILE doesn't exist.
44 Print messages at start and end of loading unless
45 optional fourth arg NOMESSAGE is non-nil.
46 Return t if file exists."
47 (if (null (file-readable-p fullname
))
49 (signal 'file-error
(list "Cannot open load file" file
)))
50 ;; Read file with code conversion, and then eval.
52 ;; To avoid any autoloading, set default-major-mode to
54 ;; So that we don't get completely screwed if the
55 ;; file is encoded in some complicated character set,
56 ;; read it with real decoding, as a multibyte buffer,
57 ;; even if this is a --unibyte Emacs session.
58 (let ((default-major-mode 'fundamental-mode
)
59 (default-enable-multibyte-characters t
))
60 ;; We can't use `generate-new-buffer' because files.el
62 (get-buffer-create (generate-new-buffer-name " *load*"))))
64 (source (save-match-data (string-match "\\.el\\'" fullname
))))
67 (message "Loading %s (source)..." file
)
68 (message "Loading %s..." file
)))
70 (push file preloaded-file-list
))
72 (let ((load-file-name fullname
)
73 (set-auto-coding-for-load t
)
74 (inhibit-file-name-operation nil
))
75 (with-current-buffer buffer
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
)))
102 (do-after-load-evaluation fullname
))
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 "Return 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
287 (let ((charset (nth 1 form
)))
288 (byte-compile-normal-call
289 (cons 'make-char-internal
290 (cons (if (charset-quoted-standard-p charset
)
291 (charset-id (nth 1 charset
))
292 (list 'charset-id charset
))
293 (nthcdr 2 form
)))))))
295 (defun charset-list ()
296 "Return list of charsets ever defined.
298 This function is provided for backward compatibility.
299 Now we have the variable `charset-list'."
302 (defsubst generic-char-p
(char)
303 "Return t if and only if CHAR is a generic character.
304 See also the documentation of `make-char'."
306 (let ((l (split-char char
)))
307 (and (or (= (nth 1 l
) 0) (eq (nth 2 l
) 0))
308 (not (eq (car l
) 'composition
))))))
310 (defun decode-char (ccs code-point
&optional restriction
)
311 "Return character specified by coded character set CCS and CODE-POINT in it.
312 Return nil if such a character is not supported.
313 Currently the only supported coded character set is `ucs' (ISO/IEC
314 10646: Universal Multi-Octet Coded Character Set), and the result is
315 translated through the translation-table named
316 `utf-translation-table-for-decode', or through the
317 translation-hash-table named `utf-subst-table-for-decode'
318 \(if `utf-translate-cjk-mode' is non-nil).
320 Optional argument RESTRICTION specifies a way to map the pair of CCS
321 and CODE-POINT to a character. Currently not supported and just ignored."
324 (or (and utf-translate-cjk-mode
325 (utf-lookup-subst-table-for-decode code-point
))
330 (make-char 'latin-iso8859-1 code-point
))
331 ((< code-point
#x2500
)
332 (setq code-point
(- code-point
#x0100
))
333 (make-char 'mule-unicode-0100-24ff
334 (+ (/ code-point
96) 32) (+ (% code-point
96) 32)))
335 ((< code-point
#x3400
)
336 (setq code-point
(- code-point
#x2500
))
337 (make-char 'mule-unicode-2500-33ff
338 (+ (/ code-point
96) 32) (+ (% code-point
96) 32)))
339 ((and (>= code-point
#xe000
) (< code-point
#x10000
))
340 (setq code-point
(- code-point
#xe000
))
341 (make-char 'mule-unicode-e000-ffff
342 (+ (/ code-point
96) 32)
343 (+ (% code-point
96) 32))))))
345 (or (aref (get 'utf-translation-table-for-decode
346 'translation-table
) c
)
349 (defun encode-char (char ccs
&optional restriction
)
350 "Return code-point in coded character set CCS that corresponds to CHAR.
351 Return nil if CHAR is not included in CCS.
352 Currently the only supported coded character set is `ucs' (ISO/IEC
353 10646: Universal Multi-Octet Coded Character Set), and CHAR is first
354 translated through the translation-table named
355 `utf-translation-table-for-encode', or through the
356 translation-hash-table named `utf-subst-table-for-encode' \(if
357 `utf-translate-cjk-mode' is non-nil).
359 CHAR should be in one of these charsets:
360 ascii, latin-iso8859-1, mule-unicode-0100-24ff, mule-unicode-2500-33ff,
361 mule-unicode-e000-ffff, eight-bit-control
362 Otherwise, return nil.
364 Optional argument RESTRICTION specifies a way to map CHAR to a
365 code-point in CCS. Currently not supported and just ignored."
366 (let* ((split (split-char char
))
367 (charset (car split
))
370 (or (and utf-translate-cjk-mode
371 (utf-lookup-subst-table-for-encode char
))
372 (let ((table (get 'utf-translation-table-for-encode
373 'translation-table
)))
374 (setq trans
(aref table char
))
376 (setq split
(split-char trans
)
377 charset
(car split
)))
378 (cond ((eq charset
'ascii
)
380 ((eq charset
'latin-iso8859-1
)
381 (+ (nth 1 split
) 128))
382 ((eq charset
'mule-unicode-0100-24ff
)
383 (+ #x0100
(+ (* (- (nth 1 split
) 32) 96)
384 (- (nth 2 split
) 32))))
385 ((eq charset
'mule-unicode-2500-33ff
)
386 (+ #x2500
(+ (* (- (nth 1 split
) 32) 96)
387 (- (nth 2 split
) 32))))
388 ((eq charset
'mule-unicode-e000-ffff
)
389 (+ #xe000
(+ (* (- (nth 1 split
) 32) 96)
390 (- (nth 2 split
) 32))))
391 ((eq charset
'eight-bit-control
)
395 ;; Coding system stuff
397 ;; Coding system is a symbol that has the property `coding-system'.
399 ;; The value of the property `coding-system' is a vector of the
401 ;; [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
402 ;; We call this vector as coding-spec. See comments in src/coding.c
405 (defconst coding-spec-type-idx
0)
406 (defconst coding-spec-mnemonic-idx
1)
407 (defconst coding-spec-doc-string-idx
2)
408 (defconst coding-spec-plist-idx
3)
409 (defconst coding-spec-flags-idx
4)
411 ;; PLIST is a property list of a coding system. To share PLIST among
412 ;; alias coding systems, a coding system has PLIST in coding-spec
413 ;; instead of having it in normal property list of Lisp symbol.
414 ;; Here's a list of coding system properties currently being used.
418 ;; The value is a coding category the coding system belongs to. The
419 ;; function `make-coding-system' sets this value automatically
420 ;; unless its argument PROPERTIES specifies this property.
422 ;; o alias-coding-systems
424 ;; The value is a list of coding systems of the same alias group. The
425 ;; first element is the coding system made at first, which we call as
426 ;; `base coding system'. The function `make-coding-system' sets this
427 ;; value automatically and `define-coding-system-alias' updates it.
429 ;; See the documentation of make-coding-system for the meanings of the
430 ;; following properties.
432 ;; o post-read-conversion
433 ;; o pre-write-conversion
434 ;; o translation-table-for-decode
435 ;; o translation-table-for-encode
439 ;; o valid-codes (meaningful only for a coding system based on CCL)
442 (defsubst coding-system-spec
(coding-system)
443 "Return coding-spec of CODING-SYSTEM."
444 (get (check-coding-system coding-system
) 'coding-system
))
446 (defun coding-system-type (coding-system)
447 "Return the coding type of CODING-SYSTEM.
448 A coding type is an integer value indicating the encoding method
449 of CODING-SYSTEM. See the function `make-coding-system' for more detail."
450 (aref (coding-system-spec coding-system
) coding-spec-type-idx
))
452 (defun coding-system-mnemonic (coding-system)
453 "Return the mnemonic character of CODING-SYSTEM.
454 The mnemonic character of a coding system is used in mode line
455 to indicate the coding system. If the arg is nil, return ?-."
456 (let ((spec (coding-system-spec coding-system
)))
457 (if spec
(aref spec coding-spec-mnemonic-idx
) ?-
)))
459 (defun coding-system-doc-string (coding-system)
460 "Return the documentation string for CODING-SYSTEM."
461 (aref (coding-system-spec coding-system
) coding-spec-doc-string-idx
))
463 (defun coding-system-plist (coding-system)
464 "Return the property list of CODING-SYSTEM."
465 (aref (coding-system-spec coding-system
) coding-spec-plist-idx
))
467 (defun coding-system-flags (coding-system)
468 "Return `flags' of CODING-SYSTEM.
469 A `flags' of a coding system is a vector of length 32 indicating detailed
470 information of a coding system. See the function `make-coding-system'
472 (aref (coding-system-spec coding-system
) coding-spec-flags-idx
))
474 (defun coding-system-get (coding-system prop
)
475 "Extract a value from CODING-SYSTEM's property list for property PROP."
476 (plist-get (coding-system-plist coding-system
) prop
))
478 (defun coding-system-put (coding-system prop val
)
479 "Change value in CODING-SYSTEM's property list PROP to VAL."
480 (let ((plist (coding-system-plist coding-system
)))
482 (plist-put plist prop val
)
483 (aset (coding-system-spec coding-system
) coding-spec-plist-idx
486 (defun coding-system-category (coding-system)
487 "Return the coding category of CODING-SYSTEM.
488 See also `coding-category-list'."
489 (coding-system-get coding-system
'coding-category
))
491 (defun coding-system-base (coding-system)
492 "Return the base coding system of CODING-SYSTEM.
493 A base coding system is what made by `make-coding-system'.
494 Any alias nor subsidiary coding systems are not base coding system."
495 (car (coding-system-get coding-system
'alias-coding-systems
)))
497 ;; Coding system also has a property `eol-type'.
499 ;; This property indicates how the coding system handles end-of-line
500 ;; format. The value is integer 0, 1, 2, or a vector of three coding
501 ;; systems. Each integer value 0, 1, and 2 indicates the format of
502 ;; end-of-line LF, CRLF, and CR respectively. A vector value
503 ;; indicates that the format of end-of-line should be detected
504 ;; automatically. Nth element of the vector is the subsidiary coding
505 ;; system whose `eol-type' property is N.
507 (defun coding-system-eol-type (coding-system)
508 "Return eol-type of CODING-SYSTEM.
509 An eol-type is integer 0, 1, 2, or a vector of coding systems.
511 Integer values 0, 1, and 2 indicate a format of end-of-line; LF,
512 CRLF, and CR respectively.
514 A vector value indicates that a format of end-of-line should be
515 detected automatically. Nth element of the vector is the subsidiary
516 coding system whose eol-type is N."
517 (get coding-system
'eol-type
))
519 (defun coding-system-eol-type-mnemonic (coding-system)
520 "Return the string indicating end-of-line format of CODING-SYSTEM."
521 (let* ((eol-type (coding-system-eol-type coding-system
))
522 (val (cond ((eq eol-type
0) eol-mnemonic-unix
)
523 ((eq eol-type
1) eol-mnemonic-dos
)
524 ((eq eol-type
2) eol-mnemonic-mac
)
525 (t eol-mnemonic-undecided
))))
528 (char-to-string val
))))
530 (defun coding-system-lessp (x y
)
531 (cond ((eq x
'no-conversion
) t
)
532 ((eq y
'no-conversion
) nil
)
533 ((eq x
'emacs-mule
) t
)
534 ((eq y
'emacs-mule
) nil
)
535 ((eq x
'undecided
) t
)
536 ((eq y
'undecided
) nil
)
537 (t (let ((c1 (coding-system-mnemonic x
))
538 (c2 (coding-system-mnemonic y
)))
539 (or (< (downcase c1
) (downcase c2
))
540 (and (not (> (downcase c1
) (downcase c2
)))
543 (defun coding-system-equal (coding-system-1 coding-system-2
)
544 "Return t if and only if CODING-SYSTEM-1 and CODING-SYSTEM-2 are identical.
545 Two coding systems are identical if two symbols are equal
546 or one is an alias of the other."
547 (or (eq coding-system-1 coding-system-2
)
548 (and (equal (coding-system-spec coding-system-1
)
549 (coding-system-spec coding-system-2
))
550 (let ((eol-type-1 (coding-system-eol-type coding-system-1
))
551 (eol-type-2 (coding-system-eol-type coding-system-2
)))
552 (or (eq eol-type-1 eol-type-2
)
553 (and (vectorp eol-type-1
) (vectorp eol-type-2
)))))))
555 (defun add-to-coding-system-list (coding-system)
556 "Add CODING-SYSTEM to `coding-system-list' while keeping it sorted."
557 (if (or (null coding-system-list
)
558 (coding-system-lessp coding-system
(car coding-system-list
)))
559 (setq coding-system-list
(cons coding-system coding-system-list
))
560 (let ((len (length coding-system-list
))
561 mid
(tem coding-system-list
))
563 (setq mid
(nthcdr (/ len
2) tem
))
564 (if (coding-system-lessp (car mid
) coding-system
)
566 len
(- len
(/ len
2)))
567 (setq len
(/ len
2))))
568 (setcdr tem
(cons coding-system
(cdr tem
))))))
570 (defun coding-system-list (&optional base-only
)
571 "Return a list of all existing non-subsidiary coding systems.
572 If optional arg BASE-ONLY is non-nil, only base coding systems are listed.
573 The value doesn't include subsidiary coding systems which are what
574 made from bases and aliases automatically for various end-of-line
575 formats (e.g. iso-latin-1-unix, koi8-r-dos)."
576 (let* ((codings (copy-sequence coding-system-list
))
577 (tail (cons nil codings
)))
578 ;; Remove subsidiary coding systems (eol variants) and alias
579 ;; coding systems (if necessary).
581 (let* ((coding (car (cdr tail
)))
582 (aliases (coding-system-get coding
'alias-coding-systems
)))
584 ;; CODING is an eol variant if not in ALIASES.
585 (not (memq coding aliases
))
586 ;; CODING is an alias if it is not car of ALIASES.
587 (and base-only
(not (eq coding
(car aliases
)))))
588 (setcdr tail
(cdr (cdr tail
)))
589 (setq tail
(cdr tail
)))))
592 (defun map-charset-chars (func charset
)
593 "Use FUNC to map over all characters in CHARSET for side effects.
594 FUNC is a function of two args, the start and end (inclusive) of a
595 character code range. Thus FUNC should iterate over [START, END]."
596 (let* ((dim (charset-dimension charset
))
597 (chars (charset-chars charset
))
598 (start (if (= chars
94)
603 (make-char charset start
)
604 (make-char charset
(+ start chars -
1)))
607 (make-char charset
(+ i start
) start
)
608 (make-char charset
(+ i start
) (+ start chars -
1)))))))
610 (defalias 'register-char-codings
'ignore
"")
611 (make-obsolete 'register-char-codings
612 "it exists just for backward compatibility, and does nothing."
615 (defconst char-coding-system-table nil
616 "This is an obsolete variable.
617 It exists just for backward compatibility, and the value is always nil.")
619 (defun make-subsidiary-coding-system (coding-system)
620 "Make subsidiary coding systems (eol-type variants) of CODING-SYSTEM."
621 (let ((coding-spec (coding-system-spec coding-system
))
622 (subsidiaries (vector (intern (format "%s-unix" coding-system
))
623 (intern (format "%s-dos" coding-system
))
624 (intern (format "%s-mac" coding-system
))))
627 (setq elt
(aref subsidiaries i
))
628 (put elt
'coding-system coding-spec
)
629 (put elt
'eol-type i
)
630 (put elt
'coding-system-define-form nil
)
631 (add-to-coding-system-list elt
)
632 (or (assoc (symbol-name elt
) coding-system-alist
)
633 (setq coding-system-alist
634 (cons (list (symbol-name elt
)) coding-system-alist
))))
637 (defun transform-make-coding-system-args (name type
&optional doc-string props
)
638 "For internal use only.
639 Transform XEmacs style args for `make-coding-system' to Emacs style.
640 Value is a list of transformed arguments."
641 (let ((mnemonic (string-to-char (or (plist-get props
'mnemonic
) "?")))
642 (eol-type (plist-get props
'eol-type
))
645 ((eq eol-type
'lf
) (setq eol-type
'unix
))
646 ((eq eol-type
'crlf
) (setq eol-type
'dos
))
647 ((eq eol-type
'cr
) (setq eol-type
'mac
)))
648 (if (setq tmp
(plist-get props
'post-read-conversion
))
649 (setq properties
(plist-put properties
'post-read-conversion tmp
)))
650 (if (setq tmp
(plist-get props
'pre-write-conversion
))
651 (setq properties
(plist-put properties
'pre-write-conversion tmp
)))
653 ((eq type
'shift-jis
)
654 `(,name
1 ,mnemonic
,doc-string
() ,properties
,eol-type
))
655 ((eq type
'iso2022
) ; This is not perfect.
656 (if (plist-get props
'escape-quoted
)
657 (error "escape-quoted is not supported: %S"
658 `(,name
,type
,doc-string
,props
)))
659 (let ((g0 (plist-get props
'charset-g0
))
660 (g1 (plist-get props
'charset-g1
))
661 (g2 (plist-get props
'charset-g2
))
662 (g3 (plist-get props
'charset-g3
))
665 (eq (cadr (assoc 'latin-jisx0201
666 (plist-get props
'input-charset-conversion
)))
668 (eq (cadr (assoc 'ascii
669 (plist-get props
'output-charset-conversion
)))
673 (eq (cadr (assoc 'japanese-jisx0208-1978
674 (plist-get props
'input-charset-conversion
)))
676 (eq (cadr (assoc 'japanese-jisx0208
677 (plist-get props
'output-charset-conversion
)))
678 'japanese-jisx0208-1978
))))
680 (if (plist-get props
'force-g0-on-output
)
684 (if (plist-get props
'force-g1-on-output
)
688 (if (plist-get props
'force-g2-on-output
)
692 (if (plist-get props
'force-g3-on-output
)
695 `(,name
2 ,mnemonic
,doc-string
697 ,(plist-get props
'short
)
698 ,(not (plist-get props
'no-ascii-eol
))
699 ,(not (plist-get props
'no-ascii-cntl
))
700 ,(plist-get props
'seven
)
702 ,(not (plist-get props
'lock-shift
))
705 ,(plist-get props
'no-iso6429
)
707 ,properties
,eol-type
)))
709 `(,name
3 ,mnemonic
,doc-string
() ,properties
,eol-type
))
711 `(,name
4 ,mnemonic
,doc-string
712 (,(plist-get props
'decode
) .
,(plist-get props
'encode
))
713 ,properties
,eol-type
))
715 (error "unsupported XEmacs style make-coding-style arguments: %S"
716 `(,name
,type
,doc-string
,props
))))))
718 (defun make-coding-system (coding-system type mnemonic doc-string
723 "Define a new coding system CODING-SYSTEM (symbol).
724 Remaining arguments are TYPE, MNEMONIC, DOC-STRING, FLAGS (optional),
725 and PROPERTIES (optional) which construct a coding-spec of CODING-SYSTEM
726 in the following format:
727 [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
729 TYPE is an integer value indicating the type of the coding system as follows:
730 0: Emacs internal format,
731 1: Shift-JIS (or MS-Kanji) used mainly on Japanese PCs,
732 2: ISO-2022 including many variants,
733 3: Big5 used mainly on Chinese PCs,
734 4: private, CCL programs provide encoding/decoding algorithm,
735 5: Raw-text, which means that text contains random 8-bit codes.
737 MNEMONIC is a character to be displayed on mode line for the coding system.
739 DOC-STRING is a documentation string for the coding system.
741 FLAGS specifies more detailed information of the coding system as follows:
743 If TYPE is 2 (ISO-2022), FLAGS is a list of these elements:
744 CHARSET0, CHARSET1, CHARSET2, CHARSET3, SHORT-FORM,
745 ASCII-EOL, ASCII-CNTL, SEVEN, LOCKING-SHIFT, SINGLE-SHIFT,
746 USE-ROMAN, USE-OLDJIS, NO-ISO6429, INIT-BOL, DESIGNATION-BOL,
747 SAFE, ACCEPT-LATIN-EXTRA-CODE.
748 CHARSETn are character sets initially designated to Gn graphic registers.
749 If CHARSETn is nil, Gn is never used.
750 If CHARSETn is t, Gn can be used but nothing designated initially.
751 If CHARSETn is a list of character sets, those character sets are
752 designated to Gn on output, but nothing designated to Gn initially.
753 But, character set `ascii' can be designated only to G0.
754 SHORT-FORM non-nil means use short designation sequence on output.
755 ASCII-EOL non-nil means designate ASCII to g0 at end of line on output.
756 ASCII-CNTL non-nil means designate ASCII to g0 before control codes and
758 SEVEN non-nil means use 7-bit code only on output.
759 LOCKING-SHIFT non-nil means use locking-shift.
760 SINGLE-SHIFT non-nil means use single-shift.
761 USE-ROMAN non-nil means designate JIS0201-1976-Roman instead of ASCII.
762 USE-OLDJIS non-nil means designate JIS0208-1976 instead of JIS0208-1983.
763 NO-ISO6429 non-nil means not use ISO6429's direction specification.
764 INIT-BOL non-nil means any designation state is assumed to be reset
765 to initial at each beginning of line on output.
766 DESIGNATION-BOL non-nil means designation sequences should be placed
767 at beginning of line on output.
768 SAFE non-nil means convert unsafe characters to `?' on output.
769 Characters not specified in the property `safe-charsets' nor
770 `safe-chars' are unsafe.
771 ACCEPT-LATIN-EXTRA-CODE non-nil means code-detection routine accepts
772 a code specified in `latin-extra-code-table' (which see) as a valid
773 code of the coding system.
775 If TYPE is 4 (private), FLAGS should be a cons of CCL programs, for
776 decoding and encoding. CCL programs should be specified by their
779 PROPERTIES is an alist of properties vs the corresponding values. The
780 following properties are recognized:
782 o post-read-conversion
784 The value is a function to call after some text is inserted and
785 decoded by the coding system itself and before any functions in
786 `after-insert-functions' are called. The argument of this
787 function is the same as for a function in
788 `after-insert-file-functions', i.e. LENGTH of the text inserted,
789 with point at the head of the text to be decoded.
791 o pre-write-conversion
793 The value is a function to call after all functions in
794 `write-region-annotate-functions' and `buffer-file-format' are
795 called, and before the text is encoded by the coding system itself.
796 The arguments to this function are the same as those of a function
797 in `write-region-annotate-functions', i.e. FROM and TO, specifying
800 o translation-table-for-decode
802 The value is a translation table to be applied on decoding. See
803 the function `make-translation-table' for the format of translation
804 table. This is not applicable to type 4 (CCL-based) coding systems.
806 o translation-table-for-encode
808 The value is a translation table to be applied on encoding. This is
809 not applicable to type 4 (CCL-based) coding systems.
813 The value is a char table. If a character has non-nil value in it,
814 the character is safely supported by the coding system. This
815 overrides the specification of safe-charsets.
819 The value is a list of charsets safely supported by the coding
820 system. The value t means that all charsets Emacs handles are
821 supported. Even if some charset is not in this list, it doesn't
822 mean that the charset can't be encoded in the coding system;
823 it just means that some other receiver of text encoded
824 in the coding system won't be able to handle that charset.
828 The value is a symbol whose name is the `MIME-charset' parameter of
831 o mime-text-unsuitable
833 A non-nil value means the `mime-charset' property names a charset
834 which is unsuitable for the top-level media type \"text\".
836 o valid-codes (meaningful only for a coding system based on CCL)
838 The value is a list to indicate valid byte ranges of the encoded
839 file. Each element of the list is an integer or a cons of integer.
840 In the former case, the integer value is a valid byte code. In the
841 latter case, the integers specify the range of valid byte codes.
843 o composition (meaningful only when TYPE is 0 or 2)
845 If the value is non-nil, the coding system preserves composition
850 If the value is non-nil, the coding system is not compatible
851 with ASCII, which means it encodes or decodes ASCII character
852 string to the different byte sequence.
854 These properties are set in PLIST, a property list. This function
855 also sets properties `coding-category' and `alias-coding-systems'
858 EOL-TYPE specifies the EOL type of the coding-system in one of the
861 o symbol (unix, dos, or mac)
863 The symbol `unix' means Unix-like EOL (LF), `dos' means
864 DOS-like EOL (CRLF), and `mac' means MAC-like EOL (CR).
866 o number (0, 1, or 2)
868 The number 0, 1, and 2 mean UNIX, DOS, and MAC-like EOL
871 o vector of coding-systems of length 3
873 The EOL type is detected automatically for the coding system.
874 And, according to the detected EOL type, one of the coding
875 systems in the vector is selected. Elements of the vector
876 corresponds to Unix-like EOL, DOS-like EOL, and Mac-like EOL
879 Kludgy features for backward compatibility:
881 1. If TYPE is 4 and car or cdr of FLAGS is a vector, the vector is
882 treated as a compiled CCL code.
884 2. If PROPERTIES is just a list of character sets, the list is set as
885 a value of `safe-charsets' in PLIST."
887 ;; For compatiblity with XEmacs, we check the type of TYPE. If it
888 ;; is a symbol, perhaps, this function is called with XEmacs-style
889 ;; arguments. Here, try to transform that kind of arguments to
892 (let ((args (transform-make-coding-system-args coding-system type
893 mnemonic doc-string
)))
894 (setq coding-system
(car args
)
896 mnemonic
(nth 2 args
)
897 doc-string
(nth 3 args
)
899 properties
(nth 5 args
)
900 eol-type
(nth 6 args
))))
902 ;; Set a value of `coding-system' property.
903 (let ((coding-spec (make-vector 5 nil
))
904 (no-initial-designation t
)
905 (no-alternative-designation t
)
906 (accept-latin-extra-code nil
)
908 (if (or (not (integerp type
)) (< type
0) (> type
5))
909 (error "TYPE argument must be 0..5"))
910 (if (or (not (integerp mnemonic
)) (<= mnemonic ?
) (> mnemonic
127))
911 (error "MNEMONIC argument must be an ASCII printable character"))
912 (aset coding-spec coding-spec-type-idx type
)
913 (aset coding-spec coding-spec-mnemonic-idx mnemonic
)
914 (aset coding-spec coding-spec-doc-string-idx
915 (purecopy (if (stringp doc-string
) doc-string
"")))
917 (setq coding-category
'coding-category-emacs-mule
))
919 (setq coding-category
'coding-category-sjis
))
920 ((= type
2) ; ISO2022
922 (vec (make-vector 32 nil
))
926 (let ((charset (car fl
)))
927 (if (and no-initial-designation
929 (or (charsetp charset
)
931 (charsetp (car charset
)))))
932 (setq no-initial-designation nil
))
933 (if (charsetp charset
)
934 (if (= i
1) (setq g1-designation charset
))
939 (setq elt
(car tail
))
941 (setq no-alternative-designation nil
)
942 (if (and elt
(not (charsetp elt
)))
943 (error "Invalid charset: %s" elt
)))
944 (setq tail
(cdr tail
)))
945 (setq g1-designation
(car charset
)))
948 (setq no-alternative-designation nil
)
949 (error "Invalid charset: %s" charset
)))))
950 (aset vec i charset
))
951 (setq fl
(cdr fl
) i
(1+ i
)))
952 (while (and (< i
32) fl
)
953 (aset vec i
(car fl
))
954 (if (and (= i
16) ; ACCEPT-LATIN-EXTRA-CODE
956 (setq accept-latin-extra-code t
))
957 (setq fl
(cdr fl
) i
(1+ i
)))
958 (aset coding-spec
4 vec
)
959 (setq coding-category
960 (if (aref vec
8) ; Use locking-shift.
961 (or (and (aref vec
7) 'coding-category-iso-7-else
)
962 'coding-category-iso-8-else
)
963 (if (aref vec
7) ; 7-bit only.
964 (if (aref vec
9) ; Use single-shift.
965 'coding-category-iso-7-else
966 (if no-alternative-designation
967 'coding-category-iso-7-tight
968 'coding-category-iso-7
))
969 (if (or no-initial-designation
970 (not no-alternative-designation
))
971 'coding-category-iso-8-else
972 (if (and (charsetp g1-designation
)
973 (= (charset-dimension g1-designation
) 2))
974 'coding-category-iso-8-2
975 'coding-category-iso-8-1
)))))))
977 (setq coding-category
'coding-category-big5
))
978 ((= type
4) ; private
979 (setq coding-category
'coding-category-ccl
)
980 (if (not (consp flags
))
981 (error "Invalid FLAGS argument for TYPE 4 (CCL)")
982 (let ((decoder (check-ccl-program
984 (intern (format "%s-decoder" coding-system
))))
985 (encoder (check-ccl-program
987 (intern (format "%s-encoder" coding-system
)))))
988 (if (and decoder encoder
)
989 (aset coding-spec
4 (cons decoder encoder
))
990 (error "Invalid FLAGS argument for TYPE 4 (CCL)")))))
992 (setq coding-category
'coding-category-raw-text
)))
994 (let ((plist (list 'coding-category coding-category
995 'alias-coding-systems
(list coding-system
))))
996 (if no-initial-designation
997 (plist-put plist
'no-initial-designation t
))
999 (or (eq properties t
)
1000 (not (consp (car properties
)))))
1001 ;; In the old version, the arg PROPERTIES is a list to be
1002 ;; set in PLIST as a value of property `safe-charsets'.
1003 (setq properties
(list (cons 'safe-charsets properties
))))
1004 ;; In the current version PROPERTIES is a property list.
1005 ;; Reflect it into PLIST one by one while handling safe-chars
1007 (let ((safe-charsets (cdr (assq 'safe-charsets properties
)))
1008 (safe-chars (cdr (assq 'safe-chars properties
)))
1011 ;; If only safe-charsets is specified, make a char-table from
1012 ;; it, and store that char-table as the value of `safe-chars'.
1013 (if (and (not safe-chars
) safe-charsets
)
1015 (if (eq safe-charsets t
)
1017 (setq safe-chars
(make-char-table 'safe-chars
))
1018 (while safe-charsets
1019 (setq charset
(car safe-charsets
)
1020 safe-charsets
(cdr safe-charsets
))
1021 (cond ((eq charset
'ascii
)) ; just ignore
1022 ((eq charset
'eight-bit-control
)
1025 (aset safe-chars i t
)
1027 ((eq charset
'eight-bit-graphic
)
1030 (aset safe-chars i t
)
1033 (aset safe-chars
(make-char charset
) t
))))
1034 (if accept-latin-extra-code
1037 (if (aref latin-extra-code-table i
)
1038 (aset safe-chars i t
))
1040 (setq l
(cons (cons 'safe-chars safe-chars
) l
))))
1042 (setq prop
(car (car l
)) val
(cdr (car l
)) l
(cdr l
))
1043 (if (eq prop
'safe-chars
)
1045 (if (and (symbolp val
)
1046 (get val
'translation-table
))
1047 (setq safe-chars
(get val
'translation-table
)))
1048 (setq val safe-chars
)))
1049 (plist-put plist prop val
)))
1050 ;; The property `coding-category' may have been set differently
1051 ;; through PROPERTIES.
1052 (setq coding-category
(plist-get plist
'coding-category
))
1053 (aset coding-spec coding-spec-plist-idx plist
))
1054 (put coding-system
'coding-system coding-spec
)
1055 (put coding-system
'coding-system-define-form nil
)
1056 (put coding-category
'coding-systems
1057 (cons coding-system
(get coding-category
'coding-systems
))))
1059 ;; Next, set a value of `eol-type' property.
1061 ;; If EOL-TYPE is nil, set a vector of subsidiary coding
1062 ;; systems, each corresponds to a coding system for the detected
1064 (setq eol-type
(make-subsidiary-coding-system coding-system
)))
1066 (cond ((or (eq eol-type
'unix
) (null eol-type
))
1072 ((or (and (vectorp eol-type
)
1073 (= (length eol-type
) 3))
1074 (and (numberp eol-type
)
1075 (and (>= eol-type
0)
1079 (error "Invalid EOL-TYPE spec:%S" eol-type
))))
1080 (put coding-system
'eol-type eol-type
)
1082 (define-coding-system-internal coding-system
)
1084 ;; At last, register CODING-SYSTEM in `coding-system-list' and
1085 ;; `coding-system-alist'.
1086 (add-to-coding-system-list coding-system
)
1087 (or (assoc (symbol-name coding-system
) coding-system-alist
)
1088 (setq coding-system-alist
(cons (list (symbol-name coding-system
))
1089 coding-system-alist
)))
1091 ;; For a coding system of cateogory iso-8-1 and iso-8-2, create
1092 ;; XXX-with-esc variants.
1093 (let ((coding-category (coding-system-category coding-system
)))
1094 (if (or (eq coding-category
'coding-category-iso-8-1
)
1095 (eq coding-category
'coding-category-iso-8-2
))
1096 (let ((esc (intern (concat (symbol-name coding-system
) "-with-esc")))
1097 (doc (format "Same as %s but can handle any charsets by ISO's escape sequences." coding-system
))
1098 (safe-charsets (assq 'safe-charsets properties
))
1099 (mime-charset (assq 'mime-charset properties
)))
1101 (setcdr safe-charsets t
)
1102 (setq properties
(cons (cons 'safe-charsets t
) properties
)))
1104 (setcdr mime-charset nil
))
1105 (make-coding-system esc type mnemonic doc
1106 (if (listp (car flags
))
1107 (cons (append (car flags
) '(t)) (cdr flags
))
1108 (cons (list (car flags
) t
) (cdr flags
)))
1113 (put 'safe-chars
'char-table-extra-slots
0)
1115 (defun define-coding-system-alias (alias coding-system
)
1116 "Define ALIAS as an alias for coding system CODING-SYSTEM."
1117 (put alias
'coding-system
(coding-system-spec coding-system
))
1118 (put alias
'coding-system-define-form nil
)
1119 (add-to-coding-system-list alias
)
1120 (or (assoc (symbol-name alias
) coding-system-alist
)
1121 (setq coding-system-alist
(cons (list (symbol-name alias
))
1122 coding-system-alist
)))
1123 (let ((eol-type (coding-system-eol-type coding-system
)))
1124 (if (vectorp eol-type
)
1126 (nconc (coding-system-get alias
'alias-coding-systems
) (list alias
))
1127 (put alias
'eol-type
(make-subsidiary-coding-system alias
)))
1128 (put alias
'eol-type eol-type
))))
1130 (defun merge-coding-systems (first second
)
1131 "Fill in any unspecified aspects of coding system FIRST from SECOND.
1132 Return the resulting coding system."
1133 (let ((base (coding-system-base second
))
1134 (eol (coding-system-eol-type second
)))
1135 ;; If FIRST doesn't specify text conversion, merge with that of SECOND.
1136 (if (eq (coding-system-base first
) 'undecided
)
1137 (setq first
(coding-system-change-text-conversion first base
)))
1138 ;; If FIRST doesn't specify eol conversion, merge with that of SECOND.
1139 (if (and (vectorp (coding-system-eol-type first
))
1140 (numberp eol
) (>= eol
0) (<= eol
2))
1141 (setq first
(coding-system-change-eol-conversion
1145 (defun autoload-coding-system (symbol form
)
1146 "Define SYMBOL as a coding-system that is defined on demand.
1148 FROM is a form to evaluate to define the coding-system."
1149 (put symbol
'coding-system-define-form form
)
1150 (setq coding-system-alist
(cons (list (symbol-name symbol
))
1151 coding-system-alist
))
1152 (dolist (elt '("-unix" "-dos" "-mac"))
1153 (let ((name (concat (symbol-name symbol
) elt
)))
1154 (put (intern name
) 'coding-system-define-form form
)
1155 (setq coding-system-alist
(cons (list name
) coding-system-alist
)))))
1157 (defun set-buffer-file-coding-system (coding-system &optional force nomodify
)
1158 "Set the file coding-system of the current buffer to CODING-SYSTEM.
1159 This means that when you save the buffer, it will be converted
1160 according to CODING-SYSTEM. For a list of possible values of CODING-SYSTEM,
1161 use \\[list-coding-systems].
1163 If CODING-SYSTEM leaves the text conversion unspecified, or if it
1164 leaves the end-of-line conversion unspecified, FORCE controls what to
1165 do. If FORCE is nil, get the unspecified aspect (or aspects) from the
1166 buffer's previous `buffer-file-coding-system' value (if it is
1167 specified there). Otherwise, leave it unspecified.
1169 This marks the buffer modified so that the succeeding \\[save-buffer]
1170 surely saves the buffer with CODING-SYSTEM. From a program, if you
1171 don't want to mark the buffer modified, specify t for NOMODIFY.
1172 If you know exactly what coding system you want to use,
1173 just set the variable `buffer-file-coding-system' directly."
1174 (interactive "zCoding system for saving file (default nil): \nP")
1175 (check-coding-system coding-system
)
1176 (if (and coding-system buffer-file-coding-system
(null force
))
1178 (merge-coding-systems coding-system buffer-file-coding-system
)))
1179 (setq buffer-file-coding-system coding-system
)
1180 ;; This is in case of an explicit call. Normally, `normal-mode' and
1181 ;; `set-buffer-major-mode-hook' take care of setting the table.
1182 (if (fboundp 'ucs-set-table-for-input
) ; don't lose when building
1183 (ucs-set-table-for-input))
1185 (set-buffer-modified-p t
))
1186 (force-mode-line-update))
1188 (defun revert-buffer-with-coding-system (coding-system &optional force
)
1189 "Visit the current buffer's file again using coding system CODING-SYSTEM.
1190 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1192 If CODING-SYSTEM leaves the text conversion unspecified, or if it
1193 leaves the end-of-line conversion unspecified, FORCE controls what to
1194 do. If FORCE is nil, get the unspecified aspect (or aspects) from the
1195 buffer's previous `buffer-file-coding-system' value (if it is
1196 specified there). Otherwise, determine it from the file contents as
1197 usual for visiting a file."
1198 (interactive "zCoding system for visited file (default nil): \nP")
1199 (check-coding-system coding-system
)
1200 (if (and coding-system buffer-file-coding-system
(null force
))
1202 (merge-coding-systems coding-system buffer-file-coding-system
)))
1203 (let ((coding-system-for-read coding-system
))
1206 (defun set-file-name-coding-system (coding-system)
1207 "Set coding system for decoding and encoding file names to CODING-SYSTEM.
1208 It actually just set the variable `file-name-coding-system' (which
1209 see) to CODING-SYSTEM."
1210 (interactive "zCoding system for file names (default nil): ")
1211 (check-coding-system coding-system
)
1212 (if (and coding-system
1213 (coding-system-get coding-system
'ascii-incompatible
))
1214 (error "%s is not ASCII-compatible" coding-system
))
1215 (setq file-name-coding-system coding-system
))
1217 (defvar default-terminal-coding-system nil
1218 "Default value for the terminal coding system.
1219 This is normally set according to the selected language environment.
1220 See also the command `set-terminal-coding-system'.")
1222 (defun set-terminal-coding-system (coding-system)
1223 "Set coding system of your terminal to CODING-SYSTEM.
1224 All text output to the terminal will be encoded
1225 with the specified coding system.
1226 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1227 The default is determined by the selected language environment
1228 or by the previous use of this command."
1230 (list (let ((default (if (and (not (terminal-coding-system))
1231 default-terminal-coding-system
)
1232 default-terminal-coding-system
)))
1234 (format "Coding system for terminal display (default %s): "
1237 (if (and (not coding-system
)
1238 (not (terminal-coding-system)))
1239 (setq coding-system default-terminal-coding-system
))
1241 (setq default-terminal-coding-system coding-system
))
1242 (set-terminal-coding-system-internal coding-system
)
1243 (redraw-frame (selected-frame)))
1245 (defvar default-keyboard-coding-system nil
1246 "Default value of the keyboard coding system.
1247 This is normally set according to the selected language environment.
1248 See also the command `set-keyboard-coding-system'.")
1250 (defun set-keyboard-coding-system (coding-system)
1251 "Set coding system for keyboard input to CODING-SYSTEM.
1252 In addition, this command enables Encoded-kbd minor mode.
1253 \(If CODING-SYSTEM is nil, Encoded-kbd mode is turned off -- see
1254 `encoded-kbd-mode'.)
1255 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1256 The default is determined by the selected language environment
1257 or by the previous use of this command."
1259 (list (let ((default (if (and (not (keyboard-coding-system))
1260 default-keyboard-coding-system
)
1261 default-keyboard-coding-system
)))
1263 (format "Coding system for keyboard input (default %s): "
1266 (if (and (not coding-system
)
1267 (not (keyboard-coding-system)))
1268 (setq coding-system default-keyboard-coding-system
))
1270 (setq default-keyboard-coding-system coding-system
))
1271 (if (and coding-system
1272 (coding-system-get coding-system
'ascii-incompatible
))
1273 (error "%s is not ASCII-compatible" coding-system
))
1274 (set-keyboard-coding-system-internal coding-system
)
1275 (setq keyboard-coding-system coding-system
)
1276 (encoded-kbd-mode (if coding-system
1 0)))
1278 (defcustom keyboard-coding-system nil
1279 "Specify coding system for keyboard input.
1280 If you set this on a terminal which can't distinguish Meta keys from
1281 8-bit characters, you will have to use ESC to type Meta characters.
1282 See Info node `Terminal Coding' and Info node `Unibyte Mode'.
1284 On non-windowing terminals, this is set from the locale by default.
1286 Setting this variable directly does not take effect;
1287 use either \\[customize] or \\[set-keyboard-coding-system]."
1288 :type
'(coding-system :tag
"Coding system")
1289 :link
'(info-link "(emacs)Terminal Coding")
1290 :link
'(info-link "(emacs)Unibyte Mode")
1291 :set
(lambda (symbol value
)
1292 ;; Don't load encoded-kbd-mode unnecessarily.
1293 (if (or value
(boundp 'encoded-kbd-mode
))
1294 (set-keyboard-coding-system value
)
1295 (set-default 'keyboard-coding-system nil
))) ; must initialize
1300 (defun set-buffer-process-coding-system (decoding encoding
)
1301 "Set coding systems for the process associated with the current buffer.
1302 DECODING is the coding system to be used to decode input from the process,
1303 ENCODING is the coding system to be used to encode output to the process.
1305 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems]."
1307 "zCoding-system for output from the process: \nzCoding-system for input to the process: ")
1308 (let ((proc (get-buffer-process (current-buffer))))
1310 (error "No process")
1311 (check-coding-system decoding
)
1312 (check-coding-system encoding
)
1313 (set-process-coding-system proc decoding encoding
)))
1314 (force-mode-line-update))
1316 (defalias 'set-clipboard-coding-system
'set-selection-coding-system
)
1318 (defun set-selection-coding-system (coding-system)
1319 "Make CODING-SYSTEM used for communicating with other X clients.
1320 When sending or receiving text via cut_buffer, selection, and clipboard,
1321 the text is encoded or decoded by CODING-SYSTEM."
1322 (interactive "zCoding system for X selection: ")
1323 (check-coding-system coding-system
)
1324 (setq selection-coding-system coding-system
))
1326 ;; Coding system lastly specified by the command
1327 ;; set-next-selection-coding-system.
1328 (defvar last-next-selection-coding-system nil
)
1330 (defun set-next-selection-coding-system (coding-system)
1331 "Make CODING-SYSTEM used for the next communication with other X clients.
1332 This setting is effective for the next communication only."
1334 (list (read-coding-system
1335 (if last-next-selection-coding-system
1336 (format "Coding system for the next X selection (default %S): "
1337 last-next-selection-coding-system
)
1338 "Coding system for the next X selection: ")
1339 last-next-selection-coding-system
)))
1341 (setq last-next-selection-coding-system coding-system
)
1342 (setq coding-system last-next-selection-coding-system
))
1343 (check-coding-system coding-system
)
1345 (setq next-selection-coding-system coding-system
))
1347 (defun set-coding-priority (arg)
1348 "Set priority of coding categories according to ARG.
1349 ARG is a list of coding categories ordered by priority."
1351 (current-list (copy-sequence coding-category-list
)))
1352 ;; Check the validity of ARG while deleting coding categories in
1353 ;; ARG from CURRENT-LIST. We assume that CODING-CATEGORY-LIST
1354 ;; contains all coding categories.
1356 (if (or (null (get (car l
) 'coding-category-index
))
1357 (null (memq (car l
) current-list
)))
1358 (error "Invalid or duplicated element in argument: %s" arg
))
1359 (setq current-list
(delq (car l
) current-list
))
1361 ;; Update `coding-category-list' and return it.
1362 (setq coding-category-list
(append arg current-list
))
1363 (set-coding-priority-internal)))
1367 (defvar ctext-non-standard-encodings-alist
1368 '(("big5-0" big5
2 (chinese-big5-1 chinese-big5-2
))
1369 ("ISO8859-14" iso-8859-14
1 latin-iso8859-14
)
1370 ("ISO8859-15" iso-8859-15
1 latin-iso8859-15
))
1371 "Alist of non-standard encoding names vs the corresponding usages in CTEXT.
1373 It controls how extended segments of a compound text are handled
1374 by the coding system `compound-text-with-extensions'.
1376 Each element has the form (ENCODING-NAME CODING-SYSTEM N-OCTET CHARSET).
1378 ENCODING-NAME is an encoding name of an \"extended segments\".
1380 CODING-SYSTEM is the coding-system to encode (or decode) the
1381 characters into (or from) the extended segment.
1383 N-OCTET is the number of octets (bytes) that encodes a character
1384 in the segment. It can be 0 (meaning the number of octets per
1385 character is variable), 1, 2, 3, or 4.
1387 CHARSET is a charater set containing characters that are encoded
1388 in the segment. It can be a list of character sets. It can also
1389 be a char-table, in which case characters that have non-nil value
1390 in the char-table are the target.
1392 On decoding CTEXT, all encoding names listed here are recognized.
1394 On encoding CTEXT, encoding names in the variable
1395 `ctext-non-standard-encodings' (which see) and in the information
1396 listed for the current language environment under the key
1397 `ctext-non-standard-encodings' are used.")
1399 (defvar ctext-non-standard-encodings
1401 "List of non-standard encoding names used in extended segments of CTEXT.
1402 Each element must be one of the names listed in the variable
1403 `ctext-non-standard-encodings-alist' (which see).")
1405 (defvar ctext-non-standard-encodings-regexp
1406 (string-to-multibyte
1408 ;; For non-standard encodings.
1409 "\\(\e%/[0-4][\200-\377][\200-\377]\\([^\002]+\\)\002\\)"
1411 ;; For UTF-8 encoding.
1412 "\\(\e%G[^\e]*\e%@\\)")))
1414 ;; Functions to support "Non-Standard Character Set Encodings" defined
1415 ;; by the COMPOUND-TEXT spec. They also support "The UTF-8 encoding"
1416 ;; described in the section 7 of the documentation of COMPOUND-TEXT
1417 ;; distributed with XFree86.
1419 (defun ctext-post-read-conversion (len)
1420 "Decode LEN characters encoded as Compound Text with Extended Segments."
1423 (let ((case-fold-search nil
)
1424 (in-workbuf (string= (buffer-name) " *code-converting-work*"))
1425 last-coding-system-used
1428 (narrow-to-region (point) (+ (point) len
)))
1430 (set-buffer-multibyte t
))
1431 (while (re-search-forward ctext-non-standard-encodings-regexp
1433 (setq pos
(match-beginning 0))
1434 (if (match-beginning 1)
1435 ;; ESC % / [0-4] M L --ENCODING-NAME-- \002 --BYTES--
1436 (let* ((M (char-after (+ pos
4)))
1437 (L (char-after (+ pos
5)))
1438 (encoding (match-string 2))
1439 (encoding-info (assoc-string
1441 ctext-non-standard-encodings-alist t
))
1442 (coding (if encoding-info
1443 (nth 1 encoding-info
)
1444 (setq encoding
(intern (downcase encoding
)))
1445 (and (coding-system-p encoding
)
1447 (setq bytes
(- (+ (* (- M
128) 128) (- L
128))
1448 (- (point) (+ pos
6))))
1450 (delete-region pos
(point))
1451 (forward-char bytes
)
1452 (decode-coding-region (- (point) bytes
) (point) coding
)))
1453 ;; ESC % G --UTF-8-BYTES-- ESC % @
1455 (delete-region pos
(+ pos
3))
1456 (decode-coding-region pos
(point) 'utf-8
))))
1457 (goto-char (point-min))
1458 (- (point-max) (point)))))
1460 ;; Return a char table of extended segment usage for each character.
1461 ;; Each value of the char table is nil, one of the elements of
1462 ;; `ctext-non-standard-encodings-alist', or the symbol `utf-8'.
1464 (defun ctext-non-standard-encodings-table ()
1465 (let ((table (make-char-table 'translation-table
)))
1466 (aset table
(make-char 'mule-unicode-0100-24ff
) 'utf-8
)
1467 (aset table
(make-char 'mule-unicode-2500-33ff
) 'utf-8
)
1468 (aset table
(make-char 'mule-unicode-e000-ffff
) 'utf-8
)
1469 (dolist (encoding (reverse
1471 (get-language-info current-language-environment
1472 'ctext-non-standard-encodings
)
1473 ctext-non-standard-encodings
)))
1474 (let* ((slot (assoc encoding ctext-non-standard-encodings-alist
))
1475 (charset (nth 3 slot
)))
1477 (cond ((charsetp charset
)
1478 (aset table
(make-char charset
) slot
))
1480 (dolist (elt charset
)
1481 (aset table
(make-char elt
) slot
)))
1482 ((char-table-p charset
)
1483 (map-char-table #'(lambda (k v
)
1484 (if (and v
(> k
128)) (aset table k slot
)))
1488 (defun ctext-pre-write-conversion (from to
)
1489 "Encode characters between FROM and TO as Compound Text w/Extended Segments.
1491 If FROM is a string, or if the current buffer is not the one set up for us
1492 by encode-coding-string, generate a new temp buffer, insert the
1493 text, and convert it in the temporary buffer. Otherwise, convert in-place."
1495 (let ((workbuf (get-buffer-create " *code-conversion-work*")))
1496 ;; Setup a working buffer if necessary.
1497 (cond ((stringp from
)
1498 (set-buffer workbuf
)
1500 (set-buffer-multibyte (multibyte-string-p from
))
1502 ((not (eq (current-buffer) workbuf
))
1503 (let ((buf (current-buffer))
1504 (multibyte enable-multibyte-characters
))
1505 (set-buffer workbuf
)
1507 (set-buffer-multibyte multibyte
)
1508 (insert-buffer-substring buf from to
)))))
1510 ;; Now we can encode the whole buffer.
1511 (let ((encoding-table (ctext-non-standard-encodings-table))
1512 last-coding-system-used
1513 last-pos last-encoding-info
1514 encoding-info end-pos
)
1515 (goto-char (setq last-pos
(point-min)))
1516 (setq end-pos
(point-marker))
1517 (while (re-search-forward "[^\000-\177]+" nil t
)
1518 ;; Found a sequence of non-ASCII characters.
1519 (setq last-pos
(match-beginning 0)
1520 last-encoding-info
(aref encoding-table
(char-after last-pos
)))
1521 (set-marker end-pos
(match-end 0))
1522 (goto-char (1+ last-pos
))
1526 (if (< (point) end-pos
)
1527 (aref encoding-table
(following-char))))
1528 (unless (eq last-encoding-info encoding-info
)
1529 (cond ((consp last-encoding-info
)
1530 ;; Encode the previous range using an extended
1532 (let ((encoding-name (car last-encoding-info
))
1533 (coding-system (nth 1 last-encoding-info
))
1534 (noctets (nth 2 last-encoding-info
))
1536 (encode-coding-region last-pos
(point) coding-system
)
1537 (setq len
(+ (length encoding-name
) 1
1538 (- (point) last-pos
)))
1540 (goto-char last-pos
)
1541 (insert (string-to-multibyte
1542 (format "\e%%/%d%c%c%s\ 2"
1547 ((eq last-encoding-info
'utf-8
)
1548 ;; Encode the previous range using UTF-8 encoding
1550 (encode-coding-region last-pos
(point) 'mule-utf-8
)
1552 (goto-char last-pos
)
1555 (setq last-pos
(point)
1556 last-encoding-info encoding-info
))
1557 (if (< (point) end-pos
)
1559 (throw 'tag nil
)))))
1560 (set-marker end-pos nil
)
1561 (goto-char (point-min))))
1562 ;; Must return nil, as build_annotations_2 expects that.
1567 (defcustom auto-coding-alist
1568 '(("\\.\\(arc\\|zip\\|lzh\\|lha\\|zoo\\|[jew]ar\\|xpi\\)\\'" . no-conversion
)
1569 ("\\.\\(ARC\\|ZIP\\|LZH\\|LHA\\|ZOO\\|[JEW]AR\\|XPI\\)\\'" . no-conversion
)
1570 ("\\.\\(sx[dmicw]\\|odt\\|tar\\|tgz\\)\\'" . no-conversion
)
1571 ("\\.\\(gz\\|Z\\|bz\\|bz2\\|gpg\\)\\'" . no-conversion
)
1572 ("\\.\\(jpe?g\\|png\\|gif\\|tiff?\\|p[bpgn]m\\)\\'" . no-conversion
)
1573 ("/#[^/]+#\\'" . emacs-mule
))
1574 "Alist of filename patterns vs corresponding coding systems.
1575 Each element looks like (REGEXP . CODING-SYSTEM).
1576 A file whose name matches REGEXP is decoded by CODING-SYSTEM on reading.
1578 The settings in this alist take priority over `coding:' tags
1579 in the file (see the function `set-auto-coding')
1580 and the contents of `file-coding-system-alist'."
1583 :type
'(repeat (cons (regexp :tag
"File name regexp")
1584 (symbol :tag
"Coding system"))))
1586 (defcustom auto-coding-regexp-alist
1587 '(("^BABYL OPTIONS:[ \t]*-\\*-[ \t]*rmail[ \t]*-\\*-" . no-conversion
)
1588 ("\\`\xFE\xFF" . utf-16be-with-signature
)
1589 ("\\`\xFF\xFE" . utf-16le-with-signature
)
1590 ("\\`\xEF\xBB\xBF" . utf-8
))
1591 "Alist of patterns vs corresponding coding systems.
1592 Each element looks like (REGEXP . CODING-SYSTEM).
1593 A file whose first bytes match REGEXP is decoded by CODING-SYSTEM on reading.
1595 The settings in this alist take priority over `coding:' tags
1596 in the file (see the function `set-auto-coding')
1597 and the contents of `file-coding-system-alist'."
1600 :type
'(repeat (cons (regexp :tag
"Regexp")
1601 (symbol :tag
"Coding system"))))
1603 (defun auto-coding-regexp-alist-lookup (from to
)
1604 "Lookup `auto-coding-regexp-alist' for the contents of the current buffer.
1605 The value is a coding system is specified for the region FROM and TO,
1609 (let ((alist auto-coding-regexp-alist
)
1611 (while (and alist
(not coding-system
))
1612 (let ((regexp (car (car alist
))))
1613 (if enable-multibyte-characters
1614 (setq regexp
(string-to-multibyte regexp
)))
1615 (if (re-search-forward regexp to t
)
1616 (setq coding-system
(cdr (car alist
)))
1617 (setq alist
(cdr alist
)))))
1620 ;; See the bottom of this file for built-in auto coding functions.
1621 (defcustom auto-coding-functions
'(sgml-xml-auto-coding-function
1622 sgml-html-meta-auto-coding-function
)
1623 "A list of functions which attempt to determine a coding system.
1625 Each function in this list should be written to operate on the
1626 current buffer, but should not modify it in any way. The buffer
1627 will contain undecoded text of parts of the file. Each function
1628 should take one argument, SIZE, which says how many
1629 characters (starting from point) it should look at.
1631 If one of these functions succeeds in determining a coding
1632 system, it should return that coding system. Otherwise, it
1635 If a file has a `coding:' tag, that takes precedence over these
1636 functions, so they won't be called at all."
1639 :type
'(repeat function
))
1641 (defvar set-auto-coding-for-load nil
1642 "Non-nil means look for `load-coding' property instead of `coding'.
1643 This is used for loading and byte-compiling Emacs Lisp files.")
1645 (defun auto-coding-alist-lookup (filename)
1646 "Return the coding system specified by `auto-coding-alist' for FILENAME."
1647 (let ((alist auto-coding-alist
)
1648 (case-fold-search (memq system-type
'(vax-vms windows-nt ms-dos cygwin
)))
1650 (while (and alist
(not coding-system
))
1651 (if (string-match (car (car alist
)) filename
)
1652 (setq coding-system
(cdr (car alist
)))
1653 (setq alist
(cdr alist
))))
1656 (put 'enable-character-translation
'permanent-local t
)
1657 (put 'enable-character-translation
'safe-local-variable
'booleanp
)
1659 (defun find-auto-coding (filename size
)
1660 "Find a coding system for a file FILENAME of which SIZE bytes follow point.
1661 These bytes should include at least the first 1k of the file
1662 and the last 3k of the file, but the middle may be omitted.
1664 The function checks FILENAME against the variable `auto-coding-alist'.
1665 If FILENAME doesn't match any entries in the variable, it checks the
1666 contents of the current buffer following point against
1667 `auto-coding-regexp-alist'. If no match is found, it checks for a
1668 `coding:' tag in the first one or two lines following point. If no
1669 `coding:' tag is found, it checks any local variables list in the last
1670 3K bytes out of the SIZE bytes. Finally, if none of these methods
1671 succeed, it checks to see if any function in `auto-coding-functions'
1674 If a coding system is specifed, the return value is a
1675 cons (CODING . SOURCE), where CODING is the specified coding
1676 system and SOURCE is a symbol `auto-coding-alist',
1677 `auto-coding-regexp-alist', `coding:', or `auto-coding-functions'
1678 indicating by what CODING is specified. Note that the validity
1679 of CODING is not checked; it's the caller's responsibility to check it.
1681 If nothing is specified, the return value is nil."
1682 (or (let ((coding-system (auto-coding-alist-lookup filename
)))
1684 (cons coding-system
'auto-coding-alist
)))
1685 ;; Try using `auto-coding-regexp-alist'.
1686 (let ((coding-system (auto-coding-regexp-alist-lookup (point)
1689 (cons coding-system
'auto-coding-regexp-alist
)))
1690 (let* ((case-fold-search t
)
1691 (head-start (point))
1692 (head-end (+ head-start
(min size
1024)))
1693 (tail-start (+ head-start
(max (- size
3072) 0)))
1694 (tail-end (+ head-start size
))
1695 coding-system head-found tail-found pos char-trans
)
1696 ;; Try a short cut by searching for the string "coding:"
1697 ;; and for "unibyte:" at the head and tail of SIZE bytes.
1698 (setq head-found
(or (search-forward "coding:" head-end t
)
1699 (search-forward "unibyte:" head-end t
)
1700 (search-forward "enable-character-translation:"
1702 (if (and head-found
(> head-found tail-start
))
1703 ;; Head and tail are overlapped.
1704 (setq tail-found head-found
)
1705 (goto-char tail-start
)
1706 (setq tail-found
(or (search-forward "coding:" tail-end t
)
1707 (search-forward "unibyte:" tail-end t
)
1708 (search-forward "enable-character-translation:"
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))))
1728 (when (re-search-forward
1729 "\\(.*;\\)?[ \t]*enable-character-translation:[ \t]*\\([^ ;]+\\)"
1731 (setq char-trans
(match-string 2)))))
1733 ;; If no coding: tag in the head, check the tail.
1734 ;; Here we must pay attention to the case that the end-of-line
1735 ;; is just "\r" and we can't use "^" nor "$" in regexp.
1736 (when (and tail-found
(or (not coding-system
) (not char-trans
)))
1737 (goto-char tail-start
)
1738 (re-search-forward "[\r\n]\^L" tail-end t
)
1739 (if (re-search-forward
1740 "[\r\n]\\([^[\r\n]*\\)[ \t]*Local Variables:[ \t]*\\([^\r\n]*\\)[\r\n]"
1742 ;; The prefix is what comes before "local variables:" in its
1743 ;; line. The suffix is what comes after "local variables:"
1745 (let* ((prefix (regexp-quote (match-string 1)))
1746 (suffix (regexp-quote (match-string 2)))
1750 ;; N.B. without the \n below, the regexp can
1752 "[ \t]*coding[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1757 "[ \t]*unibyte[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1762 "[ \t]*enable-character-translation[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1765 (concat "[\r\n]" prefix
"[ \t]*End *:[ \t]*" suffix
1768 (forward-char -
1) ; skip back \r or \n.
1769 (re-search-forward re-end tail-end
'move
)
1770 (setq tail-end
(point))
1772 (when (and set-auto-coding-for-load
1773 (re-search-forward re-unibyte tail-end t
))
1774 (setq coding-system
'raw-text
))
1775 (when (and (not coding-system
)
1776 (re-search-forward re-coding tail-end t
))
1777 (setq coding-system
(intern (match-string 1))))
1778 (when (and (not char-trans
)
1779 (re-search-forward re-char-trans tail-end t
))
1780 (setq char-trans
(match-string 1))))))
1782 ;; If the coding-system name ends with "!", remove it and
1783 ;; set char-trans to "nil".
1784 (let ((name (symbol-name coding-system
)))
1785 (if (= (aref name
(1- (length name
))) ?
!)
1786 (setq coding-system
(intern (substring name
0 -
1))
1787 char-trans
"nil"))))
1788 (when (and char-trans
1789 (not (setq char-trans
(intern char-trans
))))
1790 (make-local-variable 'enable-character-translation
)
1791 (setq enable-character-translation nil
))
1793 (cons coding-system
:coding
)))
1794 ;; Finally, try all the `auto-coding-functions'.
1795 (let ((funcs auto-coding-functions
)
1796 (coding-system nil
))
1797 (while (and funcs
(not coding-system
))
1798 (setq coding-system
(condition-case e
1800 (goto-char (point-min))
1801 (funcall (pop funcs
) size
))
1804 (cons coding-system
'auto-coding-functions
)))))
1806 (defun set-auto-coding (filename size
)
1807 "Return coding system for a file FILENAME of which SIZE bytes follow point.
1808 See `find-auto-coding' for how the coding system is found.
1809 Return nil if an invalid coding system is found.
1811 The variable `set-auto-coding-function' (which see) is set to this
1812 function by default."
1813 (let ((found (find-auto-coding filename size
)))
1814 (if (and found
(coding-system-p (car found
)))
1817 (setq set-auto-coding-function
'set-auto-coding
)
1819 ;; This variable is set in these two cases:
1820 ;; (1) A file is read by a coding system specified explicitly.
1821 ;; after-insert-file-set-coding sets this value to
1822 ;; coding-system-for-read.
1823 ;; (2) A buffer is saved.
1824 ;; After writing, basic-save-buffer-1 sets this value to
1825 ;; last-coding-system-used.
1826 ;; This variable is used for decoding in revert-buffer.
1827 (defvar buffer-file-coding-system-explicit nil
1828 "The file coding system explicitly specified for the current buffer.
1829 Internal use only.")
1830 (make-variable-buffer-local 'buffer-file-coding-system-explicit
)
1831 (put 'buffer-file-coding-system-explicit
'permanent-local t
)
1833 (defun after-insert-file-set-coding (inserted &optional visit
)
1834 "Set `buffer-file-coding-system' of current buffer after text is inserted.
1835 INSERTED is the number of characters that were inserted, as figured
1836 in the situation before this function. Return the number of characters
1837 inserted, as figured in the situation after. The two numbers can be
1838 different if the buffer has become unibyte.
1839 The optional second arg VISIT non-nil means that we are visiting a file."
1841 coding-system-for-read
1842 (not (eq coding-system-for-read
'auto-save-coding
)))
1843 (setq buffer-file-coding-system-explicit coding-system-for-read
))
1844 (if last-coding-system-used
1845 (let ((coding-system
1846 (find-new-buffer-file-coding-system last-coding-system-used
))
1847 (modified-p (buffer-modified-p)))
1849 ;; Tell set-buffer-file-coding-system not to mark the file
1850 ;; as modified; we just read it, and it's supposed to be unmodified.
1851 ;; Marking it modified would try to lock it, which would
1852 ;; check the modtime, and we don't want to do that again now.
1853 (set-buffer-file-coding-system coding-system t t
)
1854 (if (and enable-multibyte-characters
1855 (or (eq coding-system
'no-conversion
)
1856 (eq (coding-system-type coding-system
) 5))
1857 ;; If buffer was unmodified and the size is the
1858 ;; same as INSERTED, we must be visiting it.
1860 (= (buffer-size) inserted
))
1861 ;; For coding systems no-conversion and raw-text...,
1862 ;; edit the buffer as unibyte.
1863 (let ((pos-marker (copy-marker (+ (point) inserted
)))
1865 (buffer-file-name nil
))
1867 ;; If we're doing this for find-file,
1868 ;; don't record undo info; this counts as
1869 ;; part of producing the buffer's initial contents.
1870 (let ((buffer-undo-list t
))
1871 (set-buffer-multibyte nil
))
1872 (set-buffer-multibyte nil
))
1873 (setq inserted
(- pos-marker
(point)))))
1874 (restore-buffer-modified-p modified-p
))))
1877 ;; The coding-spec and eol-type of coding-system returned is decided
1878 ;; independently in the following order.
1879 ;; 1. That of buffer-file-coding-system locally bound.
1880 ;; 2. That of CODING.
1882 (defun find-new-buffer-file-coding-system (coding)
1883 "Return a coding system for a buffer when a file of CODING is inserted.
1884 The local variable `buffer-file-coding-system' of the current buffer
1885 is set to the returned value.
1886 Return nil if there's no need to set `buffer-file-coding-system'."
1887 (let (local-coding local-eol
1888 found-coding found-eol
1891 ;; Nothing found about coding.
1894 ;; Get information of `buffer-file-coding-system' in LOCAL-EOL
1895 ;; and LOCAL-CODING.
1896 (setq local-eol
(coding-system-eol-type buffer-file-coding-system
))
1897 (if (null (numberp local-eol
))
1898 ;; But eol-type is not yet set.
1899 (setq local-eol nil
))
1900 (if (and buffer-file-coding-system
1901 (not (eq (coding-system-type buffer-file-coding-system
) t
)))
1902 ;; This is not `undecided'.
1903 (setq local-coding
(coding-system-base buffer-file-coding-system
)))
1905 (if (and (local-variable-p 'buffer-file-coding-system
)
1906 local-eol local-coding
)
1907 ;; The current buffer has already set full coding-system, we
1908 ;; had better not change it.
1911 (setq found-eol
(coding-system-eol-type coding
))
1912 (if (null (numberp found-eol
))
1913 ;; But eol-type is not found.
1914 ;; If EOL conversions are inhibited, force unix eol-type.
1915 (setq found-eol
(if inhibit-eol-conversion
0)))
1916 (if (eq (coding-system-type coding
) t
)
1917 (setq found-coding
'undecided
)
1918 (setq found-coding
(coding-system-base coding
)))
1920 (if (and (not found-eol
) (eq found-coding
'undecided
))
1921 ;; No valid coding information found.
1924 ;; Some coding information (eol or text) found.
1926 ;; The local setting takes precedence over the found one.
1927 (setq new-coding
(if (local-variable-p 'buffer-file-coding-system
)
1928 (or local-coding found-coding
)
1929 (or found-coding local-coding
)))
1930 (setq new-eol
(if (local-variable-p 'buffer-file-coding-system
)
1931 (or local-eol found-eol
)
1932 (or found-eol local-eol
)))
1934 (let ((eol-type (coding-system-eol-type new-coding
)))
1935 (if (and (numberp new-eol
) (vectorp eol-type
))
1936 (aref eol-type new-eol
)
1939 (defun modify-coding-system-alist (target-type regexp coding-system
)
1940 "Modify one of look up tables for finding a coding system on I/O operation.
1941 There are three of such tables, `file-coding-system-alist',
1942 `process-coding-system-alist', and `network-coding-system-alist'.
1944 TARGET-TYPE specifies which of them to modify.
1945 If it is `file', it affects `file-coding-system-alist' (which see).
1946 If it is `process', it affects `process-coding-system-alist' (which see).
1947 If it is `network', it affects `network-coding-system-alist' (which see).
1949 REGEXP is a regular expression matching a target of I/O operation.
1950 The target is a file name if TARGET-TYPE is `file', a program name if
1951 TARGET-TYPE is `process', or a network service name or a port number
1952 to connect to if TARGET-TYPE is `network'.
1954 CODING-SYSTEM is a coding system to perform code conversion on the I/O
1955 operation, or a cons cell (DECODING . ENCODING) specifying the coding systems
1956 for decoding and encoding respectively,
1957 or a function symbol which, when called, returns such a cons cell."
1958 (or (memq target-type
'(file process network
))
1959 (error "Invalid target type: %s" target-type
))
1960 (or (stringp regexp
)
1961 (and (eq target-type
'network
) (integerp regexp
))
1962 (error "Invalid regular expression: %s" regexp
))
1963 (if (symbolp coding-system
)
1964 (if (not (fboundp coding-system
))
1966 (check-coding-system coding-system
)
1967 (setq coding-system
(cons coding-system coding-system
))))
1968 (check-coding-system (car coding-system
))
1969 (check-coding-system (cdr coding-system
)))
1970 (cond ((eq target-type
'file
)
1971 (let ((slot (assoc regexp file-coding-system-alist
)))
1973 (setcdr slot coding-system
)
1974 (setq file-coding-system-alist
1975 (cons (cons regexp coding-system
)
1976 file-coding-system-alist
)))))
1977 ((eq target-type
'process
)
1978 (let ((slot (assoc regexp process-coding-system-alist
)))
1980 (setcdr slot coding-system
)
1981 (setq process-coding-system-alist
1982 (cons (cons regexp coding-system
)
1983 process-coding-system-alist
)))))
1985 (let ((slot (assoc regexp network-coding-system-alist
)))
1987 (setcdr slot coding-system
)
1988 (setq network-coding-system-alist
1989 (cons (cons regexp coding-system
)
1990 network-coding-system-alist
)))))))
1992 (defun decode-coding-inserted-region (from to filename
1993 &optional visit beg end replace
)
1994 "Decode the region between FROM and TO as if it is read from file FILENAME.
1995 The idea is that the text between FROM and TO was just inserted somehow.
1996 Optional arguments VISIT, BEG, END, and REPLACE are the same as those
1997 of the function `insert-file-contents'.
1998 Part of the job of this function is setting `buffer-undo-list' appropriately."
2001 (let ((coding coding-system-for-read
)
2004 ;; Temporarily turn off undo recording, if we're decoding the
2005 ;; text of a visited file.
2006 (setq buffer-undo-list t
)
2007 ;; Otherwise, if we can recognize the undo elt for the insertion,
2008 ;; remove it and get ready to replace it later.
2009 ;; In the mean time, turn off undo recording.
2010 (let ((last (car-safe buffer-undo-list
)))
2011 (if (and (consp last
) (eql (car last
) from
) (eql (cdr last
) to
))
2012 (setq undo-list-saved
(cdr buffer-undo-list
)
2013 buffer-undo-list t
))))
2014 (narrow-to-region from to
)
2015 (goto-char (point-min))
2017 (setq coding
(funcall set-auto-coding-function
2018 filename
(- (point-max) (point-min)))))
2020 (setq coding
(car (find-operation-coding-system
2021 'insert-file-contents
2022 (cons filename
(current-buffer))
2023 visit beg end replace
))))
2024 (if (coding-system-p coding
)
2025 (or enable-multibyte-characters
2027 (coding-system-change-text-conversion coding
'raw-text
)))
2030 (decode-coding-region (point-min) (point-max) coding
)
2031 (setq last-coding-system-used coding
))
2032 ;; If we're decoding the text of a visited file,
2033 ;; the undo list should start out empty.
2035 (setq buffer-undo-list nil
)
2036 ;; If we decided to replace the undo entry for the insertion,
2039 (setq buffer-undo-list
2040 (cons (cons from
(point-max)) undo-list-saved
))))))))
2042 (defun recode-region (start end new-coding coding
)
2043 "Re-decode the region (previously decoded by CODING) by NEW-CODING."
2045 (list (region-beginning) (region-end)
2046 (read-coding-system "Text was really in: ")
2047 (let ((coding (or buffer-file-coding-system last-coding-system-used
)))
2049 (concat "But was interpreted as"
2050 (if coding
(format " (default %S): " coding
) ": "))
2052 (or (and new-coding coding
)
2053 (error "Coding system not specified"))
2054 ;; Check it before we encode the region.
2055 (check-coding-system new-coding
)
2057 (narrow-to-region start end
)
2058 (encode-coding-region (point-min) (point-max) coding
)
2059 (decode-coding-region (point-min) (point-max) new-coding
)))
2061 (defun make-translation-table (&rest args
)
2062 "Make a translation table from arguments.
2063 A translation table is a char table intended for character
2064 translation in CCL programs.
2066 Each argument is a list of elements of the form (FROM . TO), where FROM
2067 is a character to be translated to TO.
2069 FROM can be a generic character (see `make-char'). In this case, TO is
2070 a generic character containing the same number of characters, or an
2071 ordinary character. If FROM and TO are both generic characters, all
2072 characters belonging to FROM are translated to characters belonging to TO
2073 without changing their position code(s).
2075 The arguments and forms in each argument are processed in the given
2076 order, and if a previous form already translates TO to some other
2077 character, say TO-ALT, FROM is also translated to TO-ALT."
2078 (let ((table (make-char-table 'translation-table
))
2081 (let ((elts (car args
)))
2083 (let* ((from (car (car elts
)))
2084 (from-i 0) ; degree of freedom of FROM
2085 (from-rev (nreverse (split-char from
)))
2086 (to (cdr (car elts
)))
2087 (to-i 0) ; degree of freedom of TO
2088 (to-rev (nreverse (split-char to
))))
2089 ;; Check numbers of heading 0s in FROM-REV and TO-REV.
2090 (while (eq (car from-rev
) 0)
2091 (setq from-i
(1+ from-i
) from-rev
(cdr from-rev
)))
2092 (while (eq (car to-rev
) 0)
2093 (setq to-i
(1+ to-i
) to-rev
(cdr to-rev
)))
2094 (if (and (/= from-i to-i
) (/= to-i
0))
2095 (error "Invalid character pair (%d . %d)" from to
))
2096 ;; If we have already translated TO to TO-ALT, FROM should
2097 ;; also be translated to TO-ALT. But, this is only if TO
2098 ;; is a generic character or TO-ALT is not a generic
2100 (let ((to-alt (aref table to
)))
2102 (or (> to-i
0) (not (generic-char-p to-alt
))))
2105 (set-char-table-default table from to
)
2106 (aset table from to
))
2107 ;; If we have already translated some chars to FROM, they
2108 ;; should also be translated to TO.
2109 (let ((l (assq from revlist
)))
2116 (setq l
(cdr l
)) ))))
2117 ;; Now update REVLIST.
2118 (let ((l (assq to revlist
)))
2120 (setcdr l
(cons from
(cdr l
)))
2121 (setq revlist
(cons (list to from
) revlist
)))))
2122 (setq elts
(cdr elts
))))
2123 (setq args
(cdr args
)))
2124 ;; Return TABLE just created.
2127 (defun make-translation-table-from-vector (vec)
2128 "Make translation table from decoding vector VEC.
2129 VEC is an array of 256 elements to map unibyte codes to multibyte
2130 characters. Elements may be nil for undefined code points.
2131 See also the variable `nonascii-translation-table'."
2132 (let ((table (make-char-table 'translation-table
))
2133 (rev-table (make-char-table 'translation-table
))
2136 (setq ch
(aref vec i
))
2140 (aset rev-table ch i
))))
2141 (set-char-table-extra-slot table
0 rev-table
)
2144 (defun define-translation-table (symbol &rest args
)
2145 "Define SYMBOL as the name of translation table made by ARGS.
2146 This sets up information so that the table can be used for
2147 translations in a CCL program.
2149 If the first element of ARGS is a char-table whose purpose is
2150 `translation-table', just define SYMBOL to name it. (Note that this
2151 function does not bind SYMBOL.)
2153 Any other ARGS should be suitable as arguments of the function
2154 `make-translation-table' (which see).
2156 This function sets properties `translation-table' and
2157 `translation-table-id' of SYMBOL to the created table itself and the
2158 identification number of the table respectively. It also registers
2159 the table in `translation-table-vector'."
2160 (let ((table (if (and (char-table-p (car args
))
2161 (eq (char-table-subtype (car args
))
2162 'translation-table
))
2164 (apply 'make-translation-table args
)))
2165 (len (length translation-table-vector
))
2168 (put symbol
'translation-table table
)
2171 (setq translation-table-vector
2172 (vconcat translation-table-vector
(make-vector len nil
))))
2173 (let ((slot (aref translation-table-vector id
)))
2175 (eq (car slot
) symbol
))
2177 (aset translation-table-vector id
(cons symbol table
))
2179 (setq id
(1+ id
)))))
2180 (put symbol
'translation-table-id id
)
2183 (defun translate-region (start end table
)
2184 "From START to END, translate characters according to TABLE.
2185 TABLE is a string or a char-table.
2186 If TABLE is a string, the Nth character in it is the mapping
2187 for the character with code N.
2188 If TABLE is a char-table, the element for character N is the mapping
2189 for the character with code N.
2190 It returns the number of characters changed."
2192 (list (region-beginning)
2195 (dotimes (i (length translation-table-vector
))
2196 (if (consp (aref translation-table-vector i
))
2197 (push (list (symbol-name
2198 (car (aref translation-table-vector i
)))) l
)))
2200 (error "No translation table defined"))
2202 (setq table
(completing-read "Translation table: " l nil t
)))
2205 (let ((val (get table
'translation-table
)))
2206 (or (char-table-p val
)
2207 (error "Invalid translation table name: %s" table
))
2209 (translate-region-internal start end table
))
2211 (put 'with-category-table
'lisp-indent-function
1)
2213 (defmacro with-category-table
(table &rest body
)
2214 "Evaluate BODY with category table of current buffer set to TABLE.
2215 The category table of the current buffer is saved, BODY is evaluated,
2216 then the saved table is restored, even in case of an abnormal exit.
2217 Value is what BODY returns."
2218 (let ((old-table (make-symbol "old-table"))
2219 (old-buffer (make-symbol "old-buffer")))
2220 `(let ((,old-table
(category-table))
2221 (,old-buffer
(current-buffer)))
2224 (set-category-table ,table
)
2226 (with-current-buffer ,old-buffer
2227 (set-category-table ,old-table
))))))
2229 (defun define-translation-hash-table (symbol table
)
2230 "Define SYMBOL as the name of the hash translation TABLE for use in CCL.
2232 Analogous to `define-translation-table', but updates
2233 `translation-hash-table-vector' and the table is for use in the CCL
2234 `lookup-integer' and `lookup-character' functions."
2235 (unless (and (symbolp symbol
)
2236 (hash-table-p table
))
2237 (error "Bad args to define-translation-hash-table"))
2238 (let ((len (length translation-hash-table-vector
))
2241 (put symbol
'translation-hash-table table
)
2244 (setq translation-hash-table-vector
2245 (vconcat translation-hash-table-vector
[nil])))
2246 (let ((slot (aref translation-hash-table-vector id)))
2248 (eq (car slot) symbol))
2250 (aset translation-hash-table-vector id (cons symbol table))
2252 (setq id (1+ id)))))
2253 (put symbol 'translation-hash-table-id id)
2256 ;;; Initialize some variables.
2258 (put 'use-default-ascent 'char-table-extra-slots 0)
2259 (setq use-default-ascent (make-char-table 'use-default-ascent))
2260 (put 'ignore-relative-composition 'char-table-extra-slots 0)
2261 (setq ignore-relative-composition
2262 (make-char-table 'ignore-relative-composition))
2265 ;;; Built-in auto-coding-functions:
2267 (defun sgml-xml-auto-coding-function (size)
2268 "Determine whether the buffer is XML, and if so, its encoding.
2269 This function is intended to be added to `auto-coding-functions'."
2270 (setq size (+ (point) size))
2271 (when (re-search-forward "\\`[[:space:]\n]*<\\?xml" size t)
2272 (let ((end (save-excursion
2274 (re-search-forward "[\"']\\s-*\\?>" size t))))
2276 (if (re-search-forward "encoding=[\"']\\(.+?\\)[\"']" end t)
2277 (let* ((match (match-string 1))
2278 (sym (intern (downcase match))))
2279 (if (coding-system-p sym)
2281 (message "Warning: unknown coding system \"%s\"" match)
2285 (defun sgml-html-meta-auto-coding-function (size)
2286 "If the buffer has an HTML meta tag, use it to determine encoding.
2287 This function is intended to be added to `auto-coding-functions'."
2288 (setq size (min (+ (point) size)
2290 ;; Limit the search by the end of the HTML header.
2291 (or (search-forward "</head>" size t)
2292 ;; In case of no header, search only 10 lines.
2295 ;; Make sure that the buffer really contains an HTML document, by
2296 ;; checking that it starts with a doctype or a <HTML> start tag
2297 ;; (allowing for whitespace at bob). Note: 'DOCTYPE NETSCAPE' is
2298 ;; useful for Mozilla bookmark files.
2299 (when (and (re-search-forward "\\`[[:space:]\n]*\\(<!doctype[[:space:]\n]+\\(html\\|netscape\\)\\|<html\\)" size t)
2300 (re-search-forward "<meta\\s-+http-equiv=[\"']?content-type[\"']?\\s-+content=[\"']text/\\sw+;\\s-*charset=\\(.+?\\)[\"']" size t))
2301 (let* ((match (match-string 1))
2302 (sym (intern (downcase match))))
2303 (if (coding-system-p sym)
2305 (message "Warning: unknown coding system \"%s\"" match)
2311 ;; arch-tag: 9aebaa6e-0e8a-40a9-b857-cb5d04a39e7c
2312 ;;; mule.el ends here